CONTROL ENGINEERING As perNew Syllabus of VTU For Vill Semester Mechanical Engineering Also useful for M.E, / M.Tech. of all other Universities S.B. Halesh teen. ono) more Associate Professor Department of techanical Engineering, Sir M, Visvesvaraya Institute of Technology Bengaluru-562187 J.K, XEROX Kannada Sahitya Bhavan Road "Rear VNCE., Vilayanagara ‘Mysore-570016 Mob :9141720824,9845002201Control Engineering - For Engineering Students, by S.B. Halesh and published by ‘Sapna Book House (P) Ltd., R.O. #11, 3rd Main Road, Gandhinagar, Bangalore- 560 009 Ph: 080-40114455 Engg. - 183] ISBN 978-81-280-1361-5 © Author “No part ofthis publication whichis material protected by this copyright noice may be reproduced or ‘ransmitted or utilized or stored in any form or by any means now known of hereinafter invented, electronic, digital or mechanical, including photocopying, scanning, recording or by any information storage or retrieval system, without prior written permission from the publisher" ‘Copying ofthe book and selling it after photocopying or reselling it as second hand book is illegl and is not allowed, under the copyright at” This bookis sold subject tothe condition that it shall not, by way of trade or otherwise, be ent, resold hired out, or otherwise circulated without the publisher's prior writen consent. First Elton : Jan.,2014 Pages : xil+ 417 No.OfCopies : 1000 Paper Used : 11.2Kg(58 gsm) Deluxe Maplitho anata Book Size 418" Crown [FoR COPIES CONTACT OUR FOLLOWING DISTRIBUTORS : 1. UBS are Denture Lt, argo, Ph ens183, zz0123 2, de Bek Ht, Barat Ph: aero, 225088 4 Prats Sana, Baral, Ph: pases, 2267050 4 Wiamand Bone Dept. Hat Pe: Ra6u9 5. ayaa Agencies, Huth Ph z082 Azo elabl at ot Showroom 4 + SADASHVNAGAR, Near Bhoshyan Ciclo, Bangs. Ph (0s) 2auoue 145 “FIAYANAGAR, 4 Bock, Banglore Px: (08) e05670 ++ INDIRANAGAR, Opp. ChinnayaKishna Temple, Bangalore Pr (os) asso ++ KORAMANGALA, 808 Rood, 7h Bock, Bangor. Ph 08) aonne0 + RESIDENCY ROAD, No 2. ACR Towers, Bangalore. Ph (0s) 6098 ++ MYSORE, Nesyan Shas Ross, Devaraje Mods Ph (oat 00499 4 MANGALORE, Excel Wal, KS. Rao Rowe. Pr (089 2250 Ph (088) 2409 “+ MUBL, Lox Mal, Coen Rose ireked= eine ie eaksron ac eieaantecunecae) ‘ypeseting and Cover Design by » CSYATALINE, Bangalore -20 email: [email protected] Printed at : Poornima Printers, Bangalore Dedicated to Wy Brothers S.8. Srinivas, $B. Ravindranath & S.B. PramodSYLLABUS : 1OMES2 Exam Hrs: 03 Exam Marks: 100 ‘Total Lecture Hrs: 52 PART- A UNIT-1 INTRODUCTION: Concept of automatic controls, Open loop and closed loopsystems, Concepts of feedback, requirements of an ideal control system. Types of controllers- Proportional, Integral Proportional Integral,Proportional integral Differential controllers. 07 Hours ‘UNIT -2 MATHEMATICAL MODELS: Transfer function models, models of mechanicalsystems, ‘models of electrical circuits, DC and AC motors in controlsystems, models of thermal systems, models of yale systems pneumatic system, Analogous systems Force voltage, Force current, fours uNIT-3 BLOCK DIAGRAMS AND SIGNAL FLOW GRAPHS: Transfer Functions definition, function, biock representation of systems elements, reduction of bockdiagrams, Signal flow graphs: Mason's gain form. 7 Hours unin -4 ; ‘TRANSIENT AND STEADY STATE RESPONSE ANALYSIS: Introduction, frst ardarand second order system response to step ramp and impulse inputs, concepsof time constant and is importance in peed of response. System sablity:Routh's Hurwitz Criterion, 6 Hours PART-B unir-5 FREQUENCY RESPONSE ANALYSIS: Polar plots, Nyquist Stability Criterion, Stability “Analysis, Relative stability concepts, phase and gain margin, Mand Ncirces. 6 Hours untr-6 ‘ FREQUENCY RESPONSE ANALYSIS USING BODE PLOTS: Bode attenuation digrams, Stability Analysis using Bode plots, Simplified Bode Diagrams. 7 Hours uNIT-7 ROOT LOCUS PLOTS: Definition of root loci, general rules for constructing ro0t loc, Analysis using oot locus plots 7 Hours uNIT-8 SYSTEM COMPENSATION AND STATE VARIABLE CHARACTERISTICS OF LINEAR SYSTEMS: Series and feedback compensation, Introduction to sate conceptsstte equation oflinear continuous data system. Matrix representation of stateeqations, controllably ant observabiliy, Kalman and Gilberts tes. 7 Hours CONTENTS UNIT 1 - INTRODUCTION 1-20 1.1 Introduction 1 1.2 Concept of Automatic Control 2 1.3. Terminology 2 1.4 Classification of Control Systems 3 1.5 Open Loop Control System 5 1.5.1. Advantages 5 1.5.2 Disadvantages 6 1.6 Closed Loop Control System 6 1.6.1 Advantages 6 1.6.2. Disadvantages 7 1.6.3 Application of the Closed Loop Control System 7 1.7 Concept of Feedback Control System 8 1.7.1 Effect of Feedback on the Contro! System 10 1.8 Requirement of an Ideal Control System 10 9 Controllers " 1.9.1 Types of Controllers "1 1.9.2 Proportional - Controller 12 1.9.3 Derivative Controller 13 1.9.4 Proportional Plus Derivative, Controller 14 1.9.5 Integral - Controller 15 1.9.6 Proportional Plus Integral Controller 16 1.9.7. Proportional Plus integral Plus Differential Controller 7 1.9.8 ON - OFF Controller 17 1.9.9 “Additions 18 Review Questions 19 UNIT 2 - MATHEMATICAL MODELS 21-69 2.1 Introduction 21 2.2 Transfer Function Mode! 21 2.3 Models of Mechanical Systems 22 2.3.1. Mechanical Translational System 222.3.2 Mechanical Rotational Systems 24 2.4 Models of Electrical System 25 2.5 DC and AC Motors in Control 26 2.5.1 Direct Current Motor 26 2.8.2. Alternating Current Motors 29 2.6 Models of Hydraulic Systems a1 2.7 Models of Pneumatic System 34 2.8 Models of Thermal System 36 2.9 Liquid Level Systems 36 2.10 Anafogous Systems 39 2.10.1. Force-Current Analogy ~ [F Analogy] 39 2.10.2 Force - Voltage Analogy - F-V Analogy 4 Review Questions UNIT 3 - BLOCK DIAGRAMS AND SIGNAL FLOW GRAPHS 70-127 70 3.1 Transfer Function 3.2 Block Diagram n 3.2.1. Basic Elements of 2 Block Diagram n 3.3. Block Diagram of a Closed Loop System 72 3.3.1. Transfer Function of Closed Loop System 72 3.4 Reduetion of Block Diagram 73 3.4.1 Block Diagram Reduction Rules 73 3.5. Signal Flow Graphs 99 3.6.1 Basic Elements of a SFG 99 3.5.2 Terms Used in SFG 100 3.5.3. Properties of SFG 101 3.8.4 Procedure to Draw SFG 101 3.6 Mason's Gain Formula 102 UNIT 4 - TRANSIENT AND STEADY STATE RESPONSE ANALYSIS 128 - 187 4.1. Introduction 128 4.2 Parts of Time Response 128 4.3 Standard Test Inputs 129 4 Order and Type of the Control System 131 131 4.4.1. Order of the System 4.4.2 Types of the System nal 4.8 Steady State Error and Error Constants 134 4.6 Effect of Standard Test Inputs on Steady State Errors or Error Constants 135 4.6.1. Unit Step Input and Positional Errors 135 4.6.2 Unit Ramp Input and Velocity Error 135 4.6.3 Unit Parabolic input and Acceleration 136 4.7 Steady-state Errors of TYPE - 0 unity feedback system 137 4.8 Steady-state Errors of TYPE 1 unit feed back system 137 4.9 Steady-state Errors of TYPE 2 unity feed back system 138 4.10 Transient Response : First order system 141 4.10.1. Unit Step Input 141 4.10.2 Unit Impulse Input 143 4.10.3 Unit Ramp Input 144 4.11 Transient Response : Second order system 144 4.12 Time Response Specification 148, 4.12.1. Expression for Peak Time (tp) 149 4.12.2 Expression for Rise Time (tr) 150 4.12.3 Expression for Maximum Overshoot (Mp) 151 4.12.4 Expression for Setting Time (ts) 152 4.13. System - Stability 167 4.13.1, Concept of Stability 167 4.13.2. Terms Used 167 4.13.3 Condition for Stability Analysis 168 4.13.4 Characteristic Equation 168 4.14 Routh's Hurwitz Criterion 169 4.14.1 Routh’s Array 169 4.14.2 Special Cases im Roviow Questions 186 UNIT 5 - FREQUENCY RESPONSE ANALYSIS 188 - 234 5.1 Introduction 188 5.2 Polar ~ Plots 188 5.3. Basic Terms used 189) 5.4 Nyquist Stability Criterion 190 5.4.1. Concept of Stability Criterion 192 5.5 Cauchy's Principle of Argument 192 5.6 Generalised Nyquist Path 1935,7 Stability and Relative Stability 194 5.8 Constant Magnitude Circle - M - circle 227 5.9 N- Circle (Constant phase shift circles) 230 233 Review questions UNIT 6 - FREQUENCY RESPONSE ANALYSIS USING BODE PLOTS 238 - 292 6.1 Introduction 236 6.2 Bode Plot For Open Loop Transfer Function Gl jw) Hi jw) 236 6.2.1 Contribution of Various Factors to Bode Plot 237 6.3 Frequency Response Specification 249 6.4 Advantages of Bode Plots 250 6.5 Procedure to draw the Bode Plots 251 Review Questions 291 UNIT 7 - ROOT LOCUS PLOTS 293 - 374 7.1. Introduction 293 7.1.1, The Root Locus Concept 293 7.1.2 Conditions of Root Locus 295 Magnitude Condition 296 -1.4 General Rules For Construction of Root Locus 297 Review Questions 373 UNIT 8 - SYSTEM COMPENSATION AND STATE VARIABLE CHARACTERISTICS OF LINEAR SYSTEMS. 375-401 8.1 Need for the System Compensation 375 8.2 Types of System Compensation 378 8.2.1. Series Compensated System or Series Compensation 376 8.2.2. Parallel Compensated System or Feedback Compensation 376 8.2.3 Series - Parallel compensated system 377 8.3 Types of Compensators 377 8.3.1 Lag Compensator 377 8.3.2 Characteristics of Lag Compensator 380 8.3.3 Lead Compensator 381 8.3.4 Characteristics of Lead Compensator 384 8.3.5 Lag - Lead Compensator 385 8.3.6 Characteristics of Lag - Lead Compensator 387 8.4 Introduction to State Concepts 388 8.4.1 System State and State Variable 389 8.4.2 State Vector and State Space 390 8.5 State Equations 390 8.5.1. Matrix Representation of State Equations 391 8.5.2 Output Equation 391 8.5.3. State Model 392 8.5.4 Block Diagram Representation of Linear Systems Described by State Equations 393 8.5.5 Transfer Function from the State Equation 394 8.6 Controllability and observability 396 8.7 Kalman's and Gilbert's Test 396 8.7.1. Kalman's Test of Controllability 396 8.7.2. Kalman’s Test of Observability 397 8.7.3. Gilbert's Test of Controllability and Observability 398 8.8 State Transition Matrix 399 Review Questions 401 QUESTION PAPERS. 402 - 416 REFERENCES AND FURTHER READING 417UNIT 1.1 INTRODUCTION Control engineering is the engineering discipline that applies control theory to design systems ‘with predictable behaviors, Multi-disciplinary in nature, control systems engineering activities {focus on implementation of control systems mainly derived by mathematical modeling of systems of adiverse range. Before it emerged as a unique discipline, control engineering was practiced as apart of mechanical engineering and control theory was studied as a part of electrical engineering, since electrical circuits can often be easily described using control theory techniques. In the very first control relationships, a current output was represented with a voltage control input. However, not having, proper technology to implement electrical control systems, designers left with the option of less efficient and slow responding mechanical systems. A very effective mechanical controller that is siill widely used in some hydro plants isthe governor. Later on, previous to modern power electronics, process control systems for industrial applications were devised by mechanical engineers using pneumatic and hydraulic control devices, many of which are still in use today. There are two major divisions in control theory, namely, classical and modern, which have direct implications over the control engineering applications. The scope of classical contro! theory is limited to single-input and single-output (SISO) system design, The system analysisis carried out in ‘ime domain using differential equations, in complex-s domain with Laplace transform orin frequency domain by transforming from the complex-s domain, All systems are assumed to be second order and single variable, and higher-order system responses and multivariable effects are ignored. A controller designed using classical theory usually requires on-site tuning due to design approximations. Yet, due to easier physical implementation of classical controller designs as compared to systems designed using modem control theory these controllers are preferred in most industrial applications ‘The most common controllers designed using classical control theory are PID controtlers. In contrast, modem control theory is carried out strictly inthe complex-s or the frequency domain, and can deal with multi-input and multi-output (MIMO) systems. This overcomes the limitations of classical control theory in more sophisticated design problems, such as fighter aircraft control. In2__—Gontrol Engineering modern design, a system is represented as a set of first order differential equations defined using state variables, Nonlinear, multivariable, adaptive and robust control theories come under this division. Being fairly new, modern control theory has many areas yet tobe explored. Scholars like Rudolf E, Kalman and Alexander Lyapunoy are well-known among the people who have shaped modem control theory. Modern day control engineering is a relatively new field of study that gained a significant attention during 20° century with the advancement in technology. It can be broadly defined as practical application of control theory, Control engineering has an essential rote in a wide range of control systems, from simple household washing machines to high-performance F-16 fighter aircraft. It seeks to understand physical systems, using mathematical modeling, in terms of inputs, outputs and various components with different behaviors; use control systems design tools to develop controllers forthose systems; and implement controllers in physical systems employing available technology. A system can be mechanical, electrical, fluid, chemical, financial and even biological, and the ‘mathematical modeling, analysis and controller design uses control theory in one or many of the time, frequency and complex:s domains, depending on the nature of the design problem, 1.2 CONCEPT OF AUTOMATIC CONTROL ‘The control of an industrial process (manufacturing, production, and processing by automatic rather than manual means is often called automation. Automation is the automatic operation or control of a process, device, and steel industries, among others. The concept of automation is, central to an industrial society. Automatic machines can be used to increase the productivity of a plant and to obtain high-quality products, Automatic control of machines and processes is utilised to prodeuce a product within specified tolemces and to achieve high precision. Automatic control has played a Vital role in the advancement of engineering and science. In addition to its extreme importance in space-vehicle, missle-guidance, robotic and mechatronics systems, automatic control has become an important and integral oart of modera manufacturing and industrail processes. For example, automatic control is essential in the numerical control of machine tools in the manufacturing industries, in the design of autopilot systems in aerispace industries, and in the design of vehicles in automobile industrs. It is also essential in industrial operations that require the control of pressure, temperature, humidity, viscosity, and fluid flow. Due to the importance of ‘automatic control as a means of attaining optimal performance of dynamic systems, improving productivity and relieving the drudgery of many routine repetitive manual operations, most engineers ‘nd scientist must now have a good understanding of this field. The current revolution in computer and information technology is causing an equally momentous social change: the expansion of information gathering and information processing as computer extend the reach of the human brain, Control systems are used to achieve increased performance of a device or system. 1.3 TERMINOLOGY Following are the basic terms used in the study of control engineering: ‘© Control : The meaning control is to regulate, direct or command a system so that a desired objective is achieved, i | | | { Introduction 9 + Sytem :A system isan arangementor combination of diferent bisaton of dierent physi component ha a onecied together of related together o form an ene unto achieve n conn oecing A system canbe both psc an am abstsct one For example -A cas oom napster and an abstract system can be dynam phenome sich economies edict sal + Input: Input isthe apie siglo extra exciton signa that tapi othe eon System to geta required output + Plant; Panis ke portion ofthe sytem which tobe contol or regia + Process: Iris an operation which st eoontrlled or eulted to oban the desied ouput * Controller: 1s an cls in a contol system which generates neesay con control the plant or process. s Sessa cont action @ + Manipulate variable: i noua sina or con ston gener sina o contol ction generated bythe contra affect the value of controlled variable or output from the plant or process. 7 ‘ « Disturbane: isan insite signal which ends afer the ot ich ent atet he ouput sponse ofthe system ifsue disturbances gencrated ouside he sytem and acts addition to normal nets Kage external disturbance and ifthe disturbance generated within a system by itselfiskrownas itera ‘+ Control system : A control system is an arrangement of c -ment of components interconnected in such @ ‘ay 80 as to regulate, direct or command itself to obtain a certain objective. Reference Inpiat Controlled Plant / Process output Control system deals with cause and effect relationship between input and ou = ip iput and output for a plant or 1.4 CLASSIFICATION OF CONTROL SYSTEMS ‘Type 1 : Generally, control systems can be classified into three types. They are {Natural control system: A contol stem which eis inate incling b scald nr conol ten Ex: Haan bg ane nine lassen 2.Mademade control system : A contol sysem crewedty contol tem Ex Automobile 2.Combinationa contest sontrolsystm whichshevng combiion ofboth atu and man-made ar clled combinational cont sysem Ex Man dvi an atonobte ‘Type 2 Based oa the open coma systems can be died ino 1. Manually operated conta system 2. Automatic control system ‘ype3.; From te analysis point of view, contol sytem ean be cassie nto 1 Linear and Non-linear contra stems uman beings are called man-made ‘A control system which obeys the principle of superposition is known ali of superposition is known as liner control system. The Principle of superposition is a combination of an additive property and homogenous property4__—Controt Engineering (a) Additive Property: If x and *y’ belongs to the domain of the function ‘f° then we ean write ferry) =/O+/0) (6) Homogenous Property; For any." belongs to the domain ofthe function ‘f' and for any sealar constant &, we ean write F(x) = af) Fora linear system input/output relationships may be represented by:s near differential equation. ‘Accontal system which does not obey the principle of superposition is known as Non-Linear contol system. if function f(x) =», itis obvious that f(x-+)) = (+ y)* # G+ ODPand f (0x) ‘Therefore the function f (x) = x* is Non-linear 2. Time varying and Time invariant control systems ‘A control system in which one or mote parameters vary as a function of time in known as time varying control system. Thus, a time-variant system is a system described by a differential equation with variable coefficients and linear time variant system is described by linear differential equations with variable coefficients Its derivatives appear as linear combinations, but a coefficient or coefficients of terms may involve the independent variable. Example: A space shuitle leaving the earth as it mass reduces with time during the flight. Also, a rocket-burning fuel system is an example of time variant system since the rocket mass varies during the flight asthe fuel is burned. ‘A contol system in which all the parameters are constant with respect to time is known as tim variant control system. Thus, a time-invariant system is a system described by a differential ‘equation with constant coefficients and alinear time invariant system s described by linear differential equations with constant coefficients. ‘Example: Resistance, capacitance and inductance in an electrical network and a single degree of Freedom spring mass viscous damper system is an example ofa time-invariant system provided the ‘characteristics of all the three components do not vary with time. 3. Deterministic and Stochastic controf systems ‘A control system in which the response 10 input is predictable and repeatable is known as ot ale). deterministic control system. ‘A control system in which the response to input is unpredictable is known as stochastic control system. 4, Continuous - Time and Discrete Time control systems ‘Accontrol system in which all the system variables are defined for all the time is known asco time control systems. Example: Tacho-generator feedback used in the control of the DC motor. ‘A control system is which one or more system variables are defined only at certain discrete intervals of time, generally evenly spaced steps is known as Discrete-time control system. Example: Micro processor based systems. Introduction — 5 5. Lumped-parameter and tributed parameter control systems ‘A.control system that can be described by ordinary differential equation is known as lumped parameter control system, ‘Accontrod system that can be described by partial differential equations are known as distributed- parameter control systems. 6.Single input - Single Output [SISO] and Mu systems le Input Multiple-Output [MIMO} control A control system in which there is one input and one qutput is called single input and single output control system, ‘Acontrol system in which there are multiple input and multiple output is known as multiple input and multiple output control systems, ‘Type 4 : Based on the presence of feedback 1, Open Loop control system - A control system in which control action is independent of the Aesired output is known as open loop control system. (Feedback is absent) 2. Closed Loop-control system - A control system in which control action in dependent on the. desired output is known as closed-loop control system. (Feedback is present) 1.5 OPEN LOOP CONTROL SYSTEM -Anopen loop control system is one in which control action is independent of the desired output. It K ie ‘means the desired output is neither measured nor compared with the input, ‘The block diagram ofthe open loop control system is as shown in figure f.1 >] Conor —o| Plant or Process > Ouiput Input Fig. LL : Open Loop control system In this system, when input is applied to the controller which generates necessary control action required to control the plant or process which is to be controlled to generate desired output. The accuracy ofthese system depends entirely on the accuracy with which the input-output relationship “is designed. Ifthere are any variations in the external environment or disturbance during operation, | tron = Ctapat) Signal | fromthe [= Feedback clement : thermostat ~ Thermostat —_____] Fig. 13 : Auiomaile Elecirie Tron8 Control Engineering Figure 1.3 shows block diagram of an automatic electric Iron with temperature control. It works ‘on the principle of feedback. In an automatic electric Iron, thermostat acts as a feedback element. ‘Thermostat senses the actual temperature of iron, if temperature is beyond the particular value (desired temperature), Relay switches off the supply to iron and maintains constant temperature at the output, Thus, it isa closed loop control system. 2, Voltage Stabilizer Up ‘Variable x (f Input Yoonall Ay 5 To Ipad Voltage Down o—! Ftrereise Fixed Output Voltage N Fig. 1.4: Voliage Stabilizer Figure 1.4 shows voltage stabilizer can be treated as closed loop control system Input othe voltage stabilizer isthe variable voltage and outputis the constant voltage. Working principle of voltage stabilizer i based on the controlling the numberof secondary turns as per requirement to increase or decrease the output voltage. When input decreases, the output switch will be connected above A. On the other hand, if input increases, the output switch will be connected below A. 3. Ship Stabilization System using fin Fn { | a ‘Sea water level Sea waters, Fig, .5(a): Ship stabilization system Controller —} f= Plant —4 >| Fin Actuator >| ship Desired Roll Caproes Rol —={__ Feedback element Roll Sensor Fig, 1S(b): Ship sabilization system using fin Introduction 8 Figure 1.5(@) shows the block diagram ofa ship stabilization system using fin. In this system roll sensor acs feedback element Fin actuator acts acontolle and ship acts a plant which is 10 be controlled. The desired rll position and controlled rol (output) are compared atthe suming point to generated error or deviation (i any) based on which fin actuator initiates the necessary control action to stabilize the ship, 4. Room Heating System Controllee——~} —-|__ Plan j= Room Desired 4 t - Temperature \ectpoa Relay > switen >] Furnace — ‘Thermostat Fig. 1.6 : Room heating system ‘The block diagram of a soom heating system is as shown in figure 1.6. In this system, thermostat acts as a feedback element, its function isto sense the actual temperature and compares with the desired temperature. Based on the deviation (error) obtained controlling elements such as relay ‘and switch are activated to produce necessary signal to the plant-(furnace) so as to obtain the desired output temperature of the room, 5. Automobile speed control system — Plane Leg Desired g Eyes and - Acta .