27 views

Uploaded by avionics54g

A brief description of flight simulation

- 8.2
- Modeling and Control of Crane on Offshore Vessel
- JNTU old question papers 2007
- Frequency-domain Approach for Calculating Wind-Induced Vibration and Aeroelastic Stability Characteristics of Long-span Bridges
- Thin Airfoil Theory
- Failures of Adaptive Control Theory
- Cohen-Tannoudji Claude , Jacques Roc Gilbert Grynberg - Photons and Atoms. Introduction to Quantum Electrodynamics (509s)
- Vibrational Spectroscopy Basics
- material point method
- Notes on Aircraft Performance 2016v2
- Cartpole Eom
- XFLR
- The Proca Field and Symmetry Breaking / Mass Generation -- as seen from the vantage point of classical field theory.
- SubsonicAirfoilsPres.pdf
- MIKEL_OGUETA_GUTIERREZ.pdf
- Journal
- Automation
- Double Well in a Ring
- Modelado
- Relativistic Electrodynamics

You are on page 1of 678

com

www.ebook777.com

free ebooks ==> www.ebook777.com

free ebooks ==> www.ebook777.com

Flight Dynamics,

Simulation, and Control

For Rigid and

Flexible Aircraft

www.ebook777.com

free ebooks ==> www.ebook777.com

free ebooks ==> www.ebook777.com

Flight Dynamics,

Simulation, and Control

For Rigid and

Flexible Aircraft

Ranjan Vepa

Taylor & Francis Group, an informa business

www.ebook777.com

free ebooks ==> www.ebook777.com

MATLAB® is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks does not

warrant the accuracy of the text or exercises in this book. This book’s use or discussion of MATLAB® soft-

ware or related products does not constitute endorsement or sponsorship by The MathWorks of a particular

pedagogical approach or particular use of the MATLAB® software.

CRC Press

Taylor & Francis Group

6000 Broken Sound Parkway NW, Suite 300

Boca Raton, FL 33487-2742

CRC Press is an imprint of Taylor & Francis Group, an Informa business

Version Date: 20140707

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts

have been made to publish reliable data and information, but the author and publisher cannot assume

responsibility for the validity of all materials or the consequences of their use. The authors and publishers

have attempted to trace the copyright holders of all material reproduced in this publication and apologize to

copyright holders if permission to publish in this form has not been obtained. If any copyright material has

not been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmit-

ted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented,

including photocopying, microfilming, and recording, or in any information storage or retrieval system,

without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.

com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood

Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and

registration for a variety of users. For organizations that have been granted a photocopy license by the CCC,

a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used

only for identification and explanation without intent to infringe.

Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com

http://www.crcpress.com

free ebooks ==> www.ebook777.com

www.ebook777.com

free ebooks ==> www.ebook777.com

free ebooks ==> www.ebook777.com

Contents

Preface.................................................................................................................... xix

Author.................................................................................................................. xxiii

1.1 Introduction............................................................................................ 1

1.2 Components of an Aeroplane............................................................... 1

1.2.1 Fuselage......................................................................................1

1.2.2 Wings.......................................................................................... 2

1.2.3 Tail Surfaces or Empennage....................................................2

1.2.4 Landing Gear............................................................................. 3

1.3 Basic Principles of Flight....................................................................... 3

1.3.1 Forces Acting on an Aeroplane............................................... 3

1.3.2 Drag and Its Reduction............................................................5

1.3.3 Aerodynamically Conforming Shapes: Streamlining.........6

1.3.4 Stability and Balance................................................................ 6

1.4 Flying Control Surfaces: Elevator, Ailerons and Rudder................. 7

1.4.1 Flaps, High-Lift and Flow Control Devices........................ 10

1.4.2 Introducing Boundary Layers............................................... 12

1.4.3 Spoilers..................................................................................... 15

1.5 Pilot’s Controls: The Throttle, the Control Column

and Yoke, the Rudder Pedals and the Toe Brakes........................... 16

1.6 Modes of Flight..................................................................................... 16

1.6.1 Static and In-Flight Stability Margins.................................. 18

1.7 Power Plant........................................................................................... 19

1.7.1 Propeller-Driven Aircraft...................................................... 19

1.7.2 Jet Propulsion.......................................................................... 19

1.8 Avionics, Instrumentation and Systems........................................... 20

1.9 Geometry of Aerofoils and Wings..................................................... 21

1.9.1 Aerofoil Geometry.................................................................. 21

1.9.2 Chord Line............................................................................... 21

1.9.3 Camber.....................................................................................22

1.9.4 Leading and Trailing Edges..................................................22

1.9.5 Specifying Aerofoils............................................................... 23

1.9.6 Equations Defining Mean Camber Line.............................. 24

1.9.7 Aerofoil Thickness Distributions......................................... 24

1.9.8 Wing Geometry....................................................................... 26

Chapter Highlights......................................................................................... 30

vii

www.ebook777.com

free ebooks ==> www.ebook777.com

viii Contents

Exercises........................................................................................................... 30

Answers to Selected Exercises...................................................................... 32

References........................................................................................................ 32

2.1 Introduction.......................................................................................... 33

2.2 Continuity Principle............................................................................ 33

2.2.1 Streamlines and Stream Tubes.............................................34

2.3 Bernoulli’s Principle.............................................................................34

2.4 Laminar Flows and Boundary Layers..............................................34

2.5 Turbulent Flows.................................................................................... 35

2.6 Aerodynamics of Aerofoils and Wings............................................ 35

2.6.1 Flow around an Aerofoil........................................................ 36

2.6.2 Mach Number and Subsonic and Supersonic Flows......... 36

2.7 Properties of Air in the Atmosphere................................................. 38

2.7.1 Composition of the Atmosphere: The Troposphere,

Stratosphere, Mesosphere, Ionosphere and Exosphere..... 38

2.7.2 Air Density.............................................................................. 39

2.7.3 Temperature............................................................................. 39

2.7.4 Pressure.................................................................................... 39

2.7.5 Effects of Pressure and Temperature................................... 40

2.7.6 Viscosity................................................................................... 40

2.7.7 Bulk Modulus of Elasticity.................................................... 41

2.7.8 Temperature Variations with Altitude: The Lapse Rate.... 41

2.8 International Standard Atmosphere (from ESDU 77021, 1986)..... 41

2.9 Generation of Lift and Drag............................................................... 45

2.10 Aerodynamic Forces and Moments.................................................. 47

2.10.1 Aerodynamic Coefficients..................................................... 50

2.10.2 Aerofoil Drag........................................................................... 53

2.10.3 Aircraft Lift Equation and Lift Curve Slope.......................54

2.10.4 Centre of Pressure................................................................... 57

2.10.5 Aerodynamic Centre.............................................................. 57

2.10.6 Pitching Moment Equation.................................................... 58

2.10.7 Elevator Hinge Moment Coefficient..................................... 60

Chapter Highlights......................................................................................... 61

Exercises...........................................................................................................63

Answers to Selected Exercises......................................................................65

References........................................................................................................ 66

3.1 Introduction.......................................................................................... 67

3.2 Speeds of Equilibrium Flight............................................................. 71

3.3 Basic Aircraft Performance................................................................. 73

3.3.1 Optimum Flight Speeds......................................................... 73

3.4 Conditions for Minimum Drag.......................................................... 76

free ebooks ==> www.ebook777.com

Contents ix

3.6 Range and Endurance Estimation.....................................................77

3.7 Trim........................................................................................................ 79

3.8 Stability of Equilibrium Flight........................................................... 82

3.9 Longitudinal Static Stability...............................................................84

3.9.1 Neutral Point (Stick-Fixed)..................................................... 85

3.9.2 Neutral Point (Stick-Free)....................................................... 85

3.10 Manoeuvrability................................................................................... 86

3.10.1 Pull-Out Manoeuvre.............................................................. 86

3.10.2 Manoeuvre Margin: Stick-Fixed........................................... 87

3.10.3 Manoeuvre Margin: Stick-Free............................................. 89

3.11 Lateral Stability and Stability Criteria.............................................. 89

3.12 Experimental Determination of Aircraft Stability Margins.......... 91

3.13 Summary of Equilibrium- and Stability-Related Equations......... 92

Chapter Highlights......................................................................................... 95

Exercises........................................................................................................... 97

Answers to Selected Exercises.................................................................... 101

References...................................................................................................... 102

4.1 Introduction........................................................................................ 103

4.2 Aircraft Dynamics............................................................................. 103

4.3 Aircraft Motion in a 2D Plane.......................................................... 104

4.4 Moments of Inertia............................................................................. 109

4.5 Euler’s Equations and the Dynamics of Rigid Bodies.................. 111

4.6 Description of the Attitude or Orientation..................................... 115

4.7 Aircraft Equations of Motion........................................................... 119

4.8 Motion-Induced Aerodynamic Forces and Moments................... 122

4.9 Non-Linear Dynamics of Aircraft Motion

and the Stability Axes....................................................................... 125

4.9.1 Equations of Motion in Wind Axis Coordinates,

V T, α and β.............................................................................. 130

4.9.2 Reduced-Order Modelling: The Short Period

Approximations.................................................................... 135

4.10 Trimmed Equations of Motion......................................................... 137

4.10.1 Non-Linear Equations of Perturbed Motion..................... 139

4.10.2 Linear Equations of Motion................................................. 140

Chapter Highlights....................................................................................... 141

Exercises......................................................................................................... 142

References...................................................................................................... 143

of Motion....................................................................................................... 145

5.1 Introduction........................................................................................ 145

5.2 Small Perturbations and Linearisations......................................... 145

www.ebook777.com

free ebooks ==> www.ebook777.com

x Contents

Derivative Concept............................................................................ 148

5.4 Direct Formulation in the Stability Axis........................................ 152

5.5 Decoupled Equations of Motion...................................................... 158

5.5.1 Case I: Motion in the Longitudinal Plane of Symmetry....158

5.5.2 Case II: Motion in the Lateral Direction,

Perpendicular to the Plane of Symmetry.......................... 160

5.6 Decoupled Equations of Motion in terms of the Stability

Axis Aerodynamic Derivatives........................................................ 161

5.7 Addition of Aerodynamic Controls and Throttle......................... 164

5.8 Non-Dimensional Longitudinal and Lateral Dynamics.............. 173

5.9 Simplified State-Space Equations of Longitudinal

and Lateral Dynamics....................................................................... 179

5.10 Simplified Concise Equations of Longitudinal and Lateral

Dynamics............................................................................................ 181

Chapter Highlights....................................................................................... 182

Exercises......................................................................................................... 182

Reference........................................................................................................ 184

6.1 Introduction........................................................................................ 185

6.2 Dynamic and Static Stability............................................................ 185

6.2.1 Longitudinal Stability Analysis.......................................... 185

6.2.2 Lateral Dynamics and Stability.......................................... 196

6.3 Modal Description of Aircraft Dynamics and the Stability

of the Modes....................................................................................... 201

6.3.1 Slow–Fast Partitioning of the Longitudinal

Dynamics............................................................................... 201

6.3.2 Slow–Fast Partitioning of the Lateral Dynamics.............. 204

6.3.3 Summary of Longitudinal and Lateral Modal

Equations................................................................................ 213

6.3.3.1 Phugoid or Long Period....................................... 213

6.3.3.2 Short Period............................................................ 214

6.3.3.3 Third Oscillatory Mode........................................ 214

6.3.3.4 Roll Subsidence...................................................... 215

6.3.3.5 Dutch Roll............................................................... 215

6.3.3.6 Spiral....................................................................... 215

6.4 Aircraft Lift and Drag Estimation................................................... 216

6.4.1 Fuselage Lift and Moment Coefficients............................. 219

6.4.2 Wing–Tail Interference Effects............................................ 220

6.4.3 Estimating the Wing’s Maximum Lift Coefficient........... 220

6.4.4 Drag Estimation.................................................................... 221

6.5 Estimating the Longitudinal Aerodynamic Derivatives.............225

6.6 Estimating the Lateral Aerodynamic Derivatives......................... 232

6.6.1 Perturbation Analysis of Trimmed Flight......................... 238

free ebooks ==> www.ebook777.com

Contents xi

Flight....................................................................................... 238

6.6.3 Perturbation Analysis of Lateral Trimmed Flight........... 243

6.6.3.1 Control Settings for Steady Sideslip................... 243

6.6.3.2 Control Settings for Turn Coordination

and Banking........................................................... 245

6.6.4 Perturbations of Coupled Trimmed Flight....................... 250

6.6.5 Simplified Analysis of Complex Manoeuvres:

The Sidestep Manoeuvre..................................................... 250

Chapter Highlights....................................................................................... 252

Exercises......................................................................................................... 255

Answers to Selected Exercises.................................................................... 263

References...................................................................................................... 264

and Non-Linear Phenomenon................................................................... 265

7.1 Introduction........................................................................................ 265

7.2 Longitudinal and Lateral Modal Equations.................................. 265

7.3 Methods of Computing Aircraft Dynamic Response.................. 269

7.3.1 Laplace Transform Method................................................. 270

7.3.2 Aircraft Response Transfer Functions............................... 270

7.3.3 Direct Numerical Integration.............................................. 275

7.4 System Block Diagram Representation........................................... 277

7.4.1 Numerical Simulation of Flight Using

MATLAB®/Simulink®.......................................................... 283

7.5 Atmospheric Disturbance: Deterministic Disturbances..............284

7.6 Principles of Random Atmospheric Disturbance Modelling....... 291

7.6.1 White Noise: Power Spectrum and Autocorrelation....... 291

7.6.2 Linear Time-Invariant System with Stochastic

Process Input......................................................................... 293

7.7 Application to Atmospheric Turbulence Modelling..................... 296

7.8 Aircraft Non-Linear Dynamic Response Phenomenon............... 299

7.8.1 Aircraft Dynamic Non-Linearities and Their Analysis..... 302

7.8.2 High-Angle-of-Attack Dynamics and Its

Consequences........................................................................305

7.8.3 Post-Stall Behaviour.............................................................306

7.8.4 Tumbling and Autorotation................................................ 307

7.8.5 Lateral Dynamic Phenomenon........................................... 307

7.8.6 Flat Spin and Deep Spin......................................................308

7.8.7 Wing Drop, Wing Rock and Nose Slice.............................309

7.8.8 Fully Coupled Motions: The Falling Leaf......................... 309

7.8.9 Regenerative Phenomenon.................................................. 311

Chapter Highlights....................................................................................... 312

Exercises......................................................................................................... 312

References...................................................................................................... 330

www.ebook777.com

free ebooks ==> www.ebook777.com

xii Contents

8.1 Automatic Flight Control Systems: An Introduction.................... 333

8.2 Functions of a Flight Control System.............................................. 336

8.3 Integrated Flight Control System..................................................... 347

8.3.1 Guidance System: Interfacing to the Automatic

Flight Control System........................................................... 352

8.3.2 Flight Management System................................................. 353

8.4 Flight Control System Design..........................................................354

8.4.1 Block Diagram Algebra........................................................ 357

8.4.2 Return Difference Equation................................................ 360

8.4.3 Laplace Transform................................................................ 362

8.4.4 Stability of Uncontrolled and Controlled Systems.......... 362

8.4.5 Routh’s Tabular Method....................................................... 365

8.4.6 Frequency Response............................................................. 366

8.4.7 Bode Plots............................................................................... 369

8.4.8 Nyquist Plots......................................................................... 369

8.4.9 Stability in the Frequency Domain.................................... 369

8.4.10 Stability Margins: The Gain and Phase Margins............. 370

8.4.11 Mapping Complex Functions and Nyquist Diagrams.... 370

8.4.12 Time Domain: The State Variable Representation........... 371

8.4.13 Solution of the State Equations and the

Controllability Condition.................................................... 373

8.4.14 State-Space and Transfer Function Equivalence.............. 375

8.4.15 Transformations of State Variables..................................... 376

8.4.16 Design of a Full-State Variable Feedback Control Law..... 377

8.4.17 Root Locus Method.............................................................. 379

8.4.18 Root Locus Principle............................................................ 381

8.4.19 Root Locus Sketching Procedure....................................... 381

8.4.20 Producing a Root Locus Using MATLAB®....................... 385

8.4.21 Application of the Root Locus Method: Unity

Feedback with a PID Control Law..................................... 387

8.5 Optimal Control of Flight Dynamics.............................................. 390

8.5.1 Compensating Full-State Feedback: Observers and

Compensators........................................................................ 391

8.5.2 Observers for Controller Implementation......................... 392

8.5.3 Observer Equations.............................................................. 393

8.5.4 Special Cases: The Full- and First-Order Observers........ 393

8.5.5 Solving the Observer Equations......................................... 395

8.5.6 Luenberger Observer............................................................ 396

8.5.7 Optimisation Performance Criteria................................... 396

8.5.8 Good Handling Domains of Modal Response

Parameters............................................................................. 397

8.5.9 Cooper–Harper Rating Scale..............................................400

free ebooks ==> www.ebook777.com

Contents xiii

Systems and Autopilots..................................................................... 401

8.6.1 Design of a Pitch Attitude Autopilot Using PID

Feedback and the Root Locus Method.............................. 401

8.6.2 Example of Pitch Attitude Autopilot Design for the

Lockheed F104 by the Root Locus Method......................... 405

8.6.3 Example of Pitch Attitude Autopilot Design,

Including a Stability Augmentation Inner Loop,

by the Root Locus Method.................................................. 405

8.6.4 Design of an Altitude Acquire-and-Hold Autopilot........408

8.6.5 Design of a Lateral Roll Attitude Autopilot...................... 416

8.6.6 Design of a Lateral Yaw Damper........................................ 419

8.6.7 Design of a Lateral Heading Autopilot.............................. 421

8.6.8 Turn Coordination with Sideslip Suppression.................423

8.6.9 Application of Optimal Control to Lateral Control

Augmentation Design..........................................................425

8.7 Performance Assessment of a Command or Control

Augmentation System....................................................................... 428

8.8 Linear Perturbation Dynamics Flight Control Law Design

by Partial Dynamic Inversion.......................................................... 429

8.8.1 Design Example of a Longitudinal Autopilot Based

on Partial Dynamic Inversion.............................................434

8.9 Design of Controllers for Multi-Input Systems............................. 437

8.9.1 Design Example of a Lateral Turn Coordination

Using the Partial Inverse Dynamics Method................... 437

8.9.2 Design Example of the Simultaneously Operating

Auto-Throttle and Pitch Attitude Autopilot..................... 439

8.9.3 Two-Input Lateral Attitude Control Autopilot................. 441

8.10 Decoupling Control and Its Application: Longitudinal

and Lateral Dynamics Decoupling Control...................................446

8.11 Full Aircraft Six-DOF Flight Controller Design by Dynamic

Inversion..............................................................................................448

8.11.1 Control Law Synthesis......................................................... 459

8.11.2 Example of Linear Control Law Synthesis

by Partial Dynamic Inversion: The Fully

Propulsion-Controlled MD11 Aircraft.............................. 462

8.11.3 Example of Quasi-Non-Linear Control Law

Synthesis by Partial Dynamic Inversion: The Fully

Propulsion-Controlled MD11 Aircraft..............................464

8.11.4 Full Aircraft Orientation Control Law Design

by Dynamic Inversion.......................................................... 468

8.11.5 Aircraft Flight Control Synthesis in Wind Axes

Coordinates, V T, β and α...................................................... 471

www.ebook777.com

free ebooks ==> www.ebook777.com

xiv Contents

Exercises......................................................................................................... 475

Answers to Selected Exercises....................................................................484

References...................................................................................................... 485

9.1 Introduction........................................................................................ 487

9.2 Piloted Flight Simulation.................................................................. 488

9.2.1 Full Moving-Base Simulation: The Stewart Platform...... 491

9.2.2 Kinematics of Motion Systems............................................ 492

9.2.3 Principles of Motion Control............................................... 493

9.2.4 Motion Cueing Concepts..................................................... 493

9.3 Principles of Human Pilot Physiological Modelling..................... 497

9.3.1 Auricular and Ocular Sensors............................................ 498

9.4 Human Physiological Control Mechanisms.................................. 502

9.4.1 Crossover Model...................................................................504

9.4.2 Neal–Smith Criterion........................................................... 507

9.4.3 Pilot-Induced Oscillations...................................................508

9.4.4 PIO Categories....................................................................... 509

9.4.5 PIOs Classified under Small Perturbation Modes........... 510

9.4.6 Optimal Control Models...................................................... 510

9.4.7 Generic Human Pilot Modelling........................................ 511

9.4.8 Pilot–Vehicle Simulation...................................................... 515

9.5 Spatial Awareness.............................................................................. 516

9.5.1 Visual Displays...................................................................... 517

9.5.2 Animation and Visual Cues................................................ 518

9.5.3 Visual Illusions...................................................................... 520

Chapter Highlights....................................................................................... 522

Exercises......................................................................................................... 522

References...................................................................................................... 528

10.1 Introduction........................................................................................ 529

10.2 Flight Dynamics of Flexible Aircraft.............................................. 529

10.3 Newton–Euler Equations of a Rigid Aircraft................................. 530

10.4 Lagrangian Formulation................................................................... 536

10.4.1 Generalised Coordinates and Holonomic Dynamic

Systems................................................................................... 537

10.4.2 Generalised Velocities.......................................................... 537

10.4.3 Virtual Displacements and Virtual Work......................... 538

10.4.4 Principle of Virtual Work..................................................... 539

10.4.5 Euler–Lagrange Equations..................................................540

10.4.6 Potential Energy and the Dissipation Function...............543

10.4.7 Euler–Lagrange Equations of Motion

in Quasi-Coordinates...........................................................545

free ebooks ==> www.ebook777.com

Contents xv

10.4.9 Application of the Lagrangian Method to a Rigid

Aircraft................................................................................ 553

10.5 Vibration of Elastic Structures in a Fluid Medium..................... 559

10.5.1 Effects of Structural Flexibility in Aircraft

Aeroelasticity..................................................................... 563

10.5.2 Wing Divergence............................................................... 563

10.5.3 Control Reversal................................................................ 565

10.5.4 Wing Flutter....................................................................... 566

10.5.5 Aerofoil Flutter Analysis.................................................. 567

10.6 Unsteady Aerodynamics of an Aerofoil....................................... 575

10.7 Euler–Lagrange Formulation of Flexible Body Dynamics......... 582

10.8 Application to an Aircraft with a Flexible Wing Vibrating

in Bending and Torsion................................................................... 595

10.8.1 Longitudinal Small Perturbation Equations

with Flexibility................................................................... 595

10.8.2 Lateral Small Perturbation Equations with Flexibility........ 599

10.9 Kinetic and Potential Energies of the Whole Elastic Aircraft... 601

10.9.1 Kinetic Energy................................................................... 601

10.9.2 Simplifying the General Expression...............................604

10.9.3 Mean Axes..........................................................................604

10.9.4 Kinetic Energy in terms of Modal Amplitudes............ 605

10.9.5 Tisserand Frame................................................................ 607

10.10 Euler–Lagrange Matrix Equations of a Flexible Body

in Quasi-Coordinates...................................................................... 611

10.11 Slender Elastic Aircraft................................................................... 614

10.12 Aircraft with a Flexible Flat Body Component............................ 618

10.12.1 Elastic Large Aspect Ratio Flying Wing Model............. 618

10.12.2 Flexible Aircraft in Roll.................................................... 620

10.13 Estimating the Aerodynamic Derivatives: Modified Strip

Analysis............................................................................................. 622

Chapter Highlights....................................................................................... 627

Exercises......................................................................................................... 627

Answers to Selected Exercises....................................................................648

References...................................................................................................... 649

Index............................................................................................................... 651

www.ebook777.com

free ebooks ==> www.ebook777.com

free ebooks ==> www.ebook777.com

List of Acronyms

AC Aerodynamic centre

ADF Automatic direction finding

amc Aerodynamic mean chord

AR Aspect ratio

BDF Backward difference formula

CG Centre of gravity

CH Cooper–Harper (rating)

CM Centre of mass

CP Centre of pressure

DME Distance measuring equipment

EFIS Electronic flight information system

EIS Electronic information system

EPR Engine pressure ratio

FBW Fly by wire

FCU Flight control unit

FDAU Flight data acquisition unit

FMGS Flight management and guidance system

GPS Global positioning system

HSI Horizontal situation indicator

HUD Head-up displays

IAS Indicated airspeed

IFS In-flight simulation

ILS Instrument landing system

INS Inertial navigation system

NDF Numerical differentiation formula

NP Neutral point

PD Proportion derivative

PID Proportional, integral, derivative

PIO Pilot-induced oscillation

psfc Power-specific fuel consumption

RMI Radio magnetic indicator

SISO Single input, singe output

TCAS Traffic collision avoidance system

TR Trapezoidal rule

Tsfc Thrust-specific fuel consumption

VHF Very high frequency

VOR VHF omni-range or vestibulo-ocular reflex

xvii

www.ebook777.com

free ebooks ==> www.ebook777.com

free ebooks ==> www.ebook777.com

Preface

phenomenal levels. A plethora of low-cost airlines have made it possible for

the common man to travel between continents at relatively reasonable fares.

This has also led to the design of newer energy-efficient aircraft incorporat-

ing the principles of feedback control. These aircraft have generally tended

to be lighter and more flexible because of the use of composite structures and

other smart materials. It therefore becomes important to consider the aircraft

not as a rigid body, as has been done traditionally in the past, but as an inher-

ently flexible body. Such considerations will require a revision of a number

of traditional concepts, although many of them can be easily adapted to the

flexible aircraft.

This book addresses the core issues involved in the dynamic modelling,

simulation and control of a selection of aircraft. The principles of model-

ling and control could be applied both to traditional rigid aircraft as well

as more modern flexible aircraft. A primary feature of this book is that it

brings together a range of diverse topics relevant to the understanding of

flight dynamics, its regulation and control and the design of flight control

systems and flight simulators.

This book will help the reader understand the methods of modelling both

rigid and flexible aircraft for controller design application as well as gain

a basic understanding of the processes involved in the design of control

systems and regulators. It will also serve as a useful guide to study the simu-

lation of flight dynamics for implementing monitoring systems based on the

estimation of internal system variables from measurements of observable

system variables.

The book brings together diverse topics in flight mechanics, aeroelasticity

and automatic controls. It would be useful to designers of hybrid flight con-

trol systems that involve advanced composite structure–based components

in the wings, fuselage and control surfaces. The distinctive feature of this

book is that it introduces case studies of practical control laws for several

modern aircraft and deals with the use of non-linear model-based tech-

niques and their applications to flight control.

Chapter 1 begins with an introduction and reviews the configuration of a

typical aircraft and its components. Chapter 2 deals with the basic principles

governing aerodynamic flows. Chapter 3 covers the mechanics of equilib-

rium flight and describes static equilibrium, trimmed steady level flight, the

analysis of the static stability of an aircraft, static margins stick-fixed and

stick-free, modelling of control-surface hinge moments and the estimation of

the elevator angle for trim. Basic concepts of stability based on disturbances

to one parameter alone are discussed. The effects of a change in the angle of

xix

www.ebook777.com

free ebooks ==> www.ebook777.com

xx Preface

attack on the pitching moment and its application to stability assessment are

discussed. Also considered are steady flight at an angle to the horizontal and

the definition of flight path, incidence and pitch angles and the heading, yaw

and sideslip angles. The assessment of manoeuvrability and the application

of margins required for a steady pull-out from a dive are also introduced.

Chapter 4 is dedicated to the development of the non-linear equations of

motion of an aircraft, including simple two-dimensional dynamic models,

and the development of the aircraft’s equations of motion in three dimen-

sions. The general Euler equations of rigid body and the definition and

estimation of moments of inertia matrix are discussed. The definitions of

motion-induced aerodynamic forces and moments and the need for vari-

ous reference axes that are fixed in space, fixed to the body and fixed in the

wind as well as the definition of stability axes are clearly explained. The

non-linear dynamics of aircraft motion in the stability axes is derived both in

terms of body axis degrees of freedom and wind axis variables. The concept

of non-linear reduced order modelling is introduced, and the short period

approximation is discussed. Finally, the trimmed equations of motion as

well as the non-linear perturbation equations of motion are derived. The

concept of linearisation is also introduced, and the linear equations of air-

craft motion are briefly discussed. In Chapter 5, the small perturbation equa-

tions of motion are described in detail, and the equations are expressed as

two sets of decoupled equations representing the longitudinal and lateral

dynamics. Chapter 6 introduces the methodology of linear stability analysis

and provides a modal description of aircraft dynamics. The application of

small perturbation equations in determining the control setting angles for

executing typical manoeuvres is also discussed in this chapter.

Chapter 7 covers the evaluation of aircraft dynamic response and the

application of MATLAB®/Simulink® in determining the aircraft’s response

to typical control inputs. A basic introduction to aircraft non-linear dynamic

phenomenon is also presented in this chapter. Chapter 8 deals with aircraft

flight control, the design of control laws, stability augmentation, autopilots

and the optimal design of feedback controllers. Chapter 9 describes flight

simulators and the principles governing their design. Finally, Chapter 10 is

dedicated to the flight dynamics of elastic aircraft, including the principles

of aeroelasticity from an aircraft perspective.

I thank my colleagues and present and former students at the School of

Engineering and Material Science, Queen Mary University of London, for

their support in this endeavour.

I thank my wife Sudha for her love, understanding and patience. Her

encouragement and support provided me the motivation to complete this

project. I also thank our children Lullu, Satvi and Abhinav for their under-

standing during the course of this project.

Ranjan Vepa

London, United Kingdom

free ebooks ==> www.ebook777.com

Preface xxi

information, please contact:

3 Apple Hill Drive

Natick, MA 01760-2098 USA

Tel: 508-647-7000

Fax: 508-647-7001

E-mail: info@mathworks.com

Web: www.mathworks.com

www.ebook777.com

free ebooks ==> www.ebook777.com

free ebooks ==> www.ebook777.com

Author

Dr. Ranjan Vepa earned his PhD in applied mechanics from Stanford

University, Stanford, California, specialising in the area of aeroelasticity

under the guidance of the late Prof. Holt Ashley. He currently serves as a

lecturer in the School of Engineering and Material Science, Queen Mary

University of London, where he has also been the programme director of

the Avionics Programme since 2001. Prior to joining Queen Mary, he was

with the NASA Langley Research Center, where he was awarded a National

Research Council Fellowship and conducted research in the area of unsteady

aerodynamic modelling for active control applications. Subsequently, he

was with the Structures Division of the National Aeronautical Laboratory,

Bengaluru, India, and the Indian Institute of Technology, Chennai, India.

Dr. Vepa’s research interests include the design of flight control systems

and the aerodynamics of morphing wings and bodies with applications in

smart structures, robotics and biomedical engineering and energy systems,

including wind turbines. He is particularly interested in the dynamics and

in the robust adaptive estimation and the control of linear and non-linear

aerospace, energy and biological systems with uncertainties. The research

in the area of the aerodynamics of morphing wings and bodies is dedicated

to the study of aerodynamics and its control. This includes the use of smart

structures and their applications to the control of aerospace vehicles, jet

engines, robotics and biomedical systems. Other applications of this work

are to wind turbine and compressor control, maximum power point track-

ing, flow control over smart flaps and the control of biodynamic systems.

Dr. Vepa currently conducts research on biomimetic morphing and aero-

dynamic shape control and their applications, which include feedback con-

trol of aerofoil section shape in subsonic and transonic flow for unmanned

aerial vehicles (UAV), airship and turbomachine applications and integra-

tion of computational aeroelasticity (CFD, computational fluid dynamics/

CSD, computational structural dynamics) with deforming grids as well as

their applications to active flow control. Of particular interest are the bound-

ary layer instabilities in laminar flow arising due to various morphing-

induced disturbances. Dr. Vepa has also been studying the optimal use and

regulation of alternate power sources such as fuel cells in hybrid electric

vehicle power trains, modelling of fuel cell degradation and health monitor-

ing of aircraft structures and systems. With regard to structural health moni-

toring and control, observer and Kalman filter–based crack detection filters

are being designed and applied to crack detection and isolation in aeroelastic

aircraft structures such as nacelles, casings, turbine rotors and rotor blades.

Feedback control of crack propagation and compliance compensation in

cracked vibrating structures is also being investigated. Another issue is the

xxiii

www.ebook777.com

free ebooks ==> www.ebook777.com

xxiv Author

ysis of their plates and their interaction with unsteady aerodynamics. These

research studies are contributing to the holistic design of vision-guided

autonomous UAVs, which are expected to be extensively used in the future.

Dr. Vepa is the author of three books: Biomimetic Robotics (Cambridge

University Press, 2009), Dynamics of Smart Structures (Wiley, 2010) and

Dynamic Modeling, Simulation and Control of Energy Generation (Springer, 2013).

He is a member of the Royal Aeronautical Society, London; the Institution of

Electrical and Electronic Engineers (IEEE), New York; a fellow of the Higher

Education Academy; a member of the Royal Institute of Navigation, London;

and a chartered engineer.

free ebooks ==> www.ebook777.com

1

Introduction to Flight Vehicles

1.1 Introduction

While aerodynamics is the study of flows past and over bodies, the principles

of flight are governed by the dynamics and aerodynamics of flight vehicles.

The focus of this chapter is on the general principles of flight and the pri-

mary features of aircraft. Further details may be found in Anderson [1] and

Shevell [2]. As the aerodynamics of bodies is greatly influenced by their

external geometry, the aerodynamics of flight vehicles is entirely deter-

mined by their external geometry. The external geometry is in turn com-

pletely influenced by the entire complement of components external to the

vehicle. The basic architecture of a typical aeroplane, the simplest of flight

vehicles, is well known to any cursory observer of aeroplanes. It can be con-

sidered to be the assemblage of a number of individual components. The

principal external components are the fuselage, the left and right wings, the

power plant pods or nacelles, the tail plane unit comprising of the horizontal

and vertical stabilisers, the various control flaps and control surfaces and the

landing gear. When the components are assembled or integrated together, a

complete external picture of a typical aeroplane emerges. A typical planform

or t op-down view of an aeroplane is shown in Figure 1.1.

The primary components of an aeroplane are the fuselage, the wing, the

tail surfaces which are collectively referred to as the empennage, the power

plant, the various control surfaces used to control the flight of the aeroplane

and the landing gear.

1.2.1 Fuselage

The fuselage is the main body of any aeroplane, housing the crew and pas-

sengers or the cargo or payload and the like.

www.ebook777.com

free ebooks ==> www.ebook777.com

2 Flight Dynamics, Simulation, and Control

FIGURE 1.1

Typical planform view of an aeroplane.

1.2.2 Wings

The wings are the main lifting element of the aeroplane. They comprise of

the wing leading and trailing edges, flaps and slats that are used to aug-

ment the lift on the wing, ailerons to enable the aeroplane to bank while

turning and spoilers that are capable of reducing the wing lift during land-

ing and act as speed brakes. The high-lift devices controlled and oper-

ated below the wing permit the wing to develop the necessary lift during

take-off when a large passenger jet attains speeds of the order of 320 km/h

after accelerating down a runway of length 3–4 km. The controls and drive

mechanisms linking these devices are usually shrouded in canoe-shaped

fairings attached to the underside of the wing. The wing essentially carries

the entire aeroplane and all other associated systems. The wing is essen-

tially a single aerodynamic element although it extends symmetrically on

either side of the fuselage.

The tail surfaces are the basic elements that stabilise and control the aero-

plane. Normally, both the vertical and horizontal tail surfaces have a fixed

forward portion and a hinged rearward portion. The forward portion of the

horizontal tail surface is known as the stabiliser, while the rearward hinged

portion on the same surface is known as the elevator. On many long-haul air-

liners, the horizontal stabiliser is an all movable unit. On the vertical tail, the

fixed forward portion is known as the fin, while the hinged rearward portion

is known as the rudder. Both on the rudder and on the elevator are additional

hinged surfaces known as the trim tabs which are used to adjust the forces on

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 3

the pilot’s control column (which controls the movement of the elevator) and

rudder pedals so that these are force free. Together, the entire horizontal and

vertical tail surface assembly is known as the empennage.

To enable an aeroplane to operate from land, aeroplanes are provided with

landing gear comprising of wheels with types mounted on axles. Brakes

are integral elements while the axles are attached via supporting struts and

shock absorbers to the fuselage. To minimise drag during take-off and in

steady flight, cowlings and retractable mechanisms are provided. The lat-

ter permit the retraction of the entire landing gear to an enclosed housing

within the fuselage once the aeroplane is airborne.

1.3.1 Forces Acting on an Aeroplane

Consider the equilibrium of an aeroplane on the ground. Its weight may be

regarded as acting vertically downwards through the aeroplane’s centre of

gravity (CG) and this is balanced by two sets of reactions acting vertically

upwards, one at the points of contact of the main undercarriage and the

ground surface and the other either at the nose wheel or tail skid depend-

ing on the type of aeroplane. To maintain an aeroplane in vertical equilib-

rium during flight, the vertical reactions at the main undercarriage and nose

wheels must be replaced by equivalent upward forces: the lift components

acting on the main wing and tail plane surface. In the days of the lighter than

air balloons, which were axially symmetric about the CG axis, the reaction

was a single lift force due to the buoyancy. This force was due to the differ-

ence in the weight of the air displaced by the balloon and the gas contained

within and acted in the vicinity of the CG. However, with the arrival of the

airship, the forces were no longer acting in a single vertical line. Typically,

a steady level flight is held in balance or equilibrium by a combination of

forces (Figure 1.2a). The forces comprise

1. The lift on the aeroplane with the principal contributions being due

to the wing and horizontal tail

2. The drag which consists of two main components the profile drag

and the induced drag

3. The thrust produced by the power plants

4. The weight of the aeroplane

www.ebook777.com

free ebooks ==> www.ebook777.com

4 Flight Dynamics, Simulation, and Control

Wing

Relative wind lift, LW

Induced

drag, Di

Profile

drag, Dp

Tail

lift, LT

Thrust, T

(a) Weight, mg

Suction

Suction

Pressure

Pressure

(b)

FIGURE 1.2

(a) Forces acting on aeroplane in steady, level, equilibrium flight and (b) pressure distribution

on a wing: front and side view of a typical wing section.

In addition to the equilibrium of forces, the forces on the tail plane contrib-

ute principally towards rotational moments acting on the aeroplane. All the

rotational moments acting on the aeroplane must cancel each other to ensure

that the aeroplane is in rotational equilibrium. Rotational equilibrium is

essential so the aeroplane can maintain steady orientation during a long and

sustained flight. Thus, the attitude of the aeroplane must remain steady dur-

ing extended periods of flight.

The principal phenomenon that is responsible for holding the aeroplane in

flight is the wing lift which is caused as a result of the generation of a low-

pressure or suction region over the top surface of the wing and high-pressure

region below the lower surface of the wing (Figure 1.2b). The region of low

pressure on the top surface of the wing is caused by the flow of air over the

curved surface of the wing with a resultant increase in flow velocity and con-

sequent decrease in pressure relative to the rest of the atmosphere. Similarly,

the region of high pressure below the lower surface of the wing represents a

region where the pressure is relatively greater than in the surrounding air.

The result of these two complementary effects on the two surfaces of the

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 5

wing is the generation of lift. This generation is due to the fact that the two

flows emerging from the upper and lower surfaces at the trailing edge of the

wing result in a downwash or vortical flow. Thus, the wing experiences an

upward and opposite reaction in the form of lift.

The lift is directly proportional to the air density and also a function of the

airspeed; the higher the airspeed, the greater the lift generated by the wings.

An increase in the wing surface area increases the lift in direct proportion.

The wing camber and angle of attack are the other parameters that cause the

lift to increase.

As for the drag on the aeroplane, there are two distinct types of drag that

act to retard the aeroplane when it is forward flight. The first is profile drag

that is itself made up of two components, form drag and skin friction drag. The

former is produced due to the finite shape of the aeroplane as the result of

the streamlined flow around its body. Thus, the shape of the body is almost

always optimally streamlined to reduce this component to minimum. The

latter component is produced due to the viscous friction between the aero-

plane’s skin and the airflow around the body. The airflow results in the for-

mation of a thin boundary layer where the flow velocity reduces to zero as

one gets closer to the skin of the aeroplane. This type of drag depends to a

large extent on the thickness of the boundary layer that must be kept to a

minimum to reduce the drag. These aforementioned two components that

constitute the profile drag have one common feature: they both increase

markedly as the speed of the aeroplane increases and the increase is directly

proportional to the square of the airspeed.

The second type of drag experienced by an aeroplane is the induced drag.

Due to the pressure difference between the top and bottom side of the wing

surface, there is a spill over of air, particularly at the wing tips, from the

bottom to the top. To a large extent, the induced drag is caused by a meeting

of the airflow emerging from the upper and lower surfaces at the trailing

edge, at a finite angle, resulting in the formation of vortices, set up due to

the air spilling over. The vortices accumulate at the wing tips to produce a

rotating flow of air, rotating in the direction of the wing root and resulting

in a wing tip vortex. These wing tip vortices are the principal contributors

to the induced drag which is caused by the energy dissipated in rotating

the air. Due to the wing tip vortices being washed away at a faster rate at

higher airspeeds, there is a decrease in the induced drag with the increase

in the speed.

As a result of the different behaviours of the two types of drag as speed

increases, there is a speed at which the drag is in fact a minimum. At this

speed, the contributions to the total drag by the two types of drag are equal,

and as a result, either an increase or decrease in the airspeed causes the drag

to increase. Aeroplanes are generally flown at a cruise speed that is just above

www.ebook777.com

free ebooks ==> www.ebook777.com

6 Flight Dynamics, Simulation, and Control

the minimum drag speed as it is important to operate on the right side of the

drag curve. Operating on the wrong side implies that a small reduction in the

airspeed increases the drag substantially and unacceptably large increments

of power are required to increase the aeroplane’s speed. Operating on the

wrong side is not acceptable and unsafe especially when the power plant is

already being operated near its maximum power output.

Finally, the drag must be low and well below the thrust generated by the

aeroplane’s power plants or propulsive units to ensure that the aeroplane

may be accelerated fast enough as may be desired during various phases of

the flight.

There is a patent need to reduce the drag acting on an aeroplane. This is done

by shaping the envelope of the various components in the flow or streamlin-

ing. By appropriately shaping the envelope so that all directions tangential

to it are parallel to the directions of the flow adjacent to it, the drag could be

considerably minimised. In most cases, this is because air is able to smoothly

pass over the body generating any eddies or turbulence. The generation of a

turbulent wake behind the body could substantially increase the drag.

Streamlining is also necessary for the generation of lift. There are indeed

three effects that contribute to wing lift: (1) the shape of the aerofoil or wing

section is such that the velocity of the flow must necessarily be higher over

the upper surface than below the lower surface; (2) the velocity of the flow

field gives rise to a pressure differential or suction that is a principal contrib-

utor to the wing lift; and (3) there is the effect of the downward inclination of

the streamlines behind the aerofoil section, known as downwash, as well as

the slight upward inclination of the flow in the vicinity of the leading edge

or front of the aerofoil, known as the upwash. Together, the upwash and the

downwash are responsible for producing a curved streamlined flow with a

resulting inertia force acting outwards. This is a significant contributor to

the lift acting on the wing section.

The weight distribution on an aeroplane also plays a critical role in ensuring

a stable flight. By stability we mean the ability of the aeroplane to return to

its equilibrium orientation when disturbed by an external effect of any kind.

To ensure stability, it is essential that the CG of the aeroplane is sufficiently

forward. Thus, it is particularly important to ensure not only that the weight

is laterally balanced but also that the aeroplane is not too tail heavy.

Maintaining rotational balance is an important requirement in flight. Lift

and weight generally do not act at the same point during a particular flight

of an aeroplane. The centre of aerodynamic pressure can be expected to

change continually depending on the selection of control surfaces deployed

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 7

during the different phases of the flight. Moreover, the weight distribution

around the aeroplane is also changing due to variations in payloads and fuel

consumption. Fuel can account for up to 30%–45% of an aeroplane’s weight,

while in an airline, the total weight of the passengers and other payloads

could weigh as much as 15%–20% of the maximum take-off weight. Thus,

ensuring stability is a difficult proposition. The problem is overcome by

making the entire horizontal tail plane movable so it could be deployed as

a stabilising surface. The tail plane generates lift and as a result of its long

moment arm, it is adequate to restore the aeroplane to an equilibrium posi-

tion when a disturbing force acts at the CG. The movable or variable position

tail plane is used to rebalance the aeroplane and particularly to maintain

equilibrium when there are changes in the aeroplane’s weight and CG loca-

tion. Thus, when the CG is aft of the centre of pressure (CP), the aeroplane

is tail heavy and it is essential to stabilise the aeroplane. At this stage, the

stabiliser is moved up to decrease the lift on the tail unit and hence rebalance

the aeroplane. This process of balancing the aeroplane by movement of the

stabiliser is known as trimming. On the smaller general aviation aeroplane,

this function is performed by the trim tabs that are smaller movable control

surfaces hinged to the rear of the elevator and rudder. Aeroplanes that are

provided with trim tabs generally have fixed stabilisers. Some aeroplanes

are provided with both an all moving horizontal tail plane, for automatic

trim, and a full set of trim tabs for manual trimming. To be able to trim the

aeroplane, the pilot must have a feel of the out of balance forces. A feel unit

usually provides this feedback and the pilot usually feels the pressure of out

of balance forces on the control column. When the aeroplane is trimmed, the

control column is relieved of the out of balance feedback and is free of any

forces acting on it. Thus, the aeroplane may be flown in a stable condition

with hands off of the control column.

To understand the fundamentals of the dynamics of heavier than an aero-

plane, it is essential to first understand not only the basic principles of flight

but also its control. The aeroplane in level flight at constant speed can be

considered to be flying in equilibrium. The weight of the craft is completely

balanced by the lift generated by the wings of the aeroplane. The thrust

imparted to it by the engines is completely balanced by the drag. The lift is

generated by the flow of air over the surface of the wing that is designed to

have a special cross section. When the aeroplane loses speed, there is also

a loss of lift that must be compensated, if the aeroplane is to fly at constant

altitude. The aeroplane compensates the loss of lift by increasing its angle

of attack that results in an increased lift. However, there is a limiting angle

www.ebook777.com

free ebooks ==> www.ebook777.com

8 Flight Dynamics, Simulation, and Control

(a) (b)

(c) (d)

FIGURE 1.3

Flow separation and the onset of stall. (a) Flat plate aerofoil at 0° incidence, (b) flat plate a erofoil

at 15° incidence, (c) aerofoil at 13° incidence and (d) aerofoil at 18° incidence. Trailing edge

separation initiated.

(about 15°) beyond which any further increase in the angle of attack only

causes the aeroplane to lose lift due to flow separation over the upper surface

of the wing as illustrated in Figure 1.3. Consequently, the aeroplane stalls and

any further increase in the angle of attack or reduction in speed results in a

dramatic loss of lift. The speed at which this condition of stalling occurs is the

stalling speed that is always the same for a particular aeroplane.

The most dangerous moments in the flight of an aeroplane are during take-

off and landing. At these stages in the flight, there is demand for maximum

lift at low speeds. To generate additional lift during these low-speed stages of

the flight, the aeroplane is provided with high-lift devices such as retractable

flaps (the Fowler flaps) and movable slats in the leading edge region which can

effectively increase the curvature of the wing section or aerofoil and thus

generate the additional lift. After take-off, every effort is generally made to

reduce the aeroplane’s drag thereby increasing its flight speed. To do this, the

landing gear is retracted and held within the belly of the aeroplane, so the

shape of the aeroplane is apparently streamlined and the drag is minimised.

The flight of the aeroplane is controlled by means of the controllers within

the cockpit of the aeroplane: the control column, the throttle levers, the rud-

der pedals and the toe brakes. These controls allow a whole family of control

surfaces to be controlled indirectly using intermediate, electro-hydraulically

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 9

Spoilers edge flaps/slats

Flaps

Ailerons

Canoe fairings

Fin

Rudder Stabiliser

Elevator

FIGURE 1.4

The complete complement of controls on a typical aeroplane.

operated mechanisms, known as power control units. Figure 1.4 shows the

complete complement of controls on a typical aeroplane.

The control columns operate the elevator when moved fore and aft. When

the elevator moves down, the additional lift generated on the tail plane forces

the aeroplane to pitch nose downwards and vice versa. The elevators are

hinged at the trailing edge of the horizontal stabiliser. The elevators are gen-

erally operated by the power control units, but on most aeroplanes, there is

the option of manual reversion, so the pilot could, when necessary, take con-

trol and manually operate them. When operated by the power control units,

there is need for some form of artificial feel. The artificial feel is provided by

an actuator applying a force on the control column. The force is computed by

the feel computer which receives its inputs from the pilot, the static pressure

ports and the horizontal stabiliser setting.

The horizontal stabiliser’s function is to provide for longitudinal trim. This

is accomplished by changing the incidence angle of the horizontal stabiliser.

It may also be driven by an electro-hydraulic power control unit or manu-

ally by cables. On some aeroplanes, increase in the airspeed causes the CP

to move aft and the aerodynamic centre forwards causing the aeroplane to

tuck. In this state, the natural phugoid mode of the aeroplane is absent and

the aeroplane could come dangerously close to being unstable. To avoid this

behaviour, the horizontal stabiliser is sometimes fitted with an automatic

pitch-trim compensator. Horizontal stabilisers are generally set in motion

by switches on the pilot’s control column. Trimming may be achieved either

automatically or manually.

www.ebook777.com

free ebooks ==> www.ebook777.com

10 Flight Dynamics, Simulation, and Control

When the control column is moved or rotated from left to right, the aile-

rons at the far end of each of the two wings rotate in a differential manner,

thereby generating a rolling moment. Thus, the aeroplane banks, the angle of

bank being directly proportional to the differential moment of the ailerons.

Roll may not only be initiated by the ailerons but also be controlled by them.

On some aeroplanes, there are two ailerons on each wing. The outboard pair

is usually locked in with the wing in high-speed flight while they may be

proportionally controlled at low speeds. The outboard pair is not used when

the flaps are deployed. Like the case of the elevator, it is often possible to

revert to manual control and artificial feel is also provided. The artificial feel

unit is usually a spring loaded roller cam mechanism which is responsible

for providing a feedback force to the control column that is directly propor-

tional to the roll actuator command input.

The rudder pedals operate the rudder that generates the necessary

turning moment to turn the aeroplane. The rudder generally provides for

the control of yaw (nose right or nose left). Some aeroplanes are provided

with dual rudders, each of which is split into two separately actuated sec-

tions. To protect the vertical tail from structural damage that may result

from excessive rudder deflection, rudder travel is limited by incorporat-

ing signal limiters in the rudder control circuits. The rudder control sys-

tem also incorporates, most often, a yaw damper which receives inputs

from a yaw rate gyro and provides additional signals to the rudder power

control unit so as to move the aeroplane in the direction opposing the

yaw motion and in proportion to the yaw rate. The yaw damper is not

usually operational in the manual reversionary mode. An artificial feel

unit similar to the one fitted to the ailerons is also fitted to the rudder.

The toe brakes apply braking to the wheel assemblies on the respective

sides while allowing for differential braking to supplement the rudder on

the ground.

Various types of control tabs, balance tabs and differentially controlled

balance panels are also used in aeroplane control. These devices are gener-

ally used to balance the forces or moments acting on the control column in

the respective directions. This is achieved without adversely affecting the

control forces and moments generated by the main control surface and thus

maintaining the control column in a force-free condition. Thus, the tabs can

mechanically fly the elevator, aileron or rudder while effectively relieving

the pilot of having to provide a command input to the control column. The

pilot may then fly the aeroplane in the particular trimmed condition in a

hands-free mode. These controlling movements are illustrated in Figures 1.5

through 1.7.

There are a plethora of high-lift devices that may be used to improve

the lift characteristics of the aeroplane’s primary lifting surfaces during

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 11

Relative wind

(a)

Relative wind

(b)

FIGURE 1.5

Operation of the elevator. (a) Elevator down results in aeroplane nose down and (b) elevator up

results in aeroplane nose up.

FIGURE 1.6

Operation of the aileron: up aileron forces wing down and down aileron forces wing up, result-

ing in bank for turning left; aeroplane continues to turn left when ailerons are returned to the

normal position.

take-off or other phases of the flight. Broadly, all flow control devices fall

into five primary categories:

2. Single multi-element/multi-surface variable camber or deployable

systems

3. Blown or suction systems

4. Inflatable systems including leading edge devices

5. Active/passive vortex and circulation control systems

www.ebook777.com

free ebooks ==> www.ebook777.com

12 Flight Dynamics, Simulation, and Control

FIGURE 1.7

Operation of the rudder: moving the rudder to the left turns the aeroplane to the left and

vice versa.

High-lift devices are usually deployed to increase the lift force. However,

there is also a substantial increase in the drag accompanying any increase

in lift. During take-off, an increase in the lift is generally required to reduce

the unstick speed and take-off run, while during landing, there is need to

reduce the landing speed and to reduce the landing run. Thus, the increase

in drag can be effectively optimised in reducing the take-off and landing

runs. One of several short chord/short span passive devices is available to

reduce the wing lift over sections of the wing surface to achieve flow control.

Although there are several methods available to increase the wing lift, single

control surfaces or multi-element/multi-surface variable camber or deploy-

able systems are normally used on most aeroplanes. These generally offer

almost negligible resistance when they are not deployed and their deploy-

ment is completely controlled by the pilot.

Wing leading edge deflection, at high angles of attack, is essential to

impede stall, thus enabling to attain higher angles of attack thus generating

greater lift. Effectively, the leading edge deflection of the wing results in an

increased curvature of the wing section. This is achieved by a combination

of slats, slots and flaps (Figures 1.8 and 1.9).

The very thin layer of air in which the velocity is gradually increasing from

zero to that of the airstream is called the boundary layer. Viscous friction plays

an important part in its evolution and typically the boundary layer affects

the streamline flow, which is outside it. The separation of the boundary layer

from the surface of the wing can result in an extreme loss of lift. Boundary

layer separation due to adverse pressure gradients on lifting surfaces due to

high angles of attack or due to transonic shock effects is the primary cause for

the flow separation followed by a loss of lift. Boundary layer separation also

causes an increase in the drag. Thus, there is an increased demand for fuel

and loss of performance. The unsteady flow associated with separation leads

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 13

(a) (b)

(c) (d)

(e) (f)

FIGURE 1.8

Typical complement of trailing edge high-lift flaps. (a) Plain hinged flap, (b) slotted flap,

(c) double-slotted flap, (d) Wragg or external aerofoil flap, (e) split flap and (f) fowler flap (which

is moved down to the rear).

FIGURE 1.9

Handley page leading edge slat (which is pulled out into place by suction at high angles of

attack).

buffeting. There are several techniques used to control boundary layer sepa-

ration and these are

1. Vanes

2. Flow control rails

3. Boundary layer blowing

4. Boundary layer suction

5. Vortex mixing

6. Passive control of boundary layer

7. Control of wing camber and thickness

8. Active control techniques

Openings in the vicinity of the leading edge wing surfaces allow the flow

of air, with a higher energy, into the boundary layer of the upper surface

to blow it off and inhibit the separation of airflow at that angle of attack.

www.ebook777.com

free ebooks ==> www.ebook777.com

14 Flight Dynamics, Simulation, and Control

a much higher angle of attack. Thus, the result is an increase in the effective

lift coefficient. The slots in the vicinity of the leading edge wing surfaces are

hydraulically opened only when the trailing edge laps are down and auto-

matically closed when the flaps are up. Leading edge segments that move on

tracks and extend from the wing leading edge to form slots are essentially

movable slots. They are known as slats and produce the same effect as fixed

slots. Slats are also hydraulically operated and the deployment and exten-

sion of slats is usually synchronised with the deployment and extension of

trailing edge flaps. The coordinated movement of slots, slats and trailing

edge flaps is designed to effectively increase the camber of the wing and

thus improve the wing characteristics at low flight speeds. Leading edge

flaps, which can give the wing an additional droop when extended, may

also be deployed to produce the same effect. The deployment of leading

edge flaps, known also as Krueger flaps, is also automatically synchronised

with the deployment of trailing edge flaps by an electrically signalled and

hydraulically operated power control unit.

The deployment of trailing edge flaps is controlled by a flap handle that is

located on the pilot’s control pedestal in the cockpit. Earlier forms of trailing

edge flaps were usually split flaps although the use of plain flaps and exten-

sion flaps (Fowler flaps) is now widespread. In one form, trailing edge flaps

are usually deployed in a two-section configuration, which are designated

as the inboard and outboard sections. Each of the inboard and outboard

sections is independently signalled electrically and can be programmed to

operate, symmetrically, in one of several coordinated schedules. In many of

the older aeroplanes, the coordination of the inboard and outboard sections

and the symmetric operation of the left and right wing flaps is achieved by

mechanical torque tubes and cabling.

Wing lift may also be regulated by controlling the airflow over the wing.

Typically, a narrow jet of air passing between the wing and trailing edge

flaps blows off the boundary layer, thus providing for attached flow and

consequently a higher lift coefficient. Theoretically, the most advantageous

methods are the boundary layer blowing off and suction from the upper

surface of the wing. Suction increases the rate flow, and consequently, there

is an increase in the rarefaction close to the wing surface in the region ahead

of the suction point. By contrast, the effect of blowing is an increase in the

rarefaction close to the wing surface over the entire chord. With boundary

layer blowing or suction, the wing drag decreases with increasing lift coef-

ficient and consequently there is an increase in lift–drag ratio.

A jet flap is another means of increasing the lift force. It is essentially estab-

lished by blowing air through a special slot in the trailing edge of the wing,

at angle to the extended chord line. The jet flap extends the wing virtually as

well as its camber and there is an increase in the total lift force acting on the

wing. The magnitude of the pressure distribution in the vicinity of the trail-

ing edge area is usually substantially greater than a wing without the jet flap.

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 15

A hybrid boundary layer suction system coupled with a jet flap is considered

to be a promising high-lift generating system.

Inflatable wings are particularly suitable for compensating the aeroplane’s

wing section for the in-flight ice accretion process. Ice accretion is particu-

larly a problem in the vicinity of the aeroplane’s leading edge, and compen-

sation is achieved by designing inflatable and deflatable wing leading edges.

During the ice accretion period, an active controller is used to deflate the

leading edge and thus compensate for the ice accretion.

A circulation control system employs rearward tangential blowing over

a rounded or near-rounded trailing edge, to reinforce the boundary layer

and delay the separation. Separation is delayed due to the flow remaining

attached to the trailing edge due to the Coanda effect. The location of the

separation point may be controlled by varying the blowing rate, thus affect-

ing the wing lift. Generally in the case of flaps with circulation control, there

is substantial increase in wing lift than in the case of conventional mechani-

cal flaps. A similar approach is adopted in the wings with vortex control jets.

There are indeed several alternate methods of controlling and regulating

wing lift. In the case of most high-speed jets, particularly those capable of

flying faster than the speed of sound, wings are swept back to minimise

drag. Yet it is well known that swept wings do not perform as well as straight

wings at lower speeds. Thus, in the case of swing wing aircraft, the wings

are movable and may be deployed as swept wings at high speeds with the

ability to revert to a straight wing configuration at low speeds.

There are also some vertical take-off and landing aircraft where the

aircraft’s lift is controlled entirely by control jets. The jet’s nozzle can be

mechanically swivelled and the jet’s exhaust directed accordingly, to alter

the direction of the net thrust acting on the aeroplane. Fluidic jets, where

the jet’s directional control is based on the so-called Coanda effect, have also

been employed in some experimental programmes.

1.4.3 Spoilers

Spoilers, so called because they are employed to spoil the lift on the wing by

disrupting the streamlined airflow around it, are usually deployed at the

instant of landing to place the full weight of an aircraft on the wheels and

prevent it from bouncing back into the air after a heavy landing. They are also

deployed automatically on an abandoned take-off following the selection of

reverse thrust, again to place the full weight of the aircraft on the wheels and

to improve braking performance. In-flight spoilers are deployed as speed

brakes to slow the aircraft rapidly and to greatly increase the rate of descent

(Figure 1.10). They are also employed occasionally for enhanced roll control.

Deploying the spoilers on one side of the aircraft disrupts the lift on that side

and aids the aircraft in rolling.

Spoilers are normally actuated by electro-hydraulic power control units.

On most civil aircraft, there are a number of spoilers and groups of these are

www.ebook777.com

free ebooks ==> www.ebook777.com

16 Flight Dynamics, Simulation, and Control

FIGURE 1.10

Aircraft with spoilers deployed: spoilers function as lift dumpers or speed brakes.

fault tolerance. During operation they are designed to be adaptive; that is, the

extension is generally much lower at higher speeds. Spoiler actuators are also

designed to retract back to their unloaded position when hydraulic power to

them is lost. In the case of flight spoilers, which are used to supplement the

aileron in roll control, the spoiler inputs are generated by the aileron’s move-

ment and moderated by a spoiler mixer mechanism or a spoiler control law.

Ground spoiler actuators are normally activated only while the aircraft are

on the ground and are controlled so the entire weight of the aircraft acts on

the landing gear just before touchdown.

and Yoke, the Rudder Pedals and the Toe Brakes

A primary complement of the pilot’s control in the cockpit are the throttle

levers to control the fuel delivered to the power plant, the control column

which may be pulled back or pushed forwards to rotate an aeroplane or to

flare the aeroplane during landing, the yoke which when turned banks the

aeroplane to one side or the other, the rudder pedal that is used to change

the direction of the aeroplane’s flight path and toe brakes which allow for the

differential braking of the wheels during landing.

Speed and power are intimately connected with changes in attitude or the

change in the direction in which the nose is pointing relative to the direction

of flight. Vertical changes in the direction of flight as well as the changes in

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 17

is called steady and we have already dealt with steady horizontal flight. The

simplified force diagrams for steady flight other than horizontal are shown

in Figure 1.11. These represent climbing, power gliding and gliding flight. In

addition, there are also the cases of an aircraft, in steady spin, in a terminal

velocity dive, climbing in a turn and gliding in a turn.

Wing

Relative wind lift, LW

Induced

Profile drag, Di

drag, Dp

Tail

lift, LT

Thrust, T

Weight, mg

(a)

Wing Induced

Relative wind lift, LW drag, Di

Profile Tail

drag, Dp lift, LT

Weight, mg

Thrust, T

(b)

Wing Induced

lift, LW drag, Di Tail

Relative wind

Profile lift, LT

drag, Dp

Weight, mg

(c)

FIGURE 1.11

Modes of flight. (a) Climbing flight, (b) power gliding and (c) gliding.

www.ebook777.com

free ebooks ==> www.ebook777.com

18 Flight Dynamics, Simulation, and Control

We observe from Figure 1.11 that the direction of the airflow relative to

the aeroplane is exactly opposite to the direction of motion of the aeroplane.

The air itself is not moving and it only has velocity relative to the aeroplane.

The direction of airflow is important as it determines the directions of the

lift and drag.

Based on the figure, we may establish that the conditions for equilibrium

flight may be obtained, as in the case of steady level flight, by resolving the

forces in the directions of the lift and drag.

The problem of stability has already been discussed. Yet the overall sta-

bility of an aeroplane is particularly important and in large passenger

aeroplane a good deal of stability is desirable. An important feature of

these aeroplanes is the inherent stability in the three aeroplane attitudi-

nal degrees of freedom of pitch, roll and yaw as well the static stability in

equilibrium flight.

As already mentioned, the tail plane generates lift, and as a result of its

long moment arm, it is adequate to restore the aeroplane to an equilibrium

position when a disturbing force such as a gust of wind acts to displace the

aeroplane from its equilibrium position. A measure of this characteristic is

the distance of the aerodynamic centre, the location of the CP of all aerody-

namic forces generated when the aeroplane pitches forwards or backwards

from a position of equilibrium and the CG. This is known as the longitudinal

static stability margin.

Stability in roll is achieved due to the dihedral construction; that is, each half

of the wing is positioned at a small positive angle (4°–10°) to the h orizontal.

Thus, when the aeroplane rolls to one side, there is an increased lift on the

corresponding side of the wing resulting in a restoring moment and the

aeroplane returns to a state of equilibrium.

Stability in yawing motion is due to the tail fin. It plays the same role in

yawing motion as the horizontal tail plane does in pitching motion. Similar

to longitudinal static stability margins, one could define lateral static stabil-

ity margins. The lateral or weathercock stability margin is essential to pro-

vide the aeroplane with directional stability.

The aforementioned stability characteristics refer to the desirable static

stability margins of an aeroplane. In addition, an aeroplane must possess

certain dynamic stability characteristics; that is, although an aeroplane may

return to state of equilibrium from a disturbed position, certain motion

characteristics are essential during its return to equilibrium. These desirable

motion features are observed when the aeroplane has acceptable dynamic

stability margins that are equally important, if not more, than the static sta-

bility margins.

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 19

Thus far, we have generated a thrust by drawing an arrow in the direction

of thrust and indicated it by the letter T, but the production of thrust in real-

ity is a different matter. A forward force can only be generated by pushing

a quantity of air back, that is, by increasing the velocity of the relative air-

flow. The thrust produced is directly proportional to this increase in the rela-

tive air velocity. Thus, thrust is produced whenever energy is imparted to a

stream of air. Without exception, all powered aircraft are propelled by one

or more thrust-producing thermal engines that convert heat energy released

by fuel combustion into mechanical power. Thrust-producing power plants

used on board an aircraft may be typical supercharged piston engines driv-

ing a propeller or one of a variety of jet engines. The former class of engines

is used for the smaller purpose-built general aviation aeroplanes, while the

latter class is used on most airliners. These power plants are mounted on the

aeroplane in one of several ways such as on the wings inside specially built

enclosures known as nacelles, on the tail plane, mounted on but external to

the fuselage or integrated into the fuselage. As many as six of these power

plant units may be used to propel a single aeroplane.

These aircraft are primarily driven by typical supercharged piston engines

driving a propeller. The propeller itself is constructed just like a wing of

constant chord and a very high span to chord ratio, a uniform twist in the

spanwise direction. It acts like a screw winding its way through the air, and

the velocity of air relative to each part of the blade will be directed like a

screw thread. The blade is designed such that the aerofoil sections along the

span are inclined at the appropriate angle attack to the net airflow, and con-

sequently, the lift components at each section will combine constructively

to produce the thrust in the direction of motion of the aircraft, while the

drag components combine to form a resisting torque. When this total resist-

ing torque is less than torque of the engine, the engine speed will continue

to increase. Consequently, there is an increase in the resisting torque, and

when this torque balances the engine torque, the equilibrium engine speed

is attained. The equilibrium engine speed and the corresponding thrust

determine the conditions of equilibrium flight.

Jet propulsion is based on the production of thrust by means of the reaction

of the force due to a rapid change in momentum of a jet of gas produced

within the aircraft but directed rearwards. It is usually associated with gas

www.ebook777.com

free ebooks ==> www.ebook777.com

20 Flight Dynamics, Simulation, and Control

possible the heat energy of the fuel into mechanical motion by causing a

shaft to rotate. When this is coupled with a fan, it will draw a large mass of

air through the aircraft where it expands and thus gain in kinetic energy.

However, when the gas expands, not only is there a fall in the pressure

accompanied by an increase in the kinetic energy, but there is also a fall

in the temperature. To ensure that the combustion chamber, to which this

charge of fuel–air mixture is delivered, functions efficiently, the tempera-

ture of the gas must be within certain limits and not fall too low. For this

reason, the air delivered to the turbine is pre-compressed by a rotary com-

pressor, fitted in front of the turbine on a common shaft. After burning in

the combustion chamber where the gases acquire further energy, a jet of hot

gas directed rearwards is produced, which in turn generates the desired

thrust. Turbo-jet propulsion is particularly adaptable to aircraft, due to its

high power to weight ratio, small size, the presence of a minimal number of

moving parts, better performance at high speeds coupled with cheaper fuel

cost and the ability to redirect pats of the jet for de-icing and flow (bound-

ary layer) control. It has replaced the supercharged rotary petrol engine

driving a propeller, as the main power plant on almost all large airliners. In

fact, currently, research is already way so such propulsion systems could be

employed for purposes of controlling an aircraft, in lieu of the usual control

surfaces such as spoilers, flaps, elevators and ailerons.

Typically about 50% of a modern airliner’s cost is contributed by the entire

complement of avionics onboard. Generally, this can be classified into the

following three groups:

gation and guidance, transponders, radar, audio, autopilots, displays

and indicators, etc.

2. Cockpit instrumentation and supporting electronics

3. Supporting electronics integrated with other subsystems such as

power plants and FADEC Avionics, landing gear and brakes, flight

control systems, fuel control systems, hydraulics systems, electrical

and power systems, lighting systems and cabin systems.

generic features is well beyond the scope of this introductory section.

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 21

The primary lifting surfaces in an aeroplane that is responsible to a large

extent in maintaining the aeroplane in flight at a reasonable altitude are

the two wings attached to the fuselage. It is quite natural to expect that the

geometry of these wings, the planform shape and the cross-sectional geom-

etry in particular plays a crucial role in the generation of the lift, which in

turn is the primary force on the aeroplane that is responsible in keeping

it aloft. The aerofoil itself is the envelope of the cross section of the wing.

It essentially consists of a structural framework covered by a thin metallic or

composite skin. While the structural framework gives the wing the required

strength and stiffness, the skin is primarily responsible in shaping the aero-

foil to match a prescribed aerofoil contour.

Aerofoil geometries have evolved over the years and there are now a number

of standardised aerofoil section geometries. Examples of typical symmetric

and unsymmetric aerofoil sections are illustrated in Figure 1.12.

A principal characteristic of any aerofoil section is the chord line or chord and

it defines the length of the aerofoil. It is a line drawn from the leading edge

(a)

(b)

FIGURE 1.12

Symmetric and unsymmetric aerofoil sections. (a) Two examples of symmetric aerofoil sections

and (b) two examples of unsymmetric aerofoil sections.

www.ebook777.com

free ebooks ==> www.ebook777.com

22 Flight Dynamics, Simulation, and Control

Chord

(a)

Mean camber

(b)

FIGURE 1.13

(a) Location of chord line and the definition of aerofoil chord and (b) location of mean

camber line.

of the aerofoil near its nose to the trailing edge of the aerofoil (Figure 1.13a).

Whether it falls totally within the aerofoil section or partially outside it, it

is a primary reference for defining the various ordinates of the upper and

lower surfaces of the aerofoil. It is normally designated by the lowercase

letter ‘c’.

1.9.3 Camber

The upper and lower surfaces of an aerofoil are known as the upper and lower

cambers. The distance halfway between the upper and lower camber line is

known as the mean camber line. The mean camber line plays and significant

role in the generation of lift and is a key parameter in determining the mean

value of the section aerodynamic lift force under steady flow conditions. The

maximum camber is the maximum distance of mean camber line from the

chord line. Its magnitude and location along the chord are usually expressed

as percentages of the chord. Typically, the magnitude is usually of the order

4% in the case of non-symmetric aerofoils and located at about 30% down-

stream from the leading edge (Figure 1.13b).

The leading edge is located at the forward tip of the aerofoil and the aerofoil

chord. A circle drawn with its centre on the mean camber line and a radius

so it passes through the forward most tip of the chord line is essential in

locating the leading edge. The leading edge radius and the coordinates of its

centre are used to define the leading edge circle.

The trailing edge is defined in a far more simpler way and is the point

where the upper and lower camber lines intersect. Although it is apparently

represented as a knife edge, it is in fact a region characterised by a finite

thickness depending on the thickness of the skin used to envelope the wing

structure.

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 23

The methodology for defining the shape of aerofoils has evolved over many

years. Its early development was done exclusively at the U.S. National

Advisory Committee on Aeronautics, (NACA) Langley Field Laboratory, by

a number of contributors. Here, the NACA 4-digit, modified 4-digit, 5-digit,

6-digit and the 6A series of aerofoils were developed over a period spanning

about 50 years.

For example, the NACA 4-digit aerofoils are defined as

NACA MXTT

where

M is the maximum value of the mean line in hundredths of chord

X is the chordwise position of the maximum camber in tenths of the chord

TT is the maximum thickness ratio (t/c) in per cent chord

The NACA 2410 refers to a 10% maximum thickness aerofoil, with maximum

value of the camber of 0.02 at x/c = 0.4.

In the case of the NACA 5-digit aerofoil (e.g. 23,015), the following applies:

First digit: Twenty-thirds (20/3) times the design lift coefficient. It’s also

safe to say that it represents the maximum height of the camber line

expressed as a percentage of the aerofoil chord length.

Second and third digits combined: The horizontal location of the maxi-

mum camber line height in 200th of the chord length. Also, if the third

digit is 0, then the trailing camber line is a straight line. If it is equal

to 1, then the trailing camber line is reflex, or bowed down.

Last two digits combined: The maximum thickness of the aerofoil

expressed as a percentage of the aerofoil chord length. It occurs at

about 30% of the chord length with the NACA 5-digit series.

The NACA 6-series aerofoils are designed for laminar flow unlike the

NACA 4- or 5-digit aerofoils. In these aerofoils,

Second digit: The horizontal location of the minimum pressure coef-

ficient (i.e. maximum suction from the accelerated air) in 10th of a

chord length for the symmetrical uncambered shape

Third digit: Tells us the approximate design lift coefficient for that aero-

foil in 10th

Last two digits combined: The maximum thickness of the aerofoil

expressed as a percentage of the aerofoil chord length

Further details on the NACA series of aerofoils may be found in Abbott and

von Doenhoff [3].

www.ebook777.com

free ebooks ==> www.ebook777.com

24 Flight Dynamics, Simulation, and Control

Employing the thin aerofoil theory, we can show that a simple cubic camber

line can be expressed in terms of the zero-lift angle of attack and the moment

coefficient of the moment about the leading edge as

⎛ x ⎞⎛ 3 ⎛ C ⎞⎛ 4x ⎞ ⎞

zc = −2α L = 0 x ⎜ 1 − ⎟ ⎜ 1 + ⎜ 1 − mLE ⎟ ⎜ 1 − ⎟ ⎟ (1.1)

⎝ c ⎠⎝ 10 ⎝ α L=0 ⎠ ⎝ 3c ⎠ ⎠

where

αL=0 = −CLi/2π is the zero-lift angle of attack (i.e. CL = 2π(α − αL = 0) = 0)

CLi is the coefficient of the lift due to the aerodynamic pressure distribution

around the aerofoil surface when the angle of attack is zero

CL is the coefficient of the lift due to the aerodynamic pressure distribution

around the aerofoil surface (i.e. CL = 2π(α − αL = 0))

CmLE is the coefficient of the moment of the aerodynamic pressure distribu-

tion about the leading edge

mean camber line. The shape of the mean camber lines of NACA 4-digit

aerofoil sections can be expressed analytically as two parabolic arcs, tangent

at the position of maximum mean-line ordinate. The equations defining the

mean lines are taken as

zc m⎛ x⎞x x

= − 2 ⎜ 2q − ⎟ , ≤ q, (1.2a)

c q ⎝ c⎠c c

2

zc m ⎡ 2 ⎛x ⎞ ⎤ x

=− 2 ⎢ ( 1 − q ) − ⎜ − q ⎟ ⎥, ≥ q (1.2b)

c (1 − q ) ⎢⎣ ⎝c ⎠ ⎥⎦ c

where m is the maximum ordinate of the mean line expressed as a fraction

of the chord while q is the chordwise position of the maximum ordinate. For

example, for the NACA 6400 aerofoil, m = 0.06 and q = 0.4, and for the NACA

4400 aerofoil, m = 0.04 and q = 0.4.

The distance between the upper and lower camber lines is the thickness and

it varies along the chord. A typical aerofoil thickness distribution (NACA

4-digit aerofoil) is given by

2 3 4

⎛ x x ⎛x⎞ ⎛x⎞ ⎛x⎞ ⎞

zt = 5t ⎜ a0 − a1 − a2 ⎜ ⎟ + a3 ⎜ ⎟ − a4 ⎜ ⎟ ⎟ (1.3)

⎜

⎝ c c ⎝c⎠ ⎝c⎠ ⎝ c ⎠ ⎟⎠

where the coefficients a0, a1, …, a4 are listed in Table 1.1.

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 25

TABLE 1.1

Coefficients Defining Thickness Distribution

of NACA 4-Digit Aerofoil

a0 a1 a2 a3 a4

0.2969 0.1260 0.3516 0.2843 0.1015

The maximum thickness is at x/c = 0.3 and the leading edge radius and the

included angle at the trailing edge are

2

⎛t⎞ ⎧ ⎛ t ⎞⎫

rLE = c × 1.1019 ⎜ ⎟ , δTE = 2 tan −1 ⎨1.16925 ⎜ ⎟ ⎬ . (1.4)

c

⎝ ⎠ ⎩ ⎝ c ⎠⎭

The equations defining the upper and lower surface are then given as

xu x zt zu zc zt

= − sin θ, = − cos θ (1.5a)

c c c c c c

and

xl x zt zl zc zt

= + sin θ, = + cos θ, (1.5b)

c c c c c c

θ = −α (1.5c)

differentiating the equation for the camber line. Hence,

dzc 2m ⎛ x⎞ x

tan α = = − 2 ⎜q− ⎟, ≤ q, (1.6a)

dx q ⎝ c⎠ c

dzc 2m ⎛ x⎞ x

tan α = =− 2 ⎜q− ⎟, ≥ q. (1.6b)

dx (1 − q ) ⎝ c ⎠ c

www.ebook777.com

free ebooks ==> www.ebook777.com

26 Flight Dynamics, Simulation, and Control

The shape of the mean camber lines of the NACA 16-series aerofoil section

is given as

zc C ⎡⎛ x⎞ ⎛ x ⎞ x x⎤

= − Li ⎢⎜ 1 − ⎟ ln ⎜ 1 − ⎟ + ln ⎥ (1.7a)

c 4 π ⎣⎝ c⎠ ⎝ c⎠ c c⎦

and

dzc CLi ⎡ ⎛ x⎞ x⎤

= − ln ⎜ 1 − ⎟ + ln ⎥ . (1.7b)

dx 4π ⎢⎣ ⎝ c ⎠ c⎦

Apart from the geometry of an aerofoil section, there are a number of other

characteristic features of wings that are also extremely important in the

development of the aerodynamic forces and moments on an aeroplane. These

include wing span, root chord, tip chord, mean geometric chord and mean

aerodynamic chord, planform area and wing aspect ratio. Certain angles asso-

ciated with the geometry of wings also play an important part in the aerody-

namics of wings. These are the incidence, sweepback and dihedral angles.

The principal characteristic of a typical section of an aerofoil is its chord.

Considering a general wing planform, the horizontal distance between

the wing tips is the wing span. The root of the wing is where the wing is

attached to the fuselage and is normally different from the fuselage cen-

treline. In many practical situations, the two are relatively located so close

to each other that they are assumed to be at the same location, spanwise.

The aspect ratio is defined as the ratio of the square of the span to the refer-

ence area (usually the area of the planform but sometimes could include

the planform area of the horizontal tail plane or the horizontally projected

area of the fuselage). The mean geometric chord is the ratio of the area of

the wing planform to the span. Various integral properties of general wing

planforms including the mean aerodynamic chord are defined in Table 1.2.

Most wing planforms are trapezoidal shaped as illustrated in Figure 1.14.

The leading and trailing edges of a typical trapezoidal planform may be swept

backwards or forwards. They play a significant role in determining the maxi-

mum lift on the aerofoil and hence the stall characteristics of the aeroplane.

The principal geometrical relationships associated with trapezoidal symmet-

ric planforms are tabulated in Table 1.3. The upward slope of the wing when

viewed from the wing root is known as the dihedral or dihedral angle. The dihedral

angle is essential as it is principally responsible in making the aeroplane suffi-

ciently stable in roll. It is usually of the order of about 5°–10°. In the case of many

modern planforms, which are also kinked, the dihedral angles corresponding

to the inboard and outboard sections of the planform can be different.

A typical example of a kinked planform is illustrated in Figure 1.15. The

principal geometrical relationships associated with kinked-trapezoidal sym-

metric planforms are tabulated in Table 1.4.

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 27

TABLE 1.2

Integral Properties of General Symmetric Planforms

Property Relation

s

S = 2 c ( y ) dy

Planform area, S

∫

0

s

2 2

Mean aerodynamic chord, c c= c ( y ) dy

S ∫

0

s

2

Mean geometric chord, c c= c ( y ) dy

b ∫ 0

s

2 ⎛ c(y) ⎞

xcen = c(y)⎜ + xLE ( y ) ⎟ dy

x position of centroid of area, xcen

S ∫

⎜ 2

⎝

0

⎟

⎠

s

2

y cen = c ( y ) ydy

Spanwise position of mean geometric chord

S ∫

0

s

2

xLEcen = xLE ( y ) c ( y ) dy

Leading edge position of mean chord

S ∫

0

b 2

Aspect ratio AR =

Sref

Sref

Reference chord cref =

b

cT

Taper ratio, λ, c0 is the chord at the centreline λ=

c0

Notes: c, chord; s, semi-span; b, span = 2s.

Fuselage

centreline

Λ = Leading-edge

sweep angle

cR

Root

c(y) Tip

cT

yR

y

Semi-span = s

FIGURE 1.14

Typical trapezoidal planform.

www.ebook777.com

free ebooks ==> www.ebook777.com

28 Flight Dynamics, Simulation, and Control

TABLE 1.3

Properties of Trapezoidal Symmetric Planforms

Property Relation

Leading edge line xLE (y) = xLE (0) + y tan ΛLE (y)

Trailing edge line xTE (y) = xTE (0) + y tan ΛTE (y)

Local chord (

c ( y ) = c0 1 − (1 − λ ) η )

4 ⎡ ⎛ 1 − λ ⎞⎤

Sweepback at any element line, n, m, fractions of tan Λ n = tan Λ m − ⎢( n − m ) ⎜ ⎟⎥

local chord AR ⎣ ⎝ 1 + λ ⎠⎦

Sweepback at any element line in terms of leading tan Λn = (1 − n)tan ΛLE + n tan ΛTE

edge and trailing edge sweep

Planform area S = s × c0 × (1 + λ)

c +c

Average chord cave = T 0

2

S

Mean geometric chord c=

b

⎛ 2 ⎞ S

Centreline chord c0 = ⎜ ⎟

⎝ 1 + λ ⎠ AR

2c0 ⎛ 1 + λ + λ2 ⎞

Mean aerodynamic chord c= ⎜ ⎟

3 ⎝ 1+ λ ⎠

⎛ d⎞

Chord at fuselage junction, fuselage diameter = d cR = c ( y d ) = c0 ⎜ 1 − (1 − λ ) ⎟

⎝ b⎠

s ⎛ 4 ⎞

Aspect ratio AR = ⎜ 1+ λ ⎟

c0 ⎝ ⎠

s

2 b ⎛ 1 + 2λ ⎞

y position of centroid of area, y cen = c ( y ) ydy y cen =

S ∫

0

6 ⎜⎝ 1 + λ ⎟⎠

⎛ 1 + 2λ ⎞

Leading edge location at above spanwise position, xLEcen xLEcen = xLE0 + c0 ⎜ ⎟ AR tan Λ LE

⎝ 12 ⎠

Chord at y = ycen c

c

x position of centroid of area, xcen xcen = xLEcen +

2

2λ 2 − λ − 1

Spanwise position of mean aerodynamic chord y mac = s

(

3 λ2 − 1 )

Leading edge position of mean aerodynamic chord xLE(ymac) = xLE(0) + ymactanΛLE(ymac)

s

2 b ⎛ 1 + 2λ ⎞

Side-slip force moment arm y cen = c ( y ) ydy y cen =

S ∫

0

6 ⎜⎝ 1 + λ ⎟⎠

s

2 b ⎛ 1 + 3λ ⎞

Rolling moment arm y p = c ( y ) y 2 dy yp =

Ss ∫

0

12 ⎜⎝ 1 + λ ⎟⎠

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 29

Centreline

Kink

Tip

FIGURE 1.15

Kinked-trapezoidal planform.

TABLE 1.4

Properties of Kinked-Trapezoidal Symmetric Planforms

Property Relation

Tip chord λ2c0

Local chord at kink λ1c0

⎛ η⎞

Chord at any spanwise location c ( y ) = c0 ⎜ 1 − (1 − λ1 ) ⎟ , 0 ≤ η ≤ η1 ,

⎝ η1 ⎠

⎛ 1− η ⎞

c ( y ) = c0 ⎜ λ 2 + ( λ1 − λ 2 ) ⎟ , η1 ≤ η ≤ 1

⎝ 1 − η1 ⎠

c0

Mean geometric chord c=

2

(

λ1 + λ 2 + ( 1 − λ 2 ) η1 )

( 2

) 2

(

2c0 1 + λ1 + λ1 η1 + λ1 + λ 2 + λ1λ 2 ( 1 − η1 )

2

)

3 (

λ1 + λ 2 + ( 1 − λ 2 ) η1 )

s

1 ⎛ λ − λ 2 ⎞ 1 − η1k

(k − 1)th moment arm, y k −1 =

c0 sk ∫ c(y) y

0

k −1

dy y k −1 = ⎜ λ 2 + 1

⎝

⎟

1 − η1 ⎠ k

λ1 − λ 2 1 − η1k +1 1 + λ1k

− + η1k

1 − η1 k + 1 k ( k + 1)

Notes: c, chord; s, semi-span; b, span = 2s; η = y/s, η1 = ykink/s and centreline chord, c0.

www.ebook777.com

free ebooks ==> www.ebook777.com

30 Flight Dynamics, Simulation, and Control

Chapter Highlights

Aircraft features: The primary aircraft components are Fuselage, Wings,

Empennage or Tail plane, Landing gear and Power plant

Control surfaces and other controls: Throttle, Elevators, Trim tabs,

Horizontal Stabiliser, Aileron, Rudder, Vertical stabiliser, Flaps

(Fowler flaps), Spoilers, Slats and other High lift devices

Each control has a specific function (e.g. speed, pitch attitude, stick

force, bank and turn, added lift, lift dumping, stall delay, etc)

Exercises

1.1 The mean aerodynamic chord of symmetric wing planform is defined

by the integral

1

2s 2 ⎛ y ⎞ ⎛ y ⎞

c= c ⎜ ⎟ d ⎜ ⎟.

S

0

∫

⎝s⎠ ⎝s⎠

zoidal panels and the chord at any spanwise location is defined by the

relations

⎛ η −η ⎞ y

c ( η) = c0 ⎜ λ i +1 + ( λ i − λ i +1 ) i +1 ⎟, η = ,

⎝ ηi + 1 − ηi ⎠ s

ηi ≤ η ≤ ηi +1 , K i = 0, 2, , J − 1, η0 = 0, ηJ = 1, λ 0 = 1.

Show that the mean aerodynamic chord and the planform area are,

respectively, given by

J −1 J −1

2sc02

c=

3S ∑ (λ

i =0

2

i )

+ λ iλ i +1 + λ i2+1 ( ηi +1 − ηi ), S = c0 ∑ (λ

i =0

i +1 + λ i ) ( ηi +1 − ηi ).

free ebooks ==> www.ebook777.com

Introduction to Flight Vehicles 31

1.2 Show that for a straight tapered symmetric wing of an area of Sw, an

aspect ratio AR and a taper ratio λ, the centre chord length c0m is given by

2 Sw

c0 m = .

1+ λ AR

(i) Show from the first principles that the sweep angle of any constant-

chord fraction line can be related to that of the leading edge sweep

angle by

4n 1 − λ

tan Λ n = tan Λ 0 − ,

AR 1 + λ

1/4 for the quarter-chord line, 1 for the trailing edge).

(ii) Show that the location of any chord fraction point on the mean aerody-

namic chord, relative to the wing apex, can be determined as

s

2 3 c (1 + λ ) ⎛ ⎛ 1 + 2λ ⎞ ⎞

xn =

S ∫ ( nc

0

root + y tan Λ n ) c( y )dy = ⎜n +⎜

2 1 + λ + λ2 ⎝

( ) ⎝ 12 ⎠

⎟ AR tan Λ n ⎟ .

⎠

⎛ ⎛ 1 + 2λ ⎞ ⎛ 1 + λ ⎞ ⎞

xn = c ⎜ n + ⎜ ⎟⎜ 2 ⎟

AR tan Λ 0 ⎟ .

⎝ ⎝ 8 ⎠⎝ 1 + λ + λ ⎠ ⎠

1.4 Consider an elliptic wing planform. The root chord is given to be c0. The

semi-span is given as s.

Show that the mean aerodynamic chord and the planform area are,

respectively, given by

1 1

2s 2 8c π

c= c ( η) dη = 0 , S = 2s c ( η) dη = sc0 .

S ∫

0

3π ∫

2

0

1.5 Verify the formulas for the aspect ratio, the spanwise position of the

centroid of area, leading edge location and the chord at the spanwise

position of the centroid of area and streamwise position of the centroid

of area in Table 1.3.

www.ebook777.com

free ebooks ==> www.ebook777.com

32 Flight Dynamics, Simulation, and Control

1.6 Verify the formulas for spanwise position of the mean aerodynamic

chord, leading edge position of the mean aerodynamic chord, side-slip

force moment arm and the rolling moment arm in Table 1.3.

1.7 Verify the formulas for tip chord, local chord at kink, chord at any span-

wise location, mean geometric chord, mean aerodynamic chord and the

(k − 1)th moment arm for kinked-trapezoidal symmetric planforms in

Table 1.4.

4s2 4s

1.1 AR = = J −1 .

S c0 ∑ i=0 ( λi+1 + λi ) ( ηi+1 − ηi )

8s

1.4 AR = .

πc0

References

1. Anderson, J., Introduction to Flight, 4th ed., McGraw-Hill, New York, 2000.

2. Shevell, R., Fundamentals of Flight, 2nd ed., Prentice Hall, Englewood Cliffs, NJ,

1989.

3. Abbott, I. H. and von Doenhoff, A. E., Theory of Wing Sections: Including a

Summary of Data, Dover, New York, 1958.

free ebooks ==> www.ebook777.com

2

Basic Principles Governing

Aerodynamic Flows

2.1 Introduction

Aerodynamic flows are encountered when one is dealing with any aspect

of aeronautical engineering. Physical principles that govern aerodynamic

flows are based on the conservation of mass, momentum and energy. Just

as Euler’s equations of motion are derived from Newton’s laws of motion in

classical mechanics, secondary laws governing the dynamics of rotational

flows can be derived from the fundamental physical principles. The flows

around aerofoils are the simplest examples of aerodynamic flows. In aero-

foil theory, it is possible to idealise the flow by ignoring the influence of the

compressibility of the fluid medium. When compressibility is included, one

needs to consider three distinct cases: subsonic flow when the flow velocity

is well below the speed of sound or pressure disturbances in the flow, tran-

sonic flow when the flow velocity is in the vicinity of the speed of sound and

supersonic flow when the flow velocity is well above the speed of sound.

Furthermore, the viscous forces which result in friction at the boundaries

play a key role in aerodynamic flows. In this chapter, we review the basic

principles governing aerodynamic flows, the influence of compressibility

and viscosity, the definition of the standard properties of the atmosphere,

the flow around an aerofoil, the generation of lift, drag and moment over an

aerofoil and the basic properties of aerofoil aerodynamics.

The continuity principle is statement of the principle of conservation of mass

and states that the mass of a fluid flowing into a control volume is equal to

the mass of the fluid flowing out of it.

33

www.ebook777.com

free ebooks ==> www.ebook777.com

34 Flight Dynamics, Simulation, and Control

A streamline can be thought of as a continuous line in the fluid flow along

which the velocity of a fluid particle is tangent to it and the velocity com-

ponents in a plane perpendicular to it are zero. At any point on the stream-

line, each particle will experience the same velocity and pressure as the

preceding particles as they pass the point. Particles following a lead particle

in a streamline will experience the same velocity and pressure as the lead

particle as they pass its location in the streamline. These values of velocity

and pressure may change from point to point along the streamline. A reduc-

tion in the velocity of streamline flow is indicated by wider spacing of the

streamline, while increased velocity is indicated by a decrease in the spacing

between the streamlines. In a steady flow, therefore, the shape of a stream-

line is invariant; that is, it does not change with time. Further, the particles

in a particular streamline maintain their position relative to the particles in

another streamline. Thus, streamlines do not cross.

If the relative positions of the streamlines are identical in all parallel cross

sections of the flow, then the flow is said to be 2D.

A set of streamlines not coincident with each other constitute a stream tube.

Bernoulli’s principle is a statement of the principle of conservation of energy

and states that along any point in a uniform irrotational flow without dissipa-

tion, the sum of the pressure energy, the kinetic energy and the gravitational

potential energy is a constant. In aerodynamic flows, it is customary to refer to

both kinetic and potential energies in terms of equivalent pressure energies.

Thus, the pressure energy is distinguished from the other two by referring to

the conventional pressure as static pressure (ps). The equivalent pressure that

results in the same energy as the kinetic energy is the dynamic pressure and is

obtained by dividing the expression for the kinetic energy by the volume of

the flow. Thus, the expression for the dynamic pressure is q = ρV 2/2, while the

equivalent pressure corresponding to the potential energy is ph = ρgh.

When a thin plate is placed in a uniform flow of air, the air encounters friction as

it flows over the plate and the flow next to the air is retarded and brought to rest

adjacent to the surface of the plate. This retardation effect diminishes rapidly

when the flow is farther away from the plate’s surface. When the flow is uniform

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 35

and the retardation effect is restricted to a narrow layer close to the plate’s sur-

face, while the flow continues to be uniform beyond this layer, the flow is said

to be laminar and the layer is referred to as a boundary layer. The thickness of

the boundary layer is governed by the viscosity of the flow and the friction of

the surface. The viscous effects are only important within the layer and may be

ignored outside it. The thickness of the boundary layer may be controlled by

sucking it away. Further details on boundary layers may be found in Schetz [1].

When disturbances are initiated in the flow, which cause the flow beyond the

boundary layer to be non-uniform and disturbed, the flow is said to be turbu-

lent. In aerodynamics, turbulent flows are undesirable as it results in energy

loss due to the formation of eddies. Steady streamline flow is desirable in most

phases of flight and turbulent flow is best avoided. The transition of a laminar

flow to a turbulent is usually a multistage process. In the first instance, the

fully laminar flow region becomes partially turbulent and is characterised by a

turbulent inner layer. The point on the surface of the thin plate where this hap-

pens is the transition point. Beyond the transition point, the boundary is slightly

thicker. If and when the boundary layer separates from the plate, it causes the

main airflow to break away and become turbulent. The point where the bound-

ary layer separates from the plate, if and when it does so, is known as the sepa-

ration point. The occurrence of separation must generally be avoided in aircraft

flight as it causes the aircraft to stall.

The relative flow of air past a wing results in the development of a pressure

distribution over it. The characteristics of the aerodynamic pressure distri-

bution over a wing are functions of several factors which may be classified

into four principal groups:

2. Aerofoil and wing shape and geometry parameters

3. Size and scale effects

4. Orientation of the body relative to the flow

standing of the aerodynamics of aerofoils and wings.

www.ebook777.com

free ebooks ==> www.ebook777.com

36 Flight Dynamics, Simulation, and Control

The flow around an aerofoil has many similar characteristics to the flow

around a thin plate. The two cases of streamlined flow around a symmetric

and unsymmetric aerofoil are shown in Figure 2.1.

The higher velocity on the upper surface of the unsymmetric aerofoil causes

the pressure to fall on it, resulting in a suction which in turn is responsible

in lifting the aerofoil.

Pressure fluctuation in a flow results in sympathetic fluctuation in density.

In a fluid medium, only longitudinal waves manifesting themselves as com-

pression or rarefaction waves can propagate through the medium. These

disturbances in the flow medium are transmitted at the speed of sound

propagation of that medium. Thus, it is customary to non-dimensionalise

all flow speeds with the speed of sound. The ratio of the velocity of the free

stream relative to the aircraft to the speed of sound is known as the Mach

number and is denoted by the letter M. As the flight speed approaches the

speed of sound, that is, the velocity of the free-stream approaches the speed

of sound, the Mach number approaches unity and effects of compressibility

(a)

(b)

FIGURE 2.1

Flow around an aerofoil. (a) Flow around a symmetric aerofoil section and (b) flow around an

unsymmetric aerofoil section.

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 37

are both pronounced and noticeable. Based on the Mach number of the

flight, the types of flow fields one could encounter in flight could be broadly

grouped into one of three types: subsonic, transonic and supersonic. In the sub-

sonic case, M < 1 at all points in the flow, and the effects of the compressibil-

ity of the flow medium could be safely ignored. Thus, the air density can be

assumed to be quite independent of any pressure fluctuations. In this case,

increasing the cross section of flow results in a lower speed and a higher pres-

sure and vice versa. When the flow is supersonic, M > 1, the velocity of free

stream past the aircraft is greater than the speed of sound at all points in the

flow. This is the region of supersonic flow when any increases or decreases

in speed must be accompanied by matching increases and decreases in the

cross-sectional area of the flow. Any increases or decreases in speed are also

accompanied by matching decreases or increases in pressure and density.

Quite naturally, when M ≈ 1 at some points in the flow, the region of flow is

in transition, from a subsonic to a supersonic, from a supersonic to subsonic

or is a mixed region. It can therefore be expected that analysis of such flows

is complex and requires special consideration. When M ≈ 1 at some points

within a flow, it is said to be transonic. When the free-stream flow Mach

number is well above unity and in the regions of 4 and 5 at all points in the

flow, the flow may be considered to be hypersonic and the analysis of such

flows is amenable to certain simplifications.

When a body with an appropriate profile such as an aerofoil moves in a

flow at subsonic speeds, the stream receives a signal of the body’s impend-

ing arrival well before the body arrives at a point in the flow. In the case

of the body moving at supersonic speeds, the flow has absolutely no prior

knowledge of the body’s arrival and the body cuts through a completely

undisturbed and unsuspecting flow. The situation can be described in terms

of propagation of spherical disturbance waves in (1) an incompressible flow,

(2) compressible flow at subsonic speeds, (3) transonic speeds and (4) super-

sonic speeds (see Figure 2.2).

Considering Figure 2.2a, the case when the source of the disturbance is

at rest relative to the flow field, the disturbance propagates with uniform

velocity in all directions and wavefronts propagate in the form of concentric

circles (2D case). When the source of the disturbance is in motion with a

speed well below the speed of sound, wavefronts continue to propagate with

a speed equal to the difference between the speed of sound and the speed

of the disturbance, in the direction of the motion of the disturbance. In the

opposite direction, the disturbance propagates at the sum of the two speeds

and the wavefronts are no longer concentric circles. This situation is illus-

trated in Figure 2.2b. When the disturbance moves with the speed of sound,

as in Figure 2.2c, the wavefront is stationary with respect to the source, and

in the course of time, it evolves into a normal shock boundary, a surface of

discontinuity across which there is discrete change in pressure and density.

Finally, when the source moves faster than the speed of sound, it is always

ahead of the wavefront and the result is an oblique shock boundary as shown

www.ebook777.com

free ebooks ==> www.ebook777.com

38 Flight Dynamics, Simulation, and Control

(a) (b)

(c) (d)

FIGURE 2.2

Propagation of spherical disturbances from a point source in (a) an incompressible flow,

(b) compressible flow at subsonic speeds, (c) transonic speeds and (d) supersonic speeds.

in Figure 2.2d. The sine of the half angle of the cone, known as the Mach

angle, is equal to the inverse of the Mach number.

The surface of the cone forms a shock boundary. The shock boundary

separates the free undisturbed flow and the region of the disturbed flow.

At transonic and supersonic speeds, there is substantial increase in the drag

experienced by the body in the flow. The features of compressible flows are

discussed by Shapiro [2].

2.7.1 Composition of the Atmosphere: The Troposphere,

Stratosphere, Mesosphere, Ionosphere and Exosphere

The envelope of air surrounding the Earth, which is essentially a cosmic

boundary layer around the Earth, is known as the atmosphere. The depth

of the atmosphere is very thin in comparison with the Earth’s radius. More

than 50% of the mass of the atmosphere is within 6 km of the Earth’s surface,

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 39

75% of the mass of the atmosphere is within 10 km and 94% of the mass of

the atmosphere is within 20 km. The atmosphere begins to decompose to an

atomic state at an altitude of 120–150 km, and beyond 200 km, it is completely

in an ionic state. It is a mixture of several gases, the primary constituents

being nitrogen (78%) and oxygen (21%), while the remaining 1% is made up

of argon, hydrogen, carbon dioxide and helium. Broadly, the atmosphere is

divided into two regions: the lower atmosphere (up to 50 km) and the upper

atmosphere. The lower atmosphere is further divided into two layers of

varying thickness across the Earth’s surface: the troposphere (8 km in depth

over the poles to 16 km over the equator) and the stratosphere. Likewise, the

upper atmosphere is divided into three regions: the mesosphere (50–80 km),

the ionosphere (70–500 km) and the exosphere (from about 450 km and beyond).

The distinguishing feature between the upper and lower atmosphere is the

fact that while the lower atmosphere is practically a homogeneous mixture,

the upper atmosphere is completely inhomogeneous both spatially and tem-

porally. It is characterised by low air pressures and densities and by intense

processes of dissociation and air ionisation, resulting in the splitting of mol-

ecules and in the formation of charged particles.

Weather and thermal air currents originate in the troposphere, where con-

densation of water vapour and cloud formation are possible. There is a grad-

ual linear fall in the temperature in the troposphere followed by the region

of constant temperature in the lower stratosphere (−56.5°C) and temperature

versus altitude rise in upper regions of the stratosphere.

There is a rapid decrease in the atmospheric density and pressure with alti-

tude. In the troposphere, there is considerable non-uniform turbulent activ-

ity. The unsteadiness of the troposphere both spatially and temporally gives

rise to a number of difficulties particularly in predicting flying characteris-

tics of flight vehicles.

2.7.3 Temperature

In the troposphere, the air temperature quickly decreases with altitude. In

the stratosphere, it remains almost constant to roughly the 25–27 km level

above which it starts to rise intensely with altitude. It is approximately 0°C at

an altitude of 50 km. In the mesosphere, the temperature falls again to −80°C

at an altitude of 80 km.

2.7.4 Pressure

At an altitude of 10 km, the air pressure is 3.8 times as low as at the ground

level, while the air density is only 3 times as low. At an altitude of 25 km,

the air pressure reduces to just 2.4% of its sea level value, while air density

www.ebook777.com

free ebooks ==> www.ebook777.com

40 Flight Dynamics, Simulation, and Control

the air pressure is only a billionth of the sea level value, while the air density

drops to less than one billionth of its sea level value.

The effects of pressure and temperature on air density can be stated in

terms of the so-called universal gas law. The universal gas law, relating to

a perfect gas, is

pv = NRT (2.1)

where

p is the pressure in the gas

Absolute T is the temperature of the gas

R is the universal gas constant

v is the volume corresponding to N moles of gas

N

ρ= (2.2)

v

p = ρRT . (2.3)

As the density is the ratio of the mass and volume, the pressure and tempera-

ture affect it indirectly, as a consequence of the universal gas law.

2.7.6 Viscosity

When one layer of a fluid slides over another, there is a friction-like force and

is termed as a viscous friction force. A standard measure of these forces is the

coefficient of viscosity, μ, which is defined by considering a narrow layer of

fluid flowing over a horizontal surface. The shear stress at the top of the layer

τ is directly proportional to the rate of shear strain, that is, the rate of change

of the flow velocity component parallel to the plane and in the direction of

the flow with respect to the normal distance from the horizontal surface. The

proportionality constant is the coefficient of dynamic viscosity, μ. Hence,

∂u

τ=μ . (2.4)

∂y

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 41

The bulk modulus of elasticity, although not a critical parameter in aero-

dynamics, is important as it indirectly influences the speed of sound. It is

defined as the ratio of the stress to strain and is given as

Δp Δp dp

K = − Lt = Lt =ρ (2.5)

Δv → 0 Δv/v Δv → 0 Δρ/ρ dρ

where

Δp is the pressure difference

Δv/v is the volume strain

K

a2 = . (2.6)

ρ

unique for the entire flow, and considering the flow to be isentropic, the dif-

ferentials are evaluated from the isentropic flow condition

pv n = Constant. (2.7)

The rate of decrease of temperature with altitude is known as the lapse rate and

has a value of 6.5 K/km in the troposphere. The lapse rate is essentially dif-

ferent in different altitude bands in the troposphere and in the stratosphere.

Based largely on mean values observed over extended periods of time, cer-

tain standard models for the variation of temperature with altitude have

been developed. Amongst these standard models, the International Standard

Atmosphere (ISA) developed by the International Civil Aviation Organisation

is accepted worldwide as a typical model for the standard atmosphere.

(from ESDU 77021, 1986)

It has become a norm amongst aeronautical engineers worldwide to accept

a common standard for measuring or considering the properties of air.

Accordingly, an ISA has been established. It includes models of temperature,

www.ebook777.com

free ebooks ==> www.ebook777.com

42 Flight Dynamics, Simulation, and Control

air density and air pressure variations with altitude. Sea level conditions are

defined in the ISA as follows:

Temperature, T0 = 15°C = 288.15 K Speed of sound, a0 = 340.3 m/s

Universal gas constant, R = 287.053 J/kg K g0 = 9.806 65 m/s2

The atmospheric bands for heights up to 150 km are treated separately in the

following:

⎛α ⎞

Temperature ratio, Tratio = 1 − ⎜ ⎟ × h, (2.8c)

⎝ T0 ⎠

g

Pressure ratio exponent , n = = 5.256, (2.8d)

α×R

a1

Ratio of the speed of sound, = Tratio ( 0.5 ) , (2.8g)

a0

is a constant temperature region,

a20

= 0.8671, (2.9c)

a0

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 43

At an altitude of 11 km,

p20 ρ20

= 0.2234, = 0.2971, (2.9d)

p0 ρ0

and

p2 ρ 2

= = exp[−0.15769( h − 11)]. (2.9e)

p20 ρ20

3. For altitudes in the range 20 ≤ h (in km) ≤ 32 km, which is a linear

temperature region,

Above 20 km and up to an altitude of 32 km, the temperature rises

linearly from 216.65 to 228.65 K at 32 km, α = −1 (lapse rate):

T3 = T2 − α × ( h − hb ), hb = 20 km. (2.10)

4. For altitudes in the range 32 ≤ h (in km) ≤ 86 km, the temperature is

given by the following equation:

T4 = Tb − α × ( h − hb ), (2.11)

where Tb and hb are given in Table 2.1.

5. For altitudes in the range 86 ≤ h (in km) ≤ 91 km, which is a constant

temperature region, temperature

T = 186.8673 K. (2.12)

6. For altitudes in the range 91 ≤ h (in km) ≤ 110 km,

1

⎡ ⎛ h − 91 ⎞2 ⎤ 2

T = Tc − A ⎢1 − ⎜ ⎟ ⎥ (2.13)

⎢⎣ ⎝ hc ⎠ ⎥⎦

where

Tc = 263.1905 K

A = 76.3232 K

hc = 19.9429 km

TABLE 2.1

Table of Lapse Rates at Various Altitudes

Altitude Range (km) hb (km) Alpha Tb (K)

32–47 32 −2.8 228.65

47–51 47 0 270.65

51–71 51 2.8 270.65

71–86 71 2 214.65

www.ebook777.com

free ebooks ==> www.ebook777.com

44 Flight Dynamics, Simulation, and Control

T = T9 − Lk ,9 ( h − ha ) (2.14)

where

T9 = 240 K

Lk,9 = 12.0 K/km

ha = 110.0 km

⎡ ( h − h10 ) ( r0 + h10 ) ⎤

T = T∞ − ( T∞ − T10 ) exp ⎢ −λ ⎥ (2.15)

⎢⎣ ( r0 + h ) ⎥⎦

where

T∞ = 1000 K

T10 = 360 K

λ = 0.01875

h10 = 120 km

r0 is the Earth’s radius, r0 = 6356 km

equations:

⎡ g M ( h − hb ) ⎤

p = pb exp ⎢ − 0 0 ⎥ (2.16)

⎣ R* TM , b ⎦

⎡ g0 M0 1 ⎤

⎢ ⎥

⎡ TM , b ⎤⎣ R* LM ,b ⎦

p = pb ⎢ ⎥ (2.17)

⎢⎣ TM , b + LM , b ( h − hb ) ⎥⎦

where

pb is the pressure at the end of the previous height band

g0M0/R* = 34.16

hb is the height at the lower end of the height band

TM,b and L M,b are given in Table 2.2

TABLE 2.2

Table of Temperature and Pressure Model Parameters at Various Altitudes

Altitude Range (km) hb LM,b TM,b (K) Equation

86–91 86 0 186.946 2.16

91–110 91 0 186.870 2.16

110–120 110 12 240 2.17

120–150 120 12 360 2.17

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 45

To understand the process of the generation of lift and drag, consider a wing

section in a 2D steady flow of a perfect ideal fluid. Consequently, the fluid

is assumed to not possess any viscosity, and as a further consequence, there

is no energy dissipation, no work done and no drag and the wing cannot

develop any circulation around it. Hence, there is also no lift! When con-

sidering the streamline of the flow around such an aerofoil, ideally there

are two stagnation points which separate the flow domain into two regions,

one above and the other below the aerofoils. The streamline on the separat-

ing boundary ends on the aerofoil surface at a stagnation point, somewhere

below the leading edge for a positive angle of attack. It re-emerges at a rear

stagnation point somewhere on the upper surface of the aerofoil just before

the trailing edge and proceeds downstream along the flow field. When the

developed pressure distribution is integrated over the surface, there is no lift

and no drag.

In the case of an aerofoil in real flow field, in addition to the aforemen-

tioned flow pattern, a circulation of the flow around the aerofoil develops.

Initially, the flow pattern is as in the ideal case as discussed earlier. Yet

because of boundary layer effects, the flow on the lower surface is retarded

more than the flow on the upper surface. The two flows meet somewhere

on the upper surface very near the trailing edge and form a local shear flow

pattern that results in the formation of an eddy. Moreover, the process of

the eddy formation is unstable, and consequently, the eddy is swept down-

stream. Since the eddy contains a vortex with a finite circulation, it follows

from the fundamental physics of vortex flows that an equal and opposite cir-

culation must develop around the wing. Towards the end of the nineteenth

century, the German mathematician Wilhelm M. Kutta and the Russian

Nicolai E. Joukowski independently put forward the theory of lift genera-

tion in physical terms while also expressing it mathematically. In particular,

Kutta postulated that the strength of the circulation generated by the vis-

cous effects was just enough to move the rear stagnation point to the trailing

edge, thus maintaining a smooth flow in the vicinity of the trailing edge.

He argued that excessive or less circulation would not necessarily result in

a stable and physically meaningful flow and that the physics of the real vis-

cous flow field required the rear stagnation point to remain at the trailing

edge, as the velocity of the flow on either side of the trailing edge must be

identical as it leaves it. This would ensure that there is no jump in either

the velocity of pressure thus requiring the flow to be smooth at the trailing

edge. The condition of smooth flow at the trailing edge thus determines the

strength of the circulation and is known as Kutta condition. It allows one to

superpose a circulation of a finite and known strength over and above the

ideal flow field so as to generate the correct flow pattern on a wing in a real

viscous flow with very small amounts of viscous dissipation. Thus, the real

www.ebook777.com

free ebooks ==> www.ebook777.com

46 Flight Dynamics, Simulation, and Control

flow field tends to be an ideal flow in this case and corresponds to a situa-

tion where the Reynolds number, Re = ρUd/μ, is high. The Reynolds number is a

non-dimensional number that characterises the ratio of the inertia to viscous

forces, where the numerator is the product of the density of the fluid in the

free stream (ρ), the free-stream velocity of the fluid (U) and a characteris-

tic distance (d), while the denominator is the coefficient of viscosity (μ). The

characteristic distance is usually taken to be the aerodynamic mean chord

length c . It should be mentioned in passing that if the Reynolds number of the

flow is too high, the flow ceases to be laminar at certain points on aerofoil

and experiences a transition to turbulent flow flowed by separation and loss

of lift. Thus, it is essential that the Reynolds number of the flow is optimum

if the flow is to remain attached, which is essential for the generation of lift.

Considering Bernoulli’s principle of conservation of energy of an ideal

flow around the aerofoil, we may relate the pressure p and velocity perturba-

tion q on the upper and lower sides of the aerofoil by the equation

1 2 1 2

pu + ρ (U + qu ) = pl + ρ (U + ql ) . (2.18)

2 2

Hence,

⎛ q +q ⎞

pl − pu = ρ ( qu − ql ) ⎜ U + u l ⎟ . (2.19)

⎝ 2 ⎠

c c

⎛ qu + ql ⎞

L=

∫ ( pl − pu ) dx = ρ∫ ( qu − ql ) ⎜ U +

0 0

⎝ 2 ⎟ dx. (2.20)

⎠

c c

L=

∫ ( p − p ) dx = ρU ∫ ( q − q ) dx = ρUΓ. (2.21)

0

l u

0

u l

c 1

L=

( qu − ql ) dx = ρU 2 c Γˆ = ρUΓ. (2.22)

∫ ( p − p ) dx = ρU c ∫ 2

l u

U c

0 0

vorticity, around an aerofoil to the lift.

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 47

In Section 2.7, it was shown that for steady inviscid flows, the Kutta–Joukowski

theorem relates circulation, and therefore vorticity, around an aerofoil to the

lift L, by the equation

L = ρU 2 c

∫ uˆ ⋅ dˆl = ρU c Γˆ = ρUΓ (2.23)

2

Here, Γ̂ is the lift per unit span non-dimensionalised with respect to the prod-

uct of density of the fluid (ρ), the aerodynamic mean chord length, c, and the

square of free-stream velocity of the fluid (U). This quantity (conventionally

multiplied by two) is called the lift coefficient and is usually denoted in litera-

ture by CL. For inviscid fluids undergoing steady (non-accelerated) flows,

d ˆ d

dt

Γ=

dt ∫ uˆ ⋅ dˆl = 0 (2.24)

Σ

and

Γˆ =

∫ uˆ ⋅ dˆl = Constant. (2.25)

Σ

When an aerofoil starts from rest, the net circulation in the fluid before the

start of the motion is zero. Thus, equation for Γ̂ is simply a mathematical

expression for Kelvin’s law, which states that the total circulation (and there-

fore the total vorticity) in an ideal fluid must remain zero at all times. In

other words, if new vorticity (or circulation) is introduced in an inviscid fluid

(e.g. through an application of the Kutta condition), then it must be accompa-

nied by equal and opposite vorticity which is usually distributed in the wake

behind the trailing edge. However, as vorticity diffuses into the wake and is

distributed in, it does not generate any forces within it. Thus, the only forces

acting are those on the aerofoil itself. Yet this shed or diffused vorticity rep-

resents a loss in kinetic energy and this contributes to the induced drag. The

energy source is the energy in the circulation imposed on the aerofoil. So

although one is dealing with an energy conservative potential flow problem

and there is no energy dissipation overall, there is energy transport from a

region near the aerofoil to the far field. It is this energy transported away

from the aerofoil that manifests itself as the induced or vortex drag. Because

the domain of the flow field is practically infinite, the fact that it is energy

conservative within the domain is not very useful. In the aerofoil problem,

www.ebook777.com

free ebooks ==> www.ebook777.com

48 Flight Dynamics, Simulation, and Control

Low pressure

High pressure

(a)

Lift, L

Nose-up

pitching moment, M

U0 Drag, D

(b)

FIGURE 2.3

(a) The mechanism of wing tip vortex formation and (b) definition of lift force, pitching moment

and drag force.

the forces and moments acting on it.

The shedding of vortices is considerably enhanced in a finite wing due to

the formation of vortices at the tips of the wing. This process is illustrated

in Figure 2.3a. Paired vortices are formed in the wake of the aircraft by an

aerodynamic process that is directly related to the lift generated. The aerody-

namic flow past the leading edge of each wing establishes a boundary layer

that separated from the wing surface and rolls up into a spiral vortex sheet.

At some distance behind the trailing edge of the wing, the streamlines of the

separated flow converge resulting in a dominant flow pattern consisting of a

pair of vortices separated by a distance equal to the wingspan of the aircraft.

The diameter of each tightly bound vortex core is only about 3% of the wing

span. The vortices are remarkably stable and persist for long distances behind

the aircraft. Wing tip vortices produced by the wings of smaller aircraft have

a negligible effect in the wake, but wing tip vortices created by larger and

heavier aircraft can be extremely dangerous to aircraft trailing behind, even

at a distance of many miles. Wing tip vortices form a part of the entire wake

roll-up vortex and the turbulence generated in the wake of large aircraft can

cause buffeting, instability, uncontrollable rolls and sudden loss of altitude in

a trailing aircraft. This is due to the fact that the flow field between the vortex

pair appears as an induced downwash, while beyond the vortex pair, it is

an induced upwash, resulting in severe wind gradients in the vicinity of the

vortices. There have been many incidents, especially at lower altitudes during

landing approaches, when wake vortex effects have resulted in fatal accidents

because of the inability of pilots to regain full control of their trailing aircraft

after being buffeted violently by the powerful wake vortices.

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 49

moments acting on an aerofoil in an incompressible or in a compressible

viscous flow with only the knowledge of the velocity or vorticity field in

a finite and arbitrarily chosen region enclosing the body. Furthermore, the

forces and moments can be conveniently expressed in terms of integrals of

the moments of the vorticity and their time derivatives (see, e.g. Ashley and

Landahl [3]). They are particularly useful in understanding the nature of the

forces and moments acting on the aerofoil. For our part, it is much more con-

venient to define a set of non-dimensional aerodynamic coefficients, which

may be used to represent the forces and moments acting on an aerofoil.

Consider an isolated vortex of strength γ c located on the surface of a flat

plate of chord c and infinite span, inclined to a uniform free stream with a

velocity U at an angle α. The total circulation induced around the flat plate

aerofoil is

c

Γ=

∫ γ c δ ( x − x ) dx = γ c . (2.26)

0

γ

The total normal suction force acting vertically upwards per unit span

is N = ρU c γ cos α, while the forward propulsive force per unit span is

T = ρU c γ sin α . The lift force per unit span, normal to the direction of the

free stream, is given by

The corresponding net drag force acting in the direction of the velocity per

unit span is

D = N sin α − T cos α = 0. (2.28)

Assuming that the strength of the vortex is proportional to the velocity nor-

mal to the plate, U sin α, the lift force per unit span is

One could also perform an exact analysis of the flow around a flat plate by

mapping a circle in a uniform flow field on to a flat plate. This results in a

non-uniform distribution of vortices, γ(x), along the surface of the flat plate

and the total circulation induced is

c

Γ = γ ( x ) dx. (2.30)

∫0

www.ebook777.com

free ebooks ==> www.ebook777.com

50 Flight Dynamics, Simulation, and Control

While the complete analysis is beyond the scope of this section, the expres-

sion for the total lift force per unit span is

the non-dimensional lift force per unit span may be expressed as

L

CL = = 2Γˆ = 2π sin α. (2.32)

1

ρU c

2

Assuming the wing to be unswept, of infinite span and of uniform chord,

one may consider the forces and moment acting on a typical wing section. By

convention, the lift, pitching moment and drag force are typically assumed to

act at point along the chord of the wing section or aerofoil and in directions

normal to and parallel to the airflow relative to the wing. These directions are

illustrated in Figure 2.3b.

The lift, pitching moment and drag are typically defined in terms of

non-dimensional lift, pitching moment and drag coefficients. The relation-

ships between the non-dimensional wing section lift, wing section pitching

moment and wing section drag coefficients, CL, Cm and CD, and the wing

section lift force, L; wing section pitching moment, M; and the wing section

drag force, D, are

1

L= ρU 02 × S × CL (2.33a)

2

1

M= ρU 02 × S × cref × Cm (2.33b)

2

and

1

D= ρU 02 × S × CD (2.33c)

2

where

ρ is the density of the free stream

U0 is the velocity of the free stream relative to the aerofoil

cref is the reference aerofoil chord, usually, c , the mean aerodynamic chord

S is the reference area, usually, SW, the area of the planform of the wing

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 51

determined under assumptions of an ideal flow field (incompressible and

inviscid flow) assuming that the aerofoil is essentially a flat plate. The ana-

lytical expression under these assumptions of ideal flow is

CL = CL 0 + 2π × sin α. (2.34)

When the reference axis is also the zero-lift line, CL0 = 0; that is, CL = 0

when α = 0.

The variation of the lift coefficient with respect to the angle of attack α has

also been determined experimentally in a wind tunnel by several experi-

menters. In almost all these experiments, the relationship has been found to

be linear for values of α less than a critical value, αs, αs ≈ 12°. For values of

α < αs ≈ 12°, sin α ≈ α and the expression for CL may be expressed as

dCL dCL

CL = CL 0 + α= ( α − α0 ) (2.35)

dα α =0 dα α =0

dCL

where CL 0 ≡ − α 0 and α 0 is the zero-lift angle of attack. Analytically, it is

dα α =0

known that

dCL

= a∞ = 2π (2.36)

dα

A typical experimentally determined plot of CL versus α is shown in

Figure 2.4a. For most conventional and many other aircraft, the maximum

attainable value of the lift coefficient CL plays a critical role in the overall

design of the aircraft. The take-off and landing distances are directly related

to the aircraft minimum speed, and a high value for maximum CL, CLmax ,

will allow for low-speed flight. The maximum lift coefficient attainable by

a wing depends on the aerofoil section and, to some extent, the wing geom-

etry. Thick aerofoils with a large leading-edge radius generally are associ-

ated with large values of CLmax . However, most modern aeroplanes utilise

high-lift devices, such as flaps and slats, to increase the lift. Thick aerofoils

with a large leading-edge radius generally are associated with large values

of CLmax well beyond that attainable by a simple wing. The maximum attain-

able value of CLmax depends on the type of flap and/or slat system, the percent

chord dedicated to the flap (or leading-edge device) and the percent span

that is flapped. Most aircraft do not have full-span flaps since some of the

wing is required for aileron control surfaces.

www.ebook777.com

free ebooks ==> www.ebook777.com

52 Flight Dynamics, Simulation, and Control

CL CL CL

α α α

(a) (b) (c)

FIGURE 2.4

(a) Plot of wing lift coefficient versus angle of attack: The effect (b) slots and (c) flaps.

Flaps and slats increase the wing maximum lift coefficient in several ways

(Figure 2.4b and c):

1. Modern slats and slotted flaps extend the leading and trailing edges

of the wing, creating an effectively larger wing area. Therefore, a CL

based on wing reference area will automatically be larger when the

increase in area is taken into account.

2. Both slats and slotted flaps create a channel for air from the lower

surface of the wing to travel to the upper surface. Especially at high

angles of attack, the lower surface has much less severe pressure gra-

dients than the upper surface; thus, the boundary layer air on the

lower surface has considerably higher momentum than that of the

upper-surface boundary layer. This higher-momentum air is injected

through these channels (or slots) into the upper-surface boundary

layer, giving that air more momentum with which to overcome the

pressure gradient and remain attached.

3. Flaps add effective camber to the wing.

The maximum lift is associated with flow separation and is usually followed

by the aircraft literally stalling while in a steady flight. Within a small region

adjoining the surface of a body in a flow, there exists a boundary layer. Viscous

stresses play a very important role within this region, causing an increase in

the drag and resulting in wing stall. To illustrate this phenomenon of flow

separation and stall, consider a 2D flow over the top surface of an aerofoil. As

computed using ideal flow theory, the pressure on the aft part of the upper

surface is increasing as the flow moves towards the trailing edge. Therefore,

the momentum of the flow must be enough to overcome this adverse pressure

gradient. In ideal flow theory, the flow momentum is exactly enough to return

the flow to stagnation conditions at the trailing edge. In a realistic flow over

a wing surface, there are additional forces acting on the air as it moves along

the surface. This arises from the viscous stress (skin friction) at the surface,

and it opposes the motion of the air and has the effect of slowing the velocity

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 53

of the fluid within the boundary layer. If the combined action of the pressure

force and the friction force is enough to bring the velocity on the aerofoil sur-

face to zero, the flow will detach from the surface. This phenomenon is called

separation, and if the separation occurs at any location not directly in the vicin-

ity of the trailing edge, the forces it generates on the aerofoil tend to stall the

aircraft. Because the flow is no longer attached to the aerofoil upper surface,

the pressure distribution on that surface is not the one that is ideally required,

and the lift is decreased, sometimes abruptly. Since the separated flow region

is highly turbulent, stall also causes unsteady vibration and buffeting.

The estimation of aerodynamic drag which is the sum of three components:

form and friction (i.e. profile) drag and induced drag, is normally best esti-

mated from test data generated within the confines of a wind tunnel. A model

of the aeroplane or one of its components is suspended in a wind tunnel

by special support and coupled to an aerodynamic balance which is used to

measure the forces and moments acting on the component. The wind tunnel

experiments are usually performed for dozens of different parametric val-

ues (different angles of attack or sideslip angle) and similar configurations.

The data are then appropriately non-dimensionalised and reinterpreted to

obtain the aerodynamic drag values corresponding to the real component.

The most commonly adopted method of estimating the drag of the aircraft

is to estimate the drag of the components separately and then sum up the

individual contributions to the total drag. A typical plot of the variations of

the total drag coefficient and the profile and induced drag contributions to it

versus the equivalent airspeed at sea level is shown in Figure 2.5a.

CD

Total drag

L/D

Profile drag

Induced drag

(a) (b)

FIGURE 2.5

(a) Plot of wing drag components versus equivalent (sea level) airspeed and (b) variation of the

L/D ratio with the angle of attack.

www.ebook777.com

free ebooks ==> www.ebook777.com

54 Flight Dynamics, Simulation, and Control

Unfortunately, the minimum value of the drag and the maximum lift do

not occur for the same angle of attack; the drag is a minimum at about 1°,

while the lift is a maximum at about 15°. Thus, there is a need to compromise

and this is done by considering the variation of the ratio of the lift to the drag

with the angle of attack. A plot of this ratio is shown in Figure 2.5b which is

usually at about α ≈ 4°. The curve shows that there is a steep fall in the value

of L/D when the angle of attack exceeds the stalling angle. This is due to the

rate of increase of drag increasing sharply before stall.

The speed at which the maximum value of the L/D ratio can be attained

corresponds to minimum thrust. As the angle of attack is increased, the

coefficient of drag and equilibrium flight requires a lower flight speed. As

the stalling angle is approached, the effect of the increasing drag coefficient

outweighs that of the decreasing velocity and more thrust is required at the

stalling speed than at other speeds. On the other hand, there is a limit to the

amount of thrust that can be delivered, and as a consequence, the maximum

speed depends primarily on engine power.

The lift coefficient of an aircraft is composed of the wing-fuselage contribu-

tion and the contribution of the tail plane. Hence,

ST T

CL = CLwb + CL (2.37)

SW

where the ratio of the area of the tail plane, ST, to the area of the wing, SW,

is used as a weighting factor to scale the non-dimensional tail plane lift,

obtained by dividing the tail plane lift by the product of the tail plane area

and the dynamic pressure.

The contribution of the wing body to the lift is mainly due to the wing

itself and may be expressed as

dCLwb

CLwb ≈ a ( α w − i0 ) , = a (2.38)

dα

where the wing angle of attack is αw and i0 is the incidence angle for zero lift.

An approximate expression for a for a rectangular wing of infinite span,

finite thickness and at relatively low Mach numbers is

⎛ t⎞

a∞ = 5.65 ⎜ 1 + 0.8 ⎟ (2.39)

⎝ c⎠

where

t = tmax is the maximum thickness of the aerofoil

c is the aerofoil chord

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 55

For finite aspect ratios {aspect ratio, AR = (wing span)2/(wing area)}, and at

relatively low Mach numbers, an approximate expression for a is

πa∞ AR

a= . (2.40)

a∞ + π AR2 + 4

of different aspect ratios, in Figure 2.4a.

The scaled contribution of the tail plane to the lift coefficient, CLT , may

be expressed in terms of αT = tail plane incidence angle, η = elevator angle,

β = tab angle, a1 = tail plane lift coefficient, a2 = elevator lift coefficient and

a3 = tab lift coefficient as

(2.41)

LT

CLT = . (2.42)

1

ρU 02 × ST

2

The tail plane angle of attack, αT, or tail plane incidence angle is related to

the aircraft angle of attack, the tail plane setting angle, iT, and the down-

wash angle (ε) the tail plane experiences due to the flow over the wing by

the relation

αT = α + iT − ε (2.43)

where the aircraft angle of attack, α, is defined in Figure 2.4b. Note that the

pitch angle θ = α + γ, α = angle of attack and γ = climb angle or flight path angle.

The downwash, ε, represents the interference effect of the wing on the tail

plane. Under steady flow conditions, ∂ε/∂α is evaluated empirically in terms

of the aircraft angle of attack α which may be expressed in terms of the wing

angle of attack, αw, as

α = α w − iw (2.44)

where iw is the wing setting angle (Figure 2.6c). The downwash, ε, may then be

approximated as

∂ε

ε= ( α w − i0 ) (2.45)

∂α

www.ebook777.com

free ebooks ==> www.ebook777.com

56 Flight Dynamics, Simulation, and Control

CL

AR = infinity

AR = 10

AR = 4

AR = 2

α

(a)

x-axis

U0

Horizontal

(b) γ = Climb angle

Aircraft datum

x-axis

iw

U0

α αw

(c)

FIGURE 2.6

(a) Plot of wing lift coefficient, for wings of different aspect ratios versus angle of attack,

(b) definition of the aircraft angle of attack and the climb angle and (c) definition of the wing

setting angle.

⎛ ∂ε ⎞

αT = ( α w − i0 ) ⎜ 1 − ⎟ + iT + i0 − iw . (2.46)

⎝ ∂α ⎠

CLw ⎛ ∂ε ⎞

αT =

∂CLw ⎜1− ⎟ + iT + i0 − iw . (2.47)

⎝ ∂α ⎠

∂α w

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 57

With reference to the wing zero-lift line, the angle ηT, defined by the relation

ηT = iT + i0 − iw , (2.48)

is the angle of the tail plane zero-lift line. Hence, the tail plane angle of attack

may be expressed as

CLw ⎛ ∂ε ⎞

αT = 1− ⎟ + ηT . (2.49)

∂CLw ⎜⎝ ∂α ⎠

∂α w

With the elevator and tab fixed, the derivative of the aircraft lift coefficient

with respect to the angle of attack may be expressed as

∂CL S ⎛ ∂ε ⎞

= a + T a1 ⎜ 1 − ⎟ (2.50)

∂α S ⎝ ∂α ⎠

where

a is the wing lift curve slope = ∂CLwb ∂α w

a1 is the derivative of the tail plane lift coefficient with respect to the angle

of attack

The centre of pressure (CP) is a point along the aerofoil chord, aft of the lead-

ing edge, about which the aerodynamic pitching moment equals zero. The

algebraic summation of the aerodynamic pressures over the upper and

lower surfaces of the wing produce a resultant force which has a definite

magnitude and direction and can be deemed to act at the CP. The moment

of the force about the CP would then be zero. With increasing angle of

attack, the position of the CP moves closer to the centre of gravity (CG).

The consequence of this movement is that the wing-body lift has relatively

no influence on the overall aerodynamic moment which can only be varied

significantly by controlling the tail plane lift.

The aerodynamic centre (AC) is a point along the chord where the pitching

moment is independent of the angle of attack, that is, the derivative of the

pitching moment coefficient, about the AC, with respect to the angle of

attack is zero, ∂CM/∂α = 0. In the case of a symmetrical wing section, the posi-

tion of the CP is fixed and is coincident with the AC. In the case of a cambered

wing, the position of the CP usually lies aft of the AC. For subsonic sections,

www.ebook777.com

free ebooks ==> www.ebook777.com

58 Flight Dynamics, Simulation, and Control

the AC lies near the quarter-chord position, that is, 25% of the chord length

back from the leading edge. For supersonic sections, the position moves aft,

closer to the mid-chord position.

The aerodynamic pitching moment is the moment of all the aerodynamic

pressure forces acting on the surface of the aircraft? It is usually necessary

to specify the point about which this moment is taken. Usually, the aerody-

namic moment of a wing or aerofoil is taken about the leading edge or the

forward quarter-chord point. The distribution of pressure on an aerofoil may

generate a moment even if the lift is zero.

Considering the wing pitching moment coefficient about any point x,

where x is measured from the wing leading edge and assumed to be positive

in the direction of the trailing edge, may be expressed in terms of the wing

pitching moment about the AC and the wing lift coefficient as

C Mx | = CM0 +

( x − xac ) C wb . (2.51)

wing −body L

c

For an aircraft one has to consider not only the wing lift and moment but

also the tail plane lift. Referring to Figure 2.7 and taking moments about the

CG of the aerodynamic forces and external torques acting on the aeroplane

at the AC,

(lT is known as the tail moment arm).

The pitching moment is important for performing stability and trim anal-

yses of an aircraft. Cambered aerofoils have nose-down pitching moment

(at zero lift), while symmetric aerofoils have zero moment.

Lwb

Mwb

LT

W = mg

FIGURE 2.7

Simplified diagram of forces and moments acting on a typical aircraft.

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 59

CMcg = C M0 +

( xcg − xac ) C wb − CT ST lT

. (2.53)

L L

c SW c

But

ST

CLwb = CL − CLT (2.54)

SW

CMcg = C M0 +

( xcg − xac ) C − CLT

ST lt

(2.55)

L

c SW c

CG and lT is the tail moment arm. The quantity

ST lt

VT = × (2.56)

Sref c

is known as the tail plane volume ratio. Sometimes a different moment arm is

defined, and correspondingly, there is different volume ratio. Note that lt ≈ lT.

Hence, the aircraft pitching moment coefficient is

CMcg = C M0 + CL

( xcg − xac ) − V C , (2.57)

T LT

c

where Sref is the reference area (usually equal to SW) employed in estimating

the tail plane volume ratio, VT , and the tail plane lift coefficient, CLT , is equal

to the scaled contribution of the tail plane to the lift coefficient, CLT ,

Sref

CLT = CLT = CLT = a1αT + a2η + a3β (2.58)

SW

and

CLwb ⎛ ∂ε ⎞ CL ⎛ ∂ε ⎞

αT = ⎜1− ⎟ + ηT ≈ ⎜1− ⎟ + ηT . (2.59)

a ⎝ ∂α ⎠ a ⎝ ∂α ⎠

www.ebook777.com

free ebooks ==> www.ebook777.com

60 Flight Dynamics, Simulation, and Control

∂CL S ⎛ ∂ε ⎞ ∂CLT ⎛ ∂ε ⎞

= a + T a1 ⎜ 1 − ⎟ ≈ a and = a1 ⎜ 1 − ⎟ . (2.60)

∂α SW ⎝ ∂α ⎠ ∂α ⎝ ∂α ⎠

Hence,

∂CMcg ( xcg − x ac ) a ⎛ ∂ε ⎞ a1 ⎛ ∂ε ⎞

= − VT 1 ⎜ 1 − ⎟ = hcg − hac − VT ⎜ 1 − ⎟ ≡ hcg − hn (2.61)

∂C L c a ⎝ ∂α ⎠ a ⎝ ∂α ⎠

sents the non-dimensional position of the aircraft’s AC, which is

a1 ⎛ ∂ε ⎞

hn = hac + VT ⎜1− ⎟ (2.62)

a⎝ ∂α ⎠

Further, for a fixed position of the stick, the pitching moment equation may

be expressed as

CMcg = C M0 + CL

( xcg − xac ) − V ⎛ a1 ⎛ ∂ε ⎞ ⎞

T ⎜ ⎜1− ⎟ CL + a1ηT + a2η + a3β ⎟ (2.63)

c ⎝ a⎝ ∂α ⎠ ⎠

or as

where

ST T

CL = CLwb + CL . (2.65)

SW

moment equation and plays a key role in the static stability, stick fixed trim

and control of the aircraft.

The elevator hinge moment coefficient may be expressed in a manner similar

to the tail plane lift coefficient as

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 61

where

H

CH = , (2.67)

1

ρU 0 × S flap × c flap

2

where

H is the aerodynamic hinge moment acting on the flap

S flap is the reference area of the flap

c flap is the flap mean chord

Chapter Highlights

• The atmosphere

It has become a norm amongst aeronautical engineers worldwide to

accept a common standard for measuring or considering the proper-

ties of air. Accordingly, an ISA has been established.

• Aerodynamic lift, drag and pitching moment

The lift, pitching moment and drag are typically defined in terms of

non-dimensional lift, pitching moment and drag coefficients.

The relationships between the non-dimensional wing section lift,

wing section pitching moment and wing section drag coefficients,

CL, Cm and CD, and the wing section lift force, L; wing section p

itching

moment, M; and the wing section drag force, D, are

1 1 1

L= ρU 02 × S × CL , M = ρU 02 × S × c × Cm and D = ρU 02 × S × CD .

2 2 2

where

ρ is the density of the free stream

U0 is the velocity of the free stream relative to the aerofoil

c is the aerofoil chord

S is the reference area of the planform of the wing

• Aerodynamic lift

The principal features of aerodynamic lift are as follows:

The relationship has been found to be linear for values of α less than

a critical value.

www.ebook777.com

free ebooks ==> www.ebook777.com

62 Flight Dynamics, Simulation, and Control

Beyond this value of α = αs, the flow over the wing surface tends to

separate, resulting in a loss of lift followed by the aircraft stalling.

The lift coefficient attains a maximum which depends on the aero-

foil section geometry and, to some extent, the wing geometry.

The maximum attainable value of CLmax also depends on the type of

flap and/or slat system, the percent chord dedicated to the flap (or

leading-edge device) and the percent span that is flapped.

For some small negative value of α, the lift coefficient of the wing is

zero. This is the zero-lift angle of attack, α = α 0.

The interference effect of the wing on the tail plane is represented

by the downwash, ε. The tail plane angle of attack, αT, or tail plane

incidence angle is related to the aircraft angle of attack, the tail plane

setting angle, iT, and the downwash angle (ε) the tail plane experi-

ences due to flow over the wing.

• Aerodynamic drag

Aerodynamic drag which consists of two main components: the pro-

file drag and the induced drag.

Profile drag that is itself made up two components:

Form drag (due to the finite shape of the aircraft as the result of the

streamlined flow around its body)

Skin friction drag (due to the viscous friction between the aircraft’s

skin and the airflow around the body both of which) increases in

direct proportion to the square of the airspeed.

The induced drag, to a large extent, is caused by a meeting of the air-

flow emerging from the upper and lower surfaces at the trailing

edge, at a finite angle, resulting in the formation of vortices at the

wing tip. Due to the wing tip vortices being washed away at a faster

rate at higher airspeeds, there is a decrease in the induced drag with

the increase in the speed.

• Pitching moment

The aerodynamic pitching moment is the moment of all the aero-

dynamic pressure forces acting on the surface of the aircraft. The

distribution of pressure on an aerofoil may generate a moment even

if the lift is zero.

The moment is important for performing stability and trim analyses

of an aircraft.

• Aerodynamic centre

The AC is a point along the chord where the pitching moment is

independent of the angle of attack; that is, the derivative of the pitch-

ing moment coefficient, about the AC, with respect to the angle of

attack is zero.

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 63

Exercises

2.1 An aircraft is flying at M = 0.8 and at h = 12,000 m.

(i) What are the air density and the speed of sound at sea level?

(ii) What are the air density and the speed of sound at that altitude?

(iii) What is the air density at an altitude, h = 2000 m?

2.2 An American aircraft is flying at M = 0.8 at h = 30,000 ft.

(i) What are the air density and the speed of sound at sea level?

(ii) What are the air density and the speed of sound at that altitude?

2.3 Assume that the section lift coefficient for a wing section varies span-

wise by the equation

CL = CL 0 1 − η2 , 0 ≤ η ≤ 1, η = y s .

Show that the wing lift coefficient for a uniform wing with constant

chord is

1

2s 2sc0 π

CLw = CLc ( η) dη = CL 0 .

S ∫0

S 4

4s

yCL = .

3π

1

2s 2sc0 ⎛ π (1 − λ ) ⎞

CLw CLc0 ( 1 − η ( 1 − λ ) ) dη = CL 0 ⎜ − ⎟.

=

S ∫

0

S ⎝4 3 ⎠

2.4 A wing of elliptical planform shape (span b = 7.2 m, root chord c0 = 1.2 m)

is flying at angle of attack α = 2°, airspeed V = 30 m/s and altitude

h = 2000 m (where air density ρ = 1.0 kg/m3, sound speed a = 332.5 m/s,

viscosity μ = 1.73 × 10−5 kg/m-s).

The wing cross section is an aerofoil with the following characteristics:

Zero-lift angle of attack: α 0 = −1.2°

Lift curve slope: CLα AR=∞ = 0.112 deg

Drag coefficient at α = 2°: CD0 = 0.01

Moment coefficient: CMac = −0.04

www.ebook777.com

free ebooks ==> www.ebook777.com

64 Flight Dynamics, Simulation, and Control

CLα AR=∞ C2

C Lα = , CD = CD 0 + L

CLα AR=∞ πAR

1+

πAR

where

AR is the aspect ratio

CL is the wing lift coefficient

(ii) Determine the lift and drag coefficients of the wing at α = 2°.

(iii) Determine the lift, drag and pitching moment of the wing at α = 2°.

(iv) Determine the Mach number and the Reynolds number.

2.5 A wing has NACA4415 wing section and a typical trapezoidal plan-

form: root chord = 2.2 m, tip chord = 1.2 m, span = 8 m and leading-edge

sweptback angle = 15°.

(i) Find the mean geometric chord, planform area, aspect ratio, taper

ratio and mean aerodynamic chord.

(ii) Estimate the wing lift curve slope at a Reynolds number = 3 × 106.

(iii) Determine (and plot) the lift curve from zero-lift angle of attack up

to the stall point.

(iv) Using tabulated data, find the maximum lift-to-drag ratio and the

corresponding angle of attack for max CL/CD.

[Hint: See [4]. If these data are not available, assume that CLα AR=∞ = 0.1 deg,

CD0 = 0.0065 and CMac = −0.1, α 0 = −4° and the stall angle is αs = 10°.]

2.6 An aircraft has the following specifications:

Weight, W = 10,000 N; CG location, hcg = 0.5

Wing area, S = 20 m2; horizontal tail plane area = 2 m2

Wing lift curve slope, CLα = 0.06/deg; AC, hac = 0.25

Wing moment coefficient, CMac = −0.05; tail lift curve slope, Clαt = 0.04/deg

Downwash at zero α, ε0 = 0; horizontal tail volume ratio, VH = 0.6

Tail downwash gradient, dε/dα = 0.3; the dynamic pressure at the wing =

480 N/m2; ratio of dynamic pressure ratio at tail to that at wing, ηpr = 1

(i) Find the aircraft AC in terms of the aerodynamic mean chord.

(ii) Find CMα (= dCM/dα) and CM0 .

(iii) Find the absolute angle of attack (from zero lift) and the lift coef-

ficient for trim conditions.