| -o|musces &| +1 ag speed Brain |Accelerator Speed Speedometer Fig. 1.7 : Automubile speed control system Figure .7 shows a block diagram of an automobile speed contol system. inthis system, speedometer measures the actual speed of the engine and compares with the desired speed to generates the error (iF any). Based on the error controlling elements such as eyes and brain takes @ decision and leg muscle and accelerator is actuated to increase or decrease the speed of the engin. 1.7 CONCEPT OF FEEDBACK CONTROL SYSTEM Feedback control system are the control system in which effect of disturbance is seen as an error after comparing the output and reference input before controller takes the proper corrective action. ‘Thus, in feedback control system controller initiates the proper corrective action for the difference10 _—Contol Engineering obtained between the output of the system (contvolled variable) and the reference input due to the presence of disturbance, Thus error in the control system output is recticed due to feedback. But feedback not only reduces the error but also reduces the sensitivity of the system to variation in parameter and unwanted disturbances (Internal and external). For example, consider a room heating system as shown in figure 1.8 in which temperature of room is controlled atthe desired level. This system consists of heating system as a plant which operated by valve as a controller, and a thermal sensor as a feedback element. In this system, actual temperature of the room (output from the heating system) is sensed by the thermal sensor and is compared with reference input (desired value) to generate ertor. For the error. valve mechanism is actuated to take the proper corrective action so that desired temperature level is maintained, ‘error signal . Actual Pass - Tae »{ Heating system iperare np (controlled reference“ Freedback variable) input) signal za Thermal sensor] Fig. L8 : Room heating system 1.7.1. EFFECT oF FEEDBACK ON THE ConrRoL SvsTeM 1. Feedback in control system improves the time response. 2. By proper design and application of feedback, stability ofthe system can be effectively controlled. 3. Gain of the system can be controlled by controlling feedback. 4, Feedback in control system reduces the effect of disturbance (Internal and External) on the system and reduces the sensitivity ofthe system to variation in parameter. 1.8 REQUIREMENT OF AN IDEAL CONTROL SYSTEM Various parameters are considered in the design of a control system. All the control systems are designed to perform specific objectives. To achieve the required objective, a control systems must satisfy the following requirements 1. Stability 2.Sensitivity 3. Speed 4, Accuracy 5. Disturbance or Noise 6.Bandwidth If a control system satisfy the all requirements the system is known as ideal control system, Stability : Stability in a control system implies that small changes in the system input, in initial Introduc conditions or in system parameters do not result in large changes in the system behavious. Staility is the important characteristic of the transient response of a control system. A ideal controt system is one which gives bounded output for bounded input. A ideal control system are designed to be stable. sitivity : An ideal control system should be insensitive to the variations in parameters ofthe system but it should be sensitive to the input commands. Its a important parameter that shoul’ be considered in the design of control system, Speed : Speed of the control system means how fast the output of the system approaches te the 4esired value. This is measured interms ofthe settling time and rise time. An ideal control system should have good speed, Accuracy : Accuracy of the control system means how much the output of the control system is nearer to the input or desired value. An ideal control system must be highly accurate Disturbance : All control systems are subject to some type of extraneous signals or noise during operation. External disturbance such as wind gust, thermal noise voltage are quite common, ‘Therefore, in the design of a control system, considerations should be given so thatthe system is insensitive to noise and disturbances but sensitive to input commands. Band width : Band width of the control system means for the range of input, the output of the control system should be constant. It refers fo the frequency response of the control system. An ideal control system must give satisfactory output forthe input frequency range. 1.9 CONTROLLERS lis generally known as an automatic controllers. The controller isa basic element in a control system which compares the actual value of the plant output with the reference input ot desired ‘value to determine the error or deviation and produces proper corrective action (control signal) that will reduce the error to a smaller value or to zer0. The measurement of error is possible due to feedback. The feedback allows to compare the plant output with its reference input to generate error. Thus input to the controller is the deviation of the output from its desired value known as error and output from the controller is the corrective action known as manipulated signal, The ‘manner is which the controller produces output ie., manipulated control signal is known ascontrol action, 14 A Types oF Contno.tens Based on the control action, controllers are classified as 1. On-off or two position controllers 2. Proportional controllers 3. Derivative controller 4. Integral controllers 5. Proportional plus integral controllers 6. Proportional plus differentia controllers 7. Proportional plus integral plus differential controllers.32_—Control Engineering 1.9.2 Propormionat - Conrnowuen Figure 1.9 shows a simple block diagram of the proportional controller. In this, the output of the ‘controller i.e., manipulated of actuating signal is proportional to the input of the controller ie., ‘error signal, error detector Reference 4 Spyenersisn ; > ky > — Manipulated signal or input” r() o : 0 Nar caneetir aera ~ Fino Feedback signal Fig. 19 : Proportional controller ip between output ofthe controller Fora controller with proportional control action the relations tn(e) and error signal e() is im(t) = K, e) ‘Taking Laplace transform on both sides, we get M(s) = k, Bs) M(s) E(s) ‘where k, is the proportional gain. ‘The relation between output ofthe controller m(?) and the error signal e For the unit step int is as shown in figure 110, m(p) e(t) Fig. 110 ‘ Iishows that there exists linea relation between controller output m() and the error signal e(). Fora zero error the controller outpat sould be zero otherwise the process will come to hall. Hence ‘mathematically it can be expressed as (i) = Kyeto-+m, ‘Where m,is the controller output for zero éror Introduction — 13 Cunacrensncs A control system with proportional contro! mode has the following characteristics: I. The controlier output is equal to m, when the error is zero. 2. The system is stable. 3, Improves Closed Loop Response. 4, Provides fast response. 5. Improves steady state error and Rise time. 6. But, high gain may lead to instability due o high oscillation. Disapvanraces 1. Provides heavily damped response. 2, Provides large steady state error. ‘Example: Non-inverting operational Amplifiers 1.9.3 Derwanive Conrrouien Inthis contro! mode, the output ofthe controller., manipulated signal mit) is directly proportional to the time derivative of the input to the controler ic, error signal e(). Mathematically it can be expressed as set) mit) ao a et) my = x, Taking Laploe Transform on bot sides, we get MG) = SE) _ Ms) ‘S E(s) Where K, is the derivative gain constant. The main advantage of this control mode is that it responds tothe rate of change of error and can produce necessary corrective action before the magnitude of the error becomes too large, Thus derivative controller anticipates the error and also initiates the early corrective action for the anticipated error. For example ifthe error changes at arate of 2% per minute, and the derivative time T, =3 minutes, the predicted error is 6%. Ifthe Controller Gain, K, = 0.2, then the derivative control mode will add an additional 0.2 x 6% = 1.2% to the controller output. ‘The derivative control mode gives a controller additional control action when the error changes consistently. Italso makes the loop more sable (up toa point) which allows using a higher controller gain and a faster integral (shorter integral time or higher integral gain). These have the effect of reducing the maximum deviation of process variable from set point if the process receives and external disturbance,14 _—Contol Engineering ‘e control action can never be used alone asit takes corrective action on the rate However Derivativ igerror itself. Any noise inthe error signal amplifiesand of change of actual error rather than actuatin gives an erratic behavior. Cranacrenisrics 1. Produces significant correction befo 2. Tends to increase stability of the system. 3. Adds damping to the system hence large values of used, ye magnitude of actuating error becomes too large. gain which will improve accuracy can be 1.9.4 Propormionat. PLus Derivative ConTROLLER “This isa combination of proportional and derivative controller which is used to improve the steady state behavior ofthe system. In this control mode manipulated signal consists of proportional error Signal added with the derivative error signal. Mathematically itcan be expressed as delt) (d= K,e(0+K, det’) m0) = Kye4K, TG “Taking Laplace Transform on both sides, we get MG) = K,B)+K,T,SEO) The transfer function is given by M(s) © Fy 7 StS Figure 1.11 shows the block diagram of the transfer funetion of the PD controller co) ro) K+K,T,S -—> mo bo) Fig, L11 : PD controller Introduction — 15, Figure 1.12 shows response of the PD controller for the unit step input, Q i | Fig. L12 1.9.5 Inrecnat - Conrnouter ett) {eure 1.13 shows a simple block diagram ofthe integral contol: nth, ouput ofthe controler ies maniolated signal i changed ate proportional tothe input ofthe controler, ear signal. ‘error detector error signal O (Feedback signal Kk, > m(o Maniputated signal Reference Input ri Fig. 1.13 Integral controler Fora controller with integral control action the relationship between output ofthe controler mi and erro signal e() is " toler and?) ar : ke By integrating, we get m= ki fele)ae ‘Taking Laplace transform on both sides, we get ms) = (3)46 _—Contvol Engineering where fis the integral sensitivity = ‘The integral controller gives a output which is ramp the integral control action is also called reset control Integral control action is used whenever the steady state error is too large. Cranscremsncs 1, Slows down system Response. 2, Increases setting and Rise time. Ex: Integrator using operational Amplifier followed by sign inverter. 1 ‘Thisis a combination of proportional and integral controller which is used to improve the performance of the system. In this, the manipulated signal consists proportional error signal added with an integral of the eror signal Itis given by .6 ProporrioNaL Pius InreGrAt Contnotten md = ko) + K fle mt) koe felon Taking Laplace transform on both sides, we get M(s) ky 4, Bi) + FEES) m9) 1) ay * HT) hire é, isthe proportional gain 1s the integral time. Fig. 1.14 shows Block diagram of the transfer function givenby ® }—> mt) Fig. LI: Pl controller Introduction — 17 ‘The integral time adjusts the integral control action, while change in proportional gain affects both proportional and integral action. The inverse ofthe integral time is called reset rate. The reset rate is the number of times per minute that a proportional part of the control action is duplicated, 1.9.7 Prorormonal pLus INTEGRAL PLUS DIFFERENTIAL CONTROLLER Its the combination of proportional, integral and differential control actions so as to derive the advantages of all the control action. General, itis known as PID controllers, The equation forthe PID controller is given by elt) at im be Keo felar+k,t ‘Taking Laplace transform on both sides, Ms FQ), 1 vets nt] where, k, isthe proportional gain Tis the integral time is the derivative or differential time Fig. 1.15 shows the block diagram of the PID controler forthe transfer function, 1) — 2 bleh, fleets pmo, ~ J(e) Fig. L1S : PID controller ‘These controllers are used extensively in industrial processes. Setting of PID gain is called as tuning however tuning should be done carefully as there are three gains. 1.9.8 ON - OFF Conrnouer ‘On OFF Controller are required to switch ON or OFF the component at appropriate time. Switches relays times unitate used forthe purposes. The main disadvantage with these ON/OFF Controller are they consume significant voltage and current during their operation hard wised electronic logic controller are also used they are cheap however they are complicated. Fluid logic systems are also available to install and maintain,18_—Control Engineering 41.9.9 Avomons Semvonecnnens «clos an acceleration every common in Fedak contol stems used contol poston velocity, sin Tait an mary applications. They afro ss eromehantsms. Aseromechan Foner tpg foedbnck contol a which he onl varie sa mechanical poston mein deat of posion sucha veloiy Or accteraion An automat aera faning Gjaes amexarps of servomectanism The strat fllows x ramp tthe died ouchdown ‘Zin noir examples thconaolystem of an instal obatin which hobo arms oced Tofliow some desired pa space Recrons — i nin which he reference input of A regulator ¢reulting syst i» feedback contol se Conan constant frag periods ine general rte nt ine era rig whe he Stem s operational. Suchen pts known asset point, The objsive af the idlesped conto SSuem is Enown asa repelatr system, Anoterexapie of fegultor contol system is the Raman balogia sytem that mains the Body temperature st approximately 286° in an cavrontent thar ual ba fleet epee i | | Inocuction—_ 19 REVIEW QUESTIONS 1. Explain with an example the following (0 Open loop contot system (i) Closed foop contro sytem 2. (a) What ar the requirements ofan ideal contol system ? (b) Differentiate between opentoop and closed loop control systems. (©) What are the characteristics ofan integral controller? (@) Describe controler which will improve the transient behaviour of the system. ‘May / June 2010, 3. Explain open loop and closed loop control system with block diagrams. What ae the advantages May / June 2010 and disadvantages of a closed loop system over an open loop system ? Dec.09/ Jan.t0 4, What are the requirements of a control system ? Briefly explain. Dec.09/ Jan.t0 ‘5. Draw the block diagram of proportional integral controller and explain. -‘Dee.09/ Jan.10 6, What are the requirements of an ideal control system ? Dec.08 / Jan.09 7. Give the comparison of open loop control system with closed loop control system ? Dec-08 / Jan.09 8, Explain open loop control system and closed loop control system with an example for each, Tune / July 08 9. What is a controller ? Explain its function in a system, June / July (8 10. Explain and state the characteristic of the following controllers ( Proportional Gi) Integral ii) Proportional + Integral Sune / July 08 11, What are the requirements of an ideal control system ? Dec. 07 / Jan. 08 12, With a neat sketch, explain the working of an tank-level control system. Dec. 07/ Jan. 08 13. Write a note on proportional controller Dec. 07 / Jan. 08 14, Explain integeal controller show integral or reduce the steady state in system. Dec. 07 / Jan. 08 15, Explain the concept of feedback control system and the requirements ofan ideal control system? July 2007 16. Explain the working principle of an automatié electric iron with temperature control using block diagram. List the advantages ofthis with conventional non automatic electric iron July 2007 17. Explain proportional + integral + differential controller and their effect on stability July 2007 18, What are the requirements of an ideal control system ? Dee. 06/ Jan. 0720_—Control Engineering 19. Differentiate between open loop system and closed loop system ? Dec. 06 / Jan. 07 20, Describe an integral controller with suitable example. What are the characteristies of integral controller. Dee. 06 / Jan. 07 21, Prove that in an integral controller the steady state errOr is 2er0, Dec. 06 / Jan. 07 22, Deseribe & controller which will improve the transient behaviour of the system. Dec. 06 / Jan. 07 23, List any three major advantages ane twa disadvantages of open loop control syster. uly 2006 24. Ina liquid level control system, controller maintains liquid level in tank by comparing the actual level with the desired level and correcting any error by adjusting a pneumatic valve. The float gives the feedback of liquid level in the tank. Draw the block diagram of the Tiquid level system, July 2006 25. Give the block diagram, relationship between output of a controller u(t) and actuating error signal e(0) and the transfer function of, Proportional control action (i) Internal control aetion (ii Proportional plus Tntegal plus derivative contro action “uky 2006 26, What are the requirements ofan ideal control system ? Can they met with (3) open Toop system (ii) feed back system ? Give a comparison between the two Jan/Feb 2006 21. Distinguish between open loop system and closed loop system with examples. July / August 2006 28, Explain the requirements of a control system, ‘Jan / Feb 2006 28. Explain the terms (System, (i Disturbances (iil) Transfer function (Gv) Open loop control system (W) Block diagram fe SES ee 244 INTRODUCTION Ofter when engineers analyse a system t0 be controled or optimised, they use a mathematical model In analysis, engineers can build adesriptive model ofthe system as a hypothesis of how Ye system could work or try t0 estimate how an enforceable event could affect the system Similarly, incontrol of asystem, engineers can ry ou different contol approaches in simulations ‘A mathematical model uses matherntical language to describe «system, A mathematical mode usually describe a system by a set of variables and a set of equations that establish relationship between the variables, The variables represent some properties of the system. Generally, most of such mathematica equations ae differential equations. These differential equations describes the dynamics of system, whether the system may be mechanical, electrical, thermal, hydraulic an 0 on. Such differential equations are obtained by using physical las governing a pariclac sycem Forexample, Newtons laws forthe mechanical systems and Kichoft's las for electical system Definition : The set of mathematical equations that represents the dynamic characteristics of az system accurately or at last forty well i called mathematical model of the system ‘Thetwo most commonly used methods of modeling are transfer function mode! an state-varisble model. The transfer function mode! is valid for linear time invariant system, whereas the stte- ‘variable models ae vali for both linegr and non-linear systems 2.2 TRANSFER FUNCTION MODEL In control engineering, functions called transfer functions are used to characterise the input-output relationship ofthe systems that can be described by linear differential equations. Thus, the transfer function expressed in terms of differential equation which describe the relationship between the ‘components of the system is known 2s transfer function model. Definition : The transfer function of linear, time invariant system is defined as the ratio of the Laplace transform of the outpst 10 the Laplace transform of the input when all initial conditions22_—Contol Engineering itis given by ls) RG) where Ga) isthe Laplace transform of impulse response of the system Cs) is the Laplace transform of the output RG) is the Laplace transform of the input ‘The block diagram representation of the transfer function model in as shown in figure 2.1. ‘Transfer func . Gis) = RG) oo) 2 ae) Fe Input Output Fig. 2 ‘The transfer function model of a system can be obtained as follows 1. Write the mathematical of the system based on physical laws. This results ina set of differential ‘equation that relates input and outpat 2. Apply the Laplace transform tothe differential equation. Assuming initial conditions are zero find the ratio of Laplace transform of output to the Laplace transform of input. 2.3 MODELS OF MECHANICAL SYSTEMS Mechanical systems ean generally classified into two types 1. Translational system 2. Rotational system ‘The difference between the two is, in translational system motion of body is along a straight line ‘where as in rotational system motion of body is about its own axis To find mathematical model ofthe system, the fundamental law governing mechanical systems is Newton’s second law of motion. 2.3.1 Mechanical Trans.anional SYSTEM ‘Translational system are related with the translation motion and basically consists of mass, spring and dash-pot. Two of them mass and spring are energy storage elements whereas dashpot is an energy absorbing element. For analysis of the translation system consider. ‘Mass : Mass represents an element having inertia. This is considered as a property of an element that stores the kinetic energy of translation motion. When a force f() is applied toa body of mass, 1M, displacement x() takes place in direction of the applied force as shown in figure 2,2. Then a reaction force f,() is generated in the body, in the direction opposite to that of applied force /() ‘The rexetion fotce is proportional to the acceleration. ic, WO L > _2(0) Displacement pO 0a 40 = Ma {24 x Friconless support Fig. 2.2 Mathematical Models — 23 According, Newton's second law of motion 0 = 40 a(t) y= a) JO = MT ‘Taking Laplace transform, assuming intial conditions are zero, we have Fis) = Ms X) Dash-pot Dash-pot is a mechanical element used to provide damping in the system, It consists of piston and. cylinder arrangement with fuid medium as shown in figure 2:3, \when a force f() is applied on piston, dashpot produces a reactive or opposing force f(t) which is proportional to the velocity. _ Piston a fot 7 <\— Fluid medium a H— Cylinder LO = BE Fig. 2.3:A dash pot ‘According Newton's second law of motion, AO =f) a ae Sin = BE ‘Taking Laplace Transform and assuming initial conditions are zero, we have, Fs) = Bs X@) Spring Spring is « mechanical element which stores the potential energy. It stores energy duri 1 potential energy. It stores energy during the variation af shape due to elastic deformation resulting ftom the application of force. Consider a spring with stiffness k fixed at one end and force is applied atthe free end as shown in figure 2.4. The displacement x() is produced due to applied force. Due to this resto ju (0 applied force. Due to this restoring force f(0) is produced which depends upon the spring stiffness k and displacement x(). ie, I = ko According Newton's second law of mation, fo 10 =£0 «(0 £0 =k x0) ed Taking Laplace Transform and assuming initial conditions ase zero, we have, FG) = kX)24 —Contvo! Engineering Mathematical models of mechanical elements Mathematica Models ~_ 25, Rotational systems are related withthe torque and angular motion. Rotational systems are similar to-translational system except that torque equations are written instead of force equations displacement, velocity and acceleration interms of angular quantity. ‘The fundamental elements of the rotational systems are mass moment of inertia (J), rotational friction coefficient of the dashpot (B) and torsional spring stffuess (k ork. ‘Mathematical models of rotational systems ae Symbol time domain frequency domain 70 woe | BE | toe | ro -st0@ s a Dash pot T= oa Tis) = Bs 6 (5) 112, ; 9B now £(010)~8,0) Equation in i A | re) = B-5-(04(0)-0,(0) fre) = as [0Ys) - a, Element Sambal rein | | vo Tt z aaa quency domain | ae ' 2 Fox 25(0 ' Mass fO= uf F(s) = Ms°X (8) Spring k TK) = k(t) Tis) = k 8 (5) Ifo F Be : be Fi = BX) TH) = k (8) — 0,0] | THs) = & (0,5) ~ 0,5) | £0 Too oy) Dash pot q beso bx Fs) = Bs 1X6) ~ X)] ‘Translation motion Rotational motion —E- 1. Mass (M) Tenia, —| | [2:Damping coetticient/frieion, Damping coefficient / faction, (B) = be FO=EXE) ftin > £0 Lee a) y 5. Spring stiffness, (&) ‘Spring stiffness, ( Spring be f Foss, (F) Torque, (7) A i ysplacement, x “Angular displacement, (0) [po FO = Ha (0 ~ x10) | FE) = HXLO- XO = - eo locity, v= 3 = & iy,o=® 6. Velocity, 2 Angular velocity, o= 2.3.2 Mechanical. Rorarionat SysTEMs. ax ~ 17. Acceleration, a= @ = a “Angular acceleration, 2.4 MODELS OF ELECTRICAL SYSTEM Electrical systems can generaly classified into two types. 