(iv) Find the trim airspeed at sea level (ρ = 1.225 kg/m3).

free ebooks ==> www.ebook777.com

Basic Principles Governing Aerodynamic Flows 65

2.7 A wind tunnel model tail plane wing with a symmetric profile has the

following properties: area = 0.25 m2, flap area = 0.06 m2, flap mean aero-

dynamic chord is 5 cm and the tail plane lift and flap hinge moment

coefficients are a1 = 3.5/rad, a2 = 1.75/rad, a 3 = 0.35/rad, b1 = 0.075/rad,

b2 = −0.015/rad, b3 = −0.03/rad. The wind tunnel wind speed is 60 m/s.

The wing incidence angle is 3° and the tab is set at 5°.

(i) If the flap angle is −4°, what is the tail plane zero-lift angle? What

are the corresponding lift and hinge moment coefficients? What are

the lift force and flap hinge moment?

(ii) At what angle can the flap be expected to move freely?

(iii) What is the lift at this flap angle?

2.8 A n aircraft is flying in steady, level flight. The aircraft properties

are wing area = 40 m2, aerodynamic mean chord = 2.5 m, tail plane

area = 5 m2, tail moment arm = 10 m and the wing lift, tail plane lift

and elevator hinge moment coefficients are a = 4.5/rad, a1 = 2.8/rad,

a2 = 1.2/rad, a3 = 0.3/rad, b1 = 0.01/rad, b2 = −0.012/rad, b3 = −0.03/rad.

The flight conditions are dε/dCL = 0.1, CM0 = −0.1, ηT = −3°, h ac = 0.1,

hcg = 0.2 and LT = 0.01 mg.

(i) Determine the wing and tail plane lift coefficients.

(ii) Determine the downwash angle and the tail plane angle of attack αT.

(iii) Determine elevator setting η, and if the tab angle β, is set to zero.

(iv) Calculate η and β when CH = 0.

2.9 Repeat 2.8 if in addition the wing flaps are down providing 25% of the

lift and if CM0 = −0.2.

2.1 (i) Density, ρ 0 = 1.22505 kg/m3; speed of sound, a0 = 340.3 m/s. (ii) ρ2 =

0.3109 kg/m3, a2 = 295.0741 m/s. (iii) Air density = 1.0066 kg/m3.

2.2 (i) Air density = 23.77 × 10−4 slugs/ft3; speed of sound = 761 mph = 1116.1 ft/s.

(ii) Air density = 8.91 × 10−4 slugs/ft3; speed of sound = 678.1 mph =

994.5467 ft/s.

2.4 (i) CLα = 0.0884/deg. (ii) CL = 0.2829, CD = 0.0133. (iii) L = 863.8663 N,

D = 40.6130 N, Mac = 124.4163 N m. (iv) M = 0.09, Re = 1.8 × 106.

2.5 (i) c = 1.6228 m, S = 12.9822 m2, AR = 4.7057, λ = 0.5455, c = 1.7491 m.

(ii) a = 0.0678/deg. (iv) (L/D)max = 99, αm = 4°.

2.6 (i) hn = 0.53. (ii) dCMcg dα = −0.0186/deg, CM0 = −0.05 . (iii) CL = 1.0417,

α = 8.0131°. (iv) Vtrim ≈ 28 m/s.

www.ebook777.com

free ebooks ==> www.ebook777.com

66 Flight Dynamics, Simulation, and Control

(iii) L = 202.03 N.

2.8 (i) CL = 1.6667, CLT = 0.13336. (ii) ε = 9.5495°, αT = 8.67°. (iii) η = −13.86°.

(iv) β = 9.37°, η = −16.2°.

2.9 (i) CL = 3.3334. CLT = 0.26667 . (ii) αT = 9.73°. (iii) η = −11.96°. (iv) β = 8.02°,

η = −11.94°

References

1. Abbott, I. H. and von Doenhoff, A. E., Theory of Wing Sections; Including a

Summary of Data, Dover, New York, 1958, pp. 490–491.

2. Ashley, H. and Landahl, M., Aerodynamics of Wings and Bodies, Addison-Wesley,

Reading, MA, 1965.

3. Schetz, J. A., Foundations of Boundary Layer Theory for Momentum, Heat and Mass

Transfer, Prentice Hall, Englewood Cliffs, NJ, 1984.

4. Shapiro, A. H., The Dynamics and Thermodynamics of Compressible Fluid Flow,

Vol. 1, Ronald Press, New York, 1953.

free ebooks ==> www.ebook777.com

3

Mechanics of Equilibrium Flight

3.1 Introduction

In this chapter, we consider certain aspects of the mechanics of equilibrium

flight which involves the modelling of steady uniform flight and steady-

state aerodynamic forces. Also considered to a certain extent is the analysis

of performance of the aircraft that is relevant to equilibrium flight, and it

involves quasi-steady models of flight dynamics and aerodynamics which

can also be characterised as the very slowly varying component of the air-

craft’s dynamics. A more detailed presentation of performance aspects may

be found in books dedicated to aircraft performance analysis such as Perkins

and Hage [1], Miele [2], Russell [3], Hull [4], Anderson [5] and Vinh [6].

Consider an aeroplane flying horizontally in still air, in un-accelerated

flight, at a steady angle of attack, α. The relative wind direction is along the

flight path but opposite to the direction of flight. Considering the symmetric

motion of the aircraft, the principal forces acting on the aircraft are the lift

distribution, the drag forces, the weight, the thrust forces generated by the

engines and the moment resulting from the lift.

When the aeroplane is in a state of static equilibrium, there is no net accelera-

tion and all the forces and moments must be in balance. Thus, the sum of all

the forces in both the vertical and horizontal directions must be necessar-

ily equal to zero. Consider an aircraft of mass m and weight mg. Assuming

that the lift is acting vertically up (α = 0), the weight mg is acting vertically

down and that the thrust and drag are in the horizontal plane, and applying

Newton’s first and second laws of motion,

dy 2 dx 2

m = L − mg = 0 , m = T − D = 0. (3.1)

d 2t d 2t

In the climbing mode of flight, a component of the weight acts to oppose the

thrust and the static equilibrium equations are

67

www.ebook777.com

free ebooks ==> www.ebook777.com

68 Flight Dynamics, Simulation, and Control

where γ is the flight path or climb angle. In the case of power gliding, γ is

negative and the equations may be expressed as

In the case of gliding flight, T = 0 and the equilibrium equations are

It follows that the climb angle may be estimated from

L

cot ( − γ ) = . (3.5)

D

In order that there is no net angular acceleration, it is more important for all

the moments acting on the aircraft about its centre of gravity (CG) are equal

to zero. Hence, it follows that

CMcg = 0. (3.6)

There are different ways in which the condition given by Equation 3.6 is realised

in practice. It results in steady equilibrium or trimmed flight and we shall briefly

indicate the principal method of maintaining trimmed equilibrium flight.

A conventional aircraft is trimmed in the presence of pitching moments by

using the horizontal tail to balance the aircraft about the CG. Trailing edge

flaps effectively add camber and can generate extremely large nose-down

moments. These are essentially utilised to control the aircraft.

The centre of pressure (CP), that is, the location at which the lift may be

assumed to act effectively as a concentrated force (meaning the moment about

that point is zero, by definition), and the aerodynamic centre (AC), that is, the

point about which the moment does not change with changes in the angle of

attack, are important points along the chord that must be determined before

an aircraft may be effectively trimmed in symmetric or longitudinal flight.

To consider lateral equilibrium, that is, the equilibrium of forces and moments

in the antisymmetric modes of flight, we first define the sideslip angle, β, in

Figure 3.1a and consider the forces and moments acting on an aircraft in steady

sideslip (Figure 3.1b). The aircraft is in equilibrium in the direction of the side-

slip velocity. Hence, the force balance equation in the y direction is

Y = mg sin ( φ ) , (3.7)

where Y is the side force acting on the whole of the aircraft. The two main

trimmed flight modes in lateral flight correspond to the

1. Steady sideslip

2. Turn coordination and banking

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 69

y

β

V, velocity of

relative wind

L

N

v

z

mg

(a) (b)

FIGURE 3.1

(a) Definition of sideslip angle and (b) aircraft in steady sideslip showing the sideslip velocity, v;

roll angle, ϕ; and relevant moments.

Each of these cases will now be considered independently and conditions for

equilibrium established in each case are also discussed:

1. Steady sideslip: When an aeroplane is banked by the use of the

ailerons alone, the resultant of the inclined lift and weight causes a

sideslip coupled with a loss of height. If the lift is increased, the side-

slip may be made to be horizontal. Hence, this equilibrium mode of

flight pertains to steady level flight at a steady sideslip and a steady

bank angle (Figure 3.2). The force equilibrium conditions are

∑D

Ttrim = trim (3.8a)

trim trim trim )

and mg sin ( φ ) = ∑ Y .

trim trim (3.8b)

y Lift

β

V, velocity of

relative wind

z mg

FIGURE 3.2

Steady level flight at a steady sideslip and a steady bank angle.

www.ebook777.com

free ebooks ==> www.ebook777.com

70 Flight Dynamics, Simulation, and Control

izontal turn, the lift is inclined inwards resulting in a sideslip which

generates an adverse yawing moment due to the larger amount of

airflow ahead of the CG. Moreover, the inertial force, the centrifugal

force, acts outwards. Together with the weight, the three forces must

be in equilibrium. In a sustained and coordinated turn, an aircraft

maintains constant altitude, at a steady tangential velocity (forward

speed), U, with a steady bank angle (i.e. zero roll rate, p), zero sideslip

and a steady rate of turn, Ω. The equilibrium conditions are

Ttrim = ∑D trim , (3.9a)

mg − Ytrim sin ( φbank ) = ∑L trim cos ( φbank ) (3.9b)

and

mUΩ + ∑Y trim cos ( φbank ) = ∑L trim sin ( φbank ), (3.9c)

acting on the aircraft.

Considering the situation illustrated in Figure 3.3, and assuming

that under steady turn conditions, Ytrim = 0, and recalling that Ω = U/R,

the radius of the turn and the turn rate are, respectively, given by

U2 U g tan ( φbank )

R= , Ω= = . (3.10)

g tan ( φbank ) R U

y Lift

Radius of turn, R

mU Ω

z mg

Ω

FIGURE 3.3

Sustained turn at a steady tangential velocity (forward speed), U, with a steady bank angle

(i.e. zero roll rate, p), zero sideslip and a steady turn rate Ω.

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 71

Assuming the lift is given in terms of the aerodynamic lift coefficient as

1

L= ρU 02CLS, (3.11)

2

2mg

U0 ≈ . (3.12)

ρ C LS

2mg

U stall ≈ . (3.13)

ρ CLmax S

One may also define an equivalent sea level airspeed, based on the flow having

the same kinetic energy at sea level; that is,

1 1

ρ0U E2 = ρU 02 . (3.14)

2 2

ρ

UE = U0 . (3.15)

ρ0

Equation 3.15 determines the variation of speed with altitude when the air-

craft is flying level and at constant speed and varies linearly with the square

root of the density ratio.

Considering gliding flight, the vertically down sink speed may be found

from

1

L= ρU 02CLS = mg cos ( − γ ) . (3.16)

2

U Eg ≈ sin ( − γ ) = × . (3.17)

ρ 0 C LS ρ 0 C LS 2 3/ 2

( 1 + ( C L CD ) )

www.ebook777.com

free ebooks ==> www.ebook777.com

72 Flight Dynamics, Simulation, and Control

In the case of climbing flight, we further assume that the thrust is composed

of two components, the direct engine thrust, Te, and the thrust due to the

power delivered to a propeller, P, if and when there is one

dU ηP

L = mg cos γ , m = T − D − mg sin γ = Te + − D − mg sin γ , (3.18)

dt U

where η is the efficiency of the propeller in delivering useful power. The lat-

ter equation may be expressed as

d ⎛ U2 ⎞ d ⎛ U2 ⎞ d

m ⎜ ⎟ + mgU sin γ = U ( Te − D ) + ηP = m ⎜ ⎟ + m ( gh ) , (3.19)

dt ⎝ 2 ⎠ dt ⎝ 2 ⎠ dt

speed in terms of other parameters. In the case of climb with constant speed,

introducing the rate of climb as

uc = U sin γ , (3.20)

ηP + ( Te − D ) U 0 ηP ( Te − D ) U 0 ηP ⎛ Te D cos γ ⎞

uc 0 = = + cos γ = +⎜ − ⎟ U0 .

mg mg L mg ⎝ mg L ⎠

(3.21)

1/ 4

2 ⎞

⎛

ηP ⎜ Te CD ⎛u ⎞ 2mg ⎛ ⎛ uc 0 ⎞2 ⎞

uc 0 = + − 1 − ⎜ c0 ⎟ ⎟ ⎜1− ⎜ ⎟ . (3.22)

mg ⎜ mg CL U

⎝ 0⎠ ⎟ ρC LS ⎜ ⎝ U 0 ⎟⎠ ⎟

⎝ ⎠ ⎝ ⎠

This is an implicit equation that could be solved numerically for uc0. The

maximum achievable vertical velocity of an aircraft is dependent on the alti-

tude at which the aircraft flies. The vertical velocity of an airplane depends

on the flight speed and the inclination of the flight path or the climb angle.

In fact, the rate of climb is the vertical component of the flight path velocity.

When uc0 = 0,

+ = , (3.23)

mg mg ρCLS CL ρCLS

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 73

For the rate of climb to be a maximum, there must exist a significant dif-

ference between power available and minimum power required to climb.

This requirement means that for a given weight of the airplane, the rate of

climb depends on the difference between the power available and the power

required, or the excess power. Hence, when the excess power is zero, the rate

of climb is zero and the airplane is in steady level flight. When the power

available is greater than the power required, the excess power will permit a

rate of climb that is a function of the magnitude of excess power.

The conditions of the airplane’s maximum climb angle or maximum climb

rate occur at specific speeds, and variations in speed will produce variations

in climb performance. The climb performance of an airplane also depends

on certain other variables.

An increase in altitude also will increase the power required and decrease

the power available. Hence, the climb performance of an airplane is affected

greatly by altitude. The speeds for maximum rate of climb, maximum angle

of climb and maximum and minimum level flight airspeeds vary with alti-

tude. As altitude is increased, these various speeds finally converge at the

absolute ceiling of the airplane. At the absolute ceiling, there is no excess of

power and only one speed will allow steady level flight. Consequently, at the

absolute ceiling of the airplane, the rate of climb is zero.

3.3.1 Optimum Flight Speeds

For an aircraft in level flight, based on the total drag curve, one obtains ana-

lytical approximations to a number of related airspeeds. Three points on the

drag curve are particularly important. These are the minimum drag speed,

the minimum power speed and the speed at the maximum speed/drag ratio.

They are illustrated in Figure 3.4. The drag curve in the figure is estimated

for the case of level flight when the lift is maintained a constant. Hence, it

follows that when the drag is a minimum, the ratio of the lift to the drag

(L/D) is a maximum. We have already seen that both the lift and drag could

be expressed in terms of the lift and drag coefficients as

1 1

L= ρU 02 × S × CL and D = ρU 02 × S × CD . (3.24)

2 2

Further, as the total drag is the sum of the profile drag and induced drag, the

drag coefficient may be expressed as the sum of two components: the profile

drag coefficient, CD0, and the induced drag coefficient, CDi:

CD = CD 0 + CDi , (3.25)

www.ebook777.com

free ebooks ==> www.ebook777.com

74 Flight Dynamics, Simulation, and Control

Power Drag Speed/Drag

speed speed ratio speed

Total drag

FIGURE 3.4

Definition of various optimum flight speeds.

CL2

CD = CD 0 + = CD 0 + KCL2 , (3.26)

πeAR

where

AR is the aspect ratio

CL is the wing lift coefficient

e is a constant that is known as Oswald’s efficiency factor

K = 1/πeAR

The minimum drag speed is then obtained by maximising the (L/D) ratio

with respect to the airspeed. Now,

1

ρU 02 × S × CL C

L 2 CL

= = L = , (3.27)

D 1

ρU 02 × S × CD CD CD 0 + KCL

2

2

which is maximum when CD/CL is minimum. Since

CD CD 0 + KCL2 CD 0

= = + KCL , (3.28)

CL CL CL

CD 0

K− = 0. (3.29)

CL2

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 75

CD 0

CL = ≡ ( CL )min drag . (3.30)

K

Hence,

⎛L⎞ 1

⎜ ⎟ = . (3.31)

⎝ D ⎠max 2 KCD 0

Since

1 1

L= ρU 02CLS = ρ0U E2CLS, (3.32)

2 2

and the aircraft is flown at constant lift, the EAS corresponding to minimum

drag is

2L

U min drag = . (3.33)

ρ0 ( CL )min drag S

To obtain the minimum power speed, one must consider the minimum

power required to maintain a constant speed in level flight. The power,

which is the product of the drag and airspeed, is minimum when the drag

decreases as fast as the speed increases. If the drag decreases any faster or

slower, the product of the drag and airspeed is not a minimum when plotted

against the airspeed. Since the airspeed is proportional to 1 CL , the power

required is minimum when CD/(CL)3/2 is minimum. But

CD CD 0 + KCL2 CD 0

3/ 2 = 3/ 2 = 3/ 2 + K CL . (3.34)

( CL ) ( CL ) ( CL )

Equation 3.34 is satisfied when

3CD 0

CL = ≡ ( CL )min power = 3 ( CL )min drag . (3.35)

K

2L U min drag

U min power = = . (3.36)

ρ0 ( CL )min power S 31/4

www.ebook777.com

free ebooks ==> www.ebook777.com

76 Flight Dynamics, Simulation, and Control

To obtain the speed at which the speed to drag ratio is a maximum, one must

consider the minimum drag to speed ratio at a constant lift in level flight.

Since the airspeed is proportional to 1 CL , the drag to speed ratio is a mini-

mum when CD/(CL)1/2 is minimum. But

CD CD 0 + KCL2 CD 0 3

1/2

( CL )

=

( CL )

1/2 = 1/2

( CL )

+K ( CL ) . (3.37)

The earlier expression is a minimum when

CD 0 1

CL = ≡ ( CL )max drag /speed = ( CL )min drag . (3.38)

3K 3

2L U min drag

U min power = = 1/ 4 . (3.39)

ρ0 ( CL )max drag /speed S ⎛1⎞

⎜ ⎟

⎝3⎠

which the aircraft is expected to operate in steady and stable flight. There

are a number of other optimum flight speeds associated with best range,

endurance and cost. These speeds are investigated elsewhere, in the sections

related to the performance of the aircraft.

To minimise the drag in steady level flight, it must be recognised that it is the

sum of the profile drag and induced drag components and that the latter is

proportional to the square of the lift. Moreover, since the lift equals weight

in steady level flight, it must also be a constant. Hence, the drag and lift are,

respectively, expressed as

L2

D = qSCD0 + , L = qSCL = W . (3.40)

πeARqS

respect to the total velocity U is

dq

= ρU . (3.41)

dU

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 77

dD W2 4W 2

= SCD0 − = SC D − . (3.42)

dq πeARq2S πeARρ2U 4S

0

Hence, when the drag is a minimum,

dD 4W 2 1 1 ⎛W ⎞

= SCD0 ρU − = 0 ⇒ q = ρU 2 = ⎜ ⎟ . (3.43)

dV πeARρV S3

2 πeARCD0 ⎝ S ⎠

Hence, the minimum drag is

CD0 W 2

D=2 . (3.44)

πeAR

Moreover, the profile drag and induced drag are equal when the drag is

minimum and each equal to half the total drag. The speed at minimum drag

which is proportional to the square root of the wing loading is given by

12

⎛ 4 ⎞ W

U min drag = ⎜ ⎟ . (3.45)

⎝ πeAR ρ 2

C D0 ⎠ S

When an aircraft is operating just below the minimum drag speed illustrated

in Figure 3.4, any disturbance causing an increase in the speed will reduce

the drag and aid the disturbance in increasing the speed. Thus, the situation

is inherently unstable as the aircraft does not return to its current operat-

ing speed. On the other hand, when it is operating just above the minimum

drag speed, any disturbance causing an increase in the speed will increase

the drag and counter the disturbance to reduce the speed. Thus, the aircraft

returns to its steady operating speed and may be considered to be operating

on the stable side of the drag curve.

For most commercial and military aircraft, the ability to fly either for long

distances or long periods of time is amongst the most important of all design

requirements. The range and endurance which an aircraft can achieve

www.ebook777.com

free ebooks ==> www.ebook777.com

78 Flight Dynamics, Simulation, and Control

its propulsion system, the amount of fuel the aircraft can carry and the way

it is operated. The cruise range and endurance for a jet aircraft are given by

the Breguet equations. These equations are not derived here but by integrat-

ing the equation of motion for steady cruise at constant lift. The equation

for the range is obtained by integrating the product of the cruise velocity at

a particular instant of the flight which is not constant and the time of flight

for unit weight of fuel consumed, over the total weight of fuel consumed.

Furthermore, to estimate the range and endurance, observe that the cruise

thrust may be expressed as T = D = W/(L/D). Hence,

W0 c −W fuel W0 c −W fuel

U cruise dW ( L D )cruise U cruise dW

Rcruise = − , (3.46)

∫

W0 c

tsfccruiseT

=−

tsfccruise ∫

W0 c

W

U cruise = 2W ρSCL . The thrust-specific fuel consumption during cruise tsfccruise

is the ratio of the weight of fuel burned to the thrust delivered per unit time.

Typically the unit of time is in hours and the thrust-specific fuel consumption

during cruise tsfccruise is of the order of 0.3–1.5 h. It is a measure of jet engine

effectiveness at converting fuel to useable thrust and the range is inversely pro-

portional to it. Since the airspeed is proportional to 1 CL , the drag to speed

ratio is a minimum when CD/(CL)1/2 is minimum. It can be shown that the maxi-

mum cruise range is given by

(

2 ( CL )

0.5

CD ) 2

Rcruise =

tsfccruise

cruise

ρS

( )

W0 c − W0 c − W fuel , (3.47)

2 ( L D )cruise 2

Rcruise =

tsfccruise ρSCL*

( )

W0 c − W0 c − W fuel . (3.48)

In these equations, W0c is the initial weight of the aircraft before initiation of

cruise, Wfuel is the weight of fuel consumed and CL* is the lift coefficient dur-

ing cruise.

The endurance is obtained by integrating the time of flight for unit weight

of fuel burned over the total weight of fuel consumed. Hence,

W0 c −W fuel W0 c −W fuel

dW ( L D )cruise dW

Ecruise = − . (3.49)

∫

W0 c

tsfccruiseT

=−

tsfccruise ∫

W0 c

W

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 79

Thus,

( L D )cruise ⎛ W0 c ⎞

Ecruise = ln ⎜⎜ ⎟⎟ . (3.50)

tsfccruise ⎝ 0 c − W fuel

W ⎠

Ecruise = ⎜ − ⎟ , (3.51)

tsfccruise ⎜ Wf

⎝ W0 c ⎟⎠

where ηpr is the efficiency of the propeller.

3.7 Trim

In order to fly an aircraft in steady equilibrium flight, the forces and moments

acting on the aircraft must be in equilibrium. The process of flying an aircraft

in steady equilibrium flight is akin to navigating a ship on a steady course

and hence known as trimming, a term commonly used by ship navigators.

As most aircraft are controlled by a stick which operates a system of pulleys

and cables to control the elevator, two distinct approaches have evolved in

maintaining an aircraft in steady level flight or in trim. The first approach is

for the pilot to apply the necessary stick forces in order to effect the pitching

moment in such a way so as to drive it to zero. This is the stick-fixed approach

as the stick is held in an appropriate position to maintain the aircraft in a

trimmed state. The second approach involves the use of trim tabs, the angles

of which are adjusted so the net force required to hold the stick in a fixed

position is zero. While this requires that the elevator be fitted with trim tabs,

it provides a method for the pilots to fly the aircraft with their hands off the

stick. Hence, it is referred to as the stick-free approach.

The condition of moment equilibrium for the stick-fixed case, correspond-

ing to the case when the pilot holds the stick (and hence the elevator) in an

equilibrium position by applying an appropriate force on it, is

CMcg = 0. (3.52)

The conditions of moment equilibrium for the stick-free case, corresponding

to the case when the pilot adjusts the trim tab to ensure that there is no aero-

dynamic moment on the elevator, are

CMcg = 0, CH = 0. (3.53)

www.ebook777.com

free ebooks ==> www.ebook777.com

80 Flight Dynamics, Simulation, and Control

Thus, in the stick-free case, two conditions, requiring both the pitching

moment on the aircraft and the hinge moment on the elevator to be zero,

must be satisfied simultaneously. From a pilot’s point of view, it is the latter

case that would correspond to a true trimmed state.

Considering the stick-fixed case and employing the pitching moment coef-

ficient equation in the form

where

a1 ⎛ ∂ε ⎞

hn ≡ hac − VT ⎜1− ⎟ , (3.55)

a⎝ ∂α ⎠

η= + CL − 1 T . (3.56)

a2VT a2VT a2

C M0 + CL ( hcg − hn )

a1ηT + a2η + a3β = . (3.57)

VT

ηtrim = − + CL . (3.58)

a2VT a2 a2VT

The corresponding value for the required hinge moment to maintain the

aircraft in trim is

CHsttrim

-fixed st -fixed

= b1αT + b2ηtrim + b3β. (3.59)

Eliminating the tail plane angle of attack and substituting for the trimmed

elevator angle,

CHsttrim

-fixed

= +⎜ + ⎜ b1 − 2 1 ⎟ ⎜ 1 − ⎟ ⎟ CL

a2VT ⎜⎝ a2VT a⎝ a2 ⎠ ⎝ ∂α ⎠ ⎟⎠

⎛ ba ⎞ ⎛ ba ⎞

+ ⎜ b1 − 2 1 ⎟ ηT + ⎜ b3 − 2 3 ⎟ β. (3.60)

⎝ a2 ⎠ ⎝ a2 ⎠

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 81

If we define

b2 a1 ba

b1 ≡ b1 − , b3 = b3 − 2 3 , (3.61)

a2 a2

CHsttrim

-fixed

= + b1ηT + b3β + ⎜ + ⎜1− ⎟ ⎟ CL , (3.62)

a2VT ⎜

⎝ a2VT a⎝ α ⎠ ⎟⎠

∂α

which may be rearranged and expressed as

CHsttrim

-fixed

= + CL b1ηT + b 3β. (3.63)

a2VT a2VT

Considering the stick-free case, and again employing the pitching moment

and hinge moment equations,

⎛ a ⎛ ∂ε ⎞ ⎞

CMcg = CM0 + CL ⎜ hcg − hac − VT 1 ⎜ 1 − ⎟ ⎟ − VT ( a1ηT + a2η + a3β ) , (3.64a)

⎝ a ⎝ ∂α ⎠⎠

CH = b1αT + b2η + b3β, (3.64b)

the conditions for trim are

⎛ a ⎛ ∂ε ⎞ ⎞

C M0 + CL ⎜ hcg − hac − VT 1 ⎜ 1 − ⎟⎟

⎝ a⎝ ∂α ⎠ ⎠

a1ηT + a2η + a3β = (3.65a)

VT

and

where, from Equation 2.49,

CLwb ⎛ ∂ε ⎞ CL ⎛ ∂ε ⎞

αT = ⎜1− ⎟ + ηT ≈ ⎜1− ⎟ + ηT . (3.66)

a ⎝ ∂α ⎠ a ⎝ ∂α ⎠

Using the second condition in the pitching moment equation to eliminate the

elevator angle, the corresponding pitching moment (stick-free) equation is

⎛ a ⎛ ∂ε ⎞ ⎞

(C )

Mcg

stick -free

= CM0 + CL ⎜ hcg − hac − VT 1 ⎜ 1 −

a⎝

⎟ ⎟ − VT ( a1ηT + a3β ) , (3.67a)

∂α ⎠ ⎠

⎝

www.ebook777.com

free ebooks ==> www.ebook777.com

82 Flight Dynamics, Simulation, and Control

where

a2b1 a2b3

a1 ≡ a1 − and a3 = a3 − . (3.67b)

b2 b2

This stick-free pitching moment equation may now be set equal to zero and

solved for the stick-free trim tab angle. The stick-free tail plane lift coeffi-

cient is another important quantity of interest. The tail plane lift coefficient

is given by

(C ) T

L

stick -free

= a1αT + a3β. (3.69)

An aeroplane in steady level flight may be considered to be in a state of equi-

librium as all the forces and moments acting on it are in balance. However,

it is also continually experiencing disturbing forces and moments due to the

appearance of gusts of wind. The stability of the aircraft in an equilibrium

state is its natural or inherent ability to return to the state of equilibrium or

condition of level flight it is in without the pilot having to intervene or take

any action. Static stability is the ability of the aircraft to return to the state of

equilibrium or steady level flight after all the disturbing forces and moments

have been removed. Dynamic stability refers to the entire dynamics of the

motion of the aircraft while it returns to the state of equilibrium or steady

level flight after all the disturbing forces and moments have been removed.

Thus, stability on the whole is concerned with the motion of the aircraft after

all the disturbing forces and moments have been removed. Positive stability

indicates a tendency to return to the original equilibrium state prior to the

initiation of the disturbances.

A high degree of stability makes the aircraft resistant to change and thereby

tends to reduce the aircraft’s controllability, the ease with which the pilot can

manoeuvre the aircraft using the control surfaces. Thus, an aircraft with

extremely good stability characteristics, that is, an ability to return to its undis-

turbed equilibrium state very fast following a transient disturbance, makes it

equally harder for the pilot to control and manoeuvre the aircraft using the

control surfaces. On the other hand, an aircraft that is not unstable but only

marginally stable is a lot easier to fly than an unstable aeroplane that has a

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 83

the aircraft in each and all the operating flight conditions is a primary require-

ment, particularly if hands-off flying of the aircraft in one or several of the

steady level flight conditions over extended periods of time is desirable.

Just as the motion of a rigid body may be conveniently described by ref-

erence to translation along and rotation about three mutually perpendicu-

lar reference axes, the study of the motion and stability of an aircraft can be

conveniently partitioned into three domains. The aircraft longitudinal axis

traditionally runs fore and aft through the aircraft’s CG and is an axis of geo-

metrical symmetry of the aircraft. Rotation of the aircraft about this axis is

referred to as rolling, the axial aerodynamic force acting on the aircraft about

this axis is denoted as X and the corresponding moment about the same axis,

the rolling moment, as L. The stability of the aircraft with reference to this

axis is the lateral stability of the aircraft; it is concerned with the movement

of the aircraft in the lateral plane and rolling about the longitudinal axis. The

lateral axis is the axis passing through the aircraft’s CG and running from

one side of the aircraft to the other. It is perpendicular to the longitudinal axis

and remains in the lateral plane when the aircraft is in a state of equilibrium.

Rotation about this axis is known as pitching, the aerodynamic side force act-

ing along this axis is denoted as Y and the corresponding moment about the

lateral axis, the pitching moment, is denoted as M. Stability with reference

to the lateral axis is referred to as the longitudinal stability and is concerned

with motion perpendicular to the lateral plane and pitching rotation about

the lateral axis. The normal axis, like the other two axes, passes through the

CG but is mutually perpendicular to the other two. Rotation of the aircraft

about this axis is referred to as yawing, the axial normal aerodynamic force

acting on the aircraft about this axis is denoted as Z and the correspond-

ing moment about the same axis, the yawing moment, as N. Stability about

the normal axis is directional stability and is concerned with the directional

stability of the aircraft. It is normally coupled with several aspects of lateral

stability and is usually treated as a special mode of lateral stability.

An aircraft displaced by an external disturbing pitching moment about a lat-

eral axis that is naturally or inherently capable of returning to its undisturbed

steady equilibrium position is said to be longitudinally stable. Although the

lift due to the wing lifting surface is the predominant component of the total

lift acting on the aircraft, the lift on the tail plane is the predominant con-

tributor to the pitching moment. The moment arm of the tail plane lift force

is the distance between the tail plane and the CG of the aircraft. This dis-

tance is responsible for generating an aerodynamic moment when there is

a nose-up disturbing pitching moment acting on the aircraft. However, the

weight of the aircraft functions as a restoring moment only when the CG is

ahead of the AC, that is, the point along the chord where the derivative of

the pitching moment coefficient with respect to the angle of attack is zero,

CM,α ≡ ∂CM/∂α = 0. The total lift and moment acting on the aircraft may be

assumed to act at this point. As any change in the angle of attack does not

www.ebook777.com

free ebooks ==> www.ebook777.com

84 Flight Dynamics, Simulation, and Control

aircraft provides a restoring moment to the aircraft about the AC only when it

is behind the CG and returns the aircraft to its undisturbed equilibrium state.

When an aircraft displaced by an external disturbing yawing moment

about a normal axis is naturally or inherently capable of returning to its

undisturbed steady equilibrium position, it is said to be directionally stable.

Directional or weathercock stability is greatly aided by the vertical fin or

stabiliser, which when displaced generates a side force. The moment of this

side force acts as a restoring moment to return the aircraft to its steady undis-

turbed equilibrium position.

Lateral stability is the natural or inherent ability of the aircraft to recover

from a disturbing torque acting about the longitudinal axis and causing the

aircraft to roll about this axis without any input from the pilot. The dihedral

of the wings, the angle by which each of the two wings on either side of

the fuselage is raised above the horizontal plane, generates a net differen-

tial restoring aerodynamic moment which returns the aircraft to its steady

undisturbed equilibrium position. When one of the wings dips down rela-

tive to the other, it experiences a greater lift as a result of the larger angle

of attack it offers to the airflow, that is, the dihedral effect. Thus, there is a

restoring aerodynamic moment acting on the aircraft.

Returning to the case of longitudinal stability and considering the aerody-

namic moment acting about the CG, the moment acts as a restoring moment

(the disturbance being a change in the angle attack) only when

∂CMcg

< 0. (3.70)

∂α

Since

∂CMcg ∂C Mcg ∂CL

= (3.71)

∂α ∂CL ∂α

and ∂CL/∂α < 0, the condition of ∂C Mcg /∂α < 0 is satisfied when ∂C Mcg/∂CL < 0.

Again, since

∂C Mcg /∂CL < 0 when hcg − hn < 0, that is, when the aircraft’s CG is ahead of the

location of the aircraft’s AC.

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 85

The condition for longitudinal static stability requires that the AC of the air-

craft is behind the CG of the aircraft. The AC of the aircraft is known as the

neutral point (NP) and is defined as the location of the CP of all the incremental

aerodynamic forces generated by both the wing and tail plane when the aircraft

pitches forwards or backwards from a position of equilibrium. The distance of

the NP from the CG is known as the longitudinal static stability margin or sim-

ply the stability margin which must be positive for longitudinal static stability.

The condition for stick-fixed longitudinal stability has been shown in the pre-

vious section to be hcg − hn < 0 and it follows that hcg < hn for stability. Thus,

when the CG of the aircraft is located such that hcg = hn, the aircraft is neutrally

stable and the corresponding location of the CG is known as the stick-fixed NP.

The quantity Hn = hn − hcg is known as the stick-fixed stability margin. It is usu-

ally about 5%–20% of the mean aerodynamic chord in most modern airliners.

Similar to the stick-fixed case, one may define the stick-free NP. To do this,

consider the stick-free pitching moment:

⎛ a ⎛ ∂ε ⎞ ⎞

(C )

Mcg

stick -free

= CM0 + CL ⎜ hcg − hac − VT 1 ⎜ 1 −

a⎝

⎟ ⎟ − VT ( a1ηT + a3β ) . (3.73)

∂α ⎠ ⎠

⎝

a1 ⎛ ∂ε ⎞

hcg − hac − VT ⎜1− ⎟ < 0. (3.74)

a⎝ ∂α ⎠

Corresponding to the stick-free case and as in the stick-fixed case, we may

define

a1 ⎛ ∂ε ⎞

hʹn ≡ hac + VT ⎜1− ⎟ , (3.75)

a⎝ ∂α ⎠

a2b1

a1 ≡ a1 − .

b2

The condition for stick-free longitudinal stability is then given as hcg − hʹn < 0

and it follows that hcg < hʹn for stability. Thus, when the CG of the aircraft is

located such that hcg = hʹn, the aircraft is neutrally stable and the correspond-

ing location of the CG is known as the stick-free NP. The quantity H ʹn = hʹn − hcg

is known as the stick-free stability margin.

www.ebook777.com

free ebooks ==> www.ebook777.com

86 Flight Dynamics, Simulation, and Control

3.10 Manoeuvrability

A major application of the elevator in an aircraft is to manoeuvre the aircraft

in a variety of manoeuvres, some of which are relatively easy to execute, while

others may be extremely complex and difficult to execute. Manoeuvrability

refers to the pilot’s ability to deploy the control surfaces and effectively

manoeuvre the aircraft. Apart from steady level equilibrium flight, there

are other forms of steady equilibrium flight modes involving steady rotation

such as steady level turns and steady pull-outs. The conditions of equilibrium

for these standard manoeuvres can be modified by employing the principle

of d’Alembert and replacing the rotational acceleration terms by equivalent

inertial forces. Thus, the equilibrium equations are modified by replacing g

by (n + 1)g as the aircraft is assumed to be pulling additional ng’s (in addition

to its weight).

Further, the rotational motion gives rise to an additional tail plane inci-

dence angle or roll angle depending on whether it is a pull-out or a steady

horizontal turn. We shall consider the typical case of the pull-out manoeuvre.

The pull out manoeuvre is the principal manoeuvre by which the pilot pulls

up from a dive and then enters a climb or alternatively pushes over a climb

and enters a dive. The basic manoeuvre includes entering a climb or a dive

from steady level flight and is illustrated in Figure 3.5.

R

L

T D

qlT

U0

W

ΔαT

qlT

FIGURE 3.5

The pull out manoeuvre.

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 87

As the aircraft is in rotation there is a net change in the tail plane incidence

angle given by:

⎛ ql ⎞ qlT

ΔαT = tan −1 ⎜ T ⎟ ≈ U , (3.76)

⎝ U0 ⎠ 0

follows that

U 02

ng = , (3.77)

R

U0

q= , (3.78)

R

and we obtain

nglT

ΔαT = . (3.79)

U 02

where Lpo is the total lift in a pull-out manoeuvre. Hence, it follows that

It follows that

CLpo = ( n + 1) Cw = ( n + 1) CL , (3.82)

The manoeuvre margin is a measure of the manoeuvrability of the aircraft. Like

the stability margin, it can be expressed as a distance from the aircraft’s CG.

Reconsider the pitching moment equation derived earlier for the case of

steady level flight. It is

www.ebook777.com

free ebooks ==> www.ebook777.com

88 Flight Dynamics, Simulation, and Control

and η by η + Δη. Hence,

⎛ ⎛ ngl ⎞ ⎞

CMcg = CM0 + ( n + 1) CL ( hcg − hn ) − VT ⎜ a1 ⎜ ηT + 2T ⎟ + a2 ( η + Δη) + a3β ⎟ . (3.84)

⎝ ⎝ U0 ⎠ ⎠

Subtracting the pitching moment equation for steady level flight from the

equation earlier and rearranging,

Δη CL ( hcg − hn ) gl

= − a1 T 2 . (3.85)

n a2VT a2U 0

Hence,

Δη a2VT gl a

= hcg − hn − V T2 1 . (3.86)

n CL U 0 CL

But

1

L= ρU 02CLSW = mg , (3.87)

2

Δη a2VT gl a 1 ρSlT

= hcg − hn − VT T2 1 ρU 02S = hcg − hn − VT a1 . (3.88)

n CL U 0 mg 2 2m

The quantity on the right defines the additional margin required in the

pull-out manoeuvre. Unlike the steady flight equilibrium case, this addition

margin can be provided by appropriately deploying the elevator.

Let μ = m/ρSlT be a non-dimensional relative mass parameter and the ear-

lier equation reduces to

Δη a2VT Va

= hcg − hn + T 1 ≡ hm (3.89)

n CL 2μ

based on the definition of the manoeuvre point and is the distance of the CG

ahead of the manoeuvre point. The manoeuvre point is the farthest NP loca-

tion when the aircraft is pulling the maximum number of g’s. Thus, when

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 89

cg ac nx ne mx me

FIGURE 3.6

Relative location of NPs, CG, the AC and the manoeuvre points.

an extreme manoeuvre, it should remain at least as stable as in steady flight.

Thus, the manoeuvre point must be behind the NP.

Following similar arguments, in the stick-free case, we obtain a similar mod-

ified margin which is given as

Δη a2V Va

− = hʹn − hcg + 1 ≡ H ʹm . (3.90)

n CL 2μ

For a statically stable aircraft, the relative positions of the stick-fixed NP (nx),

the stick-free NP (ne), the stick-fixed manoeuvre point (mx), the stick-free

manoeuvre point (me), the wing–body AC (ac) and the aircraft’s CG (cg) are

illustrated in Figure 3.6.

Spiral or lateral stability of an aircraft may be examined by trimming the

aircraft to level flight then setting it into a 15°–20° bank and observing its ten-

dency to increase or decrease the bank over a period of time, usually under

a minute or two. When both left and right banked turns show no significant

change in bank, the airplane may be deemed to have exhibited neutral spi-

ral stability. If during the test the bank angle decreases at a steady rate, it is

considered to be stable and unstable when there is a steady increase in the

bank angle.

Although the analysis of lateral stability is quite similar in principle to

the longitudinal case, the lateral dynamic features are patently different

from the longitudinal case. In the first instance, the weight and hence the

CG have no direct role to play in lateral stability. Furthermore, the appar-

ent axial symmetry of the aircraft about its longitudinal axis implies that a

www.ebook777.com

free ebooks ==> www.ebook777.com

90 Flight Dynamics, Simulation, and Control

disturbing rolling moment does not generate a large restoring moment. The

influence of the dihedral angle, discussed earlier, breaks this symmetry and

contributes to the restoring moment. Another contributing factor is the fact

that a disturbing rolling moment generates a sideslip due to orientation of

the lift vector, which in turn generates an adverse yawing moment due to the

larger effect of the airflow ahead of the CG. This in turn results in a yaw rate

response which in turn generates a rolling moment. Thus, there is an aero-

dynamic feedback effect. It is clear that, for stability, this effect must not be

regenerative, and hence the condition for lateral stability could be stated as

Lr <

( − N r ) × −L , (3.91)

v

Nv

where Lr, Nr are the derivatives of the dimensionless rolling and yawing

moments with respect to the yaw rate, while Lv, Nv are the derivatives of the

dimensionless rolling and yawing moments with respect to the sideslip. It

will be seen later that this condition is directly related to the stability of the

spiral mode.

Directional or weathercock stability is provided by the fin, although a

larger fin could increase Nv and reduce the lateral stability of the aircraft.

Since Nv is normally positive and Lv is normally negative (hence the negative

signs in the lateral stability condition stated earlier), the condition for direc-

tional stability is

N r < 0. (3.92)

pendent roll rate response, and the requirement that the resulting aerody-

namic moment be a restoring moment implies that

Lp < 0. (3.93)

This particular derivative is influenced by the dihedral effect and the wing

sweep, which break the symmetry and have a stabilising effect, and by the

fin, while the roll–yaw coupling is only influenced by the fin. Moreover, the

fin is also the main contributing factor to the directional stability.

Finally, there is one other parameter that affects the static stability: the

location of the CG in the vertical plane. When the CG is relatively high, in

any free-standing body, it tends to fall down when disturbed from a position

of equilibrium. Aircraft suffer from a similar effect, and the higher the CG in

the vertical plane, the less stable is the aircraft.

A much more detailed analysis of dynamic stability in Chapter 6 shows

that when the aforementioned conditions are met, the aircraft is laterally

stable although the conditions are not sufficient.

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 91

st -fixed

The stick-fixed trim characteristics (ηtrim vs. CL ) can give an indication of

the static stability or instability of an aircraft in steady level flight. These

characteristics can form the basis for an experimental procedure for deter-

mining the location of the stick-fixed NP for a typical general aviation aircraft.

The pitching moment coefficient equation is

st -fixed

Solving for ηtrim , we have

CM0 1 C H

st -fixed

ηtrim = − ( a1ηT + a3β ) − L n , H n = ( hn − hcg ) . (3.96)

a2VT a2 a2VT

d st -fixed H

ηtrim = − n < 0. (3.97)

dCL a2VT

Thus, one could conclude that for a given aircraft and a particular set of

st -fixed

flight conditions, ηtrim is a linear function of the lift coefficient, CL. Hence,

when Hn = 0,

d st -fixed

ηtrim = 0, (3.98)

dCL

d st -fixed H

ηtrim = − n > 0. (3.99)

dCL a2VT

fly the aircraft straight and level at various speeds, recording the elevator

angle to trim. This is repeated for various different positions of the CG. Thus,

to experimentally determine the NP of an aircraft, in the first instance, keep-

ing the total weight of the aircraft a constant, some weights on the aircraft

www.ebook777.com

free ebooks ==> www.ebook777.com

92 Flight Dynamics, Simulation, and Control

are moved from one known location to another so as to change the position

of the CG. The position of the aircraft’s CG is estimated in each case.

When the elevator setting is altered to increase the angle of attack in trim,

the airspeed must be reduced to maintain trim and the corresponding CL may

be estimated since the total weight of the aircraft is unchanged. For each of

the CG locations, the trim characteristics are determined, and these corre-

spond to a plot of CL versus a range of stick-fixed trim settings of the elevator,

st -fixed

st -fixed

ηtrim . A plot of the inverse of the gradient dηtrim dCL of each characteristic

along the y-axis versus the location of the CG along the x-axis is then made.

The plot is usually a straight line and must be extrapolated so it intersects the

x-axis. The intersection on the x-axis corresponds to the CG location for neu-

tral stability, that is, when the CG coincides with the NP. Thus, the NP may be

experimentally obtained. The stability margin is then estimated.

1. Pitching moment equation

CMcg = CM0 + CL

( xcm − xac ) − V C .

T LT

c

CLwb ⎛ ∂ε ⎞ CL ⎛ ∂ε ⎞

αT = ⎜1− ⎟ + ηT ≈ ⎜1− ⎟ + ηT .

a ⎝ ∂α ⎠ a ⎝ ∂α ⎠

b1 ⎛ ∂ε ⎞

CH = b1αT + b2η + b3β, CH = CL ⎜1− ⎟ + b1ηT + b2η + b3β.

a⎝ ∂α ⎠

CMcg = CM0 + CL

( xcg − xac ) − V ⎛ a1 ⎛ ∂ε ⎞ ⎞

T ⎜ ⎜1− ⎟ CL + a1ηT + a2η + a3β ⎟

c ⎝ a⎝ ∂α ⎠ ⎠

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 93

or

where

a1 ⎛ ∂ε ⎞

hn ≡ hac + VT ⎜1− ⎟.

a⎝ ∂α ⎠

η= + CL − 1 T .

a2VT a2VT a2

CH =

(

b2 C M0 − C Mcg ) +C b2 ( hcg − hʹn )

+ b1ηT + b3β,

L

a2VT a2VT

where

b2 a1 b2 a3 a ⎛ ∂ε ⎞

b1 ≡ b1 − and b3 = b3 − ; hʹn ≡ hac + VT 1 ⎜ 1 − ⎟

a2 a2 a⎝ ∂α ⎠

and

a2b1 a2b3

a1 ≡ a1 − and a3 = a3 − .

b2 b2

CH b1 T b3

Rearrange CH in (7) for η, η = − α − β, and eliminate αT,

b2 b2 b2

CH b1 CL ⎛ ∂ε ⎞ b1 b3

η= − ⎜1− ⎟ − ηT − β.

b2 b2 a ⎝ ∂α ⎠ b2 b2

Solve for β:

CH b1 CL ⎛ ∂ε ⎞ b1 b2

β= − ⎜1− ⎟ − ηT − η.

b3 b3 a ⎝ ∂α ⎠ b3 b3

www.ebook777.com

free ebooks ==> www.ebook777.com

94 Flight Dynamics, Simulation, and Control

a2

CMcg = CM0 − VT CH + CL ( hcg − hʹn ) − VT ( a1ηT + a3β ) .

b2

9. Trim

Stick-fixed: CMcg = 0.

ηtrim = − + CL ,

a2VT a2 a2VT

C stH trim

-fixed

= + CL + b1ηT + b3β.

a2VT a2VT

Stick-free: CMcg = 0, CH = 0.

b1 ⎛ ∂ε ⎞

( CH )stick -free = 0 = CL ⎜1− ⎟ + b1ηT + b2η + b3β

a⎝ ∂α ⎠

= + CL + b1ηT + b3β,

a2VT a2VT

st -free b1 ⎛ ∂ε ⎞ b1 b3

ηtrim = −CL ⎜1− ⎟ − ηT − β,

ab2 ⎝ ∂α ⎠ b2 b2

βtrim =− − CL − ηT ,

a2b3VT a2b3VT b3

(C )

Mcg

stick - free

= CM0 + CL ( hcg − hʹn ) − VT ( a1ηT + a3β ) = 0,

(C ) T

L

stick -free

= a1αT + a3β.

a1 ⎛ ∂ε ⎞

Stick-fixed: hcg − hn < 0, hn ≡ hac + VT ⎜1− ⎟.

a⎝ ∂α ⎠

a1 ⎛ ∂ε ⎞

Stick-free: hcg − hʹn < 0, hʹn ≡ hac + VT ⎜1− ⎟.

a⎝ ∂α ⎠

11. Stability margins

Stick-fixed: Hn = hn − hcg.

Stick-free: H ʹn = hʹn − hcg .

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 95

Δη a2V Va

Stick-fixed: H m ≡ − = hn − hcg + 1 .

n CL 2μ

Δη a2V Va

Stick-free: H ʹm ≡ − = hʹn − hcg + 1 .

n CL 2μ

Chapter Highlights

• Equilibrium flight

In order to fly an aircraft in steady equilibrium flight, the forces and

moments acting on the aircraft must be in equilibrium.

The condition of moment equilibrium for the stick-fixed case, corre-

sponding to the case when the pilot holds the stick (and hence the

elevator) in an equilibrium position by applying an appropriate force

on it, is CMcg = 0.

The conditions of moment equilibrium for the stick-free case, cor-

responding to the case when the pilot adjusts the trim tab to

ensure that there is no aerodynamic moment on the elevator, are

CMcg = 0, CH = 0.

The AC of the aircraft (or the NP) is the location of the CP of all aero-

dynamic forces generated by both the wing and tail plane when the

aircraft pitches forwards or backwards from a position of equilib-

rium, and the CG, in the stick-fixed case, is the stick-fixed NP, while in

the stick-free case, it is the stick-free NP.

• Stability

Stability is the inherent ability of aircraft to return to initial equi-

librium position (uniform motion) sufficiently quickly after being

disturbed.

Static stability is the ability of the aircraft to return to the state of

equilibrium or steady level flight after all the disturbing forces and

moments have been removed.

Dynamic stability refers to the entire dynamics of the motion of the

aircraft while it returns to the state of equilibrium or steady level

flight after all the disturbing forces and moments have been removed.

Disturbances (such as atmospheric turbulence) change the angle of

attack (Δα) or angle of sideslip, resulting in changes in the pressure

distribution on airplane’s component lifting surfaces (wing, fuselage,

horizontal tail and vertical tail) and consequently change the lift, side

force, pitching moment, rolling moment and yawing moment.

www.ebook777.com

free ebooks ==> www.ebook777.com

96 Flight Dynamics, Simulation, and Control

location (which determines arm lengths with respect to CG and

local dynamic pressure) of airplane component lifting surfaces are

designed to achieve stability.

A measure of the stability characteristic of an aircraft is the distance of

the AC of the aircraft (or the NP), the location of the CP of all aerodynamic

forces generated by both the wing and tail plane when the aircraft pitches

forwards or backwards from a position of equilibrium, and the CG.

This is known as the longitudinal static stability margin or simply the

stability margin.

For a statically stable aircraft, the relative locations of the stick-fixed

NP (nx in Figure 3.6), the stick-free NP (ne), the stick-fixed manoeu-

vre point (mx), the stick-free manoeuvre point (me), the wing–body

AC (ac) and the aircraft’s CG (cg) are important.

• Speed stability

When an aircraft is operating just below the minimum drag speed,

any disturbance causing an increase in the speed will reduce the drag

and aid the disturbance in increasing the speed. Thus, the situation

is inherently unstable as the aircraft does not return to its current

operating speed. On the other hand, when it is operating just above

the minimum drag speed, any disturbance causing an increase in the

speed will increase the drag and counter the disturbance to reduce

the speed. Thus, the aircraft returns to its steady operating speed and

may be considered to be operating on the stable side of the drag curve.

• Control

Control is affected by the pilot through deflection of control surfaces.

Deflecting control surface changes lift or side force and produces

moment about the CG.

A high degree of stability makes the aircraft resistant to change and

thereby tends to reduce the aircraft’s controllability, the ease with

which the pilot can manoeuvre the aircraft using the control surfaces.

Thus, an aircraft with extremely good stability characteristics, that is,

an ability to return to its undisturbed equilibrium state very fast fol-

lowing a transient disturbance, makes it equally harder for the pilot

to control and manoeuvre the aircraft using the control surfaces.

On the other hand, an aircraft that is not unstable but only margin-

ally stable is a lot easier to fly than an unstable aeroplane that has a

natural tendency to diverge from the trimmed equilibrium state.

Manoeuvrability refers to the pilot’s ability to deploy the control sur-

faces and effectively manoeuvre the aircraft, and the manoeuvre mar-

gin is a measure of the manoeuvrability of the aircraft. Like the stability

margin, it can be expressed as a distance from the aircraft’s CG.

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 97

Exercises

3.1 (i) Consider an American flying wing aircraft made using a National

Advisory Committee on Aeronautics (NACA) 2412 aerofoil with

a wing area of 250 ft 2, a wing span of 50 ft and a span efficiency

factor of 0.9. If the aircraft is flying at a 6° angle of attack and with

a Reynolds number of approximately 9 × 106, what are the CL and

CD for the flying wing?

(ii) If the flying wing is flying at sea level at V∞ = 280 ft/s, how much lift

is it generating and how much drag is it experiencing?

(Hint: If no data sheets are available for the NACA 2412 aerofoil sec-

tion, use CD0 = 0.0098, α 0 = −2° and CLα AR =∞ = 0.105 deg .)

3.2 An American aircraft with CD0 = 0.02 and K = 0.12 is flying at M = 0.8 at

h = 30,000 ft. If the aircraft has a wing area of 375 ft2 and is generating

25,000 lb of lift force, what is its drag coefficient and how much drag is it

generating?

3.3 (i) Starting with the most general expression for the range,

W0 −W fuel

UdW

R=− ,

∫

W0

tsfcT

W0 −W fuel

dW

E=− ,

∫

W0

tsfcT

show that the maximum range and endurance at constant speed for

jet-powered aircraft are given by

W0 −W fuel

U ⎛ CL ⎞ dW U ⎛ CL ⎞ ⎛ W0 ⎞

Rmax log e ⎜⎜

=−

tsfc ⎜⎝ CD ⎟⎠max ∫

W0

W

= ⎜ ⎟

tsfc ⎝ CD ⎠max ⎝ W0 − W f

⎟⎟

⎠

and that

W0 −W fuel

1 ⎛ CL ⎞ dW 1 ⎛ CL ⎞ ⎛ W0 ⎞

Emax log e ⎜⎜ ⎟⎟ .

=−

tsfc ⎜⎝ CD ⎟⎠max ∫

W0

W

= ⎜ ⎟

tsfc ⎝ CD ⎠max ⎝ W0 − W f ⎠

www.ebook777.com

free ebooks ==> www.ebook777.com

98 Flight Dynamics, Simulation, and Control

fuel consumption rate, psfc. In this case, the general expressions for

the range and endurance are

W0 −W fuel

ηpr dW

R=− ,

W0

∫ psfcT

W0 −W fuel

ηpr dW

E=− .

∫

W0

UpsfcT

Show that the maximum range and endurance at constant speed for a

propeller-driven aircraft are given by

W0 −W fuel

ηpr ⎛ CL ⎞ dW ηpr ⎛ CL ⎞ ⎛ W0 ⎞

Rmax log e ⎜⎜

=−

psfc ⎜⎝ CD ⎟⎠max ∫

W0

W

=

psfc ⎜⎝ CD ⎟⎠max ⎝ W0 − W f

⎟⎟

⎠

and that

W0 −W fuel

ηpr ⎛ CL ⎞ dW ηpr ⎛ CL ⎞ ⎛ W0 ⎞

Emax log e ⎜⎜ ⎟⎟ .

=−

Upsfc ⎜⎝ CD ⎟⎠max ∫

W0

W

= ⎜ ⎟

Upsfc ⎝ CD ⎠max ⎝ W0 − W f ⎠

aircraft has a mass of 60,000 kg of which 20% is fuel. The wing area is

250 m2 . The drag coefficient is given by

CD = 0.021 + 0.051CL2 .

fuel consumption is 10−6 N/J. Determine the maximum range and

endurance for constant velocity flight.

(b) If the thrust-specific fuel consumption of the same aircraft pow-

ered by a jet engine is 1.4 × 10−5/s, determine the maximum range

and endurance for constant velocity flight.

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 99

its weight is 6000 lb (500 lb of this is usable fuel), the thrust-specific

fuel consumption of its jet engine is tsfcsl = 0.9/h, and its wing area is

S = 184 ft 2. Find

(

(i) The value of (L/D)max and the value of CL0.5 CD ) max

for this aircraft

(ii) The max endurance and the corresponding airspeed both at sea

level and at 20,000 ft

(iii) The max range and the corresponding airspeed both at sea level and

at 20,000 ft

3.5 An aircraft has the following specifications:

Weight, W = 10,000 N; CG location from wing root leading edge, hcg = 0.5

Wing area, S = 20 m2; wing lift curve slope, CLα = 0.06/deg

AC, hac = 0.25; wing moment coefficient, CMac = −0.05

Wing setting angle, iw = 0; downwash at zero α, ε0 = 0

Horizontal tail volume ratio, VH = 0.6; tail lift curve slope, Clαt = 0.04/deg

Zero-lift angle of attack, α 0 = −1.1°; tail downwash gradient, dε/dα = 0.3;

the dynamic pressure at the wing = 480 N/m 2; ratio of dynamic pres-

sure at tail to that at wing, η pr = 1; tail setting angle, iT = 3° (leading

edge down):

(i) Estimate the angle of the tail plane zero-lift line.

(ii) Determine the tail plane angle of attack.

(iii) Find the aircraft stick-fixed NP and the stick-fixed stability margin.

Is the aircraft stable?

(iv) Determine the elevator angle in the trimmed state given that

a2 = 1.8/rad.

3.6 An aircraft has the following specifications:

The CG is located 0.45c behind the leading edge of the wing, the AC of

the wing–body is at 0.25 c , the tail volume ratio is 0.4, the wing setting

angle is zero, zero-lift angle of attack is α 0 = −1.1°, the wing lift curve slope

is 0.08/deg, the tail lift curve slope is 0.07/deg, ∂ε/∂α = 0.3, CMac = −0.05,

the tail setting angle is 3°, and the downwash angle at zero lift is zero.

The weight is 12,000 N-s, the wing area is 21 m2, and the aircraft is flying

at sea level conditions.

(i) Calculate the NP.

(ii) Calculate the static margin. Is this aircraft stable?

3.7 The wing is known to have an area of Sw = 16 m2, aspect ratio AR = 4 and

an aerodynamic rolling moment derivative with respect to the roll rate,

Lp = −0.85. Assuming a uniform lift distribution along the span and a lift

www.ebook777.com

free ebooks ==> www.ebook777.com

100 Flight Dynamics, Simulation, and Control

coefficient slope of a∞ = 6.1/rad, calculate the wing taper ratio λ and the

centre chord length c0m. You should assume that

s

a

Lp = − ∞ 2

2Sw s ∫ c y dy,

0

2

3.8 An aircraft has the following characteristics: CM0 = −0.03; tail plane

volume coefficient, VT = 0.6, a = dCLwb dα = 4.6 (per radian); the rate of

change of the tail plane lift coefficient with tail plane incidence in radians,

a1 = 3.0; the rate of change of tail plane lift coefficient with elevator angle

in radians, a2 = 1.5; the rate of change of the mean downwash over the

tail plane with wing incidence, dε/dα = −0.5; the wing zero-lift incidence

angle, α0 = −2°; and the tail plane setting relative to wing datum line = 0°.

The position of the AC of the aircraft without the tail plane is 0.05c in

front of the aircraft’s CG.

Find the elevator angle required to trim at a wing incidence of 5° with

the tab set, such that tab angle is 0°.

3.9 An aircraft has a total mass of 10,000 kg and is flying at sea level at a

speed of 150 m/s. The wings’ span is 12 m and the aspect ratio is 5. The

aerodynamic mean chord (amc) is 1.5 m, the tail plane area is 12 m2 and

the tail arm lT is 5 m. The wing lift curve slope is 5.7/rad. The coefficients

in the pitching moment equation and the coefficients defining the tail

plane lift and the hinge moment are given in Tables 3.1 and 3.2.

(i) Find the wings’ area Sw and the total lift coefficient CL.

(ii) Find the tail plane lift coefficient CLT .

(iii) Find the downwash angle and the tail plane incidence assuming the

ground effect reduces the downwash angle ε by 1°.

(iv) Assuming a stick-free trim, find the elevator angle η, the hinge

moment coefficient CH and the tab angle β.

TABLE 3.1

Coefficients in the Pitching Moment Equation

dε/dα C M0 ηT hcg hac

0.5 −0.01 2 0 0.2 0.05

TABLE 3.2

Coefficients Defining the Tail Plane Lift and Hinge Moment

a1 a2 a3 b1 b2 b3

3.1/rad 1.2/rad 0.5/rad 0.1/rad −0.5/rad −0.01/rad

free ebooks ==> www.ebook777.com

Mechanics of Equilibrium Flight 101

3.10 Show that the tab angle to trim an aircraft in the stick-free condition is

CM0 a1 C Hʹ

βtrim = − ηT − L n .

VT a3 a3 VT a3

3.11 An aircraft is flying in steady equilibrium or in trim. The lift coefficient

CL = 0.3 and the elevators are set in the neutral position, η = 0, and the tab

angle is β = 0. The tail plane volume ratio is VT = 0.48; the lift curve slope

a = 4.5, a1 = 2.8/rad, a2 = 1.2/rad, dε/dα = 0.4; the AC position is hac = 0.18;

the CG position is hcg = 0.27; and CM0 = −0.016. The elevator has travel

limits of ±30°.

(i) Find the tail plane zero-lift line angle, ηT.

(ii) Show that the most forward position of the CG for which the

trimmed lift coefficient is 1.265 is 0.1.

3.12 An aircraft has the following characteristics:

Wing area S = 10 m2; the wing lift curve slope is a = 4.8, hcg−hac = 0.18;

the amc c = 1.5 m, dε/dα = 0.47; the tail moment arm lT = 10 m; and the

coefficients of the tail plane lift and the hinge moment are a1 = 3/rad,

a2 = 1.8/rad, a3 = 0.3/rad, b1 = 0.01/rad, b2 = −0.06/rad, b3 = −0.03/rad.

Show that the tail plane area required to give a stick-free CG margin

of 0.15 is 1.36 m2.

3.13 At an altitude where the relative density to sea level is 0.61, and the

aircraft speed is 180 m/s, the stick-free tab angle to trim is zero. Given

the speed of the aircraft is 85 m/s, the wing loading per unit area is

3000 N/m2, the tail plane volume ratio is V T = 0.55 and that a2 = 2.3/rad,

a3 = 0.5/rad, b2 = −0.15/rad, b3 = −0.003/rad, show that the stick-free tab

angle to trim for a 5% stick-free CG margin where the relative density

to sea level is 0.74 is −7.67°.

3.14 An aircraft performs a pull-out manoeuvre at 150 m/s at a low level

while pulling an excess normal acceleration of ng’s. The aircraft’s mass is

65,000 kg; the wing lift curve slope is a = 4.5; the wing area is 190 m2; the tail

plane volume ratio is VT = 0.49; the tail moment arm lT = 14 m, dε/dα = 0.49;

the AC position is hac = 0.16; and the CG position is hcg = 0.25, a1 = 3.7/rad,

a2 = 2/rad. Show that the change in the elevator angle from level flight

is −2.29°/g.

3.1 (i) CL = 0.7319, CD = 0.03476. (ii) L = 17049.32 lb, D = 792.34 lb.

3.2 CD = 0.0267, D = 2824.34 lb.

www.ebook777.com

free ebooks ==> www.ebook777.com

102 Flight Dynamics, Simulation, and Control

3.3c (i) Rmax = 1302.9 km, Emax = 7.38 h. (ii) Rmax = 5189.3 km, Emax = 29.376 h.

3.4 (i) ( CL CD )

max

= 14.81, ( C L CD ) max

= 21.93. (ii) Ecruise = 1.432 h. (iii) At sea

level, Rcruise = 234.31 miles, Ucruise = 262.59 ft/s = 179.039 mph. At 20,000 ft,

Rcruise = 320.94 miles, Ucruise = 359.67 ft/s = 245.23 mph.

3.5 (i) ηT = −4.1°. (ii) αT = 8.05°. (iii) hn = 0.53; Hn = 0.03. The aircraft is stable.

(iv) η = −14.01°.

3.6 (i) hn = 0.495. (ii) Hn = hn − hcg = 0.045; stable.

3.7 λ = 0.506.

3.8 ηtrim = −0.364( = −20.9°).

3.9 (i) Sw = 28.8 m2; CL = 0.2472. (ii) CLT = 0.1948 . (iii) ε = 0.004234 = 0.2426°,

αT = 0.074046 = 4.2425°. (iv) η = 0.0170163 = 0.975°, β = 6.326°.

References

1. Perkins, C. and Hage, R., Aircraft Performance, Stability and Control, John Wiley &

Sons, London, U.K., 1949.

2. Miele, A., Flight Mechanics: Theory of Flight Paths, Addison-Wesley, New York,

1962.

3. Russell, J. B., Performance and Stability of Aircraft, Arnold, London, U.K., 1996.

4. Hull, D. G., Fundamentals of Airplane Flight Mechanics, Springer International,

Edition, Springer, Berlin, Germany, 2007.

5. Anderson, J., Aircraft Performance and Design, McGraw Hill, New York, 1999.

6. Vinh, N., Flight Mechanics of High Performance Aircraft, Cambridge University

Press, New York, 1993.

free ebooks ==> www.ebook777.com

4

Aircraft Non-Linear Dynamics:

Equations of Motion

4.1 Introduction

Although several excellent books have been written on the subject of air-

craft dynamics—including Etkin and Reid [1], Seckel [2], McRuer, Ashkenas

and Graham [3], Smetana [4], Nelson [5], Cook [6] and Schmidt [7]—in this

chapter, the analysis of aircraft dynamics is revisited to bridge the gap

between the dynamics analysis and control. Moreover, modern aircraft are

generally more flexible than older aircraft which were essentially rigid, and

this requires that certain key principles are restated so that they are relevant

in the wider context.

The flight stability problems may be classified into two major groups: static

and dynamic. When the aircraft is statically stable and a static model of the

forces and moments acting on the aircraft is adequate to analyse the prob-

lem, the associated stability problem may be considered to be static. On the

other hand, the analysis of dynamic stability often requires an extensive

dynamic model. The aircraft’s dynamic response may be further classified

as longitudinal or lateral. Longitudinal responses are those which are con-

fined to the plane of symmetry of the aircraft while the lateral responses

are those which displace the plane of symmetry. In fact, as far as the lat-

eral motions are concerned, a sideslip velocity or a yawing angular veloc-

ity perturbation from equilibrium flight causes both yawing and rolling

103

www.ebook777.com

free ebooks ==> www.ebook777.com

104 Flight Dynamics, Simulation, and Control

moments, and lateral motions of the aircraft are thus characterised by two

distinct occasionally unstable motions: a spiralling, slow, non-oscillatory

motion known as the spiral mode and a yawing oscillation accompanied

by sustained rolling oscillations, called the Dutch roll. In addition, there is

also a stable, short period, pure roll subsidence mode. Motions in the plane

of symmetry or longitudinal aircraft responses are characterised in the

main by a lightly damped long-period oscillation or phugoid motion with

the aircraft’s orientation to the velocity vector, being almost a constant.

Also plausible is a short period motion normally characterised by a damped

high-frequency oscillation which may be considered to be the longitudinal

counterpart of the Dutch roll.

One of the simplest models of the aircraft’s motion in a 2D plane is based on

modelling the aircraft as a point mass with its translational motion described

in natural coordinates and as a rigid body, as far as rotational motion in the

plane. The motion of the aircraft is assumed to be restricted to pitching and

displacement in the plane of symmetry.

Referring to the forces and moments and the orientation of the reference

axes as shown in Figure 4.1, we may apply the laws of force and moment

equilibrium to obtain the equations of motion:

= M + Td (4.1b)

Iθ

Lift

Thrust

Ue

θ

γ

Drag

mg

FIGURE 4.1

Reference axes for motion in a plane.

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 105

where L, D and M are the aerodynamic lift, drag and nose-up pitching moment

acting on the aircraft, given in terms of the corresponding aerodynamic

coefficients by

⎛1⎞ ⎛1⎞

L = ⎜ ⎟ ρU 2SCL ( α , η) , D = ⎜ ⎟ ρU 2SCD ( α , η) (4.2a)

⎝2⎠ ⎝2⎠

and

⎛1⎞ ⎛ ⎛ qc ⎞ ∂ ⎛ α c ⎞ ∂ ⎞

M = ⎜ ⎟ ρU 2Sc ⎜ Cm ( α , η) + ⎜ ⎟ Cm ( α , η ) + ⎜ ⎟ Cm ( α , η) ⎟.

⎝2⎠ ⎜

⎝ ⎝ U ⎠ ∂ ( qc U ) ⎝ U ⎠ ∂ (αc U ) ⎟

⎠

(4.2b)

In Equations 4.1 and 4.2,

I is the moment of inertia of the aircraft about the pitch axis

γ is the flight path angle

θ is the pitch angle

U is the magnitude of the velocity vector

T is the thrust

δ is the direction of the thrust vector relative to the horizontal

Td is the pitching moment due to the thrust

The difference between the pitch angle and the flight path angle is the angle

of the aircraft’s forward axis relative to the wind velocity vector and is the

angle of attack. The angle of attack, the pitch rate and height rate are

The simplified lateral kinematic and dynamic equations for the lateral

motion of the aircraft in a plane perpendicular to the plane of symmetry

take the form

The east and north velocities of the aircraft may then be expressed in terms

of the aircraft velocity, the aircraft attitudes, the wind velocity and the wind

direction as

Veast = V cos γ cos ψ − Vwind cos χ wind , Vnorth = V cos γ sin ψ − Vwind sin χ wind . (4.5)

steady level flight, that is,

δ = γ = 0, L = mg , D = T , M = 0. (4.6)

www.ebook777.com

free ebooks ==> www.ebook777.com

106 Flight Dynamics, Simulation, and Control

in use. These are

2. The equilibrium model

The equilibrium model is usually quite adequate in cruise while the non-

equilibrium model provides useful and complete information during the

climb and descent phases of the flight. The non-equilibrium model exploits

the concept of energy height, he, which is defined as the sum of the potential

and kinetic energies of the aircraft per unit weight and is given by

⎛ 1 2⎞

E ⎜ mgh + mU ⎟

⎠ = h + U (4.7)

2

2

he = =⎝

mg mg 2g

where

m is the aircraft mass

U is the true airspeed

h is the altitude

The rate of change of the energy height is known as the specific excess power,

Pe, and is

dhe dh U dU

Pe = = + . (4.8)

dt dt g dt

Using the velocity U and the flight path angle γ, as natural coordinates, δ = γ,

and referring to Figure 4.1, the equations of motion may be written as

dU

m = T − D − mg sin γ (4.9a)

dt

dγ

mU = L − mg cos γ. (4.9b)

dt

U dU ⎛T −D⎞

+ U sin γ = U ⎜ ⎟ . (4.10)

g dt ⎝ mg ⎠

But the rate of climb is given by

dh

= U sin γ. (4.11)

dt

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 107

dhe dh U dU ⎛T −D⎞

Pe = = + =U⎜ ⎟ . (4.12)

dt dt g dt ⎝ mg ⎠

The concepts of energy height and specific excess power are useful in determin-

ing how best to climb to a predetermined altitude and airspeed.

Assuming level flight, that is, γ = 0 and that T = D,

dU

m = mx = 0 (4.13a)

dt

dγ

mU = mΔh = L − mg = L − Le (4.13b)

dt

1

where Le = mg. If we express the lift as L = ρU 2 ACL, assuming that the lift

2

1

coefficient, CL, is constant and that Le = ρU e2 ACL , we may write

2

or

ρgACL ρg 2 ACL

Δh = − Δh = − Δh = −ω2ph Δh. (4.15)

m Le

model, the rotational kinetic energy of the aircraft is assumed to be negli-

gible and the following energy balance equation may be obtained:

1 1

mU 2 + mgΔh = mU e2 . (4.16)

2 2

Further, resolving the forces acting on the aircraft in the vertical and hori-

zontal directions,

www.ebook777.com

free ebooks ==> www.ebook777.com

108 Flight Dynamics, Simulation, and Control

mΔh = L − mg = L − Le , (4.18a)

mx = 0, (4.18b)

1

L= ρU 2 ACL , (4.19)

2

1

Le = ρU e2 ACL , (4.20)

2

we may write

or

ρgACL ρg 2 ACL

Δh = − Δh = − Δh = −ω2ph Δh, (4.22)

m Le

This equation gives a first approximation to the natural frequency of the

aircraft and corresponds to the so-called phugoid mode which is

2g

ωph = , (4.23)

Ue

where

2mg

U e2 = . (4.24)

ρACL

flight path is sinusoidal. The sum of the potential energy due to the air-

craft’s altitude and the kinetic energy is conserved. Thus, when the aircraft

gains altitude, the velocity of the aircraft reduces and vice versa.

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 109

The definition and evaluation of the moments of inertia are primarily impor-

tant in the development of the governing equations of rotational motion of a

rigid body such as an aircraft.

Consider a continuous body B as shown in Figure 4.2.

The moment of inertia of the body B about an axis OP is defined by the

integral

∫

V

where

dm is an element of mass at an arbitrary point in the body with coordinates

x, y, z

h is the perpendicular distance from the axis OP to the point

The integral is over the entire volume, V, of the body B.

2

h 2 = ( r ⋅ r ) − ( r ⋅ e ) , (4.26)

where

r = xi + yj + zk (4.27)

r dm

B

y

O

p q

FIGURE 4.2

Axis system and notation for evaluating moments of inertia.

www.ebook777.com

free ebooks ==> www.ebook777.com

110 Flight Dynamics, Simulation, and Control

e = e x i + e y j + e z k (4.28)

2

h 2 = ( r ⋅ r ) ( e ⋅ e ) − ( r ⋅ e ) = e ⋅ ⎡⎣e ( r ⋅ r ) − r ( r ⋅ e ) ⎤⎦ = e ⋅ ( r × ( e × r ) ) . (4.29)

∫ (( r ⋅ r ) (e ⋅ e) − ( r ⋅ e) ) dm. (4.30)

2

IOP =

V

Hence, the moment of inertia of the body about the axis OP may be

expressed as

IOP =

∫ (( x

2

)( ) 2

)

+ y 2 + z 2 e x2 + e y2 + e z2 − ( xe x + ye y + ze z ) dm. (4.31)

V

To reduce the integral further, we define the six moments of inertia integrals

as in Table 4.1.

In terms of the integrals defined in Table 4.1, we may define the moment of

inertia of the body about the axis OP as

IOP = I xx e x2 + I yy e y2 + I zz e z2 − 2I xy e x e y − 2I xz e x e z − 2I yz e y e z . (4.32)

TABLE 4.1

Moment of Inertia Integrals

Moment of Inertia Product of Inertia (PI)

Integral (MI) Reference Axis Integral Reference Axes

I xx =

∫(y )

+ z 2 dm I xy = xydm

2

V

MI about the Ox axis

∫

V

PI about the Ox, Oy axes

I yy =

∫ (x )

+ z 2 dm I xz = xzdm

2

V

MI about the Oy axis

∫

V

PI about the Ox, Oz axes

I zz =

∫ (x )

+ y 2 dm I yz = yzdm

2

V

MI about the Oy axis

∫

V

PI about the Oy, Oz axes

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 111

The moment of inertia of the body about the axis OP may also be expressed

as the matrix product

⎡ I xx −I xy −I xz ⎤ ⎡ e x ⎤ ⎡ ex ⎤

e z ⎤⎦ I ⎢⎢ e y ⎥⎥ (4.33)

⎢ ⎥

IOP = ⎡⎣ e x ey e z ⎤⎦ ⎢ −I xy I yy −I yz ⎥ ⎢⎢ e y ⎥⎥ = ⎣⎡ e x ey

⎢⎣ −I xz −I yz I zz ⎥⎦ ⎢⎣ e z ⎥⎦ ⎢⎣ e z ⎥⎦

where

⎡ I xx −I xy −I xz ⎤

⎢ ⎥

I = ⎢ −I xy I yy −I yz ⎥ (4.34)

⎢⎣ −I xz −I yz I zz ⎥⎦

The equations of motion of an aircraft in flight consist of a set of dynami-

cal equations obtained by the application of Newton’s laws and Euler’s rigid

body equations and a set of kinematical relationships relating the different

coordinates in the selected reference frames. However, in order to apply the

methods of Newton and Euler, some basic assumptions must be made. These

are the following:

1. The mass of the aircraft and its distribution remain unchanged dur-

ing the duration of the motion.

2. The aircraft is treated as a rigid body during the duration of the

motion.

3. The aircraft has a symmetrical distribution of mass relative to a ver-

tical plane passing through the geometrical axis of the aircraft that

passes through the centre of mass (CM) and is a principal axis of the

aircraft.

4. The rotation of the Earth in space and the curvature of the Earth’s

surface are considered negligible.

The motion of a single rigid body has six independent degrees of freedom,

three of which represent translational motions of a reference point (usu-

ally the CM of the rigid body) along three mutually orthogonal reference

www.ebook777.com

free ebooks ==> www.ebook777.com

112 Flight Dynamics, Simulation, and Control

directions while the other three represent the orientation of the body relative

to the reference directions. To define the equations of motion, we begin with

definition of the moment of momentum vector.

The relative moment of momentum of a system of particles is defined by

h=

∫ ( r × v ) dm = ∫ ρ ( r × v ) dV , (4.35)

V V

where

r is the position vector of a mass particle

v is the particle velocity relative to a moving frame of reference

Assuming that the relative motion is purely rotational, we may define the

moment of momentum vector of a rigid body as

h = r × ( ω × r ) dm = ⎡⎣ω ( r ⋅ r ) − r ( r ⋅ ω) ⎤⎦ dm = Iω,

∫

V

∫

V

(4.36)

where

I is the moment of inertia matrix

ω is the angular velocity vector in a reference frame rigidly attached to the

body at a point fixed in the body given as

T

ω = ⎡⎣ p q r ⎤⎦ (4.37)

or as

ω = pi + qj + rk, (4.38)

body axes.

If the axes along which h is resolved are defined to be coincident with the

physical principal axes of the body, then I is a diagonal matrix. Thus, when h

is not resolved along principal body axes, we get

⎡ I xx −I xy −I xz ⎤ ⎡ p ⎤

⎢ ⎥

h = Iω = ⎢ −I xy I yy −I yz ⎥ ⎢⎢ q ⎥⎥ , (4.39)

⎢⎣ −I xz −I yz I zz ⎥⎦ ⎢⎣ r ⎥⎦

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 113

dinate system (i.e. a coordinate system in which the reference axes rotate

and accelerate linearly) is treated in most textbooks on advanced dynamics.

The rate of change of vector p (such as the translational velocity or angular

momentum) with its components defined in a rotating reference frame is

obtained as

dp dp

= + ω × p, (4.40)

dt inertial dt body

where

the symbol × denotes the vector cross product and the body derivative implies

that the derivatives are taken as if the body axes are inertially fixed

ω is the angular velocity vector of the noninertial reference frame

motion of a rigid body are

d ( mv ) ⎛ dv ⎞

F = ma O = + ω × ( mv ) = m ⎜ + ω × v ⎟ , (4.41)

dt body ⎝ dt ⎠

where

d/dt represents the derivative taken as if the axes are inertially fixed

aO is the acceleration vector of the origin of the reference frame fixed in the

body

v is the velocity vector in the body-fixed frame given by

⎡U ⎤

⎢ ⎥

v = U bi + Vb j + Wb k = Ui + Vj + Wk or as v = ⎢V ⎥ , (4.42)

⎢ ⎥

⎢⎣W ⎥⎦

where we have ignored the subscripts ‘b’ for brevity and F is the three-

component external force vector in the same frame given by

⎡X ⎤

⎢ ⎥

F = Xi + Yj + Zk or as F = ⎢Y ⎥ . (4.43)

⎢ ⎥

⎢⎣ Z ⎥⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

114 Flight Dynamics, Simulation, and Control

⎡i j k⎤

⎢ ⎥

ω× v = ⎢ p q r ⎥ = i ( qW − rV ) − j ( pW − rU ) + k ( pV − qU ) , (4.44)

⎢⎣U V W ⎥⎦

which may also be expressed in matrix notation as

⎡0 −r q ⎤ ⎡ U ⎤ ⎡ qW − rV ⎤

− p ⎥ ⎢⎢ V ⎥⎥ = ⎢ rU − pW ⎥ . (4.45)

⎢ ⎥ ⎢ ⎥

ω× v = ⎢ r 0

⎢⎣ −q p 0 ⎥⎦ ⎢⎣W ⎥⎦ ⎢⎣ pV − qU ⎥⎦

Thus, we have the three scalar equations governing the translational

motion:

( )

m U + qW − rV = X. (4.46a)

( )

m V + rU − pW = Y. (4.46b)

( )

m W + pV − qU = Z. (4.46c)

defined as

H=

∫ ( r × v ) dm, (4.47)

V

a

where va is the absolute particle velocity. It then follows that

H = h+

∫ (r × v

V

a0 ) dm = h + m ( rCM × v a0 ) , (4.48)

where va0 is the velocity of the CM in the same reference frame.

We may also show that

d ⎛ d ⎞ ⎛ d ⎞ ⎛ d ⎞

h=

dt ∫ ⎜⎝ r × dt v ⎟⎠ dm − ∫ ⎜⎝ r × dt v ⎟⎠ dm = ∫ ⎜⎝ r × dt v ⎟⎠ dm − m ( r

V V

0

V

CM × aO ) , (4.49)

where

v0 is the velocity of the CM in the moving reference frame

rCM is the position vector of the CM relative to the origin of the body-fixed

frame

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 115

dh

M = rCM × F = m ( rCM × aO ) + + ω × h = m ( rCM × aO ) + h + ω × h, (4.50)

dt body

where M is the three-component external torque vector in the body-fixed

frame which is obtained by taking moments of all external forces about the

origin of the body-fixed frame and may be written as

⎡L ⎤

⎢ ⎥

M = Li + Mj + Nk or as M = ⎢ M ⎥ . (4.51)

⎢ ⎥

⎢⎣ N ⎥⎦

In Equation 4.50, h is the moment of momentum vector for the rigid body.

The three-component external torque vector is given by evaluating the sum

of the moments of the forces acting on the rigid body. The previous equations

are the famous Euler equations and describe how body-axis components of

the angular velocity vector evolve in time in response to torque components

in body axes.

The translational dynamics of a rigid body can be represented by a set of

relatively simple equations. The rotational dynamics of the rigid body on the

other hand are much more complicated for several reasons: the mass of the

rigid body, m, is a scalar while the moment of inertia is a 3 × 3 matrix. When

the origin of the body axes coincides with the CM, rCM = 0 and the accelera-

tion vector of the origin of the body frame does not influence the rotational

dynamics. If the body axes can also be chosen to coincide with the principal

axes of the body, the moment of inertia is diagonal; in all practical situations,

the matrix I has off-diagonal terms. This is the lesser of the complications in

rigid body attitudinal dynamics.

The main complication is the description of the attitude or orientation of the

body in space. To define the orientation of the body in space, we begin by

defining three mutually perpendicular axes fixed in the body at its centre

of gravity (CG). The body axes are a right-handed triple of orthogonal axes,

Obxbybzb, fixed to the body with

2. The Obyb axis pointing to the starboard, that is, to the right of the

forward-pointing body (lateral axis)

3. Obzb pointing down (normal axis)

www.ebook777.com

free ebooks ==> www.ebook777.com

116 Flight Dynamics, Simulation, and Control

The inertial frame fixed in space is defined by reference axis system. The

reference axes are a right-handed triple of orthogonal axes, Oexeyeze, in the tan-

gent plane of the Earth’s surface with

2. The Oeye axis pointing east

3. The Oeze pointing towards the centre of the Earth

a yawing motion about the z-axis, followed by a pitching motion about

the resulting y-axis and a rolling motion about the final x-axis, the axes

aligned to the frame fixed in the body. Not only are they difficult to depict

in a 2D diagram, but they also cannot be defined uniquely. However, the

yaw–pitch–roll sequence of rotation angles ψ, θ and ϕ from the space-fixed

to the body-fixed reference frame is an accepted standard sequence in air-

craft dynamics, and the rotational angles generated by it are known as the

Euler angles. They give rise to a basic set of coordinates for defining the

attitude of the aircraft.

Thus, the transformation relating the body axes to the space-fixed inertial

reference axes is

⎡ xI ⎤ ⎡ xB ⎤

⎢ ⎥ ⎢ ⎥

⎢ y I ⎥ = TIB × ⎢ yB ⎥ , (4.52)

⎢ ⎥ ⎢ ⎥

⎢⎣ zI ⎥⎦ ⎢⎣ zB ⎥⎦

where

⎢ ⎥⎢ ⎥⎢ ⎥

TIB = ⎢ sin ψ cos ψ 0⎥ ⎢ 0 1 0 ⎥ ⎢0 cos φ − sin φ ⎥

⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ − sin θ 0 cos θ ⎥⎦ ⎢⎣0 sin φ cos φ ⎥⎦

φ cψsθcφ + sψsφ⎤

⎢ ⎥

= ⎢ sψcθ sψsθsφ + cψcφ sψsθcφ − cψsφ ⎥ , (4.53)

⎢⎣ −sθ cθsφ cθcφ ⎥⎦

and the subscript B refers to the body axes while the subscript I refers to

the space-fixed inertial axes. Each of the three-component matrices in TIB is

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 117

defined by

⎡ xB ⎤ ⎡ xI ⎤ ⎡ xI ⎤

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ yB ⎥ = TIB × ⎢ y I ⎥ = TBI × ⎢ y I ⎥ , (4.54)

−1

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢⎣ zB ⎥⎦ ⎢⎣ zI ⎥⎦ ⎢⎣ zI ⎥⎦

where

⎢ ⎥⎢ ⎥⎢ ⎥

TBI = ⎢0 cos φ sin φ ⎥ ⎢ 0 1 0 ⎥ ⎢ − sin ψ cos ψ 0 ⎥ = TIBT

⎢⎣0 − sin φ cos φ ⎥⎦ ⎢⎣ sin θ 0 cos θ ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦

⎡ cψcθ sψcθ − sθ ⎤

⎢ ⎥

= ⎢ cψsθsφ − sψcφ sψsθsφ + cψcφ cθsφ ⎥ . (4.55)

It is important to recognise that the order of rotations implicit in the TBI and

TIB is important as the three matrices in each of these transformations do not

commute. However, when the three Euler angles are assumed to be small

and when the small angle approximations, cθ = cos θ ≈ 1, c ϕ = cos ϕ ≈ 1,

cψ = cos ψ ≈ 1, sθ = sin θ ≈ θ, sϕ = sin ϕ ≈ ϕ, and sψ = sin ψ ≈ ψ, hold, the three

matrix transformations commute and

⎡1 −ψ θ⎤

⎢ ⎥

TIB = ⎢ ψ 1 −φ ⎥ . (4.56)

⎢⎣ −θ φ 1 ⎥⎦

Since any vector, p, in space can be resolved into its components either in the

body-fixed axes or in the space-fixed inertial axes, the transformations may

be used to the components in one frame given the components in the other

and vice versa. The components in the body-fixed frame are identified by

using the additional subscript B while the inertial components are identified

by the subscript, I. Thus, when the vector p is resolved into components in

the body frame, it is denoted as pB, while the same vector is denoted as pI

when it is resolved into the inertial frame.

Thus,

This relationship can be applied to both the translational and rotational

motion of a rigid body.

www.ebook777.com

free ebooks ==> www.ebook777.com

118 Flight Dynamics, Simulation, and Control

In an axis frame fixed in space, that is, an inertial reference frame, the rate

of change of moment of momentum is

dh I d ( TIBhB )

= = MI . (4.58)

dt dt

Since the transformation, TIB, is not constant, this equation may be written as

dhB dTIB

TIB + hB = MI (4.59)

dt dt

or as

dhB dT

+ TBI IB hB = TBI MI = MB . (4.60)

dt dt

⎡ 0 −rB qB ⎤ ⎡ 0 −rB qB ⎤

dTIB ⎢ ⎥ ⎢ ⎥

TBI = TBI TIB × ⎢ rB 0 − pB ⎥ = ⎢ rB 0 − pB ⎥ . (4.61)

dt

⎢⎣ −qB pB 0 ⎥⎦ ⎢⎣ −qB pB 0 ⎥⎦

reference frame and the inertial axes or space-fixed reference frame, it is

important to establish the relationship between the angular velocity of the

body-fixed reference frame, that is, the angular velocity components, pB, qB

and rB, of the body and the rates of change of the roll, pitch and yaw angles,

and ψ.

ϕ, θ and ψ, that is, the attitude rates φ, Recall the inertial axes were

θ

transformed to the body axes by three successive rotations: the yaw angle ψ

about the z-axis, the pitch angle θ about the yaw rotated inertial axis and the

roll angle ϕ about the yaw and pitch rotated inertial axes. The angular veloc-

ity in the body axes is the sum of the yaw rate about the space-fixed z-axis,

the pitch rate about the yaw rotated inertial y-axis and the roll rate about the

body x-axis. Transforming space-fixed z-axis and the yaw rotated inertial

y-axis to the body frame, the body-axis angular velocity components may

be expressed as

−1

⎡ pB ⎤ ⎡ 1 ⎤ ⎡1 0 0 ⎤ ⎧ ⎡0 ⎤ ⎡ cos θ 0 sin θ ⎤ ⎡0 ⎤ ⎫

−1

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎪⎪ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎪⎪

⎢ qB ⎥ = ⎢0 ⎥ φ + ⎢0 cos φ − sin φ ⎥ ⎨⎢1⎥ θ + ⎢ 0 1 0 ⎥ ⎢0 ⎥ ψ ⎬ .

⎢⎣ rB ⎥⎦ ⎢⎣0 ⎥⎦ ⎢⎣0 sin φ cos φ ⎥⎦ ⎪ ⎢0 ⎥ ⎢⎣ − sin θ 0 cos θ ⎥⎦ ⎢⎣ 1 ⎥⎦ ⎪

⎪⎩ ⎣ ⎦ ⎪⎭

(4.62)

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 119

Solving the equations defining pB, qB and rB, for the attitude rates, φ , θ and ψ,

one obtains

⎢ ⎥ ⎢ ⎥ (4.63)

⎢ θ ⎥ = ⎢0 cos φ − sin φ ⎥ ⎢⎢ qB ⎥⎥ .

⎢ψ ⎥ ⎢0 sin φ cos θ cos φ cos θ ⎥⎦ ⎣⎢ rB ⎦⎥

⎣ ⎦ ⎣

It can be expressed as

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ (4.64)

⎢ θ ⎥ = ⎢0 1 0 ⎥ ⎢⎢ qB ⎥⎥ + ⎢0 cos φ − 1 − sin φ ⎥ ⎢ qB ⎥

⎢ψ ⎥ ⎢0 1 ⎥⎦ ⎣⎢ rB ⎦⎥ ⎢⎣0

⎣ ⎦ ⎣ 0 sin φ cos θ ( cos φ cos θ ) − 1⎥⎦ ⎢⎣ rB ⎥⎦

Hence, the Euler equations take the form

⎡ 0 −rB qB ⎤

dhB ⎢ ⎥

+ rB 0 − pB ⎥ hB = MB . (4.65)

dt ⎢

⎢⎣ −qB pB 0 ⎥⎦

given by

⎡ hx B ⎤ ⎡ pB ⎤ ⎡ I xx −I xy −I xz ⎤ ⎡ pB ⎤

⎢ ⎥ ⎢ ⎥

hB = ⎢ hy B ⎥ = I ⎢⎢ qB ⎥⎥ = ⎢ −I xy I yy −I yz ⎥ ⎢⎢ qB ⎥⎥ . (4.66)

⎢ ⎥ ⎢⎣ rB ⎥⎦ ⎢⎣ −I xz −I yz I zz ⎥⎦ ⎢⎣ rB ⎥⎦

⎢⎣ hz B ⎥⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

120 Flight Dynamics, Simulation, and Control

Thus,

⎡ h x ⎤ ⎡ hx ⎤

⎢ B⎥ ⎡ 0 −rB qB ⎤ ⎢ B ⎥ ⎡ L ⎤

⎢ ⎥ ⎢ ⎥

⎢ hy B ⎥ + ⎢ rB 0 − pB ⎥ ⎢⎢ hy B ⎥⎥ = ⎢⎢ M ⎥⎥ , (4.67)

⎢ ⎥ ⎢⎣ −qB pB 0 ⎥⎦ ⎢ ⎥ ⎢⎣ N ⎥⎦

⎢ hz B ⎥ ⎢⎣ hz B ⎥⎦

⎣ ⎦

where

hx B = I xx pB − I xy qB − I xz rB (4.68a)

hy B = −I xy pB +I yy qB − I yz rB (4.68b)

hz B = −I xz pB − I yz qB + I zz rB (4.68c)

y-axis (the pitch axis in the case of an aircraft) is a principal axis of the body

and that as a consequence, the moment of inertia matrix satisfies

⎡ I xx 0 −I xz ⎤

⎢ ⎥

I=⎢ 0 I yy 0 ⎥ . (4.69)

⎢⎣ −I xz 0 I zz ⎥⎦

Substituting for the components of the h B vector,

⎡ I xx −I xy −I xz ⎤ ⎡p B ⎤ ⎡ 0 −rB qB ⎤ ⎡ I xx −I xy −I xz ⎤ ⎡ pB ⎤

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢q ⎥ = M .

⎢ −I xy I yy −I yz ⎥ ⎢⎢q B ⎥⎥+ ⎢ rB 0 − pB ⎥ ⎢ −I xy I yy −I yz ⎥ ⎢ B⎥ B

⎢ −I xz −I yz I zz ⎥⎦ ⎢⎣ rB ⎥⎦ ⎢⎣ −qB pB 0 ⎥⎦ ⎢⎣ −I xz −I yz I zz ⎥⎦ ⎢⎣ rB ⎥⎦

⎣

(4.70)

Solving for p B, q B and rB, we obtain

I zz I

p B =

Δ

( ) ( )

L + I xz pBqB + ( I yy − I zz ) qBrB + xz N − I xz qBrB + ( I xx − I yy ) pBqB , (4.71a)

Δ

1

q B =

I yy

( ( ))

M − ( I xx − I zz ) pBrB − I xz pB2 − rB2 , (4.71b)

I xx I

rB =

Δ

( ) ( )

N − I xz qBrB + ( I xx − I yy ) pBqB + xz L + I xz pBqB + ( I yy − I zz ) qBrB , (4.71c)

Δ

where Δ = I xx I zz − I xz

2

.

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 121

motion of a rigid body in the body frame are

⎛ ⎡ 0 −rB qB ⎤ ⎞

⎜ ⎢ ⎥ ⎟

m ⎜ v + ⎢ rB 0 − pB ⎥ v ⎟ = FB , (4.72)

⎜ ⎢⎣ −qB pB 0 ⎥⎦ ⎟⎠

⎝

that is,

( )

m U + qBW − rBV = XB , (4.73a)

( )

m V + rBU − pBW = YB , (4.73b)

( )

m W + pBV − qBU = ZB . (4.73c)

rotation angles ϕ, θ and ψ to the body-axis angular velocity components,

T

[ pB qB rB ] . It can be shown that

⎢ ⎥ ⎢

⎢ θ ⎥ = ⎢0 cos φ − sin φ ⎥⎥ ⎢⎢ qB ⎥⎥ . (4.74)

⎢ψ ⎥ ⎢0 sin φ cos θ cos φ cos θ ⎥⎦ ⎢⎣ rB ⎥⎦

⎣ ⎦ ⎣

P ≡ [x y z ] . (4.75)

u = U + zq − yr , (4.76a)

v = V + xr − zp (4.76b)

and

w = W + yp − xq. (4.76c)

www.ebook777.com

free ebooks ==> www.ebook777.com

122 Flight Dynamics, Simulation, and Control

for the inertial position of the point P:

⎡ xi ⎤ ⎡u ⎤

d ⎢ ⎥ ⎢ ⎥

⎢ yi ⎥ = TIBv P = TIB ⎢v ⎥ , (4.77)

dt ⎢ ⎥ ⎢ ⎥

⎢⎣ zi ⎥⎦ ⎢⎣ w ⎥⎦

T

where [ u v w ] are the components of the aircraft’s velocity vector in the

body axes at the point P.

The aerodynamic forces and moments on aircraft are produced by the motion

of the vehicle through the air and are obtained, in principle, by integrat-

ing the aerodynamic pressure over the entire surface of the aircraft. They

depend to large extent on the velocity of the air mass relative to the aircraft.

They may be conveniently expressed in an axis system of three mutually

orthogonal axes, one of which is aligned in the direction of the negative of

the velocity vector of the wind relative to the aircraft. Such a wind axis system

may be obtained in much the same way as the space-fixed inertial axes are

obtained from the body axes.

Thus, the transformation relating the body axes to the wind axes is

defined as

⎡ xW ⎤ ⎡ xB ⎤

⎢ ⎥ ⎢ ⎥

⎢ yW ⎥ = TWB ( α , β ) × ⎢ yB ⎥ , (4.78)

⎢ ⎥ ⎢ ⎥

⎢⎣ zW ⎥⎦ ⎢⎣ zB ⎥⎦

with

⎢ ⎥⎢ ⎥

TWB ( α , β ) = ⎢ − sin β cos β 0⎥ ⎢ 0 1 0 ⎥ , (4.79)

⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ − sin α 0 cos α ⎥⎦

where

α is known as the instantaneous angle of attack

β is the instantaneous sideslip angle

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 123

inertial frame with roll angle set to zero; the pitch angle equal to the angle

of attack, α; and the yaw angle equal to the negative of the sideslip angle, β.

The velocity of the wind relative to the aircraft has three components,

u, v and w, in the negative x, negative y and negative z directions. When

the wind is still in an inertial frame, the aircraft’s body velocity has three

components, U b = u, V b = v, W b = w, in the positive x, positive y and positive

z directions. Let

VT = u2 + v 2 + w 2 , (4.80)

u2 + w 2 v u w

cos β = , sin β = , cos α = and sin α= ,

VT VT u +w

2 2

u + w2

2

(4.81)

so

⎛ v ⎞ ⎛w⎞

β = tan −1 ⎜⎜ ⎟ and a = tan −1 ⎜ ⎟ . (4.82)

⎝ u +w ⎠

2 2 ⎟ ⎝u⎠

Hence,

VT =

( )

( uu + vv + ww ) , β = v u + w − v ( uu + ww )

2 2

and α =

wu

− uw

.

VT

(

VT2 u2 + w 2 ) u2 + w 2

(4.83)

⎢v⎥ = T ⎢ ⎥ and ⎢ v ⎥ = T ⎢ ⎥ (4.85)

⎢ ⎥ BW ( α , β ) × 0

⎢ ⎥ ⎢ ⎥ BW ( α , β ) × ⎢ VTβ ⎥ .

⎢⎣ w ⎥⎦ ⎢⎣ 0 ⎥⎦ ⎢⎣ w ⎥⎦ ⎢V α cos β ⎥

⎣ T ⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

124 Flight Dynamics, Simulation, and Control

The aerodynamic forces and moments in the wind axes transform to the

body axes according to the transformation

where

⎡ CD ⎤ ⎡ −bCl ⎤

1 ⎢ ⎥ 1 ⎢ ⎥

FAW = − ρSV 2 ⎢CY ⎥ and M AW = ρSV 2 ⎢ cCm ⎥ . (4.87)

2 ⎢ ⎥ 2 ⎢ ⎥

⎢⎣CL ⎥⎦ ⎢ −bCn ⎥

⎣ ⎦

Hence, the Euler equations and the Newtonian equations of motion governing

the rotational and translational motion of a rigid body take the form

⎡ 0 −rB qB ⎤

dhB ⎢ ⎥

+ rB 0 − pB ⎥ hB = M NB + TBW ( α , β ) M AW (4.88a)

dt ⎢

⎢⎣ −qB pB 0 ⎥⎦

and

⎛ ⎡ 0 −rB qB ⎤ ⎞

⎜ ⎢ ⎥ ⎟

m ⎜ v + ⎢ rB 0 − pB ⎥ v ⎟ = FNB + TBW ( α , β ) FAW + Fg , (4.88b)

⎜ ⎢⎣ −qB pB 0 ⎥⎦ ⎟⎠

⎝

where

Fg are the gravitational forces

FNB and M NB are the other non-aerodynamic forces and moments defined

in the body axes including those due to control, thrust and power and

atmospheric effects

Since the gravitational forces, Fg, are easily defined in inertial axes, they may

be included in the equations of motion quite easily. Moments due to gravity

gradient effects are neglected. Thus, the translational equations of motion

including forces due to the acceleration due to gravity may be expressed as

⎛ ⎡ 0 −rB qB ⎤ ⎞ ⎡0 ⎤

⎜ ⎢ ⎥ ⎟ ⎢ ⎥

m ⎜ v + ⎢ rB 0 − pB ⎥ v ⎟ = FNB + TBW ( α , β ) FAW + mgTBI ⎢0 ⎥ . (4.89)

⎢⎣ −qB ⎢ ⎥

⎜

⎝ pB 0 ⎥⎦ ⎟⎠ ⎢⎣1 ⎥⎦

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 125

and the Stability Axes

To obtain the equations of motion about trimmed or equilibrium flight,

aerodynamicists adopt yet another set of axes, the stability axes. It is a

body-fixed axis system with one of the axes in this mutually orthogonal

triple being aligned to the direction of aircraft’s trimmed velocity vector.

Since the aircraft’s trimmed velocity vector, by hypothesis, is a constant,

the trimmed angle of attack and the trimmed sideslip are also constants.

Thus, the stability axis is equivalent to the trimmed wind axes and is also

fixed in the body.

Thus, the transformation relating the body axes to the stability axes is

defined as

⎡ xS ⎤ ⎡ xB ⎤

⎢ ⎥ ⎢ ⎥

⎢ yS ⎥ = TWB ( α e , βe ) × ⎢ yB ⎥ , (4.90)

⎢ ⎥ ⎢ ⎥

⎢⎣ zS ⎥⎦ ⎢⎣ zB ⎥⎦

with

⎢ ⎥⎢ ⎥

TWB ( α e , βe ) = ⎢ − sin βe cos βe 0⎥ ⎢ 0 1 0 ⎥

⎢ ⎥⎢ ⎥

⎣ 0 0 1 ⎦ ⎣ − sin α e 0 cos α e ⎦

⎢ ⎥

= ⎢ − cos α e sin βe cos βe − sin α e sin βe ⎥ , (4.91)

⎢ ⎥

⎣ − sin α e 0 cos α e ⎦

where

α e is known as the trimmed angle of attack

βe is the trimmed sideslip angle

In particular, when α e and βe are identically equal to zero, the stability axes

are identical to the body axes.

www.ebook777.com

free ebooks ==> www.ebook777.com

126 Flight Dynamics, Simulation, and Control

With the introduction of the stability axes, the Euler equations and the

Newtonian equations of motion governing the rotational and translational

motion of a rigid body in these axes take the form

⎡0 −rB qB ⎤

dhB ⎢ ⎥

+ rB 0 − pB ⎥ hB = M NB + TBW ( α e , βe ) M AS (4.92a)

dt ⎢

⎢⎣ −qB pB 0 ⎥⎦

and

⎛ ⎡ 0 −rB qB ⎤ ⎞ ⎡0 ⎤

⎜ ⎢ ⎥ ⎟ ⎢ ⎥

m ⎜ v + ⎢ rB 0 − pB ⎥ v ⎟ = FNB + TBW ( α e , βe ) FAS + mgTBI ⎢0 ⎥ , (4.92b)

⎢⎣ −qB ⎢ ⎥

⎜

⎝ pB 0 ⎥⎦ ⎟⎠ ⎢⎣1 ⎥⎦

where FAS and M AS are the aerodynamic forces and moments defined in the

stability axes

⎢ ⎥⎢ ⎥

TBW ( α e , βe ) = ⎢ 0 1 0 ⎥ ⎢sin βe cos βe 0⎥

⎢⎣ sin α e 0 cos α e ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦

⎢ ⎥

=⎢ sin βe cos βe 0 ⎥ (4.93)

⎢⎣ sin α e cos βe − sin α e sin βe cos α e ⎥⎦

and

⎡0 ⎤ ⎡ − sin θ ⎤

⎢ ⎥ ⎢ ⎥

mgTBI ⎢0 ⎥ = mg ⎢sin φ cos θ ⎥ . (4.94)

⎢ ⎥ ⎢ ⎥

⎢⎣1 ⎥⎦ ⎢⎣cos φ cos θ ⎥⎦

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 127

dynamics of aircraft flight. The translational equations of motion are

⎡ m ( u + qb w − rbv ) ⎤ ⎡0 ⎤

⎢ ⎥ ⎢ ⎥

⎢ m ( v + r u − p w ) ⎥ = F + T ( α , β ) F + mgT ⎢0 ⎥ , (4.95)

b b NB BW e e AS BI

⎢ ⎥ ⎢ ⎥

⎢ ⎥ ⎢⎣1 ⎥⎦

⎣⎢ m ( w

+ p b v − q b u ) ⎦⎥

where

FAS and M AS are the aerodynamic forces and moments defined in the sta-

bility axes

TBW(α e, βe) is defined by Equation 4.93 and the gravity force vector by

Equation 4.94

tions are

⎡ m ( u + qb w − rbv ) ⎤

⎢ ⎥ ⎡ − siin θ ⎤

⎢ m ( v + r u − p w ) ⎥ − mg ⎢ sin φ cos θ ⎥ = F + T ( α , β ) F (4.96a)

b b ⎢ ⎥ NB BW e e AS

⎢ ⎥

⎢ ⎥ ⎢⎣cos φ cos θ ⎥⎦

⎢⎣ m ( w + pbv − qbu ) ⎥⎦

and

⎡ I xx

s

−I xy

s

−I xz

s

⎤ ⎡ p B ⎤ ⎡ 0 −rB qB ⎤ ⎡ I xx

s

−I xy

s

−I xz

s

⎤ ⎡ pB ⎤

⎢ s s ⎥⎢ ⎢ ⎥ ⎢ s ⎥⎢

⎢ −I xy I yy

s

−I yz ⎥ ⎢ q B ⎥⎥ + ⎢ rB 0 − pB ⎥ ⎢ −I xy

s

I yy

s

−I yz ⎥ ⎢ qB ⎥⎥

⎢ −I xz −I yz I zz ⎦ ⎣⎢ rB ⎦⎥ ⎢⎣ −qB pB 0 ⎥⎦ ⎢⎣ −I xz −I yz I zz ⎦ ⎣⎢ rB ⎦⎥

s s s ⎥ s s s ⎥

⎣

= M NB + TBW ( α e , βe ) M AS . (4.96b)

The matrix,

⎡ I xx

s

−I xy

s

−I xz

s

⎤

⎢ s s ⎥

I = ⎢ −I xy I yy

s

−I yz ⎥ , (4.97)

⎢ −I xz

s

−I yz

s

I zz

s ⎥

⎣ ⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

128 Flight Dynamics, Simulation, and Control

I is the moment of inertia matrix in the stability axes. From symmetry consid-

erations, it is often assumed that, in the case of symmetric trimmed conditions,

⎡ I xx

s

0 −I xz

s

⎤

⎢ ⎥

I=⎢ 0 I yy

s

0 ⎥ . (4.98)

⎢ −I xz

s

0 I zz

s ⎥

⎣ ⎦

If we let

T T

M AS = [ LA M A N A ] , M NB + TBW ( α e , βe ) M AS = [ Le Me N e ] (4.99a)

T T

FAS = [ X A YA ZA ] , FNB + TBW ( α e , βe ) FAS = [ X e Ye Ze ] (4.99b)

and drop the superscript ‘s’ from the components of the inertia matrix, we obtain

⎢ v ⎥ + ⎢ r ⎥⎢ ⎥ ⎢ ⎥ 1 ⎢ ⎥ (4.100)

⎢ ⎥ ⎢ b 0 − pb ⎥ ⎢ v ⎥ − g ⎢ sin φ cos θ ⎥ = ⎢ Ye ⎥

m

⎢⎣ w ⎥⎦ ⎢⎣ −qb pb 0 ⎥⎦ ⎢⎣ w ⎥⎦ ⎢⎣cos φ cos θ ⎥⎦ ⎢⎣ Ze ⎥⎦

and

−1

⎡ p B ⎤ ⎡ I xx 0 −I xz ⎤

⎢ q ⎥ = ⎢ 0 I yy 0 ⎥

⎥

⎢ B⎥ ⎢

⎢⎣ rB ⎥⎦ ⎢⎣ −I xz 0 I zz ⎥⎦

⎡ ⎡ Le ⎤ ⎡ 0 −rB qB ⎤ ⎡ I xx 0 −I xz ⎤ ⎡ pB ⎤ ⎤

⎢⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎥

× ⎢ ⎢ Me ⎥ − ⎢ rB 0 − pB ⎥ ⎢ 0 I yy 0 ⎥ ⎢⎢ qB ⎥⎥ ⎥ (4.101)

⎢ ⎢ N e ⎥ ⎢ −qB pB 0 ⎥⎦ ⎢⎣ −I xz 0 I zz ⎥⎦ ⎢⎣ rB ⎥⎦ ⎥⎦

⎣⎣ ⎦ ⎣

or

I zz I

p B =

Δ

( ) (

Le + I xz pBqB + ( I yy − I zz ) qBrB + xz N e − I xz qBrB + ( I xx − I yy ) pBqB ,

Δ

)

(4.102a)

1

q B =

I yy

( ( ))

Me − ( I xx − I zz ) pBrB − I xz pB2 − rB2 , (4.102b)

I xx I

rB =

Δ

( ) (

N e − I xz qBrB + ( I xx − I yy ) pBqB + xz Le + I xz pBqB + ( I yy − I zz ) qBrB ,

Δ

)

(4.102c)

where Δ = I xx I zz − I xz

2

.

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 129

the rotation angles ϕ, θ and ψ to the body-axis angular velocity components,

T

[ pB qB rB ] . It can be shown that

⎡ φ ⎤ ⎡ 1 sin φ tan θ cos φ tan θ ⎤ ⎡ pB ⎤

⎢ ⎥ ⎢

⎢ θ ⎥ = ⎢0 cos φ − sin φ ⎥⎥ ⎢⎢ qB ⎥⎥ . (4.103)

⎢ψ ⎥ ⎢0 sin φ cos θ cos φ cos θ ⎥⎦ ⎢⎣ rB ⎥⎦

⎣ ⎦ ⎣

P ≡ [x y z ] . (4.104)

The body components of the velocity, vP, at this point are

v = V + xrB − zpB (4.105b)

and

Finally, the aforementioned equations must be complemented by equations

for the inertial position of the point P:

⎡ xi ⎤ ⎡u ⎤

d ⎢ ⎥ ⎢ ⎥

⎢ yi ⎥ = TIBv P = TIB ⎢v ⎥ (4.106)

dt ⎢ ⎥ ⎢ ⎥

⎢⎣ zi ⎥⎦ ⎢⎣ w ⎥⎦

where [u v w]T are the components of the aircraft’s velocity vector in the

body axes at the point P and

TIB = ⎢⎢ sin ψ cos ψ 0 ⎥⎥ ⎢⎢ 0 1 0 ⎥⎥ ⎢⎢0 cos φ −sin φ ⎥⎥

⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ − sin θ 0 cos θ ⎥⎦ ⎢⎣0 sin φ cos φ ⎥⎦

⎡cψcθ cψsθsφ − sψcφ

φ cψsθcφ + sψsφ⎤

⎢

= ⎢ sψcθ sψsθsφ + cψcφ sψsθcφ − cψsφ ⎥⎥ , (4.107)

⎢⎣ −sθ cθsφ cθcφ ⎥⎦

and the subscript B refers to the body axes while the subscript I refers to the

space-fixed inertial axes.

www.ebook777.com

free ebooks ==> www.ebook777.com

130 Flight Dynamics, Simulation, and Control

The three body-centred velocity components with the origin at the centre of

mass are given in terms of V T, α and β as

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ v ⎥ = TBW ( α , β ) × ⎢ 0 ⎥ = ⎢ sin β ⎥ VT , (4.108)

⎢⎣ w ⎥⎦ ⎢⎣ 0 ⎥⎦ ⎢⎣ sin α cos β ⎥⎦

⎡ u ⎤ ⎡ VT ⎤ ⎡1 0 0

⎤ ⎡VT ⎤

⎢⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

0 ⎥ ⎢ β ⎥ , (4.109)

⎢ v ⎥ = TBW ( α , β ) × ⎢ VTβ ⎥ = TBW ( α , β ) ⎢0 VT

⎢⎣ w ⎥⎦ ⎢ ⎥

VT α cos β ⎥ ⎢⎣0 0 VT cos β ⎥⎦ ⎢⎢ α ⎥⎥

⎣⎢ ⎦ ⎣ ⎦

⎢ ⎥⎢ ⎥

TBW ( α , β ) = ⎢ 0 1 0 ⎥ ⎢ sin β cos β 0⎥

⎢⎣ sin α 0 cos α ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦

⎡cαcβ −cαsβ − sα ⎤

⎢ ⎥

= ⎢ sβ cβ 0 ⎥. (4.110)

⎢⎣ sαcβ − sα sβ cα ⎥⎦

Thus,

⎡1 0 0 ⎤ ⎡VT ⎤ ⎡ u ⎤

⎢ ⎥ ⎥

⎢ ⎢ ⎥

⎢0 VT 0 ⎥ ⎢ β ⎥ = TWB ( α , β ) ⎢ v ⎥ , (4.111)

⎢⎣0 0 ⎢ ⎥

VT cos β ⎥⎦ ⎢ α ⎥ ⎢⎣ w ⎥⎦

⎣ ⎦

where

⎢ ⎥⎢ ⎥

TWB ( α , β ) = ⎢ − sin β cos β 0⎥ ⎢ 0 1 0 ⎥

⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ − sin α 0 cos α ⎥⎦

⎡ cαcβ sβ sαcβ ⎤

⎢ ⎥

= ⎢ −cαsβ cβ −sαsβ ⎥ . (4.112)

⎢⎣ −sα 0 cα ⎥⎦

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 131

d ⎢ ⎥ d ⎢ ⎥ d ⎢ ⎥ ⎢ ⎥⎢ ⎥ F

v = v + ⎢ v ⎥ = − ⎢ −w 0 u ⎥ ⎢ qB ⎥ + B . (4.113)

dt ⎢ ⎥ dt ⎢ ⎥ dt m

⎢⎣ w ⎥⎦ ⎢⎣ w ⎥⎦ ⎢⎣ w ⎥⎦ ⎢⎣ v −u 0 ⎥⎦ ⎢⎣ rB ⎥⎦

p,q,r F

Furthermore,

⎡u⎤

VT =

1

[

d ⎢ ⎥ v u + w − v ( uu + ww )

u v w] ⎢ v ⎥ , β =

2 2

( ) and α =

wu

− uw

.

VT dt

⎢⎣ w ⎥⎦ VT2 u2 + w 2 ( ) u + w2

2

(4.114)

Thus,

⎡ VT ⎤ ⎡ u v w ⎤ ⎡u⎤

⎢ 2 ⎥ ⎢ ⎥ d ⎢ ⎥

VT ⎢ VT cos β × β ⎥ = ⎢ −uv (u 2

+ w2 ) −wv ⎥ ⎢ v ⎥ , (4.115)

⎢V cos 2 β × α ⎥ ⎢ ⎥ dt

⎣⎢

T

⎦⎥ ⎣ −w 0 u ⎦ ⎢⎣ w ⎥⎦

and

⎡ u v w ⎤⎡ 0 w −v ⎤ ⎡⎢ 0 0 0 ⎥⎤

⎢ ⎥⎢ ⎥

⎢ −uv (u 2

+ w2 ) −wv ⎥ ⎢ −w 0 u ⎥ = ⎢ −wVT2

⎢

0 uVT2 ⎥ ,

⎥

u ⎦ ⎢⎣ v −u 0 ⎥⎦ ⎢ uv

⎢ ⎥

⎣ −w 0

⎣ (

− u2 + w 2 ) vw ⎥

⎦

(4.116)

⎡ 0 0 0 ⎤⎥ ⎡ 1 0 0 ⎤

⎢ ⎢ ⎥

⎢ −wVT2 0 uVT2 ⎥ = ⎢0 V cos β

T

3

0 ⎥

⎢ ⎥ ⎢

⎢⎣ uv (

− u2 + w 2 ) vw ⎥ ⎣0 0 VT2 cos β ⎥⎦

⎦

⎡ 0 0 0 ⎤

⎢ ⎥

× ⎢ − sin α 0 cos α ⎥ . (4.117)

⎢⎣cos α sin β − cos β sin α sin β ⎥⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

132 Flight Dynamics, Simulation, and Control

Thus,

0 ⎤ ⎡VT ⎤

−1

⎡1 0 ⎡1 0 0 ⎤

⎢0 ⎥ ⎢ ⎥ 1 ⎢

VT 0 ⎥⎢ β ⎥ = 0 VT cos β 0 ⎥⎥

⎢

⎢ ⎥ VT ⎢

⎢⎣0 0 VT cos β ⎥⎦ ⎢ α ⎥ ⎢⎣0 0 cosβ ⎥⎦

⎣ ⎦

⎡ u v w ⎤ ⎡ u ⎤ ⎡ u ⎤

⎢ ⎥⎢ ⎥ ⎢ ⎥

× ⎢ −uv u + w2

2

−wv ⎥ ⎢ v ⎥ + TWB ( α , β ) ⎢ v ⎥ .

⎢⎣ −w 0 u ⎥⎦ ⎣⎢ w ⎥⎦ ⎢⎣ w ⎥⎦

p,q,r F

(4.118)

Using both formulations for transforming the body velocity component deriv-

atives and aerodynamic forces to wind axis component rates, it follows that

⎡ VT ⎤ ⎡ 0 0 0 ⎤ ⎡ pB ⎤ ⎡XA ⎤

⎢

⎥ ⎢ ⎥⎢ ⎥ 1 ⎢ ⎥

⎢ VTβ ⎥ = −VT ⎢ −sα 0 cα ⎥ ⎢ qB ⎥ + ⎢ YA ⎥

⎢V α cos β ⎥ m

T ⎢⎣cαsβ −cβ sαsβ ⎥⎦ ⎢⎣ rB ⎥⎦ ⎢⎣ ZA ⎥⎦

⎣⎢ ⎦⎥

1⎢ ⎥⎪ ⎢ ⎥ ⎢ ⎥⎪

+ ⎢ −cαsβ cβ −sαsβ ⎥ ⎨T ⎢ 0 ⎥ + mg ⎢ sin φ coss θ ⎥ ⎬ .

m

⎢⎣ −sα 0 cα ⎥⎦ ⎪⎩ ⎢⎣ − sin σ ⎥⎦ ⎢⎣cos φ cos θ ⎥⎦ ⎪

⎭

(4.119)

coordinates are

I zz I

p B =

Δ

( ) (

Le + I xz pBqB + ( I yy − I zz ) qBrB + xz N e − I xz qBrB + ( I xx − I yy ) pBqB ,

Δ

)

(4.120a)

1

q B = (

I yy

( ))

Me − ( I xx − I zz ) pBrB − I xz pB2 − rB2 , (4.120b)

I xx I

rB =

Δ

( ) (

N e − I xz qBrB + ( I xx − I yy ) pBqB + xz Le + I xz pBqB + ( I yy − I zz ) qBrB ,

Δ

)

(4.120c)

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 133

where Δ = I xx I zz − I xz

2

and the moments of inertia evaluated in the body axes

are assumed to be constant. In the previous text, the external moments are

defined as

T T

⎡L Me N e ⎤⎦ = M NB + TBW ( α , β ) M AW , M AW = [ LA M A N A ] . (4.121)

⎣ e

The relationships between the rotation angle rates φ , θ and ψ to the body-axis

angular velocity components [pB qB rB]T are

⎢ ⎥ ⎢

⎢ θ ⎥ = ⎢0 cos φ − sin φ ⎥⎥ ⎢⎢ qB ⎥⎥ , (4.122)

⎢ψ ⎥ ⎢0 sin φ cos θ cos φ cos θ ⎥⎦ ⎢⎣ rB ⎥⎦

⎣ ⎦ ⎣

ψ =

( sin φqB + cos φrB ) . (4.123c)

cos θ

Observe that the last equation for ψ is decoupled from the first two and can

be integrated once ϕ and θ are known explicitly.

Finally, the previous equations must be complemented by equations for

the inertial position of the point P, which take the form

d ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

yi = TIBv P = TIB ⎢ v ⎥ = TIBTBW ( α , β ) × ⎢ 0 ⎥ = TIB ⎢ sin β ⎥ VT

dt ⎢ ⎥

⎢⎣ zi ⎥⎦ ⎢⎣ w ⎥⎦ ⎢⎣ 0 ⎥⎦ ⎢⎣ sin α cos β ⎥⎦

i.e.

d ⎢ ⎥ ⎢ ⎥⎢ ⎥

yi = sψcθ sψsθsφ + cψcφ sψsθcφ − cψsφ ⎥ ⎢ sin β ⎥ VT .

dt ⎢ ⎥ ⎢

⎢⎣ zi ⎥⎦ ⎢⎣ −sθ cθsφ cθcφ ⎥⎦ ⎢⎣ sin α coss β ⎥⎦

(4.124)

www.ebook777.com

free ebooks ==> www.ebook777.com

134 Flight Dynamics, Simulation, and Control

Thus, the complete set constitutes a set of eight coupled equations and the

four remaining equations that are decoupled from these.

The aircraft motion depends on the aerodynamic and engine thrust forces

and moments acting on the vehicle. The aerodynamic forces consist of the

lift force, the drag force and the side force. The aerodynamic moments

are described by the nose-up pitching moment, the rolling moment and

the yawing moment. The aerodynamic forces and moments depend on the

wind velocity magnitude; the wind axis direction angles, α and β; the body

T

angular rates [ pB qB rB ] and control surface deflections. The generalised

aerodynamic forces and moments in the wind axes may be expressed in

terms of the non-dimensional lift, drag and side force coefficients and the

non-dimensional rolling, nose-up pitching and yawing moment coefficients.

Thus, the engine thrust force and moment and the wind axis aerodynamic

forces and moments are defined, respectively, as

⎡T ⎤ ⎡ T0 ⎤ ⎡ TδT ⎤

⎢ ⎥=⎢ ⎥+⎢ ⎥ δT , (4.125a)

⎢⎣ MT ⎥⎦ ⎣ MT 0 ⎦ ⎣ MδT ⎦

⎡XA ⎤ ⎡ CD ⎤ ⎡ LA ⎤ ⎡ −bCl ⎤

⎢Y ⎥ ⎢C ⎥ ⎢M ⎥ ⎢ ⎥

⎢ A ⎥ = − 1 ρSVT2 ⎢ Y ⎥ and ⎢ A ⎥ = 1 ρSVT2 ⎢ cCm ⎥ . (4.125b)

⎢ ZA ⎥ 2 ⎢ CL ⎥ ⎢ NA ⎥ 2 ⎢ −bCn ⎥

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

T

⎣ ⎦ C

⎣ T⎦ M

⎣ T⎦ ⎣ cCmT ⎦

The aerodynamic coefficients are further expanded and expressed as

Lift coefficient

⎛ u ⎞ ⎛ c ⎞

CL = CL0 + CLα α + CLu ⎜ ( )

⎟ + CLη η + CLq q + CLα α ⎜ ⎟ (4.126a)

⎝ VT∞ ⎠ ⎝ 2VT∞ ⎠

Drag coefficient

CL2 ⎛ u ⎞

CD = CD0 + , CD0 = CD00 + CD0 α α + CD0 u ⎜ ⎟ + CDη0 η (4.126b)

πeA ⎝ VT∞ ⎠

Side force coefficient

⎛ b ⎞

( )

CY = CY0 + CYβ β + CYζ ζ + CYp p + CYr r ⎜ ⎟ (4.126c)

⎝ 2VT∞ ⎠

Rolling moment coefficient

⎛ b ⎞

( )

Cl = Cl0 + Clβ β + Clξξ + Clζ ζ + Clp p + Clr r ⎜ ⎟ (4.126d)

⎝ 2VT∞ ⎠

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 135

⎛ u ⎞ ⎛ c ⎞

Cm = Cm0 + Cmα α + Cmu ⎜ (

⎟ + Cmη η + Cmq q + Cmα α ⎜ ) ⎟ (4.126e)

⎝ VT∞ ⎠ ⎝ 2VT∞ ⎠

⎛ b ⎞

(

Cn = Cn0 + Cnβ β + Cnξξ + Cnζ ζ + Cnp p + Cnr r ⎜ )

⎟ (4.126f)

⎝ 2VT∞ ⎠

Moreover, the coefficients CLx, CDx , CYx , Clx , Cmx and Cnx may be either constants

in the case of a linear model or non-linear functions of α, β, p, q, r and the

control surface deflections. The basic aircraft aerodynamic coefficients are

made up of contributions from the fuselage, wing, horizontal tail plane and

the vertical fin.

The short period approximation is obtained by assuming that VT ⊕0 and that

ϕ ≈ 0 and θ ≈ 0. Hence, the five short period equations of motion are

⎡ pB ⎤

⎡ β ⎤ ⎡ − sin α 0 cos α ⎤ ⎢ ⎥

⎢ ⎥ = −⎢ qB

⎢⎣α cos β ⎥⎦ ⎣cos α sin β − cos β sin α sin β ⎥⎦ ⎢ ⎥

⎢⎣ rB ⎥⎦

⎧ ⎡ cos σ ⎤ ⎡0 ⎤ ⎫

1 ⎡ Ya ⎤ 1 ⎡ −cαsβ cβ −sαsβ ⎤ ⎪ ⎢ ⎥ ⎢ ⎥⎪

+ ⎢Z ⎥ + ⎢ ⎨T 0 ⎥ + mg ⎢0 ⎥ ⎬

mVT ⎣ a ⎦ mVT ⎣ −sα 0 cα ⎥⎦ ⎪ ⎢

⎢⎣ 1 ⎥⎦ ⎪

⎩ ⎣ − sin σ ⎦

⎢ ⎥

⎭

(4.127)

with

I zz I

p B =

Δ

( ) (

La + I xz pBqB + ( I yy − I zz ) qBrB + xz N a − I xz qBrB + ( I xx − I yy ) pBqB ,

Δ

)

(4.128a)

1

q B = (

I yy

( ))

Ma + MT − ( I xx − I zz ) pBrB − I xz pB2 − rB2 , (4.128b)

I xx I

rB =

Δ

( ) (

N a − I xz qBrB + ( I xx − I yy ) pBqB + xz La + I xz pBqB + ( I yy − I zz ) qBrB

Δ

)

(4.128c)

www.ebook777.com

free ebooks ==> www.ebook777.com

136 Flight Dynamics, Simulation, and Control

and

d ⎢ ⎥ ⎢

yi = sψcθ sψsθsφ + cψcφ sψsθcφ − cψsφ ⎥⎥ ⎢⎢ sin β ⎥⎥ VT . (4.128d)

dt ⎢ ⎥ ⎢

⎢⎣ zi ⎥⎦ ⎢⎣ −sθ cθsφ cθcφ ⎥⎦ ⎢⎣ sin α cos β ⎥⎦

The two longitudinal short period equations are obtained by further assum-

ing that β ≈ 0 and considering only the longitudinal states. Thus,

⎛ ρSVT ⎞

α = qB − ⎜

⎝ 2m ⎠

( )

⎟ ( CT sin ( α + σ ) + CL − CW cos α ) (4.129)

and

q B =

( Ma + MT ) , (4.130)

I yy

where

2mg

CW = . (4.131)

ρSVT2

The longitudinal downrange and the vertical inertial position are given by

dxi dzi dh

= cos ( θ − α ) VT , =− = − sin ( θ − α ) VT . (4.132)

dt dt dt

The three lateral short period equations are (roll subsidence and Dutch roll

motion)

ρSVT

β = −rB − ( CY + CT cos σ sin β ) , (4.133a)

2m

p B =

( I zz La + I xz N a ) (4.133b)

Δ

and

rB =

( I xx N a + I xz La ) . (4.133c)

Δ

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 137

⎡cos β ⎤

d ⎡ xi ⎤ ⎡cψ −sψcφ sψ sφ ⎤ ⎢ ⎥

⎢ ⎥=⎢ sin β ⎥ VT . (4.134)

dt ⎣ yi ⎦ ⎣ sψ cψcφ −cψsφ ⎥⎦ ⎢

⎢⎣ 0 ⎦⎥

When the bank angle ϕ is almost zero,

d ⎡ xi ⎤ ⎡cos ( ψ + β ) ⎤

⎢ ⎥=⎢ ⎥ VT . (4.135)

dt ⎣ yi ⎦ ⎢⎣ sin ( ψ + β ) ⎥⎦

For purposes of linear stability analysis or flight control system design, the

aircraft dynamic models are frequently linearised about some operating

condition or flight regime, in which it is assumed that the aircraft velocity and

altitude are constant. The control surfaces and engine thrust are trimmed, to

these conditions, and the control system is designed to maintain them.

Next, we consider the process of trimming. It can be recognised that the

aircraft may be trimmed either in the space-fixed inertial frame or in a body-

fixed frame. Assuming only the case of the former, the conditions of trimmed

flight, assuming that there are no steady rotations (e.g. steady level flight), are

M NBe + TBW ( α e , βe ) M ASe = 0 (4.136a)

and

⎡0 ⎤

⎢ ⎥

FNBe + TBW ( α e , βe ) FASe + mgTBIe ⎢0 ⎥ = 0, (4.136b)

⎢ ⎥

⎢⎣1 ⎥⎦

where the subscript ‘e’ refers to equilibrium conditions corresponding to the

trimmed state and is used to differentiate the trimmed quantities from the

corresponding general quantities.

Subtracting the previous steady-state trim equations, the Euler equations

and the Newtonian equations of motion governing the rotational and trans-

lational motion take the form

⎡ 0 −rB qB ⎤

dhB ⎢ ⎥

+ rB 0 − pB ⎥ hB = ( M NB − M NBe ) + TBW ( α e , βe ) ( M AS − M ASe )

dt ⎢

⎢⎣ −qB pB 0 ⎥⎦

(4.137a)

www.ebook777.com

free ebooks ==> www.ebook777.com

138 Flight Dynamics, Simulation, and Control

and

⎛ ⎡0 −rB qB ⎤ ⎞

⎜ ⎢ ⎥ ⎟

m ⎜ v + ⎢ rB 0 − pB ⎥ v ⎟

⎜ ⎢⎣ −qB pB 0 ⎥⎦ ⎟⎠

⎝

⎡0 ⎤

⎢ ⎥

= ( FNB − FNBe ) + TBW ( α e , βe ) ( FAS − FASe ) + mg ( TBI − TBIe ) ⎢0 ⎥ , (4.137b)

⎢ ⎥

⎢⎣1 ⎥⎦

where

⎡0 ⎤ ⎡ − ( sin θ − sin θe ) ⎤

⎢ ⎥ ⎢ ⎥

mg ( TBI − TBIe ) ⎢0 ⎥ = mg ⎢sin φ cos θ − sin φe cos θe ⎥ . (4.137c)

⎢ ⎥ ⎢ ⎥

⎢⎣1 ⎥⎦ ⎢cos φ cos θ − cos φ cos θ ⎥

e e

⎣ ⎦

When resolved in the body axes, the moment of momentum vector is given by

⎡ hx ⎤

⎢ B⎥

hB = ⎢⎢ hy B ⎥⎥ , (4.138)

⎢ ⎥

⎢⎣ hz B ⎥⎦

where

hx B =I xx pB − I xy qB − I xz rB (4.139a)

hy B = −I xy pB +I yy qB − I yz rB (4.139b)

hz B = −I xz pB − I yz qB +I zz rB (4.139c)

When these components are substituted into the governing equations, the

translational equations of motion are then unchanged and are

⎡ m ( u + qb w − rbv ) ⎤ ⎡0 ⎤

⎢ ⎥ ⎢ ⎥

⎢ m ( v + r u − p w ) ⎥ = F + T ( α , β ) F + mgT ⎢0 ⎥ , (4.140)

b b NB BW e e AS BI

⎢ ⎥ ⎢ ⎥

⎢ ⎥ ⎢⎣1 ⎥⎦

⎣⎢ m ( w

+ p b v − q b u ) ⎦⎥

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 139

⎤ ⎡ pB ⎤ ⎡ 0 ⎤ ⎡ pB ⎤

⎡ I xx

s

−I xy

s

−I xz

s

⎢ ⎥ ⎢ −rB qB ⎤ ⎡ I xx

s

−I xy

s

−I xz

s

⎢ ⎥

⎢ s s ⎥ ⎥⎢ s s ⎥

⎢ −I xy I yy

s

−I yz ⎥ ⎢ q B ⎥ + ⎢ rB 0 − pB ⎥ ⎢ −I xy I yy

s

−I yz ⎥ ⎢ qB ⎥

s ⎥⎢ ⎥ ⎢ s ⎥⎢ ⎥

⎢ −I xz

⎣

s

−I yz

s

I zz ⎦ ⎢⎣ rB ⎥⎦ ⎣ −qB pB 0 ⎥⎦ ⎢⎣ −I xz

s

−I yz

s

I zz ⎦ ⎢⎣ rB ⎥⎦

= M NB + TBW ( α e , βe ) M AS . (4.141)

of motion governing the translational motion may also be expressed in terms

of the aerodynamic forces and moments in the stability axes as

⎡ m ( u + qb w − rbv ) ⎤ ⎡0 ⎤

⎢ ⎥ ⎢ ⎥

⎢ m ( v + r u − p w ) ⎥ = ( F − F ) + T ( α , β ) ( F − F ) + mg ( T − T ) ⎢0 ⎥ .

b b NB NBe BW e e AS ASe BI BIe

⎢ ⎥ ⎢ ⎥

⎢ ⎥ ⎢⎣1 ⎥⎦

⎢ m ( w + pbv − qbu ) ⎥⎦

⎣

(4.142)

⎡ I xx

s

−I xy

s

−I xz

s

⎤ ⎡ p B ⎤ ⎡ 0 −rB qB ⎤ ⎡ I xx

s

−I xy

s

−I xs z ⎤ ⎡ pB ⎤

⎢ s s ⎥⎢ ⎢ ⎥⎢ s s ⎥⎢

⎢ −I xy I yy

s

−I yz ⎥ ⎢ q B ⎥⎥ + ⎢ rB 0 − pB ⎥ ⎢ −I xy I yy

s

−I yz ⎥

⎥ ⎢ qB ⎥

⎢ −I xz −I yz I zz ⎦ ⎣ rB ⎦ ⎣ −qB pB 0 ⎥⎦ ⎢⎣ −I xz −I yz I zz ⎦ ⎢⎣ rB ⎥⎦

s s s ⎥ s s s ⎥

⎢ ⎥ ⎢

⎣

= ( M NB − M NBe ) + TBW ( α e , βe ) ( M AS − M ASe ) . (4.143)

These equations may be linearised to form the linear equations for the

perturbation states which are presented in the next section.

The non-linear equations of perturbed motion about an equilibrium state are

given by

⎢ v ⎥ + ⎢ r ⎥⎢ ⎥ ⎢ ⎥ 1⎢ ⎥

⎢ ⎥ ⎢ b 0 − pb ⎥ ⎢ v ⎥ − g ⎢ sin φ cos θ − sin φe cos θe ⎥ = ⎢ Ye − Ye 0 ⎥

m

⎢ w ⎥ ⎢ −qb pb 0 ⎥⎦ ⎢⎣ w ⎥⎦ ⎢cos φ cos θ − cos φe cos θe ⎥ ⎢⎣ Ze − Ze 0 ⎥⎦

⎣ ⎦ ⎣ ⎣ ⎦

(4.144a)

www.ebook777.com

free ebooks ==> www.ebook777.com

140 Flight Dynamics, Simulation, and Control

and

−1

⎡ p B ⎤ ⎡ I xx 0 −I xz ⎤

⎢ q ⎥ = ⎢ 0 I yy 0 ⎥

⎥

⎢ B⎥ ⎢

⎢⎣ rB ⎥⎦ ⎢⎣ −I xz 0 I zz ⎥⎦

⎡ ⎡ Le − Le 0 ⎤ ⎡ 0 −rB qB ⎤ ⎡ I xx 0 −I xz ⎤ ⎡ pB ⎤ ⎤

⎢⎢ ⎥

0 ⎥ ⎢⎢ qB ⎥⎥ ⎥ . (4.144b)

⎥ ⎢ ⎥⎢ ⎥

× ⎢ ⎢ Me − Me 0 ⎥ − ⎢ rB 0 − pB ⎥ ⎢ 0 I yy

⎢ ⎢ N e − N e 0 ⎥ ⎢ −qB pB 0 ⎥⎦ ⎢⎣ −I xz 0 I zz ⎥⎦ ⎢⎣ rB ⎥⎦ ⎥⎦

⎣⎣ ⎦ ⎣

Linearising the aerodynamic loads,

⎡ X e − X e 0 ⎤ ⎡ X Au Δu + X Aα Δα + X Aq q + X Aηη + X AτΔτ ⎤

⎢ ⎥ ⎢ ⎥

⎢ Ye − Ye 0 ⎥ = ⎢ YAβΔβ + YAp p + YAr r + YAξξ + YAζ ζ ⎥ , (4.145a)

⎢⎣ Ze − Ze 0 ⎥⎦ ⎢⎣ ZAu Δu + ZAα Δα + ZAq q + ZAηη + ZAτΔτ ⎦⎥

and

⎢ ⎥ ⎢ ⎥

M

⎢ e − M e0 ⎥ = M

⎢ Au u

Δu + M Aα Δα + M Aq q + M Aα Δα + M Aηη + MTτ Δτ ⎥ . (4.145b)

⎢ N e − N e 0 ⎥⎦ ⎢⎣ N AβΔβ + N Ap p + N Ar r + N Aξξ + N Aζ ζ ⎥⎦

⎣

steady forward flight in the stability axes are given by the following:

Δθ = q. (4.146d)

h = U 0 ( Δθ − Δα ) . (4.146e)

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 141

Δφ = p. (4.146i)

Δψ = r. (4.146j)

In these equations, the first five of which represent the longitudinal equa-

tions or equations of symmetric motions and the last five represent the lateral

equations or equations of asymmetric motion, the variables are defined as

follows:

U0 is the steady forward speed.

Δu is the forward speed perturbation.

Δα is the perturbation in the angle of attack.

Δβ is the perturbation in the sideslip angle.

h is the aircraft altitude.

p is the roll rate in the body-fixed axes.

Δφ is the perturbation roll angle.

q is the pitch rate in the body-fixed axes.

Δθ is the perturbation pitch angle.

r is the yaw rate in the body-fixed axes.

Δψ is the perturbation yaw angle.

ξ is the aileron.

η is the elevator.

ζ is the rudder.

Δτ is the perturbation in the engine thrust or throttle position.

Chapter Highlights

• Axis systems and motions

• Axis systems

Space-fixed reference axes: Fixed in space, these are a typical set of

inertial axes.

Body-fixed reference axes: Fixed in the aircraft, normally at the

CG of the aircraft.

Wind axes: Fixed with reference to wind direction relative to aircraft.

www.ebook777.com

free ebooks ==> www.ebook777.com

142 Flight Dynamics, Simulation, and Control

the steady wind; it is also a body-fixed system when aircraft is fly-

ing in equilibrium (trim).

• Body-fixed axes

Origin at airplane CG, x-axis forward along fuselage, y-axis to

pilot’s right and z-axis downward (right-hand system)

Axis x Y z

Velocity u V w

Aero force X (axial) Y (side) Z (normal) Cy = Y/qS

Aero moment L M N

Pos. direction Right wing down Nose up Nose to right

Moment coefficient CL = L/qSb CM = M/qSc CN = N/qSb

(S = gross wing area, B = span, c = MAC, q = dynamic pressure)

Angles Bank, ϕ α = tan−1w/u β = sin−1v/V

Exercises

4.1 Show that the relationship between the components of the body angu-

lar velocity vector and the rates of change of the orientation angles cor-

responding to a yaw–pitch–roll rotational sequence from a space-fixed

reference frame to a body-fixed reference frame is

−1

⎡1 0 0 ⎤ −1

⎡ pB ⎤ ⎡ 1 ⎤ ⎢ ⎥

⎧ ⎡0 ⎤ ⎡ cos θ 0 sin θ ⎤ ⎡0 ⎤ ⎫

⎢ ⎥ ⎢ ⎥ ⎢ ⎪⎪ ⎢ ⎥ ⎢ ⎢ ⎥ ⎪⎪

⎢ qB ⎥ = ⎢0 ⎥ φ + ⎢0 coos φ − sin φ ⎥ ⎨⎢1⎥ θ + ⎢ 0 1 0 ⎥⎥ ⎢0 ⎥ ψ ⎬ ,

⎥ ⎪ ⎢0 ⎥

⎢⎣ rB ⎥⎦ ⎢⎣0 ⎥⎦ ⎢⎣0 sin φ cos φ ⎥⎦ ⎢⎣ − sin θ 0 cos θ ⎥⎦ ⎢⎣ 1 ⎥⎦ ⎪

⎩⎪ ⎣ ⎦ ⎭⎪

which is

⎡ pB ⎤ ⎡ 1 ⎤ ⎡ 0 ⎤ ⎡ − siin θ ⎤

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

q

⎢ B⎥ ⎢ ⎥= 0 φ + ⎢ cos φ ⎥ θ + ⎢ sin φ cos θ ⎥ ψ.

⎢⎣ rB ⎥⎦ ⎢⎣0 ⎥⎦ ⎢⎣ − sin φ ⎥⎦ ⎢⎣cos φ cos θ ⎥⎦

Hence, obtain the inverse relation, relating the rates of change of the ori-

entation angles to the components of the body angular velocity vector.

free ebooks ==> www.ebook777.com

Aircraft Non-Linear Dynamics 143

space form.

4.3 Express the aircraft’s linear lateral equations of motion in state-space form.

4.4 Linearise the aircraft’s two longitudinal short period equations in α and

qB by introducing linear approximations to the aerodynamic and thrust

coefficients, and express these equations in state-space form.