1. R-L-C series cizcuit 2. RL-C parallel circuit, ‘To find mathematical model of the system, the fundamental laws governing electrical circuits are Kirchoff's Current Law (KCL) and Kirchoff's Voltage Law (KVL). ‘The basic clements of an electrical circuits are : Resistor, Inductor and Capacitor. Resistor : Resistor is an energy dissipative element. The relationship between the current and voltage for the circuit shown in figure 2.5 is given by= Control Engineering vu) = RAD) wR ‘Q where R = Resistance ba i) = The current through the resistor 25: Resistor wo Inductor : An inductor is an electrical element which stores energy in magnetic field. The relationship between the current and voltage forthe circuit shown in figure 2.6 is given by ai ‘The voltage across the resistor fa L wo oH ——$9 -—— 10 4 Fig, 2.6: Inductor Capacitor : A capacitor is an electrical element which stores energy in electrical form. The relationship between current and voltage for the circuit shown in figure 2.7 is given by 1 O ¢ wo go Jiat a(t) “dt Fig. 2.7 : Capacitor or 2.5 DC AND AC MOTORS IN CONTROL ‘A motors an electromechanical system in which output is some mechanical variable like postion, velocity or acceleration for a voltage input. Direct current (D.C) motors are most widely used is control system to control various variables suchas speed, displacement, acceleration etc, while alternate current (A.C) motor are more difficult tocontrol especially for position control and due to non linear output behaviour. 2.5.1 Dinect Current Moron D.C. motors are generally used in open loop and closed loop control system to control various variables like speed, torque, displacement etc, Motor with series, compound or shunt connections are used in control systems depending upon the nature of operation. Principle D.C. motor is basically a torque transducer that converts electrical energy into mechanical energy. ‘The torque developed on the motor shaft is directly proportional to the field flux and armature D.C motors are used in two different control modes. They are |. Armature control mode 2. Field control mode Br Fig. 28 : Armature controlled DC motor Figure 2.8 shows a schematic diagram of armature - controlled D.C, motor. Let the speed of motor is controlled by armature supply voltage e, and field voltage e, is Kept constant. Here, input tothe system is armature supply voltage e, and output of the system is angular displacement 0, ‘The transfer function of the armature controlled D.C. motor can be found by dividing system into three parts as the Armature circuit, the D.C. motor and the Mechanical system, Hence, the differential equations are For Armature circuit Rio bat 6 eo at whee 6, Taking Laplace Transform on both sides, we bave EL) = RL fs) +L, 1 (3) + b, 8.(9) E(9) = (9) [R,+L, 3+ k, 58) -o Where ¢, = back emf B= back emf constant For D.C, motor, TO) = ki, -@ where k,, =k, fyi, = motor torque constant k,_ = proportionality constant ky i, = $= airgap flux of the motor k= field constant of the motor For Mechanical system (Rotational) w+ Bd = TO From equation (2), we have J+ 36 =i,Contr Engineering a 2 Matematand8s —_20 ‘Taking Laplace transform i E,(s) = R,Ifs) + L,s (3) Js¥0(8) + Bs O(3) = k, 18) i EAs) = B+, 19) 0 =k, L(3) + (Ist + Bs) s) = 0 {For D.C. motor 2 Te) = ki, 1s? + Bs)p(3) vy @ a © % 16+ Bo = TH) Substituting value /(s) in equation (1), we get From equation (2), we have (4s? + Bs)0(s) JO+ BO = ki, Bs) = (Ras) 06) Taking Laplace tasformon both sides, we have i, Es) = Us? + Bs) (R, + L, 8) 6s) + kf, (5) Js°0(s) + Bs (3) = ky 1,68) ‘Thus, transfer function of the system is, . nk, (5) + Us? + Bs) (6) = 0 —@) 3) E,@) ky (Re + Ls)(ds? + Bs) + kgs ‘Transfer function of field - controlled D.C. motor T.. 2 8 Fig, 29 Fed controled DC motor Figure 29 shows a schematic diagram of field-contolled D.C. motor. In this ystems, the amature cuuten (or voltage) is meinained aa constant value and motor speed is controlled by varying the field eurent (voltage). The transfer function ofthe ield controlled D.C. motor canbe found by dividing system ino three pars as the Fel circuit, D.C. motor and the Mechanical sytem. Hence ditferetial equations are For Field eircit, a ef = Ril Ly Gr ‘Taking Laplace transform on both sides, we have (4 + B5)(s) % 19) ‘Substituting value /,(s) in equation (1), we get [R, +L,s\(Js? + B(s))®(s) Bes (% ze +8(9)9(6) Thus, transfer function of the systems 64s) Fo ‘Thus, the ioput to the system is fleld supply voltage Z,(s) and output of the system is angular displacement 63). kp 25.2 Avrenunne Cunnent Motors A.C, motors re used in control system where low power is tobe controlled. Since A.C. motors are light in weight, more rugged and have no brush contact, they can deliver low power. ‘The simplest form of the A.C. motors used in the control system is two phase induction motor. Figure 2.10 shows the two phase induction motor with servo amplifier ‘Two phase induction motor consists of a rotor and a stator with two distriboted windings which are displaced electrically 90° from each other. These are known as reference winding and control windings. This 90° phase shift between the reference and control winding voltage can also be obiained by connecting a capacitor in series with the reference winding. The main field is supplied continuously to produce high torque, while the contol windings are supplied with controlling voltage ‘which is generally applied froma servo amplifier. By changing the control voltage the motor output20 voltage shaft rotation isin direction can be controlled. When the control voltage leads tothe fixed voltage while it lags the direction of rotation is reversed. The diameter of rotor is kept small to have fess, inertia and better acceleration characteristics. Control Engineering A.C. Supply wPReference voltage ' J<—Main or Reference [Resonate] | capacior | winding pina Servo Amplifier Fig. 2.10: Two phase induction motor ; ‘The transfer function of induction motor is obtained with following assumption i. The torque - speed characteristic of the motor are parallel and equally spaced for rated voltages. ‘Torque at ero speed slope m = "No toad speed , ji, The torque-control voltage characteristic for motor ae linear. ‘Torque at zero speed at rated voltage Rated voltaged E, for the fixed winding iii, The inductance L., and resistance R, of control winding in the motor are negligible. ‘Consider the simplified diagram of induction motor as shown in figure. ‘The torque equation of the motor is Balt) at where K,, is the motor torque constant and 8,(i) is the angular displacement of shaft Td) = Ke()+ m Mathematical Mogole— 91 ‘The total torque produced by the motor is utilized in rotating the motor with load and to over come motor friction Hence Q where J, and B, are moment of inertia of load with motor shaft and coefficient of viscous friction Eating etn 1) and 2 west keeliyam fall) «4,0, 5 al) dT gee ‘Taking Laplace transform on both sides, we get K, E,(3) + Ms 0,(3) = ‘The motor transfer function is S570 os) + By 8p 8) 8n(9) Ky E.(s) ~ av Substituting the value of M and ‘T- in the equation (a) we get hs ! wheres = & Hydraulic capacitance “The energy inthe liquid isthe hydraulic capacitance. To derive the expression for the hydraulic capacitance, considera figure 2.11 as shown 1 cuca mamaria ener yon em Mathematical Models — 99 Rate of accumulation = Inflow - Outflow és AG TI-% but “p= pk eo te 1 é "dt Pea A(#) - A(®) is analogous to spring stiffness (¥) Flux $6) is analogus to displacement X(s), Analogous Quantities in Force (Torque) ~ Current Analogy ‘Mechanical System Electrical System ‘Translational System Rotational System P(s) = Mstx(s) + Bsx(s) + kx(s) Fs) = X(s) (Ms? + Bs + Kk] =) 11, Force (F) ‘Torque (1) Current (C) To obtain the F-1 analogy, consier the electrical system as shown in igure 2.14, 2. Mass (t) mero Capacienos (C) Applying, Kirchoff's current law, we have : i = i +504 160 Torsional fiction ; Expressing curent interms of node voltage, 5. Friction Coefficient (8) Resistance of resistance () a : C Co Coefficient (B) R vid we get, ig 2 2+ 8 = RL 4, suttnes (&) Torsional Stiffness (k) | Reciprocal of Inductance (3) Taking Laplace transform sides, we get L e 5, Displacement =) “Angular displacement (0) | Magnetic Aux @) = ae Angular velocity To express the equation (2) similar to equation (1), consider “ ~ sey: vas (et) = > where 6 = flux J ) d w=F (0-8-0 tuking Laplace transform on both side, we get poem ee 5 vl 2.10.2 Force - Vottace Anavoay - F-V ANaLoay VG) = 5469) 0F 96) = = In this analogy voltage in the electrical system is considered as the analogus to force in mechanical system. To obtain F-V analogy, consider the electial system as shown in figure 2.15, Thus, equation (2), becomes Me) 1 t Applying Kirchoff's Voltage Law, we have 1s) = 2 + Tae cd i ai 1 ee v0 = oR LEP 4A file ‘Rearranging, we get ‘Taking Laplace transform both sides,we have 1 = sce stss -@) ny = afer +542 _— Control Engineering L 163) R = is ye ya Ve) = HOR¥LID+ TE De dq a consider, )= GE where q = chang 4 ‘Taking Laplace transform, x0 ein 16) = 50) Thus, equation (4) becomes Os) Vis) = s069R + L#O) + =F Fig. 215 Rearranging, we have ol is? + Rs+ 4] -©) Mo) = oof # +R+ 2] ‘Comparing equation (1) and (5), we get 1, Voltage, V(s) is analogous to force, Fis) 2 ndutanee, Lis analogous to mas, 3. Resistance, Ris analogous to tional coeficient or damping oeticient, B. 4. Reciprocal of capacitance, 4 is analogous to spring stiffness, 5. Charge, 0() is analogous io displacement, X) Analogous Quantities in Force (Torque) - Voltage Analogy Late Electrical System ‘Translational System Rotational System Toe Tove @ Torage 2. Mass (M) Inertia (J) Inductance (L) cr a rriction Coettcient(B) | coemetnt sistance 7 a Suites ‘Torsional Siti () | Reciprocal of Capacitance 5. Displacement (2) ‘Angular displacement (0) Charge (@) 6. voy ( Angular velocity (i-22-0) 3 Mathematical Models — 49 General Steps Involved 1. For the applied force, determine the numberof displacement variables required 2. Identify the elements which are under the influence of different displacement. 3. Using Nodsl analysis, represent each displacement by a separate node Note : Total numbers of node is equal to total number of displacements 4, Show all the elements in parallel under the respective node which are under the influence of respective displacements. 5. Show all the elements causing same displacement in parallel in between the respective node. 6.Complete he equivalent mechanical system and write the equation of equilibrium using Newton's law of motion. 7. For F-V analogy (loop Analysis}, simulate equation of equilibrium using following analogous quantities. Foavinestipagst biases ti 8, ForF-Ianalogy (Node Analysis), simulate equation of equilibrium using following analogous quantities. 1 ko Tix ei id avore the equivalent electrical analogous network of the equivalent mechanical system,44 —Controt Engineering WORKED EXAMPLES Example : 1 Obtain the differential equation of equilibrium for the mechanical system shown in figure 2.16 and draw the equivalent mechanical system, atso draw the analogous electrical network based on (@ Force - Voltage analogy ; ) Force - Current analogy Ullé 4, i, Yao Tro Fie 216 yao Solution = Due to applied force,f(@) mass, m, willbe displaced by x(*) and due toB, and , the mass, will be displaced by x,(. Hence equivalent mechanical system is as shown figure 2.16(a) k Fig. 216(a) Hence differential equations are Far Node mi, + (x, -%)* BR) = LO. raga 1 FV Analogy | PV; MOL BOR b> _ Mathematical Models — _4s ais EG Sal Bib) ane Fel Analogy : F+1;M—+C;B—> are ; OBE Slums For Node x, : mia + Bois thn + hima) + (iy -) = 0 BV Analogy: F3 ViMoL BR: bse gyi si girta diy LT Rint fica ‘Ge flare Ri = drags: P31: 86:85 birt ia vosd ae So ote dst ual fey m7 wee El -1) <0 Analogous Electrical network are as shown in figure 2.16(b) and 2.16¢¢) F-V analogy Here element in parafef under node are connected in seties L, eC R, wa) Fie, 2.1606) FA Analogy ‘ Fig. 216)48 —Control Engineering Example + 2 For the mechanical system shown in figure 2.17 @) Draw the mechanical network i) Write the differential equation of the system Fig.27 Solution : _ Due to applied force f(t), mass m, willbe displaced by x, Now due to B,, the mass m, is displaced by x, and due to B,, the mass ms displaced by x,. where as £, is under displacement x, and x, Bis under displacement x, and x, fis under displacement x, and x, 3B, is under diplacement x, nd x, 4s under displacement x, B, is under displacement x, 1 Heenee equivalent mechanical system is as shown in figure 2.17(a) Fig. 2.17(a) Mathematical Models ~ 47 ‘The differential equation of equilibrium are For Node x,: md. + BCs, —%)+h(—n) = £0 For Node x,! mk + Bibi i) +ha(xq—%4) = 0 For Node x,: Bi(ij- 2) 4h (3%) = 0 For Node x, : 1h + By + byXy + By( ig ~ i) +b, (4, Example + 3 For the mechanical system shown in figure 2.18 @ draw mechanical equivalent network Gi) obtain the equation of equilibria (il) draw analogous electrical network based on FV analogy 4 3 k, 4,80 0 ky FO 0) Fig. 2.18 Solution : Due to force is applied atthe spring, it will absorb some energ theta kee ee aie sng y and is under displacement, Now by x, and due tok, the mass m, is displaced by x,48 _—Contiol Engineering Hence equivalent mechanical network is as shown in figure 2.18(8) FO! qi k 5, Fig. 2.18(a) Hence equation of equilibrium are For Node kh @-x) =f 1 P-V Analogy : F>ViM>L;BAR:k> Git ait Fi elena =) For Node x, ims ht — x) thal 9) + Bila = a pot FAV Analogy : PF VjM-» L:B>R3k> Git 943 idee a flee) Reh) =O For Node x, gy + hy + Bis + B35 — 2) #0 — screenees anise Analogous electrical network : vo einem Fig. 2.18( "Analogy Example : 4 For the mechanical system shown in figure 2.19 @ Draw the equivalent mechanical network i) Obtain the differential equation of equilibrium i) Draw the analogous electrical network based on F-V analogy. Cl, ___ Solution : Fig 2.19 Due to force /() is applied atthe spring, it will absorb some energy and is under displacement x, ‘where as m, is under displacement x, and my is under displacement x, and fis between x, and x, 5, is between x, and x, is under displacement x, Bis under displacement x, Hence, equivalent mechanical network is as shown in figure 2.19(a) E k By AWW K 0 ci 3 Fig 2.190)'50__—Contol Engineering ‘The differential equation of equilibrium are For Node 3, Ka =s0 1 FeV Analogy FVM L:B Rsk bixoas 88 1 aJe-aa = For Node 3 amiiy + By (a — ia) + (a —%) PV amg: Fs ViMoabibo Riko Lie vad di, wa + R(iy-4)+ aJl ihe ° For Node x, 5s + Byig + gx + By(s a) 1 ane FLV Analogy : F> Vi M9 LiBO Rik Gix aid i seen et Lee Ris + Z fides R(b-i) = 0 ‘Hence analogous electrical netwouk is as shown i figure 2.19(b) e Fig. 2.1966): F-V Analogy Example : $ For the mechanical system shown in figure 2.20 draw the (i) mechanical equivalent system Gi) analogous electrical network Also, obtain the differential equation of equilibrium, Mathematical Models —_ 51 £0 Fig. 2.20( fechanical equivalent sytem ‘Analogous Electrical network : It is as shown in figure 2.20(b) EV analogy : FV; M1589 R:k-> Liq: i R, 4, c Fig. 2.2006) The differential equations of equilibrium are52_—Conrol Engineering For Node x, ~ Matematica Models _ 53 mi + By, + kx, + By(% —%)+ B(%-%) = FO al fee 1 rf a ee ee 4 1 diy thee nao fue 8 =) Hi) +R li,-t For Node x; my + By (i ~ Hy) + Bul —) = 0 PAV Analogy: FV; MLB > Rsk bjx oq: k 4 2s myly-i)#Rlis-4) For Node x, mks + B,( iy — 2) + Bulg) = 0 FeV Analogy : FV; ML;BIR;k9 Giz qik ot LG + Rally) #Rlo~i) = 0 Example : 6 Obtain the differentiat equations for the torsional system shown in figure 2.21 also draw the equivalent mechanical system and analogous electrical network. Fig. 221 Solution: ‘As the torque T(0 is applied 1 J, $0 J, and B, under the angular displacement @, ; k and B, tewveen and @, Sirarly J, and B, under dhe angular displacement 0, and Bs between, and 6, Finally J, 8, and, are under, Since here are three angular displacement, 3 nodes are wed ithe equivalent mechanical system as showin igure 221(a) 6 od, : Hence differential equation are | Ror Node 1: 4, + 8,8, + B,(6,~8,) + &,(0, -0,) TV Analogy: T- V;J4L:B3R;k >t é ne reaming 7-9 fis-9cs-+ Este it For Node 2 J + 0, + B,(8, ~0,) +88, -6,)+4@-0,) <0 TY Analogy: To V;J 91; B9R:k> 20g: § 9154 _—Contl Engineering For Node 3 J8,+80,+8,(0,-0,) +0, = 0 1 TV- Analogy : 13 ViJ 9 LiB Rik G09 93691 is. Rig + Rll) +> [igtt = 0 1,94 Ris + Blly~i) Jide Torque - voltage Analogy : ‘Analogous Quantities are L SR ko G09 q:b 7! Tov JL BARRO G04 Here element in parallel under node are connected in series LR, ee do a ») f ‘Torque-current Analogy: Analogous quantities are : f rds TI JAC Bg ik 7109058 na Mathematical Modals —_ 55 Example :7 For the electro mechanical system shown in figure 2.22 find the transfer functi —>Magnet Fig. 222 Solution : This electromechanical system consists of wo pars (i) Electrical (ii) Mechanical In this system, voltage, e() isthe input to the system and displacement, (isthe output from the system, The flow of current in the coil due to the applied voltage, e(0) produces @ magnetic fieke Which produces a force, @) proportional to the current, i) inthe coil AO =i, $0) =k i, (given) ‘where &, isthe proportionality constant depends on the number of turns inthe coil and strength of the maghetic field ‘The equivalent electrical and mechanical systems are as shown in figure 2.2%(a) and 2.22(b) FO=k, 0 Fig, 222(6): Mechanical Sytem Fig, 222(a) + Electrical System56 —Control Engineering Considering electrical system For loop 1 co = Ride Zflale)-a(O “Taking Laplace transform on both sides, we get is) = mG) 216) Z 50) Aa (r+ Lhe 2()+ 18) 5 Mathematica! Models — 57 ) #3) __ A) 19) Tees)” CRCE=) t o- ((cc) z() co al Genes |) Rese) YC) t : +4,9Y(s) (c¢8? +1)(Res+1)-1 - (eae, i) ged sent) 0 = [[ece +ifixce+1)—I] 469) Bice + Co eRCe+ DEMO) . 1s) = fi -o ft +s eo = eelacer tld) 2t9] For loop 2 a I-(Ler+i(Rcs+i)] = Eftuen-atppes ee Substituting vale of 6) nthe equation 3), we get ‘Taking Laplace transform on bth sides, we get tease 24 cron + Beem eiiRcs+ o= En ()- Lube tn ()ern() -2) ) Considering mechanical system Bh, CP RES + 1) Mo) —K, Hts) Co= W) [s? + Bok] [1+ (LCs + )(RCS-+ I) f= mi+B)+(t) ‘Thus, Transfer function, Since fi) = ki,Ct Ye) | kgs Lilo = m+ B50) Bl)” [a eae fia cer RCo Dom K CRC) “Taking Laplace transform on both sides, we get K, 19) = Ys) (Ms! + Bs 8) -® Substituting the value of J, (s) in equation (2), we get 1 Ans o- 0 aarte) i j 2) ea ae +e58_—Conto Engineering A thermometer is dipped in a vessel containing liquid at a constant temperate of 90. The Solution : ‘Thermometer Fig. 223 The rate of heat flow into thermometer is given by aq _ A{t)~2,(¢) a Rise in indicated temperature is given by de, Lda dt at 0, _ 1 (0)(*)-0,(4)) die Camean ‘Taking Laplace transform on both sides, we get H[0.0)-0,6)) (+ zp 400 ‘Transfer function of system, 0,6) Mathoratical Models — 59 Example : 9 Obtain the transfer fun ion for the hydraulic system shown in fj Os) where q~ flow rate C— Hydraulic capacitance R= Hydraulic resistance h-Mead June/July 08 (10 M> —— Tank ls, kz tH _ Tank? h Solution : le 224 ‘The differential equations of the hydraulic system are given by For Tank 1 ah, RC, TE hy = Ria ‘Taking Laplace transform, we get R,C,sH,() +H, (3) = R, Os) Qs) a)2_— Control Engingering ‘or Tank 2 act aay Coking Laplace transform, we get R,C,5H,() +H, (9) = R,Q9) +h, > Raid Ry os 0) HO = Rey Buc R= 2 4 ore 28 Lows His) -3) ** (3) - Alo R, = 2 % (3) = &= 06) = HY) = R019) “@ Substituting equation (3) and (4) in equation (2), we get Ay 2.0 = Reeth Simplifying and substituting value of H,(s) from equation (1), we get 1 ty RO) O89 = Reset)” (RCs +) ‘Transfer fonction of the system is 2(s) Ans O() > (RGeFIKRE e+) Mathematical Models —_ 61 Example : 10 a jon E25) ‘ 5 Obtain the transfer function 77> of the electrical crsit shown in igure 2.25 R R july 2006 (10 Wp r July 2006 (1 1) Fig 225 Solution : bs Rewriting the given elecical network we have Toop Woop Toop For Loop 1: Diffecential equation is 00 = Ry 00% LIlG(~410) & “aking Laplace transform on both sides Ly 21) = 1+ TAH()-LO) Cs, i Rs+D 1) 16) =) For loop 2 : Differential equation is 0 Rie 4 falar ‘Taking Laplace transform on both sides o=R,o+tt a hor En) 143] 0 GH9)+ [CRC + GJS) @62_—Contol Engineering For foop 3 Differetial equation is i eto = Gfulee Taking Laplace transform on both sides La EDT Sls) 18) = Cs B(8) Substituting the value of £5) in equation (1), we get CEs) = (C\Rs+D 16) 8i(5)+ C585) (Gae+t) -@) 1) = ‘Also, substituting the value J) in equation 2), we get 0 = -Gh()+(CCRS+C + ICSE) 19) = [GGRes+G#G),(5) a) Egquating equation (3) and (4), we have C6 (s)+ GsE,(s) (CRs) CORs+C,+C)E,(s) [68+ +GKGRs+1)-C] E66) =C, 0) Transfer function of system is Eels) _ cq Es) Gariaarsaea)-q “™ Example : Obtain the transfer function of the mechnical system shown in figure, writing the physical Dec’ 2011 (8 M) . be Le s(oupu \ HH, Lore. hws fhe Teh vcore mens aise wenn = nO Mathematica Models ~ 63 Solution The Equivalent mechanical system is as shown in figure Fig. 2.26(@) The differential equation of equilibrium are For Node x, : m4 Sidr thay + Fi ~3) 48-2) =0 ‘Taking Laplace Transform on both sides, we get mE K+ hs Kil) +b (5) 4 F 6X (o)~f2X(0) 4k Xy(s)-¥X (=O (mits hers seh +8) X= (Fs2Ox() (rs+8)x(6) Se eacnicnen —(@ For Node x: msi he + Simi) Hama) = Fd) ‘Taking Laplace Transform on both sides, we get mas X()+ fee X()*F 5X) fsXi(s)k X(s)-k X (0) F(s) (mst fees s+k)X() ~ (F548) Xs) =F (0) Substituting value of X,(s), we get (ms? + iss fs+k, +8) (mat fase Fok) [rue fsnfeeky ek) (Fone ANS64 _=Contol Engineering Example : 12 Write the differential equations governing the behaviour of the mechanical system shown in figure 2.27. Also obtain the analogous electrical cireuit based on force voltage analogy and loop equations Dec'2011 (12 M), Dee'2013/Jan.2014(8M) Fig, 227 Solution = ‘The Equivalent mechanical system is as shown in figure OB, Fig. 227 (a) ‘The differential equation of equilibrium are For Node x, = $00, = mig + Bok, + BH — ip) 1 FLV Analogy : F ViM > L:BORiE> Gixoagik oi ai v= Gy t ait Rl ia) Mathematical Models ~ 65, For Node x,: 0 = mi, + Bilis ~%) +b PvAmbgy:F->Vsit- LB Rt-> Exot 4 Fig. 22706) Example : 13 Relate liquid outflow of the second tank with the liquid in flow of the first tank for the hydraulic system shown in figure 2.28 Dec'2013/Jan.2014 (6M) 1 Tank 1 A, % Fig, 22868 _—Control Engineoring Solution + Relation between inflow and out low is given by For Tank 1 aRcih = oF ROG For Tank? a = arRc, a= arene, ay rap = & Eq (1) become 4 = 4+ RCPg, = a #RCP) 4 4 = Gage) Bq (2) become 4 = 9+ RCFa, Equating Eqs.(4) and (5), we get 4 r CREP) = Vt RPS, ‘Thus, relation between inflow and out flow is i, at Re, & sre, Fh #4 =RoRC, SE = wo e a o Example : 14 For the mechanical system shown in figure 2.29 write down the differential equ: ematical mode and Dec'2013/2014(6M) Fe x(outpus) ts f, ___Mathematical Models —_ 67 Solution: ‘The equivalent mechanical system is as shown in figure 2,29 (a) Fig. 2.29 (a) The differential equation of equilibrium are For Node x, : mB + Chi ha, (3 ~ 8) +h) =0 Taking Laplace Transform on both sides, we get ms? Xy(5) + Cys Xi(s)+X4(9) + G 5X (8) ~C, sXa(5)+ 6 Xa(9) 6X43) =0 (rms? + C54 Cy 54 hy + 4) X(9) = (G.s4h)Xa(s) (Gh) fs) X(s) = (ns. G 54 G44, +h) mo For Node x, : m* + Ci — i) +(e ~m)= F(t) ‘Taking Laplace Transform on both sides, we get Im Xal)+ C8 X48) — Cys Xi(5)s hy Xals)~ ky Xs) = F(s) (ms + Gos, )¥260) ~ (C548) Xi(5}= B(s) Substituting value of X0), we get or Xe (ms? +654 Gs +4) FO” (iss cro by nate Oae Gave )a Answer38_—Contol Engineering REVIEW QUESTIONS |. Forthe mechanical system shown in figure 2.30, Draw the mechanical equivalect andthe differential 2quation of equilibrium. Also draw the analogous eletrical nework using F-V Analogy. i eS Ede 1 KS fo Zh 5 I 2, Show thatthe two systems shown in figure 2.31 (a) and (b) are analogous system by comparing thir eansfefonstion, 4 | k ¥, : Fig, 231 () 3. For the mechanical system shown in figure 2.28. Was 5 ¥ aty 1 3 Fig. 232 Draw the equivalent mechanical network and differential equation of motion. Alsodraw the electrical analogy network based on the F-1 and F-V analogy. Mathematical Models —_ 69 ee Matomatical Modis — X,(3) bea ro Pn 8 . Fig. 233 5. Obtain the electrical analogous network forthe mechanical system shown in figure 2.34 based nodal analysis and loop analysis, 4 k By Yas By I + ¥x, fo ig. 2.4 6. Derive the expression for the electrical analogy considering suitable example based on F-V analogy and F-I analogy. 7. Derive the transfer function of a simple liquid level system. 8. Write a note on the model of thermal system, models of hydraulic and pneumatic system. 9. What is a mathematical model ? Explain with an example, 410. Obtain the governing differential equation for a D.C, motor (Armature controlled) Dec.08 / Jan.09 (08 M) I Derive the transfer function of an armature controlled DC motor, where output parameter is ‘the angle tumed by the motor sheft and input is the applied voltage to the armature circuit ‘Dee.07 / Jan.08 (10M) 12, Derive the TF for an armature controlled DC motor: The field current is maintained constant produces a torque T'= f,/ on the rotor. k, and k, are the back emf constant and motor torque constant respectively. Dec.06/Jan.07 (10M)3.1 TRANSFER FUNCTION Any physical system is mathematically described by a set of linear differential equations which are ‘known as system equations. These equations are simplified by using Laplace transformations and ‘aking the system inital conditions to be zero. Initial conditions are taken to be zero, asin actual analysis or design of control system, the system is assumed 1o be initially a esti. initially they have no energy stored in their elements. These simplified system equations are manipulated to determine the ratio of Laplace transform of system output to the Laplace transform of the system input to investigate the system properties. This ratio is known as transfer function of the system. ‘The complete transfer function of the system is obtained by the determining the transfer function of -vatious components in the system and then combining them according to their connections, Definition : The transfer function of a linear time invariant system is defined as the ratio of the Laplace transform of the system output to the Laplace transform of the system input with all initial condition assumed to be zero. Consider a control system represented in time - domain as shown in figure 3.1 i) ——>) of) Fig. 3. In Laplace domain as shown in figure 3.2 (s-plane) Ry——>| Gs) -— co) Fig. 32 1F G(s) be the transfer function of system, mathematically Laplace tansform of output! GS) = “Teplace transform of inp | Block Diagrams and Signal Flow Graghs— 74 a G0) = gy 8 the transfer function ofthe system. 3.2 BLOCK DIAGRAM ‘A Control system can be simple or complicated, but it consists of number of components. Each components performs some specific functions. One ofthe easy and convenient Way of representing ‘the fonction of each component is by the unidirectional blocks. All the blocks are interconnected by lines with arrows indicating flow of signals from output of one block to the input of ancther. Suc block diagram gives an overall idea ofthe inter relationship that exist among various components ‘Thus, block diagram is a pictorial representation of the given control system. 9.2.1 Basic ELewents oF « BLock Diacram ‘Generally, any block diagram bas following five basic elements associated with it {) Block - Its a rectangular box, ofa symbol that explain the mathematical operation on the input to produce the corresponding output. The flow of information is unidirectional. 4i) Transfer function - The mathematical function of each block is shown by inserting corresponding transfer function of the element inside the block. iif) Summing points - Itis 2 symbol that shows the algebraic summation of two or more signal. ‘Theplus or minus sign at each arrow head indicates wether that signa ito be added ar subtracted, iv) Take off point - It is a point from which the signal is taken for the feedback purpose or distribution tothe other blocks, ¥) Arrows - These are the symbols that indicates the direction of flow of signal from one block to the another, ‘Transfer function R(s) eo cs) Od eS ome Input Output Arrows Biock RO), RG) a) __ RO BO) Summing point Take off point Fig. 33 : Base element of Block diagram72_—Contral Engineering 3.3 BLOCK DIAGRAM OF ACLOSED LOOP SYSTEM ‘A block diagram in which forward path contain only one block, feedback path contains only one block, one summing point and one take off point is known as closed loop system or simple or canonical form of a closed loop system, For example, consider the block diagram ofa closed loop system in Laplace domain as shown in. figure3.4 7 > a a > Cs) FBO) He) ke Fig, 34: Block diggram of a closed loop system Where, R(s) = Reference Input C{s) = Controlled variable (output signal) B(s) = Feedback signal E(s) = Actuating signal (error signal) G(s) = Forward path transfer function Hs) = Feedback path transfer function 3.3.1 Transren FuncTion oF Ctoseo Loop System “The mathematical function relating C(s) to R(s) called the transfer Function of closed loop system. In figure (4, C(s) - controlled variable (output signal) and R(s) - reference input are related follows: Cle) = EG) Where Els) = ROS) ¥ BOS) but Bis) = Cts) * E(s) = RG) = CHS) ‘Thus G(s) = ERG) = C(IMOIGS) Cs) = RWG) ¥ CHING) C{s) F Ch) HI)GIB)._ = R(SIG) CG) [1 £ HGS] = REGIS) [2 nese sem ee eae wwhce rate ltr mpi eck Negative (~) sign is for positive feedback, Block Diagrams and Signal Flow Graphs — 73 The closed loop TF. can be represented as in figure 3.5 RG) a a TEG@HO ) Tiga 3.4 REDUCTION OF BLOCK DIAGRAM ‘The transfer function of a block diagram can be obtained easily when it is represented in its simple closed loop form in which forward path consists of one block, feedback path consists of one block, ‘one summing point and one take off point. But, block diagram of control system, generally consists ‘number of blocks, interconnected to each other. These are brought to the simple closed loop form by the reduction of block diagram, this is achieved by the use ofthe block diagram reduction rules. 3.4.1 Bock Discrau Repucrion RULES: Rulel: Combining blocks in cascade (series) ‘The transfer function of the blocks which are in cascade get multiplied with each other RG) 9 ) 6 © ce Loo LS . cago LY ‘Hint: If there is a summing point or take off point in between the blocks, the blocks ‘cannot be said to be in cascade (series). Rule 2: Combining blocks in parallel The transfer function of blocks which are in parallel get added algebraically. 1 6a te Re o9 ‘ e Ot Go QO = & | G9) ~ Gis) + 6, LE 1 oo Rule 3: Moving a summing point after a block sac Rs ao fA = 20.1 x COs) = REG) + GO CG) = RG) +9 GIS) Cis) = (RIS) +) GES) Gs)_ fox74 Control Engineering _ ‘Hint: The output must remain same, while moving the summing point Rule 4: Moving a summing point before a block a2 o | Rs) 4 sures | C) 04) = RNG +« | ' evn [RO + eye] Ce) = RIGS) + x Rule 5: Moving take off point after a block 4 Re A, >| Ge) an = “3c C13) <_ RS CGS) = RESIGLS) 7 roLG_| Row Hint: After moving take off, value of the signal taken off must remain same Rule 6: Moving take off point before a block ros = FO, as) > cs) _ <-f Gis 1 R(s)G(s) R(s)G(s) Rule 7: Eliminating a feedback loop 04 ce) no {a6 9 ee TAAL aw 3) |e Caxtysz Caxtety Rule 9: Splitting a summing point A 2 Caytits Carthee General Procedure The General steps involved in the reduction of block diagram are 1) Combine all he blocks in cascade 2)Comibine al the blocks in prale! 3) limite al minor feedback loops 4) Shift Summing points and takeoff points if necessary 5) Repeat steps 1 to 4 until canonical form has been obtained $9 Using sander ante: ston of simple cheep steno closed op TF ofthe overall system, Hint: As for as possible try to shift take off points towards right and summing points te the left.= Control Engineering WORKED EXAMPLES ample: 1 educe the block diagram shown is figure 3.6 to its simplest possible form and find its sed loop transfer function, Dec.09/Fan.10 (106) 1 3 Ps Ci OY G, Fig. 36 lution tep 1: Move the take off point (4) after the block G, a 7.0) Block Diagrams and Signal Flow Graph Step 2: Move the summing point (3) before the block G, 6 8) [seperate the path, Step 3: Inter change the summing point 2 and 3 ‘Combine the parallel blocks aa ants G, combine the cascade blocks ‘Separate the path along the dotted tine and H, Also G, and G, Minor feedback loop Step 4: Eliminate minor feedback an¢ Conte te arts ant 2rol Engineering 6 —GGs _ GE TGGHh cs) Minor feed ‘back loop Combine the cascade blocks GGx a THGGA)G, ~ U+GGt) ‘Eliminating the minor feedack loop (G(G:G,+6)) Uaraem ) —___G(66,+6)__ i G.8Th) * (GGA) HCG, +ENG+ A) Tha, rp os) R)—> Gan )HG Closed loop TF Example : 2 Obtain the over all transfer function of the block diagram shown, technique. G6,6,+ 66 FGG,G,+ GG, +6, figure 3.7 by reduction Jan/Feb 2005 (10M) Eas Solution : Step L : Combine the two summing points 2) and (3) Block Diagrams and Signal Flow Graphs ~_79 > CIs) RG) a > Cs) > Cs) ‘Step 4 : Combine the cascade blocks + RO) > C(3)0 —Control Engineering Block Diagrams and Signal Flow Graphs — a1 Step 2: Separate the paths along the dotted line step 5 : Eliminate the minor feedback loop -— 4 (G+G)G, roa 2 By * GH) +G)GH, cw EG) gen, O“GUNDHG +) G | . ne 7, RO) G ets) TO GHG GG ao Berio ais Step 3: Move the summing point (4) before block , and inter change summing points (3) and (@) cw) GG, +6,6, RG)” HGH +GG,H, + GG Example : 3 a pu ew) Reduce the block diagram shown in figure 3.8 to its canonical form and determine its y control ratio RR 4k ie ] Step 4 : Move the take off point (1) after the block G, and Combine the minor feedback loop Step 1 : Move the take off point (4) before summing point (3) RG) z ea 7 ! HLH,82 —Con Engineering ‘Step 5 : Combine the cascade blocks Split the summing point (2) “pr Parallel <_< “Minor feedback lop \ ! \ 1 A T Ti 1+6,H, Lot, 6G.+G, GG, Rs) GG.+G, GG, Step 7 : Comibine the cascade blocks G, (6G,+6) (56,8) * (GH, GH) Block Diagrams and Signal How Graphs — 83 GG,+6, | cw (4+G.H,\l+ GH) “HA, Step 8 : Eliminate the minor feedback loop Ga+e (FGHNGH) (GG, +) GGG) * (FGA \I+G,H,)-(6,0,4 6) mi, Ore (ie ayaa) MM) RG) |__ (GG, +G,) | as), = (1+ GH, I+ GH) (GG, +6), ~ _Blnint moe edb op OG I+ GH) (GG +G NH, te GGG) (HGH UG A)-(G.G, +6,)HH, (GG. +6.) Hy) ~ (GG, +G,) HH, + (GG, + Ganj 4°.) Taman24 _—Contol Engineering Control ratio is given by ow |_____ Ci ee RG) ~ 1G, + GH, + GG, Mill, ~G,GH Hy ~ GH Hy + GGH, + Gull as) GG, +6. RQ) “1G, GH, - GH, + GGT, +GH, “"* Example + 4 Reduce the given block diagram shown in figure 3.9 and then obtain the transfer funetion of the system if G, and H,= H,= 1, H, July 2007 (10M) cy 5 a 3 Fi, Solution = Step 1: Split the summing point (3) and separate the peth’s G, RO), Step 2: Eliminate the minor feedback loop = 7, G 7 ‘Move the summing point (3) before block G, and interchange the summing point (2) and (3) Block Diagrams and Signal Flow Graphs ~ 85 Cascade oo), a GtGG, ) _ - GH, 2 [> 1+GyHy GG, (G.+6,G,)G, Step 5: Combine the cascade blocks = 75 ey ga 2 " ide blocks = 13 GH, + GG\H26 _~Controt Engineering (GsG6,)o,_ a TG, + GG a Ht |e Step 6: Eliminate the minor feed back loop (G,+G,G.)Gs u (G.+GG,)6, hGH, +O GH, + 6,G,H, +G,G.GH, RG) (GA GGG, ce OF) FGM FOG + GCM +GGGR, ‘Transfer function of the system is, is) | (G+ G,G.)G, RW) ~ 1+G,H, +GGH, +G,G,H, +6,G,GH, 1G,=6,=1 2: and H, = Hy=1, =? Then os) 6 Rom AN Example : 5 Reduce the given block diagram and write the overall transfer function of the system shown in figure 3.10 July-August 2005 (8 M) 7, 4 . Co 5 Block Diagrams and Signal Flow Graphs —_ 87 Solution Step 1: Move the take off point (5) to before block G, and interchange point (4) and (5) 1 2 RE + as GIS i, Step 2: Combines the cascade blocks = H,6, G Eliminate the minor feedback loop = 2 Sem aalEG a G ecw) Step 3:88 _—Contral Engineering Rosy T & ots) s C GG, Step 4: Combine the eascade blocks = Tae SG, Eliminate the minor feed back oop GG, 1+G,G,H, +6, GG, Me GG og GG + Git, +EGH, 156,61, +6,H, GG, cw, —_ 9G |), 1¥G,G,H, +G,, +6,G,H, & Step 5: Combine the cascade blocks RG GG.6, cw, Sl EGR GH GG)” ‘Transfer funtion ofthe system CW) | __GGG, agg RO j,G,H, + GH, + GG, H, ae eee Example : 6 a Obtain C° ofthe system in figure 3.11 by lock diagram simplification, July 2006 (10 M) Block Diagrams and Signal Flow Graphs — a9 Hi, Solution : Step fove the take off point (4) after block G, and interchange point (4) and S Step 2 : Combine the cascade blocks = G,G, GG,_ ‘Eliminate the minor feed back loop = 1+G,G,H, RG) HIG, ‘Minor feedback loop Step 3 Combine the cascade blocks = 22%, ep 3 : Combin looks = Eliminate the minor feed back loop >Cls)90_—Contol engineering TGGth GG GGG Hh ~ E.G GGA, T+ G,G,Hh,* G, Step 4 : Eliminating unity feedback loop x1 GG.H,+ 6,6, RG) GGG, %)- 156,G,H, -GGM, peeeeene Gye 1-G.G,H, +6,G,H, +G,6,G, Rs) __ G66, | 1 GiGi, + 6G, #GG,G, co) _ Gi eC Cee Ri) ~ 1-G.G,H, +G,G,1, +,6,G, Example : 7 ion © Reduce the block diagram, shown in figure 3.12 and find the overall transfer function =. Dec. 07/Jan.08 (10 M) RS) Block Diagrams and Signa’ Fow Graghs ~ 91 Solution = Step 1: Combine the parallel blocks G, and G,= G, + G, cs, Step 2 : Combine the eascade blocks = G(G, + G)G, Separate the paths along the dotted ines 6) 6,046, + GF ~—rersy GH, Hy Step 3: Splitthe summing point cis) ‘Step 4 : Combine the parallel block = I- GH Eliminate the minor feedback loop GG(G, +65) 1+G,G(G,+G)-H) iGu(Ga + Gs) . a sa} a Cos) Step + Combine the cascade block RS) GGG, +6)92_—Control Engineering ‘Overall transfer function is, CQ) _ (I Gslt,)G,G,(G, + G,) Ro) = 1-G,G,HH,(G, +6) Example: 8 Reduce the block diagram shown in figure 3.13 to canonical form and determine transfer fon £® function 2, , G,= Hy June/July 08. RG) 1 2 = . RQZ 7 6 cy) G i - : 3 7, #, ks RO) Step 2 : Combine the cascade blocks = G,G, GG, Eliminate the minor feed back 0” = 7 G.G 7, Block Diagrams and Signal Fiow Graphs —_ 93 > C3) Cascade & Step 4: Combine the cascade Block = TG. HG ~iminaing the minor feedback oop 66.6, GGG, GGG, XH, GH +G,GH+G.G.GH, ps)94_—Contol Engineering Brock Diagrams and Signal Flow Graghe—~ 95 ‘Transfer function is, ok Diagrams and Signal Flow Graphs — 98 Step 2: Eliminate the minor Feecback foop 2 GGG, RO) ~ 14G,H,+6,6,4,+G.G,G4, =) (54 1(6+2)(0? 1) +4542) 1x2x3 - T42x242x3x341%2x3x1 Example : 9 For the system shown in figure 3.14 Determine Closed loop transfer function fi) Characteristic equation ‘System type “ iv) System differential equation L + T T + T R(s) — s) > sea os z >cts) 1 = Step 3 : Combine the cascade blocks = yp ez P Snianmatummmneisl oa Kata) e4sr0 Eliminate the minor feedback loop 1 2 . (646+ 2 I eais+2) a i Grier Ie ease? Solution : ‘Step 1: Combine'the cascade block x( +25) 1 Eliminate the unity feed back loop 2s irae 2ee 25 Splitthe summing point 1 Os) > Gea Ry 1 6 442 (3) cs) S43P HIS SS Closed Transfer function is, oe \ R(s) s+ 39 +257 43516 _—Conteob Engineering Block Diagrams and Signal Flow Graphs — 97 Srample : 10 , Step: Eliminate minor eed back oop = 5 ine the overall transfer function of the black diagram shown in figure Determine the overall transfer f ar Seu wom B : : c k x _ep—fa-$ ; Step 4: liminate minor fed back loop ie. overall ‘Transfer function a GG6, ‘ 19G.GH,+GR, NS. Ts Solution: Step 1: (@) Combine blocks G, and G, = 6,6, (b) After combining move ie takeoff point 1 ahead of GG, R EE i (©) Rearrange the system | aGRram -—— a i GG, >C e {G, +O, | Example + 11 = Reduce given block diagram into canonical form and determine closed loop transfer function. Also represent in open loop form. Dee? 2013/2014 (10M) x toop «21 se 2) Eliminate minor eed back oop : Siep2+ (Bint oem - Go,6, (©) After eliminating combine with +G.GH, GGG, TGGt, > Fig38_—Contiol Engineering 28_—Contol Engineering Step 1: (a) Eliminating unity feedback path = $7 1 (b) Combining blocks in parallel = 24 2 04s) 1 1 25? +8543) +09) uy stl] |s+2| s(s+3) 2 Fig Step 2: (2) Combining the blocks in Cascade (Series) 1o(2s? +8543) © S(s4I)(s+2)(5+3) 5) 10(2s? +85-+3) 10) a(t Iyer 2)(o+3) Step3 (0) Eliminating feedback path toast s8s+3) Geese _ wos +83) Set Nes 2\e+3) WoQt+843) © S(o4)(e+2)(0+3)+ 2028 +85+3) Block Diagrams and Signal Flow Graphs — 99 1o(2s8 +8643) 8° 465 +4957 + 1625-466 £- opentoop form is 102" +8543) 6) —— eee S657 +495? $1625 4 66 +——> 8, «s) ‘Thus, Transfer function is (3) 20s! 480543 544659 +495? +1625 +66 () 3.5 SIGNAL FLOW GRAPHS A signal flow graph is an another way of representing a linear control systems, when any control system is described by a set of linear algebraic equations having the form a= Soy Where i212, n ‘The signal flow graph (SFG) was developed by S.J. Mason for the control system described by aa algebraic equations to represent the cause and effect relationship among the various variables of the system, Thus signal flow graph can be defined as The diagramatic or graphical representation of the variables of a set of linear algebraic ‘equations representing the linear control system is called signal flow graph (SFG). 3.5.1 Basic Evewents oF a SFG Im SFG, variable plays a vital role, these are represented first. Thus variables ofthe set of equations representing the control system are considered as the first basic element and are represented by the small circles called nodes in SFG. The control system consists of number o° Variables, which are both dependent and independent in nature. Each (dependent and independent) variables are presented by separated node. The nodes are connected by directee line segments called branches, according tothe cause and effect equations. The branch is associated with the transfer function and an arrow. The transfer function represents (branch gain) mathematical “ operation on one variable to produce the other variable. The direction of the atrow indicates the flow of signal For instance, consider that a linear system is represented by the simple algebraic equation100_—Contol Engineering ‘where x, isthe input, x, is the output and a1, is the gain between the two variables. 9.5.2 Tenus useo iv SFG In addition tothe nodes and branches defined earlier, the following terms are useful forthe analysis of SEG ‘Consicera signal flow graph shown in the figure 3.15 Pig. 315 i) Node : Nodes are the variables of the system represented by small circles. ii) Input Node : The node that has only outgoing branches is known, as Input or source node, Example : x, is Input node. i) Output Node: The node that has only incoming branches is known as output or sink node Example x, in output node. iv) Mixed Node : The node that having both incoming and outgoing branches is known as mixed ‘or chain node. Example : x5.) Xp %, and X, '¥) Beanch : Directed line segment joining two nodes is known as branch. vi) Path : It isa traversal from one néde to-anothex node in the direction of te branch arrow, such that no node traverse more than once. vii) Branch gain : The gain between nodes is kno\ are expressed in terms of transfer functions. viii) Forward path : The path that stats from an input node and ends at an output node and along hich no node is traversed more than once is known as forward path. yn as branch gain or transmittance, such gains Example: BTN BOAT HOHOE ix) Path Gain ; The product ofthe branch gains encountere« pati is known as path gain or forward path gain. 1d while going through the a forward Block Diagrams and Signal Flow Graphs 103 Example : Consider a forward path 13 4X yg X gg XI x) Feedback loop : A path which starts from a particular node and ends at the same node, travelling through atleast one other node, and along which no node is traversed more than once is known as feedback loop or closed loop. Example: %,-x,-x,-x, xi) Self loop : A path which stats from 2 particular néde and ends atthe same node. Example : x,—, Hint : A self loop should not be considered while defining the forward path, i) Non - touching loop : If there is no node common between the two Or more loops, such loops are said to be non - touching loops. . xiii) Loop Gain : The product of all the gains ofthe branches forming loop is known as loop gain. 4 ts path gain is 3.5.3 \PRopennes oF SFG {) The SFG can be applied only to linear systems. ii) The equations for which SFG is drawn must be a linear algebraic equations in the form of cause and effect, il) Nodes ate used fo represent variables and line connecting between nodes in represented by branches. iv) The signals along the branches represented by arrow. ‘v) The signal traveling along branch is multiplied by the gain of the branch. vi) The branch indicates the functional dependence of one signal'on another. vii) At each node all the signals of in coming branches added and transmitted to the out going branches, Hint : The value of variable at each node is an algebraic sum ofall the signals of incoming branches. vii) A mixed node may be treated as an outputnode by adding an outgoing branch with unit branch ain 3.54 Proceoune 10 Draw SFG Case I: From block diagram 1 Replace all the summing points by nodes, 2. Replace all the take off points by nodes. 3, Replace all the blocks by branches, indicating block 4, IFthe branch connecting a summing point and take off point has unity gain, then summing and take off point can be combined and represented by a single node.192 _—Contol Engineering H, 5. If there are more take off points froma the same signal then all take off points can be combined and represented by a single node. 6. If the gain of the Link connecting two summing points is one, then two summing points can be ‘combined and can be replaced by a single node, 4 Case II; From the linear algebraic equations 1. Replace all the variables by a separate aode. 2. Coefficients of the variables are represented as the branch gains. 3. To complete'the SPG, show separate input and output nodes. 