4.5 Linearise the aircraft’s three lateral short period equations for roll sub-

sidence and Dutch roll motions in β, pB and rB by introducing linear

approximations to the aerodynamic and thrust coefficients, and express

these equations in state-space form.

References

1. Etkin, B. and Reid, L. D., Dynamics of Flight: Stability and Control, 3rd ed., John

Wiley & Sons, New York, 1998.

2. Seckel, E., Stability and Control of Airplanes and Helicopters, Academic Press, New

York, 1964.

3. McRuer, D., Ashkenas, I., and Graham, D., Aircraft Dynamics and Automatic

Control, Princeton University Press, Princeton, NJ, 1973.

4. Smetana, F. O., Computer Assisted Analysis of Aircraft Performance, Stability and

Control, McGraw-Hill College, New York, 1984.

5. Nelson, R. C., Flight Stability and Automatic Control, 2nd ed., McGraw-Hill, New

York, 1998.

6. Cook, M. V., Flight Dynamics Principles, Arnold, London, U.K., 1997.

7. Schmidt, L. V., Introduction to Aircraft Flight Dynamics, AIAA Education Series,

Reston, VA, 1998.

www.ebook777.com

free ebooks ==> www.ebook777.com

free ebooks ==> www.ebook777.com

5

Small Perturbations and the Linearised,

Decoupled Equations of Motion

5.1 Introduction

Although we have already derived the linear dynamic equations of motion

about steady equilibrium flight, as discussed in Chapter 4, we shall examine

the derivation process in some detail in this chapter. This is essential par-

ticularly as it is important to understand the limitations of such approxima-

tions and the context in which they may be applied.

The small perturbation approximations, compatible with conditions of trim,

are now introduced. The aircraft’s velocities and angular velocities are per-

turbed about a steady operating condition. Thus, the perturbations satisfy

⎡u ⎤ ⎡ U e + Δu ⎤ ⎡ pB ⎤ ⎡ Δp ⎤ ⎡φ ⎤ ⎡ φe + Δφ ⎤

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢v ⎥ = ⎢ Ve + Δv ⎥ , ⎢ qB ⎥ = ⎢ Δq ⎥ , ⎢θ ⎥ = ⎢ θe + Δθ ⎥ (5.1)

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢⎣ w ⎥⎦ ⎢⎣We + Δw ⎥⎦ ⎢⎣ rB ⎦⎥ ⎢⎣ Δr ⎥⎦ ⎢⎣ψ ⎥⎦ ⎢⎣ψ e + Δψ ⎥⎦

where Ue, Ve and We represent the steady trimmed velocities of the aircraft

and the steady-state body axis angular velocities are each assumed to be

equal to 0. They are related to the trimmed angle of attack and the trimmed

sideslip angle by the relations

⎛ We ⎞ ⎛V ⎞

α e = tan −1 ⎜ ⎟ and βe = tan −1 ⎜ e ⎟ . (5.2)

⎜ U e2 + Ve2

⎝

⎟

⎠ ⎝ Ue ⎠

145

www.ebook777.com

free ebooks ==> www.ebook777.com

146 Flight Dynamics, Simulation, and Control

Hence,

2

Ve = U e tan ( βe ) and We = U e2 + Ve2 tan ( α e ) = U e 1 + ( tan ( βe ) ) tan ( α e ) .

(5.3)

⎢ ⎥

⎢ m ( Δv + ΔrU − ΔpW ) ⎥

e e

⎢ ⎥

⎢ ⎥

⎢⎣ m ( Δw + ΔpVe − ΔqU e ) ⎥⎦

⎡0 ⎤

⎢ ⎥

= ( FNB − FNBe ) + TBW ( α e , βe ) ( FAS − FASe ) + mg ( TBI − TBIe ) ⎢0 ⎥ (5.4)

⎢ ⎥

⎣⎢1 ⎥⎦

and

⎡ I xx

s

0 −I xz

s

⎤ ⎡ Δp ⎤

⎢ ⎥⎢ ⎥

⎢ 0 I yy

s

0 ⎥ ⎢ Δq ⎥ = ( M NB − M NBe ) + TBW ( α e , βe ) ( M AS − M ASe ) . (5.5)

⎢ −I xz I zz

0 ⎦ ⎣ Δr ⎦

s s ⎥

⎢ ⎥

⎣

⎡0 ⎤ ⎡ 0 −cos θe 0 ⎤ ⎡ Δφ ⎤

0 ⎥ ⎢ Δθ ⎥ . (5.6)

⎢ ⎥ ⎢ ⎥⎢ ⎥

mg ( TBI − TBIe ) ⎢0 ⎥ ≈ mg ⎢ cos φe cos θe −sin φe sin θe

⎢⎣ 1 ⎥⎦ ⎢⎣ −sin φe cos θe −cos φe sin θe 0 ⎥⎦ ⎢⎣ Δψ ⎥⎦

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎢ Δθ ⎥ = ⎢ 0 cos φe − sin φe ⎥ ⎢ Δq ⎥ . (5.7)

⎢ Δψ ⎥ ⎢ sin φ cos θ cos φe cos θe 0 ⎥⎦ ⎢⎣ Δr ⎥⎦

e e

⎢⎣ ⎥⎦ ⎣

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 147

tion equations for the vehicle position:

⎡ Δx ⎤ ⎡ U e + Δu ⎤ ⎡U e ⎤

d ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

Δy = TIB ( ψ e + Δψ , θe + Δθ, φe + Δφ ) ⎢ Ve + Δv ⎥ − TIB ( ψ e , θe , φe ) ⎢ Ve ⎥

dt ⎢ ⎥

⎢⎣ Δz ⎥⎦ ⎢⎣We + Δw ⎥⎦ ⎢⎣We ⎥⎦

⎡ Δu ⎤ ⎡U e ⎤

⎢ ⎥ ⎢ ⎥

= TIB ( ψ e , θe , φe ) ⎢ Δv ⎥ + ΔTIB ( ψ e , θe , φe ) ⎢ Ve ⎥ , (5.8)

⎢⎣ Δw ⎥⎦ ⎢⎣We ⎥⎦

where

⎢ ⎥⎢ ⎥⎢ ⎥

TIB ( ψ , θ, φ ) = ⎢ sin ψ cos ψ 0⎥ ⎢ 0 1 0 ⎥ ⎢0 cos φ −sin φ ⎥ ,

⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ − sin θ 0 cos θ ⎥⎦ ⎢⎣0 sin φ cos φ ⎥⎦

(5.9)

T

and ⎡⎣ Δu Δv Δw ⎤⎦ are the components of the aircraft perturbation veloc-

ity vector in the body axes.

In the stability axes, the aerodynamic forces and moments can be expressed

in terms of the aerodynamic forces and moments in the wind axes:

and

www.ebook777.com

free ebooks ==> www.ebook777.com

148 Flight Dynamics, Simulation, and Control

Δβ = β − βe, and it follows that

⎡ 1 −Δβ −Δα ⎤

⎢ ⎥

TBW ( α − α e , β − βe ) = TBW ( Δα , Δβ ) = ⎢ Δβ 1 0 ⎥ . (5.12)

⎢⎣ Δα 0 1 ⎥⎦

Hence,

⎢ ⎥ ⎢ ⎥

FAS = ⎢ Δβ 1 0 ⎥ FAW and M AS = ⎢ Δβ 1 0 ⎥ M AW .

⎢⎣ Δα 0 1 ⎥⎦ ⎢⎣ Δα 0 1 ⎥⎦

(5.13)

Further, since FASe = FAWe and M ASe = M AWe, the aerodynamic forces and

moments in the stability axes may be conveniently expressed in terms of the

forces and moments in the wind axes.

and Moments: Stability Derivative Concept

The aerodynamic forces and moments in the wind axes all are functions of

the dynamic pressure that is defined as

1 1

q=

2

( )

ρ u2 + v 2 + w 2 = ρU 2 ,

2

(5.14)

and it has the units of pressure. Then the aerodynamic forces and moments

in the wind axes can be expressed in terms of certain dimensionless aero-

dynamic force coefficients, CWX, CWY and CWZ, and dimensionless aerodynamic

moment coefficients, CWL, CWM and CWN, a reference area (usually the area of

the wing planform), S, and certain reference lengths, lL, lM and lN:

⎡ XW ⎤ ⎡CWX ⎤ ⎡ CD ⎤

⎢ ⎥ ⎢ ⎥ 1 2 ⎢ ⎥

FAW = ⎢ YW ⎥ = qS ⎢ CWY ⎥ = − ρV S ⎢ CY ⎥ (5.15)

2

⎢⎣ ZW ⎥⎦ ⎢⎣ CWZ ⎥⎦ ⎢⎣ CL ⎥⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 149

and

⎡ LW ⎤ ⎡ lLCWL ⎤ ⎡ cCm ⎤

⎢ ⎥ ⎢ ⎥ 1 ⎢ ⎥

M AW = ⎢ MW ⎥ = qS ⎢lMCWM ⎥ = ρV 2S ⎢ −bCl ⎥ (5.16)

2

⎢⎣ NW ⎥⎦ ⎢⎣ lN CWN ⎥⎦ ⎢ −bCn ⎥

⎣ ⎦

where

c is the mean aerodynamic chord

b is the span

The coefficient CWX is equal to the negative of the drag coefficient, CD, that

is, CWX = −CD, and the coefficient CWL is equal to the negative of the lift

coefficient, CL, that is, CWZ = −CL.

Conceptually, one may expand the wind axis forces and moments in the form

of Taylor’s series in terms of the wind axis perturbations and retain the most

relevant terms. As a result of this, we may write the forces and moments as

⎡ XW ⎤ ⎡ XWe ⎤ ⎡ ⎡ XW ⎤ ⎡ XW ⎤ ⎡ XW ⎤ ⎤ ⎡ Δu ⎤ ⎡ XW ⎤

⎢ ⎥ ⎢ ⎥ ⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥⎢ ⎥ ∂ ⎢ ⎥

⎢ YW ⎥ ≈ ⎢ YWe ⎥ + ⎢ ∂u ⎢ YW ⎥ ∂v ⎢ YW ⎥ ∂w ⎢ YW ⎥ ⎥ ⎢ Δv ⎥ +

∂w ⎢

YW ⎥ Δw wind

⎢⎣ ZW ⎥⎦ ⎢⎣ ZWe ⎥⎦ ⎢ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎥ ⎣⎢ Δw ⎥⎦ ⎢⎣ ZW ⎥⎦

⎣ e e e⎦ wind e

⎡ ⎡ XW ⎤ ⎡ XW ⎤ ⎡ XW ⎤ ⎤ ⎡ Δ ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥⎢ ⎥

+⎢ ⎢ YW ⎥ ∂ ⎢ YW ⎥ ∂ ⎢ YW ⎥ ⎥ ⎢ Δ ⎥

⎢∂ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ e e e⎦ wind

⎡ ⎡ XW ⎤ ⎡ XW ⎤ ⎡ XW ⎤ ⎤ ⎡ Δ ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

⎢ YW ⎥ ∂ ⎢ YW ⎥ ∂ ⎢ YW ⎥ ⎥ ⎢ Δ ⎥

+⎢ (5.17a)

⎢∂ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ e e e⎦ wind

⎡ LW ⎤ ⎡ LWe ⎤ ⎡ ⎡ LW ⎤ ⎡ LW ⎤ ⎡ LW ⎤ ⎤ ⎡ Δu ⎤ ⎡ LW ⎤

⎢ ⎥ ⎢ ⎥ ⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥⎢ ⎥ ∂ ⎢ ⎥

M ≈ M + M M M

⎢ W ⎥ ⎢ We ⎥ ⎢ ∂u ⎢ W ⎥ ∂v ⎢ W ⎥ ∂w ⎢ W ⎥ ⎥ ⎢ ⎥ Δv + MW ⎥ Δw wind

∂w ⎢

⎢⎣ NW ⎥⎦ ⎢⎣ NWe ⎥⎦ ⎢ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎥ ⎣⎢ Δw ⎥⎦ ⎢⎣ NW ⎥⎦

⎣ e e e⎦ wind e

⎡ ⎡ LW ⎤ ⎡ LW ⎤ ⎡ LW ⎤ ⎤ ⎡ Δ ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥⎢ ⎥

+⎢ ⎢ MW ⎥ ∂ ⎢ MW ⎥ ∂ ⎢ MW ⎥ ⎥ ⎢ Δ ⎥

⎢∂ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ e e e⎦ wind

⎡ ⎡ LW ⎤ ⎡ LW ⎤ ⎡ LW ⎤ ⎤ ⎡ Δ ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

⎢ MW ⎥ ∂ ⎢ MW ⎥ ∂ ⎢ MW ⎥ ⎥ ⎢ Δ ⎥

+⎢ (5.17b)

⎢∂ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ e e e⎦ wind

where the derivatives are evaluated for the conditions for trimmed flight.

www.ebook777.com

free ebooks ==> www.ebook777.com

150 Flight Dynamics, Simulation, and Control

velocities,

⎡ XW ⎤ ⎡ XWe ⎤ ⎡ ⎡ XW ⎤ ⎡ XW ⎤ ⎡ XW ⎤ ⎤ ⎡ Δu ⎤

⎢ ⎥ ⎢ ⎥ ⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

⎢ YW ⎥ ≈ ⎢ YWe ⎥ + ⎢ ∂u ⎢ YW ⎥ ∂v ⎢ YW ⎥ ∂w ⎢ YW ⎥ ⎥ TWB ( α , β ) ⎢ Δv ⎥

⎢⎣ ZW ⎥⎦ ⎢⎣ ZWe ⎥⎦ ⎢ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎥ ⎢⎣ Δw ⎥⎦

⎣ e e e⎦

⎡ ⎡ XW ⎤ ⎡ XW ⎤ ⎡ XW ⎤ ⎤ ⎡ Δp ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

+ ⎢ ⎢ YW ⎥ YW ⎥ YW ⎥ ⎥ TWB ( α , β ) ⎢ Δq ⎥

⎢ ∂p ⎢ ZW ⎥ ∂q ⎢ ZW ⎥ ∂ ⎢ ZW ⎥ ⎥

⎢ ⎢

⎢⎣ Δ ⎥⎦

⎣ ⎣ ⎦e ⎣ ⎦e ⎣ ⎦e ⎦

⎡ XW ⎤ ⎡0 ⎤

∂ ⎢ ⎥ ⎢ ⎥

+ YW ⎥ TWB ( α , β ) ⎢0 ⎥ Δw

∂w ⎢

⎢⎣ ZW ⎥⎦ ⎢⎣ 1 ⎥⎦

e

⎡ ⎡ XW ⎤ ⎡ XW ⎤ ⎡ XW ⎤ ⎤ ⎡ Δp ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

+ ⎢ ⎢ YW ⎥ YW ⎥ YW ⎥ ⎥ TWB ( α , β ) ⎢ Δq ⎥ (5.18a)

∂p

⎢ ⎢ ZW ⎥ ∂q⎢ ∂ ⎢

⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ ⎣ ⎦e e e⎦

and

⎡ LW ⎤ ⎡ LWe ⎤ ⎡ ⎡ LW ⎤ ⎡ LW ⎤ ⎡ LW ⎤ ⎤ ⎡ Δu ⎤

⎢ ⎥ ⎢ ⎥ ⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

M ≈ M + M M

⎢ W ⎥ ⎢ We ⎥ ⎢ ∂u ⎢ W ⎥ ∂v ⎢ W ⎥ ∂w ⎢ W ⎥ ⎥ WBM T ( α , β ) ⎢ Δv ⎥

⎢⎣ NW ⎥⎦ ⎢⎣ NWe ⎥⎦

⎣ ⎣ NW ⎦ e

⎢ ⎢ ⎥ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎥ ⎢⎣ Δw ⎥⎦

e e⎦

⎡ ⎡ LW ⎤ ⎡ LW ⎤ ⎡ LW ⎤ ⎤ ⎡ Δp ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

+ ⎢ ⎢ MW ⎥

∂p ∂q ⎢ MW ⎥ ∂ ⎢ MW ⎥ ⎥ TWB ( α , β ) ⎢ Δq ⎥

⎢ ⎢ NW ⎥ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ ⎣ ⎦e e e⎦

⎡0 ⎤

⎡ LW ⎤ ⎢ ⎥

∂ ⎢ ⎥

+ MW TWB ( α , β ) ⎢0 ⎥ Δw

∂w ⎢ ⎥ ⎢ ⎥

⎢⎣ NW ⎥⎦ ⎢⎣1 ⎥⎦

e

⎡ ⎡ LW ⎤ ⎡ LW ⎤ ⎡ LW ⎤ ⎤ ⎡ Δp ⎤

⎢∂ ⎢ ⎢ ⎥

⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥

+ ⎢ ⎢ MW ⎥ MW MW ⎥ TWB ( α , β ) ⎢ Δq ⎥ (5.18b)

∂p

⎢ ⎢ NW ⎥ ∂q⎢ ⎥ ∂ ⎢ ⎥ ⎢ ⎥

⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎥

⎣ ⎣ ⎦e e e⎦ ⎢⎣ Δ ⎥⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 151

T

⎡ 1 −Δβ −Δα ⎤

⎢ ⎥

TWB ( α , β ) = TWB ( α e , βe ) TWB ( α − α e , β − βe ) = TWB ( α e , βe ) ⎢ Δβ 1 0 ⎥ .

⎢⎣ Δα 0 1 ⎥⎦

(5.19)

Hence, retaining only the linear terms, the aerodynamic forces and moments

are approximated as

⎡ XW ⎤ ⎡ XWe ⎤ ⎡ ⎡ XW ⎤ ⎡ XW ⎤ ⎡ XW ⎤ ⎤ ⎡ Δu ⎤

⎢ ⎥ ⎢ ⎥ ⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

⎢ YW ⎥ ≈ ⎢ YWe ⎥ + ⎢ ∂u ⎢ YW ⎥ ∂v ⎢ YW ⎥ ∂w ⎢ YW ⎥ ⎥ TWB ( α e , βe ) ⎢ Δv ⎥

⎢⎣ ZW ⎥⎦ ⎢⎣ ZWe ⎥⎦ ⎢ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎥ ⎢⎣ Δw ⎥⎦

⎣ e e e⎦

⎡ ⎡ XW ⎤ ⎡ XW ⎤ ⎡ XW ⎤ ⎤ ⎡ Δp ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

+ ⎢ ⎢ YW ⎥ Y Y T

⎢ W ⎥ ∂ ⎢ W ⎥ ⎥ WB ( e e ) ⎢ Δ ⎥ α , β

∂p

⎢ ⎢ ZW ⎥ ∂

⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ ⎣ ⎦e e e⎦

⎡ XW ⎤ ⎡0 ⎤

∂ ⎢ ⎥ ⎢ ⎥

+ YW ⎥ TWB ( α e , βe ) ⎢0 ⎥ Δw

∂w ⎢

⎢⎣ ZW ⎥⎦ ⎢⎣ 1 ⎥⎦

e

⎡ ⎡ XW ⎤ ⎡ XW ⎤ ⎡ XW ⎤ ⎤ ⎡ Δp ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

+ ⎢ ⎢ YW ⎥ YW ⎥ YW ⎥ ⎥ TWB ( α e , βe ) ⎢ Δ ⎥ (5.20a)

∂p

⎢ ⎢ ZW ⎥ ∂ ⎢ ∂ ⎢

⎢⎣ ZW ⎥⎦ ⎢⎣ ZW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ ⎣ ⎦e e e⎦

and

⎡ LW ⎤ ⎡ LWe ⎤ ⎡ ⎡ LW ⎤ ⎡ LW ⎤ ⎡ LW ⎤ ⎤ ⎡ Δu ⎤

⎢ ⎥ ⎢ ⎥ ⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

⎢ MW ⎥ ≈ ⎢ MWe ⎥ + ⎢ ∂u ⎢ MW ⎥ ∂v ⎢ MW ⎥ ∂w ⎢ MW ⎥ ⎥ TWB ( α e , βe ) ⎢ Δv ⎥

⎢⎣ NW ⎥⎦ ⎢⎣ NWe ⎥⎦ ⎢ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎥ ⎢⎣ Δw ⎥⎦

⎣ e e e⎦

⎡ ⎡ LW ⎤ ⎡ LW ⎤ ⎡ LW ⎤ ⎤ ⎡ Δp ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

+ ⎢ ⎢ MW ⎥

∂ p ∂ ⎢ MW ⎥ ∂ ⎢ MW ⎥ ⎥ TWB ( α e , βe ) ⎢ Δ ⎥

⎢ ⎢ NW ⎥ ⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ ⎣ ⎦e e e⎦

⎡ LW ⎤ ⎡0 ⎤

∂ ⎢ ⎥ ⎢ ⎥

+ MW ⎥ TWB ( α e , βe ) ⎢0 ⎥ Δw

∂w ⎢

⎢⎣ NW ⎥⎦ ⎢⎣1 ⎥⎦

e

⎡ ⎡ LW ⎤ ⎡ LW ⎤ ⎡ LW ⎤ ⎤ ⎡ Δp ⎤

⎢∂ ⎢ ⎥ ∂ ⎢ ⎥ ∂ ⎢ ⎥ ⎥ ⎢ ⎥

+ ⎢ ⎢ MW ⎥ MW ⎥ MW ⎥ ⎥ TWB ( α e , βe ) ⎢ Δ ⎥ (5.20b)

∂p

⎢ ⎢ NW ⎥ ∂ ⎢ ∂ ⎢

⎢⎣ NW ⎥⎦ ⎢⎣ NW ⎥⎦ ⎥ ⎢⎣ Δ ⎥⎦

⎣ ⎣ ⎦e e e⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

152 Flight Dynamics, Simulation, and Control

derivatives of aerodynamic forces and moments evaluated under equilibrium

conditions and referred to the stability axes. It is normally much more

convenient to derive the linearised equations of motion directly in the stabil-

ity axes.

The stability axis is in fact a body-fixed axis system although it is not aligned

with the principal axis of the aircraft. To transform the moments of inertia

to the stability axis, it is essential to recognise the fact that rotational kinetic

energy of a rigid body is invariant in any axis system. Thus,

1 T 1 1

Trot = ωb Ibωb = ωTs Isωs = ωTs TWB ( α e , βe ) Ib TBW ( α e , βe ) ωs . (5.21)

2 2 2

⎡ I xx

s

−I xy

s

−I xz

s

⎤

⎢ s s ⎥

Is = ⎢ −I xy I yy

s

−I yz ⎥ = TWB ( α e , βe ) Ib TBW ( α e , βe )

⎢ −I xz

s

−I yz

s

I zz

s ⎥

⎣ ⎦

⎡ I xx 0 −I xz ⎤

⎢ ⎥

= TWB ( α e , βe ) ⎢ 0 I yy 0 ⎥ TBW ( α e , βe ) . (5.22)

⎢⎣ −I xz 0 I zz ⎥⎦

⎡ u s + qs ws − rsvs ⎤

⎢ ⎥

m ⎢ v s + rsus − ps ws ⎥ = TWB ( α e , βe ) ( FNB − FNBe ) + ( FAS − FASe )

⎢⎣ w s + psvs − qsus ⎥⎦

⎡0 ⎤

⎢ ⎥

+ mgTWB ( α e , βe ) ( TBI − TBIe ) ⎢0 ⎥ (5.23a)

⎢⎣ 1 ⎥⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 153

⎡ I xx

s

−I xy

s

−I xz

s ⎡ ( ) ( s 2

)

⎤ ⎡ p s ⎤ ⎢ qs rs I zz − I yy − I xz ps qs + I xy ps rs + I yz rs − qs ⎥

s s s s 2 ⎤

⎢ s s ⎥⎢ ⎥ ⎢

⎢ −I xy I yy

s

−I yz ( s

)

s s s

( s 2

)

⎥ ⎢ q s ⎥ + ⎢ rs ps I xx − I zz − I xy qs rs + I yz qs ps + I xz ps − rs ⎥ .

2 ⎥

⎢ −I xz −I yz I zz ⎦ ⎣ rs ⎦ ⎢ ps qs I ys y − I xx

s s s ⎥

⎢ ⎥

⎣

⎣ ( )

s

− I yz

s

rs ps + I xz

s

(

qs rs + I xy

s

)

qs2 − ps2 ⎥

⎦

= TWB ( α e , βe ) ( M NB − M NBe ) + ( M AS − M ASe )

(5.23b)

The small perturbation approximations, compatible with conditions of trim

in the stability axes, are considerably simpler. The aircraft’s velocities and

angular velocities are perturbed about a steady operating condition. Thus,

the perturbations satisfy

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ vs ⎥ = ⎢ Δvs ⎥ , ⎢ Δvs ⎥ = TWB ( α e , βe ) ⎢ Δv ⎥ , ⎢ qs ⎥ = ⎢ Δqs ⎥ = TWB ( α e , βe ) ⎢ Δq ⎥ ,

⎢ ws ⎥ ⎢ Δws ⎥ ⎢⎣ Δws ⎥⎦ ⎢⎣ Δw ⎥⎦ ⎢⎣ rs ⎥⎦ ⎢⎣ Δrs ⎥⎦ ⎢⎣ Δr ⎥⎦

⎣ ⎦ ⎣ ⎦

(5.24)

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ θs ⎥ = TWB ( α e , βe ) ⎢ θ ⎥ , ⎢ θse ⎥ = TWB ( α e , βe ) ⎢ θe ⎥ and ⎢ Δθs ⎥ = TWB ( α e , βe ) ⎢ Δθ ⎥

⎢ψ s ⎥ ⎢⎣ψ ⎥⎦ ⎢⎣ψ se ⎦⎥ ⎢⎣ψ e ⎥⎦ ⎢⎣ Δψ s ⎥⎦ ⎢⎣ Δψ ⎥⎦

⎣ ⎦

(5.25)

where

⎡ cos βe sin βe 0 ⎤ ⎡ cos α e 0 sin α e ⎤

⎢ ⎥⎢ ⎥

TWB ( α e , βe ) = ⎢ − sin βe cos βe 0⎥ ⎢ 0 1 0 ⎥ (5.26)

⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ − sin α e 0 cos α e ⎥⎦

and U es represents the steady trimmed velocity of the aircraft and the steady-

state stability axis angular velocities are each assumed to be equal to 0.

The linearised equations of motion in the stability axes are then given by

⎡ mΔu s ⎤

⎢ ⎥

(

⎢ m Δv s + ΔrsU es

⎢

) ⎥ = TWB ( α e , βe ) ( FNB − FNBe ) + ( FAS − FASe )

⎥

⎣ (

⎢ m Δw − Δq U s

s s e ) ⎥

⎦

⎡0 ⎤

⎢ ⎥

+ mgTWB ( α e , βe ) ( TBI − TBIe ) ⎢0 ⎥ (5.27)

⎢⎣ 1 ⎥⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

154 Flight Dynamics, Simulation, and Control

or

⎡ mΔu s ⎤

⎢ ⎥

(

⎢ m Δv s + ΔrsU es

⎢

) ⎥ = TWB ( α e , βe ) ( FNB − FNBe ) + ( FAS − FASe )

⎥

⎣ (

⎢ m Δw − Δq U s

s s e ) ⎥

⎦

⎡ 0 −cos θe ⎤

⎢ ⎥ ⎡ Δφ ⎤

+ mgTWB ( α e , βe ) ⎢ cos φe cos θe −sin φe sin θe ⎥ ⎢ ⎥ ,

Δθ

⎢⎣ −sin φe cos θe −cos φe sin θe ⎥⎦ ⎣ ⎦

(5.28)

and

⎡ I xx

s

−I xy

s

−I xz

s

⎤ ⎡ Δp s ⎤

⎢ s s ⎥⎢ ⎥

⎢ −I xy I yy

s

−I yz ⎥ ⎢ Δq s ⎥ = TWB ( α e , βe ) ( M NB − M NBe ) + ( M AS − M ASe ) .

⎢ −I xz −I yz I zz ⎦ ⎣ Δrs ⎦

s s s ⎥

⎢ ⎥

⎣

(5.29)

in the stability axes are defined as

⎡ ⎤ ⎡ ⎤

⎢XW ⎥ ⎢ LW ⎥

⎢ ⎥ ⎢ ⎥

FAS − FASe = ⎢ YW ⎥ and M AS − M ASe = ⎢ MW ⎥ . (5.30)

⎢ ⎥ ⎢ ⎥

⎢ ZW ⎥ ⎢ NW ⎥

⎣ ⎦ stability axes ⎣ ⎦ stability axes

vectors in the stability axes may be expressed in terms of the stability

derivatives:

⎡ ⎤ ⎡ ⎤ ⎡ ⎤

⎢ X W

⎥ ⎢ X W

⎥ ⎢XW ⎥

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ YW ⎥ = ⎢ YW ⎥ + ⎢ YW ⎥ (5.31)

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ ZW ⎥ ⎢ ZW ⎥ ⎢ ZW ⎥

⎣ ⎦ stability ⎣ ⎦ without controls ⎣ ⎦ controls

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 155

and

⎡ ⎤ ⎡ ⎤ ⎡ ⎤

⎢ LW ⎥ ⎢ LW ⎥ ⎢ LW ⎥

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ MW ⎥ = ⎢ MW ⎥ + ⎢ MW ⎥ , (5.32)

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ NW ⎥ ⎢ NW ⎥ ⎢ NW ⎥

⎣ ⎦ stability ⎣ ⎦ without controls ⎣ ⎦ controls

with the contributions of the basic aircraft with the controls locked as

⎡ ⎤ ⎡

⎤ ⎡ ⎤

⎢XW ⎥ ⎢X u Xv Xw ⎥

⎡ Δus ⎤ ⎢ ⎥

X w

⎢ ⎥ ⎢ ⎥

⎢ ⎥ ⎢ ⎥

⎢ YW ⎥ ≈ ⎢Yu Yv Y w ⎥ ⎢ Δvs ⎥ + ⎢ Y w ⎥ Δw s

⎢ ⎥ ⎢ ⎥ ⎢ Δw ⎥ ⎢ ⎥

⎢ ZW ⎥ ⎢ Zu Zv Z w ⎥ ⎣ s ⎦ ⎢ Z w ⎥

⎣ ⎦ without controls ⎣ ⎦ ⎣ ⎦

⎡ ⎤ ⎡ ⎤

⎢X p Xq Xr ⎥

⎡ Δps ⎤ ⎢

X p X q X r ⎥

⎡ Δp s ⎤

⎢ ⎥

⎢ ⎥ ⎢ ⎥

⎢ ⎥

+ ⎢Y p Yq Y r ⎥ ⎢ Δqs ⎥ + ⎢ Y p Y q Y r ⎥ ⎢ Δq s ⎥

(5.33)

⎢ ⎥ ⎢ Δr ⎥ ⎢ ⎥ ⎢ Δr ⎥

⎢ Zp Zq Z r ⎥ ⎣ s ⎦ ⎢ Z p Z q Z r ⎥ ⎣ s ⎦

⎣ ⎦ ⎣ ⎦

and

⎡ ⎤ ⎡

⎤ ⎡ ⎤

⎢ LW ⎥ ⎢ Lu Lv Lw ⎥

⎡ Δus ⎤ ⎢

Lw ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢ MW ⎥ ≈ ⎢ Mu Mv M w ⎥ ⎢ Δvs ⎥ + ⎢ M w ⎥ Δw s

⎢ ⎥ ⎢ ⎥ ⎢ Δw ⎥ ⎢ ⎥

⎢ NW ⎥ ⎢ Nu Nv N w ⎥ ⎣ s ⎦ ⎢ N w ⎥

⎣ ⎦ without controls ⎣ ⎦ ⎣ ⎦

⎡

⎤ ⎡

⎤

⎢ Lp Lq Lr ⎥

⎡ Δps ⎤ ⎢

L p Lq Lr ⎥

⎡ Δp s ⎤

⎢ ⎥

⎢ ⎥ ⎢ ⎥

⎢ ⎥

+ ⎢Mp Mq M r ⎥ ⎢ Δqs ⎥ + ⎢ M p M q M r ⎥ ⎢ Δq s ⎥ ,

⎢ ⎥ ⎢ Δr ⎥ ⎢ ⎥ ⎢ Δr ⎥

⎢Np Nq N r ⎥ ⎣ s ⎦ ⎢ N p N q N r ⎥ ⎣ s ⎦

⎣ ⎦ ⎣ ⎦

(5.34)

www.ebook777.com

free ebooks ==> www.ebook777.com

156 Flight Dynamics, Simulation, and Control

⎡ ⎤ ⎡

⎤

⎢XW ⎥ ⎢X η Xτ Xξ X ζ ⎥ ⎡ Δη⎤

⎢ ⎥

⎢ ⎥ ⎢ ⎥ Δτ

⎢ YW ⎥ ≈ ⎢Yη Yτ Yξ Y ζ ⎥ ⎢ ⎥ (5.35)

⎢ Δξ ⎥

⎢ ⎥ ⎢ ⎥

⎢ ⎥

⎢ ZW ⎥ ⎢ Zη Zτ Zξ Z ζ ⎥ ⎣ Δζ ⎦

⎣ ⎦ controls ⎣ ⎦

and

⎡ ⎤ ⎡

⎤

⎢ LW

⎥ ⎢ Lη Lτ Lξ Lζ ⎥ ⎡ Δη⎤

⎢ ⎥ ⎢ ⎥ Δτ

⎢ ⎥

⎢ MW ⎥ ≈ ⎢ Mη Mτ Mξ M ζ ⎥ ⎢ ⎥ . (5.36)

⎢ Δξ ⎥

⎢ ⎥ ⎢ ⎥

⎢ ⎥

⎢ NW ⎥ ⎢ Nη Nτ Nξ N ζ ⎥ ⎣ Δζ ⎦

⎣ ⎦ controls ⎣ ⎦

moments of the most relevant perturbations are included.

There is an important relationship that is useful in understanding the trans-

formation between the wind axis aerodynamic force and moment perturbation

vector and the corresponding vectors in the stability axes. This is particularly

useful in estimating the aerodynamic stability derivatives in the stability axes,

as the aerodynamic forces and moments are generally derived in the wind axes.

In the stability axes, the aerodynamic forces and moments can be expressed in

terms of the aerodynamic forces and moments in the wind axes:

and

However, considering trimmed flight, FASe = FAWe and M ASe = M AWe. Subtracting

the latter pair of equations from the former pair, it follows that

= FAW − FAWe + ( TBW ( α − α e , β − βe ) − I ) FAW , (5.38a)

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 157

(5.39a)

(5.39b)

and

But

⎢ ⎥⎢ ⎥

TBW ( α e , βe ) = ⎢ 0 1 0 ⎥ ⎢ sin βe cos βe 0 ⎥ , (5.41)

⎢⎣ sin α e 0 cos α e ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦

⎢ ⎥⎢ ⎥

TWB ( α e , βe ) = ⎢ − sin βe cos βe 0⎥ ⎢ 0 1 0 ⎥ (5.42)

⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ − sin α e 0 cos α e ⎥⎦

and

⎡ 1 −Δβ −Δα ⎤

⎢ ⎥

TBW ( α − α e , β − βe ) = TBW ( Δα , Δβ ) = ⎢ Δβ 1 0 ⎥ , (5.43)

⎢⎣ Δα 0 1 ⎥⎦

⎡ 1 Δβ Δα ⎤

⎢ ⎥

TWB ( α − α e , β − βe ) = TWB ( Δα , Δβ ) = ⎢ −Δβ 1 0 ⎥ . (5.44)

⎢⎣ −Δα 0 1 ⎥⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

158 Flight Dynamics, Simulation, and Control

⎡ 1 Δβ Δα ⎤ ⎡ 1 Δβ Δα ⎤

⎢ ⎥ ⎢ ⎥

FAW = ⎢ −Δβ 1 0 ⎥ ( FAS − FASe ) + ⎢ −Δβ 1 0 ⎥ FAWe , (5.45a)

⎢⎣ −Δα 0 1 ⎥⎦ ⎢⎣ −Δα 0 1 ⎥⎦

and

⎡ 1 Δβ Δα ⎤ ⎡ 1 Δβ Δα ⎤

⎢ ⎥ ⎢ ⎥

M AW = ⎢ −Δβ 1 0 ⎥ ( M AS − M ASe ) + ⎢ −Δβ 1 0 ⎥ M AWe . (5.45b)

⎢⎣ −Δα 0 1 ⎥⎦ ⎣⎢ −Δα 0 1 ⎥⎦

⎢ ⎥ ⎢ ⎥

FAS − FASe = ⎢ Δβ 1 0 ⎥ ( FAW − FAWe ) + ⎢ Δβ 1 0 ⎥ FAWe ,

⎢⎣ Δα 0 1 ⎥⎦ ⎢⎣ Δα 0 1 ⎥⎦

(5.46a)

⎢ ⎥ ⎢ ⎥

M AS − M ASe = ⎢ Δβ 1 0 ⎥ ( M AW − M AWe ) + ⎢ Δβ 1 0 ⎥ M AWe .

⎢⎣ Δα 0 1 ⎥⎦ ⎢⎣ Δα 0 1 ⎥⎦

(5.46b)

5.5.1 Case I: Motion in the Longitudinal Plane of Symmetry

Decoupled longitudinal motion is motion in response to a disturbance that

is confined to the longitudinal plane of symmetry, that is, the Oxz plane. The

only forces that are responsible for this motion are the axial force X and the

normal force Z. The pitching moment M is the only moment that is respon-

sible for angular motions in the plane of symmetry. Since it is assumed that

there is no lateral motion, all the lateral motion variables, p, r and v, and the

derivatives of X, Z and M with respect to these variables are assumed to be

zero. The equations of longitudinal symmetric motion are therefore obtained

by extracting the axial and normal force equations and the pitching moment

equation from the complete dynamical equations of motion and setting all

coupling terms to zero. In particular, both βe = 0 and Ve = 0.

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 159

mΔu s ⎡ ⎤

⎡ ⎤ ⎡ cos α e sin α e ⎤ ⎡ ⎡ X NB ⎤ ⎡ X NBe ⎤ ⎤ ⎢ X W ⎥

⎢ ⎥=⎢ ⎢ − ⎥+

(

⎢⎣ m Δw s − ΔqU e

s

) ⎥⎦ ⎣ − sin α e cos α e ⎥⎦ ⎢⎣ ⎢⎣ ZNB ⎥⎦ ⎢⎣ ZNBe ⎥⎦ ⎥⎦ ⎢ ⎥

⎢⎣ ZW ⎥⎦

+ mg ⎢ Δθ (5.47a)

⎣ − sin α e cos α e ⎥⎦ ⎢⎣ sin θe ⎥⎦

I yy

s

Δq s = MNB − MNBe + MW (5.47b)

where the components of the aerodynamic force perturbation vector and the

pitching moment in the stability axes are related to the corresponding wind

axis components by the relations

⎡ ⎤

⎢ X W ⎥ ⎡ XW ⎤ ⎡ XWe ⎤ ⎡ 0 0 −Δw ⎤ ⎡ XW ⎤

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ 1 ⎢ ⎥⎢ ⎥

⎢ Y W ⎥ = ⎢ YW ⎥ − ⎢ YWe ⎥ + ⎢ 0 1 0 ⎥ ⎢ YW ⎥ (5.48a)

⎢ ⎥ ⎢ Z ⎥ ⎢ Z ⎥ Ue ⎢⎣ Δw 0 0 ⎥⎦ ⎢⎣ ZW ⎥⎦

⎢ ZW ⎥ ⎣ W ⎦ ⎣ We ⎦

⎣ ⎦

MW = MW − MWe , (5.48b)

and perturbation pitch angle in the stability axis is equal to the perturbation

pitch angle in the wind axes, Δθs = Δθ. The perturbation of the angle of attack

in the body axes is

Δw Δw

Δα ≈ ≈ . (5.49)

U +V 2

e e

2 Ue

Further,

⎡ Δu ⎤ ⎡ Δus ⎤ ⎡ Δφ ⎤ ⎡ Δφs ⎤

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ Δv ⎥ = TBW ( α e , βe ) ⎢ Δvs ⎥ , ⎢ Δθ ⎥ = TBW ( α e , βe ) ⎢ Δθs ⎥ (5.50)

⎢⎣ Δw ⎥⎦ ⎢⎣ Δws ⎥⎦ ⎢⎣ Δψ ⎥⎦ ⎢⎣ Δψ s ⎥⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

160 Flight Dynamics, Simulation, and Control

where

⎢ ⎥⎢ ⎥

TBW ( α e , βe ) = ⎢ 0 1 0 ⎥ ⎢ sin βe cos βe 0⎥

⎢⎣ sin α e 0 cos α e ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦

⎢ ⎥

= ⎢ sin βe cos βe 0 ⎥ (5.51)

⎢⎣ sin α e cos βe − sin α e sin βe cos α e ⎥⎦

The aerodynamic

force and moment perturbation vectors in the stability axes,

X W, ZW and MW , may be expressed in terms of the stability axis aerodynamic

derivatives introduced earlier.

Perpendicular to the Plane of Symmetry

In the case of motion in the lateral direction, the motion is adequately

described by the side force Y and the rolling and yawing moments L and N.

Since the longitudinal motion variables, Δus, Δws, Δw s and Δqs, are assumed

to be zero, the derivatives of Y, L and N with respect to these motion coor-

dinates are also assumed to be zero. The equations of lateral asymmetric

motion are therefore obtained by extracting the side force and the rolling

and yawing moment equations from the complete dynamical equations of

motion and setting all coupling terms to zero.

It is also customary to write these equations in terms of the stability axis

perturbation side force, Y W , and the stability axis perturbation rolling and

yawing moments, LW and N W , as

( )

m Δv s + ΔrsU es = cos βe ( YNB − YNBe ) + Y A + mg cos βe cos θe cos φe Δφ (5.53a)

⎡ ⎤

⎡ I xx

s

−I xz

s

⎤ ⎡ Δp s ⎤ ⎡cos βe 0 ⎤ ⎡ cos α e sinα e ⎤ ⎡ ⎡ LNB ⎤ ⎡ LNBe ⎤ ⎤ ⎢ LW ⎥

⎢ s s ⎥⎢ ⎥

=⎢ ⎢ − ⎥+

⎣ −I xz I zz ⎦ ⎣ Δrs ⎦ ⎣ 0 1 ⎥⎦ ⎢⎣ − sin α e cos α e ⎥⎦ ⎢⎣ ⎢⎣ N NB ⎥⎦ ⎢⎣ N NBe ⎥⎦ ⎥⎦ ⎢ ⎥

⎢⎣ N W ⎥⎦

(5.53b)

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 161

stability axes are related to the corresponding wind axis moments by the relation

⎡ ⎤

⎢ LW ⎥ ⎡ LW ⎤ ⎡ LWe ⎤

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ MW ⎥ = ⎢ MW ⎥ − ⎢ MWe ⎥ . (5.54)

⎢ ⎥ ⎢N ⎥ ⎢N ⎥

⎢ N W ⎥ ⎣ W ⎦ ⎣ We ⎦

⎣ ⎦

Axis Aerodynamic Derivatives

The aerodynamic force and moment components may be expressed in terms

of the stability axis aerodynamic derivatives as

⎡ ⎤ ⎡

⎤ ⎡ ⎤

⎢XW ⎥ ⎢X u Xv Xw ⎥

⎡ Δus ⎤ ⎢ ⎥

X w

⎢ ⎥ ⎢ ⎥

⎢ ⎥ ⎢ ⎥

⎢ YW ⎥ ≈ ⎢ Y u Yv Y w ⎥ ⎢ Δvs ⎥ + ⎢ Y w ⎥ Δw s

⎢ ⎥ ⎢ ⎥ ⎢ Δw ⎥ ⎢ ⎥

⎢ ZW ⎥ ⎢ Z u Zv Z w ⎥ ⎣ s ⎦ ⎢ Z w ⎥

⎣ ⎦ ⎣ ⎦ ⎣ ⎦

⎡

⎤ ⎡

⎤

⎢X p Xq Xr ⎥

⎡ Δps ⎤ ⎢

X p X q X r ⎥

⎡ Δpp s ⎤

⎢ ⎥

⎢ ⎥ ⎢ ⎥

⎢ ⎥

+ ⎢Y p Yq Y r ⎥ ⎢ Δqs ⎥ + ⎢ Y p Y q Y r ⎥ ⎢ Δq s ⎥ . (5.55a)

⎢ ⎥ ⎢ Δr ⎥ ⎢ ⎥ ⎢ Δr ⎥

⎢ Zp Zq Z r ⎥ ⎣ s ⎦ ⎢ Z p Z q Z r ⎥ ⎣ s ⎦

⎣ ⎦ ⎣ ⎦

and

⎡ ⎤ ⎡

⎤ ⎡ ⎤

⎢ LW ⎥ ⎢ Lu Lv Lw ⎥ L w ⎥

⎡ Δus ⎤ ⎢

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢ MW ⎥ ≈ ⎢ M u Mv M w ⎥ ⎢ Δvs ⎥ + ⎢ M w ⎥ Δw s

⎢ ⎥ ⎢ ⎥ ⎢ Δw ⎥ ⎢ ⎥

⎢ NW ⎥ ⎢ N u Nv N w ⎥ ⎣ s ⎦ ⎢ N w ⎥

⎣ ⎦ ⎣ ⎦ ⎣ ⎦

⎡

⎤ ⎡

⎤

⎢ Lp Lq Lr ⎥

⎡ Δps ⎤ ⎢

L p Lq Lr ⎥

⎡ Δpp s ⎤

⎢ ⎥

⎢ ⎥ ⎢ ⎥

⎢ ⎥

+ ⎢Mp Mq M r ⎥ ⎢ Δqs ⎥ + ⎢ M p M q M r ⎥ ⎢ Δq s ⎥ . (5.55b)

⎢ ⎥ ⎢ Δr ⎥ ⎢ ⎥ ⎢ Δr ⎥

⎢Np Nq N r ⎥ ⎣ s ⎦ ⎢ N p N q N r ⎥ ⎣ s ⎦

⎣ ⎦ ⎣ ⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

162 Flight Dynamics, Simulation, and Control

controls omitted or fixed) are

⎡ mΔu s ⎤ ⎡ cos α e sin α e ⎤ ⎡ ⎡ X NB ⎤ ⎡ X NBe ⎤ ⎤

⎢ ⎥=⎢ ⎢ − ⎥

(

⎢⎣ m Δw s − ΔqU e

s

) ⎥⎦ ⎣ − sin α e cos α e ⎥⎦ ⎢⎣ ⎢⎣ ZNB ⎥⎦ ⎢⎣ ZNBe ⎥⎦ ⎥⎦

⎡ ⎤

⎢ Δus X u + Δws X w + Δw s X w + Δqs X q + Δq s X q ⎥

+

⎢ ⎥

⎢⎣ Δus Z u + Δws Z w + Δw s Z w + Δqs Z q + Δq s Z q ⎥⎦

+ mg ⎢ Δθs (5.56a)

⎣ − sin α e cos α e ⎥⎦ ⎢⎣ sin θe ⎥⎦

I yy

s

Δq s = MNB − MNBe + Δus M u + Δws M w + Δw s M w + Δqs M q + Δq s M q (5.56b)

with

Δh ≈ Uγ = U ( Δθ − Δα ) ≈ U e ( Δθ − Δα ) , (5.57b)

Δw Δw

Δα ≈ ≈ , (5.57d)

U +V

2

e e

2 Ue

The lateral equations of motion are

( )

m Δv s + ΔrsU es = cosβe ( YNB − YNBe ) + Δvs Y v + Δps Y p + Δrs Y r + Δp s Y p + Δrs Y r

+ mg cos βe cos θe cos φe Δφ (5.58a)

⎡ I xx

s

−I xz

s

⎤ ⎡ Δp s ⎤ ⎡cos βe 0 ⎤ ⎡ cos α e sin α e ⎤ ⎡ ⎡ LNB ⎤ ⎡ LNBe ⎤ ⎤

⎢ s s ⎥⎢ ⎥

=⎢ ⎢ − ⎥

⎣ −I xz I zz ⎦ ⎣ Δrs ⎦ ⎣ 0 1 ⎥⎦ ⎢⎣ − sin α e cos α e ⎥⎦ ⎢⎣ ⎢⎣ N NB ⎥⎦ ⎢⎣ N NBe ⎥⎦ ⎥⎦

⎡

⎤

⎢ Δvs Lv + Δps L p + Δrs Lr + Δp s L p + Δrs L r ⎥

+ (5.58b)

⎢ ⎥

⎢⎣ Δvs N v + Δps N p + Δrs N r + Δp s N p + Δrs N r ⎥⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 163

with

and

⎡ Δu ⎤ ⎡ Δus ⎤ ⎡ Δφ ⎤ ⎡ Δφs ⎤

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ Δv ⎥ = TBW ( α e , βe ) ⎢ Δvs ⎥ , ⎢ Δθ ⎥ = TBW ( α e , βe ) ⎢ Δθs ⎥ (5.60)

⎢⎣ Δw ⎥⎦ ⎢⎣ Δws ⎥⎦ ⎢⎣ Δψ ⎥⎦ ⎢⎣ Δψ s ⎥⎦

where TBW(α e, βe) and Δw are defined by Equations 5.51 and 5.52 and

The moments of inertia in the stability axes are related to the moments of

inertia in the body axes (principal axes) according to the relations

⎡ I xx

s

−I xy

s

−I xz

s

⎤

⎢ s s ⎥

Is = ⎢ −I xy I yy

s

−I yz ⎥ = TWB ( α e , βe ) Ib TBW ( α e , βe )

⎢ −I xz

s

−I yz

s

I zz

s ⎥

⎣ ⎦

⎡ I xx

b

0 0⎤

⎢ ⎥

= TWB ( α e , βe ) ⎢ 0 I xx

b

0⎥ TBW ( α e , βe ) . (5.62)

⎢0

⎣ 0 I xx ⎦

b ⎥

I xx

s

= I xx

b

cos 2 ( α e ) + I zz

b

sin 2 ( α e ) , (5.63a)

I zz

s

= I xx

b

sin 2 ( α e ) + I zz

b

cos 2 ( α e ) , (5.63b)

and

sin ( 2α e )

I xz

s

(

= I xx

b

− I zz

b

) 2

. (5.63c)

www.ebook777.com

free ebooks ==> www.ebook777.com

164 Flight Dynamics, Simulation, and Control

2. All perturbations may be specified either in the body or stability

axes, but the preferred axis system is the stability axis system.

3. The terms involving gravity must be treated with care in the stabil-

ity axes.

The effects of aerodynamic and thrust controls on the forces and moments

acting on the aircraft may now be included. To this end, we expressed the

forces and moments acting on the aircraft as

⎡ ⎤ ⎡ ⎤ ⎡ ⎤

⎢XW ⎥ ⎢XW ⎥ ⎢XW ⎥

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ YW ⎥ = ⎢ YW ⎥ + ⎢ YW ⎥ (5.64a)

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ ZW ⎥ ⎢ ZW ⎥ ⎢ ZW ⎥

⎣ ⎦ ⎣ ⎦ without controls ⎣ ⎦ controls

and

⎡ ⎤ ⎡ ⎤ ⎡ ⎤

⎢ LW ⎥ ⎢ LW ⎥ ⎢ LW ⎥

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ MW ⎥ = ⎢ MW ⎥ + ⎢ MW ⎥ (5.64b)

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ NW ⎥ ⎢ NW ⎥ ⎢ NW ⎥

⎣ ⎦ ⎣ ⎦ without controls ⎣ ⎦ controls

where

⎡ ⎤ ⎡

⎤ ⎡ ⎤

⎢XW ⎥ ⎢X u Xv Xw ⎥

u

⎡ s⎤ ⎢ ⎥

Δ

X w

⎢ ⎥ ⎢ ⎥

⎢ ⎥ ⎢ ⎥

⎢ YW ⎥ ≈ ⎢Yu Yv Y w ⎥ ⎢ Δvs ⎥ + ⎢ Y w ⎥ Δw s

⎢ ⎥ ⎢ ⎥ ⎢ Δw ⎥ ⎢ ⎥

⎢ ZW ⎥ ⎢ Zu Zv Z w ⎥ ⎣ s ⎦ ⎢ Z w ⎥

⎣ ⎦ without controls ⎣ ⎦ ⎣ ⎦

⎡

⎤ ⎡

⎤

⎢X p Xq Xr ⎥

⎡ Δps ⎤ ⎢

X p X q X r ⎥

⎡ Δp s ⎤

⎢ ⎥

⎢ ⎥ ⎢ ⎥

⎢ ⎥

+ ⎢Y p Yq Y r ⎥ ⎢ Δqs ⎥ + ⎢ Y p Y q Y r ⎥ ⎢ Δq s ⎥ (5.65a)

⎢ ⎥ ⎢ Δr ⎥ ⎢ ⎥ ⎢ Δr ⎥

⎢ Zp Zq Z r ⎥ ⎣ s ⎦ ⎢ Z p Z q Z r ⎥ ⎣ s ⎦

⎣ ⎦ ⎣ ⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 165

and

⎡ ⎤ ⎡

⎤ ⎡ ⎤

⎢ LW ⎥ ⎢ Lu Lv Lw ⎥

⎡ Δus ⎤ ⎢

Lw ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢ MW ⎥ ≈ ⎢ Mu Mv M w ⎥ ⎢ Δvs ⎥ + ⎢ M w ⎥ Δw s

⎢ ⎥ ⎢ ⎥ ⎢ Δw ⎥ ⎢ ⎥

⎢ NW ⎥ ⎢ Nu Nv N w ⎥ ⎣ s ⎦ ⎢ N w ⎥

⎣ ⎦ without controls ⎣ ⎦ ⎣ ⎦

⎡

⎤ ⎡

⎤

⎢ Lp Lq Lr ⎥

⎡ Δps ⎤ ⎢

L p Lq Lr ⎥

⎡ Δp s ⎤

⎢ ⎥ ⎢ ⎥

⎢ ⎥ ⎢ ⎥

+ ⎢Mp Mq M r ⎥ ⎢ Δqs ⎥ + ⎢ M p M q M r ⎥ ⎢ Δq s ⎥ .

⎢ ⎥ ⎢ Δr ⎥ ⎢ ⎥ ⎢ Δr ⎥

⎢ Np Nq N r ⎥ ⎣ s ⎦ ⎢ N p N q N r ⎥ ⎣ s ⎦

⎣ ⎦ ⎣ ⎦

(5.65b)

The additional effects of the controls may also be expanded in a similar man-

ner. Thus, the forces and moments due to the additional aerodynamic and

thrust controls may be expressed as

⎡ ⎤ ⎡

⎤

⎢XW ⎥ ⎢X η Xτ Xξ X ζ ⎥ ⎡ Δη⎤

⎢ ⎥

⎢ ⎥ ⎢ ⎥ Δτ

⎢ YW ⎥ ≈ ⎢Yη Yτ Yξ Y ζ ⎥ ⎢ ⎥ (5.66a)

⎢ Δξ ⎥

⎢ ⎥ ⎢ ⎥

⎢ ⎥

⎢ ZW ⎥ ⎢ Zη Zτ Zξ Z ζ ⎥ ⎣ Δζ ⎦

⎣ ⎦ controls ⎣ ⎦

and

⎡ ⎤ ⎡

⎤

⎢ LW ⎥ ⎢ Lη Lτ Lξ Lζ ⎥ ⎡ Δη⎤

⎢ ⎥

⎢ ⎥ ⎢ ⎥ Δτ

⎢ MW ⎥ ≈ ⎢ Mη Mτ Mξ M ζ ⎥ ⎢ ⎥ (5.66b)

⎢ Δξ ⎥

⎢ ⎥ ⎢ ⎥

⎢ ⎥

⎢ NW ⎥ ⎢ Nη Nτ Nξ N ζ ⎥ ⎣ Δζ ⎦

⎣ ⎦ controls ⎣ ⎦

where

Δη is the angular displacement of the elevator

Δτ is the incremental total engine thrust or equivalently the throttle

displacement

Δξ is the angular displacement of the ailerons which are assumed to be

symmetrically deployed

Δζ is the angular displacement of the rudder

www.ebook777.com

free ebooks ==> www.ebook777.com

166 Flight Dynamics, Simulation, and Control

controls and throttle included are

⎡ mΔu s ⎤ ⎡ cos α e sin α e ⎤ ⎡ ⎡ X NB ⎤ ⎡ X NBe ⎤ ⎤

⎢ ⎥=⎢ ⎢ − ⎥

(

⎢⎣ m Δw s − ΔqU e

s

) ⎥⎦ ⎣ − sin α e cos α e ⎥⎦ ⎢⎣ ⎢⎣ ZNB ⎥⎦ ⎢⎣ ZNBe ⎥⎦ ⎥⎦

⎡

⎤

⎢ Δus X u + Δws X w + Δw s X w + Δqs X q + Δq s X q + Δη X η + Δτ X τ ⎥

+

⎢ ⎥

⎢⎣ Δus Z u + Δws Z w + Δw s Z w + Δqs Z q + Δq s Z q + Δη Z η + Δτ Z τ ⎥⎦

+ mg ⎢ Δs (5.67a)

⎣ − sin α e cos α e ⎥⎦ ⎢⎣ sin e ⎥⎦

and

I yy

s

Δq s = MNB − MNBe

+ Δus M u + Δws M w + Δw s M w + Δqs M q + Δq s M q + Δη M η + Δτ M τ (5.67b)

with

Δθ s = Δqs , (5.68a)

The lateral equations of motion are

( )

m Δv s + ΔrsU es = cosβe ( YNB − YNBe )

+ Δvs Y v + Δps Y p + Δrs Y r + Δp s Y p + Δrs Y r + Δξ Y ξ + Δζ Y ζ

+ mg cos βe cos θe cosφe Δφ (5.69a)

⎡ I xx

s

−I xz

s

⎤ ⎡ Δp s ⎤ ⎡cos βe 0 ⎤ ⎡ cos α e sin α e ⎤ ⎡ ⎡ LNB ⎤ ⎡ LNBe ⎤ ⎤

⎢ s s ⎥⎢ ⎥

=⎢ ⎢ − ⎥

⎣ −I xz I zz ⎦ ⎣ Δrs ⎦ ⎣ 0 1 ⎥⎦ ⎢⎣ − sin α e cos α e ⎥⎦ ⎢⎣ ⎢⎣ N NB ⎥⎦ ⎢⎣ N NBe ⎥⎦ ⎥⎦

⎡ s L p + Δrs L r + Δξ Lξ + Δζ Lζ ⎤

Δ v s L v + Δps L p + Δrs L r + Δp

+⎢ ⎥

⎢ ⎥

⎢⎣ Δvs N v + Δps N p + Δrs N r + Δp s N p + Δrs N r + Δξ N ξ + Δζ N ζ ⎥⎦

(5.69b)

with

Δφ s = Δps , (5.70a)

Δψ s = Δrs . (5.70b)

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 167

In the case when the trim angles βe and α e are equal to zero or almost equal

to zero, the longitudinal equations of motion with the effects of aerodynamic

controls and throttle included reduce to

⎢ ⎥ = ⎢⎢ ⎥ − ⎢ Z ⎥ ⎥ + mg ⎢ sin θ ⎥ Δθs

(

⎢⎣ m Δw s − ΔqU e

s

) Z

⎥⎦ ⎢⎣ ⎣ NB ⎦ ⎣ NBe ⎦ ⎥⎦ ⎣ e ⎦

⎡ ⎤

⎢ Δus X u + Δws X w + Δw s X w + Δqs X q + Δq s X q + Δη X η + Δτ X τ ⎥

+

⎢ ⎥

⎢⎣ Δus Z u + Δws Z w + Δw s Z w + Δqs Z q + Δq s Z q + Δη Z η + Δτ Z τ ⎥⎦

(5.71)

I Δq s = MNB − MNBe

s

yy

+ Δus M u + Δws M w + Δw s M w + Δqs M q + Δq s M q + Δη M η + Δτ M τ (5.72)

with

Δh = U es ( Δθ − Δα ) . (5.73b)

( )

m Δv s + ΔrsU es = ( YNB − YNBe )

+ Δvs Y v + Δps Y p + Δrs Y r + Δp s Y p + Δrs Y r + Δξ Y ξ + Δζ Y ζ

⎡ I xx

s

−I xz

s

⎤ ⎡ Δp s ⎤ ⎡ ⎡ LNB ⎤ ⎡ LNBe ⎤ ⎤

⎢ s s ⎥⎢ ⎥

= ⎢⎢ ⎥−⎢ ⎥⎥

⎣ −I xz I zz ⎦ ⎣ Δrs ⎦ ⎢⎣ ⎣ N NB ⎦ ⎣ N NBe ⎦ ⎥⎦

⎡ ⎤

Δ v s L v + Δp s L p + Δrs L r + Δp s L p

+ Δrs L r + Δξ L ξ + Δζ L ζ

+⎢ ⎥

⎢ ⎥

⎢⎣ Δvs N v + Δps N p + Δrs N r + Δp s N p + Δrs N r + Δξ N ξ + Δζ N ζ ⎥⎦

(5.74b)

with

Δφ s = Δps , (5.75a)

www.ebook777.com

free ebooks ==> www.ebook777.com

168 Flight Dynamics, Simulation, and Control

s

, the longitudinal equations may be written as

⎢ Δw − ΔqU s ⎥ = ⎢ ⎢ Z ⎥ − ⎢ Z ⎥ ⎥ + g ⎢ sin θ ⎥ Δθs

⎣ s e⎦ m ⎢⎣ ⎣ NB ⎦ ⎣ NBe ⎦ ⎥⎦ ⎣ e ⎦

⎡ Δus Xu + Δws X w + Δw s X w + Δqs X q + Δq s X q + ΔηXη + ΔτX τ ⎤

+⎢ ⎥

⎣ Δus Zu + Δws Zw + Δw s Zw + Δqs Zq + Δq s Zq + ΔηZη + ΔτZτ ⎦

(5.76a)

1

Δq s = s (

MNB − MNBe ) + Δus Mu + Δws Mw + Δw s Mw + Δqs Mq + Δq s Mq

I yy

+ ΔηMη + ΔτMτ (5.76b)

with

Δθ s = Δqs , (5.77a)

Δh = U es ( Δθ − Δα ) , (5.77b)

where the various derivatives are referred to as the state-space derivatives.

Rearranging the equations, they may be written as

⎡1 −X w −X q 0 0 ⎤ ⎡ Δu s ⎤

⎢ ⎥

⎢0

⎢ 1 − Zw −Zq 0 0 ⎥⎥ ⎢ Δw s ⎥

⎢0 − Mw 1 − Mq 0 0 ⎥ ⎢ Δq s ⎥

⎢ ⎥⎢ ⎥

⎢0 0 0 1 0 ⎥ ⎢ Δθ s ⎥

⎢0

⎣ 0 0 0 1 ⎥⎦ ⎢⎢ Δh ⎥⎥

⎣ ⎦

⎡ Xu Xw Xq − g cos θe 0 ⎤ ⎡ Δus ⎤

⎢Z Zw Zq + U es − g sin θe 0 ⎥⎥ ⎢⎢ Δw

ws ⎥⎥

⎢ u

= ⎢ Mu Mw Mq 0 0 ⎥ ⎢ Δqs ⎥

⎢ ⎥⎢ ⎥

⎢ 0 0 1 0 0 ⎥ ⎢ Δθs ⎥

⎢ 0

⎣ −1 0 U es 0 ⎥⎦ ⎢⎣ Δh ⎥⎦

⎡ 1

⎢ X − X e ⎥⎤

⎡ Xη Xτ ⎤ ⎢ ⎥

⎢Z Zτ ⎥⎥ ⎢ 1 Z − Z e

⎥

⎢ η ⎡ Δη⎤ ⎢ ⎥

+ ⎢ Mη Mτ ⎥ ⎢ ⎥ + ⎢ 1 ⎥ (5.78)

0 ⎥ ⎣ ⎦ ⎢ s M − M

⎢ ⎥ Δτ ⎢ e

⎥

⎢ 0 ⎥

⎢ 0 0 ⎦⎥ ⎢ ⎥

⎣ ⎢ 0 ⎥

⎢⎣ 0 ⎥⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 169

write the equations as

⎡1 −X w −X q 0 ⎤ ⎡ Δu s ⎤ ⎡ Xu Xw Xq − g cos θe ⎤ ⎡ Δus ⎤

⎢0 1 − Zw −Zq 0 ⎥ ⎢ Δw s ⎥ ⎢ Zu Zw Zq + U es − g sin θe ⎥ ⎢ Δws ⎥

⎢ ⎥ ⎢ Δqq ⎥ = ⎢ ⎥⎢ ⎥

⎢0 − Mw 1 − Mq 0 ⎥ ⎢ s ⎥ ⎢ Mu Mw Mq 0 ⎥ ⎢ Δqs ⎥

⎣⎢0 0 0 1 ⎦⎥ ⎣⎢ Δθ s ⎦⎥ ⎢⎣ 0 0 1 0 ⎥⎦ ⎢⎣ Δθs ⎥⎦

⎡ 1 ⎤

⎢ (X − X e )

⎥

⎡ Xη Xτ ⎤ ⎢ 1 ⎥

⎢Z Zτ ⎥ ⎡ Δη⎤ ⎢ (Z − Z e) ⎥

+⎢ η ⎥ +

⎢ Mη Mτ ⎥ ⎢⎣ Δτ ⎥⎦ ⎢⎢ 1 ⎥

⎥

⎢⎣ 0 0 ⎥⎦ ⎢ s (M − M e )⎥

⎢ ⎥

⎢⎣ 0 ⎥⎦

(5.79a)

In addition, the equation for the height is

In most real situations, the derivatives X q , Zq and Mq are small and negligible.

Hence, we multiply the previous set of coupled equations by M−1 where

⎡1 0 0 0⎤

⎢0 1 0 0⎥

M=⎢ ⎥ , (5.80)

⎢0 − Mw 1 0⎥

⎣⎢ 0 0 0 1 ⎥⎦

and the resulting equations are

⎡1 0 0 −X q ⎤ ⎡ Δu ⎤

⎢0 1 0 0

⎢ ⎥ ⎢ Δθ ⎥

⎥⎢ ⎥

⎢0 0 1 −Zq ⎥ ⎢ Δw ⎥

⎢⎣0 0 0 1 − Mq − Mw Zq ⎥⎦ ⎢⎣ q B ⎥⎦

⎡ Xu − g cos θ0 Xw 0⎤ ⎡ Δu ⎤

⎢ 0 0 0 1⎥⎢ ⎥

=⎢ ⎥ ⎢ Δθ ⎥

⎢ Zu − g sin θ0 Zw U e + Zq

s

⎥ ⎢ Δw ⎥

⎢ Mu + Mw Zu − Mw g sin θ0 Mw + Mw Zw Mq + Mw U e + Zq ⎥ ⎢⎣ qB ⎥⎦

s

( )

⎣ ⎦

⎡ 1 ⎤

⎢ ( X B − X Be ) ⎥

⎡ Xη Xτ ⎤ ⎢ 1 ⎥

⎢ 0 0 ⎥ ⎡ Δη⎤ ⎢ ( Z B − Z Be ) ⎥

⎥⎢ ⎥+M ⎢

−1 (5.81)

+⎢ ⎥

⎢ Z η Zτ ⎥ ⎣ Δτ ⎦ ⎢ 1 ⎥

M

⎢⎣ η + M w Zη Mτ + Mw Zτ ⎥⎦ ⎢ s ( M B − M Be ) ⎥

⎢ ⎥

⎢⎣ 0 ⎦⎥

www.ebook777.com

free ebooks ==> www.ebook777.com

170 Flight Dynamics, Simulation, and Control

been called the concise form by Cook [1] and are given by

⎢0 ⎢ ⎥

⎢ 1 − zq 0 ⎥⎥ ⎢ Δw s ⎥ ⎢⎢ zu zw zq zθ ⎥⎥ ⎢⎢ Δws ⎥⎥

=

⎢0 0 1 − mq 0 ⎥ ⎢ Δq s ⎥ ⎢ mu mw mq mθ ⎥ ⎢ Δqs ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣0 0 0 1 ⎦ ⎣⎢ Δθ s ⎦⎥ ⎣ 0 0 1 0 ⎦ ⎣ Δθs ⎦

⎡ 1 ⎤

⎢ ( − ) ⎥

m

⎡ xη xτ ⎤ ⎢ ⎥

⎢z 1

zτ ⎥⎥ ⎡ Δη⎤ ⎢ ( − ) ⎥⎥

M −1 ⎢

m

η

+⎢ +

⎢ mη mτ ⎥ ⎢⎣ Δτ ⎥⎦ ⎢ 1 ⎥

⎢ ⎥ ⎢ ( − )⎥

⎣0 0⎦ ⎢

s

⎥

⎢ 0 ⎥

⎣ ⎦

(5.82)

1

Δv s + ΔrsU es = ( YNB − YNBe )

m

+ ΔvsYv + ΔpsYp + ΔrsYr + Δp sYp + ΔrsYr + ΔξYξ + ΔζYζ

⎡ I xz

s

⎤

⎢ 1 − s ⎥

I xx ⎥ ⎡ Δp s ⎤ ⎡ I xx ⎤ ⎡ ⎡ LNB ⎤ ⎡ LNBe ⎤ ⎤

−1

s

⎢ =⎢ ⎥−⎢

⎢ ⎥ s ⎥ ⎢⎢ ⎥⎥

⎢ I xz

s ⎥ Δrs ⎦ ⎣ I zz ⎦ ⎢⎣ ⎣ N NB ⎦ ⎣ N NBe ⎦ ⎥⎦

⎢− s 1 ⎥⎣

⎣ I zz ⎦

⎡ Δvs Lv + Δps Lp + Δrs Lr + Δp s Lp + Δrs Lr + Δξ Lξ + Δζ Lζ ⎤

+⎢ ⎥ (5.83b)

⎣ Δvs N v + Δps N p + Δrs N r + Δp s N p + Δrs N r + ΔξN ξ + ΔζN ζ ⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 171

with

where again the various derivatives are referred to as the state-space derivatives.