3.6 MASON'S GAIN FORMULA, Mason's gain formula is used for the determination of overall transfer function (Input - output telaonhip) of assem, Iisa convenient and ext a of Finding he relation eve he inp and output variabtes of a system. The number of steps involved in the block diagram reduction are ‘more anditisa time consuming procedure and also the ts of solving forte input oust elaonship by algebraic manipulation could be quite tedious. An advantage of Mason's gain formulas that system transfer functions are readily obtained without manipulation of the graph, Block Diagrams and Signal Flow Graphs Mason's gain formula is given by 1 x La Number of forward path Overall TF, Where k P, = Path gain of & forward path A = Determinant of the graph = 1 (atm of individual loop gain) + (sum of gain products ofall combinations of two non touching oops) ~ (sum of gain products of all Combinations of three non touching loops) + ‘Value of by eliminating all loop gains and associated products which are touching to the k* forward path Example : 12 cs) For the system shown in igure 3.16 determine £1" using Mason's gain formala, a Dec. 09/Jan.10 (10 M) Rs) Solution : Fig. 3.16 Step 2: Identify the number of forward path and their gain Forward path gains are 1x1 G,xG,xGX1 = 6.6, | x 1% 6, xG,x Gxl = GGG, |THE me wo Forward paths. k= 2 Step 2 : Identify the individual loops and their loop gain, <= Loop gains are ‘There are four individual loops104 —Convral Engineering ‘Step 3: Find the Combination of non touching loops 1. Combination of two non touching loops LL, = © GGA) x ( ‘Noother esmbination of two of more non touching loops Step. : Find the value of determinant A A = Determinant of the graph ~ (um of individual loop gain) + (sum of gain products ofall combinations of two non touching loops) ~ (sum of gain products of all Combinations of three non touching lops) + 1-4 Ltt L)+ dy) ~ GG.H,~ Gl, - 66,0,~ 66,61 + 66.64", 1+ GGH,+ GH, + 6,0,0,+ 6.,6,+ 6G,G HH, Step 5 : Find the value of 4, 4, \, = Value of A by eliminating all loop gains and associated products Which are touching to the k* forward path ‘Since from the SFG, itis seen that all the loops are touching all the forward paths, wehave A, = ‘Thus, From Mason’s Gain Formula, Sinek=2 ie, Overall wansfer function, ls) RG) “TOG, + GH, + 6.6.6, +6,6 Ans. jG G,6,HH, Example : 13 Using signal flow grapk and Mason's gain formula, obtain the overall transfer function of the system depicted in figure 3.17. Dec. 07/Jan.08 (10 M) Re Fig. 317 Bock Diagrams and Signal Flow Graphs ~ 105 Solution : - Step 1: Identify the numberof forward pot and their pain ~. Forward path gains are 1x6, x6, 6.x 6, 1xG,x6,x 6x6, ‘Step 2+ ldentfy the individual loops and thei oop gain + Loop gain are = 6H, GH, 1, = 66,66, -) 1, = G,6,66,-1) Step3: Find the Combination of non touching loops 1.Combination of two non touching loops LL, = © Git) x — GH) = 6.6.48, Nother combination of two or more non touching loops Step 4: Find the value of determinant { = Determinant ofthe praph ~ (samo individual loop gan) + (Gum of gan products of ll combinations of two non touching loops) ~ (sam of gain products of all Combinations of tree non touching loops) + » Aw IAL, +L +L +h) +b) 1 AE GM, ~ Of, 6.6.6,6,- 0,6,6,6, + 10,6,44) 21s Gil OF,+ 6.6.00,» 6006+ sorte: Step 5 : Find the value of a, ° 4, = Value of by eliminating al loop gains ad associated products which ae touching tothe & forward path Since from the SFG, itis sen that ll he loops are touching al the forward paths, we have 4, =Aj=1 Thus, From Mason's Gain Formula, 12 ayt | Overall transfer funtion, GE,66, G,G,6,G, | TE HE two forward paths. k ‘There are four individual loops r CG) GGGGAGEO0, RG) 1G H+ Gl, +6,0,6,6, +6,6,6,6, +0.0,1,M, Ans.106 —Control Engineering Example : 14 is c(s) FFor the signal flow graph shown in figure 3.18, Determine the transfer function using Mason's gain formula. Dec'08/Jan’09 (10 M) © CW) Solution : cee ‘Step 1 : Identify number of forward path and their gain Forward path gains are xG,xG,x1=GG, x6,xG,x1=6,G, XG,xG,x G,x1= 66,6, = 1X G, x Gx G, x1 = GGG, ‘Step 2: Identify individual loops and ther gain loop gain +. Four forward paths ie., k= 4 P P P, Gag, P. Loop gains are LG, l= - GH, . L= 6G, _ } Therwarefiveingividuat loops Ls ~ GHG, = GHG, ‘Step 3 : Find the combination of non touching loops and their gain 1. Combination of two non touching oops are [by = GH Git, 2. There is no Combination three or more non touching loops Step 4 : Find the value of A ‘A = Determinant ofthe graph = 1-(um of individual loop gain) + (sum of gain produets ofa combinations of two non touching Loops) ~ (sum of gain products of all Combinations of three non touching loops) + Hs tht bt blt hh) 107 Signal Flow Graph = 1 GH Gt,~ G,G,+ CH,G,+ 6.6,44,+ A= 1+ GH OH,- 6,6, GH6,+ 6.G,H,+ ind the value of A, HH) GH Step A, = Value of by eliminating all loop gains and associated products Which are touching tothe K* forward path Since from figure itis seen that forall the forward pats all the loops are touching A= Ase d=1 — fork=1,2,3,4) ‘Thus, From Mason's gain formula ___ OG, +6,6,+ 6,0,0,+ 26,6, GH,+ Gull, ~ GG, + GHG, + G.GH, OGM, 8 a For the signal low graph in figure 3.19 find {using Mason's gain formula, Tone/July'08 (10 M) Fig, 3.19 Solution : Step 1 : Identify the number of forward paths and their gains Forward path gains are x G,% 6, G,x G,x G,* G,= 6.G,6,6,6,G, 1k G,xG.xG,x6,= 66,66, ‘There are 3 forward 1x Gx G,x Gx Gx G, COGG, | paths eves108 —Contro! Engineering Step 2: Identify the individual loops and their loop gain s Loops gain are E- -Gn, = -G,G,H, 1 COGGH, een There are six individual loops - 6, 6G,66H, = - 6, G,6,G,4, ind the Combination of non touching loops 1, Combination of two non touching loops Lib, =~ G, Hy X- G, 6G,G,= 6, CGC CHM, ‘There are no combination of three of more non touching loops Step 4 : Find the value of determinant & 4 = Determinant of the graph 1 ~(Gum of individual loop gain) + (sum of gain products of i combinations ef two non touching loops) ~ (sun of gain products ofall Combinations of three non touching loops) + Aa Ht Let Lot Let byt + ibd Step S: Find the value of 4, ‘A, = Value of by eliminating all loop gains and associated products which ate touching to the K* forwand path Toffind A, : For P, forward path, all the loops are touching « &=l ‘Tofind A, : For P, forward path only L, is non touching, 2A, = 1—L,isnonsouching ‘Tofind A, For P, forward path, all loops are touching s Ae “Thus, From Mason's gain formula, ix Pact Pda + P, 3S na, o Bbc ae 4 7 4G, G,GsG + GE,G)Ge(1+ Geta) + G\G2CsOGe 4 1 Gut 6.G,Hy- G,6,6,6.G,6H,~ Gls GBBGEM, ~ G.GG.GH) + IC GH) + © 6,6,6,64)) = 14 Gl + Gh, + COG EGGH, + GH,+ GO.C,GGM, + 6,G,6,G,H,+ 6,6,6,6,6, 4H, Were ‘lock Diagrams and Signal Fw Graphs — 108 Example : 16 a c Obtain the overall transfer function < from the si fer function > from the signal flow graph shown in figure 3.20. July/August 2003 (10 M) Dec’2012 (WOM) Solution Fig, 320 Step 1 : Identify the number of forward path and forward path gain ‘The forward path gains are ?, = 6,6,6, There are six forward path. k= 6 + ind inivival loops an heir lop gin + Loop guns are E64, -GR, -G.H1.6,H, | Step3:Find the combination nos touching oops nd their op gains Combination two non touching loops are bs = GG,H, Ps two aon touching loop and 0 combination thre and more non Step 4: Find the vale ofA “Phere are 3 individual Loops 4. = 1-[sumof individual loop gain] + {sum of gain product of all combination of two non touching loops} — {sum of gain product of all combination of tree non touching loops] +1 8 = a fhl th) + bl 1- EG, G+ GHG) + (G,H,G,Ft) = 1+ Gt CH GULGG, + QU GSH, Step 5: Find the value of A, = Value of by eliminating all loop ssn and associated produets which are touching to the k forward path Since k= ‘Tofind A, :Consider P, check which is the loop non touching tot. For P, Lis non touching 4, Simitary, ‘Toffind A,: For P,,L, is non touching A= 1-L,=14 GU, ‘Tofind A, For P,, all loops are touching it. Fa Al 4, To find 4; For P, all loops are touching it n=l Tofind A,: For P, all loops are touching it. A=1 ‘Toffind A, For P, all loops are touching it. en Thus, From Mason's gain formula by ag GGG + GsHl,) + GyGsGy( + Gut) + GG Gel 1) + GGG,(1)~6,6,HsG,G«(1) - GGG, 1+ GH + Gill = GiHG,G, + GyF.GsH, Example: 17 Explain Mason’s gain formula, use it to determine the transmittance of the flow graph 1 =: Jan/Feb 2006 (10 M) sl shown in figure 3.21 given A = B= Block Diagrams and Signal Flow Graphs — 111 Fig, 3.20 Solution : Step 1 : Identify number of forward path and their gain - Forward path gains are P, = 1x2xSx1=10 Py = 1x3x6x1=18 Py = Lx4x7x1=28 2 1x2xAx6x1=12A=— Hl + Six forward path ie, k=6 Py = 1X2KAKBx7x1=14AB 1x3xBx7x1=21B ‘Step 2: Find individual loops and loop gains There are no loops Step 3: Find the value of & ‘Since there are no loops asl and A, = Lfork=1,2,3,4,5,6 Thus, Mason's gain formula is Pa. z PAL + Bas + Ba: Bast BA, = Ty AA Bat Bas + Bast Fay r ay roriss28+ 124 #1" Gai) u Hi Ss) +12(s+1)+14421(541) (s+1) a142 _—Conto! Engineering 565+ 14514103 Thus, transmittance = = = 7 Ans, « (eH Example : 18 Find the transfer funetion ofthe system shown in figure 3.22 using Mason's gain formula. July-2007 (10M) RS) Fig. 3.22 Solution ‘Step 1 : Identify the number of forward paths and their gains «+ Forward path gains are P,=1XG,x Gx G,x Gx G,x1= G.G,6,6,G,| «. Two forward paths ‘Step 2: Identity the individual loops and loop gains Loop gains are «Three individual loops Step 3 : Find the combination of touching loops and loop gains 1. Combination of two non touching loops are LL, = G,G,H,H, 2. There is no combination of three or more non touching loops Step 4 : Find the value of & A = Determinant of the graph = 1 (sum of individual loop gain) + (sum of gain products of all ‘combinations of two non touching loops) — (sum of gain products of all Combinations of three non touching loops) + a= 1-[hthth)}+[hh) I-[-G.H, - Gilly ~ 6,645] +[6,6,H Hy] = 1+ GH, + GH, + OGGH,+ 6GHM, Block Diagrams and Signat Flow Graphs — Step $ : Find the value of 4, 4A, = Value of & by eliminating all loop gains and associated products which touching to the £* forward path ‘To find D, : For Pll loops are touching Ae ‘To find D, : For P., L, is on touching loop : 4,2 1-L,=14+6,H, Thus, Masor’s gain formulais given by ly Pat PBA, Gin = $3 Ad ,= ABM ob) GG,G,G,G,+G,G,\1+ GH.) aa RG)” 1+G,1,+6,H,+6,6,6,H,+6,G,i,H, 8° Example + 19 ct July 2006 (10 M15 Fig. 323 Step 1 : Identify the number of forward paths and their gains Forward path gains are P, = 1x1xG,x1%G,xG,x1= 6G, P, = 1x1xG,x1*G,x1=6,6, ‘Step 2 : Identify the individual loops and loop gains +: Loop gains are ‘There are 5 individual loops GG, G,-1) = 66,6, GING, Ch = 6G,114 _—Control Engineering ‘Step 3 : Find the combination of non touching loops ‘There are no combination non touching loops Step 4: Find the value of A {8 = Determinant ofthe praph = 1 (Gum of individual loop gain) + (sum of gain products of all combinations of two non touching loops) ~ (sum of gain ‘products of all Combinations of three non touching loops) + A= [het tht] Step 5 : Find the value of 4, 4, Value of A by eliminating all loop gains and associated products which are touching to the & forward path Since all the loops are fouching the all the Forward paths we have A= 421 Thus, from Mason's Gain formula als) Rs) ——GiGGj+GGe____ 15 GH, + G,G,H, + GH, + GGG, +6G, Ans. Example : 20 Draw the signal flow graph for the block diagram shown in figure 3.24 and find its control ratio using Mason's formula, Dec’06/Jan'7 (10 M) ee es) 6, > | RG) i) Block Diagrams and Signal Fiow Graphs —_115 Solution + Step 1 : Represent all the summing points and take off points each by a separate node and show ow of signal as shown in figure 3.24), Fig. 3240) Step 2: Identify the number of forward path and their gain + Forward path gains are p, ee) 2. There are Two forward path PL = Ix1x1KG,x1x126, ie be? Step3 : ett the individual loops and thee gin eee GH, L1G, x Gx CH) =~ 6,6, | 2 There ate Sindvidual oops L, 1 = LL GLX CH) = = GH, Ly = 1x Gx 1H) x1 x CHD = GH, Hy Step 4 :Find the combination of on touching loops : There io combination of non touching loops Step 8 : Find the value of & 4 = Determinant ofthe graph = 1-(sum of individual loop gain) + (sum of gain products of all combinations of wo non touching tops) ~ (Gum of gain products ofall Combinations of three non touching tops) + Ae 1[hththth+h] = IAG, - GH, ~ G64, ~G,t, + 6,001] = 1+ GH, + GH, + GH, + GH, GH, Step 6 : Find the value of 4, 4, = Value of & by eliminating all loop gains and associated products Which touching to the k forward path ‘Since all the loops are touching all the forward paths, A= A=16 — Control Engineering hhus, From Masori’s gain formula 1y Ray+ Roy 1y pa, - Ba re rn Gain GG.+G, RG) > 1G, +G,H, + GG, + GH, GH, EEE Fxample + 21 Ans. ie, R in X inthe SFG shown in fig. 3.25, obtain 5. Solution = _ ‘Step 1 : Identify the number of forward paths and their gains :, Forward path gains are P, GGG, | P= 6,6, |, rourforwarpatsie, t= ees rece | suey: eels np see i 4, | There are 2 individual loops L, = G,G,H, ‘Step 3: Find the combination of non touching loops ‘There are no combination non touching Loops Step 4 : Find the value of & ‘A = Determinant of the graph Block Diagrams and Signal Flow Graghs ~ 117 (sum of individual loop gain) + (sum of gain products ofall combinations of two non touching loops) —(sum of gain products of all Combinations of three non touching loops) + [4+4] — [H+ GG Hy Step 5 : Find the value of A, 4, = Value of by eliminating all loop guns and associated products which are touching to the k* forward path wwe have 4, = LA=1-L,=1- Hd, ‘Thus, from Mason's Gain formula rida, GG:GG, + GG H,) + 6,G.G,0-H,)+ 6.6.6, 1-(H, +G,G,H,) Fal + Rs + RAs + Pade a zxample : \ M . A, Fig. 3.26 Solution : Step 1 : Identify the nuinber of forward path and their gain “Forward path gains are P.=G6G, Pix G,G,G,_ | THe ae two forward paths. k= 2118 Control Enginoering ‘Step2: Identify the individual loops and their loop gain Loop gains are ‘There are four individual loops Step: ind the Combination of non touching loops No combination of two or more non touching loops Step 4 : Find the value of determinant A ‘A = Determinant ofthe graph (sum of individual loop gai) + (sum of guin products of at ‘combinations of two non touching loops) ~ (sum of gain products of «ll Combinations of three non touching loops) + 1-4 +h+L) - GHy- OG.H,~ 6.G1,+ 6.6.64.) A, = Value of by eliminating all loop gains and associated products which touching to the * forward path ‘Since from the SG, it is seen that all the loops are touching all the forward paths, we have A,=1 ‘Thus, From Mason’s Gain Formula, Fai + Ped Daya Mat Ble GGrG,+ G.G,G,(1+G.t, Th Gi + GG:H, +G.G,Hl, GGG, ie, Overall transfer function, CC GG.6,466,6,+6,6,6,0.H, R ” 1FG,H+G,GH,+GGH,— tos Block Bi yams and Signal Flow Graphs ~ 119 23 Example Use Mason's gain formula for determining the overall transfer function of the system shown in figure 3.27. Dec'2011 (10 m) G, Fig. 327 + en the numberof forward pth and thee gan Forward path gins are 6a, A=G, Step2 Identity the individual loops and their top gains 2. lop gins are 12 =] sa L=-H, Ga, con, | ‘There are two forward paths <. 2 There are Four individuat loops Step3+ Find the combination of non touching loops ad ther gin 1. Combination two non touching loops are LL, = Wi, + LL,= 4,6, Ther is other combination of to or more nos touring loops Step 4: Find he vale of B= 1[Lt hth th )e[b+ hb] 1-[-th~ H,~ GH, ~G,6,H,] +[,, + 1,641] Step 5: Find the value of 4, A= 1 for k=t A, = 1-L,-L+QL, +H, + H+ HH,20_—Contot Engineering “hus, from mason's gain formula “hus, overall transfer function c GGG.+G.[h+ H+ Hy +H, Hh] Ro 1s [i +H, +GH,+6,G,H,] +[HH, + Gl] Ans. Gxample : 24 ‘ind the control ratio for the signal flow graph shown in figure 3.28 by using Mason's gain ‘ormula, Dee? 2013/Jan.2014 (10M) Solution Step 1 : Mentfy the number of forward path and their gain - Forward path gains, 6G6,6, 6G, Step 2 : Identify the individual loops and their loop gain ‘There are two forward paths -. k= 2 Loop gains are 1 =- Ga, ~ G,6,G,G,#, There arethree individual loops b= - CGH, Step 3: Find the combination of non touching loops and their gain 1. Combination two non touching loops are LL, = G,G,G,H, H, ‘There is other combination of two or more non touching loops Step 4 : Find the value of determinant 4 4 = Determinant ofthe graph = 1 (bum of individual loop gain) + (sum of gain products of all combinations of two non touching loops) ~ (sum of gain products of «ik Combinations of three non touching loops) + Block Diagrams and Signal Flow Graphs —_121 a+r Ll +L A= 1+ GH,+ G,G,G,G,H,+ 6.GH,+ GCGHH) Step 5 : Find the value of A, A, = Value of by eliminating all loop gains and associated products Which touching t0 the E* forward path, ‘Since from the SF, itis seen that all the loops are touching all the forward paths, we have a=1 a, = 1 L214 GH, Thus, From Mason's Gain Formula, 1y Fa + Bs Gain a kee i GOGG.+6,0(1+ 6,H1,) T= TEGH+C6.6,G.H,+ 6,64, ie, Overall transfer function,122 —Controt Engineering REVIEW QUESTIONS 1. Determine the transfer function C(s)R(s) for the system given in the figure using block diagram reduction techniques. mS) © »[e, }-— G > cw) 7 2. Obtain the signal flow graph ofthe System whose block diagram is given in the figure and hence ‘determine the transfer function using Mason's gain formula, Rs) of, G G > CW) Hi, 3. Obiain the signal flow graph ofthe system whose block diagram is shown in figure. Also determine the transfer function using Mason's gain formula, RG) * Os) 4, Obtain the total gain of the SFG shown in Block Diagrams and Signal Flow Graphs —_ 129 5, Use block diagram reluson methods to obtain he ove taster fnction OR er diagram shown in figure, “om CR For the losk >C 6. For the system shown in figure use block diagram reduction methods to obtain the equivalen: transfer function and check the result using Mason's gain formula, ee > c H,124 —Control Engineesing 7. Use Mason’ rule to find the transfer function of the following system. Hy ' OE a > H, 8, Using Mason's gain formula, find the transfer function of the system whose signal flow graph is shown in figure. 9. Simply the block diagram shown in the figure and obtain the transfer function C(S)R) my RQ) ofa cs) i, 10, Determine the overall transfer function of the system whose signal flow graph is given in figure. Block Diagrams and Signal Flow Graphs — 125 11, Reduce the block diagram shown in figure 12, Obtain the taster function ©) of the system shown in igure x 5) (3) Xe Xo) 13, Reduce the block diagram in igure to its simplest possible form and hence obtain the closed Toop transfer function. 7 iA G > Cs) i, RQ)= Control Engineering Using block diagram reduction techiques, obtain the transfer function C(s)R() forthe system wo in the figure. >= OR {R}--EH- loop | A, — wo A, Obtain the system gain in SFG shown in figure. For the block diagram given in figure find overall transfer function ofthe system. #, |e RO), G, 3 + O15) + ‘Verify the above using signal flow graph analysis. 17. Determine C/R using Mason's gain formula, Block Diagrams and Signal low Grape — 12711 INTRODUCTION most conto ystems, ine fused as an independent variable, so contol systems are ihren ite domain system, Therefore, response analysis ofa congo system has become an important tsorforte design and aaa of onto sytem, Response analysis means stay of tse ‘put behaiourasafunetionof ime, when subjected © known input. This also known sim “poe Tr oaprhhnar hres ie miei int cig atsfatry perfomance, The time response analysis of conl ste helps in evaluating te ‘lio be yee ssn of hesystem. Treforest coal seme complet “nal evaluation of the performance of the system is based on the time response responding reals. es neton Defintion : The time response of a control system is the output of the system as a func time, when subjected to known input 4.2 PARTS OF TIME RESPONSE “The ime response ofa control system is usually consist of two pasts: The Transient response and the Steady-State response 1, Transient Response Definition : Transient response is defined as the part of diesout or goes to Zero over a period of time or as time of time, It is denoted by y,(0. Mathematically itis given by 4 the total time response which decays oF ‘becomes very large or after large interval lim yi = 0 Transiont and Steady State Responce Analysis — 129 2. Steady State Response Definition : The steady state response is defined as the part of the total time response which xemains after the transient response has diesou, It is denoted by y,, Thus, the total time response (9), of a control system can written as VO = (04 ¥, In general, all the control systems exhibit transient behaviour to some extent before a steady state is attained because of energy storing elements such as, spring mass etc, which are always part of ‘acontrol system and cannot be avoided. Due to these energy storing elements, the response of the control system cannot follow sudden changes in the input instantaneously and transients are observed, ‘Thus, transient response analysis is necessarily important, a it plays significant role inthe dynamic behaviour of the system and infact, deviation of output response and input must be control before steady state is reached, ‘The steady-state response of a control system is also very important as it indicates final output of the system. This points towards the accuracy of the system. In general, if the steady state response of the system does not agree with the desired reference exactly, then, the system is sAid to be in steady state error. Thus, the difference in the desired output and the actual output of the system is called steady state error and is denoted by ¢,,. All these definition can be shown in the wave form as shown in figure 4.1 0) Desired time () jp 1 steady state Response time Tansien’ Response ume SY see Fig. 4.1 : Response of the system and steady state error 4.3 STANDARD TEST INPUTS In evaluating performance of a system, itis necessary to assume some basic standard test input signals. By assuming these standard test signals, not only mathematical analysis of the system is made easy but also these input signals response allows to prediet the system's performance to ‘other more complex inputs. So forthe ease ofthe time-domain analysis, following standard test signals are used.4120_—Contral Engineering Step input ‘A Step iat sa andar est np whos value changes from oe postion 1 ano ero tne tis also non a pain inp This Step input represents an instantaneous change inteinpatasshownin fue 42 Mathematically, Step input i represented as eae HO =A forr20 =0 forr<0 ei where A is real constant If A= |, then itis known as unit Step function and denoted by 1) = Au = ro = Wl ‘The Laplace transform of Step input is Fig, 4.2 + Step Input A RO) = 2. Ramp Input ‘A ramp input is a standard input signal which changes with time linearly. It is also known as Yelocity input. Graphical epresentation is as shown infigure 4.3. ‘Mathematically, Ramp input is represented as ro) = At fort20 0 ferr forrz0 on =0 — forrs0 where A is areal constant ' ‘The Laplace transform of Parabolic input is Fig, 44 : Parabolic Input “Transient and Steady State Response Analysis — 131 A a) = IFA 1, then the Parabolic signal is known as unit Parabolic signal. 4. Impulse Input A impulse input is a standard input signal whose value is zero everywhere except 1 = 0. Ibis ‘generally called 8-function. ae ea Mathematically impulse inputs given by ) 1) = ABE for t=0 7 =O forreo Graphically itis as shown in figure 45. (If area of the impulse is unit), A =1 itis known as unit impulse input. Note that area of the impulse in nothing but magnitude of impalse signal, ‘The Laplace transform of impulse signal is ' ‘gf LIAB] = RO)=A Fig. 45 : Impulse Input Jn general, the mathematica equation representing standard test inputs are piven’ Tb 1 Table 4.1 Inputs ro Re) Seep nA 130 4 20 <0 é Ramp nO=At 120 a =0 to Parabolic | r= 4 139 a P 2 e & > 120 ff i Impulse 120) 1 sf, © n bese q gS, 44 ORDER AND TYPE OF THE CONTROL: ‘SYSTEM epee F441 Onoen oF tHe System x Sg. ‘$F Consider a feedback control system as shown in figure 46,192 —Gontol Engineering Rs) Gs) > CG) Hs) Fig. 4.6 : Feed back control system “The closed loop transfer function is given by o_o) RG) ~ T+G()AG ‘The denominator 1 + GXeiA\s) is known as the characteristic function, When characteristic function in equated to zero is known as characteristic equation i.., 1 + G(s)H{() = 0. ‘The characteristic equation, 1 + G(s)H(G) = 0 decides the poles of the transfer function. The degree of the characteristic equation is known as the order of the control system. Definition : The highest power of ‘s' present in the characteristic equation of a closed loop transfer function of a linear control system is known as order of a system. 4.4.2 Tvoes oF THe SysTeM The open loop transfer for the system shown in figure 4.6 s obtained by breaking the feedback loop atthe summing point i. pen lop taser ncn, $2) «eos Rs) “The numberof poles of the open loop transfer function decides the type of the system. In general, i'n’ is the number of poles atthe origin then the system is TYPE ‘n’ system. “Let open foop transfer function can be expressed as A(T,s+1XT5+1) OWE = FereeIKieH) ‘This is known Time constant form. ‘where & = constant umber of poles at the origin "The value of min the denominator polynomial decides the type of the system. It n=0, TYPE - 0- system 1, TYPE - 1 - system =2, TYPE - 2- system “Transient and Stoady Stato Response Analysis —_199 Definition : The TYPE of a system is defined as the number of potes at the origin or number of integrators present in she system, Thussorderis the prope ofthe closed TF — Gl) Te Gaye) Mlle TYPE isthe propery of pen oop TF G(s) His). For example, consider a control system as shown in figure 4.7. 4 3k Gs) | (® Fi0s+39) > C(s) ‘The open loop TF is G(s) His) * 3100035) ‘Ter are two poles at he origin (5 ‘Thus, the type of the system is TYPE - 2 system. 3 Theclosed loop tris 4) _- _ O19) F¥10s+ 39 RG) ~ 1+G(s)H(6) 3k * 3+ 100438) ) as RG) ~ Pe TOs 430s" + 3e Sf +105 4-39¢ 43k = O isa characteristic equation ‘Therefore, highest ofthe characteristic equation is 5. Thus, order of the system is 5,134 —Contol Engineering 4.5 STEADY STATE ERROR AND ERROR CONSTANTS: “The steady state error e,, isthe difference between the input or desired value and the output oF actual of a closed loop system for a known input as response tends to infinity, The steady state error is an important aspect of system behaviour by which system accuracy can be measured, Definition : Steady state error is the difference between the input value (destred value) and the output (actual response) of the control system. Expression for error constant and steady state error: Consider a closed loop linear contro! system as shown in figure 4.8. Fis), Gs) co) HG) Fig. 48 : Feedback control system From figure 4.8 Exvor signal, (3) = RQ) ~ BUS) Bu, Feedback signal, BG) = Cs) His) (6) G(s) Hs) £ E(9) = R()~ Els) G(s) HIS) ie, Eis) + E19) G(s) HU) = RO) Als) + + GS) HON = RG) ‘Thus, for non unity feedback [Ty CG) = £9) GO) a) * FO) = Treo] Dutforuniy feedback HUG) = 1 + Error signal) = 2 But steady state error e(?) corresponds 0 time domain and the condition forthe steady state is > + Steady sate exon, ¢, = HR ep Using Final value theorem i.e. im (¢ = lim sF(s) Transtont and Steady State Response Analysis —_195 ¢, = time(s = lim sets) a tig, sO i , = Mt STG Gy forum festback stem im 2A) eu = HS Ty Gif Mon unity Feedback aytem, 4.6 EFFECT OF STANDARD TEST INPUTS ON STEADY STATE ERRORS OR ERROR CONSTANTS 4.6.1 Un Srep input auio Posimionat. Ennons ‘The steady state error of the System for unit step input is called posit For unit step input, =u) (Cr A=1) R65) _ tim “ 66) ~ og ale ~ Trlime(s) where 4, = limG(s) is known asthe positional error constant 4.6.2 Uner Raup Input ano Vetociry Enron ‘The steady state error ofthe system for unit ramp input is called velocity error. For unitramp input, nse 1 RG) = But steady state error in Laplace domain for unity feedback system is given by136 —Contol Engineering > Bae *TigsoQ) * where Ji sG() is known as velocity error constant. 418.3 Unrr Paragouc INPUT AND ACCELERATION The steady sate error of the system for unit parabolic input is called acceleration error. 2 z Forunitparsbolicinput,§—r®) = RG) But steady state error in Laplace domain for unity feedback system is given by 5R(3) ME T+ G(s) = 20 Tol) = tim 8 F865 where ,= lit) Gi) is known as acceleration error constant, So, static error constants forthe different standard input are given in Table 4.2. Table 42 ‘Type of Input | Error Constants unit Step Jim Gis) unitRamp c= ti G6) unit Parabolic = lim ea(s) “Transient and Steady State Response Anahysis — 197 4,7 STEADY-STATE ERRORS OF TYPE - 0 UNITY FEEDBACK SYSTEM For the TYPE - 0 unity feedback system, the open loop transfer function i given by K+ T(t) OO = TM HS) The steady state error for unity feedback system is given by 2, = tims 8) = TS GE) ‘The steady state error for standard inputs are Positional Error, (unit Step input) e(p) = lim —s Velocity Error, (unit Ramp input) e,) = 4.8 STEADY-STATE ERRORS OF TYPE 1 UNIT FEED BACK SYSTEM Forthe TYPE- unity feedback system, the open loop transfer function is given by Ale Tes B9) GO) = “eT Yl Ta)138 _—Contol Engineering ‘The steady sate error for unity feedback sytem is given by = tim #80) oF AT Gl ‘The steady state error for standard inputs are Positional Error, (unit Step input) ©) = aT EE 1 Tia) Velocity Error, (unit Ramp input) é,fo) = lim = Tra) L erty * Tims ~ ¥, ~ Acceleration Error, (unit Parabolic input) e,(a) = Tima ioe) -— inrats 4.9 STEADY-STATE ERRORS OF TYPE 2 UNITY FEED BACK SYSTEM For the TYPE-2 unity feedback system, the open loop iransfer function is given by (1+ Za (1+ G3). FOTN Ts) ‘The steady state error for unity feedback system is given by Gs) = “Transiont and Steady State Response Analys The steady-state errors for standard inputs are Positional Error, (unit step input) 0) = lim ree) 1 * TaimG() ‘Velocity Evror, (unit Ramp input) 2,0) = So steatly state errors for different types of system when subjected to standard test inputs ate given Table 4.3 Table 43 Steady ~ state errors TYPE of Input TESS TPES THT Unit Step 0 ° Unit Ramp » ° Unit Parabolic ~ 740_—Contvoi Engineering WORKED EXAMPLES 1 the type and order ofthe following system for which open loop transfer function are given as follows x 1 GOH = Teas 105)(1 20 100(5~1) ) GloMe) = Seesyera) Solution : Case i : Consider open loop TF. (OLTF) Goomey = ear Time content form of OLTFis given by R(t #I) ooo (si E51) Comparing bth equation, we have n= 1 «system i TYPE 1 system ‘Also, the characteristic equation of system of the system is given by 1+ Gis) His) = 0 GHG) = Ans, Tsar 1030+ 209) 200 423089 +31 8+5+K = 0 . Since highest power of inthe characteristic equation is 4, The oder of the system i 4. Consider the OLT Case 5 1oo(s1) G(s)H(s) Sir er 6) [Express in time constant form, ie, ( fi 109 __(s-1)__ GOMES) = “39 F(O2s+ NO1e7s+1) Ans. “Transient and Steady Slate Response Analysis 141 3330-1) F(O2sNOI67s+1) Comparing with standard Time constant form, we have n = 2, System is TYPE -2 system Ans, Also, The characteristic equation of the system of the system is given by 1+ Gio) = 0 100451) les PT) F(t SIe+8) 118+ 1305-100 = 0 Since highest power of sin the CEis 4, :. Order of the system is 4 ee 4.10 TRANSIENT RESPONSE : FIRST ORDER SYSTEM. 4.40.1 Unir Srep tnpur Consider the first order linear unity feedback control system whose block diagram is shown in figure 4.9, RO 420 1 ow) Fig. 49 + Feedback System ‘The transfer function of the system is given by Gs) R(s) &) Rs) ~ 147s Thus, the output response C(s) is given by142. —Contral Enginosring 1 For unit step input, R(s) = Using partial fracto Tr ars ‘Taking inverse - Laplave transform, we have Output response, C(i) = (1-e"), for 20 a Error Response : It is given by eh =~ CO (ee) Steady state extor: I is given by lime(e) = lime” 2 eT -@ ‘The equation (1) states that the initially output response, C(x) is zero and it becomes unity finally. increases exponentially as shown in igure ‘That is, the output response C(O starts at zero for 4.10 as # increases Transiont and Steady State Response Analysis — 143 «For t= T, Ctt) = Ie"! = 0.632 = 63.2% of the final value, ‘The value ‘Tis known as time constant of the system. The time constant indicates how fast the system reaches the final value Definition : The time taken for the step response to reach 63.2% of the final value is known «8 ime constant ofthe systom % error T OTST aT Fig. 4.10 : Unit step response of first order system 4.10.2 Uniruputse input For Impulse input, Rs) = 1 \ 6) = 2 in ‘Taking inverse-Laplace Transform, we have, co cw = Le" xi This is the expression for unit impulse response for ‘he First order system, This states that as¢= 0, output 4 response C(t) = 7 and decreases as input (1) increases ic., CU) =0 as ¢= oo. This response of the ' unit impulse input is known as weighing function Fig, 411 : Unit impulse and is as shown in figure 4.11, response of first order system ‘The equation (3), states thatthe steady error reaches to zero as time “tends to infinity144 —Contiol Engineering 4.10.3. Unsr Rane Input 1 For Ramp input, RS) = as) A Using partial fraction = Solving we get, ‘king inverse - Laplace transform, we have Oo) =1-T+ Tet -@ Equation (4) represents the response for unit Ramp input of the First order system, This equation (4) states that for ¢ = 0, output response (0 and increases linearly ast > © ie. (C(0) == as shown in figure 4.12. ow Fig. 4.12 :Unit Ramp response of First order system 4.11 TRANSIENT RESPONSE : SECOND ORDER SYSTEM Consider the second order linear unity feedback control system whose block diagram is as shown infigure 4.13, Transient and Steady State Response Analysis — 145 es) Fig. 4.13 + Feedback System ‘The transfer function of the system is given by k cy) __ as) a7) k R(s) ~ TeO(s)i@) ~ 1, Fd T)ok +73) PLE aa r ‘ ak 54S. The denominator, s*-+1.5+ is simitar to 6+&6+4% (Equation of motion for the dary pte 5+ Fo+% i ped torsional vibratory system) where 2 0, and Equation (A) can be written in following form as ls) a Rs) ~ 3 42€0,s+02 ~@) Where & is the known as the damping factor (or) damping ratio which can be defined as the ratio ‘of damping coefficient to critical damping coefficient. Damping Coefficient Critical damping Coefficient ~ * ge ‘Comparing equation (A) and (B)= and ©} 2ko, = 7 and oF WF “Thus, equation (B) is known as the standard form of the second order system. Where, denominator polynomial, s*+ 2Eo, ¢+ 02 =0 is known as the characteristic equation ‘The roots ofthe characteristic equation infact are the poles ofthe transfer function, The nature of roots ofthis equation characterises the time response of any system, “Thus, equation (B) can be represented as shown in igre 4.14, RG) +, Bo) ot ce Fig. 4.14 Als, the response C(s) ofa second order system depends on the value of & Tha is Case I: when & < 1, system is called under damped system Case 2: when &= 1, system i called critically damped system Case3 : when & > I, system is called over damped system Case 4: when = 0, system is called undamped system. Case 1 : Under damped system with unit step input (For & < 1) Let unit step input is applied to the second order system, ; 1 ie, RG) = while, second order system is represented by a) _ ot RG) ~ Feiiosro 1 o P+ tkas+oF as) By applying partial fraction expansion Transient and Steady Stato Fes Bs+D FF2b0,.60 we get, A s+ 20, s 3 +28o,s+05 Adding and subiracting 2 to the second term of the denomizator, we get as) = 2) il (s4260,) 5 (s+€0,)' +03(1-€) _ 1 (s42%0,) © # (s#60,) +0; where ©, = ©, f1-E? incalled damped natural frequency. Rearranging, we get S460, go, 1 OO =F Grto,F eal Gebo,) rol Taking inverse Laplace transforms ow S¥ sina gt Since @,48 —Control Engineering ew! i-8 {eos F)+bsino,] -etcos @= Gand sin @= /1—E* and rearranging we get, Cte) = 1-2 sine ,te0s 0+ cos o,tsind] Ae under damped system ‘where @ = tan’ in under damped system subjected to isthe exression forte ine eons a second order under damp. rn ep ipa canbe sen rm tit expression tet the response clon Bu ma persistent as shown in figure 4.14(a) a Fig, 4.14(a) : Time response of a 2 order under damped system 4.12 TIME RESPONSE SPECIFICATION ‘The time response specification are the performance characteratis of a control system. Ths is common thatthe performance characteristics are speifiedinterms of he transient response & iit step-input because ifthe response to aunit step input is known, itis mathematically ps determine the response to any type of input ‘The commonly specified response specification are vane 1. Delay time (1) It is the time required for the response (output) to reach 50% of its final val inthe very first time. Transient and Steady State Response Analysis — 149 2. Rise time (¢) : Its the time required forthe response (output) to reach 100% of its final value inthe very first time. 43. Peak time (t,) Tis the time required for the response (output) to reach first peak of the over shoot. 4, Maximum overshoot (¥,): Its the maximum peak value of the response curved measuted from unit tis detined intessts of maximum percent overshoot as 100 This indicates the relative stabitity ofthe system. Thus w= oe 5. Setting time (¢) : It is the time required for the response (output) to reach and stay within a range of #2% to 5% specified by tolerance band of is final value. 4.42.1 Expression ron Peak Tine (r,) ‘The transcat response of a second order under damped system when subjected to unit step input isgivenby Hor --<—*sinfo +8) irs cw ac(e)} Tooblain peak ime, | fe50 —Contrl Engineering “Transient and Steady State Response Anaisie —_151 4.12.3 Expression Fon Maximus Overstoor (M) ‘The transient response of second order underdamped system when subjected to unit step input is given by co For maximum overshoot, orm ——sin(a,t, +8)=1 5 ©,” @,f1-8 vi-§ ‘his i the expression for peak time, 1, 112.2. Expaession ron Rise Time (1,) ‘he transient response of second order underdamped system when subjected to unit step input is sivenby But, For Rise time Cl. 1 sin(o4t, +0) ( ) =0 From figure sind But sin (n+ 8) = since om 20 aaa sin (0J,+® = 0 1 nm forn=0,1,2 O48 for Aso, eM, = This the expression for maximum overshoot.52_—Control Engineering 12.4 Expression ror Serna Tate (t,) “he transient tesponse of second order under damped system when subjected to unit step input is ivenby and other transient oscillatory “his comprises of two parts: one is exponentially decaying ie. x sin (0+ 0) which vanishes completely. Thus seting ime is obtained by considering only ‘xponcntial term, neglecting all other ters. rom the definition of setling time, we have For tolerance band of 2%, at ¢ ie for lower value of & ee = 0.02 = Ee, = In (0.02) aon * bo, . 4 car fo, Chus, setting 1 where Tis exponential time constant = =~ inere Tis exp bo, Similarly, For 5% tolerance band, =r = bo, Settling time, This is expression for setting time, ‘Transient and Stoady Stato Response Analysis 153 Example : 2 A unity feedback system is characterised by an open loop transfer function Gs) Determine the following, when the system is subjected to a unit step input 1. undamped natural frequently 2. Damping r: 3. peak overshoot 4. peak 5. settling time Solution Dec.09/Jan.10 (12 M) - The characteristic equation is of the system is 14 Gis) Hi) = 0 His) ie Pett 16 = comparing with + 2E@, s+ a =0 1. Undamped naturel frequency @, = 16 + @,=4rad/sec Ans 2. Damping rat, _ 2s Ans Mes 3, Peak over shoo, % 2 MOON ouass Ang 4, Peak ti t= ey = me = 0815 see a : me * ali-oae 44 5, Settling time, see (2% tolerance band) Ans Ew, ~ 025x4154 —Control Engineering Example : Figure 4.15(a) shows a mechanical vibratory system. When 1 KN force is applied to the system, the mass oscillates as shown. Determine the mass M, damping coefficient B and stiffness k of the system from this response curve. Dec.06/Jan.07 (10 M) 0 ab t > 1 (sec) v 24 Fig. 4.15(a) Fig. 4.1510) Solution : Data : F(® = 1000 N ; x, = 40 mm ; ¢, = 2.4, Max, peak = 3.8 mm The differential equation of motion forthe spring-mass-damped system subjected to external force isgivenby ni Bes =f ie, mE none of ting Lap aso, Ms? X(s) + Bs X(s) + & X(s) = Fis) X(6) (Ms? + Bs + K) = Fls) —() «Theta ton i gienby x6) 1 tt -@ Fs)” Wes Bewk ‘2 “The standard TE of 2* order system is given by oeeO eee F32o,s+0 Transient and Steady Siato Responee Analys's —_185 ‘comparing denominator, wehave of radsec -B 20, CO) Since, FW) = Thus Xs) From figure, for Steady - State value of x(0 is X, = 40 Jima sx(6) = lim | ws Bu, Sto ww) 0 = 100 ik =25Ninm 25000 Nim Ans Maximum overshoot, 095° fom figure 4.15(b) max, peak = 38) ie, 0.095 Taking Natural logarithm on both sides, we get EE = in 0.085) =~ 2.354 he56 _—Contro! Engineering Squaring on both sides, we get Tay = 0561 & = 0561-0561 & 1.561 = 0561 (0561 & = (Ee =0599~06 Toke & = 06 ‘Ne know that, Peak time, 1,24 (given) o, ; 2avi-o6 1,636 raaisec %xom equation (3), je, 1.636 = yy vu Mass, M = 9340.5 ke Ans. 4 ne Zromequation (4, 8: m7 __ 8 06 = 3 f5000% Damping coefficient, B= 51987976 Nsectm Ans. Transient and Steady State Response Analysis —_ 157 Example : 4 ‘The open \oop transfer function of a unity negative feedback control system is ‘maximum overshoot, peak time, rise time, and settling time, Jan/Feb 2005 (05 M) ‘The characteristic equation of the system is given by 1+.) His) = 0 Baro aH ase) ** = Ks) = . #455425 =0 comparing with, $°+ Em, s+ 02 = 0 vee of =2 = a, = Sradbce f ss and 2, = ee —- 5 50, 7 7x5 75 1 Maximum over shoot, Pi 67657 016303 Ans. % M, = 16303% 2. Peak time, 7 = =0.755 sec Ans. a,ji-8 sVi-03* 4 3. Sealing ime, = yoy —_For2broteance band Baxg 716 00 Ans. 4, Rise time, oeas158_—Contol Engineering Example : 5 Unity feedback control system is characterised by GINS) = the system is subjected to a step of IV. Sotutlon The characteristic equation ofthe system is piven by 1+G()M) = 0 k Ho S410sek=0 0 Comparing with s' +280, 5+ 02 = 0 we have, obek = 0, and 2o, = 10 3 @, ©, = 10 sadisex s. System, gain k = 100 Form - step ; Rise time, where 6 = For step of IV, n= 4 (given) Forn- step ; peak time a s(s+10) Determine the system gain k, so that the system will have a damping ratio of 0.5. For this, value of k, find the rise time, peak time, settling time and peak over shoot. Assume that Dec.07/Jan.08 (08M) pen loop transfer function =0.5 given) sient and Stoady State Response Analysis — 159 For step IV, n= 4 (given) = 143 secs Ans pat nese, = <768 e Has = 0.16303 Ans Settling ti 1, . For 2% tol band = 8 05x10 eae Example + 6 Unity feedback control system is characterised by an open-loop transfer function G(s) k MG) = egg): Petermine the system ga f 9 thatthe system will have a damping ratio of €5, For this value of k, find the rise time, peak time, setting time and peak over shoot fora unit step input Dec'tt (8 M) July 2006 (10 1) Solution ‘The characteristic equation ofthe system is given by 1+ G(o) Hts) = Eke +10) #41004 =0 comporingwith, +20, s+ 02 = 0 o 2k, = 10 > @, rad ns, ko, = 10 = 5g = 10 radser Ai k = 10=100 Ans, Peak time, hee sy = 0:3627 sec Ans. * oi-@ 1ovi-0F Seating time, t160_—Control Engineering 4 05x10 Beak overshoot, Mi, = ME a Moat % M, = 16303 % Ans = 08 sec Ans (2% tolerance band) = 0.16303 Example : 7 k An unity feedback system has an open loop transfer Tunetion G18) = Say 1. By what factor the amplifier gain k should be mul from 0.2 t0 0.8 2. By what factor the time constant T should be multiplied so that the damping ratio is, reduced from 0.6 to 0.3, Solution = The characteristic equation of the system is given by 1+ G69) His) = 0 Since H(s) = 1, we have so that damping ratioisinereased k not. ser ~° ST+sek=0 Lik Srterk oo Comparing with s? +2E0,5-+003, we have qe esp > u Ao and to 5 = SF 1.For&=£,=02 letk=k,and For§=6,=08 let k=, and T= constant 102 = Tipe and O8 = ee 1 02 War 1 os * 2ur > 4 vk A Squaring on bot sides, we get Je Sunn 8 ‘Transient and Steady State Response Analysis 164 ‘Thos tinwase damping ratio from 0.21008, mpfr sin X shou be muted by 2. Let &= 6, =06, '=T, and & = &, = 03, Take k = constant 0s ae 1 ayeR m403= 96 03 squaring on both side, we get Thus, to reduce damping ratio from 0.6 to 03, time constant T'should be multiplied by 4 Example : ‘The response of a system subjected to a unit step input is, CW = 1402 6-12 e-™ Obtain the expression for the closed loop transfer function. Also determine the undamped natural frequency and damping ratio of the system, Solution : For unit step input, ROS But, response is Cl) = 1402 e®—12 e! Taking Laplace transform, we get 2 +10 __ (6+ 60)(s+10) +024(s-+10) ~129{5-+ 60) “(5+ 60)(s+10) cs)162 —Contral Engineering $F +.605+ 10+ 600+ 02; - 5 +605" +105 , 600 CO = Fee 705+600) But coed op vans fc gvenby $y since, 9) = + forunitstep inp, we have 0 seater 7 < C00) ane Comparing 9th, we have Fikko,sr0 0} = 60-0, = 24.4948 radlsce Ans -. to, = 10 b= aia = 1.428 Ans Thus, system is over damped as & > 1 Example : 9 For the system shown in figure 4.16. Determine the 1. damping ratio and undamped natural frequency in the absence of derivative feedback @=0) 2, constant ‘a? which will increase the damping to 0.7 3. overshoot in above two eases 8 we RO) 4, > Cs) Fig. 4.16 ‘Trancignt and Steady State Response Analysis — 163 Sotuiion Case 1 : when «=, the block diagrisn reduces (0 RO) + 8 +2) oo 8 O19) = Fee amd HO) ‘The characteristic equation is given by 1+ Gs) His) = 0 8 ie, 1+ x1 20 542) 842048 =0 Comparing wth, *+ 26, s+ a = 0, we have @! = 8 => @,=2.828 radisec Ans, 1 = 03535 Ans. 2828 ae Case 2: when ‘a’ is present, ‘There are two feesback paths which are parallel 8 O) = Gay and M=1 tas The characteristic equation is given by 1+ GOH) = 0 8 5 (1+as) qeentt™) S420484 Bar = 0 Pear 1+ Comparing with s? +260, s+ oF = 0, we have164 —Contol Engineesng a 2o, = 2480 but 240.7 2828 = 2+ 80 a= 02452 Ans. 7 given Case 3 : Overshoot, For & = 0.3535 From Case 1) M, = 030507 Ans. % M, = 30507 % For & = 0.7 (Case 2 given) M, = 00459 Ans. % M, = 459% Example : 10 ‘The block diagram of a simple serve mechanism is shown in figure 4.17. Determine the values of & and T to give an over shoot of 15% and a peak time of 3 sec for a unit step input. Also find the settling for 2% and 5% tolerance band ping “eae oe Sotutio From figure 4.17 +. The characteristic equation is given by 1+ G(s) His) = 0 ies hd eT = 0 PB +k=0 comparing with # +28 +0! = 0, wehave = -@ ‘Transient and Steady Stas Response Analysis —_ 165 For overshoot of 15% ie., M,=0.15 (given) Bu, iM, = @F ons = Me ‘Taking natural logarithm on both sides, we have In (0.15) = ving 15169 For peak time of 3 sec, ie.,1,=3 (given) On solving, we get, b= 7 Bu for nts input ieee Splapus 1, 1-05169" 1.223 rad/see ‘Thus, from (1) From 2) is Settling time, 1, Also, = augi@extaas 747455 see Ans166 —Control Engineering Seample 211 120 and His) = ©, obtain the Zora simple feedback control system with G6) = -—yg3) AM HO G sharacteristic equation ofthe system, Determine its undamped natural frequency, damping catio and maximum over shoot. Solution: The characteristic equation is given by 1+ Gls) Hs) = 0 120 fe) 6 1+ o ie, (8435422) =0 2a, s+ 02 = 0, we have ‘Comparing with, of = 22 = 0,= VB = 4.600 ndlsec e,23 = be Maximum overshoot, = 034632 Ans. 34.632 % ‘Transion and Steady State Response Analysis 167 4.13 SYSTEM- STABILITY For a control system designer main concer is whether designed system is stable or unstable, Because, stability is a very important characteristic of any control system. Among the many forms of performance specification used in design, the most important requirementis the system stability Almost every working system is designed tobe stable. An unstable system is generally considered. tobe useless. Thus, study of whether the system is stable or unstable is known as stability analysis, Definition + A system is said to stable if its output is bounded for any bounded input 4.13.1 Concept oF Sraaiuiry 1¢can be illustrated by considering circular object (ball) resting an different types of surfaces under the presence of gravitational force, as shown in figure 4.18 \ Ammer Stable Unstable Fig. 4.18 For analysis and design purposes, stability of the system can be classified as 1. Absolute stability Absolut stability refers to the condition of whether the system is stable or unstable it isa YES or NO answer. 2, Relative stability Relative stability i the quantitative measure of how stable isthe system and this is the degree of stability, Marginally stable 4.13.2 Tens Useo 1. Stable system : A system is said (o be stable if the output is bounded for any bounded input, ‘Thus A systems said o be in equilibrium ifthe output eventually comes back to its equilibrium state ‘when the system is subjected to a disturbance. 2. Unstable system : A system is said to be unstable ifthe output is unbounded for a bounded input. Itmeans with zero input the output may increases indefinitely 3. Absolute stable : A system is said to be absolutely stable with respect to a parameters of the system if it is stable forall values of these parameters 4, Conditionally stable : A system is said to be conditionally stable with respect to a parameters of the system if itis table only for a certain bounded range of values of these parameters. 5. Marginally stable [critically stable] : A system is said to be marginally stable ithe system is {just on the verge of becoming unstable. 6. Relative stability : It is the quantitative measure of te degree of stability, determines how stable isthe system and how close it o being unstable.168_—Contol Engineering 4.13.3 Conomion ror Stasiuiry ANaLYsis The stability of the system depends on the 1008 ofthe characteristic equation of the system. Nature of roots Location of roots 1. Real, negative ice, All the roots are in : lefthalf ofthe 5-plane 2. Complex conjugate with = negative real parts ice, All the roots are in eft half ofthe s-plane i ‘Stable 3, Real, postive ive, ifany one rootin the right half ofs-plane Unstable 4, Complex conjugate with positive real part Unstable 5. Repeated pair roots on imaginary axis Unstable 6. Non repeated pair roots on imaginary axis 6 ‘Marginally stable 4.13.4 Cnanactenisnc Eavanion ‘The characteristic equation ofa contol systems obtained by setting the denominator of the transfer function ofa closed loop system equal to zer. Itis defined by the following equation 14+ GMS) = 0 where G(s) is the overall feed forward path transfer function. LHA(s) is the over all feedback path transfer function, Teansiont and Steady State Response Anaiys's — 108 Note : H(s)= 1 for unity feedback system, Example : Ifa system having transfer function A) 542 - Ris) ~ fe +2842)(e? +7541) ‘The characteristic equation is given by SG? +2642) (2475412) = 0 4.14 ROUTH'S HURWITZ CRITERION Routh’s Hurwitz criterion is an algebraic method of determining the stability of the linear time invariant system. This criterion provides information about whether systems stable or not depending ‘upon the position of the roots of the characteristic equation whether they lies in the leftchalf of = plane or right haf ofthe s-plane. Apart from ascertaining stability ofthe contro system, this method ceanalso be used determine the limiting values of a variable parameter within which system is stable and beyond which system would became unstable. ‘The characteristic equation of system can be written as ays 4a, st4 a, s+ taysta, Where a, dy dy..d, are the coefficients typically positive real numbers. ‘The R-H (Routh’s Hurwitz) criterion is based on the arranging the coefficient ofthe characteristic equation in the form an array known as Routh’s Array. 