Rearranging the equations, they may be written as

⎡1 −Yp −Yr 0 0⎤

⎢ ⎥ ⎡ Δv s ⎤

I s

xz

⎢0 1 − Lp − − Lr 0 0⎥ ⎢ ⎥

⎢ I s

xx ⎥ ⎢ Δps ⎥

⎢ ⎥ ⎢ Δr ⎥

⎢0 I xz

s

s

− − N p 1 − N r 0 0⎥ ⎢ ⎥

⎢ I zz

s ⎥ ⎢ Δφs ⎥

⎢ ⎥

⎢0 0 0 1 0 ⎥ ⎢⎣ Δψ s ⎦⎥

⎢0 0 0 0 1 ⎥⎦

⎣

⎢ ⎥⎢ ⎥ ⎢

⎢ Lv Lp Lr 0 0 ⎥ ⎢ Δps ⎥ ⎢ L L ⎥⎥

⎡ Δ⎤

= ⎢ Nv Np Nr 0 0 ⎥ ⎢ Δrs ⎥ + ⎢ N N ⎥ ⎢ ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥ Δ

⎢0 1 0 0 0 ⎥ ⎢ Δφs ⎥ ⎢ 0 0 ⎥⎣ ⎦

⎢0

⎣ 0 1 0 0 ⎥⎦ ⎢⎣ Δψ s ⎥⎦ ⎢⎣ 0 0 ⎥⎦

⎡ 1

⎢ YN − YN e ⎥⎤

⎢ ⎥

⎢ 1 L − L ⎥

N N e

⎢ I xxs ⎥

+⎢ ⎥ (5.85)

⎢ 1 N N − N N e ⎥

⎢ I zz

s ⎥

⎢ ⎥

⎢ 0 ⎥

⎢⎣ 0 ⎥⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

172 Flight Dynamics, Simulation, and Control

independent of the first four. Hence, the aforementioned set may be written as

⎡1 −Yp −Yr 0⎤

⎢ ⎥ ⎡ Δv s ⎤

I s

xz

⎢0 1 − Lp − − Lr 0⎥ ⎢ ⎥

⎢ I s

xx ⎥ ⎢ Δps ⎥

⎢ ⎥ ⎢ Δr ⎥

⎢0 I xz

s

s

− − N p 1 − N r 0⎥ ⎢ ⎥

⎢ I zz

s ⎥ ⎢ Δφs ⎥

⎢ ⎥⎣ ⎦

⎣⎢0 0 0 1 ⎥⎦

⎢ ⎥⎢ ⎥ ⎢

Lv Lp Lr 0 ⎥ ⎢ Δps ⎥ + ⎢ L L ⎥⎥ ⎡ Δ⎤

=⎢

⎢ Nv Np Nr 0 ⎥ ⎢ Δrs ⎥ ⎢ N N ⎥ ⎢⎣ Δ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢⎣ 0 1 0 0 ⎥⎦ ⎣ Δφs ⎦ ⎣ 0 0 ⎦

⎡ 1 ⎤

⎢ m YNB − YNBe ⎥

⎢ ⎥

⎢ 1 L − L ⎥

NB NBe

+ ⎢ I xx

s ⎥ (5.86a)

⎢ ⎥

⎢ 1 N NB − N NBe

⎥

⎢ I zz

s ⎥

⎢ ⎥

⎣ 0 ⎦

and

Yq , Lq , N q , Yr , Lr and N r are small and negligible. Hence, we multiply the

aforementioned set of coupled equations by M−1 where

⎡1 0 0 0⎤

⎢ ⎥

⎢0 I s

xz

1 − 0⎥

⎢ I s

xx ⎥

M=⎢ ⎥ , (5.87)

⎢0 I xz

s

− 1 0⎥

⎢ I zz

s ⎥

⎢ ⎥

⎣0 0 0 1⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 173

and write the set of first-order equations in the so-called concise form

given by

⎡1 − y p − y r 0 ⎤ ⎡ Δv s ⎤

⎢0 ⎢ ⎥

⎢ 1 − lp −lr 0 ⎥⎥ ⎢ Δp s ⎥

⎢0 −np 1 − nr 0 ⎥ ⎢ Δrrs ⎥

⎢ ⎥⎢ ⎥

⎣0 0 0 1 ⎦ ⎣⎢ Δφ s ⎦⎥

⎡ 1 ⎤

⎢ ( − ) ⎥

⎡ yv yp yr yφ ⎤ ⎡ Δvs ⎤ ⎡ yξ yζ ⎤ ⎢ ⎥

⎢l lp lr lφ ⎥⎥ ⎢⎢ Δps ⎥⎥ ⎢⎢ lξ lζ ⎥⎥ ⎡ Δξ ⎤ ⎢ 1 ( − ) ⎥⎥

v −1 ⎢ s

=⎢ + + M

⎢ nv np nr nφ ⎥ ⎢ Δrs ⎥ ⎢ nξ nζ ⎥ ⎢⎣ Δζ ⎥⎦ ⎢ ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ 1 ( )⎥⎥

⎣0 1 0 0 ⎦ ⎣ Δφs ⎦ ⎣ 0 0⎦ −

⎢ s

⎢ 0 ⎥

⎣ ⎦

(5.88)

The concise equations are the standard decoupled linear longitudinal and

lateral equations of motion often quoted in most textbooks on flight

dynamics.

It is customary not to use the superscript or subscript ‘s’ as these equations

are normally expressed in the body-fixed stability axes. Also the perturba-

tion quantities are all denoted without explicitly using the symbol ‘Δ’.

In many applications, either the state-space derivatives or the concise deriva-

tives in the left-hand side of the equations are zero, and in these cases, these

equations are relatively easier to analyse. The taxonomy (classification) of

stability derivatives is illustrated in Figure 5.1. The aerodynamic coefficients

and non-dimensional stability derivatives are primarily defined to facilitate

the estimation and measurement of the relevant aerodynamic parameters

essential for the calculation of the stability derivatives.

On the other hand, the state-space and concise derivatives facilitate the

determination of the natural frequencies and damping ratios of the air-

craft’s modes of oscillation. The examples in Chapter 6 illustrate the appli-

cation of these various groups of stability derivatives. The transformation of

www.ebook777.com

free ebooks ==> www.ebook777.com

174 Flight Dynamics, Simulation, and Control

Stability derivatives

dimensional coefficients

Reduce to

state-space form

State-space

derivatives

Divide by

‘inertial coupling’ matrix

Concise

derivatives

(real time and distance)

FIGURE 5.1

Taxonomy of stability derivatives.

in the following.

Consider the longitudinal and lateral small perturbation equations of

motion in dimensional form in terms of the dimensional stability derivatives,

X u , X w , Z u , Z w , Z q , M u , M w , M q , M w etc.:

⎡ ⎤

⎢m 0 0 − X q ⎥ ⎡ Δu ⎤

⎢0 1 0 0 ⎥ ⎢ Δθ ⎥

⎢ ⎥ d ⎢ ⎥

− Z q ⎥ dt ⎢ Δw ⎥

⎢0 0 m

⎢ ⎥ ⎢ ⎥

⎢

⎥ ⎣ qB ⎦

⎣0 0 − M w I yy − M q ⎦

⎡

⎤ ⎡

⎤

⎢ Xu −mg Xw 0 ⎥ ⎡ ⎤ ⎢ Xη

Δu Xτ ⎥

⎢ 0 0 0 1 ⎥ ⎢ Δθ ⎥ ⎢ 0 0 ⎥ ⎡ Δη⎤

⎥⎢ ⎥+⎢

=⎢ ⎢ ⎥

⎥ (5.89a)

mU e + Z q ⎥ ⎢ Δw ⎥ ⎢ Z η

⎢ Zu 0 Zw s

Z τ ⎥ ⎣ Δτ ⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢

⎥ ⎣ qB ⎦ ⎢

⎥

⎣ Mu 0 Mw Mq ⎦ ⎣ Mη Mτ ⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 175

and

⎡ ⎤

⎢m − Y p − Y r 0 ⎥ ⎡ Δv ⎤

⎢ ⎥ ⎢ ⎥

0 ⎥ d ⎢ Δp ⎥

⎢0 I xx − L p −I xz − Lr

⎢ ⎥ dt ⎢ Δr ⎥

⎢0 −I xz − N p I zz − N r 0⎥ ⎢ ⎥

⎢ ⎥ ⎣ Δφ ⎦

⎣0 0 0 1⎦

⎡

⎤ ⎡

⎤

⎢ Yv Yp Y r − mU e mg ⎥ ⎡ Δv ⎤ ⎢ Y ξ Yζ ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥

0 ⎥ ⎢ Δp ⎥ + ⎢ Lξ

= ⎢ Lv Lp Lr Lζ ⎥ ⎡ Δξ ⎤

⎢ ⎥ (5.89b)

⎢ ⎥ ⎢ Δr ⎥ ⎢ ⎥ ⎣ Δζ ⎦

⎢Nv Np Nr 0 ⎥ ⎢ ⎥ ⎢Nξ Nζ ⎥

⎢ ⎥ ⎣ Δφ ⎦ ⎢ ⎥

⎣ 0 1 0 0 ⎦ ⎣ 0 0 ⎦

One may define the standard non-dimensional parameters, τ, μ1, μ2, ix, iy, iz,

and explain their physical significance. Hence, or otherwise, the relationship

between the dimensional states, Δu, Δw, Δv, Δp, qB, Δr, and the dimensionless

states, ∆ u, ∆w, ∆v, ∆ p, qB , ∆ r , may be established.

To non-dimensionalise the longitudinal equations, let τ = t t̂ ,

m mU es CwU es 1 2

tˆ = = = ( )

, qs = ρ U es Sw . (5.90)

1 qs g 2

ρU esSw

2

∂ 1 ∂

= . (5.91)

∂t tˆ ∂τ

Then let

Δu Δw

Δu = , Δw = s (5.92)

U es Ue

and

c ∂ c ∂ 1 ∂ tˆ

qB = θ= s θ= θ= qB , (5.93)

U e ∂t

s

U e tˆ ∂τ μ1 ∂τ μ1

www.ebook777.com

free ebooks ==> www.ebook777.com

176 Flight Dynamics, Simulation, and Control

where

U estˆ (5.94)

μ1 = .

c

less quantities,

U estˆ I I Δv

μ2 = , ix = xx2 , iz = zz2 , Δv = s , (5.95)

b mb mb Ue

b ∂ b ∂ 1 ∂ tˆ

Δp = pB = φ= s φ= φ= pB (5.96)

U e ∂t

s ˆ

U e t ∂τ μ 2 ∂τ μ2

and

b ∂ b ∂ 1 ∂ tˆ

Δr = rB = ψ= s ψ= ψ= rB . (5.97)

U e ∂t

s

U e tˆ ∂τ μ 2 ∂τ μ2

Note that t̂ is an aerodynamic time parameter and is the ratio of the inertial

momentum to the aerodynamic force, while the scaling factor μ1 is the ratio of

the aerodynamic distance to the geometric distance (length scale) in the chordwise

direction. It may also be interpreted as the ratio of the kinetic energy to the

pressure energy. The scaling factor μ2 is the ratio of the aerodynamic distance

to the geometric distance in the spanwise direction. The ratios

I yy I I zz

iy = , ix = xx2 and iz = (5.98)

mc 2 mb mb 2

Hence, one may reduce the longitudinal and lateral perturbation equations

to the standard dimensionless form as follows:

⎡μ1 0 0 −X q ⎤ ⎡ Δu ⎤

⎢ ⎥ ⎢ Δθ ⎥

1 ⎢0 μ1 0 0 ⎥ ∂ ⎢ θ⎥

μ1 ⎢ 0 0 μ1 −Zq ⎥ ∂τ ⎢ Δw ⎥

⎢ ⎥ ⎢ ⎥

⎢⎣ 0 0 − Mw μ1iy − Mq ⎥⎦ ⎣ qB ⎦

⎡ Xu −Cw Xw 0 ⎤ ⎡ Δu ⎤ ⎡ Xη Xτ ⎤

⎢ 0 0 0 μ1 ⎥⎥ ⎢⎢ Δθ ⎥⎥ ⎢⎢ 0 0 ⎥⎥ ⎡ Δη⎤

=⎢ + (5.99a)

⎢ Zu 0 Zw μ1 + Zq ⎥ ⎢ Δw ⎥ ⎢ Zη Zτ ⎥ ⎢⎣ Δτ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢⎣ Mu 0 Mw Mq ⎥⎦ ⎣ qB ⎦ ⎣ Mη Mτ ⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 177

and

⎡μ 2 −Yp −Yr 0⎤

⎢ I xz ⎥ ⎡ Δv ⎤

⎢0 μ 2ix − Lp −μ 2ix − Lr 0⎥ ⎢ ⎥

1 ⎢ I xx ⎥ ∂ ⎢ Δp ⎥

μ2 ⎢ I xz ⎥ ∂τ ⎢ Δr ⎥

⎢0 −μ 2iz − N p μ 2iz − N r 0⎥ ⎢ ⎥

⎢ I zz ⎥ ⎣ Δφ ⎦

⎢0 0 0 μ 2 ⎥⎦

⎣

⎡ Yv Yp Yr − μ 2 Cw ⎤ ⎡ Δv ⎤ ⎡ Y Y ⎤

⎢L

v Lp Lr 0 ⎥⎥ ⎢⎢ Δp ⎥⎥ ⎢⎢ L L ⎥⎥ ⎡ Δ ⎤

=⎢ + (5.99b)

⎢ Nv Np Nr 0 ⎥ ⎢ Δr ⎥ ⎢ N N ⎥ ⎢⎣ Δ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣0 μ2 0 0 ⎦ ⎣ Δφ ⎦ ⎣ 0 0 ⎦

Δu, Δw, qB and t, we obtain

⎡ ⎤

− X q ⎥ ⎡ U e Δu u⎤

s

⎢m 0 0

⎢ ⎥

⎢0 1 0 0 ⎥ ⎢ Δθ ⎥

⎢ ∂

⎥ ⎢ s ⎥

⎢0

U e Δw

0 m − Z q ⎥ ∂τ ⎢ ⎥

⎢ ⎥ ⎢ μ1 ⎥

q

I yy − M q ⎦ ⎢⎣ tˆ

⎢ ⎥ B

⎣0 0 − M w ⎥⎦

⎡

⎤ ⎡ U s Δu ⎤ ⎡

⎤

⎢ Xu −mg Xw 0 ⎥ ⎢ e ⎥ ⎢ X Xτ ⎥

⎢ 0 0 0 1 ⎥ ⎢ Δθ ⎥ ⎢ 0 0 ⎥ ⎡ Δ⎤

= tˆ ⎢

⎥ ⎢ s ⎥ + t̂ ⎢

U e Δw

⎥

⎢ ⎥

⎢ Zu 0 Zw mU e + Z q ⎥ ⎢

s

⎥ ⎢ Z Z τ ⎥ ⎣ Δτ ⎦

⎢ ⎥ ⎢ μ1 ⎥ ⎢ ⎥

qB

M q ⎦ ⎢⎣ tˆ

⎢

⎥

⎥⎦ ⎢ M

⎥

⎣ Mu 0 Mw ⎣ Mτ ⎦

(5.100)

Substituting for

m mU es CwU es μ1 U es

tˆ = = = , = , (5.101)

1

ρU esSw qs g tˆ c

2

www.ebook777.com

free ebooks ==> www.ebook777.com

178 Flight Dynamics, Simulation, and Control

− X q ⎥ ⎡ U e Δu ⎤

⎡ ⎤ s

⎢m 0 0

⎢ ⎥

mU e ⎥ ⎢ Δθ ⎥

1 ⎢0 0 0

⎢ s

∂

⎥ ⎢ s ⎥

U e Δw

mU es ⎢ 0

0 m − Z q ⎥ ∂τ ⎢ s ⎥

⎢ ⎥ ⎢ Ue ⎥

⎢

⎥ qB ⎥

⎣0 0 − M w I yy − M q ⎦ ⎢⎣ c ⎦

⎡

⎤ ⎡ U es Δu ⎤

X mg X 0 ⎡

⎤

⎢ u − w

⎥⎢ ⎥ ⎢ X Xτ ⎥

⎢ 0 0 0 mU es ⎥ ⎢ Δθ ⎥ 1 ⎢ 0 0 ⎥ ⎡ Δ⎤

1

= ⎢ ⎥ ⎢U s Δw ⎥ + ⎢

⎥ ⎢ ⎥ (5.102)

qs ⎢ Z u mU e + Z q ⎥ ⎢ s ⎥ qs

e

0 Zw s ⎢ Z Z τ ⎥ ⎣ Δτ ⎦

⎢ ⎥ ⎢ Ue ⎥ ⎢ ⎥

⎢

⎥ qB ⎥ ⎢ M

⎣ Mu 0 Mw M q ⎦ ⎢⎣ c ⎦ ⎣ M τ ⎥⎦

Introducing the dimensionless derivatives and dividing the last equation by

c , we obtain

⎡μ1 0 0 −X q ⎤ ⎡ Δu ⎤

⎢ ⎥ ⎢ Δθ ⎥

1 ⎢0 μ1 0 0

⎥ ∂ ⎢ ⎥

μ1 ⎢ 0 0 μ1 −Zq ⎥ ∂τ ⎢ Δww⎥

⎢0 0 − Mw ⎥

μ1iy − Mq ⎦ ⎣ qB ⎥⎦

⎢

⎣

⎡ Xu −Cw Xw 0 ⎤ ⎡ Δu ⎤ ⎡ Xη Xτ ⎤

⎢ 0 0 0 μ1 ⎥ ⎢ Δθ ⎥ ⎢ 0 0 ⎥ ⎡ Δη⎤

=⎢ ⎥⎢ ⎥+⎢ ⎥ (5.103)

⎢ Zu 0 Zw μ1 + Zq ⎥ ⎢ Δw ⎥ ⎢ Zη Zτ ⎥ ⎢⎣ Δτ ⎥⎦

⎢M 0 Mw Mq ⎥⎦ ⎢⎣ qB ⎥⎦ ⎢⎣ Mη Mτ ⎥⎦

⎣ u

where

I yy

iy = . (5.104)

mc 2

Following the similar steps as the longitudinal case, we obtain the dimension-

less equations as

⎡μ 2 −Yp −Yr 0⎤

⎢ I ⎥ ⎡ Δv ⎤

⎢0 μ 2ix − Lp −μ 2ix xz − Lr 0⎥ ⎢ ⎥

1 ⎢ I xx ⎥ ∂ ⎢ Δp ⎥

μ2 ⎢ I xz ⎥ ∂τ Δr

⎢0 −μ 2iz − N p μ 2iz − N r 0 ⎥ ⎢⎢ ⎥⎥

⎢ I zz ⎥ ⎣ Δφ ⎦

⎢⎣ 0 0 0 μ 2 ⎥⎦

⎡ Yv Yp Yr − μ 2 Cw ⎤ ⎡ Δv ⎤ ⎡ Y Y ⎤

⎢L Lp Lr 0 ⎥ ⎢ Δp ⎥ ⎢ L L ⎥ ⎡ Δ ⎤

v

=⎢ ⎥⎢ ⎥+⎢ ⎥ (5.105)

⎢ Nv Np Nr 0 ⎥ ⎢ Δr ⎥ ⎢ N N ⎥ ⎢⎣ Δ ⎥⎦

⎢0 μ2 0 0 ⎥⎦ ⎢⎣ Δφ ⎦⎥ ⎢⎣ 0 0 ⎥⎦

⎣

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 179

erivatives, X u , X w , Z u , Z w , Z q , M u , M w , M q and M w , and the corresponding

d

dimensionless stability derivatives, Xu, Xw, Zu, Zw, Zq, Mu, Mw, Mq and Mw , may

now be obtained.

Comparing the left-hand sides of the last two sets of the longitudinal equa-

tions term by term,

M w U es Mw M w M w m M w

= , Mw = μ1 = = , (5.106)

mU e c

s

μ1 mc 1

mc ρS c 1

W ρSW c 2

2 2

2

Mw =

( )

M w U es

=

M w

. (5.107)

qSw c 2 1

ρSw c 2

2

Comparing the right-hand sides of the last two sets of the longitudinal equa-

tions term by term,

X u U es X w U es Z u U es Z w U es Z q U es

Xu = ; Xw = ; Zu = ; Zw = ; Zq = ;

qSw qSw qSw qSw qSw c

(5.108)

M u U es M w U es M q U es

Mu = ; Mw = ; Mq = . (5.109)

qSw c qSw c qSw c 2

These are the desired relationships between the nine dimensional stability

derivatives, X u , X w , Z u , Z w , Z q , M u , M w , M q and M w , and the corresponding

dimensionless stability derivatives, Xu, Xw, Zu, Zw, Zq, Mu, Mw, Mq and Mw .

Longitudinal and Lateral Dynamics

The longitudinal dynamics of aircraft motion is governed by the system of

equations describing the evolution of the forward and normal velocity per-

turbations, the pitch attitude of the aircraft and the pitch rate. These equations

www.ebook777.com

free ebooks ==> www.ebook777.com

180 Flight Dynamics, Simulation, and Control

cruise velocity and a set of dimensional stability derivatives. In the process

of analysing the governing equations, they are usually recast in state-space

form. The translation equations of motion are normalised by the aircraft’s

mass and the equation governing the pitch dynamics by the moment of iner-

tia of the aircraft about the pitch axis. As a result of these operations, the

dimensional stability derivatives are modified. These new stability derivatives

are quasi-dimensional and are referred to as the state-space derivatives as they

are dimensionally equivalent to the corresponding coefficients of the state-

space equations. The longitudinal equations, referred to body-fixed axes, in

state-space form in terms of state-space derivatives are

⎡1 0 0 −X q ⎤ ⎡ Δu ⎤

⎢0 ⎢ ⎥

⎢ 1 0 0 ⎥⎥ ⎢ Δθ ⎥

⎢0 0 1 −Zq ⎥ ⎢ Δw ⎥

⎢ ⎥⎢ ⎥

⎢⎣0 0 − Mw 1 − Mq ⎥⎦ ⎢⎣ q B ⎥⎦

⎡ Xu −g Xw 0 ⎤ ⎡ Δu ⎤ ⎡ Xη Xτ ⎤

⎢ 0 0 0 1 ⎥⎢ θ⎥ ⎢ 0 0 ⎥⎥ ⎡ Δη⎤

=⎢ ⎥⎢Δ ⎥+ ⎢ (5.110)

⎢ Zu 0 Zw U e + Zq ⎥ ⎢ Δw ⎥ ⎢ Zη

s

Zτ ⎥ ⎢⎣ Δτ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢⎣ Mu 0 Mw Mq ⎥⎦ ⎣ qB ⎦ ⎣ Mη Mτ ⎦

where

Δη is the elevator angular deflection from the trimmed state

Δτ is the thrust perturbation beyond the trimmed value of thrust

⎡1 −Yp −Yr 0⎤

⎢ I xz ⎥ ⎡ Δv ⎤

⎢0 1 − Lp − − Lr 0⎥ ⎢ ⎥

⎢ I xx ⎥ d ⎢ Δp ⎥

⎢ I xz ⎥ dt ⎢ Δr ⎥

⎢0 − − N p 1 − N r 0⎥ ⎢ ⎥

⎢ I zz ⎥ ⎣ Δφ ⎦

⎣0 0 0 1 ⎥⎦

⎢

⎡ Yv Yp Yr − U e g ⎤ ⎡ Δv ⎤ ⎡ Yξ Yζ ⎤

⎢L Lp Lr 0 ⎥⎥ ⎢⎢ Δp ⎥⎥ ⎢⎢ Lξ Lζ ⎥⎥ ⎡ Δξ ⎤

v

=⎢ + . (5.111)

⎢ Nv Np Nr 0 ⎥ ⎢ Δr ⎥ ⎢ N ξ N ζ ⎥ ⎢⎣ Δζ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣0 1 0 0 ⎦ ⎣ Δφ ⎦ ⎣ 0 0 ⎦

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 181

Longitudinal and Lateral Dynamics

Ignoring the q derivatives in the left-hand side, longitudinal equations and

all the rate derivatives in the left-hand side of the lateral equations reduces

the coefficient matrices on the left-hand side of these two sets of the equa-

tions to Equations 5.80 and 5.87, respectively.

Multiplying the two sets of equations by the inverses of these two matri-

ces, respectively, results in the concise equations of motion. The longitudinal

equations, referred to body-fixed axes, in concise form are

⎡1 0 − xq 0 ⎤ ⎡ u ⎤ ⎡ xu xw xq xθ ⎤ ⎡ u ⎤ ⎡ xη xτ ⎤

⎢0 1 − zq 0 ⎥⎥ d ⎢⎢ w ⎥⎥ ⎢⎢ zu zw zq zθ ⎥⎥ ⎢⎢ w ⎥⎥ ⎢⎢ zη zτ ⎥⎥ ⎡η⎤

⎢ = +

⎢0

⎢ 0 1 − mq 0 ⎥ dt ⎢ q ⎥ ⎢ mu mw mq mθ ⎥ ⎢ q ⎥ ⎢ mη mτ ⎥ ⎢⎣ τ ⎥⎦

⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

0 0 0 1⎦ ⎣ θ ⎦ ⎣ 0 0 1 0 ⎦⎣θ⎦ ⎣ 0 0⎦

⎣

(5.112)

where

η is the elevator angular deflection from the trimmed state

τ is the thrust perturbation beyond the trimmed value of thrust

⎡1 − y p − y r 0 0⎤ ⎡ v ⎤

⎢0 1 − lp −lr 0 0 ⎥⎥ ⎢⎢ p ⎥⎥

⎢ d

⎢0 −np 1 − nr 0 0⎥ ⎢ r ⎥

⎢ ⎥ dt ⎢ ⎥

⎢0 0 0 1 0⎥ ⎢ φ ⎥

⎢0 0 0 0 1 ⎥⎦ ⎢⎣ψ ⎥⎦

⎣

⎡ yv yp yr yφ yψ ⎤ ⎡ v ⎤

⎢l ⎡ yξ yζ ⎤

⎢ v lp lr 0 0 ⎥⎥ ⎢⎢ p ⎥⎥ ⎢

lξ lζ ⎥⎥ ⎡ ξ ⎤

= ⎢ nv np nr 0 0 ⎥⎢r⎥+⎢ (5.113)

⎢ ⎥⎢ ⎥ ⎢ n nζ ⎥ ⎢⎣ζ ⎥⎦

⎢0 1 0 0 0 ⎥ ⎢φ⎥ ⎢

ξ

⎥

0 0⎦

⎢0

⎣ 0 1 0 0 ⎥⎦ ⎢⎣ψ ⎥⎦ ⎣

where ξ and ζ are the aileron and rudder angles, respectively. No appar-

ent distinction is made between the perturbation and original quantities in

these equations.

www.ebook777.com

free ebooks ==> www.ebook777.com

182 Flight Dynamics, Simulation, and Control

Chapter Highlights

• Small perturbation equations of motion

The aircraft equations of motion can be partitioned into two decou-

pled sets usually referred to as the longitudinal equations of motion

for the symmetric modes of motion of the aircraft and the lateral equa-

tions of motion for the asymmetric modes of motion of the aircraft.

The equations may be expressed in a concise form as a set of first-

order equations in state-space notation.

• Control about pitch, roll and yaw axes

For longitudinal control (pitch axis), the elevators are employed:

∂C ∂CM

Control effectiveness derivatives: CLη ≡ L , C Mη ≡ .

∂η ∂η

Linear equations: CL = CL0 + CL αα + CL ηη, C M = C M 0 + CM αα + C M ηη.

(CL refers to the lift coefficient for longitudinal motions.)

For roll control (roll axis), the ailerons are employed:

∂C

Control effectiveness derivatives: CLξ ≡ L .

∂ξ

Linear equations: CL = CLββ + CLξξ.

(CL refers to the roll coefficient for lateral motions.)

For directional control (yaw axis), the rudder is employed:

∂C ∂C

Control effectiveness derivatives: CYζ ≡ Y , Cnζ ≡ n .

∂ζ ∂ζ

Linear equations: CY = CYββ + CYζζ, Cn = Cnββ + Cnζζ.

Exercises

5.1 Consider the relationship between moments of inertia in the stability

axes to the moments of inertia in the body principal axes. Derive explicit

expressions for the moments of inertia in the stability axes in terms of

moments of inertia in the body principal axes, the equilibrium angle of

attack and the equilibrium sideslip angle.

5.2 Consider the derivation of the non-dimensional lateral stability deriva-

tives in Section 5.7. Obtain explicit expressions for the non-dimensional

lateral stability derivatives, Yv, Yp, Yr, Lv, Lp, Lr, Nv, Np, Nr, in terms of the

corresponding dimensional lateral stability derivatives and other rel-

evant parameters.

free ebooks ==> www.ebook777.com

Small Perturbations and the Linearised, Decoupled Equations of Motion 183

terms of the state-space derivatives, are

⎡1 0 0 −X q ⎤ ⎡ Δu ⎤

⎢0 ⎢ ⎥

⎢ 1 0 0 ⎥⎥ ⎢ Δθ ⎥

⎢0 0 1 −Zq ⎥ ⎢ Δw ⎥

⎢ ⎥⎢ ⎥

⎢⎣0 0 − Mw 1 − Mq ⎥⎦ ⎢⎣ q B ⎥⎦

⎡ Xu −g Xw 0 ⎤ ⎡ Δu ⎤ ⎡ Xη Xτ ⎤

⎢ 0 0 0 1 ⎥⎢ ⎥ ⎢ 0 0 ⎥⎥ ⎡ Δη⎤

=⎢ ⎥ ⎢ Δθ ⎥ + ⎢ .

⎢ Zu 0 Zw U e + Zq ⎥ ⎢ Δw ⎥ ⎢ Zη

s

Zτ ⎥ ⎢⎣ Δτ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢⎣ Mu 0 Mw Mq ⎥⎦ ⎣ qB ⎦ ⎣ Mη Mτ ⎦

⎡ Δu ⎤ ⎡ xu xθ xw x q ⎤ ⎡ Δu ⎤ ⎡ x η xτ ⎤

⎢ ⎥ ⎢

⎢ Δθ ⎥ = ⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθ ⎥⎥ ⎢⎢ 0 0 ⎥⎥ ⎡ Δη⎤

+ ⎢ ⎥.

⎢ Δw ⎥ ⎢ zu zθ zw z q ⎥ ⎢ Δw ⎥ ⎢ z η zτ ⎥ ⎢ Δτ ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎣ ⎦

⎢⎣ q B ⎥⎦ ⎢⎣ mu mθ mw mq ⎥⎦ ⎣ qB ⎦ ⎣ mη mτ ⎦

derivatives.

5.4 The perturbation equations of lateral motion of an aircraft are

⎡1 0 0 0 ⎤ ⎡ Δv ⎤

⎢ I ⎥⎢ s⎥

⎢0 1 − xz 0 ⎥ ⎢ Δp s ⎥

⎢ I xx ⎥⎢ ⎥

⎢ I xz ⎥⎢ ⎥

⎢0 − 1 Δ

0⎥ ⎢ s ⎥r

⎢ I zz ⎥⎢

⎥

⎢0

⎣ 0 0 1 ⎥⎦ ⎣ Δφs ⎦

⎡ Yv Yp Yr − U e g ⎤ ⎡ Δvs ⎤ ⎡ Yξ Yζ ⎤

⎢L Lp Lr 0 ⎥⎥ ⎢⎢ Δps ⎥⎥ ⎢⎢ Lξ Lζ ⎥⎥ ⎡ Δξ ⎤

v

=⎢ + ⎢ ⎥.

⎢ Nv Np Nr 0 ⎥ ⎢ Δrs ⎥ ⎢ N ξ N ζ ⎥ ⎢ Δζ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥⎣

⎣0 1 0 0 ⎦ ⎣ Δφs ⎦ ⎣ 0 0 ⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

184 Flight Dynamics, Simulation, and Control

⎡ Δv s ⎤ ⎡ yv yp yr yφ ⎤ ⎡ Δvs ⎤ ⎡ yξ yζ ⎤

⎢ Δp ⎥ ⎢

⎢ s ⎥ = ⎢ lv lp lr 0 ⎥⎥ ⎢⎢ Δps ⎥⎥ ⎢⎢ lξ lζ ⎥⎥ ⎡ Δξ ⎤

+ ⎢ ⎥.

⎢ Δrs ⎥ ⎢ nv np nr 0 ⎥ ⎢ Δrs ⎥ ⎢ nξ nζ ⎥ ⎢ Δζ ⎥⎦

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎣

⎢⎣ Δφ s ⎥⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφs ⎦ ⎣ 0 0⎦

derivatives.

5.5 (i) Express the concise longitudinal stability derivatives in terms of the

non-dimensional longitudinal stability derivatives.

(ii) Express the concise lateral stability derivatives in terms of the non-

dimensional lateral stability derivatives.

Reference

1. Cook, M. V., Flight Dynamics Principles, Arnold, London, U.K., 1997.

free ebooks ==> www.ebook777.com

6

Longitudinal and Lateral Linear

Stability and Control

6.1 Introduction

To define the concept of stability of aircraft’s motion, it is essential to begin

with the notion of equilibrium flight which refers to a steady motion of flight

of the aircraft. When an aircraft is flying in one such equilibrium flight path,

it is said to be dynamically stable if a small perturbation within a defined

measure or norm applied to the aircraft in the unperturbed state results in

a deviation from this state which does not exceed an established measure or

norm. From a practical standpoint if the aircraft’s motion following a small

perturbation is divergent, no matter how slowly, the aircraft is said to be

dynamically unstable and is considered to be dynamically stable if it returns

to its equilibrium flight path at some time in the future.

When necessary conditions for stability can be established entirely from

the static conditions for equilibrium flight without recourse to the dynami-

cal conditions for equilibrium and the aircraft’s unperturbed state is stable

when these conditions are satisfied, it is said to be statically stable. However,

the conditions for static stability are only a subset of the conditions of

dynamic stability.

To establish the conditions of longitudinal dynamic stability, we assume that

the state vector admits a solution of the form

T T

⎡ Δu ( t ) Δθ ( t ) Δw ( t ) qB ( t ) ⎤⎦ = ⎡⎣ Δu0 Δθ0 Δw0 qB0 ⎤⎦ exp ( λt ) .

⎣

(6.1)

185

www.ebook777.com

free ebooks ==> www.ebook777.com

186 Flight Dynamics, Simulation, and Control

The choice of the assumed solution is most appropriate for stability analy-

sis. If a solution of this form is found and the real part of λ is greater

than zero, then that solution would diverge exponentially. The aircraft’s

unperturbed state may then be considered unstable as one of the perturba-

tion states is exponentially divergent. On the other hand, all the solutions

found are such that the real part of λ is less than zero, then that solution

would not diverge exponentially; rather, all the solutions may be consid-

ered to be converging transients and the aircraft’s unperturbed state may

be considered to be stable. If some of the solutions found are such that the

real part of λ is equal to zero, while the remaining are such that the real

part of λ is less than zero, then the perturbation would neither diverge

exponentially nor converge. Thus, in this case, the aircraft’s unperturbed

state may be considered to be neutrally stable. Thus, the stability of the

aircraft’s unperturbed state can be entirely determined by examining the

sign of the real part of λ.

When this assumed solution is substituted in the equations of motion,

with the controls fixed,

⎡1 0 0 −X q ⎤ ⎡ Δu0 ⎤ ⎡ Xu −g Xw 0 ⎤ ⎡ Δu0 ⎤

⎢0 1 0 ⎥ ⎢

0 ⎥ ⎢ Δθ0 ⎥ ⎥ ⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθ0 ⎥⎥

λ⎢ exp ( λt ) = ⎢ exp ( λt ) ,

⎢0 0 1 −Zq ⎥ ⎢ Δw0 ⎥ ⎢ Zu 0 Zw U es ⎥ ⎢ Δw0 ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥

⎢⎣0 0 − Mw − Mq ⎥⎦ ⎣ qB0 ⎦ ⎢⎣ Mu 0 Mw Mq ⎥⎦ ⎣ qB0 ⎦

(6.2)

⎡ ⎡ Xu −g Xw 0 ⎤ ⎡1 0 0 −X q ⎤ ⎤ ⎡ Δu0 ⎤ ⎡0 ⎤

⎢⎢ ⎥

⎢⎢ 0 0 0 1 ⎥⎥ ⎢0

⎢ 1 0 0 ⎥⎥ ⎥ ⎢⎢ Δθ0 ⎥⎥ ⎢⎢0 ⎥⎥

− λ = (6.3)

⎢ ⎢ Zu 0 Zw U es + Zq ⎥ ⎢0 0 1 −Zq ⎥ ⎥ ⎢ Δw0 ⎥ ⎢0 ⎥

⎢⎢ ⎥ ⎢ ⎥⎥ ⎢ ⎥ ⎢ ⎥

⎢ ⎢ Mu 0 Mw Mq ⎥⎦ ⎢⎣0 0 − Mw − Mq ⎥⎦ ⎥⎦ ⎣ qB0 ⎦ ⎣0 ⎦

⎣ ⎣

tions, it follows by eliminating any three of the four unknowns that the

determinant

⎡ ⎡ Xu −g Xw 0 ⎤ ⎡1 0 0 −X q ⎤ ⎤

⎢⎢ ⎥ ⎢0 ⎥

⎢⎢ 0 0 0 1 ⎥ 1 0 0 ⎥⎥ ⎥

−λ⎢ = 0. (6.4)

⎢ ⎢ Zu 0 Zw U es + Zq ⎥ ⎢0 0 1 −Zq ⎥ ⎥

⎢⎢ ⎥ ⎢ ⎥ ⎥

⎢⎣ ⎢⎣ Mu 0 Mw Mq ⎥⎦ ⎢⎣0 0 − Mw 1 − Mq ⎥⎦ ⎥⎦

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 187

generality, it may be expressed as

Xu − λ −g Xw 0

0 −λ 0 1

= 0. (6.5)

Zu 0 Zw − λ U e + Zq

s

Mu 0 Mw − λMw Mq − λ

equation in λ, known as the characteristic equation, which has the form

Dividing through by a4, the coefficient of the highest power term in the pre-

vious polynomial, we obtain

where

ai

bi = , i = 0, 1, 2 and 3. (6.8)

a4

This is the characteristic equation in its final form, and the roots of this equa-

tion determine the longitudinal stability of the aircraft’s unperturbed state

representing a particular equilibrium flight condition. Considering the case

of any aircraft’s longitudinal dynamics, the roots of the characteristic equa-

tion exhibit certain general features. They may be expressed in the form

( )

λ k = −ωk ζ k ± −1 1 − ζ 2k , k = 1 and 2. (6.9)

( ) ( )

Δ ( λ ) = λ 2 + 2ζ 1ω1λ + ω12 × λ 2 + 2ζ 2ω2λ + ω22 = 0 (6.10)

where the coefficients bi, i = 0, 1, 2 and 3 are related to the damping ratios,

ζk and natural frequencies, ωk, k = 1, 2, by the relations b3 = 2ζ1ω1 + 2ζ2ω2,

b2 = ω12 + ω22 + 4ζ 1ζ 2ω1ω2 , b1 = 2ω1ω2 ( ζ 1ω2 + ζ 2ω1 ), and b0 = ω1 ω2 . Provided

2 2

ω12 > 0 and ω22 > 0 , the stability is entirely dependent on the sign of ζ k for

k = 1 and 2.

www.ebook777.com

free ebooks ==> www.ebook777.com

188 Flight Dynamics, Simulation, and Control

d T T

⎡u w q θ ⎤⎦ = ⎡⎣0 0 0 0 ⎤⎦ (6.11)

dt ⎣

reduces to

Xu − λ −g Xw 0

0 −λ 0 1

= 0. (6.12)

Zu 0 Zw − λ U e + Zq

s

Mu 0 Mw − λMw Mq − λ λ = 0

Since the equilibrium flight corresponds to the trivial solution, the condition

for the trivial solution being the only solution is

Xu − λ −g Xw 0

0 −λ 0 1

> 0. (6.13)

Zu 0 Zw − λ U e + Zq

s

Mu 0 Mw − λMw Mq − λ λ = 0

Thus, for static stability, we require that b0 > 0. This is equivalent to requiring

that ω12 × ω22 > 0, a condition that is implicitly assumed to be satisfied. When

this condition is not satisfied, the assumption that the quartic admits two

pairs of complex roots breaks down and one of the solutions always has a

positive real part. The system is then always unstable. In what follows, we

shall assume that the system is statically stable, that is, ω12 × ω22 > 0. The two

natural frequencies ω1 and ω2 are usually well separated. The lower one cor-

responds to the phugoid mode of oscillation discussed earlier. Hence, in this

case, ω12 is directly proportional to CL. The higher one corresponds to the

so-called short period mode of oscillation of the aircraft. It will be shown later

that in this case, ω22 is directly related to

∂Cm ∂C ∂C

− = − m × L . (6.14)

∂α ∂CL ∂ α

Thus, −∂Cm/∂CL is a measure of the static stability margin of the aircraft and must

be greater than zero. This ensures that the aerodynamic pitching moment

generated is a restoring moment. Once the two pairs of solutions for λ, λ = λk ± ,

k = 1 and 2, are found, we may find the solutions for the state vector:

T

x 0 = ⎣⎡ Δu0 Δθ0 Δw0 qB0 ⎤⎦ . (6.15)

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 189

The vector

1 T

x0 = ⎡ Δu0 Δθ0 Δw0 qB0 ⎤⎦ (6.16)

Δθ0 ⎣

2 in the equation

⎡ ⎡ Xu −g Xw 0 ⎤ ⎡1 0 0 −X q ⎤ ⎤ ⎡ Δu0 ⎤ ⎡0 ⎤

⎢⎢ ⎥

⎢⎢ 0 0 0 1 ⎥⎥ ⎢0 1 0 0 ⎥⎥ ⎥ ⎢⎢ Δθ0 ⎥⎥ ⎢⎢0 ⎥⎥

− λ ⎢ = .

⎢ ⎢ Zu 0 Zw U es + Zq ⎥ ⎢0 0 1 −Zq ⎥ ⎥ ⎢ Δw0 ⎥ ⎢0 ⎥

⎢⎢ ⎥ ⎢ ⎥⎥ ⎢ ⎥ ⎢ ⎥

⎢⎣ ⎢⎣ Mu 0 Mw Mq ⎥⎦ ⎢⎣0 0 − Mw 1 − Mq ⎥⎦ ⎥⎦ ⎣ qB0 ⎦ ⎣0 ⎦

(6.17)

To solve these equations, typically one of the states in the unknown vector,

such as Δθ 0, is assumed to be unity and all the others are solved for. Thus, one

solves for the three ratios, Δu0/Δθ 0, Δw0/Δθ 0 and qB0/Δθ 0, from the equations

⎡ Xu − λ Xw 0 ⎤ ⎡ Δu0 ⎤ ⎡ g ⎤

⎢ ⎥ 1 ⎢ ⎥ ⎢ ⎥

⎢ Zu Zw − λ U + Zq ⎥ ⎢ Δw0 ⎥ = ⎢ 0 ⎥ .

s

e

Δθ0

⎢⎣ Mu Mw − λMw Mq − λ ⎥⎦ ⎣⎢ qB0 ⎥⎦ ⎢⎣ 0 ⎥⎦

These solutions represent the relative motions of the states for each of val-

ues of λ, λ = λk± , k = 1 and 2. These correspond to the modes of oscillation

of the aircraft, and the general motion is composed of these two pairs of

individual modes, that is, the general motion is obtained by multiplying

each of the modes by an arbitrary constant and summing these solutions

for the two pairs of modes. The first pair corresponds to the phugoid mode

(lower natural frequency) and the second (higher natural frequency) to the

short period mode.

Several case studies relating to the calculation of the characteristic polyno-

mial as well as the phugoid and short period damping and natural frequen-

cies are presented first. A complete qualitative discussion is deferred to a

latter section.

In many situations, the original source data for the relevant stability

derivatives that are essential for the longitudinal stability analysis are in

American imperial units, and for purposes of comparison, these data must

be converted to standard SI units. A complete set of the essential conversion

factors are listed in Table 6.1.

www.ebook777.com

free ebooks ==> www.ebook777.com

190 Flight Dynamics, Simulation, and Control

TABLE 6.1

Table of Conversion Factors and Constants

Parameter/Constant Symbol American Units SI Equivalent

Mass m Slug 14.594 kg

Length l Foot 0.3048 m

Velocity V ft/s 0.3048 m/s

Acceleration a ft/s2 0.3048 m/s2

Force F Lb 4.448 N

Moment M lb–ft 1.356 N m

Density ρ Slug/ft3 515.383 kg/m3

Inertia I Slug ft2 1.3558 kg m2

Pressure p lb/ft2 47.8778 N/m2

1 knot kt 1.689 ft/s 0.515 m

Sea level ρ (air) ρ0 0.00238 slug/ft3 1.225 kg/m3

Sea level speed of sound a0 1116.44 ft/s 340.29 m/s

Radian rad 57.3° 57.3°

Accn. due to gravity g 32.17 ft/s2 9.81 m/s2

OF A DC 8 IN HORIZONTAL LEVEL FLIGHT

The relevant stability derivative data for this airliner are provided in

Tables 6.2 and 6.3. For the purposes of this exercise, we consider an air-

craft (DC 8) for which the following stability derivatives are assumed

to be zero:

The governing linear equations of motion are

⎡1 0 0 0 ⎤ ⎡ Δu ⎤ ⎡ Xu −g Xw 0 ⎤ ⎡ Δu ⎤

⎢ ⎥

⎢0 1 0 0 ⎥⎥ ⎢ Δθ ⎥ ⎢⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθ ⎥⎥

⎢ = , (6.19)

⎢0 0 1 0 ⎥ ⎢ Δw ⎥ ⎢ Zu 0 Zw U es ⎥ ⎢ Δw ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣0 0 − Mw 1 ⎦ ⎢⎣ q B ⎥⎦ ⎢⎣ Mu 0 Mw Mq ⎥⎦ ⎣ qB ⎦

⎛ Δw ⎞

h = U es ⎜ Δθ − s ⎟ . (6.20)

⎝ Ue ⎠

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 191

OF A DC 8 IN HORIZONTAL LEVEL FLIGHT

TABLE 6.2

DC 8 (4 Engine Jetliner) Stability Derivative Data

Geometry Data (SI Units)

Span 43.37304 cm 7.0104 AR 7.788188462

Wing Area 241.547904

Flight Conditions

8001 8002 8003 8004

Approach Holding Cruise V-ne

h (m) 0 4572 10058.4 10058.4

M 0.219 0.443 0.84 0.88

a (m/s) 340.4616 322.4784 299.3136 299.3136

Rho (kg/m/m/m) 1.225 0.770647603 0.409535324 0.409535324

Vt0 (m/s) 74.2188 142.70736 251.21616 263.182608

Dynamic pr. 3400.286578 7850.534923 12927.02585 14189.0866

(N/m/m)

Inertial Data

8001 8002 8003 8004

Approach Holding Cruise V-ne

Weight (N) 845120 845120 1023040 1023040

M (kg) 86040.73837 86040.73837 104167.6261 104167.6261

Ixx (kg-m-m) 4186398.073 4213494.5 5107676.613 5107676.613

Iyy 3983174.865 3983174.865 4823164.123 4823164.123

Izz 7559903.316 7966349.731 9659876.459 9659876.459

Ixz 37934.99872 −87385.97919 60966.96223 72753.90826

Xcg/c 0.15 0.15 0.15 0.15

Trim Conditions

8001 8002 8003 8004

Approach Holding Cruise V-ne

q0 0 0 0 0

U0 (m/s) 74.2188 142.70736 251.21616 263.182608

W0 0 0 0 0

Delta-fl (deg) 35 0 0 0

(continued)

www.ebook777.com

free ebooks ==> www.ebook777.com

192 Flight Dynamics, Simulation, and Control

OF A DC 8 IN HORIZONTAL LEVEL FLIGHT

TABLE 6.2 (continued)

DC 8 (4 Engine Jetliner) Stability Derivative Data

Longitudinal State-Space Derivatives

8001 8002 8003 8004

Approach Holding Cruise V-ne

Tu (1/s) −0.000595 −0.0000846 0.000599 0.000733

Xu (1/s, T = 0) −0.02851 −0.00707 −0.0145 −0.0471

Xu (total) −0.0291 −0.00714 −0.014 −0.0463

Xw (1/s) 0.0629 0.0321 0.0043 −0.0259

Xelev 0 0 0 0

Zu (1/s, T = 0) −0.2506 −0.1329 −0.0735 0.0622

Zu (total) −0.2506 −0.1329 −0.0735 0.0622

Zwdot 0 0 0 0

Zw (1/s) −0.6277 −0.756 −0.806 −0.865

Zele (m/s/s/rad) −3.105912 −7.22376 −10.54608 −11.76528

Mu (1/s/m, T = 0) −2.52625E−05 −0.000206693 −0.00257874 −0.008333333

Mu (total) −2.52625E−05 −0.000206693 −0.00257874 −0.008333333

Mwdot (1/m) −0.003503937 −0.002362205 −0.001673228 −0.001706037

Mw (1/s/m) −0.028543307 −0.035104987 −0.036417323 −0.045603675

Mq (1/s) −0.7924 −0.991 −0.924 −1.008

Mele (1/s/s) −1.35 −3.24 −4.59 −5.12

8001 8002 8003 8004

Approach Holding Cruise V-ne

Yv (1/s) −0.1113 −0.1008 −0.0868 −0.0931

Yail (m/s/s/rad) 0 0 0 0

Yrud (m/s/s/rad) 1.764792 4.108704 5.586984 6.132576

VLv (1/s/s) −1.335 −2.68 −4.43 −5.05

Lp (1/s) −0.95 −1.233 −1.18 −1.289

Lr 0.612 0.391 0.336 0.35

Lail (1/s/s) −0.726 −1.62 −2.11 −2.3

Lrud −0.1848 0.374 0.559 0.63

VNv (1/s/s) 0.763 1.271 2.17 2.47

Np (1/s) −0.1192 −0.048 −0.01294 −0.00744

Nr −0.268 −0.252 −0.23 −0.252

Nail (1/s/s) −0.0496 −0.0365 −0.0519 −0.0615

Nrud −0.39 −0.86 −1.168 −1.282

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 193

OF A DC 8 IN HORIZONTAL LEVEL FLIGHT

TABLE 6.3

Values of the Non-Zero Longitudinal Stability Derivatives of the DC 8

in a Cruise Condition

U es (m/s) 251.21616

Xu (1/s) −0.014 Zu (1/s) −0.0735 Mu (1/s/m) −0.00257874 Mw (1/s/m) −0.036417323

Xw (1/s) 0.0043 Zw (1/s) −0.806 Mw (1/m) −0.001673228 Mq (1/s) −0.924

⎢0 ⎥ ⎢ ⎥ ⎢

⎢ 1 0 0 ⎥ ⎢ Δθ ⎥ ⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθ ⎥⎥

=

⎢0 0 1 0 ⎥ ⎢ Δw ⎥ ⎢ −0.0735 0 −0.806 251.22 ⎥ ⎢ Δw ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣0 0 0.001673 1 ⎦ ⎢⎣ q B ⎦⎥ ⎣ −0.002578 0 −0.03642 −0.924 ⎦ ⎣ qB ⎦

(6.21)

which reduces to

⎢ ⎥ ⎢

⎢ Δθ ⎥ = ⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθ ⎥⎥

. (6.22)

⎢ Δw ⎥ ⎢ −0.0735 0 −0.806 251.22 ⎥ ⎢ Δw ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎢⎣ q B ⎥⎦ ⎣ −0.0025 0 −0.0351 −1.3443 ⎦ ⎣ qB ⎦

The resulting phugoid and short period damping and natural frequencies

are tabulated in Table 6.5.

www.ebook777.com

free ebooks ==> www.ebook777.com

194 Flight Dynamics, Simulation, and Control

DYNAMICS IN HORIZONTAL LEVEL FLIGHT AT AN

ALTITUDE OF 20,000 ft; FLIGHT VELOCITY, Ue, OF

830 ft/s (MACH 0.8) AND WEIGHT OF 637,000 lb

The stability derivatives for the Boeing 747 and several other aircraft

may be obtained from Heffley and Jewell [1]. The governing longitudi-

nal coupled equations of motion are

⎢ ⎥ ⎢ 0 ⎥⎥ ⎢⎢ w ⎥⎥ ⎢⎢ 32.7 ⎥⎥

d ⎢ w ⎥ ⎢ −0.0941 −0.624 820

= − η,

dt ⎢ q ⎥ ⎢ −0.000222 −0.00153 −0.668 0 ⎥ ⎢ q ⎥ ⎢ 2.08 ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣θ⎦ ⎣ 0 0 1 0 ⎦⎣θ⎦ ⎣ 0 ⎦

(6.24)

⎛ w ⎞

h = U e ⎜ θ − ⎟ . (6.25)

⎝ U e ⎠

The resulting phugoid and short period damping and natural fre-

quencies for both flight conditions are tabulated in Table 6.5. The

last two examples also show that the attitude equation, which

allows us to determine the height by direct integration once the

other states are found, does affect the stability of the other longitu-

dinal dynamic modes.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 195

DYNAMICS OF THE F104 STARFIGHTER

The stability derivatives for the F104 Starfighter are presented in

Table 6.4. The linear dynamics of the F104 Starfighter is therefore

governed by

⎢0 ⎢ ⎥

1 0 0 ⎥⎥ ⎢ Δθ ⎥ ⎢⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθ ⎥⎥

⎢ = .

⎢0 0 1 0 ⎥ ⎢ Δw ⎥ ⎢ −0.21 0 −0.44 305 ⎥ ⎢ Δw ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣0 0 0.0156 1 ⎦ ⎢⎣ q B ⎦⎥ ⎣ −0.00 0 −0.00056 −0.279 ⎦ ⎣ qB ⎦

(6.28)

⎢ ⎥ ⎢

⎢ Δθ ⎥ = ⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθ ⎥⎥

. (6.29)

⎢ Δw ⎥ ⎢ −0.21 0 −0.44 305 ⎥ ⎢ Δw ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎢⎣ q B ⎥⎦ ⎣ −0.0033 0 0.0063 −5.0370 ⎦ ⎣ qB ⎦

The resulting phugoid and short period damping and natural frequencies

are tabulated in Table 6.5.

TABLE 6.4

Values of the Non-Zero Longitudinal Stability Derivatives of the F 104

Starfighter in a Cruise Condition

U es (ft/s) 305

Xu (1/s) −0.0352 Zu (1/s) −0.21 Mu (1/s/ft) −0.0 Mw (1/s/ft) −0.00056

Xw (1/s) 0.107 Zw (1/s) −0.44 Mw (1/ft) −0.0156 Mq (1/s) −0.279

(continued)

www.ebook777.com

free ebooks ==> www.ebook777.com

196 Flight Dynamics, Simulation, and Control

DYNAMICS OF THE F104 STARFIGHTER

TABLE 6.5

Phugoid and Short Period Damping and Natural Frequencies

Aircraft/Flt. Cond. Mode Damping Natural Frequency

DC 8 Phugoid 0.2476 0.0237

Short period 0.3420 3.1474

Boeing 747 Phugoid 0.2890 0.0102

h = 20,000 ft, M = 0.8 Short period 0.4994 1.2940

Boeing 747 Phugoid 0.0267 0.0653

h = 20,000 ft, M = 0.65 Short period 0.4730 1.2600

F 104 Starfighter Phugoid 0.1944 0.1431

Short period Two real roots

−0.0341 −5.4225

In the lateral case, the state vector is assumed to be of the form

T T

⎡ Δv ( t ) Δp ( t ) Δr ( t ) Δφ ( t ) ⎤⎦ = ⎡⎣ Δv0 Δp0 Δr0 Δφ0 ⎤⎦ exp ( λt ) .

⎣

(6.31)

When this assumed solution is substituted in the equations of motion, with

the controls fixed and the rate derivatives ignored in the left-hand side of the

equations, one obtains,

⎡1 0 0 0⎤

⎢ I ⎥ ⎡ Δv0 ⎤ ⎡ Yv Yp Yr − U e g ⎤ ⎡ Δv0 ⎤

⎢0 1 − xz 0⎥ ⎢ ⎥ ⎢L

⎢ I xx Δ

⎥⎢ 0⎥p v Lp Lr 0 ⎥⎥ ⎢⎢ Δp0 ⎥⎥

λ⎢ ⎥ ⎢ Δr ⎥ exp ( ) ⎢⎢

λt = exp ( λt ) .

I xz N v Np Nr 0 ⎥ ⎢ Δr0 ⎥

⎢0 − 1 0⎥ ⎢ 0 ⎥ ⎢ ⎥⎢ ⎥

⎢ I zz ⎥ ⎣ Δφ0 ⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφ0 ⎦

⎣0 0 0 1 ⎥⎦

⎢

(6.32)

It follows that the following set of homogeneous equations

⎡ ⎡1 0 0 0⎤ ⎤

⎢ ⎡ Yv Yp Yr − U e g⎤ ⎢ ⎥ ⎥ ⎡ Δv0 ⎤ ⎡0 ⎤

⎢0 I xz

⎢⎢ 1 − 0⎥ ⎥ ⎢

⎢ ⎢ Lv Lp Lr 0 ⎥⎥ ⎢ I xx

⎥ ⎢ ⎥

⎥ ⎥ ⎢ Δp0 ⎥ ⎢0 ⎥

⎢⎢N −λ⎢ ⎥ ⎢ Δr ⎥ = ⎢0 ⎥

⎥ (6.33)

⎢⎢ v Np Nr 0 ⎥ I xz

⎥ ⎢0 − 1 0⎥ ⎥ ⎢ 0 ⎥ ⎢ ⎥

⎢⎣ 0 1 0 0⎦ ⎢ I zz ⎥ ⎥ ⎣ Δφ0 ⎦ ⎣0 ⎦

⎢ ⎢0 0 0 1 ⎥⎦ ⎥⎦

⎣ ⎣

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 197

follows by eliminating any three of the four unknowns that the determinant

⎡1 0 0 0⎤

⎡ Yv Yp Yr − U e g⎤ ⎢ I ⎥

⎢L ⎥ ⎢0 1 − xz 0⎥

⎢ v Lp Lr 0⎥ ⎢ I xx ⎥

⎢ Nv Np Nr 0⎥

−λ⎢

I xz ⎥ = 0. (6.34)

⎢ ⎥ ⎢0 − 1 0⎥

⎣0 1 0 0⎦ ⎢ I zz ⎥

⎢0 0 0 1 ⎥⎦

⎣

The determinant equation may be expressed as

Yv − λ Yp Yr − U e g

I xz

Lv Lp − λ Lr + λ 0

I xx

= 0. (6.35)

I xz

Nv Np + Nr − λ 0

I zz

0 1 0 −λ

Evaluating this determinantal equation, we obtain a quartic polynomial

equation in λ, known as the characteristic equation, which has the form

Dividing through by a4, the coefficient of the highest power term in the pre-

vious polynomial, we obtain

Δ ( λ ) = λ 4 + b3λ 3 + b2λ 2 + b1λ + b0 = 0 (6.37)

where

ai

bi = , i = 0, 1, 2 and 3. (6.38)

a4

This is the characteristic equation in its final form, and the roots of this equa-

tion determine the directional (yawing) and lateral motion (sideslip) stabil-

ity of the aircraft’s unperturbed state representing a particular equilibrium

flight condition. Considering the case of any aircraft’s lateral dynamics, the

roots of the characteristic equation exhibit certain general features. They

may be expressed in the form

−1 −1

(

λ d = −ωd ζ d ± −1 1 − ζ 2d , λ 0 = ) T0

and λ s =

Ts

. (6.39)

⎡ 1 ⎤⎡ 1⎤

Δ ( λ ) = ⎢λ + ⎥ ⎢λ + ⎥ ⎡⎣λ 2 + 2ζ dωdλ + ω2d ⎤⎦ = 0. (6.40)

⎣ Ts ⎦ ⎣ To ⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

198 Flight Dynamics, Simulation, and Control

The complex root of the characteristic equation corresponds to the Dutch roll

mode that in its simplest form is a coupled mode involving yaw and roll.

When viewed across the plane of symmetry of the aircraft, the wing tip of an

aircraft performing Dutch roll oscillations traces an elliptic trajectory.

Of the remaining two roots, both of which correspond to real roots, the

one that corresponds to the smaller time constant, To, corresponds to the

roll subsidence mode (predominantly roll) and is always stable (To is always

positive). The remaining root corresponding to a larger time constant, Ts,

corresponds to the spiral divergence mode (predominantly yaw) and can some-

times be unstable (negative time constant). There is usually a small rolling

moment associated with this mode and the stability is determined by the

sense of the rolling moment generated; the mode is stable if the net rolling

moment opposes the disturbing torque. Hence, the stability of the mode is

determined by an equation involving only the roll coordinate.

For directional and lateral static stability, as in the longitudinal case, we require

⎡ 1 ⎤⎡ 1⎤ 2

⎢λ + T ⎥ ⎢λ + T ⎥ ⎡⎣λ + 2ζ dωdλ + ωd ⎤⎦ > 0. (6.41)

2

⎣ s⎦⎣ o⎦ λ =0

For directional static stability, this reduces to the requirement that the aerody-

namic damping in roll and yaw and the aerodynamic stuffiness in yaw must

be sufficiently large. For lateral static stability, a restoring rolling moment is

required and so a negative rolling moment ensures this.

The lateral dynamics of the DC 8 in terms of the state-space derivatives

(Table 6.6) is considered.

Considering the case when the controls are fixed, the perturbation

equations of motion are

⎡1 0 0 0 ⎤ ⎡ Δv ⎤

⎢0 1 −0.0119 0 ⎥⎥ d ⎢⎢ Δp ⎥⎥

⎢

⎢0 −0.0063 1 0 ⎥ dt ⎢ Δr ⎥

⎢ ⎥ ⎢ ⎥

⎣0 0 0 1 ⎦ ⎣ Δφ ⎦

⎢ −0.0176 −1.18 0.336 0 ⎥⎥ ⎢⎢ Δp ⎥⎥

=⎢ (6.42)

⎢ 0.00864 −0.01294 −0.23 0 ⎥ ⎢ Δr ⎥

⎢ ⎥⎢ ⎥

⎣ 0 1 0 0 ⎦ ⎣ Δφ ⎦

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 199

DYNAMICS OF THE DC 8

TABLE 6.6

Values of the Non-Zero Lateral Stability Derivatives of the DC 8

in a Cruise Condition

U es (m/s) 251.21616 Ixz/Ixx 0.0119 Ixz/Izz 0.0063

Yv −0.0868 Lp −1.18 Nv 0.0086379 Nr −0.23

Lv −0.017634 Lr 0.336 Np −0.01294

which reduces to

⎢ Δp ⎥ ⎢ −0.0175 −1.1802 0.333 0 ⎥⎥ ⎢⎢ Δp ⎥⎥

d ⎢ ⎥ ⎢

= . (6.43)

dt ⎢ Δr ⎥ ⎢ 0.0085 −0.0204 −0.2279 0 ⎥ ⎢ Δr ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣ Δφ ⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφ ⎦

The characteristic polynomial is

Δ ( λ ) = λ 4 + 1.4949λ 3 + 2.5408λ 2 + 2.8142λ + 0.0112. (6.44)

The resulting Dutch roll damping and natural frequencies and the roots

corresponding to the roll subsidence and spiral modes are tabulated in

Table 6.7.

TABLE 6.7

Dutch Roll Damping and Natural Frequencies and the Roots Corresponding

to the Roll Subsidence and Spiral Modes

Damping or Natural Frequency or

Aircraft/Flt. Cond. Mode Root Time Constant

DC 8 Dutch roll 0.0794 1.4957

Roll subsidence −1.2534 0.8 s

Spiral −0.0040 249.63 s

Boeing 747 Dutch roll 0.0095 0.9463

h = 20,000 ft, M = 0.8 Roll subsidence −0.5603 1.78 s

Spiral −0.0073 136.55 s

Boeing 747 Dutch roll 0.0823 1.07

h = 20,000 ft, M = 0.65 Roll subsidence −0.9130 1.0953 s

Spiral −0.0108 92.5861 s

Fighter aircraft Dutch roll 0.2110 2.1202

Roll subsidence −0.7653 1.31 s

Spiral −0.0062 161.68 s

www.ebook777.com

free ebooks ==> www.ebook777.com

200 Flight Dynamics, Simulation, and Control

HORIZONTAL LEVEL FLIGHT AT AN ALTITUDE OF 40,000 ft

AND FLIGHT VELOCITY, Ue, OF 774 ft/s (MACH 0.8)

The governing lateral coupled equations of motion are

⎢ ⎥ ⎢ 0 ⎥⎥ ⎢⎢ p ⎥⎥ ⎢⎢ 0.153 ⎥⎥

d ⎢ p ⎥ ⎢ −3.05 −0.4650 0.388

= + ζ.

dt ⎢ r ⎥ ⎢ 0.598 −0.0318 −0.115 0 ⎥ ⎢ r ⎥ ⎢ −0.475 ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣φ⎦ ⎣ 0 1 0.0805 0 ⎦ ⎣φ⎦ ⎣ 0 ⎦

(6.45)

The characteristic polynomial is

At M = 0.65, h = 20,000 ft,

The resulting Dutch roll damping and natural frequencies and the roots

corresponding to the roll subsidence and spiral modes for both the flight

conditions are tabulated in Table 6.7.

HIGHLY MANOEUVRABLE FIGHTER AIRCRAFT

The governing lateral coupled equations of motion with controls fixed are

⎢ ⎥ ⎢ 0 ⎥⎥ ⎢⎢ p ⎥⎥

d ⎢ p ⎥ ⎢ −12.9 −0.746 0.3877

= . (6.48)

dt ⎢ r ⎥ ⎢ 4.31 0.024 −0.174 0 ⎥ ⎢r ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣φ⎦ ⎣ 0 1 0 0 ⎦ ⎣φ⎦

The resulting Dutch roll damping and natural frequencies and the roots

corresponding to the roll subsidence and spiral modes are tabulated in

Table 6.7.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 201

and the Stability of the Modes

To fully appreciate the usefulness of the modal description of aircraft

dynamics, consider the aircraft’s longitudinal equations of motion when it

is in steady trimmed flight. It is assumed that control forces are present, but

there are no other external forces or moments. Under these circumstances,

the equations of motion are

⎡1 0 − xq 0 ⎤ ⎡ Δu s ⎤ ⎡ xu xw xq xθ ⎤ ⎡ Δus ⎤ ⎡ xη xτ ⎤

⎢0 ⎢ ⎥

⎢ 1 − zq 0 ⎥⎥ ⎢ Δw s ⎥ ⎢⎢ zu zw zq zθ ⎥⎥ ⎢⎢ Δws ⎥⎥ ⎢⎢ zη zτ ⎥⎥ ⎡ Δη⎤

= +

⎢0 0 1 − mq 0 ⎥ ⎢ Δq s ⎥ ⎢ mu mw mq mθ ⎥ ⎢ Δqs ⎥ ⎢ mη mτ ⎥ ⎢⎣ Δτ ⎥⎦

⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣0 0 0 1 ⎦ ⎣⎢ Δθ s ⎦⎥ ⎣ 0 0 1 0 ⎦ ⎣ Δθs ⎦ ⎣ 0 0⎦

(6.50)

and

The last equation is independent of all the others and represents the height

integration mode.

It is convenient to partition the dynamics into fast and slow groups, and to

do this, we assume that the fast dynamics is instantaneously satisfied com-

pared to the slower dynamics. Further, we may associate the slower dynam-

ics with Δus and Δθs as in steady trimmed flight, variations in these two

quantities may be considered to be not only small but also slow. Hence,

setting Δw s = Δq s = 0, we obtain

⎢0 ⎢ ⎥

1 − zq 0 ⎥⎥ ⎢ 0 ⎥ ⎢⎢ zu zw zq zθ ⎥⎥ ⎢⎢ Δws ⎥⎥ ⎢⎢ zη zτ ⎥⎥ ⎡ Δη⎤

⎢ = + .

⎢0 0 1 − mq 0 ⎥ ⎢ 0 ⎥ ⎢ mu mw mq mθ ⎥ ⎢ Δqs ⎥ ⎢ mη mτ ⎥ ⎢⎣ Δτ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

0 0 0 1 ⎦ ⎣⎢ Δθ s ⎦⎥ ⎣ 0 0 1 0 ⎦ ⎣ Δθs ⎦ ⎣ 0 0⎦

⎣

(6.52)

Rearranging the equations,

⎡ Δu s ⎤ ⎡ xu xθ xw xq ⎤ ⎡ Δus ⎤ ⎡ xη xτ ⎤

⎢ ⎥ ⎢

⎢ Δθs ⎥ = ⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθs ⎥⎥ ⎢⎢ 0 0 ⎥⎥ ⎡ Δη⎤

+ . (6.53)

⎢ 0 ⎥ ⎢ zu zθ zw zq ⎥ ⎢ Δws ⎥ ⎢ zη zτ ⎥ ⎢⎣ Δτ ⎥⎦

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢⎣ 0 ⎥⎦ ⎢⎣ mu mθ mw mq ⎥⎦ ⎣ Δqs ⎦ ⎣ mη mτ ⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

202 Flight Dynamics, Simulation, and Control

⎢ ⎥=⎢ + +

⎢⎣ Δθs ⎥⎦ ⎣ 0 0 ⎥⎦ ⎢⎣ Δθs ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎢⎣ Δqs ⎥⎦ ⎢⎣ 0 0 ⎥⎦ ⎢⎣ Δτ ⎥⎦

(6.54)

⎡ zu zθ ⎤ ⎡ Δus ⎤ ⎡ zw zq ⎤ ⎡ Δws ⎤ ⎡ zη zτ ⎤ ⎡ Δη⎤ ⎡0 ⎤

+ + =

⎢m

⎣ u mθ ⎥⎦ ⎢⎣ Δθs ⎥⎦ ⎢⎣ mw mq ⎥⎦ ⎢⎣ Δqs ⎥⎦ ⎢⎣ mη mτ ⎥⎦ ⎢⎣ Δτ ⎥⎦ ⎣⎢0 ⎥⎦

−1

⎢ Δq ⎥ = − ⎢ m ⎢⎢ + ⎥ . (6.55)

⎣ s⎦ ⎣ w mq ⎥⎦ ⎢⎣ ⎣ mu mθ ⎥⎦ ⎢⎣ Δθs ⎥⎦ ⎢⎣ mη mτ ⎥⎦ ⎢⎣ Δτ ⎥⎦ ⎥⎦

−1

⎢ ⎥=⎢ −

⎢⎣ Δθs ⎥⎦ ⎣ 0 0 ⎥⎦ ⎢⎣ Δθs ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎢⎣ mw mq ⎥⎦ ⎢⎣ mu mθ ⎥⎦ ⎢⎣ Δθs ⎥⎦

−1

⎡ xη xτ ⎤ ⎡ Δη⎤ ⎡ xw x q ⎤ ⎡ zw z q ⎤ ⎡ zη zτ ⎤ ⎡ Δη⎤

+⎢ − (6.56)

⎣0 0 ⎥⎦ ⎣⎢ Δτ ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎢⎣ mw mq ⎥⎦ ⎢⎣ mη mτ ⎥⎦ ⎢⎣ Δτ ⎥⎦

−1

⎡ Δu s ⎤ ⎡ ⎡ xu xθ ⎤ ⎡ x w x q ⎤ ⎡ zw z q ⎤ ⎡ zu zθ ⎤ ⎤ ⎡ Δus ⎤

⎢ ⎥ = ⎢⎢ − ⎢ ⎥

⎢⎣ Δθs ⎥⎦ ⎢⎣ ⎣ 0 0 ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎣ mw mq ⎥⎦ ⎢⎣ mu mθ ⎥⎦ ⎥ ⎢⎣ Δθs ⎥⎦

⎦

−1

⎡ ⎡ xη xτ ⎤ ⎡ xw x q ⎤ ⎡ zw z q ⎤ ⎡ zη zτ ⎤ ⎤ ⎡ Δη⎤

+ ⎢⎢ − ⎥ . (6.57)

⎢⎣ ⎣ 0 0 ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎢⎣ mw mq ⎥⎦ ⎢⎣ mη mτ ⎥⎦ ⎥ ⎢⎣ Δτ ⎥⎦

⎦

which has several interesting interpretations. The other two components of

the mode are found from the constraint equation relating the two pairs of

components. The characteristic equation of the previously mentioned pair

is a second-order polynomial in s and for most aircraft represents a lightly

damped oscillation. In fact, there is an interesting interpretation, based on

point mass approximations, which allows one to estimate the natural

frequency of phugoid oscillations quite easily.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 203

Now one may also consider the fast subsystem of the longitudinal

dynamics. To derive the equations governing the fast subsystem, consider

the governing longitudinal equations in their rearranged form given by

⎢0 ⎢ ⎥

1 0 0 ⎥⎥ ⎢ Δθ s ⎥ ⎢⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθs ⎥⎥ ⎢⎢ 0 0 ⎥⎥ ⎡ Δη⎤

⎢ = + .