4.14.1 Roum's Array ‘The method of arranging the coefficients of the characteristic equation into an array is called Routh’s Array. The Routh’s Array is formed as follows = ‘Consider the characteristic equation of system as asta sea stat taste ed Where ay, ayy dyn, are the coefficients Step I : Arrange al the coefficients of the characteristic equation in two rows as follows Rowl ay a ay Rowds? a, a, as Step II : Form 3" row from Row 1 and Row 2 Rowl sla, a, a4 4% Row2 sa, ay as ra a170 _—Control Engineering 2-8 od s000 tes 55 where b= [= This process is continued till we get a zero as coefficient in3* row ‘Step IIL : Form 4 row from row 2 and row 3 Row s\a a ay a Rowd sla, a, as a Rows s71b, bh by by Rows Sq GS and so on Step IV : Consider the characteristic equation of a control system as ays! 44,8 bays" +038? 40,5? 40s! Hay Solution: Routh’s array va, a; a 0 sh & 0 eee sla, d, 0 0 sle 0 0 0 sl, 0 0 0 Koy points Rout’s array is also known as Routh’s stability criterion or Routh - Hurwite’s method (RH criterion). Thus, R-H criterion states that : The necessary and sufficient condition for system stability is that all the terms inthe first column of Routh’ array must have same sign. If these are any sign changes exist then, the system is unstable and the number of sign changes in the elements of first column of Routh's array is equal to the number of roots of the characteristic equation in the right half of the s-plane Transient and Steady Slate Response Analysis — Example : 12 ‘The characteristic equation of the control system is s° + 9s? + 26s + 24 = 0, Using Routh - Hurnitz criterion ascertain its stability, Solution: ‘The Rous ary forthe given hacer equations Ppt % el 9m 1 (3) |(2) 0 elu o ‘To forms! row : , = 2X29)= 4K!) _ 70, 5 3 1 (Ries -x0 Toto trong BE 3 Since there are no sign changes in te first column of the Routh array and are positive, Hence the system is stable. 4.14.2 Speciat Cases Case I: First element in any row of the Routh’s array is zero while remaining row contains at least, ‘one non-zero element, Due to this first element of any row is zero, the elements in next row will all become infinite ‘Therefore Routh’s criterion to ascertaining system stability fails ‘The following methods can be used to overcome this difficulty, ‘Method 1 (For Special case 1) Replace the zero element by a small positive number € and then proceed to complete the Routh aay and examine the sign change by taking fim Example : 13 Examine the stability of the system having characteristic equation as S424 39 4684284120 4 Solution "The Routh’s array for the given characteristic equation is1 Enginearing ° 1 32 2 6 ? wl sO new < iso oe | be 1 oo € s o 0 s° 1 0 0 To examine sign change, consider 3 3 tin{6~2) = 6-tim? oe = —e (change in sign) (+ve to -ve) 146 2 lim —\ oT = +15 (Change in sign) (ve to +ve) AAs there are two sign changes, system is unstable, Method 2 (For special case) Modify the given characteristic equation by replacing s by 1 and rearrange the characteristic squation, Then form the Routh’s array from the new characteristic equation in x and examine Stability with this array. From the previous example, Consider the characteristic equation 42st 43846242641 = 0 Replace s by + (Method 2 - For special case 1) Teansiont and Steady Stato Response Analysis 173 SS SOP Bee OE ‘The Routh’s array for this new equation is rl i 62 alee oe x) 45 15 0 er) x'[-0429 0 0 Go oo Calculation 2X6) -EX0) ten tow; CREE) 3, 190-08) nen ‘Since from Routh's array. I is found that in the first column there are 2 sign changes Le., +ve to ~ ve and -ve to +ve, v. system is unstable. Case IT: All the elements in a row of @ Routh's array are 2270 Due to this, all elements ofa row in a Routh’s array are zero, the elements for the next row cannot be calculated. Therefore, Routh’s criterion to ascertain the system stability Fail, ‘The following method can be used to overcome this difficulty, Method 1 (For Special Case) 1. Form an Auxiliary Equation A(6), using the coefficients of a row just above the row of zeros 2. Take the derivative of an Auxiliary equation with respect ro sie, 2A(8) as 3. Replace the row of zeros with the coefficients a) 4. Complete the Routh’s array in the usual manne. 5. Examine the sign changes if any inthe first column of the Routh's array to ascertain stability,174 —Contiol Engineering Example : 14 Determine the range of & for stability of the system, as shown in figure 4.19 Re) + 6-2) > ot > +| (ais +65+25) aD Solution: From gvenbock diagram ‘The characteristic equation ofthe sytem is given by 1+ Gt) He) = 0 te Hs=2) (oF 65-25) P4784 E4315 05-2h = 0 Thus, Routh amy is given by f # |as-2 For stability L.From stow; 92499 5 p> 3 2. rom °row ; 25-2k>0 = 252k. ke 12S AS > 0, Thus range of for stablity 0<-< 12.5 Ans. Example #15 Ascertain the stability of the system given by the characteristic equation, F434 SAIS + BS + Os + Dec.09/Fan.10 (08 M) ‘Transient and Steady Stato Response Analysis — 175 Solution ‘The Routh’s Atray for the given characteristic equation is 8 6 4 0 Catetation For Blemens of ow BB)-O)0), 3 For Elements of ow 290)-(00) 3 ‘This is a special case where all the elements in a row are zero, In such cases, form Auxiliary equation considering s# row as ACs) = 2st + 6844 Differentiating with respect tos, we get MO sein . we — s|3 9 60 s]2 640 s/8 200 s[3 400 133 0 0 0 s4 000176 —Control Engineering Calculation For new Routh's Array (8X6)—(12)(2) For Elements of # row FacBeens otto A=) 35 (133)(4) (013) For Elements of s row ae Since there are no sign changes in the first column of the new Routh’s Array. Hence there are no poles in right - half ofthe s-plane Solving Auxiliary equation, we get 2st + 6e44 = 0 Roots are s = tiand s= (2 ‘These roots are also the roots ofthe characteristic equation and are non repeated with no positive real part and hence the system is marginally stable Example : 16 Check the stability of the system having characteristic equation PDS Bot + 128 +205? + 166 + 16 = 0 Solution : ‘The Routh's Array forthe given characteristic equation is 1 8 2 16 212 16 0 212 16 so 0 0 ‘This is a special case where all the elements in a row are zero. In such cases, form Auxiliary equation considering s* row as Als) = 2+ 12s +16 Differentiating with respect to s, we get als) ds 8+ 24s Array as 20-16 16 0 16 0 0 0 00 oo Since there are no sign changes in the first column of the new Routh’s Array. Hence there are no poles in right-half ofthe s-plane Solving Auxiliary equation we get 2 +12 +16 = 0 Let # = x We have 224 12x +16 Roots are x Thus,s = iY2 and 2 ‘These roots are also the roots ofthe original characteristic equation and are non-repeated with no positive real parts and hence the system is marginally stable, Example : 17 Using R-H criterion investigate the stability of the system sf + 2s + Is? + 185 + 18 = 0. Comment on the stability of the system. Also find the frequency of sustained oscillation. ‘Tune/July 08 (10M) Solution : The Routh's array for the given characteristic equation is s|l 8 0 18 0 0 ‘This isa special case whefe all the terms in row are zero,78 ‘ontrol Engineering ases, form the Auxiliary Equation considering row (ust above the zers element row)as Als) = 2s'+ 18 Differentiating with respect to s, we get mnsuch aa(s) “ds 4s Replace the zer0s of the s! row by coefficients of as :. The modified Routh’s Array is s{l 18 sj2 18 0 e218 0 s|4 0 sis Since there are no sign changes inthe first column ofthe new Routh’s Array. Hence these are no poles in right-Half ofthe s-plane Solving Auxiliary equation, we have 2418 =0 vRootare s = £3 ‘There roots are also the roots of the original characteristic equation and are non repeated with no -4ve real pasts and the system is marginally stable. The frequency of sustained oscillation are found as follows Puts = fointhe AG), jen 2Gay +18 = 0 20? +18, @ = £3 1ad/ sec ‘The frequency of sustained oscillation are + 3 rad/see Example : 18 Characteristic equation of a servo system is given by asta, +a,s4a,s+a,=0 s for stability, use Routh’s exiterion July 2006 (5M) ‘The Routh’s Array forthe given characteristic equation is given by Transient and Steady State Response Analysis — 179 sila a a, sla, a, 0 sb & sl 0 al, sid, is = a9 24 ~ (0) where b= Se, For stability, 1.a,>0 24,>0 3.b>0=>4,4,-4,,>0 4.620 => b,a,-b,a,>0 5.d,>0=b,>0 pa Example : 19 Find the range of & so that the system with following characteris SHS 4S 4 bse k=O Solution : ‘The Routh’s Array for the given characteristic equation is ic equation will be stable. Jan/Feb 2005 (10 M) aes) e{ 5 4 0 es] 42 & , | 198-56 “42” el ok Forsiabilty01 Engineering rom s* row k>0 : 168~Sk rom s! row ao 168 - 5k > 0 k < 336 5s, range of k for stability is 0 k<336 Ans. ‘imple : 20 applying Routh criterion, discuss the stability of the closed loop system as a function for the following open loop transfer function K(s+3) cia Je. 06/Fa.07 (0 ME Eas) eer a ) Hs) ston: characteristic equation ofa system is given by 1449 Hi) = 0 by Het) Ho ife +4516) #14304 125+ {k-16)+ k= Ois the characteristic equation s,Routhis Array is gvenby s 1 ieee s 3 -16) 0 s a) st 0 ° k Tran nt and Steady State Response Analy ere Forstability L From #° rows k> 0 (224) 2. From s row s{ 25" Jae 16) ~ 34> 0 (52-0 e~ 16) - 9k > 0 = + 68k ~ 832-9 <0 B59 +932 <0 +, Roote are > 23,3153 and k < 35.6846 3. From # row; S2-k>0 ke ‘Thus For 0< & < 23.3153 system is unstable For 23.3153 35.6846 system is unstable 23,3153 and k= 35.6846, the system is marginally stable, Example : 21 Investigate the stability of the system. Using Routh Hurwitz criteri characteristic equation SP Ast + 12 + 205% + 305 + 100 = 0 Solution : The Routh’s array for the given characteristic equation is For mn having the following July/August 2005 (8 M1) s 1 2 3 st] 4 20 100 eto 5 0 st] 17.1428 100 0 st | 398334 0 el 10 Calculation + (2)(12)=(20)(1) 4 (720) -(5)4) For Elements of row ; (7){100)~(4)(0) = 17.1428 = 100 For Blement of row182_—Contral Engineering (17.1428)(5) (10007) _ or Element of s! row : =~ 358334 For Element of ae 35.633 (-358334)(100)-(171428)(0) For Elements of row ; a = 100 ‘Thus, from Routh's array itis clear chat, there are two sign changes in first column of the Routh's array. Hence system is unstable with two roots located in right half of s-plane. Example : 22 Explain Routh Hurwitz criterion for stability of a control system and examine the stability of + 25! + 3s! + 85 + 2 = 0 using same. ‘July 2007 (10 M) Solution : ‘Toexamine stability, consider ‘The Routh’s Array for the given characteristic equation is syr 32 e}2 8 0 e]a 2 0 sR 0 Calculation = : ForBlenens ot # row; CD=(00) 2 z For Bement of'ow; MBH) a {12)2)~ (10) For Elements of s row ; 2 ‘Thus, from Routh’s Array itis clear that, there are two sign changes in * column ofthe Routh's ‘Array. Hence sysem is unstable with wo roots located in right half of s-plane Brample = 23 ‘The open loop TE. of a feed back control system is given by k 60) H) = ——*__ = Seeayst+2s+2) Find the range of values of k for stability, also determine the stability of the system when kod “Transient and Steady State Response Anal Solution : ‘The characteristic equation of the systems given by 14) He) = 0 + k Heras? 2542) St 6s'+ 100 + BF +k = 0, ‘The Routh’s array for the characteristic equation is given by # eee s Geen 0) ko 0 0 o 0 Forstability, Lk>0 36k or k< 11.556 2 52 ‘System wall be stable for the range 0O0>k>0 Also 10k=1 ie = 600 + 100 k~ 10 > 0. Root ere 0.08 ~ 0.09, 0.08 + 0.09 complex roots, hence system is unstable Example : 25 [A system has a following characteristic equation. Determine the value of K so that system is stable by Routh's-Hurwitz criterion 2s'+ 3s'+ ds'+ 5 + k Dec? 2013/Jan.2014(5M) Solution : Routh Array forte given characteristic equation is sf 2 ak | 3 10 0 5 e| Bo 105 s | 32 09 0 0 3 For system to stable ko 10 3k 3 10 WA PO k= F 3 2 The vabeotstotdbeo << 2 Transient and Sioady Stale Response Analysis — 18585_—Control Engineering REVIEW QUESTIONS Derive the response of a first order system, subjected to unit step function input. [Explain the importance of time constant ox speed of response. i, Define steady state response and steady state error. |. With a neat sketch explain all the time domain specification. i, Detive the expressions for static error coefficients. i, Derive the expressions for peak time and rise time interms of & ad «, for a second order control ystem, 10 ". A unity feedback system is characterised by an openloop transfer function G(s) — Determine the fllowing, when the system is subjected to a unit step input. i. Undamped natural frequency {i Damping ratio i, Peak overshoot iv. Peak time ¥. setling time 4. Ascertain the stability ofthe system given by the characteristic equation SHE SAAD + BEES HE ), by Routt Hurwitz criterion 1 Derive the response of a second order system subjected to unit step input for underdamped case. (0. Weite a note on standard test inputs 1, Explain Routh Hurwite's stability criterion. 2. Write a note on the special cases in RH eiteron. 2(s+8) aer4) Find the closed loop transfer function and also determine the response of the system for unit step put. 3. A feedback system with negative unity feedback has a plant G(s) = 4, Comment on the stability ofthe systems whose characteristic equations are given below. Also Jetermine the number of roots, if any, in the right-half of the s-plane Ltt 4284 10548 =0 Wseste2e est “Transint and Steady State Rosconse Analysis — 187 15, Consider the transfer fureton cl) if RG) ~ F 4D DP aS a8 using the RH criterion, determine the system is stable or not. 16. Using RH criterion, determine the range of K for stability of the system having characteristic equation s* +25 +(K+4) s%+95+25=01 INTRODUCTION hen we want to use or control rea system we must know how it behaves when different snals reapplied tot. This wil give a measure of the dynamic response ofthe system. One Way find the response ofa systems to aply atest signal othe input and 100K at the output see witresponds. Many test signals ae possible, buta simple and useful test signal isthe sine wave rusoida est signal). This is because the output of a system with a sine wave input is alsoasine ve, but witha different amplitude and phase. By measuring the output amplitude and phase ofa stem over a range of frequencies of the input sine wave, a particular version ofthe dynamic sponse is built ~ this is called the frequency response. By frequency response we mean the sponse characteristic ofthe system wen subjected to sinusoidal inputs. Te input frequency s tied, and output characteristics are computed or represented as a function of ce frequency quency response analysis provide useful in sighs nt stability and performance characteristics the contol system. There are a number of ways to represent the frequency response of & acess we wll use Polar plots and Nyquist plc in this chapter 2 POLAR-PLOTS 2 know that frequency response isthe graphical representation ofthe magnitude and its phase gle versus the range of frequencies of the input sinusoidal test signal. Thus, the plot of agnitude |G( jo) and phase angle {G(,0) 28a function w in the polar coordinates gives apolar ot for sinusoidal input. Therefor, pola plot sa plot ofthe magnitude Gj) and its phase angle the frequency is taken over its fll range from zero to infinity. Consider a continuous system ansfer function G(s) represented in the frequency domain as a sinusoidal transfer function. The Sulting GGo) is acompex function of single variable Ttmay be plotted ether fromthe magnitude ‘pase obtained directly orby ploting the eal and imaginary pars of G(jo) aso varies, withthe sar values of magnitude and phase angle taken from the Certesian plot, A positive phase angle motes a phase advance through the system while a negative value a phase lg, Inthe frequency sponse we have Frequency Response Analysis —_ 189 Magnitude, M = [G(jo)H(jo) Phase angle, 9 = |G(ja)HCjo) Procedure to sketch the polar plot 1. Consider the given transfer function Gis)/#(s) 2, Substitute s =o in Gis)H1(s), obtain the frequency response function G(ja)HGia) 3. Obtain the expression for magnitude M(a) and phase angle (0) 4. Tabulate values of M(a), and («) for different values of «(varied from 0 to =) 5. Separate the real and imaginary part by rationalising the complex frequency function G{ jo)H je) to find the intersection poinis, 6. Equate imaginary part of Gj) to zero to find intersection with ral axis and equate real part of G(ja) H() to zer0 to find intersection wth imaginary axis 7. Determine the frequencies atthe interseetion point 8, Sketch the polar plot G( jw) jaa) according to the values obtained in step 4. 5.3 BASIC TERMS USED 1, Encircled: A point is said to be encircled by a closed path if it is found to be inside the closed path. The closed path can be of any contour need nat be a citcle always. Clockwise encirclement are treated as negative. Anticlock wise encirclements treated as positive. 2, Enclosed: AA point is said to be enclosed by a closed path if a point in found tole tothe left of the path when the path is transvesed inthe prescribed direction as shown in figure 5.1 {| nf | ey U LT Fig. 5. Point xis led by closed path T(s) _| 1. Point x is encircled by closed path Ts) Point y is not encircled by closed path F(=) | 2. Point yis not encircled by closed path Fs) Point x is enclosed 73. pointy isenclosed 4, point yis not enclosed “4. point» is not enclosed 3. Closed patl 4, Analytic function : A function is said tobe analytic ina plane ifits value and its derivative have finite existence at that point, is the continuous curve have the same starting and ending point180 —Control Engineering k For example, If F(s) = then F(s) in analytic at all points in s-plane except at s= 0 and aeray =~ a because, as =0 and s =~ 4, the value of F (s) is. Therefore, ples of the function are is singularities, 5, Single valued function ¢ separate value of If F(s) = V5, for s = 49, the values of F(s) are ~7 and +7 -. function F(s) = vs is not a single valued function, On the other hand, the function F(s) = isa single - valued function. function is said to be single valued if it has one value for each 5.4 NYQUIST STABILITY CRITERION ‘The Nyquist Subility Criterion, named after Harry Nyquist, provides a simple test for stability of closed loop control system by examining the open loop system's Nyquist path. Nyquist analysis is a frequency response method. Itis basically a graphical procedure for determining the absolute and relative stability of closed-loop control systems. Stability information obtained directly from a graph of the open-loop frequency response function GH(a). It should be noted here thatthe Routh-Hurwitz stability method can only be used for determining absolute stability and is applicable to systems whose characteristic equation is a finite polynomial in s. Nyquist method is also useful for obtaining information about transfer functions of systems from the experimental frequency response data. The Nyquist analysis can be used for systems with time delays without the need for approximations and gives exact results about both absolute and relative stability of the system. We know that, for stability all roots ofthe characteristic equation, 1 + G(s)H{(s) =O must lie on the left hand side of the s-plane. But, Nyquist suggested that rather than analysing whether all roots ofthe characteristic equation are located in let hand side of s-plane, i s better to find the presence of any one root of the characteristic equation in right hand side of the s-plane which makes the system unstable. For example consider a open loop transfer function Gans) = is called open loop poles of G{s)Hs) 5 =~2is called open loop zeros of G(s)H(s) "Now consider closed loop transfer function as cls) Gls) Res) * THO()A) Frequency Resaense Analysis —_ 191 Now Consider, Bree ereese 0M Ise + Gone) = 145 = 5 } iscalled zer00f1+ — GWH) 4=~2is called pole of 1 + G(s)H(5) Now consider the characteristic equation, ie., 1+ Gi)Ms) = 0 ett? 29 +33 245-0 is the root of the characteristic equation Ths, we can conclude that 1. Poles of 1 + G(s)H(3) = poles of Gis)H{s) i., open loop poles ofthe system 2. Zetos of 1 + G(s)H(s) = roots of the characteristic equation 1 + G(s)H(s) = 0 i.e, closed loop poles ‘Therefore Nyquist suggested that, system is stable if all the zeros of 1 + G(s)A() ie, poles of closed loop system are located inthe left half ofs-plane, Thus Nyquist stability criterion relates the ‘number of zero and poles of | + G(s)H(s) that lie on the RHS of splane tothe open loop frequency response G(s)H(), Nyguist plot is obtained from the polar plot by drawing its mirror image, as @ varies from coto ~ eo and number of encirclements of point ~1 + 0 is observed. Anticlock wise encirclements ate taken aspositive Finally, Nyquist stability criterion can be stated as "For a closed - loop system to be stable, the Nyquist plot of G(s)H(s) must encircle the point (-1 + j0) as many times as the number of poles of G(s)H{(3) that are in the right half of splane. Anticlockwise encirclements are taken as positive and clockwise encirclements are taken as negative”182_—Contol Engineering Mathematically. N=P-z ‘Where NV = Number of encirclements of point (-1, 0) Number of poles of G(s)/{s) that are on the right half of s-plane Z-= Number of zeros of 1 + G(s)H(s) that lie on the right half of splane If Z=0, the closed loop system in stable If P=0, the open loop system in stable 5.4.1 Concert oF Stasiuity CRITERION ‘A Stability test for time invariant linear system can be derived inthe frequency domain. Itis known ‘ss Nyquist stability criterion. tis based on the complex analysis result known as Cauchy’s principle of argument. Note that the system transfer function isa complex function. By applying Cauchy's principle of argument to the open loop system transfer function we will get information about stability of the closed loop system transfer function and arrive atthe Nyquist, stability criterion, ‘The importance of Nyquist stability lies inthe fact that it can also be used to determine he relative degree of system stability by producing the so - called phase and gain stability margins. These stability margins are needed for frequency domain controller design technigues. 