⎢0 0 1 − zq ⎥ ⎢ Δw s ⎥ ⎢ zu zθ zw zq ⎥ ⎢ Δws ⎥ ⎢ zη zτ ⎥ ⎢⎣ Δτ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢0 0 0 1 − mq ⎥⎦ ⎢⎣ Δq s ⎥⎦ ⎢⎣ mu mθ mw mq ⎥⎦ ⎣ Δqs ⎦ ⎣ mη mτ ⎦

⎣

(6.58)

Considering the last two equations and recognising that the contributions of

Δus and Δθs may be ignored, we have

⎢0 = + (6.59)

⎣ 1 − mq ⎥⎦ ⎢⎣ Δq s ⎥⎦ ⎢⎣ mw mq ⎥⎦ ⎢⎣ Δqs ⎥⎦ ⎢⎣ mη mτ ⎥⎦ ⎢⎣ Δτ ⎥⎦

or

⎡ zq ⎤ ⎡ zq ⎤

Δ w ⎢1 1 − mq ⎥ ⎡ zw zq ⎤ ⎡ Δws ⎤ ⎢

1

1 − mq ⎥ ⎡ zη zτ ⎤ ⎡ Δη⎤

⎡ s⎤ ⎢

⎢ Δq ⎥ = ⎢

⎥⎢ +⎢ ⎥⎢ .

⎣ s⎦ 1 ⎥ ⎣ mw mq ⎥⎦ ⎢⎣ Δqs ⎥⎦ ⎢ 1 ⎥ ⎣ mη mτ ⎥⎦ ⎢⎣ Δτ ⎥⎦

⎢0 1 − mq ⎥⎦ ⎢0 1 − mq ⎥⎦

⎣ ⎣

(6.60)

If we let

⎡ zq ⎤ ⎡ zq ⎤

1 1

⎡ zʹw zʹq ⎤ ⎢ 1 − mq ⎥ ⎡ zw zq ⎤ ⎡ zʹη zʹτ ⎤ ⎢ 1 − mq ⎥ ⎡ zη zτ ⎤

⎢ mʹ =⎢ ⎥⎢ and ⎢ =⎢ ⎥ ,

⎣ w mʹq ⎥⎦ ⎢ 1 ⎥ ⎣ mw

⎥

mq ⎦ ⎣ mʹη mʹτ ⎥⎦ ⎢ 1 ⎥ ⎣ mη

⎢ mτ ⎥⎦

⎢0 1 − m ⎥ ⎢0 1 − m ⎥

⎣ q ⎦ ⎣ q ⎦

(6.61)

⎢ Δq ⎥ = ⎢ mʹ + . (6.62)

⎣ s⎦ ⎣ w mʹq ⎥⎦ ⎢⎣ Δqs ⎥⎦ ⎢⎣ mʹη mʹτ ⎥⎦ ⎢⎣ Δτ ⎥⎦

tem and it can be reduced to a single second-order equation in Δws. These are

www.ebook777.com

free ebooks ==> www.ebook777.com

204 Flight Dynamics, Simulation, and Control

motion. It is the body contribution to the longitudinal dynamics and depends

to a large extent on the stability of the orientation of the aircraft.

Consider next the aircraft’s lateral equations of motion when it is in steady

trimmed level flight. Again, it is assumed that control forces are present, but

there are no other external forces or moments. Under these circumstances,

the equations of motion are

⎡1 − y p − y r 0 ⎤ ⎡ Δv s ⎤ ⎡ yv yp yr yφ ⎤ ⎡ Δvs ⎤ ⎡ yξ yζ ⎤

⎢0 ⎢ ⎥

⎢ 1 − lp −lr 0 ⎥⎥ ⎢ Δp s ⎥ ⎢⎢ lv lp lr lφ ⎥⎥ ⎢⎢ Δps ⎥⎥ ⎢⎢ lξ lζ ⎥⎥ ⎡ Δξ ⎤

= +

⎢0 −np 1 − nr 0 ⎥ ⎢ Δrrs ⎥ ⎢ nv np nr nφ ⎥ ⎢ Δrs ⎥ ⎢ nξ nζ ⎥ ⎢⎣ Δζ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣0 0 0 1 ⎦ ⎢⎣ Δφ s ⎥⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφs ⎦ ⎣ 0 0⎦

(6.63)

and

Like the height integration mode in the longitudinal case, the last equation

represents the heading integration mode. Further, the second of the four

equations represents a fast equation and can be assumed to decay so rapidly

that it is instantaneously satisfied. Thus,

Δp s = 0 (6.65)

and

⎡ Δvs ⎤

⎢ Δp ⎥

s ⎡ Δξ ⎤

lr Δrs + ⎡⎣lv lp lr lφ ⎤⎦ ⎢ ⎥ + ⎡lξ lζ ⎤⎦ ⎢ ⎥ = 0. (6.66)

⎢ Δrs ⎥ ⎣ ⎣ Δζ ⎦

⎢ ⎥

⎣ Δφs ⎦

⎡ Δvs ⎤

l 1 ⎢ ⎥ 1 ⎡ Δξ ⎤

Δps = − r Δrs − ⎡⎣lv lr lφ ⎤⎦ ⎢ Δrs ⎥ − ⎡⎣lξ lζ ⎦⎤ ⎢ ⎥ . (6.67)

lp lp l ⎣ Δζ ⎦

⎢⎣ Δφs ⎥⎦ p

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 205

⎡ lr ⎤

⎢1 − y r + yp 0⎥

lp ⎥ ⎡ Δvv s ⎤ ⎡ yv

⎢ yr yφ ⎤ ⎡ Δvs ⎤ ⎡yp ⎤ ⎡ Δvs ⎤

⎢ l ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ 1 ⎢ ⎥ ⎢ ⎥

⎢0 1 − nr + r np 0 ⎥ ⎢ Δrs ⎥ = ⎢ nv nr nφ ⎥ ⎢ Δrs ⎥ − ⎢ np ⎥ ⎡⎣lv lr lφ ⎤⎦ ⎢ Δrs ⎥

⎢ lp ⎥⎢ ⎥ ⎢ 0 lp

0 0 ⎥⎦ ⎣⎢ Δφs ⎥⎦ ⎢⎣ 1 ⎥⎦ ⎢⎣ Δφs ⎥⎦

⎢ lr ⎥ ⎣ Δφs ⎦ ⎣

⎢0 1⎥

⎢⎣ lp ⎥⎦

⎡ yp ⎤ ⎡ y y ⎤

1⎢ ⎥ ⎡ Δ⎤ ⎢ ⎥ ⎡ Δ⎤

− ⎢ np ⎥ ⎡⎣l l ⎤⎦ ⎢ ⎥ + ⎢ n n ⎥ ⎢ ⎥ (6.68)

lp ⎣ Δ⎦ ⎢ 0 Δ

⎢⎣ 1 ⎥⎦ ⎣ 0 ⎥⎦ ⎣ ⎦

⎡ lr ⎤

⎢1 − y r + yp 0⎥

lp ⎥ ⎡ Δvv s ⎤ ⎡ ⎡ yv

⎢ yr yφ ⎤ ⎡ yp ⎤ ⎤ ⎡ Δvs ⎤

⎢ l ⎥⎢ ⎥ ⎢⎢ ⎥ 1 ⎢ ⎥ ⎥⎢ ⎥

⎢0 1 − nr + r np 0 ⎥ ⎢ Δrs ⎥ = ⎢ ⎢ nv nr nφ ⎥ − ⎢ np ⎥ ⎡⎣lv lr lφ ⎤⎦ ⎥ ⎢ Δrs ⎥

⎢ lp ⎥ ⎢ ⎥ ⎢⎢ 0 lp

0 0 ⎥⎦ ⎢⎣ 1 ⎥⎦ ⎥ ⎢ Δφs ⎥

⎢ lr ⎥ ⎣ Δφs ⎦ ⎣ ⎣ ⎦⎣ ⎦

⎢0 1⎥

⎢⎣ lp ⎥⎦

⎡ ⎡ yξ yζ ⎤ ⎡ yp ⎤ ⎤

⎢⎢ ⎥ 1 ⎢ ⎥ ⎥ ⎡ Δξ ⎤

+ ⎢ ⎢ nξ nζ ⎥ −

lp ⎢ np ⎥ ⎡⎣lξ lζ ⎤⎦ ⎥ ⎢ ⎥

Δζ

(6.69)

⎢⎢ 0 0 ⎥⎦ ⎢⎣ 1 ⎥⎦ ⎥⎣ ⎦

⎣⎣ ⎦

We now observe that the last row can be ignored. This last mode is the spiral

mode which represents the effect of gravity during a rolling motion and

causes the aircraft to sideslip and yaw. However, it is often unstable although

pilots are able to control the aircraft relatively easily when this is the case.

Thus, we obtain the approximate equations for the Dutch roll mode which are

⎡ lr ⎤

⎢1 − y r + yp ⎥

lp Δv ⎡ y yr ⎤ 1 ⎡ yp ⎤ ⎤ ⎡ Δvs ⎤

⎢ ⎥ ⎡⎢ s ⎤⎥ = ⎢ ⎡⎢ v − ⎢ n ⎥ ⎡⎣lv lr ⎤⎦ ⎥ ⎢ ⎥

⎢ lr ⎥ ⎣ Δrs ⎦ ⎢⎣ ⎣ nv nr ⎥⎦ lp ⎣ p⎦ ⎥⎦ ⎣ Δrs ⎦

⎢0 1 − nr + np ⎥

⎣ lp ⎦

⎡ ⎡ yξ yζ ⎤ 1 ⎡ yp ⎤ ⎤ ⎡ Δξ ⎤

+ ⎢⎢ − ⎢ n ⎥ ⎡⎣lξ lζ ⎤⎦ ⎥ ⎢ ⎥ (6.70)

⎢⎣ ⎣ nξ nζ ⎥⎦ lp ⎣ p⎦ ⎥⎦ ⎣ Δζ ⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

206 Flight Dynamics, Simulation, and Control

lr

−1

⎡ ⎤

⎢ 1 − y r + yp ⎥

Δ v

⎡ s⎤ Δ v

⎡ s⎤ ⎢ lp ⎡ y yζ ⎤ 1 ⎡ yp ⎤ ⎤ ⎡ Δξξ ⎤

⎢ Δr ⎥ = A ⎢ Δr ⎥ + ⎢ ⎥ ⎢ ⎡⎢ ξ − ⎢ n ⎥ ⎡⎣ lξ lζ ⎤⎦ ⎥ ⎢ ⎥

⎣ s⎦ ⎣ s⎦ 0 l ⎥ n nζ ⎥⎦ lp ⎣ p⎦ ⎥⎦ ⎣ Δζ ⎦

⎢ 1 − nr + r np ⎥ ⎢⎣ ⎣ ξ

⎣ lp ⎦

(6.71)

where

−1

⎡ lr ⎤

⎢1 − y r + yp ⎥

lp ⎡ y yr ⎤ 1 ⎡ y p ⎤ ⎤

A=⎢ ⎥ ⎢ ⎡⎢ v − ⎢ ⎥ ⎡lv lr ⎤⎦ ⎥ , (6.72)

⎢ l ⎥ n nr ⎥⎦ lp ⎣ np ⎦ ⎣

⎢0 1 − nr + r np ⎥ ⎢⎣ ⎣ v ⎥⎦

⎣ lp ⎦

and formulae for evaluating the natural frequency and damping follow from

the standard form for a second-order coupled equation, that is, 2ζ = Trace (A)

and ω2n = det A . The damping is predominantly determined by the deriva-

tive nr, and hence, for stability, it is required that nr < 0.

The equation for the roll subsidence mode is

⎡ Δξ ⎤

(1 − lp ) Δp s = lpΔps + ⎡⎣lξ lζ ⎤⎦ ⎢ ⎥ . (6.73)

⎣ Δζ ⎦

The Dutch roll is coupled roll yaw motion that is often unstable while the

roll subsidence is caused by the high roll damping of the wings. A relatively

accurate time constant for the motion may be obtained from the stability

derivative, lp, and is almost always very stable.

To derive the approximation for the spiral mode, we exploit the fact that both

the roll subsidence and Dutch roll modes are relatively faster than it and set

Δv s = Δp s = Δrs = 0 in the lateral equations. Eliminating Δvs, Δps and Δrs in the

last equations results in the approximate equation for the spiral mode. Thus,

⎡1 − y p − y r 0⎤ ⎡ 0 ⎤ ⎡ 0 ⎤

⎢0 ⎢ ⎥ ⎢ ⎥

⎢ 1 − lp −lr 0 ⎥⎥ ⎢ 0 ⎥ ⎢ 0 ⎥

=

⎢0 −np 1 − nr 0⎥ ⎢ 0 ⎥ ⎢ 0 ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣0 0 0 1 ⎦ ⎣⎢ Δφ s ⎥⎦ ⎢⎣ Δφ s ⎥⎦

⎡ yv yp yr yφ ⎤ ⎡ Δvs ⎤ ⎡ yξ yζ ⎤

⎢l lp lr lφ ⎥⎥ ⎢⎢ Δps ⎥⎥ ⎢⎢ lξ lζ ⎥⎥ ⎡ Δξ ⎤

v

=⎢ + , (6.74)

⎢ nv np nr nφ ⎥ ⎢ Δrs ⎥ ⎢ nξ nζ ⎥ ⎢⎣ Δζ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣0 1 0 0 ⎦ ⎣ Δφs ⎦ ⎣ 0 0⎦

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 207

⎡ yv yp y r ⎤ ⎡ Δvs ⎤ ⎡ yφ ⎤ ⎡ yξ yζ ⎤

⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎡ Δξ ⎤

⎢ lv lp lr ⎥ ⎢ Δps ⎥ = − ⎢ lφ ⎥ Δφs − ⎢ lξ lζ ⎥ ⎢ ⎥ . (6.75)

Δζ

⎢⎣ nv np nr ⎥⎦ ⎢⎣ Δrs ⎥⎦ ⎣⎢ nφ ⎥⎦ ⎢⎣ nξ nζ ⎥⎦ ⎣ ⎦

Thus,

−1

⎡ Δvs ⎤ ⎡ yv yp yr ⎤ ⎡ yφ ⎤ ⎡ yv yp yr ⎤ ⎡ yξ yζ ⎤

−1

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎡ Δξ ⎤

⎢ Δps ⎥ = − ⎢ lv lp lr ⎥ ⎢ lφ ⎥ Δφs − ⎢ lv lp lr ⎥ ⎢ lξ lζ ⎥ ⎢ ⎥ .

Δζ

⎢⎣ Δrs ⎥⎦ ⎢⎣ nv np nr ⎥⎦ ⎢⎣ nφ ⎥⎦ ⎢⎣ nv np nr ⎥⎦ ⎢⎣ nξ nζ ⎥⎦ ⎣ ⎦

(6.76)

−1

⎡ yv yp yr ⎤ ⎡ yφ ⎤

⎢ ⎥ ⎢ ⎥

Δφ s + ⎡⎣0 1 0 ⎤⎦ ⎢ lv lp lr ⎥ ⎢ lφ ⎥ Δφs

⎢⎣ nv np nr ⎥⎦ ⎢⎣ nφ ⎥⎦

−1

⎡ yv yp yr ⎤ ⎡ yξ yζ ⎤

⎢ ⎥ ⎢ ⎥ ⎡ Δξ ⎤

+ ⎡⎣0 1 0 ⎤⎦ ⎢ lv lp lr ⎥ ⎢ lξ lζ ⎥ ⎢ ⎥ = 0. (6.77)

Δζ

⎢⎣ nv np nr ⎥⎦ ⎢⎣ nξ nζ ⎥⎦ ⎣ ⎦

When the axes are assumed to be the stability axes, lϕ = nϕ = 0. Further, when

the controls are fixed,

−1

⎡ yv yp yr ⎤ ⎡ yφ ⎤

⎢ ⎥ ⎢ ⎥

Δφ s + ⎡⎣0 1 0 ⎤⎦ ⎢ lv lp lr ⎥ ⎢ 0 ⎥ Δφs = 0 (6.78)

⎢⎣ nv np nr ⎥⎦ ⎢⎣ 0 ⎥⎦

which reduces to

Δφ s +

( y (l n

φ r v − lv nr ) )

Δφs = 0 (6.79)

Δ

where

Δ = ⎡⎣ yv ( lp nr − lr np ) + y p ( lr nv − lv nr ) + y r ( lv np − lp nv ) ⎤⎦ ≈ y r ( lv np − lp nv ) . (6.80)

www.ebook777.com

free ebooks ==> www.ebook777.com

208 Flight Dynamics, Simulation, and Control

⎡ yv yp yr ⎤ ⎡ yξ yζ ⎤

−1

Δφ s +

( yφ ( lr nv − lvnr ) ) Δφ + ⎡0 ⎢

0 ⎤⎦ ⎢ lv lp lr ⎥

⎥ ⎢ ⎥ ⎡ Δξ ⎤

Δ

s ⎣ 1 ⎢ lξ lζ ⎥ ⎢ ⎥ = 0.

Δζ

⎢⎣ nv np nr ⎥⎦ ⎢⎣ nξ nζ ⎥⎦ ⎣ ⎦

(6.81)

Thus, the condition for stability of the spiral mode depends entirely on the

rolling and yawing moment derivatives. Since the derivative yr is negative,

the derivative yϕ is positive and the condition for stability is

−nr

lr < ( −lv ) , (6.82)

nv

for lateral static stability.

Once the roll angle component is found, the other components of the mode

are found from

⎡ yv yp y r ⎤ ⎡ Δvs ⎤ ⎡ yφ ⎤ ⎡ yξ yζ ⎤

⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎡ Δξ ⎤

⎢ lv lp lr ⎥ ⎢ Δps ⎥ = − ⎢ lφ ⎥ Δφs − ⎢ lξ lζ ⎥ ⎢ ⎥ . (6.83)

Δζ

⎢⎣ nv np nr ⎥⎦ ⎢⎣ Δrs ⎥⎦ ⎣⎢ nφ ⎥⎦ ⎢⎣ nξ nζ ⎥⎦ ⎣ ⎦

Thus, eliminating the spiral mode from the lateral equations,

⎡1 − y p − y r ⎤ ⎡ Δv s ⎤ ⎡ yv yp y r ⎤ ⎡ Δvs ⎤ ⎡ yξ yζ ⎤

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎡ Δξ ⎤

⎢0 1 − lp −lr ⎥ ⎢ Δp s ⎥ = ⎢ lv lp lr ⎥ ⎢ Δps ⎥ + ⎢ lξ lζ ⎥ ⎢ ⎥ . (6.84)

Δζ

⎢0 −np 1 − nr ⎥⎦ ⎢⎣ Δrs ⎥⎦ ⎢⎣ nv np nr ⎥⎦ ⎢⎣ Δrs ⎥⎦ ⎢⎣ nξ nζ ⎥⎦ ⎣ ⎦

⎣

⎡1 − y r − y p ⎤ ⎡ Δv s ⎤ ⎡ yv yr y p ⎤ ⎡ Δvs ⎤ ⎡ yξ yζ ⎤

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎡ Δξ ⎤

⎢0 −nr 1 − np ⎥ ⎢ Δrs ⎥ = ⎢ nv nr np ⎥ ⎢ Δrs ⎥ + ⎢ nξ nζ ⎥ ⎢ ⎥ . (6.85)

Δζ

⎢0 1 − lr −lp ⎥⎦ ⎢⎣ Δp s ⎦⎥ ⎢⎣ lv lr lp ⎥⎦ ⎢⎣ Δps ⎥⎦ ⎢⎣ lξ lζ ⎥⎦ ⎣ ⎦

⎣

2

) = 0 (6.86)

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 209

with the roll subsidence mode. The previous sets of lateral equations are of

primary importance when roll rates are generally very high and when the

effects of gravity can be considered to be relatively small. However, since this

is a relatively stable mode, when the roll rates can be assumed to be small,

for stability and bifurcation analysis, the approximation involving the Dutch

roll and spiral modes is of primary importance.

OF A DC 8 IN HORIZONTAL LEVEL FLIGHT

Reconsider Case Study 6.1 and obtain the phugoid and short period

damping ratios and natural frequencies using the previous approxima-

tions to the modal dynamics.

The equations of motion are

⎢0 ⎥ ⎢ ⎥ ⎢

1 0 0 ⎥ ⎢ Δθ ⎥ ⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθ ⎥⎥

⎢ = .

⎢0 0 1 0 ⎥ ⎢ Δw ⎥ ⎢ −0.0735 0 −0.806 251.22 ⎥ ⎢ Δw ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥

0 0 0.001673 1 ⎦ ⎢⎣ q B ⎦⎥ ⎣ −0.002578 0 −0.03642 −0.924 ⎦ ⎣ qB ⎦

⎣

(6.87)

dynamics which is

⎢0.001673 = (6.88)

⎣ 1 ⎥⎦ ⎢⎣ q B ⎥⎦ ⎣⎢ −0.03642 −0.924 ⎥⎦ ⎢⎣ qB ⎥⎦

or

⎢ q ⎥ = ⎢ −0.03507 . (6.89)

⎣ B⎦ ⎣ −1.3443 ⎥⎦ ⎢⎣ qB ⎥⎦

sponding to a damping ratio of 0.348 and a natural frequency of 3.0897.

The corresponding results obtained in Case Study 6.1 are 0.342 and 3.1474.

(continued)

www.ebook777.com

free ebooks ==> www.ebook777.com

210 Flight Dynamics, Simulation, and Control

OF A DC 8 IN HORIZONTAL LEVEL FLIGHT

To obtain the phugoid, we assume that the fast equations are satisfied

instantaneously. Thus,

⎡ 1 0 ⎤ ⎡ Δw ⎤ ⎡ −0.806 251.22 ⎤ ⎡ Δw ⎤ ⎡ −0.0735 ⎤ ⎡0 ⎤

⎢0.001673 = +⎢ Δu = ⎢ ⎥

1 ⎥⎦ ⎢⎣ q B ⎥⎦ ⎣⎢ −0.03642 ⎥ ⎢ ⎥ ⎥

−0.924 ⎦ ⎣ qB ⎦ ⎣ −0.002578 ⎦ ⎣0 ⎦

⎣

(6.91)

and

−1

⎡ Δw ⎤ ⎡ −0.806 251.22 ⎤ ⎡ 0.0735 ⎤

⎢ q ⎥ = ⎢ −0.03642 −0.924 ⎥⎦ ⎢⎣0.002578 ⎥⎦

Δu

⎣ B⎦ ⎣

⎡ −0.924 −251.22 ⎤ ⎡ 0.0735 ⎤ ⎡ −0.0723 ⎤

= 0.1011 ⎢ Δu = ⎢ ⎥ Δu.

⎣0.03642 −0.806 ⎥⎦ ⎢⎣0.002578 ⎥⎦ ⎣ 0.000061⎦

(6.92)

From the first two equations for Δu and Δθ,

⎢ ⎥ = ⎢ + . (6.93)

⎢⎣ Δθ ⎥⎦ ⎣ 0 0 ⎥⎦ ⎢⎣ Δθ ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎢⎣ qB ⎥⎦

Eliminating the fast states, it reduces to

⎡ Δu ⎤ ⎡ −0.014 −9.81⎤ ⎡ Δu ⎤ ⎡0.0043 0 ⎤ ⎡ −0.0723 ⎤

⎢ ⎥ = ⎢ + Δu (6.94)

⎢⎣ Δθ ⎥⎦ ⎣ 0 0 ⎥⎦ ⎢⎣ Δθ ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎢⎣ 0.000061⎥⎦

and simplifies as

⎡ Δu ⎤ ⎡ −0.014 −9.81⎤ ⎡ Δu ⎤

⎢ ⎥ = ⎢ (6.95)

⎢⎣ Δθ ⎥⎦ ⎣0.000061 0 ⎥⎦ ⎢⎣ Δθ ⎥⎦

Hence, the characteristic polynomial is

Δ ( λ ) = λ 2 + 0.014λ + 0.0005941. (6.96)

The roots of the characteristic polynomial are −0.0007 ± i0.0233 cor-

responding to a damping ratio of 0.2872 and a natural frequency of

0.0244. The corresponding results obtained in Case Study 6.1 are 0.2476

and 0.0237. The modal amplitude ratio is given by

Δu0 λ

= . (6.97)

Δθ0 0.000061

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 211

Reconsider Case Study 6.4 and obtain the damping ratios and natural

frequencies corresponding to the lateral modes, using the previous

approximations to the modal dynamics.

The equations of motion are

⎡1 0 0 0 ⎤ ⎡ Δv ⎤

⎢0 1 −0.0119 0 ⎥⎥ d ⎢⎢ Δp ⎥⎥

⎢

⎢0 −0.0063 1 0 ⎥ dt ⎢ Δr ⎥

⎢ ⎥ ⎢ ⎥

⎣0 0 0 1 ⎦ ⎣ Δφ ⎦

⎢ −0.0176 −1.18 0.336 0 ⎥⎥ ⎢⎢ Δp ⎥⎥

=⎢ . (6.98)

⎢ 0.00864 −0.01294 −0.23 0 ⎥ ⎢ Δr ⎥

⎢ ⎥⎢ ⎥

⎣ 0 1 0 0 ⎦ ⎣ Δφ ⎦

⎢ ⎥ ⎢ 0 ⎥⎥ ⎢⎢ Δp ⎥⎥

d ⎢ Δp ⎥ ⎢ −0.0175 −1.1802 0.333

= . (6.99)

dt ⎢ Δr ⎥ ⎢ 0.0085 −0.0204 −0.2279 0 ⎥ ⎢ Δr ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣ Δφ ⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφ ⎦

Δv = Δr = Δϕ = 0, and we obtain

obtained in Case Study 6.4 are −1.2534 for the pole and 0.8 s for the time

constant.

Assuming that the roll subsidence satisfied instantaneously,

⎡ Δv ⎤

⎢ Δp ⎥

d

Δp = ⎡⎣ −0.0175 −1.1802 0.333 0 ⎤⎦ ⎢ ⎥ = 0. (6.101)

dt ⎢ Δr ⎥

⎢ ⎥

⎣ Δφ ⎦

(continued)

www.ebook777.com

free ebooks ==> www.ebook777.com

212 Flight Dynamics, Simulation, and Control

DYNAMICS OF THE DC 8

Hence,

0.333Δr − 0.0175Δv

Δp = = 0.282Δr − 0.0148Δv. (6.102)

1.1802

⎡ Δv ⎤

⎡ Δv ⎤ ⎡ −0.0868 0 −251.22 9.81⎤ ⎢ ⎥

d ⎢ ⎥ ⎢ ⎥ 0. 282 Δr − 0.0148Δv

Δr = 0.0085 −0.0204 −0.2279 0 ⎥⎢ ⎥.

dt ⎢ ⎥ ⎢ ⎢ Δr ⎥

⎢⎣ Δφ ⎥⎦ ⎢⎣ 0 1 0 0 ⎥⎦ ⎢ ⎥

⎣ Δφ ⎦

(6.103)

Setting Δϕ = 0 gives

⎡ Δv ⎤

d ⎡ Δv ⎤ ⎡ −0.0868 0 −251.22 ⎤ ⎢ ⎥

⎢ ⎥=⎢ 0.282Δr − 0.0148Δv ⎥ ,

dt ⎣ Δr ⎦ ⎣ 0.0085 −0.0204 −0.2279 ⎥⎦ ⎢

⎢⎣ Δr ⎥⎦

(6.104)

which simplifies to

d ⎡ Δv ⎤ ⎡ −0.0868 −251.22 ⎤ ⎡ Δv ⎤

⎢ ⎥=⎢ . (6.105)

dt ⎣ Δr ⎦ ⎣ 0.0088 −0.2337 ⎥⎦ ⎢⎣ Δr ⎥⎦

sponding to a damping ratio of 0.1073 and a natural frequency of 1.4937.

The corresponding results obtained in Case Study 6.4 are 0.0794 and

1.4957. The modal amplitude ratio is given by

Δv0 λ ( λ + 0.2337 )

= . (6.107)

Δψ 0 0.0088

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 213

DYNAMICS OF THE DC 8

To establish the approximation for the spiral mode, we assume that

the first three equations are satisfied instantaneously. Hence,

d ⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥

Δp = −0.0175 −1.1802 0.333 ⎥ ⎢ Δp ⎥ + ⎢ 0 ⎥ Δφ = ⎢0 ⎥

dt ⎢ ⎥ ⎢

⎢⎣ Δr ⎥⎦ ⎢⎣ 0.0085 −0.0204 −0.2279 ⎥⎦ ⎢⎣ Δr ⎥⎦ ⎢⎣ 0 ⎦⎥ ⎢⎣0 ⎥⎦

(6.108)

and

−1

⎡ Δv ⎤ ⎡ −0.0868 0 −251.22 ⎤ ⎡9.81⎤

⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ Δp ⎥ = − ⎢ −0.0175 −1.1802 0.333 ⎥ ⎢ 0 ⎥ Δφ. (6.109)

⎢⎣ Δr ⎥⎦ ⎢⎣ 0.0085 −0.0204 −0.2279 ⎥⎦ ⎢⎣ 0 ⎥⎦

Hence,

−1

⎡ −0.0868 0 −251.22 ⎤ ⎡9.81⎤

⎢ ⎥ ⎢ ⎥

Δφ = − ⎡⎣0 1 0 ⎤⎦ ⎢ −0.0175 −1.1802 0.333 ⎥ ⎢ 0 ⎥ Δφ = −0.0043Δφ.

⎢⎣ 0.0085 −0.0204 −0.2279 ⎥⎦ ⎢⎣ 0 ⎥⎦

(6.110)

obtained in Case Study 6.4 are −0.004 for the pole and 249.63 s for the

time constant.

The linear modes of motion corresponding to small perturbations from a

state of equilibrium in a uniform uncontrolled flight with constant forward

speed and their principal features are summarised in this section.

This is basically an energy-conserving mode, where the sum of the aircraft’s

translational and rotational energies and the aircraft’s gravitational potential

energy is always maintained at a constant value. Observe that the chemical

energy generated is completely and continuously balanced by the energy dissi-

pated in flight. For an observer flying in steady formation with the aircraft, the

phugoid has an elliptic pattern with the vertical amplitude of about 2 times

www.ebook777.com

free ebooks ==> www.ebook777.com

214 Flight Dynamics, Simulation, and Control

the horizontal. Furthermore, for a positive rate of climb, the forward accelera-

tion is negative while the pitch attitude is positive and aligned with the path,

indicating that the motion around the ellipse is counterclockwise. The phugoid

oscillation becomes a non-oscillatory convergence and a divergence, especially

in high subsonic flight. The unstable diverging component is known as the tuck

mode because it manifests itself as a slow increase in speed and nose-down

pitch attitude. It is essentially a static instability due to the derivative of the

pitching moment with longitudinal perturbation in velocity being negative,

and as speed increases, the nose has a tendency to tuck under. An understand-

ing of the tuck mode is essential as it provides the primary mechanism for

tumbling about the pitch axis, a fundamental non-linear mode.

One may also summarise the governing reduced-order phugoid modal

equations as

−1

⎡ Δu s ⎤ ⎡ ⎡ xu xθ ⎤ ⎡ x w x q ⎤ ⎡ zw z q ⎤ ⎡ zu zθ ⎤ ⎤ ⎡ Δus ⎤

⎢ ⎥ = ⎢⎢ − ⎢ ⎥

⎢⎣ Δθs ⎥⎦ ⎢⎣ ⎣ 0 0 ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎣ mw mq ⎥⎦ ⎢⎣ mu mθ ⎥⎦ ⎥ ⎢⎣ Δθs ⎥⎦

⎦

−1

⎡ ⎡ xη xτ ⎤ ⎡ xw x q ⎤ ⎡ zw z q ⎤ ⎡ zη zτ ⎤ ⎤ ⎡ Δη⎤

+ ⎢⎢ − ⎥ . (6.111)

⎢⎣ ⎣ 0 0 ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎢⎣ mw mq ⎥⎦ ⎢⎣ mη mτ ⎥⎦ ⎥ ⎢⎣ Δτ ⎥⎦

⎦

The short period mode is a longitudinal pitch axis mode of free oscillations

that can have decaying or growing amplitude. In the case of statically stable

but uncontrolled aircraft, the oscillations have a short period and grow or

decay slowly. In the case of statically unstable aircraft, the growth is rapid,

as in a pilot-induced oscillation. In general, it is a coupled high-frequency

oscillation involving the velocity perturbation in the vehicle’s heave axis and

the pitch rate. The aerodynamic coupling plays a key role either as an energy

sink resulting in energy dissipation and a damped response or as a regenera-

tive energy source leading to instability:

⎡ zq ⎤ ⎡ zq ⎤

⎢1 1 − mq ⎥ ⎡ zw zq ⎤ ⎡ Δws ⎤ ⎢

1

1 − mq ⎥ ⎡ zη

⎡ Δw s ⎤ ⎢ zτ ⎤ ⎡ Δη⎤

⎢ Δq ⎥ = ⎢

⎥⎢ + ⎢ ⎥⎢ .

⎣ s⎦ 1 ⎥ ⎣ mw mq ⎥⎦ ⎢⎣ Δqs ⎥⎦ ⎢ 1 ⎥ ⎣ mη mτ ⎥⎦ ⎢⎣ Δτ ⎥⎦

⎢0 1 − mq ⎥⎦ ⎢0 1 − mq ⎥⎦

⎣ ⎣

(6.112)

For fighter aircraft, the centre of gravity (CG) is often located behind the

neutral point or the aerodynamic centre. When this happens, the aircraft

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 215

is either unstable or just stable. The characteristic equation has four real

roots. When the CG is moved further backwards, one pair of real roots

become a complex-conjugated pair which can be associated neither with

the phugoid nor with the short period. This is usually referred to as the

third oscillatory mode.

The roll subsidence mode is caused by the high roll damping effect of the

wing and is directly related to the slope of the lift coefficient versus the

effective angle-of-attack curve. This mode is also affected by sideslip motions

induced by aileron yaw:

⎡ Δξ ⎤

(1 − lp ) Δp s = lpΔps + ⎡⎣lξ lζ ⎤⎦ ⎢ ⎥ . (6.113)

⎣ Δζ ⎦

The Dutch roll is a coupled dynamic lateral and directional mode of motion.

The heading and sideslip angles in a Dutch roll are out of phase with each

other with the heading and sideslip motions being consistent with those in

a relatively flat yawing oscillation, implying that rolling component is rela-

tively less significant. However, the bank angle cannot be totally ignored

and it leads the sideslip and lags behind the yaw, indicating that the sideslip

follows the roll motion, which follows the yaw motion:

⎡ lr ⎤

⎢1 − y r + yp ⎥

lp Δv ⎡ y yr ⎤ 1 ⎡ yp ⎤ ⎤ ⎡ Δvs ⎤

⎢ ⎥ ⎡⎢ s ⎤⎥ = ⎢ ⎡⎢ v − ⎢ n ⎥ ⎡⎣lv lr ⎤⎦ ⎥ ⎢

⎢ l ⎥ Δr n nr ⎥⎦ lp ⎣ p⎦

⎥

⎥⎦ ⎣ Δrs ⎦

⎢0 1 − nr + r np ⎥ ⎣ s ⎦ ⎢⎣ ⎣ v

⎣ lp ⎦

⎡ ⎡ yξ yζ ⎤ 1 ⎡ yp ⎤ ⎤ ⎡ Δξ ⎤

+ ⎢⎢ − ⎢ n ⎥ ⎡⎣lξ lζ ⎤⎦ ⎥ ⎢ ⎥ (6.114)

⎢⎣ ⎣ nξ nζ ⎥⎦ lp ⎣ p⎦ ⎥⎦ ⎣ Δζ ⎦

6.3.3.6 Spiral

The spiral mode is a mode reflecting the coupling of yaw with sideslip and

roll. When an uncontrolled aircraft rolls, gravity causes it to sideslip and

consequently to yaw. The yaw motion again generates rolling moments that

could be considered to be an aerodynamic feedback. However, provided the

aerodynamic rolling moment is not regenerative (positive feedback due to

the fin, negative feedback due to the dihedral effect), the mode is stable. The

motion of the spiral is characterised by banking and turning motion with

www.ebook777.com

free ebooks ==> www.ebook777.com

216 Flight Dynamics, Simulation, and Control

almost negligible side acceleration. When unstable, the aircraft also descends

and the motion resembles a tightening spiral dive:

⎡ yv yp yr ⎤ ⎡ yξ yζ ⎤

−1

Δφ s

( y (l nr v − lv nr ) ) ⎢

0 ⎤⎦ ⎢ lv lp lr ⎥

⎥ ⎢ ⎥ ⎡ Δξ ⎤

Δφs + ⎡⎣0 1 ⎢ lξ lζ ⎥ ⎢ ⎥ = 0

φ

+

Δ Δζ

⎢⎣ nv np nr ⎥⎦ ⎢⎣ nξ nζ ⎥⎦ ⎣ ⎦

(6.115)

where

Δ = ⎡⎣ yv ( lp nr − lr np ) + y p ( lr nv − lv nr ) + y r ( lv np − lp nv ) ⎤⎦ ≈ y r ( lv np − lp nv ) .

(6.116)

For completeness, we observe that there are in addition two other modes

in the linear description of aircraft dynamics, which may be represented as

direct integration modes resulting in the height, in the longitudinal case and,

the heading, in the lateral case.

So far in this chapter, the calculation or estimation of the aerodynamic sta-

bility derivatives has not been discussed although it is an important aspect

in the assessment of the stability of an aircraft. In Chapter 2, some of the

important aerodynamic features of aerofoils were discussed. However, it is

now essential that some of the key finite span effects are presented in order

to discuss the estimation or calculation of the aerodynamic stability deriva-

tives relevant to the assessment of dynamic stability.

The fundamental feature of finite wings is that they induce the roll up

of the spanwise flow which in turn is responsible for the generation of two

powerful vortices that are shed by the wing at the wing tips. The roll up

of the spanwise flow results in an induced flow that effectively reduces the

angle of attack by a finite amount. The reduction in the angle of attack due

to the induced downwash is termed the induced angle of attack, αin, and can

be shown to be directly proportional to the lift coefficient, CL, and inversely

proportional to the wing aspect ratio, AR. Thus it can be expressed as

CL w

α in = ≈ in . (6.117)

πeAR U e

and αin a constant along the span, for a straight, unswept wing with an

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 217

sible in principle to adopt Prandtl’s lifting-line theory and show that αin is

almost constant in the spanwise direction. In Chapter 2, it was shown that

for an aerofoil, the lift coefficient may be expressed in terms of the angle

of attack as

CL = a∞ ( α − α 0 ) . (6.118)

CL = a∞ ( α − α 0 − α in ) = a∞ ( α − α 0 − win U e ) . (6.119)

⎛ ⎞

⎜ a∞ ⎟

CL = ⎜ ⎟ ( α − α 0 ) . (6.120)

a

⎜ 1+ ∞ ⎟

⎝ πeAR ⎠

a∞

a= . (6.121)

1 + a∞ πeAR

Prandtl assumed that, at each span station, all the vorticity over the wing

may be lumped at a single point, at the quarter-chord as a point vortex Γ. The

magnitude of Γ was not assumed to be a constant and was allowed to vary

depending on the spanwise station being considered. Thus, the strength of

the local circulation Γ is a function of y. Prandtl used the Biot–Savart law to

estimate the induced downwash. Thus, it is expressed as

y =+ b/2

1 1 dΓ

win = dy. (6.122)

4π ∫

y =− b/2

( y1 − y ) dy

Consequently, the lift coefficient CL is also expressed as

⎛ y =+ b/2 ⎞

2L ⎜ α − α0 − 1 1 dΓ ⎟ 2Γ

CL = a dy = . (6.123)

ρU e c

2

= ∞

⎜

⎝

4πU e ∫

y =− b/2

( y1 − y ) dy ⎟ Ue c

⎠

Prandtl solved this integral equation for Γ. Thus, the solution is correct only

for very large aspect ratio wings and for purposes of predicting the lift.

www.ebook777.com

free ebooks ==> www.ebook777.com

218 Flight Dynamics, Simulation, and Control

For low aspect ratios, AR → 0, low aspect ratio wing theory predicts a = πAR/2.

To ensure that the lift coefficient behaves exactly for both low and high

aspect ratios, the expression is modified empirically as

a∞ AR

CLα = a = , (6.124)

2

(

2 + 4 + ( 2πAR a∞ ) 1 − M∞2 + tan 2 Λ c/2 )

where Λc/2 is the wing mid-chord sweep angle.

Finally, the influence of compressibility may be included by generalising

the Prandtl–Glauert similarity rule for aerofoils which is

a∞ M = 0

a∞ ( M ) = . (6.125)

1 − M∞2

For large aspect ratio wings, the similarity rule was generalised by Von

Karman and Tsien (Shapiro [2]) as

a∞

a= . (6.126)

M∞2 a∞

1+

2 1 + 1 − M∞2

a∞

a= . (6.127)

γ −1 2

2 1+ M

M 2

∞

1+ ∞ a∞

2 1 + 1 − M∞2

Another empirical generalisation [4] for general aspect ratios takes the form

a∞ AR

a = K wb . (6.128)

2 2

⎛ 2π 1 − M∞2 AR ⎞ ⎛ ⎛ ⎞ ⎞

2+ 4+⎜ ⎟ ⎜ 1 + ⎜ tan Λ c/2 ⎟ ⎟

⎜ a∞ ⎟ ⎜ ⎜ 1 − M∞2 ⎟ ⎟

⎝ ⎠ ⎝ ⎝ ⎠ ⎠

In the equation, Kwb is a correction factor that is a function of the ratio of the

fuselage projected area to the wing projected area which varies between the

limits 0.93 and 1.0065. Finally, it is important to recognise that although

the expressions for the wing lift coefficient for finite aspect ratio wings were

discussed earlier, it is the wing pitching moment coefficient at the neutral

point that is important for stability calculations.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 219

The contribution of the horizontal tail (ht) to the aircraft’s lift curve slope

can be expressed as

⎛S ⎞ ⎛ dε ⎞

CLα ( ht ) = ⎜ ht ⎟ ηhte ⎜ 1 − dα ⎟ CLα , ht . (6.129)

⎝ Sw ⎠ ⎝ ⎠

In Equation 6.129, Sht/Sw is the ratio of the horizontal tail area to the wing sur-

face area, ηhte is the horizontal tail effectiveness factor, dε/dα is the derivative

of the downwash angle with respect to the wing angle of attack and CLα , ht is

the horizontal tail’s lift curve slope.

The contribution of the fuselage to the wing–body pitching moment is

affected by the interference by the wing flow field and can have a destabilis-

ing effect on aircraft. In accordance with slender-body theory, the distribu-

tion of lift along the fuselage may be expressed as

dL dS f

= ρU e2α , (6.130)

dx dx

where

S f = πwd2 4 is the cross-sectional area of the fuselage based on its width

wd is the fuselage width

x is the streamwise variable

Generally, the total lift of the fuselage is taken as zero while the lift distribu-

tion contributes to the pitching moment about the nose of the fuselage (x = 0)

which is found to be

xf xf

dL dS f

Mf = − x dx = −ρU e2α x dx. (6.131)

∫

dx

0

dx ∫

0

xf xf

M f = ρU α S f dx = ρU αVf , Vf =

∫ S dx. (6.132)

2 2

e

∫

0

e

0

f

2M f 2Vf

Cmf = = α. (6.133)

ρU e2Sw c Sw c

www.ebook777.com

free ebooks ==> www.ebook777.com

220 Flight Dynamics, Simulation, and Control

Thus, the pitching moment derivative with respect to the angle of attack is

increased, bringing the aircraft neutral point closer to the CG. Consequently,

the fuselage has a destabilising effect on the overall stability of the aircraft.

The interference effect of the wing flow field on the tail was modelled by

introducing the downwash angle, in Chapter 2, in evaluating the tail plane

lift. The downwash angle can be shown to be proportional to the induced

downwash and consequently can be expressed as

ε = κ CL πeAR , 0 ≤ κ ≤ 2. (6.134)

The upper limit corresponds to the case when the tail plane is relatively far

downstream but in the downwash field of the wing. For a high tail plane not

affected by the wing, κ = 0.

The downwash angle is generally assumed to be directly proportional to

the angle of attack and, rather than using Equation 6.134, is calculated from [4]:

dε 1.19

dα

(

= 2 × π K a K λ K H cos Λ c/2 ) , (6.135a)

Ka =

1

−

1

, Kλ =

10 − 3λ

, KH =

(1 − hH b ) . (6.135b)

AR 1 + AR1.7

7 3 2l b

t

In Equation 6.135b, AR is the wing aspect ratio, λ is the wing taper ratio, hH/b

is the ratio of the horizontal tail plane height above the fuselage centre line

to the wing span and lt/b is the ratio of the distance of the tail plane aerody-

namic centre from the wing aerodynamic centre to the wing span.

To estimate the increase in CLmax contributed by modern high-lift devices, the

following simple methodology may be employed. Firstly, the change in CLmax

is determined from

⎛ S flap ⎞

ΔCLmax = Δclmax ⎜⎜ ⎟⎟ cos ( Λ hl ) (6.136)

⎝ Sref ⎠

where

∆clmax is the increment in aerofoil maximum cl obtainable from the particu-

lar flap system

Sflap is the wing area spanned by the flaps

Λhl is the sweep angle of the hinge line

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 221

TABLE 6.8

Increase in Aerofoil Maximum Lift Coefficient

for High-Lift Devices

High-Lift Device ∆clmax

Trailing edge flaps

Plain 0.9

Single slotted 1.3

Fowler (single slotted) 1.3 (cf/c)

Double slotted 1.6 (cf/c)

Triple slotted 1.9 (cf/c)

Leading Edge Devices

Fixed slot 0.2

Leading edge droop 0.3

Kruger 0.3

Slat 0.4 (cf/c)

Source: Data from Raymer, D.P., Aircraft Design: A

Conceptual Approach, AIAA education series,

American Institute of Aeronautics and

Astronautics (AIAA), Washington, DC, 1989.

∆clmax can be obtained from several sources including test data. It may also be

approximated from Table 6.8. It should be noted that the table gives values

for the maximum increase attainable at the optimum angle of attack and flap

deflection. Because of the increased drag associated with such extreme flap

deployment (flap angular deflection equal to approximately 50°), this setting

is normally used for landing only. Take-off flap angles are usually in the

range of 25°, with ∆clmax equal to about 70% of the landing value.

It can be shown that induced drag is a component of the 3D lift in the drag

direction:

For αin small, sin αin = αin, and

CL C2

CDi = CLα in = CL = L . (6.138)

πeAR πeAR

The value of e the span efficiency factor is 1 for elliptical wings and between .5

and 1 for most common wing shapes. Oswald’s span efficiency factor is often

expressed as

1

e= (6.139)

k

www.ebook777.com

free ebooks ==> www.ebook777.com

222 Flight Dynamics, Simulation, and Control

0.15

( )

k = 4.61 1 − 0.045 AR0.65 ( cos Λ LE ) − 3.1. (6.140)

For an aircraft with multiple components, the value of k for the aircraft is

the sum of the values of k for the individual components. A wide variety of

methods for estimating Oswald’s span efficiency factor are discussed by Niţă

and Scholz [6].

The total drag of the wing is the sum of the profile drag and the induced drag:

CL 2

CD = CD 0 + . (6.141)

πeAR

A variety of devices have been used on aircraft to reduce induced drag. The

winglet is the most effective and most widely used of these devices. In addi-

tion, jet fighter aircraft which carry additional fuel tanks or stores on their

wing tips experience a small reduction in induced drag when such wingtip

stores are in place. All of these devices inhibit the formation of the wingtip

vortices and therefore reduce downwash and induced drag.

Recall, however, from Chapter 3 that profile drag is composed of skin fric-

tion drag and the form drag. The primary contribution to the profile drag is

due to skin friction. An element on the surface of an aerofoil, in a flow field,

experiences shear stress tangential to the surface and a pressure normal to

it. The shear stress multiplied by the area of the element gives the tangential

force. The component of this tangential force in the free-stream direction

when integrated over the profile gives the skin friction drag. In general, the

equivalent flat plate area of the ith component of the skin friction drag coef-

ficient can be computed from

Swet

CD0 = C f FQ , (6.142)

Sref

where

F is the form or shape factor

Q is the interference factor

Cf is the body-averaged skin friction coefficient

Sref = Sw, the wing planform area

grating the surface (wall) shear stresses over the entire wetted surface area.

Alternatively, the friction coefficient, which is the wall shear stress non-

dimensionalised by the dynamic pressure, may be integrated over the entire

wetted surface area to estimate the skin friction drag coefficient.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 223

The skin friction coefficient for a turbulent boundary layer (Re > 107) is gen-

erally found experimentally. The results have been stated in the literature

with approximated curves fitted to the data. Two such popular approxima-

tions, the accurate logarithmic curve fit due to Von Karman which is valid

for a larger range of Reynolds numbers and the power law fit for the case of

incompressible flow are given, respectively, by

0.455 0.074

Cf = and C f = , (6.143a)

⎡( log Re )2.58 ⎤ ⎡( Re )0.2 ⎤

⎣ ⎦ ⎣ ⎦

( C f )incompressible

Cf = 0.65

, (6.143b)

⎡ 1 + 0.144 M 2 ⎤

(

⎢⎣ ) ⎥⎦

although the Spalding and Chi [7] approximation employed in the Engi

neering Services Data Unit (ESDU) datasheets [8,9] is definitely a better

approximation.

For the laminar case (Re < 5 × 105), including the compressibility cor-

rection, it is

1.328

Cf = 0.5 0.12 . (6.143c)

( Re ) (1 + 0.12M ) 2

For wings, the Reynolds number is based on the exposed portion of the

mean aerodynamic chord, while for fuselages or nacelles, it would be

the length. The flow field around most aircraft is largely turbulent, so

the friction coefficient can be computed from the formula for turbulent

boundary layers.

The form factor, F, is a measure of how streamlined a particular shape is. It

therefore has a major influence on the profile drag since thin bodies exhibit

lower adverse pressure gradients and, therefore, less boundary layer thick-

ening near the trailing edge. The form factor is a function of the component

thickness-to-length ratio. For wings, this function is the thickness-to-chord

ratio, t/c. In general, the lower the thickness ratio, the lower the form factor,

though some shapes (blunt trailing edges) have higher pressure drag than

others. The following gives the form factor for a typical wing:

t ⎛ t ⎛ t ⎞⎞

Fwing = ⎡⎣( F * − 1) cos 2 Λ 0.5c ⎤⎦ + 1, F * = 1 + max ⎜ 3.3 − max ⎜ 0.008 − 27 max ⎟ ⎟ .

c ⎝ c ⎝ c ⎠⎠

(6.144a)

www.ebook777.com

free ebooks ==> www.ebook777.com

224 Flight Dynamics, Simulation, and Control

TABLE 6.9

Form Factor Estimation

Name LW f2 F0∗

Hoerner [11] 2 60 1

Torenbeek [12] 2.7 100 1

⎛x⎞ 0.28

Nicolai [13,14], Raymer [5] 0.6 max ⎜ ⎟

⎝c⎠

100 (

1.34 M 0.18 cos ( Sw _ max_ thickline ) )

For unswept low speeds 2 100 1

(x/c) for max. thick. = 0.3

Source: Data from Gur, O. et al., J. Aircraft, 47(4), 1356, 2010.

In general, there are several models of the form factor based on the general

expression

2

⎧⎪ t ⎛ t ⎞ ⎫⎪

F * = F0* ⎨1 + LW max + f 2 ⎜ max ⎟ ⎬ (6.144b)

⎩⎪ c ⎝ c ⎠ ⎪⎭

The typical expression for estimating the form factor of a fuselage body of

length l and depth d is

−3

⎧⎪ l ⎛ l ⎞ ⎫⎪

F * = ⎨1 + 0.0025 + 60 ⎜ ⎟ ⎬ . (6.145)

⎩⎪ d ⎝ d ⎠ ⎭⎪

The profile drag depends not only on the component size and shape but also

on aerodynamic interference between the component and its surrounding

components. For example, the dynamic pressure can be increased or reduced

at a junction between a wing and winglet surface, which alters the drag of

the winglet relative to its isolated drag. Interference factors tend to have val-

ues ranging from about 1 for the fuselage and well-filleted wing to about 1.5

for fuselage-mounted nacelles.

The air can only influence and be influenced by surfaces that it touches, so

the relevant area over which the friction or the pressure will act is the wetted

area – the actual area exposed to the air. This value is completely determined

by the geometry of the aircraft and, in actuality, is quite difficult to estimate.

A simple assumption that is often employed is

Other contributions to the profile drag are the flap effects, windmilling,

engine and propeller effects, base drag effects and the effects of leakages

and protuberances.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 225

To assist the estimation of stability derivatives, an accepted methodology

is to try and relate the unknown derivatives to a minimal set of aerody-

namic coefficients. These aerodynamic coefficients are obtained from the

fundamental aerodynamic forces and moments that can be derived in the

wind axes and then non-dimensionalised according to standard practice. To

assist this process, a number of equilibrium or trim conditions are defined in

the first instance. The trim equations may then be used to reduce the num-

ber of independent unknown aerodynamic coefficients to a small number.

Generally, this basic set of coefficients is chosen to be those which can either

be computed (theoretically or numerically) or experimentally measured with

some degree of reliability.

We shall briefly discuss this process considering the case of uniform level

trimmed flight with the controls fixed.

Consider the case uniform level flight with the further assumption that θ

and iE are initially zero. The conditions of equilibrium flight are

0

Now, considering the case of equilibrium flight with the climb angle equal

to zero, that is, when αE = θ and θ is small but non-zero, the net perturbation

forces in the two directions, in terms of the perturbation lift and drag, are

ρV 2Sw

X = L sin ( α E ) − D cos ( α E ) + T = ( CL sin ( αE ) − CD cos ( αE ) + CT )

2

(6.147a)

ρV 2Sw

Z = −L cos ( α E ) − D sin ( α E ) = − ( CL cos ( α E ) + CD sin ( α E ) ) . (6.147b)

2

1

M= ρV 2Sw cCm (6.148)

2

The lift coefficients are then expressed as the sum of three components:

www.ebook777.com

free ebooks ==> www.ebook777.com

226 Flight Dynamics, Simulation, and Control

⎛ ∂C ⎞ ⎛ ∂C ⎞

CMac = ( C Mac )wing + ⎜ Mac ⎟ α fuselage + ⎜ Mac ⎟ αT . (6.149c)

⎝ ∂α ⎠ fuselage ⎝ ∂α ⎠tailplane

The tail plane angle of attack is usually eliminated from the equations as it may

be expressed in terms of wing angle of attack, as defined in Equation 2.46.

To find the expression stability derivatives, the previous equations for X,

Y and M are differentiated with respect to the perturbation degrees of free-

dom. Thus, the stability derivatives with respect to the forward velocity per-

turbation may be found this way, and we have

1 ∂X ⎛ ∂CL ∂θ ∂CD ∂θ ⎞

=⎜ θ + CL − + CD θ ⎟

1

ρV 2Sw ∂u ⎝ ∂u ∂u ∂u ∂u ⎠

2

2 ∂V 1 ∂T

+ ( C L θ − CD ) + 1 . (6.150)

V ∂u ρV 2Sw ∂u

2

1 ∂X ∂C 1 ∂T

= −V D − 2CD + . (6.151)

1 u u 1

ρVSw ∂ ∂ ρVSw ∂u

2 2

In terms of the thrust coefficient and including the effect of the steady

trimmed component of αE, which is assumed to be small,

1 ∂X ∂C ∂C

= −V D − 2CD + 2CT + 2CLα E + V T . (6.152)

1 ∂u ∂u ∂u

ρVSw

2

1 ∂Z ∂C

= −V L − 2CL − 2CDα E . (6.153)

1 ∂u ∂u

ρVSw

2

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 227

may be noted that

∂ ∂ M∞2 ∂

V ≈ M∞ =− . (6.154)

∂u ∂M∞ 1 − M∞2 ∂ 1 − M∞2

∂CD ∂C 2CL ∂C

V = V D0 + V L . (6.155)

∂u ∂u πeAR ∂u

∂CL M∞2

V ≈ CL . (6.156)

∂u 1 − M∞2

The derivative of the lift coefficient with respect to the free-stream Mach

number may be evaluated from the expressions for the lift curve slope given

in the preceding section.

Furthermore, in the case of a modern jet-powered constant thrust engine,

T

CT = . (6.157)

1

ρV 2S

2

P

CT = . (6.158)

1

ρV 3S

2

Thus,

∂CT

V ≈ −nCT (6.159)

∂u

where

n = 2 for a constant thrust engine

n = 3 for a constant power engine

www.ebook777.com

free ebooks ==> www.ebook777.com

228 Flight Dynamics, Simulation, and Control

Further, since ∂V/∂w ≈ 0, ∂w/∂α ≈ V and ∂T/∂w = 0, we may also show that

1 ∂X ∂C

= − D + CL (6.161)

1 ∂w ∂α

ρVSw

2

and that

1 ∂Z ∂C

= − L − CD . (6.162)

1 ∂w ∂α

ρVSw

2

To find the derivative of the lift coefficient with respect to angle of attack

(this derivative is generally extremely important), a number of corrections

for wing aspect ratio, compressibility effects and partial or full span flaps are

made. The derivative of the drag coefficient with respect to the angle of attack

is made up of two contributions: one due to the profile drag and the other due

to the induced drag. For these purposes, the drag coefficient is expressed as

CL2

CD = CD0 + CDi , CDi = (6.163)

eπA

where e = 1/k = 1/(1 + δ) is Oswald’s span efficiency factor, and it follows that

∂CD C ∂CL

=2 L . (6.164)

∂α eπA ∂α

flight with respect to the perturbation velocity components, we may express

the derivative of the pitching moment coefficient as

1 ∂M 2 ∂C

= Cm + m . (6.165)

1

ρV 2Sw c ∂u V ∂u

2

To include the effect of the variation of thrust with velocity, we consider the

moment of this derivative and modify the previous pitching moment deriva-

tive as

1 ∂M 2 ∂C mr c ∂CT

= Cm + m + T (6.166)

1 ∂u V ∂u I yy ∂u

ρV Sw c

2

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 229

When the equilibrium value of Cm is equal to zero, the derivative of the pitch-

ing moment is obtained from the derivative of the pitching moment coefficient.

The moment coefficient, at the aircraft aerodynamic centre, is expressed as

∂CL

CMac = − ( H n ) α (6.167)

∂α

where

−∂CMac

Hn = (6.168)

∂CL

is the stability margin. The moment coefficient is also split up into three com-

ponents for the wing, fuselage and tail plane and summed up. Hence,

1 ∂Mac ∂C Mac

= . (6.169)

1

ρVSw c ∂w ∂α

2

separately consider the contributions due to the wing and tail plane. The tail

plane perturbation angle of attack due to steady pitch rotation may be eas-

ily estimated. The downwash at the tail plane due to the rotation in pitch is

approximately given by w = ltθ = ltq, and therefore, the tail plane perturbation

angle of attack is αT = qlt/V. Hence, the derivative

∂αT lt

= . (6.170)

∂q V

1

X = − ρV 2ST CDT . (6.171)

2

Hence,

1 ∂C S l qc

X = − ρV 2Sw DT T t . (6.172)

2 ∂αT SW c V

ST lt

Thus, in terms of the tail plane volume ratio, VT = × and q,* where

Sw c

qc 1 ∂X ∂C

q= , = −VT DT . (6.173)

V 1 ∂ q ∂αT

ρV Sw

2

www.ebook777.com

free ebooks ==> www.ebook777.com

230 Flight Dynamics, Simulation, and Control

Similarly,

1 ∂Z ∂CLT

= −VT (6.174)

1

ρV 2Sw ∂q ∂αT

2

and

1 ∂Mac l ∂CLT

CMac , q = = −VT t . (6.175)

1 ∂ q c ∂αT

ρV Sw

2

ing in pitch to the aircraft.

In the case of perturbations in acceleration-related degrees of freedom, the

downwash lag effect may be employed to estimate the derivative of the tail

plane angle of attack with respect to the normal acceleration at the wing. The

tail plane is assumed to be immersed in the downwash field. However, when

there is change in the wing angle of attack, this change is felt at the tail plane

after a certain time delay which is equal to the time the disturbance takes to

transport to the tail plane. Thus, the current downwash at the tail plane is

related to the angle of attack changes at some previous time.

To estimate the downwash lag, we consider the downwash and expand it

in a Taylor’s series in terms of the wing angle of attack α = αw as follows:

2

∂ε ( t − Δt ) ⎛ ( Δt ) ⎞

ε ( t − Δt ) = ε0 + ⎜ α − Δtα +

− ⎟ . (6.176)

α

∂α ( t − Δt ) ⎜ 2 ⎟

⎝ ⎠

ing that Δt = lt/V, it follows that Δτ = 2lt c , and the expression for the down-

wash at the tail plane may be written as

2

∂ε ⎛ 2lt 1 ⎛ 2l ⎞ ⎞

ε = ε0 + ⎜α − αʹ + ⎜ t ⎟ αʹʹ − ⎟ (6.177)

⎜

∂α ⎝ c 2⎝ c ⎠ ⎟

⎠

∂ε ∂

where ε0 = − i0 , as there is no downwash at zero lift and ( ʹ ) = .

∂α ∂τ

But it is known from Equations 2.43 and 2.44 that

αT = α w − iw + iT − ε, (6.178)

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 231

2

⎛ ∂ε ⎞ ∂ε ⎛ 2lt 1 ⎛ 2l ⎞ ⎞

αT = ( α w − i0 ) ⎜ 1 − ⎟ + iT + i0 − iw + ⎜ αʹ − ⎜ t ⎟ αʹʹ ⎟ . (6.179)

⎝ ∂α ⎠ ∂α ⎜⎝ c 2⎝ c ⎠ ⎟

⎠

coefficient,

dCm ⎞ ⎛ dCL ⎞ lt ST

( Cm )tail = ⎛⎜ ⎟tail αT = − ⎜ ⎟ αT , (6.180)

⎝ d α ⎠ part ⎝ dα ⎠tail c Sw

part

= = Cmαʹ = − ( CLα )tail t T × = 2CMac , q . (6.181)

dαʹ dαT ∂αʹ c Sw ∂αT c ∂α

= = Cmαʹ = CMac , q = . (6.182)

1 2 w 1

2 2 ρV Sw ∂q ∂α

∂ ∂αʹ ∂α

ρ c Sw 2

2 2

are given by

1 ∂X 1 ∂X ∂ε

= (6.183)

1 w 1

ρV cSw ∂q ∂α

∂

ρ cSw

2 2

and

1 ∂Z 1 ∂Z ∂ε

= . (6.184)

1 w 1

ρV cSw ∂q ∂α

∂

ρ cSw

2 2

ing forces and the pitching moment with respect to the velocity components

u and w, the pitch rate and the normal acceleration w have been related to

aerodynamic coefficients in Equations 6.152, 6.153, 6.161, 6.162, 6.166, 6.169,

6.173 through 6.175 and 6.182 through 6.184.

www.ebook777.com

free ebooks ==> www.ebook777.com

232 Flight Dynamics, Simulation, and Control

To obtain the lateral derivatives, we consider the sideslip angle, β, in Figure 6.1

and consider the forces and moments acting on an aircraft in steady sideslip

(Figure 6.2). The aircraft is in equilibrium in the direction of the sideslip

velocity. Hence, the force balance equation in the y direction is

Y = mg sin ( φ ) . (6.185)

The side force due to the wing or the lateral drag is mainly due to the fin,

the fuselage, the wing dihedral and the engine nacelles. The perturbation

sideslip force is assumed to be mainly contributed by the fuselage and the

vertical tail (fin). Thus, the side force, Y, is

1 1 ⎛ dC ⎞

Y= ρV 2SsCYB β − ρV 2Sv ⎜ L ⎟ β (6.186)

2 2 ⎝ dα ⎠vt −tail

where

Ss is the body side area

Sv is the area of the vertical tail

CYB is a body side force coefficient

introduce the concept of induced sidewash to allow for the interference of the

V, velocity of

relative wind

FIGURE 6.1

Definition of sideslip angle.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 233

N

v

mg

FIGURE 6.2

Aircraft in steady sideslip showing the sideslip velocity, v, roll angle, ϕ, and relevant moments.

wing wake and the vertical tail. Thus, the sidewash angle, σ, induced by the

wake is assumed to be given as

⎛ ∂σ ⎞

σ=⎜ ⎟ β. (6.187)

⎝ ∂β ⎠

where ∂σ/∂β is a constant. From the properties of vortex flows, it is expected

that this effect would effectively reduce the sideslip angle. Thus, the side force, Y,

1 1 ⎛ dC ⎞ ⎛ ∂σ ⎞

Y= ρV 2SsCYB β − ρV 2Sv ⎜ L ⎟ ⎜1− ⎟ β. (6.188)

2 2 ⎝ dα ⎠vt −tail ⎝ ∂β ⎠

Ss S ⎛ dC ⎞ ⎛ ∂σ ⎞

CYβ = CYB − v ⎜ L ⎟ ⎜1− ⎟ . (6.189)

Sw Sw ⎝ dα ⎠vt −tail ⎝ ∂β ⎠

The rolling moment derivative due to sideslip, with the controls fixed, is due

to the cumulative effect of the dihedral of the wings, the wing and the ver-

tical tail. Fuselage effects may be ignored in the first instance. The rolling

moment due to a sideslip perturbation is obtained by modelling the wing as

a number of spanwise strips and is

⎡s⎛ ⎤

1 2 ⎛ 2v ⎞ ⎞

L = − ρV ⎜ ⎟ ⎢ ⎜ ( Clα )local Γ + 2CL tan Λ1 4 ⎟ c ( y ) ydy ⎥ . (6.190)

2 ⎝ V ⎠⎢0 ⎝

⎣

∫ wing ⎠ ⎥

⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

234 Flight Dynamics, Simulation, and Control

The contribution of the vertical tail (fin) to the rolling moment is obtained

by taking moments of the side force on the vertical tail about the roll axis

and is

1 z ⎛ ∂σ ⎞ v SV lv

Lvt −tail = − ρV 2Sb ( Clα )vt −tail Vvt −tail v ⎜ 1 − ⎟ , Vvt −tail = (6.191)

2 lv ⎝ ∂β ⎠ V Sb

where

Vvt −tail is the vertical tail volume ratio

zv/lv is the ratio of the distance of centre of pressure (CP) of the vertical tail

from the aeroplane’s x-axis to the horizontal distance from the aircraft’s

CG to the vertical tail quarter-chord

The yawing moment due to the sideslip is entirely due to the vertical tail

and is obtained by taking moments of the side force on the vertical tail about

the yaw axis:

1 ⎛ ∂σ ⎞ v

N= ρV 2Sb ( Clα )vt −tail Vvt −tail ⎜ 1 − ⎟ . (6.192)

2 ⎝ ∂ ⎠V

To estimate the rolling moment due to roll rate, one adopts 3D lifting sur-

face or strip theories. To estimate the side force and yawing moment deriva-

tives, one may either consistently adopt 3D lifting surface or strip theories or

exploit the fact that these derivatives are mainly estimated from the vertical

tail loads and modify the yawing moment due to sideslip to estimate them.

Similar options are available to estimate the yaw rate derivatives.

The lateral derivates due to roll rate are the side force derivative, the rolling

moment derivative and the yawing moment derivative. Considering first the

side force due to the roll rate, one first obtains the incidence due to a steady

rate of roll at any vertical station along the vertical tail as

pz

α ≈ tan ( α ) = . (6.193)

V

Considering a chordwise strip along the vertical tail, the increment in the

side force due to the strip is

1 pz

δY = − ρV 2 ( Clα )vt −tail c ( z ) dz. (6.194)

2 V

zF

1 p

Y = − ρVSb ×

2 Sb ∫ (C

0

)

lα vt − tail c ( z ) zdz. (6.195)

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 235

Considering rolling moment due to the roll rate, one first obtains the inci-

dence due to a steady rate of roll at any spanwise station along the wing as

py

α ≈ tan ( α ) = . (6.196)

V

Considering a chordwise strip along the right wing, the increment in the

vertical force in terms of the total lift and drag increments on the strip is

⎛ py ⎞ ⎛ py ⎞ ⎛ py ⎞

δZ = −δLift cos ⎜ ⎟ − δD sin ⎜ ⎟ = −δLift − δD ⎜ ⎟ (6.197a)

⎝ V ⎠ ⎝ V ⎠ ⎝V ⎠

where

1 ⎛ py ⎞

δLift = ρV 2Clα ⎜ α e + ⎟ c ( y ) dy (6.197b)

2 ⎝ V ⎠

and

1

δD = ρV 2CDc ( y ) dy. (6.197c)

2

Hence, considering the incremental rolling moment due to the roll rate, it is

given by

1 ⎛ py ⎞

δL = − ρV 2 ⎜ Clα α e + ( Clα + CD ) ⎟ c ( y ) ydy. (6.198)

2 ⎝ V ⎠

Considering only the contribution due to the roll rate and integrating over

the entire wing span,

s

1 2p

L = − ρV 2Sb 2 ×

2 VSb 2 ∫ (C

0

lα + CD ) c ( y ) y 2dy. (6.199)

Considering yawing moment due to the roll rate, one first obtains the inci-

dence due to a steady rate of roll at any spanwise station along the wing

which is as mentioned before. Then considering a chordwise strip along the

right wing, the increment in the axial force in terms of the total lift and drag

increments on the strip is

⎛ py ⎞ ⎛ py ⎞ ⎛ py ⎞

δX = δLift sin ⎜ ⎟ − δD cos ⎜ ⎟ = δLift ⎜ ⎟ − δD (6.200)

⎝V ⎠ ⎝V ⎠ ⎝V ⎠

www.ebook777.com

free ebooks ==> www.ebook777.com

236 Flight Dynamics, Simulation, and Control

where

1 ⎛ py ⎞

δLift = ρV 2Clα ⎜ α e + ⎟ c ( y ) dy. (6.201)

2 ⎝ V ⎠

Since it is only the differential drag between the right and left wings that

gives rise to the yawing moment, we write the drag increment as

1 ∂C ⎛ py ⎞

δD = ρV 2 D ⎜ α e + c ( y ) dy. (6.202)

2 ∂α ⎝ V ⎟⎠

Hence, considering the incremental yawing moment due to the roll rate is

1 ⎛ ∂C ⎞ ⎛ py ⎞

δN = − ρV 2 ⎜ Cl − D ⎟ ⎜ ⎟ c ( y ) ydy. (6.203)

2 ⎝ ∂α ⎠ ⎝ V ⎠

Considering only the contribution due to the roll rate and integrating over

the entire wing span,

s

1 2p ⎛ ∂CD ⎞

N = − ρV 2Sb 2 ×

∫ ⎜⎝ C − ∂α ⎟⎠ c ( y ) y dy. (6.204)

2

l

2 VSb 2

0

The derivation of the expressions for the side force, rolling and yawing

moments due to a steady yaw rate is similar to those due to a steady roll rate.

Considering first the side force due to the yaw rate, one first obtains the

incidence due to a steady rate of yaw at any vertical station along the vertical

tail as

rlv

α ≈ tan ( α ) = (6.205)

V

where lv is the horizontal moment arm of the vertical tail about the centre of

rotation in yaw.

To evaluate the side force due to the yaw rate, considering a chordwise

strip along the vertical tail, the increment in the side force on the strip is

1 rl

δY = ρV 2 ( Clα )vt −tail v c ( z ) dz. (6.206)

2 V

zF

1 rl

Y= ρVSb × v

2 Sb ∫ (C

0

)

lα vt − tail c ( z ) dz. (6.207)

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 237

Thus,

1 S l 1

Y= ρVSb ( Clα )vt −tail × V v r = ρVSb ( Clα )vt −tail × Vvt −tail r. (6.208)

2 Sb 2

For estimating rolling moment due to the yaw rate, consider a chordwise

strip along the right wing. The velocity of the strip is given by

Vr = V − ry , (6.209)

1 2

δLift = ρ (V − ry ) CLy c ( y ) dy. (6.210)

2

Hence, considering the incremental rolling moment due to the yaw rate, it is

given by

1 ⎛ 2ry ⎞

δL = − ρV 2 ⎜ 1 − ⎟ CLy c ( y ) ydy. (6.211)

2 ⎝ V ⎠

Considering only the contribution due to the yaw rate and integrating over

the entire wing span,

s

1 4r

L= ρVSb 2 × 2

2 Sb ∫C

0

Ly c ( y ) y 2dy. (6.212)

When the lift coefficient, CLy , is a constant, the integral on the right-hand side

depends purely on the geometry of the wing and may be evaluated exactly.