5.5 CAUCHY'S PRINCIPLE OF ARGUMENT Let F(s) be an analytic function ina closed region of the complex plane s given in figure 5.2 except ata finite number of points (namely the poles of (3). Iti also assumed that F(s) is analytic at every points on the contour, Then, as s travels around the contour in the s-plane, the function F(s) encircles the origin in the complex plane inthe same direction W times with N given by N=Z-P Where Zand P are the number of zeras and poles ofthe function F(s) inside the contour ALL) AalFODD s-plane RIPON) RUS) 223 p=6 F(s)= plane Ne suchy’s principe of argument Frequency Response Analysis — 193 5.6 GENERALISED NYQUIST PATH ‘The generalised Nyquist path is a closed path in the s-plane which completely encloses the tig! half ofthe s-plane i nn Case 1: IF the open = loop transfer function does not certain any poles on the jc axis or at the origin, the Nyquist path is choosen as shovn in figure 5.3 a Joes orate ome) 180" 0 © (Real) Nyquist path 90° Fig. 53 Case 2: Ifthe open loop transfer function has poles on the Yj axis and atthe origin, the Nyquist path inthe s-plane is as shown i figure 5.4 180°194 —Control Enginosring Case 3: If the open loop transfer function has poles at the origin, the Nyquist path in the s-plane is as shown in figure 5.5. Nyquist path ot" Fig. 5.5 5.7 STABILITY AND RELATIVE STABILITY Stability and Relative Stability are the important characteristics of any control system and are considered as the most fundamental design requirement. Meaning of the stability depends upon the kind of system or the point of view. Because, almost every working system is designed tobe stable. Inmost general terms stability refers to the reaching steady state after passing through the transient period where as relative stability refers to the degree of stability ofa control system. The stability and relative stability can be determined from the frequency response plots using Gain margin and Phase margins. Gein margin indicates a value by which the gain of the system can be increased without affecting the stability ofthe system, while phase margin gives the amount of phase angle ‘hat can be added tothe system without disturbing the stability Gain Margin (GM) Itis a factor by which the system gain ean be inereased to bring the system to the verge of ‘instability. Mathematically it isthe reciprocal of the interception of the ofthe curve on the negative real axis from the origin. i. L Ginga. GM = Fey} when phase anal, (0) egal 0 180" 1 OM =o (eGo), 1 Indecibels, OM = 2008 Gay Freqoncy Response Anayis ~ 195 = =2010g (Gj), [Note : Gain margin must be > 1 forthe system to be sable Phase Margin (PM) {tis the amount of addtional phase - og at the gain crossover frequency required to being the system to the verge of instability. Mathematically, its the 180° + (G(J0)|., Note : Phase angle is measured with " tured with respect to the ~ 180° ie., from negative real axis. Negative phase margin indicates that the system is unstable, * Phase Cross - Over Frequency (a) The frequency at which the phase angle) ofthe n angle 1 open Toop transfer function is equal to - 180° is called phase crossover frequency. Iis given by, aus ©, = (GUO) =— 180° Gain Cross- Over Frequency (a, ) The frequency at which the magnitude M(o) of th tude M(@) of the open loop transfer function called gain cross over frequency. Its given by ©, = [Gayl =1 Following figures 5.6 shows te sign convention used equal to unity is Positive gain margin 210° : 210° gh ph img Uniteirete Unitcircle Negative phase margin margin Unstable System Suble Sytem Fig. 56196 —Controt Engineering Frequency Response Anaiysis—_ 197 Procedure to determine the relative stability of a system using polar plot 1. Draw a unit cirele with origin as centre (radius = 1) 2. Check whether the curve encloses point (-1,j0) oF not 3. Ifthe curve encloses the point (-1, 0). dhe the system in unstable 4. lf the curve does not encloses the point 1,0) then the system is stable ‘To find gain cross over frequency (0,) 1. Draw a unit circle with origin as centre (radius = 1) 2. Find the intersection point on the polar plot (mark asp) 3. Measure the frequency corresponding to this point (= 0,) ‘To find phase cross over frequency (0,) 1. Draw a unit circle with origin as centre (radius = 1) 2. Find the intersection point on the polar plot (mark asp) 3. Draw a line from the origin to meet the point p 4. Measure the angle with respect to ~ 180° (Le, from -ve real axis) te -ngle with resect to negative real ais nthe counter clocwiv dtecton ts ken os SSS it a a fs fap is eh WORKED EXAMPLES ‘Fo find number of encitclements (N) Example: [br ae (vet fom a pint of nee! whose ecrements ar tobe counted though he Sketch the pla plot of he sytem having transfer uncon Cine pain any ot nr erdon ott mert clea ph 2, Cou the mute f nection of is ine wih closed path a = oti 3, Mark these intersection on the same line with small arow indicating direction of closed part at the intersection 4, Cancel the oppositely directed arrows, The remaining arrows gives us the mimber of encirclement fl peed Sano Given ransferfunetionis G(s) Solution: otha point | rec) 5. Clockwise encirclement treated as Negative and Antcock wise encitclements are treated 5 poting sf, positive i For example: I io) = Toe pI je Gio) = T01 jo jo Maine, Mo) = [6 ft — cused 1 5 mh : : i= FO%a) outward vector, 7 + (010) Outward ‘vector SS 0 Phase angle, $e) = (G(jm) = tar! 7 - arr! 0.10 =~ tar (0.10) Ne-141=0198 —Control Engineering For the different values of o (varied between 0 to ), the different values of A(«) and (0) are as given intable o $= - tan! (0.10) 0 1 ° 1 0.98 51° 2 098 1131 3 089 26.56 10 0707 30 0.196) 78.69" 0 907 ‘Since phase angle varies between ~0" to 90° it does not interseets positive real axis, Polar plot forthe transfer function Gls) = nas showin in figure 5.7 10s G(s) Ks) plane o=0 4 oO ‘ois increasing - 90° Fig Example : 2 Sketch the polar plot of the system having open loop transfer function 10s GOH) = Tay Solution: 10s Given transfer funtion, GHG) = TO Frequency Response Analysis puting s= joo, we get ojo | 0+(100)" Magnitude, (a) = [6( 0) jo) 7 gnitude, M(w) = |O(jo)¥L40) = |r (oF 100 Mo) 60 hase angle, a) = JHGa)HCJ0) = tan"! EO ay AO = 90140 Forte different valves (varied between 0 too), we get 10 o M@= (0) = 90° tar 40 Tete? | 0 oO cd 01 0928 68.198" 02 1361 S13 05 2236 2656 1 2425 1403" 5 2496 286° 10 2299 ia 5 25 e When: 105 ‘The polar plot for the transfer function G(s) = “°° is as shown in figure polar pl afer funtion Gls) = 5 ias shown in figure 58200_—Contol Engineering Example: ‘Sketch the polar plot for the system having open loop transfer function Solution: put s= jo, in Gs), we get Gio) = (j0)*(1+ Jo) Magnitude, M(w) = [6i0), Phase angle, ¢(@) = 170) taro ‘When is varied between 0 tos, we have ° Me va) T Sj =F] eS v TTF ‘To find Intersection point Consider Gi jo) ie. 1 6) = S79) Frequency Response Analysis 201 Consider imaginary arto find intersection with real ais and eqeate fo 2er0 ore thus at © = 0, (a) Ho) Similarly consider real poiat co find Intersection with imaginary axis and equate to zer0. ie, 1 Fig. 5.9 are) = => Mo) (0) = 20° Polar plot for the transfer function, Gt Ae 5 i848 shown infigue 59 Example : 4 1 ‘ketch the polar plot forthe transfer function G(s) = —— Polar plot 8) = Te sy(ie 2s) Solution: put s=jooin Go), we gt G( ja) = (5 jo 270) Magnitude, Mea) =[6( a) (Glia) = un" @- tan"'20 ‘When w is varied between 0 (0 ©, we get phase angle, (0)202 —Contiol Engingsting Mo) IO) 0 r e el T — 180 Since phase angle varies Consider etween 0 o -180°, find the intersection point with real and imaginary 00 = jap) (1-204) +i(-30) (ee"40") “To ind poiot of Intersection on real axis, equate imaginary part f0 zero. 30 be ratte’ ) = 00 at @=0 Mo) eo) ‘To find point of Intersection on (1-207) = 0.707 rad/sec: Intersection point frequency at @=0.707 ‘M(o) = 04 0) = -90" Frecuer Response Araysis — 208 The polar plot forthe transfer function Gfe) = is as shown in figure 5.10 ae en tern Ges) HES) plane o=0 = 180° So Fig 510 Example: § Plot the polar diagram for the open loop transfer function G(s) aeeh Solution: ; 2 Given er sons GOH = Np pulting s=ja, we get (je) Ho), = —— HO.= Ta Ga+ Gor?) -o Mo) =|6( jo) (jo) 2 Separate the real and imaginary part of Gi jo) jo) from equation (1) +90) = GLi0}4( 0) = -90~tan"" tan! 2 UOT) = Tae e+ I0) 36 + 12)(2-0") ~afl+o" f+")Frequoncy Response Anaysis ~ 208 Example : 6 204 —Contol Engineering ‘Sketch the polar plot for the transfer fumetion G(s) + i)(s+2) Bee09/Fan.10 (8M) | Siton: oe 10 FGiven transfer function is G(s) = =! put so, we got Pa 2@-0) =0 i ionis GOO) = Tayesay Puts#io. wes i = o- 8 10 =2 ad aw an = = 0 Mo) and §(@) (jo\(i+ 7o)(2+ jo) ‘Thus, for varies values of «, we get f - - Magside, (a) =[6( 0) rad/sec Ma) eo) i 2 = a Phase angle, g(a) = 60) 1 B e a 7 ‘ = Tefen pot & 2 i pare the real nd imaginary past of Go rom Equation (1) 5 i ae — 10 ill ja) ja) ~ 0 20° Gol f0y2 70) j= joY2= 70) Polar plot of G(s)Ft(s) aay Bas showin gue 5.1 200+ 10j0(2-0") =o rors -200° 7 whoo? * [refire) ! (0 fare") Equating imaginary part to zero, we get 180° @=0a0 0° - 90" Fg. 5.10208_—Contol Engineering o (0) radisec | — M : 7 5 5H 7 3166 Tie i 1665 —180" z aH =F 5 006665 236° = 0 “210° 1 Polar potof = ygsay Bast i ie 5.12 180° ° Brumple 77 2 The transfer function of «contol system is G8) = Sa} Sketch the polar plot, Solution : puts =o in Gls), we get Rn 22 G0) = FoF (ios Nuw+2) ~ o(l+ Ja) J) Magnitude,M(o) = |G(o} 2 Vive Varo’ Frequency Response Analyse — 207 bo-tan!® 180°—tan'@~ tan! © 2 ‘When «in varied 0 and «>, we get o Ma) %) oT Ss =F I 7 Se To find Intersection point Consider 2 (1 jo}(2- jo) G(jo) = -12(2-0' 30) orale") ‘To find intersection point with real axis, equate imaginary part to zero 360 orto" fara") oo = M(@) = o and $(@) = -180° “To find intersection point with imaginary axis, equate real part to zero, -12(2-0? -3/0) ie (ise? fare) *° - 2a =0 o= fi = 1a ao=1414, Moy = 1414 a)208_—Control Engineoring The polar plot for transfer function, 2 (ee?) is as shown in figure 5.13. Fig. 5.11 Soe Example ‘Sketch the polar plot for a system whose open loop transfer function is G(s)H{s) Solution : Given transfer function is G(s) (6) GL Jo) Hjo) From Equation (1), Magnitude, Mf (0) phase angle, 6( LT" Dect 2013/Jan.2014 (8M) =“w Frequency Response Analysis — 209 = tan (0)-[360-tan" (ro)] (0) = 360+ tan" (ro) : For the diferent values of « (varied between 0 to <>), the different values M (C9) and 6 («) are sivenin Table @ Ma) #0) ° K 360° yy 707k ais a 0 = Polar plot forthe transfer function G(s)H(s) = is as shown in figure 5.12 270° Img =180° 0 on] @=0" Real ig. $12 Example : 9 ‘Sketch the polar plot for a system whose OLTF G(s) isTs Solution Given transfer function is G(s) Puts =jo, we get210 ontrol Engineering Freeney Response Analysis 211 From Equation (1), Exanple : 10 leo), Obtain the Nyquist diagram forthe system shown in Fig. and ascertain its stability Dec’ 2013/2014 (12 Magnitude, M (©) Js Solution : = ager 100 JY +(roy i Open op ans funtion is OH) = Teas i 5 1 i w= alee (pen oop poles are at s= -0.5, No poles the right hand side of real axis and no poles at he O* Tare ign since are the ples ies othe ll sdeof thes plane, system in stable. The Nyauist pth a : } the s-plane is as shown in figure (eis) a For different values of (varied between Oto «), the different values Mf(o) and 9 (@) are given phase angle, 6(0) intable i t o Mo) $(@) i 0 1 o ' A 0.707 ase t > 7 Sar | Polar plots givenby Pig 514 =270°F Img i Given that G(s)H(s) = G25) a) Pas =f, we get 100+ 0 atta) = Te 0) Magnitode, Me) 6 +40? se) 5 10 deo resn Phase egle (0) = [6(i0) 1 (ja) = tn > tan = tar! (40)212_—Control Engneer ‘To map section ab @ Mo) oo) a a 50 0 100 o ‘To map section be Itis defined by s = ~j, where co varies from Oto ~e, Hence it is the mirror image of section ab ‘To map section eda Itis defined by 8 = pS, Re® whee 0 varies from -90° 1 +80° tinough 0° lim G¢Re wa a fim, GeRetuner = in | apo = lim a Re =0e# e o=0 oF Br & e oy 30° ‘To Find the Intersection point with real axis “Thats, 100 1-20 GLH) = T3570" T=2;0 Equating imaginary part=0 200 _ Tea 7 w=0 ‘Thus Nyquist plot does not intersect the vegaive real axis. Thus, for o=0 : eee FUER sponse Ana's 213 From equation (3), Mf (a) = 100 o=0 0 Real Example : 11 i i K(s+3) ‘The open loo transfer function of «conte sytem is Gia) = “7° Determine the value of K for the system to be stable using Nyquist criterion. Solution : ‘Open loop poles are at s = 0, s = 41. since open loop transfer function has a pole atthe origin, ‘Nyquist path includes a small cirele arounds=Oas shown in figure 5.16, () For mapping ftom s-plane to G(s)#(s) plane put = join GH(s), Thus, we have K(ia+3) __KG+j0) o(J0=1) ~ jo(-1+ 50) ‘To find magnitude and phase angle G(Ja)H jo) Q Magnitude, M(o) = {G(jo)H(jo)) 90°-{180--tan 4)214 Control Engineering To map section ab Itis defined by 5 = je, where o varies from » to 0 “Thus, substituting in equation (3) and (4) we get o | Mo oO) S 2 = 90" 0 = 2 To Wap section bed Its defined by 5= fre @ varying from 90° 0-90" through O° Z K(re*} lim Gy) s= re = lim a A im 3K =m re) = tim 3K ah am = lim 3K = Sie s-plane G(s)HG) e $= (180+ 8) 307 a0" 0 =180" oF 30" ‘When s-plane is mapped into G(s)H() plane moves from -270° to ~180° to -90° in the anticlock iretion with an infinite magnitude ToMap section de Ikis defined by 5 =, oo, 9 various from Oo, It isthe mirror image of section ab Tomap sectioneta Ieisdefined by += lM. Ret, 0 varying fom 90° to + 90° through O° Frequency Response Anaysie— 215 K(te* +3) ] Bi ol0h ae = Bo | 2 OOM =H | tae] AS Ris very large, Re® +3 — Re*as R > o and Re*-1 - Re®as R > 0 [ate ] = te [pana | GOH) Plot rotates inthe clock wise direction. ‘To find Intersection with real axis seperate real and imaginary parts GjoyHt jo) = # ~iek(3+ jo\-1- ja) or(i+0%) 403K + jox(3- 0") 10%) ‘To find intersection with real as equating imaginary part equal to zero * Gilera) @=0 of 3-0 ony . From equation), (3) = From equation (5), 3 = 180°216 — Control Engineering K(s+3) Polar plot of G(s)A(s) = is as shown in figure 5.17 5-1) Example : 12 Using Nyquist criterion, investigate the stability of a system whose open loop transfer k function is COMO) = Taye asa) Solution : Open foop poles are ats = -1, ~2,~3, No poles on the right hand side of real axis and no poles at the origin. since ae the poles lies to the left side of the s-plane, system in stable, The Nyquist path inthe s-plane is as shown in figure 5.18 Given that ae Gre 2er3) For mapping from s-plane to. G(s)H(s) plane put $= in GISAKS) we get GOH) = ay k CU oHO= Texas Gers) “© Magnitude, M(o) =|G( jo) H( je) k Fig. 518 3) Viva Varo e+? t Frequency Response Analysis ~ 2 Phase angle, @(@) = ya) #LJa) = ~tae@— tn ta ‘To map section ab It is defined by s = fo, where @ vaties from 2 to 0. Thus o Mo) oo) = 0 ~210" K 0 f v To map section be Itis defined by s = jo, where @ vatis from 0 too, Hence itis the mirror Image of section ab ‘To map section eda Ieis deine bys = Re where 0 vais fom -90° 0 490" trough 0 me Rene =0e0 @ oF o o 90" 270" ‘To Find the Intersection point with real axis Seyarat real and imaginay pont adequate imaginary pat 0 20. Tats, k li lla 3 ja) F0)GIa]R a)" F= 10)2> 9-70) GC jot) K(6—60*)- jko(11-0") (eotyroryore) Equating imaginary part=0218 —Control Engineering oftt-o?) o(il-o) o ‘Thus, for @= IT from equation (3) oorvii K Magnitude M(o) = i te itersetion point + Foc bly ofthe system, the curve should not enciclethe pint (1,0) kK E a ke 60 kJ Forlarge value of the sytem tobe stable is 0< K <60 of K = 60 then te system in marginally stable Kk Polar plotof G(s) = pipes Dfeeay 885 shown in gare 5.19. Fig, 519 Example: 13 ‘The open loop function of a control system is Gis)H(s) Sketch the Nyquist plot and ascertain the stability Solution : July 2007 (109 Given GEA) = a J #042) Frequency Responss Analysis ~_ 219 0, -2, since poles are at the origin, Nyquist path includes a smal circle igure 5.20 Open loop poles ae at around s = Os shown in plane Fig. 520 For mapping from s-plane to Gis}E1(s) plane putting s= in G(s)H(s), we get GOH) =§ 2 = CaF io#2) Ths, ‘Magnitude, M(w) = |@(jo)H(jo)| = @) phase, ¢(0) = BUOHe) 4) Tomap section ab Itis defined by s = jc, whece © varies from ee to 0 o [M@ | %@ asta 0 =20° 0 =| 180° Tomap section bed Iti defined by »= lim re, where @ varies from 90° to -90° through 0°220 —Contral Engineering ee li Geretytre®) Thus ssplane | G(s)H() plane @ q=-20 OF 130" 0 0 = +180 Tomap section de {tis defined by s = jo, where @ various from 0 10 = It isthe, mirvar image of section ab ‘To map section of efa im ReP, where 0 varying from -90° co 90° through O° kis defined by s= 1 AM GeRENHERe) = FI RTs RO) | = ew ‘Thus 8 g=-30 =90"| ae 0 0 90° 21 To find the intersection point with negative rel axis separate real and imaginary part and equating imaginary part to zero. Thats, (=e) GG@)H{ ja) = (io) @= 8) _ io [tus mo I « ——9) 7 Splane | G(HG) plane ; 7 c= a ir * e d er cy Y I To Map section de Fig. 522 | It is defined by s = ja, where @ varies from 0 to =. It is the mirror image of section ab_ Fox mying om lane GC) ple, ting =f GM We Tolle secon ste GU jovi jay = Ot) _ ® tis defined by s = JIM Rel where 8 varies from 90° to 90° through 0° (oF 0+ jo) +2}0) Ths eee wits _. im tenner, = in | : iw (a) Rei zRe) Monte, Ma = fio) ja} = = site Mo) = 0a) = a es Phase angie, (a) = [GLJo)H{ jo) = tan"! 40~-180°-tan"'o—tan“"(20) (4) To Map ston ab = Wisden bys, hae © vies fom = 100 Ts @ M(@) 4) We +270 ar - ope pw] op [a ee Ted be neon pan wipe als sept rel and imagine nasi at i It is defined by s= lim re, where @ varies from 90° to -90° through O° a (1440) (1- jo\(1-2/0) td oF ja)-2;0) *O= fk) lim G¢rettre) = tig | —_E*E 1+3j0 +407} . 2 CAV erehiedre - (estoN-2i0) Ue) o(I+0")+40") 14 jo+ 100-8?224 —Contret Engineering Equating imaginary partto 2210 it-80"| © = Oorl-8a?'=0 © = 03835 madisec Thus, for @=03535, from equation (3) Magnitude, (w) = 1067 Since plot encircles the point (-1, 0), system is unstable (1448) Fle) is as shown in igure 5.23. Polar plot of G(s) Fig. 5.23, Example : 15 Plot the Nyquist diagram for the open loop trans function G(s)H() determine the nature of stability . arin June/July'08 (10M) Frequency Response Anaiysis~ 225 2 Given Cts = Tea ~) Open loop poles are s=0, circle around s= 038 shown ~2 since one pole is present at origin, Nyquist path includes a small figure 5.24, Fig. 5:24 For mapping from s-plane to G(s)H{s) plane, put s= in G(s)H(s), we get R G0) = Ca RTT a Thus, Magnitude, M(o) = [Gt jo) (jc ele ~ (3) Phase angle, (0) = [3(J0)4 (ia) = -90°tan"!@~ tan = Tomap section ab Itis defined by s =, where © varies from =t0 0 o | Mo 40) = 0 210" 0 = 90° Tomap section bed Its defined by s= 5 rl, where @ varies from 90° 0 90° through O°226 —Control Engineering 2 Jim trent = i ae splane | G(s)H(6) plane é e=-0 90 30" 0 0 “30° oF Tomap section de tis defined by ¢~j, where © varies from 0 to -e,TLs the miror image of section ab Tomap section eta Tis defined by s= lim feo where 8 varies from -90° to 90° through 0° ~ n Hnornennnes Slewtne i To find intersection point with negative eal axis Separate the real and imaginary part of G{ jo)H{Jo), from equation (2) we have B Ai j0\2- jo) IHS) = TOT jaya Jo) * i(1 70)2— j0) Frequency Response Analy 360122 -of1+0°)(4+0%) 36 slz{2-0*) (ofere) oor iror quating imaginary pat to zero, we get 122-0?) afore) 122-@) = 0 2 = VE = Léldndsee 414, rom equation @) Magnitude, M(@) Since plot encircle the point (-1,j0) system is unstable 2 er e+2) Thus at Nyquist plot of G(s)H(s)= is as shown in figure 5.2. 270° § 525 5.8 CONSTANT MAGNITUDE CIRCLE - M- CIRCLE Mccice are the constant magnitude circles or constant magnitude loci of a losed loop transfer function obtained from the polar plo to determine the closed loop frequency response. Consider smity feedback contol system, we know that closed loop transfer function given by Go) Ry 7 Tea) Ho)228 —Contol Engineering ‘Since ‘sis the complex quantity et |C(ieo)} Gio) thy +x 5 Squaring both sides, we get w[ieay sy] eee miles s2eey']-e- x <0 2 (MEA) + y (1) + Mx HP = 0 Dividing by (1), we get wx eee 7) * Wea 9 2M7x wt FP oa) = ae © (lex) +3" =O Frequency Response Analysis — 229 For= in the equation (1) +1 This iste eatin of sph eprint psig tour (10) nthe Go) plane. Thus, for diferent values of M, a family of circles are obtained as shown in figure 5.26. These family of circles for different values of M are called M -eirele orconstant magnitude circles or constant magnitude loc. ‘The different values of Mare as shown in the table 5.1 Tee 5.1 Centre a 2 xe Meso a Ga, 0 Cog 03 COHRBD, 04 C1940) 05 (03330) 7 = = 2 330) 1353 5 SR) O28 0 CLOTOI Tro ‘The Following Conclusions are made regarding these circles Case 4: For M< 1 ‘As the M decreases, the radii of constant M - circle reduces and centres are located on positive real axis shifts towards the ori Case 2: For M = 0 -M circle has the radius zero and centre at the origin Case 3: For M= 1 -M-circle has the infinite radivs with centre a the infinity on the real axis. Le, itis a steight tine allel ote imaginary axis it intersect the rel axis at Case 4: For M>1 As the M is increases, the radit of constant M - circle reduces and centres ae located on negative real axis shift towards the origin.230_—Contol Engineering Thus, Finally, Whether Mf Increases or decreases. the radius becomes smaller and smaller and finally cirles converge © the critical point -1.j0) M=12 x Fig. 526A family of consant Mirles 5.9 N- CIRCLE [CONSTANT PHASE SHIFT CIRCLES] cies ae the constant phase shift circles orconsan pase shift oct of a close oop transfer ‘sed fo determine the closed loop frequency response The phase shift of a closed loop system may be written as 6 = angle of (r+) ~angle (1 +x+ 5) y stant 2 ~ tart ww 6 La N=ung Adng 4,4. (ow 2 1 a4 seretayZ : noth sides we get 1 L 2° 7"2N For different values of Nin the equation, family of circles re obtained areas shoim in figure 5.27 ‘his equation represents equation of circles with centrex = with radius, known as N «circles. These ae also known as constant phase shift circles or constant Circles. ase shift Fig. 527 A family of constant N crctes232 Control Engineering The different values of NV are as shown in table 52 Table 5.2 | Centre | 6 N=tan 50° Oo os as 4 o70T 0° 03773 1 ° ° sr 0373 1 145 1 0707 2 as ‘The following conclusions are made regarding the circles 1 1. The centre is lying always at distance x= and y part depends on the phase shit 2. The plots are reflection onthe real axis of a G plane for values of @ and ~ 0 3. All the circles pass through ‘1" as well as! Frequency Response Analysis 243 REVIEW QUESTIONS 1. Whats fequency response Analysis ? Explain 2. Whats polar plots ? 3 Defi the following terms ‘gain margin ». phase margin «. gain crossover frequency 4. phase crossover frequency e.enclosed path £ enczclement 4. Explain the concept Nyquist stability criterion 5. State and explain Cauchy's principle of Argument 6 Sketch the Nyquist plot forthe sytem having open loop wansfr function pees O18) = Trailer 2yor3) and analyse the stability ofthe system, May/June 2010 7. The open loop transfer funtion ofa unity Feeback system is K GUS) = Tre A+ 01255) Draw the polar plot and determine 1. The value of K so that gain margin is 32 dB 2. The value of K so that phase margin i 6O° May/Sune 2010 6. Sketch the polar plat ofthe open lop transfer function a OW) = Sia 25) 7. Sketch the polar plot of the open loop transfer function K(s+7) GS) = 510) 8. Sketch the polar plot of the system with open lop transfer Tenction 3 Gs) = (+ 05)61)234_—Control Engineering 9. Draw the polar plot for GOH = Fos \When o varies from 0 =, determine the values of K for which the close loop system is stable. 10, Fora unity feedback syster the open loop transfer function is given by s(02s+1) orN002s-+1) Gis) i. Draw the Nyguist plot ii Is the system stable ? ii, What are phase and gain margins? 11, Aunity feedback system has a loop transfer function (5+2) Gi. using Nyquisterterion Determine whether the closed loop system is stable or no. Gs) = 6.1 INTRODUCTION Frequency response ofa system is the sponse aalyss forthe sinusoidal input signal of various {tequencies. In general frequency response analysis means plo of the magnitude (M) and phase angle (6) against the inpot frequency of oscillations (a), The magnitude and phase angle of the frequency response can be plotted against the input frequency directly on the linear stale. But when the input frequency is varied between Oto there i wide range of variations in magnitude and phase angle, hence it becomes cubersome to accommodate ll such variations with near pot but these variations can be convient shown i the logarithmic scale. Thus, representation ofthe variations inthe magaitude andthe phase angle against the inpat frequency in the logrthmic scale is known as logarithmic plot or Bode plot named after Hendrik Wade Bode (1905 - 1982) Bode suggested frequency response ofthe linear time invariant system can be studied by ploting ‘helogarithmie values of magnitude in decibels against the logariuic values of fequency and alse ‘phase angle in degrees agains the logarithmic values of fequency. ‘Thus, Bode plots consists of two parts: Magnitude Plot: Ii «plot in which magnitude of Gt ja) ja) expressed in decibels ploted agains frequency © in logarithmic scale Magnitude indB = 2010g,, |G(0) (jo) a Phase Plot: Iti «plot in which phase angle expressed in degrees ploted ageinst frequency © inlogarthmic scale Phase angle @ = (G(Jo)#Uio) Nove: The magnitude plot and phase plot normally drawn on the Semi-log Sheet. n semi-log sheet, X-axis ‘s used to represent the logariimic values af frequency @ which is non linens one, so X-axis is divided inin cycles called decades as One, Two, Three and s0 om, While Years is used o represent the values of magnitude n dB and phase angle @ in degrees which ‘are linear one as shown in figure 6.1296 _—Contel Engineering Non-Linear Kane Fig. 6.1: Semi log- sheet 6.2 BODE PLOT FOR OPEN LOOP TRANSFER FUNCTION G{ joo) H{ jo) ‘Consider the open loop transfer function inthe standard time constant form as coo = Betas) - s(legy(l+ Bs) (0 ee ‘where Kis the Constant system gain and mis the type of the system Tey T pT Tyrone 26 the time constant of different zeros and poles. Convert the equation (1) which is in the time domain to the frequency domain transfer function by substituting s=jo x(n Jo\l tio) ie, GC JoyH jo) = eTnGah (aor orien ty[ read) E ‘Ths, equation (2) cons of seven diferent kinds of ators which conti tothe Boeplot ‘They are 1 Contant sytem Gain (K) eel 2-Poles athe origin, 5 3. Zeros atthe origin, jo Note: Zeros and Poles at the origin cannot exist at the same time Frequency response analysis using bode plots — 297 Tei 4. Sings (1 5 Simple Zeros, (1+ fo) Note: Simple Zeros and Poles are known as First onder factors. Scots 2). ) 7. Complex conse a0, ( sae} 22) Bo plos canbe ound verycalying te nya eigt ine erence 6.2.1 Conraisution oF Various Factors to Boo PLor Factor 1: Constant system gain K To Maga Pat Let G(s) Hs) = K G(ja) H{ ja) = K+j0 [oUio)a(ial = SRO =x agnitude, [GH jo) (Real)? +(img)” Magnitude, ‘Magnitude in 8, [Gio)HJo},g = 20108, K = Constant Thus, the contribution ofthe constant gainXis a horizontal line parallel tothe X-axis atthe distance =20 log,, K in dB as shown in figure 6.2. fn } a f cots K-10 Zo 2 20 og K<1 5-20 = K-04 or i ‘o logscaley > " Fig, 6.2: Bode magnitude plot for constant gala K