For a rectangular wing, the expression reduces to

1 CL

L= ρVSb 2 × y r. (6.213)

2 6

Additionally, one may compute the rolling moment generated by the fin

from the expression for the side force. Thus, the additional rolling moment

due to the fin is

1

Lr − fin = ρVSb ( Clα )vt −tail × Vvt −tail zv r (6.214)

2

where zy is the distance of the CP of the vertical tail from the aeroplane’s x-axis.

www.ebook777.com

free ebooks ==> www.ebook777.com

238 Flight Dynamics, Simulation, and Control

For estimating yawing moment due to the yaw rate, again consider a chord-

wise strip along the right wing. The velocity of the strip is given as before by

Equation 6.209. The increment in the drag force on the strip is

1 2

∂D = ρ (V − ry ) CDy c ( y ) dy. (6.215)

2

Hence, considering the incremental yawing moment due to the yaw rate, it

is given by

1 ⎛ 2ry ⎞

∂N = ρV 2 ⎜ 1 − ⎟ CDy c ( y ) ydy. (6.216)

2 ⎝ V ⎠

Considering only the contribution due to the yaw rate and integrating over

the entire wing span,

s

1 4r

N = − ρVSb 2 × 2

2 Sb ∫C

0

Dy c ( y ) y 2dy. (6.217)

As in the case of the rolling moment, the contribution of the vertical tail or

fin could easily be added by taking moments of the side force with respect to

the centre of rotation in yaw. Thus,

s

1 4r 1

N = − ρVSb 2 × 2

2 Sb ∫C

0

Dy c ( y ) y 2dy − ρVSb ( Clα )vt −tail × Vvt −taillv r. (6.218)

2

Trimmed flight and conditions for trimmed flight are extremely important

to determine steady, stable, equilibrium states of flight. An equilibrium state

may be associated with any manoeuvre, and for this reason, it is important to

understand the basic manoeuvres associated with aircraft flight. In this section,

the various equilibrium or trim conditions for only the primary longitudinal

and lateral flight manoeuvres are also considered, and valid small perturba-

tion equations are established from consideration of the dynamics of the flight.

These equilibrium states are assumed stable here. However, it is also often

essential to establish conditions of stability and ensure that they are met before

employing the conditions of trimmed flight in any other design calculations.

The principal longitudinal equilibrium flight conditions are

2. Steady climb or descent

3. Transition pull up manoeuvre from level flight to climb or descent

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 239

2. Force equilibrium, z-axis mg cos ( θ ) = T sin ( i ) + ∑ L cos ( θ ) + D sin ( θ )

E

3. Moment equilibrium

a. About tail plane AC (including contributions from Drag which

could be ignored under certain circumstances)

(6.219)

b. About aircraft CG

where

θ = α + γ is the pitch angle

α is the angle of attack

γ is the climb angle or flight path angle

iE is the angle of incidence of engine thrust line to aircraft x-axis

lw, lT are the horizontal distances of the wing and tail plane aero-

dynamic centres, respectively, to the aircraft CG

lT is known as the tail moment arm

correspondingly, there is a different volume ratio. It may be noted that lt ≈ lT,

zwt c are vertical heights of the tail plane AC to aircraft CG (positive up) and

zDw c , zDp c are vertical heights of the wing and parasite drag component centres.

The contribution of the thrust vector to the nose-up pitching moment is

ignored as the moment arm is generally small and negligible. It is usual to

assume that the thrust vector is aligned with the x-axis, that is, iE = 0.

Alternately and referring to Figure 6.3, we may write the conditions for

force equilibrium as follows:

T cos ( θtrim ) = ∑ D (6.221)

mg = T sin ( θtrim ) + ∑ L. (6.222)

www.ebook777.com

free ebooks ==> www.ebook777.com

240 Flight Dynamics, Simulation, and Control

lift, Lw

x axis Wing

drag, Dw

Parasite

Mac drag, Dp

α Tail

lift, LT

γ

Thrust, T

Horizontal

z axis Weight, mg

FIGURE 6.3

Forces acting on aircraft in flight relative to wind axes.

From these equilibrium conditions, one may establish formulae for θtrim =

αtrim as well as the elevator angle ηtrim. Based on this, one may define control

margins such as the elevator trim margin.

In the case of steady climb or descent, the force equilibrium conditions are

modified to the following:

Ttrim cos ( αtrim ) = mg sin ( γ trim ) + ∑D trim (6.223)

mg cos ( γ trim ) = Ttrim sin ( αtrim ) + ∑L trim . (6.224)

vre from level flight to climb or descent may be expressed in terms of

the trim angles and forces and perturbations to these angles and forces.

Similar to the elevator trim margin, one may define a corresponding control

margin, the elevator manoeuvre margin and the so-called manoeuvre point

where the elevator manoeuvre margin is equal to zero. For a positive elevator

manoeuvre margin, the manoeuvre point must be located aft of the centre of

mass. For a detailed discussion of these aspects, the reader is referred to

the book by Hancock [15].

Considering the transition pull up manoeuvre from level flight to climb or

descent and assuming that these trim conditions are small perturbations to

those corresponding to steady level flight, we may use the small perturba-

tion equations to solve for these perturbation states. Thus, considering the

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 241

control surface deflections,

⎡ Xu −g Xw 0 ⎤ ⎡ Δu ⎤ ⎡ Xη Xτ ⎤ ⎡0 ⎤

⎢ 0 0 0 1 ⎥⎥ ⎢⎢ Δθ ⎥⎥ ⎢⎢ 0 0 ⎥ ⎡ Δη⎤ ⎢⎢ 0 ⎥⎥

⎥

⎢ + = (6.225)

⎢ Zu 0 Zw U es ⎥ ⎢ Δw ⎥ ⎢ Zη Zτ ⎥ ⎢⎣ Δτ ⎥⎦ ⎢ 0 ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢⎣ Mu 0 Mw Mq ⎥⎦ ⎣ qB ⎦ ⎣ Mη Mτ ⎦ ⎣0 ⎦

⎡ Xu −g X w ⎤ ⎡ Δu ⎤ ⎡ Xη Xτ ⎤

⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎡ Δη⎤

⎢ Zu 0 Zw ⎥ ⎢ Δθ ⎥ = − ⎢ Zη Zτ ⎥ ⎢ ⎥ . (6.226)

Δτ

⎢⎣ Mu 0 Mw ⎥⎦ ⎢⎣ Δw ⎥⎦ ⎢⎣ Mη Mτ ⎥⎦ ⎣ ⎦

⎡ Δu ⎤ ⎡ Zu Zw ⎤ ⎡ Zη Zτ ⎤ ⎡ Δη⎤

−1

⎢ Δw ⎥ = − ⎢ M Mw ⎥⎦ ⎢⎣ Mη Mτ ⎥⎦ ⎢⎣ Δτ ⎥⎦

⎣ ⎦ ⎣ u

1 ⎡ Mw − Zw ⎤ ⎡ Zη Zτ ⎤ ⎡ Δη⎤

=− ⎢−M (6.227)

( Zu Mw − Zw Mu ) ⎣ u Zu ⎥⎦ ⎢⎣ Mη Mτ ⎥⎦ ⎢⎣ Δτ ⎥⎦

and

−1

⎡ Δη⎤ ⎡ Zη Zτ ⎤ ⎡ Zu Zw ⎤ ⎡ Δu ⎤

⎢ Δτ ⎥ = − ⎢ M Mτ ⎥⎦ ⎢⎣ Mu Mw ⎥⎦ ⎢⎣ Δw ⎥⎦

⎣ ⎦ ⎣ η

1 ⎡ Mτ − Zτ ⎤ ⎡ Zu Zw ⎤ ⎡ Δu ⎤

=− , (6.228)

( η τ τ η ) ⎢⎣ − Mη

Z M − Z M Zη ⎥⎦ ⎢⎣ Mu Mw ⎥⎦ ⎢⎣ Δw ⎥⎦

1 T

Δθ = ⎡ Xu Xw Xη X τ ⎤⎦ ⎡⎣ Δu Δw Δη Δτ ⎤⎦ . (6.229)

g⎣

Thus, given the desired increments in the forward and normal velocity com-

ponents, one can estimate the increments in the elevator setting angle and

the throttle setting.

www.ebook777.com

free ebooks ==> www.ebook777.com

242 Flight Dynamics, Simulation, and Control

DYNAMICS IN HORIZONTAL LEVEL FLIGHT AT AN

ALTITUDE OF 20,000 ft; FLIGHT VELOCITY, Ue, OF

830 ft/s (MACH 0.8) AND WEIGHT OF 637,000 lb

The governing longitudinal coupled equations of motion are

⎢ ⎥ ⎢ 0 ⎥⎥ ⎢⎢ w ⎥⎥ ⎢⎢ 32.7 ⎥⎥

d ⎢ w ⎥ ⎢ −0.0941 −0.624 820

= − η

dt ⎢ q ⎥ ⎢ −0.000222 −0.00153 −0.668 0 ⎥ ⎢ q ⎥ ⎢ 2.08 ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣θ⎦ ⎣ 0 0 1 0 ⎦⎣θ⎦ ⎣ 0 ⎦

(6.230)

The increment in the elevator setting angle for a decrease in the pitch

attitude by 8° (0.14 rad) is to be determined.

Consider a unit increase in the elevator setting angle. The corre-

sponding increments in the forward and normal perturbation velocity

components are given by

⎡u⎤

⎡ −0.0941 −0.624 820 0 ⎤ ⎢⎢ w ⎥⎥ ⎡ 32.7 ⎤

⎢ −0.000222 = η (6.231)

⎣ −0.00153 −0.668 0 ⎥⎦ ⎢ 0 ⎥ ⎢⎣ 2.08 ⎥⎦

⎢ ⎥

⎣θ⎦

and

−1

⎡ u ⎤ ⎡ −0.0941 −0.624 ⎤ ⎡ 32.7 ⎤

⎢ w ⎥ = ⎢ −0.000222 η. (6.232)

⎣ ⎦ ⎣ −0.00153 ⎥⎦ ⎢⎣ 2.08 ⎥⎦

1 ⎡u⎤

θ= ⎡ −0.00643 0.0263 ⎤⎦ ⎢ ⎥ . (6.233)

32.2 ⎣ ⎣w ⎦

−1

1

θ= ⎡ −0.00643 0.0263 ⎤⎦ ⎢ η.

32.2 ⎣ ⎣ −0.000222 −0.00153 ⎥⎦ ⎢⎣ 2.08 ⎥⎦

(6.234)

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 243

DYNAMICS IN HORIZONTAL LEVEL FLIGHT AT AN

ALTITUDE OF 20,000 ft; FLIGHT VELOCITY, Ue, OF

830 ft/s (MACH 0.8) AND WEIGHT OF 637,000 lb

Inverting this relationship,

−1

−1

⎡ ⎡ −0.0941 −0.624 ⎤ ⎡ 32.7 ⎤ ⎤

η = 32.2 ⎢ ⎡⎣ −0.00643 0.0263 ⎤⎦ ⎢ ⎥ θ

⎢⎣ ⎣ −0.000222 −0.00153 ⎥⎦ ⎢⎣ 2.08 ⎥⎦ ⎥

⎦

(6.235)

which simplifies to

1

η=− θ = −0.0135θ. (6.236)

74.036

setting.

The two main trimmed flight modes in lateral flight correspond to

1. Steady sideslip

2. Turn coordination and banking

A steady sideslip is employed to lose height without an increase in airspeed.

However, whenever the wing is lowered, the greater airflow ahead of the

CG causes the aircraft to yaw towards the lower wing. The corresponding

parasitic aerodynamic moment is known as the adverse yawing moment. In

a sideslip, the wing is lowered while the yaw is inhibited by the use of the

rudder to counteract the yaw generated by the movement of the aileron. This

equilibrium mode of flight pertains to steady level flight at a steady sideslip

and a steady bank angle but no yaw (Figure 6.4).

The steady equilibrium conditions are

Ttrim = ∑D trim (6.237)

mg = Ttrim sin ( φtrim ) + ∑L

trim (6.238)

www.ebook777.com

free ebooks ==> www.ebook777.com

244 Flight Dynamics, Simulation, and Control

y Lift

β

V, velocity of

relative wind

z mg

FIGURE 6.4

Steady level flight at a steady sideslip and a steady bank angle.

and

mg sin ( φtrim ) = ∑Y trim . (6.239)

Assuming that these trim conditions are small perturbations to those corre-

sponding to steady level flight, we may use the small perturbation equations

to solve for these perturbation states. Thus, considering the lateral steady small

perturbation state-space equations, including the control surface deflections,

⎡ Yv Yp Yr − U e g ⎤ ⎡ Δv ⎤ ⎡ Yξ Yζ ⎤ ⎡0 ⎤

⎢L Lp Lr 0 ⎥⎥ ⎢⎢ Δp ⎥⎥ ⎢⎢ Lξ Lζ ⎥ ⎡ Δξ ⎤ ⎢⎢0 ⎥⎥

⎥

⎢ v + = (6.240)

⎢ Nv Np Nr 0 ⎥ ⎢ Δr ⎥ ⎢ N ξ N ζ ⎥ ⎢⎣ Δζ ⎥⎦ ⎢0 ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎣0 1 0 0 ⎦ ⎣ Δφ ⎦ ⎣ 0 0 ⎦ ⎣0 ⎦

⎡ Yv g⎤ ⎡ Yξ Yζ ⎤ ⎡0 ⎤

⎢ ⎥ ⎡ Δv ⎤ ⎢ ⎥ ⎡ Δξ ⎤ ⎢ ⎥

⎢ Lv 0 ⎥ ⎢ ⎥ + ⎢ Lξ

Δφ

Lζ ⎥ ⎢ =

Δζ ⎥⎦ ⎢ ⎥

0 . (6.241)

⎢⎣ N v 0 ⎥⎦ ⎣ ⎦ ⎢⎣ N ξ N ζ ⎥⎦ ⎣ ⎢⎣0 ⎥⎦

−1

⎡ Δξ ⎤ ⎡ Lξ Lζ ⎤ ⎡ Lv ⎤ 1 ⎡ Nζ −Lζ ⎤ ⎡ Lv ⎤

⎢ Δζ ⎥ = − ⎢ N Δv = − Δv

⎣ ⎦ ⎣ ξ N ζ ⎥⎦ ⎢⎣ N v ⎥⎦ ( Lξ Nζ − Lζ Nξ ) ⎢⎣ −Nξ Lξ ⎥⎦ ⎢⎣ N v ⎥⎦

(6.242)

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 245

1 T

Δφ = − ⎡Yv Yξ Yζ ⎤⎦ ⎡⎣ Δv Δξ Δζ ⎤⎦ . (6.243)

g⎣

−1

⎡ Lξ Lζ ⎤

Δ ra = ⎢ , (6.244)

⎣ Nξ N ζ ⎥⎦

one could ignore the off-diagonal derivatives when these are small.

When the control column is moved to the left, the aircraft will bank to the left,

accompanied by a sideslip towards the lower wing which is caused by a com-

ponent of the lift in that direction. As a consequence of the sideslip, the greater

airflow ahead of the CG also causes the aircraft to turn towards the lower wing.

On its own, this would culminate in a spiral dive towards the lower wing. On

the other hand, when the left rudder pedal is moved forward, the nose of the

aircraft not only swings to the left, but also the higher speed of the outer right

wing will produce an additional lift force on the right wing thus causing the

aircraft to bank. Thus, it is not only clear that the aileron and rudder input

responses are coupled, but it also means that in order to achieve a smooth turn

without any sideslip, the control column and rudder pedals should both be

moved in a coordinated way. This is the basis of a coordinated turn.

In a sustained turn, an aircraft maintains constant altitude, at a steady

tangential velocity (forward speed), U, with a steady bank angle (i.e. zero roll

rate, p), zero sideslip and a steady turn rate Ω.

The equilibrium conditions are

Ttrim = ∑D trim (6.245)

mg − Ytrim sin ( φtrim ) = ∑L trim cos ( φbank ) (6.246)

and

mUΩ + ∑Y trim cos ( φbank ) = ∑L trim sin ( φbank ) (6.247)

where mUΩ is the d’Alembert force (so-called centrifugal force) acting on the

aircraft.

www.ebook777.com

free ebooks ==> www.ebook777.com

246 Flight Dynamics, Simulation, and Control

y Lift

Radius of turn, R

mU Ω

z mg

FIGURE 6.5

Sustained turn at a steady tangential velocity (forward speed), U, with a steady bank angle

(i.e. zero roll rate, p), zero sideslip and a steady turn rate Ω.

Considering the situation illustrated in Figure 6.5 and assuming that under

steady turn conditions, Ytrim = 0, and recalling that Ω = U/R, the radius of the

turn and the turn rate are

U2 U g tan ( φbank )

R= and Ω = = . (6.248)

g tan ( φbank ) R U

The body components of the angular velocity vector are then given by

⎡ pB ⎤ ⎡ 1 0 0 ⎤⎡0⎤

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎢ qB ⎥ = ⎢0 cos φ −sin φ ⎥ ⎢ 0 ⎥ . (6.249)

⎢⎣ rB ⎥⎦ ⎢⎣0 sin φ cos φ ⎥⎦ ⎢⎣Ω ⎥⎦

corresponding to steady level flight, we may also use the small perturba-

tion equations to solve for these perturbation states. Thus, considering the

steady small perturbation state-space equations, including the control surface

deflections,

⎡ Yv Yp Yr − U e g ⎤ ⎡ Δv ⎤ ⎡ Yξ Yζ ⎤ ⎡0 ⎤

⎢L Lp Lr 0 ⎥⎥ ⎢⎢ Δp ⎥⎥ ⎢⎢ Lξ Lζ ⎥ ⎡ Δξ ⎤ ⎢⎢0 ⎥⎥

⎥

⎢ v + = (6.250)

⎢ Nv Np Nr 0 ⎥ ⎢ Δr ⎥ ⎢ N ξ N ζ ⎥ ⎢⎣ Δζ ⎥⎦ ⎢0 ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎣0 1 0 0 ⎦ ⎣ Δφ ⎦ ⎣ 0 0 ⎦ ⎣0 ⎦

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 247

the equations reduce to

⎡Yr − U e g⎤ ⎡ Yξ Yζ ⎤ ⎡0 ⎤

⎢ ⎥ ⎡ Δv ⎤ ⎢ ⎥ ⎡ Δξ ⎤ ⎢ ⎥

⎢ Lr 0 ⎥ ⎢ ⎥ + ⎢ Lξ

Δφ

Lζ ⎥ ⎢ = 0 . (6.251)

Δζ ⎥⎦ ⎢ ⎥

⎢⎣ N r 0 ⎥⎦ ⎣ ⎦ ⎢⎣ N ξ N ζ ⎥⎦ ⎣ ⎢⎣0 ⎥⎦

From the second and third equations,

−1

⎡ Δξ ⎤ ⎡ Lξ Lζ ⎤ ⎡ Lr ⎤ 1 ⎡ Nζ −Lζ ⎤ ⎡ Lr ⎤

⎢ Δζ ⎥ = − ⎢ N ⎥ ⎢ ⎥ Δr = − Δr ,

⎣ ⎦ ⎣ ξ Nζ ⎦ ⎣ N r ⎦ ( Lξ Nζ − Lζ Nξ ) ⎢⎣ − Nξ Lξ ⎥⎦ ⎢⎣ N r ⎥⎦

(6.252)

and from the first equation,

⎛1⎞ T

Δφ = − ⎜ ⎟ ⎡⎣Yr − U e Yξ Yζ ⎤⎦ ⎡⎣ Δr Δξ Δζ ⎤⎦ . (6.253)

⎝g⎠

−1

⎡ Lξ Lζ ⎤

Δ ra = ⎢ , (6.244)

⎣ Nξ N ζ ⎥⎦

one could ignore the off-diagonal derivatives when these are small.

Find the lateral control setting and bank angle ϕ that are needed to fly

the DC 8 in a well-banked turn along a circular path of 10 km radius

and a forward speed, U = 251.5 m/s.

The lateral dynamics of the DC 8 in terms of the state-space deriva-

tives (Table 6.10) is considered.

TABLE 6.10

Values of the Non-Zero Lateral Stability Derivatives of the DC 8

in a Cruise Condition

U es ( m/s ) 251.21616 Ixz/Ixx 0.0119 Ixz/Izz 0.0063

Yv −0.0868 Lp −1.18 Nv 0.0086379 Nr −0.23

Lv −0.017634 Lr 0.336 Np −0.01294

Controls Yξ 0.0 Lξ −2.11 Nξ −0.0519

Yζ 5.58698 Lζ 0.559 Nζ −1.168

(continued)

www.ebook777.com

free ebooks ==> www.ebook777.com

248 Flight Dynamics, Simulation, and Control

TURNING OF THE DC 8

Considering the case when the controls are not fixed, the perturba-

tion equations of motion are

⎡1 0 0 0 ⎤ ⎡ Δv ⎤

⎢0 1 −0.0119 0 ⎥⎥ d ⎢⎢ Δp ⎥⎥

⎢

⎢0 −0.0063 1 0 ⎥ dt ⎢ Δr ⎥

⎢ ⎥ ⎢ ⎥

⎣0 0 0 1 ⎦ ⎣ Δφ ⎦

⎡ −0.0868 0 −251 9.81⎤ ⎡ Δv ⎤ ⎡ 0.0 5.587 ⎤

⎢ −0.0176 −1.18 0.336 0 ⎥⎥ ⎢⎢ Δp ⎥⎥ ⎢⎢ −2.11 0.559 ⎥⎥ ⎡ ξ ⎤

=⎢ +

⎢ 0.00864 −0.01294 −0.23 0 ⎥ ⎢ Δr ⎥ ⎢ −0.0519 −1.168 ⎥ ⎢⎣ζ ⎥⎦

⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣ 0 1 0 0 ⎦ ⎣ Δφ ⎦ ⎣ 0 0 ⎦

(6.254)

T

Δp = −1.18Δp + ⎡⎣ −2.11 0.559 ⎤⎦ ⎡⎣ξ ζ ⎤⎦ . (6.255)

Assuming the controls are fixed and that the roll equation is satisfied

instantaneously,

T

Δp = ⎣⎡ −0.0176 −1.18 0.336 0 ⎤⎦ ⎡⎣ Δv Δp Δr Δφ ⎤⎦ = 0.

(6.256)

Hence,

0.336Δr − 0.0176Δv

Δp = = 0.285Δr − 0.0149Δv. (6.257)

1.18

⎡ Δv ⎤

⎡ Δv ⎤ ⎡ −0.0868 0 −251 9.81⎤ ⎢ ⎥

d ⎢ ⎥ ⎢ ⎥ 0.285Δr − 0.0149Δv ⎥

Δr = 0.00864 −0.01294 −0.23 0 ⎥⎢

dt ⎢ ⎥ ⎢ ⎢ Δr ⎥

⎢⎣ Δφ ⎥⎦ ⎢⎣ 0 1 0 0 ⎦⎢

⎥ ⎥

⎣ Δφ ⎦

(6.258)

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 249

TURNING OF THE DC 8

and setting Δϕ = 0 gives

⎡ −0.0868 0 −251 ⎤ ⎡ Δv ⎤

d ⎡ Δv ⎤ ⎢ ⎥⎢ ⎥

⎢ ⎥ = 0.00864 −0.01294 −0.23 ⎥ ⎢0.285Δr − 0.0149Δv ⎥ (6.259)

dt ⎣ Δr ⎦ ⎢

⎢⎣ 0 1 0 ⎥⎦ ⎢⎣ Δr ⎥⎦

which simplifies to

d ⎡ Δv ⎤ ⎡ −0.0868 −251 ⎤ ⎡ Δv ⎤

⎢ ⎥=⎢ . (6.260)

dt ⎣ Δr ⎦ ⎣ 0.0088 −0.2337 ⎥⎦ ⎢⎣ Δr ⎥⎦

Hence, the characteristic equation (Dutch roll modes) for this subsys-

tem (calculated as a check) is

sponding to a damping ratio of 0.1073 and a damped natural frequency

of 1.485 rad/s. The undamped natural frequency is 1.4937 rad/s.

Assuming the aircraft is initially in steady level flight, Δr = Ω =

251/10,000 = 0.0251. In addition to assuming steady level flight, also

assuming that the increments Δv = 0 and Δp = 0, from the second and

third of the previously mentioned equations, we obtain

⎢ −0.0519 ⎥ ⎢

−1.168 ⎦ ⎣ζ ⎦⎥ = −⎢ ⎥ Δr , (6.262)

⎣ ⎣ −0.23 ⎦

1 ⎡ Δr ⎤

Δφ = − ⎡ 251 5.587 ⎤⎦ ⎢ ⎥ . (6.263)

9.81 ⎣ ⎣ Δζ ⎦

(continued)

www.ebook777.com

free ebooks ==> www.ebook777.com

250 Flight Dynamics, Simulation, and Control

TURNING OF THE DC 8

Hence, for the aileron and rudder setting angles, we obtain

−1

⎡ξ ⎤ ⎡ −2.11 0.559 ⎤ ⎡ 0.336 ⎤

⎢ζ ⎥ = − ⎢ −0.0519 Δrr

⎣ ⎦ ⎣ −1.168 ⎥⎦ ⎢⎣ −0.23 ⎥⎦

= −⎢ Δr (6.264)

⎣ 0.0208 −0.8462 ⎥⎦ ⎢⎣ −0.23 ⎥⎦

which simplifies to

⎢ζ ⎥ = ⎢ −0.2016 ⎥ Δr = ⎢ −0.00508 ⎥ rad = ⎢ −0.291°⎥ . (6.265)

⎣ ⎦ ⎣ ⎦ ⎣ ⎦ ⎣ ⎦

Δφ = − ⎢ ⎢ ⎥ = −⎢ ⎢ ⎥ = −0.64 rad

⎣ 9.81 9.81 ⎥⎦ ⎣ ζ ⎦ ⎣ 9.81 9.81 ⎦⎥ ⎣ −0.00508 ⎦

(6.266)

or

Coupled trimmed flight is extremely important for aircraft performing

motions in high angular rates or flying under abnormal or upset condi-

tions. In these cases, the coupled equations of motion in steady or per-

turbed flight must be considered with some of the relevant perturbations

not assumed to be small. Decoupled trimmed flight is also a starting point

for most of these analyses.

The Sidestep Manoeuvre

To bring about the sideways displacement of an aircraft’s flight path, to align

the aircraft with the runway’s centre line, a transverse force must be applied

across the initial flight path in order to accelerate the aircraft towards the

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 251

runway. This must be followed as it nears the correct path by a force in the

opposite direction so that the aircraft aligns with the centre line with negli-

gible transverse velocity. There are two possible ways to generate the neces-

sary forces and execute the manoeuvre. One approach involves coordinated

turns and banking to generate the side forces while the other is based on a

slipping turn.

Considering the coordinated turns approach, assuming a truly banked,

steady turn, the yaw rate and the aircraft’s sideslip velocity are, respectively,

given by

dψ g g dy

r= = sin φ ≈ φ, = V sin ψ ≈ Vψ. (6.268)

dt V V dt

a coordinated turn. The first turn is assumed to be completed in time T1

with a steady banked turn and the maximum bank angle equal to ϕ1 while

the corresponding time and maximum bank angle for the second turn are

assumed to be T2 and ϕ2. Assuming that the bank angle variation with time

is sinusoidal and integrating with respect to time, the equation

πt

φ = φi sin , (6.269)

Ti

the heading angles ψi, the lateral velocities and positions are, respectively,

given, with ψ0 = 0, by

g Ti ⎛ πt ⎞ dy g ⎛ Ti πt ⎞

ψ i = φi 1 − cos ⎟ , = Vψ and y i = Vψ ( i −1)T t + φi Ti t − sin ⎟ .

V π ⎜⎝ Ti ⎠ dt π ⎜⎝ π Ti ⎠

(6.270)

At time t = Ti, the heading angles and the lateral distances travelled are,

respectively, given as

2 g Ti g

ψ iT = φi , yiT = Vψ ( i −1)T Ti + φi Ti2 . (6.271)

V π π

2g

ψ T = ψ1T + ψ 2T = ( φ1T1 + φ2T2 ) = 0. (6.272)

πV

www.ebook777.com

free ebooks ==> www.ebook777.com

252 Flight Dynamics, Simulation, and Control

The total time for the manoeuvre T = T1 + T2. The total lateral distance

travelled is

g g

yT = y1T + y1T = φ1T12 + Vψ1T T2 + φ2T22

π π

g 2g g g

= φ1T12 + φ1T1T2 + φ2T22 = φ1T1T .

π π π π

But the total time for the manoeuvre satisfies the relation

⎛ φ −φ ⎞

T = T1 + T2 = T1 ⎜ 2 1 ⎟ . (6.273)

⎝ φ2 ⎠

Hence, the total sidestepping distance is related to the total time and the

maximum bank angles during the two phases of the manoeuvre:

g ⎛ φ1φ2 ⎞ 2

yT = T . (6.274)

π ⎜⎝ φ2 − φ1 ⎟⎠

The relation is useful in estimating the total time given the required side-

stepping distance and the maximum bank angles during the two phases of

the manoeuvre. When the manoeuvre is completely asymmetric, ϕ2 = −ϕ1

and the required bank angle may be expressed as

2π y T

φ1 = . (6.275)

T2 g

Chapter Highlights

• Longitudinal stability

Motion in the plane of symmetry (xz plane) about y-axis. Disturbance

changes α.

For stability, if α increases, pitching moment (about CG) must be nega-

tive (nose-down moment to reduce α), that is, Cmα < 0(∂Cm ∂α ≡ Cmα ,

longitudinal stability derivative.

Vehicle aerodynamic centre (i.e. neutral point) must lie aft of (behind)

aeroplane CG for stability. Generally, fuselage is destabilising, wing is

(slightly) destabilising and horizontal tail is highly stabilising (that’s

its function!).

Horizontal tail design parameters: tail volume ratio; incidence, it; and

tail location from downwash point of view. Wing downwash decreases

effective α at the horizontal tail.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 253

most CG location critical for control (most stable case).

• Directional stability

Motion in the xy plane about z-axis (asymmetric plane). Disturbance

(such as side gust) changes β (angle of sideslip).

For stability, aeroplane behaves like a weathercock, and yawing

moment (about CG) must be positive (nose to right), that is, the direc-

tional stability derivative, ∂Cn ∂β ≡ Cnβ > 0.

Forward fuselage is highly destabilising, the wing has very small effect

(low α’s), and the vertical tail is highly stabilising (that’s its function!).

Vertical tail design parameters: vertical tail volume ratio and tail loca-

tion from sidewash or dynamic pressure point of view. Wing sidewash

increases effective angle of sideslip at vertical tail.

• Roll stability

Motion in the yz plane about x-axis (asymmetric plane). Aeroplane roll

leads to sideslip due to weight component.

For stability, if â is positive (i.e. aeroplane sideslips to right), rolling

moment (about CG) must be negative (right wing up, left wing down),

that is, the roll stability derivative, ∂Cl ∂β ≡ Clβ < 0.

Fuselage has no effect (low α’s), and the vertical tail is slightly stabilis-

ing. Main contributors to roll stability are high wing position, dihedral

angle and sweepback.

Wing dihedral increases α on the down-going wing during a sideslip.

Strip theory can be used to approximately compute Clβ due to dihedral

for large aspect ratio, low-sweep wings.

• Natural modes: Longitudinal motion

Generally, longitudinally stable aircraft have two distinct oscillatory

motions.

Phugoid – Low damping and long time period. Characterised by changes

in pitch attitude, altitude and forward speed at a nearly constant α.

Short period mode – Heavily damped with a very short time period.

Characterised by constant speed and rapid changes in α and pitch

attitude.

• Natural modes: Lateral motion

Coupled roll, sideslip and yaw motions

Aeroplane responses characterised by three distinct types of motions

Roll mode – A heavily damped roll subsidence motion

Dutch roll mode – Primarily sideslipping and yawing motion

Spiral mode – A slowly convergent or divergent motion. Essentially, a

yawing motion with small β

www.ebook777.com

free ebooks ==> www.ebook777.com

254 Flight Dynamics, Simulation, and Control

• Estimation of derivatives

(i) Speed derivatives

Increase in forward speed (u) changes lift and drag (or axial

force and normal force). Thrust assumed constant with speed

for jet engines

Axial force derivative (speed damping), ∂X ∂u = −ρu0SCD0 ,

where subscript 0 refers to initial or ref flight condition

Normal force derivative, ∂Z ∂u = −ρu0SCL0

Pitching moment derivative, ∂M/∂u = 0 (low speed)

(ii) Pitch rate derivatives

q = pitch rate (rate of change of pitch angle, which is the rotation

of aeroplane about y-axis while α remains constant).

Pitch rate changes effective α at tail (Δαt = qlt/u0) and thereby

generates normal force (or lift) and pitching moment.

Normal force derivative (horizontal tail only),

∂Cz

C zq ≡ = −2CLαt ηVH .

∂ ( qc 2u0 )

Pitch damping derivative (horizontal tail only),

∂Cm l

Cmq ≡ = −2CLαt ηVH t .

∂ ( qc 2u0 ) c

small per cent (very approximate empirical result).

(iii) Rate of change of α ( α ) derivatives

When α is changed suddenly, pressure distribution on wing or

tail does not change instantaneously (i.e. there is a time lag),

⎛ ∂ε lt ⎞

which induces a change in effective α ⎜ Δαt = α .

⎝ ∂α u0 ⎟⎠

Normal force derivative (horizontal tail only),

∂Cz ∂ε

Czα ≡ = −2CLαt ηVH .

∂ ( α c 2u0 ) ∂α

∂Cm l ∂ε

Cmα ≡ = −2CLαt ηVH t .

∂ ( α c 2u0 )

c ∂α

For complete aeroplane, increase α derivatives by a small per

cent (very approximate empirical result).

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 255

Roll rate (p, angular motion about the x-axis) creates a linear

velocity distribution over wing (and horizontal tail) leading to

change in local α. Rolling moment is generated due to difference

in lift on left and right wings.

∂Cl 4CLαw b/2 2

Roll damping, Clp ≡ cy dy , where c is

∂ ( pb 2u0 )

=−

Sb 2 0 ∫

the wing chord.

Rolling motion generates yawing moment (hence roll and yaw

are coupled).

∂Cn C

Cnp ≡ = − L (very approximate value) where lift coef-

∂ ( pb 2u0 ) 8

ficient depends on the flight condition. (Use L = W for straight,

level flight.)

Side force due to roll rate, Cy p, is generally very small and may

be ignored.

(v) Yaw rate derivatives

Yaw rate (r, angular motion about the z-axis) causes change in

side force on vertical tail due to local change of angle of sideslip

rl

(β), Δβ = − v

u0

∂Cy l

Side force derivative, Cyr ≡ ≈ −2 ( Cyβ )VT v .

∂ ( rb 2u0 ) b

∂Cn l

Yaw damping, Cnr ≡ ≈ −2 ( Cnβ )VT v .

∂ ( rb 2u0 ) b

∂Cl C z l

Rolling moment derivative, Clr ≡ ≈ L − 2 v ( Cyβ )VT v .

∂ ( rb 2u0 ) 4 b b

Exercises

6.1 The perturbation longitudinal dynamics of a F15 fighter aircraft in steady

uniform level flight with subsonic velocity of 556.3 ft/s is given by

⎢ Δα ⎥ ⎢ ⎥ ⎢ Δα ⎥ ⎢ −0.1497 ⎥

⎢ ⎥ = ⎢ −0.0002 −1.2763 1 0 ⎥⎢ ⎥+⎢ ⎥ Δη.

⎢ Δq ⎥ ⎢ 0.0007 1.0218 −2.4052 0 ⎥ ⎢ Δq ⎥ ⎢ −14.0611⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢ Δθ ⎥ 0 0 1 0 ⎦ ⎣ Δθ ⎦ ⎣ 0

⎣ ⎦ ⎣ ⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

256 Flight Dynamics, Simulation, and Control

period mode of the aircraft with the elevator fixed, and calculate the

characteristic polynomial and its roots corresponding to the mode.

(ii) Hence, or otherwise, obtain a pair of first-order equations to

approximate the phugoid oscillations of the aircraft with the ele-

vator fixed, and calculate the damped natural frequency and the

damping ratio corresponding to these oscillations.

(iii) Obtain the exact characteristic polynomial and its roots, and calculate

the damped natural frequency and the damping ratios of the longitu-

dinal oscillations using MATLAB® and the appropriate m-functions.

6.2 The perturbation longitudinal dynamics of a future SST airliner in

steady uniform level flight with subsonic velocity of 420 ft/s is given by

the equations

⎢ Δw ⎥ ⎢

⎢ ⎥ = ⎢ −0.1443 −0.8102 −11.5737 −104.2678 ⎥⎥ ⎢⎢ Δw ⎥⎥ ⎢⎢ −0.5304 ⎥⎥

+ Δη.

⎢ Δq ⎥ ⎢ 0.0002 −0.0029 −0.5811 −0.3672 ⎥ ⎢ Δq ⎥ ⎢ −0.0258 ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎢ Δθ ⎥ 0 0 1 0 ⎦ ⎣ Δθ ⎦ ⎣ 0.0000 ⎦

⎣ ⎦ ⎣

(i) Show that the characteristic polynomial is given by

( )(

= λ 2 + 0.0034λ + 0.0251 λ 2 + 1.4072λ + 0.7940 . )

(ii) Hence, or otherwise, calculate the damped natural frequencies and

the damping ratios, corresponding to the short period and the phu-

goid oscillations, respectively, of the aircraft with the controls fixed.

(iii) Determine the change in the setting of the elevator angle for a

decrease in the pitch angle by 4°.

6.3 The perturbation lateral dynamics of a F15 fighter aircraft in steady uniform

level flight with subsonic velocity of 556.3 ft/s is given by the equations

⎡ Δβ ⎤ ⎡ −0.2720 0.0032 −1 0.0578 ⎤ ⎡ Δβ ⎤

⎢ ⎥ ⎢

⎢ Δp ⎥ = ⎢ −43.3660 −2.4923 1.8964 0 ⎥⎥ ⎢⎢ Δp ⎥⎥

⎢ Δr ⎥ ⎢ 6.5529 −0.0573 −0.7759 0 ⎥ ⎢ Δr ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎢⎣ Δφ ⎥⎦ ⎣ 0 1 0.0032 0 ⎦ ⎣ Δφ ⎦

⎡ −0.0003 0.0420 ⎤

⎢ 8.5397 0.7107 ⎥⎥ ⎡ Δξ ⎤

+⎢ .

⎢ 0.0849 −3.4512 ⎥ ⎢⎣ Δζ ⎥⎦

⎢ ⎥

⎣ 0 0 ⎦

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 257

Calculate the time constant associated with this mode of

oscillation.

(ii) Hence, or otherwise, obtain three first-order equations to approxi-

mate the coupled Dutch roll and spiral motion of the aircraft with

the controls fixed, and find the characteristic cubic polynomial

corresponding to these motions.

(iii) Determine whether the spiral mode is stable or unstable.

(iv) Obtain the exact characteristic polynomial and its roots, and cal-

culate the damped natural frequency and the damping ratios

of the lateral oscillations using MATLAB and the appropriate

m-functions.

6.4 The perturbation lateral dynamics of a future SST airliner in steady

uniform level flight with subsonic velocity of 420 ft/s is given by the

equations

⎢ Δp ⎥ ⎢

⎢ ⎥ = ⎢ −0.01611 −1.5392 0.5677 −0.2403 ⎥⎥ ⎢⎢ Δp ⎥⎥

⎢ Δr ⎥ ⎢ 0.0017 −0.0654 −0.1583 0.0249 ⎥ ⎢ Δr ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣⎢ Δφ ⎦⎥ ⎣ 0

1 0.117 0 ⎦ ⎣ Δφ ⎦

⎡ −0.0601 0.0495 ⎤

⎢ 0.0664 0.0022 ⎥⎥ ⎡ Δξ ⎤

+⎢ .

⎢ 0.0063 −0.0044 ⎥ ⎢⎣ Δζ ⎥⎦

⎢ ⎥

⎣ 0 0 ⎦

Calculate the time constant associated with this mode of

oscillation.

(ii) Hence, or otherwise, obtain three first-order equations to approxi-

mate the coupled Dutch roll and spiral motion of the aircraft with

the controls fixed, and find the characteristic cubic polynomial

corresponding to these motions.

(iii) Determine whether the spiral mode is stable or unstable.

(iv) Determine the relationship between the steady bank angle and the

steady sideslip velocity assuming that Δp = Δr = 0. Also deter-

mine the setting angles for the aileron and rudder per unit sideslip

velocity change.

www.ebook777.com

free ebooks ==> www.ebook777.com

258 Flight Dynamics, Simulation, and Control

TABLE 6.11

Values of the Non-Zero Lateral Stability Derivatives of the Fighter

yv −0.0768 Ulv −12.9 lr 0.387 np 0.024

yp/U 0.006 lp −0.746 Unv 4.31 nr −0.174

yr/U −0.999 yξ/U 0.0 lξ −1.61 nξ −0.0346

yϕ/U 0.0369 yζ/U 0.02918 lζ 0.42 nζ −0.78

6.5 The lateral dynamics of a typical fighter aircraft in terms of the concise

derivatives (Table 6.11) is considered.

The perturbation equations of motion are

⎡ β ⎤ ⎡ yv yp U yr U yφ U ⎤ ⎡ β ⎤ ⎡ yξ U yζ U ⎤

⎢ ⎥ ⎢ 0 ⎥⎥ ⎢⎢ Δp ⎥⎥ ⎢⎢ lξ lζ ⎥⎥ ⎡ Δξ ⎤

d ⎢ Δp ⎥ ⎢ Ulv lp lr

= + .

dt ⎢ Δr ⎥ ⎢Unv np nr 0 ⎥ ⎢ Δr ⎥ ⎢ nξ nζ ⎥ ⎢⎣ Δζ ⎥⎦

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣ Δφ ⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφ ⎦ ⎣ 0 0 ⎦

Calculate the time constant associated with this mode of oscillation.

(ii) Hence, or otherwise, obtain a pair of first-order equations to

approximate the Dutch roll oscillations of the aircraft with the con-

trols fixed, and calculate the damped natural frequency and the

damping ratio corresponding to these oscillations.

(iii) Find the lateral perturbation control setting and the correspond-

ing bank angle Δϕ that are needed to fly the aircraft in a well-

banked turn along a circular path of 10 km radius and a forward

speed, U = 280 m/s.

6.6 The lateral dynamics of an AFTI-16 fighter aircraft, a modified version

of the F-16, with the concise state-space derivatives given in Table 6.12, is

considered.

TABLE 6.12

Values of the Non-Zero Lateral Stability Derivatives of the AFTI-16

in a Cruise Condition with a Steady Level Trim Velocity of 597 ft/s

Derivative Value Derivative Value Derivative Value

yϕ 32.17 lv −0.032202 nv 0.0038456

yv −0.154099 lp −0.893601 np −0.000888

yp 49.185039 lr 0.318845 nr −0.278676

yr −595.998

Note: 1 ft = 0.3048 m. However, there is no need to change the units as the data pro-

vided are consistent.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 259

steady level flight with a steady level trim velocity of 597 ft/s and the

controls fixed may be expressed as

⎡ Δv s ⎤ ⎡ yv yp yr yφ ⎤ ⎡ Δvs ⎤

⎢ Δp ⎥ ⎢

⎢ s ⎥ = ⎢ lv lp lr 0 ⎥⎥ ⎢⎢ Δps ⎥⎥

.

⎢ Δrs ⎥ ⎢ nv np nr 0 ⎥ ⎢ Δrs ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣⎢ Δφs ⎦⎥ ⎣ 0 1 0 0 ⎦ ⎣ Δφs ⎦

(i) Obtain an approximate equation for the roll subsidence mode.

Calculate the time constant associated with this mode of oscillation.

(ii) Hence, or otherwise, obtain a pair of first-order equations to

approximate the Dutch Roll oscillations of the aircraft with the con-

trols fixed and calculate the damped natural frequency and the

damping ratio corresponding to these oscillations.

(iii) Determine whether the spiral mode is stable or unstable.

6.7 The AFTI-16 fighter aircraft is an unconventional aircraft as both the

left and right wing ailerons and elevators may be independently actu-

ated. The linearised (small perturbation) lateral dynamics of an AFTI-16

in steady level flight with a forward trim speed of 597 ft/s are

⎢ Δp ⎥ ⎢ ⎡ Δξ ⎤

⎢ s ⎥ ⎢ lv lp lr 0 0 ⎥⎥ ⎢⎢ Δps ⎥⎥ ⎢⎢ lξ lηd lca lζ ⎥⎥ ⎢

Δηd ⎥⎥

⎢ Δrs ⎥ = ⎢ nv np nr 0 0 ⎥ ⎢ Δrs ⎥ + ⎢ nξ nηd nca nζ ⎥ ⎢

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ Δca ⎥

⎢ Δφ s ⎥ ⎢ 0 1 0 0 0 ⎥ ⎢ Δφs ⎥ ⎢ 0 0 0 0 ⎥⎢ ⎥

⎢ Δψ ⎥ ⎢ 0 ⎣ Δζ ⎦

⎣ s⎦ ⎣ 0 1 0 0 ⎥⎦ ⎢⎣ Δψ s ⎥⎦ ⎢⎣ 0 0 0 0⎦ ⎥

The controls are the aileron (left and right wing ailerons constrained

to an antisymmetric deflection), elevons (left and right wing eleva-

tors deflected differentially or antisymmetrically), canard and rudder,

respectively. The concise state-space derivatives of the aircraft are given

in Table 6.12 and the control derivatives are in Table 6.13.

TABLE 6.13

Values of the Non-Zero Lateral Control Derivatives of the AFTI-16

Derivative Value Derivative Value Derivative Value

yηd 8.595606 lηd −13.5832 nηd −1.50547

yξ 0.213129 lξ −17.4468 nξ −0.268303

yca 4.378995 lca 0.414519 nca 1.51008

yζ 12.635505 lζ 3.92325 nζ −1.96651

Note: 1 ft = 0.3048 m. However, there is no need to change the units as the data pro-

vided are consistent.

www.ebook777.com

free ebooks ==> www.ebook777.com

260 Flight Dynamics, Simulation, and Control

(i) Assume the elevons (the left and right elevator) and the canards

to be fixed, and find the lateral perturbation aileron and rudder

settings (the first and fourth controls) and the corresponding bank

angle ϕ that are needed to fly in a well-banked turn along a circu-

lar path of radius, 32,835 ft.

(ii) Assume that the aileron and rudder are not available due to a fault,

and determine if the same manoeuvre could be flown using the

differentially actuated elevons and the canard (the second and

third controls).

6.8 The linearised (small perturbation) lateral dynamics of an AFTI-16 in

steady level flight with a steady level trim velocity of 597 ft/s and the

controls fixed may be expressed as

⎡ β ⎤ ⎡ yv yp U yr U yφ U ⎤ ⎡ β ⎤ ⎡ yξ U yζ U ⎤

⎢ ⎥ ⎢ 0 ⎥⎥ ⎢⎢ Δp ⎥⎥ ⎢⎢ lξ lζ ⎥⎥ ⎡ Δξ ⎤

d ⎢ Δp ⎥ ⎢ Ulv lp lr

= + .

dt ⎢ Δr ⎥ ⎢Unv np nr 0 ⎥ ⎢ Δr ⎥ ⎢ nξ nζ ⎥ ⎢⎣ Δζ ⎥⎦

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥

⎣ Δφ ⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφ ⎦ ⎣ 0 0 ⎦

(i) Using the derivative data defined in the previous exercises, deter-

mine the relationship between the steady bank angle and the

steady sideslip velocity assuming that Δp = Δr = 0. Also deter-

mine the setting angles for the aileron and rudder per unit sideslip

velocity change.

(ii) Find the lateral perturbation control setting and the corresponding

bank angle ϕ that are needed to fly the aircraft in a well-banked

turn along a circular path of 30,000 ft radius and a forward speed

of 597 ft/s.

6.9 The complete longitudinal dynamic equations including the

dynamics of the elevator-servo actuator of a typical fighter can be

expressed as

⎡ Δu s ⎤ ⎡⎢ xu U se xw 0 −g 0 xη ⎤

⎥ ⎡ Δus ⎤ ⎡ 0 ⎤

⎢ Δα ⎥ ⎢ zu zη ⎥ ⎢

⎢ s⎥ ⎢ e zw 1 0 0 e ⎢

Δα s ⎥⎥ ⎢⎢ 0 ⎥⎥

⎢ Δq s ⎥ ⎢ U s Us ⎥

⎥ ⎢ Δqs ⎥ ⎢ 0 ⎥

⎢ ⎥ = mu U se mw mq 0 0 mη ⎥ ⎢ ⎥ + ⎢ ⎥ Δηc

⎢ Δθ s ⎥ ⎢⎢ ⎢ Δθs ⎥ ⎢ 0 ⎥

⎢ ⎥ ⎢ 0 0 1 0 0 0 ⎥

⎥ ⎢ Δh ⎥ ⎢ 0 ⎥

⎢ Δh ⎥ ⎢ 0 U se 0 U se 0 0 ⎥⎢ ⎥ ⎢ ⎥

⎢ Δη ⎥ ⎢ ⎢ Δη ⎥⎦ ⎢⎣ pη ⎥⎦

⎣ ⎦ ⎣ 0 0 0 0 0 − pη ⎥⎦ ⎣

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 261

⎢ −0.00117 −9.5164 1 0 0 −0.0717 ⎥⎥ ⎢ 0 ⎥

⎢ ⎢ ⎥

⎢ −0.000129 1.4168 −0.4932 0 0 −1.645 ⎥ ⎢ 0 ⎥

A=⎢ ⎥ B=⎢ ⎥.

⎢ 0 0 1 0 0 0 ⎥ ⎢ 0 ⎥

⎢ 0 −234.61 0 234.61 0 0 ⎥ ⎢ 0 ⎥

⎢ ⎥ ⎢ ⎥

⎢⎣ 0 0 0 0 0 −20.2 ⎥⎦ ⎢⎣ 20.2 ⎥⎦

period mode of the aircraft with the elevator fixed, and calculate

the characteristic polynomial and its roots corresponding to the

mode.

(ii) Hence, or otherwise, obtain a pair of first-order equations to

approximate the phugoid oscillations of the aircraft with the ele-

vator fixed, and calculate the damped natural frequency and the

damping ratio corresponding to these oscillations.

6.10 The complete lateral equations of the AFTI-16 fighter may be expressed as

⎡ Δvs ⎤ ⎡ yv yp y ra − U es yφ 0 ⎤ ⎡ Δvs ⎤

⎢ Δp ⎥ ⎢ ⎥⎢ ⎥

s

⎥ ⎢ lv lp lr 0 0 ⎥ ⎢ Δps ⎥

d ⎢

⎢ Δrs ⎥ = ⎢ nv np nr 0 0 ⎥ ⎢ Δrs ⎥

dt ⎢ ⎥ ⎢ ⎥⎢ ⎥

⎢ Δφs ⎥ ⎢ 0 1 0 0 0 ⎥ ⎢ Δφs ⎥

⎢ Δψ s ⎥ ⎢ 0 0 1 0 0 ⎥⎦ ⎢⎣ Δψ s ⎥⎦

⎣ ⎦ ⎣

⎡ yξ y ηd yca yζ ⎤

⎢l ⎡ Δξ ⎤

lηd lca lζ ⎥⎥ ⎢

⎢ ξ Δηd ⎥⎥

+ ⎢ nξ nηd nca nζ ⎥ ⎢ .

⎢ ⎥ ⎢ Δca ⎥

⎢0 0 0 0 ⎥⎢

Δζ ⎦

⎥

⎢0 0 0 0⎦ ⎥ ⎣

⎣

Under what conditions is this mode stable?

(ii) Hence, or otherwise, obtain a pair of first-order equations to

approximate the Dutch roll oscillations of the aircraft. Establish the

conditions for stability for the Dutch roll oscillations.

6.11 (i) Show that for a straight, untapered wing (i.e. one having a rect-

angular planform) having a constant spanwise load distribution

www.ebook777.com

free ebooks ==> www.ebook777.com

262 Flight Dynamics, Simulation, and Control

yaw rate and the wing lift, in terms of the wing lift coefficient CLw, is

CLw

CLr wing = .

6

Estimate the wing rolling moment due to yaw rate in terms of

the wing lift coefficient CLw for a tapered wing with a taper ratio

equal to λ.

(ii) Show that for a straight, untapered wing (i.e. one having a rectan-

gular planform) having an elliptical spanwise load distribution,

the wing rolling moment due to yaw rate and the wing lift, in

terms of the wing lift coefficient CLw, is

CLw

CLr wing = .

8

Estimate the wing rolling moment due to yaw rate in terms of

the wing lift coefficient CLw for a tapered wing with a taper ratio

equal to λ.

(iii) Show that for a straight, untapered wing (i.e. one having a rectan-

gular planform), the contribution to roll damping of a wing due

to the wing lift, having an elliptical span loading in terms of the

wing lift curve slope CLw,α, is

−CLw , α

CLp wing = .

16

Estimate the wing roll damping in terms of the wing lift curve

slope CLw,α for a tapered wing with a taper ratio equal to λ.

(iv) Show that for a straight, untapered wing (i.e. one having a rectan-

gular planform) having an elliptical spanwise load distribution,

the wing yawing moment due to roll rate and the wing lift, in

terms of the wing lift coefficient CLw, is

−CLw

CNp wing = .

16

Estimate the wing yawing moment due to roll rate in terms of the

wing lift coefficient CLw for a tapered wing with a taper ratio equal to λ.

6.12 Show from the first principles that the non-dimensional lateral yawing

moment derivative with respect to the roll rate is given by

12 fin height

l

Np = −

∫ ( Cl − ∂CD ∂α ) ( c ( η) c ) η dη + S Fb 2

2

∫ aF cF ( z ) zdz.

W

−1 2 0

State all the assumptions made in deriving this approximate expression.

free ebooks ==> www.ebook777.com

Longitudinal and Lateral Linear Stability and Control 263

tion, the aspect ratio is 6.89, the wing span to mean aerodynamic

chord ratio is 6.8, the taper ratio of the wing is 0.3, the ratio of vertical

height of the horizontal tail above the wing to the aerodynamic mean

chord is 1.5, the sweep angle of the wing at mid-chord is 5°, the wing

maximum thickness-to-chord ratio is 0.12 and the fuselage fineness

ratio is 2.12, e = 0.7766, VT = 0.6, lt c = 5.68 and CD0 = 0.1. Make any

other suitable assumptions and estimate the primary longitudinal

non-dimensional stability derivatives of the aircraft.

⎡ Δα ⎤ ⎡ −1.2763 1 ⎤ ⎡ Δα ⎤

6.1 (i) ⎢ ⎥ = ⎢

⎣ Δq ⎦ ⎣ 1.0218 −2.4052 ⎥⎦ ⎢⎣ Δq ⎥⎦

Δ ( λ ) = λ 2 + 3.6815λ + 2.0480 = 0. Approximate short period roots are

−0.683, −2.9985.

(ii) Phugoid: damping ratio is 0.0555; natural frequency 0.0982 rad/s.

6.2 (ii) ω nsp = 0.8911, short period circular natural frequency in rad/s;

ζ sp = 0.7896, short period damping ratio; ωdsp = 0.5468, short period

circular damped natural frequency in rad/s; ωnph = 0.1584, phugoid cir-

cular natural frequency in rad/s; ζph = 0.0107, phugoid damping ratio;

ωdph ≈ 0.1584, phugoid circular damped natural frequency in rad/s.

6.3 (i) Δp = −2.4923Δp. Trs = 0.4012 s.

(ii) Δ(λ) = λ 3 + 1.1465λ2 + 8.8057λ + 0.4907 = 0.

(iii) Spiral root = −0.0561, stable.

6.4 (i) Δp = −1.5392Δp. Trs = 0.6497 s.

(ii) Δ(λ) = λ 3 + 0.4410λ2 + 0.1284λ + 0.0073 = 0.

(iii) Spiral root = −0.0718, stable.

T

6.5 (i) Δp = −0.746Δp + ⎡⎣ −1.61 0.42 ⎤⎦ ⎡⎣ξ ζ ⎤⎦ . Trs = 1.34 s .

(iii) Damping ratio, 0.0865; damped natural frequency, 1.9696 rad/s.

(iii) Δϕ = 0.8092 rad = 46.4°.

6.6 (i) Δp = −0.893601Δp. Trs = 1.12 s.

Δ(λ) = λ2 + 2.2038λ + 2.7930 = 0. Damped natural frequency 1.26 and

(ii)

damping ratio 0.659 (natural frequency = 1.67 rad/s).

(iii) Δφ = −0.0997 Δφ, stable.

6.7 (i) Δϕ = 0.33581 rad = 19.24°.

(ii) The manoeuvre can be executed with the alternate pair of controls.

www.ebook777.com

free ebooks ==> www.ebook777.com

264 Flight Dynamics, Simulation, and Control

References

1. Heffley, R. K. and Jewell, W. F., Aircraft handling qualities data, National

Aeronautics and Space Administration, NASA Contractor Report, NASA

CR-2144, December 1972.

2. Shapiro, A. H., The Dynamics and Thermodynamics of Compressible Fluid Flow,

Vol. 1, Ronald Press, New York, 1953.

3. Laitone, E. V., New compressibility correction for two-dimensional subsonic

flow, Journal of the Aeronautical Sciences (Institute of the Aeronautical Sciences),

18(5), 350–350, 1951.

4. Anonymous, USAF Stability and Control DATCOM, Flight Control Division,

Air Force Flight Dynamics Laboratory, Wright-Patterson Air Force Base, OH,

October 1960, Revised April 1976.

5. Raymer, D. P., Aircraft Design: A Conceptual Approach, AIAA education series,

American Institute of Aeronautics and Astronautics (AIAA), Washington, DC,

1989.

6. Niţă, M. and Scholz, D., Estimating the Oswald Factor from basic aircraft geo-

metrical parameters, Deutscher Luft- und Raumfahrtkongress, Berlin, Germany,

2012.

7. Spalding, D. and Chi, S., The drag of a compressible turbulent boundary layer

on a smooth flat plate with and without heat transfer, AIAA Journal, 18(1), 117–

143, 1964.

8. ESDU 69011, Heat transfer by forced convection between a two-dimensional

turbulent boundary layer and a smooth flat plate, with application to wedges,

cylinders and cones, Engineering Services Data Unit (ESDU), IHS Inc., London,

U.K., 1969.

9. ESDU 68020, The compressible two-dimensional turbulent boundary layer,

both with and without heat transfer, on a smooth flat plate, with application to

wedges, cylinders, and cones, Engineering Services Data Unit (ESDU), IHS Inc.,

London, U.K., 1988.

10. Gur, O., Mason, W. H., and Schetz, J. A., Full-configuration drag estimation,

Journal of Aircraft, 47(4), 1356–1367, 2010.

11. Hoerner, S. F., Fluid Dynamic Drag, Otterbein Press, Dayton, OH, 1951.

12. Torenbeek, E., Synthesis of Subsonic Airplane Design, Springer, The Netherlands,

1982.

13. Nicolai, L. M., Estimating R/C model aerodynamics and performance, Lockheed

Martin Aeronautical Company Paper, Marietta, GA, June 2009.

14. Nicolai, L. M., Fundamentals of Aircraft Design, E.P. Domicone, Fairborn, OH,

1975.

15. Hancock, G. J., An Introduction to the Flight Dynamics of Rigid Aeroplanes, Ellis

Horwood, New York, 1995, Section III.5–III.6.

free ebooks ==> www.ebook777.com

7

Aircraft Dynamic Response: Numerical

Simulation and Non-Linear Phenomenon

7.1 Introduction

The focus of this chapter is the aircraft dynamic response due to inputs to

the controls provided by the pilot as well as the response due to disturbance

inputs that are generated by atmospheric gusts and turbulence. The latter

inputs tend to exert disturbance forces and moments on the aircraft and it

is important that the influence of such disturbances is realistically assessed.

Simulation of aircraft dynamics has also been discussed at length by Stevens

and Lewis [1] and by Zipfel [2].

The longitudinal small perturbation aircraft equations of motion (EOMs)

were shown in Chapter 5 to be

⎢0 ⎢ w ⎥ ⎢

⎢ 1 − Zw −Zq 0 ⎥⎥ ⎢ Δw s

⎥ = ⎢ Zu Zw Zq + U es g sin θe ⎥⎥ ⎢⎢ Δws ⎥⎥

⎢0 − Mw 1 − Mq 0 ⎥ Δqs ⎥ ⎢ Mu

⎢ Mw Mq 0 ⎥ ⎢ Δqs ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣0 0 0 1 ⎦ ⎢⎣ Δθ s ⎥⎦ ⎣ 0 0 1 0 ⎦ ⎣ Δθs ⎦

⎡ Xη Xτ ⎤ ⎡ X NB − X NBe m ⎤

⎢Z ⎢ ⎥

η Zτ ⎥⎥ ⎡ Δη⎤ ⎢ ZNB − ZNBe m ⎥

+⎢ ⎢ ⎥+

⎢ Mη Mτ ⎥ ⎢ Δτ ⎥ ⎢MNB − MNBe s ⎥

⎢ ⎥⎣ ⎦ ⎢ ⎥

⎣ 0 0 ⎦ ⎢⎣ 0 ⎥⎦

(7.1)

265

www.ebook777.com

free ebooks ==> www.ebook777.com

266 Flight Dynamics, Simulation, and Control

mum power. A throttle setting of Δτ > 1 corresponds to the case when the

power exceeds the maximum.

One may multiply the aforementioned set of coupled equations by M−1

where

⎡1 −X w 0 0⎤

⎢0 1 − Zw 0 0 ⎥⎥

M=⎢ . (7.2)

⎢0 − Mw 1 0⎥

⎢ ⎥

⎣0 0 0 1⎦

Write the set of first-order equations in the so-called concise form given by

⎢0 ⎢ ⎥

⎢ 1 − zq 0 ⎥⎥ ⎢ Δw s ⎥ ⎢⎢ zu zw zq zθ ⎥⎥ ⎢⎢ Δws ⎥⎥

=

⎢0 0 1 − mq 0 ⎥ ⎢ Δq s ⎥ ⎢ mu mw mq mθ ⎥ ⎢ Δqs ⎥

⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣0 0 0 1 ⎦ ⎣⎢ Δθ s ⎦⎥ ⎣ 0 0 1 0 ⎦ ⎣ Δθs ⎦

⎡ xη xτ ⎤ ⎡ ( − )m ⎤

⎢z ⎢ ⎥

zτ ⎥⎥ ⎡ Δη⎤ ( − )m

⎢ ⎥ + M−1 ⎢ ⎥

η

+⎢

⎢ mη mτ ⎥ ⎢ Δτ ⎥ ⎢( − ) s ⎥

⎢ ⎥⎣ ⎦ ⎢ ⎥

⎣0 0⎦ ⎢⎣ 0 ⎦⎥

(7.3)

Ignoring the smaller derivatives in the left side of the equation and all other

external forces except the elevator and throttle inputs on the right-hand side,

the simplified equations are

⎡ Δu s ⎤ ⎡ xu xw xq xθ ⎤ ⎡ Δus ⎤ ⎡ xη xτ ⎤

⎢ Δw ⎥ ⎢

⎢ s ⎥ = ⎢ zu zw zq zθ ⎥⎥ ⎢⎢ Δws ⎥⎥ ⎢⎢ zη zτ ⎥⎥ ⎡ Δη⎤ (7.4)

+ ⎢ ⎥

⎢ Δq s ⎥ ⎢ mu mw mq mθ ⎥ ⎢ Δqs ⎥ ⎢ mη mτ ⎥ ⎢ Δτ ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎣ ⎦

⎣⎢ Δθs ⎦⎥ ⎣ 0 0 1 0 ⎦ ⎣ Δθ s ⎦ ⎣ 0 0⎦

⎡ Δu s ⎤ ⎡ xu U se xw xq xθ ⎤ ⎡ Δus ⎤ ⎡ xη xτ ⎤

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢

⎢ Δα s ⎥ ⎢ zu zq zθ ⎥ ⎢ Δα s ⎥ ⎢ zη zτ ⎥⎥ ⎡ Δη⎤

zw

⎢ ⎥ = ⎢ U se U se U se ⎥ ⎢ ⎥ + ⎢ U se U se ⎥ ⎢ ⎥

⎢ Δq s ⎥ ⎢

mθ ⎥ ⎢ Δqs ⎥ ⎢⎢ mη

⎥⎢ ⎥

mτ ⎥ ⎢⎣ Δτ ⎥⎦

⎢ ⎥ ⎢ mu U se mw mq ⎥

⎢ Δθ ⎥ ⎢ 0

⎣ s⎦ ⎣ 0 1 0 ⎥⎦ ⎢⎣ Δθs ⎥⎦ ⎢⎣ 0 0 ⎥⎦

free ebooks ==> www.ebook777.com

Aircraft Dynamic Response 267

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ Δα s ⎥ ⎢ Δα s ⎥ ⎡ Δη⎤ ⎢ Δw s ⎥ ⎢ Δws ⎥ ⎡ Δη⎤

⎢ ⎥ = A⎢ ⎥ + B ⎢ ⎥ or ⎢ ⎥ = A⎢ ⎥ + B ⎢ ⎥ , (7.5)

⎢ Δq s ⎥ ⎢ Δqs ⎥ ⎢⎣ Δτ ⎥⎦ ⎢ Δq s ⎥ ⎢ Δqs ⎥ ⎢⎣ Δτ ⎥⎦

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ Δθ ⎥ ⎢ Δθs ⎥ ⎢ Δθ ⎥ ⎢ Δθs ⎥

⎣ s⎦ ⎣ ⎦ ⎣ s⎦ ⎣ ⎦

that is, as

⎡ Δus ⎤ ⎡ Δus ⎤

⎢ ⎥ ⎢ ⎥

⎢ Δα s ⎥ ⎢ Δws ⎥ ⎡ Δη⎤

x = Ax + Bu with x = ⎢ ⎥ or x = ⎢ ⎥ and u = ⎢ ⎥ . (7.6)

⎢ Δqs ⎥ ⎢ Δqs ⎥ ⎣ Δτ ⎦

⎢ ⎥ ⎢ ⎥

⎢ Δθs ⎥ ⎢ Δθs ⎥

⎣ ⎦ ⎣ ⎦

⎢0 ⎢ ⎥

⎢ 1 − L p − I xz I xx − Lr

s s

0 ⎥⎥ ⎢ Δp s ⎥

⎢0 − I xz

s

I zz

s

− N p 1 − N r 0 ⎥ ⎢ Δrs ⎥

⎢ ⎥⎢ ⎥

⎣0 0 0 1 ⎦ ⎣⎢ Δφ s ⎦⎥

⎡ Yv Yp Yr − U es g cos φe coss θe ⎤ ⎡ Δvs ⎤

⎢ ⎥⎢ ⎥

=⎢

Lv Lp Lr 0 ⎥ ⎢ Δps ⎥

⎢ Nv N p Nr 0 ⎥ ⎢ Δrs ⎥

⎢ ⎥⎢ ⎥

⎢⎣ 0 1 0 0 ⎥⎦ ⎣ Δφs ⎦

⎡Y Y ⎤ ⎡( YN − YN e ) ⎤

⎢L ⎥ ⎢ ⎥

L ⎥ ⎡ Δ⎤ ⎢( LN − LN e ) I xx ⎥

s

+⎢ + (7.7)

⎢ N N ⎥ ⎢⎣ Δ⎥⎦ ⎢( N N − N N e ) I zz

s ⎥

⎢ ⎥ ⎢ ⎥

⎣0 0 ⎦ ⎢

⎣ 0 ⎥

⎦

One may multiply the aforementioned set of coupled equations by M−1 where

⎡1 0 0 0⎤

⎢0 1 −I s

I s

0 ⎥⎥

xz xx

M=⎢ (7.8)

⎢0 −I s

xz I s

zz 1 0⎥

⎢ ⎥

⎣0 0 0 1⎦

www.ebook777.com

free ebooks ==> www.ebook777.com

268 Flight Dynamics, Simulation, and Control

and write the set of first-order equations in the so-called concise form given by

⎡1 − y p − y r 0 ⎤ ⎡ Δv s ⎤ ⎡ yv yp yr yφ ⎤ ⎡ Δvs ⎤

⎢0 ⎢ ⎥

⎢ 1 − lp −lr 0 ⎥⎥ ⎢ Δp s ⎥ ⎢⎢ lv lp lr lφ ⎥⎥ ⎢⎢ Δps ⎥⎥

=

⎢0 −np 1 − nr 0 ⎥ ⎢ Δrrs ⎥ ⎢ nv np nr nφ ⎥ ⎢ Δrs ⎥

⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥

⎣0 0 0 1 ⎦ ⎢⎣ Δφ s ⎥⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφs ⎦

⎡ yξ yζ ⎤ ⎡ ( − ) ⎤

⎢l ⎥ ⎢ ⎥

lζ ⎥ ⎡ Δξ ⎤ ( − ) s

⎢ ⎥ + M−1 ⎢ ⎥

ξ

+⎢

⎢ nξ nζ ⎥ ⎢ Δζ ⎥ ⎢( − ) s ⎥

⎢ ⎥⎣ ⎦ ⎢ ⎥

⎣0 0⎦ ⎢⎣ 0 ⎥⎦

(7.9)

Ignoring the smaller derivatives and all other external forces except the

aileron and rudder inputs, the simplified equations are

⎡ Δv s ⎤ ⎡ yv yp yr yφ ⎤ ⎡ Δvs ⎤ ⎡ yξ yζ ⎤

⎢ Δp ⎥ ⎢

⎢ s ⎥ = ⎢ lv lp lr lφ ⎥⎥ ⎢⎢ Δps ⎥⎥ ⎢⎢ lξ lζ ⎥⎥ ⎡ Δξ ⎤ (7.10)

+ ⎢ ⎥

⎢ Δrs ⎥ ⎢ nv np nr nφ ⎥ ⎢ Δrs ⎥ ⎢ nξ nζ ⎥ ⎢ Δζ ⎥

⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎣ ⎦

⎢⎣ Δφ s ⎥⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφs ⎦ ⎣ 0 0⎦

or as

yp yr yφ ⎤ ⎡y yζ ⎤

⎡ Δβ s ⎤ ⎡⎢ yv s⎥⎡

Δβs ⎤ ⎢ ξs

⎢ ⎥ U es U es Ue ⎢ U U es ⎥ Δξ

⎢ Δp s ⎥ = ⎢⎢ U sl ⎥ Δps ⎥ ⎢ e ⎥⎡ ⎤

lp lr lφ ⎥ ⎢ ⎥ + ⎢ lξ lζ ⎥ ⎢ ⎥ (7.11)

⎢ Δrs ⎥ ⎢ se v ⎢ Δrs ⎥ ⎢

⎢ ⎥ U e nv np nr nφ ⎥ ⎢ n nζ ⎥ ⎢⎣ Δζ ⎥⎦

⎢⎣ Δφ s ⎥⎦ ⎢⎢ ⎥ Δφs ⎥ ⎢ ξ ⎥

⎣ 0 1 0 0 ⎥⎦ ⎣ ⎦ ⎢⎣ 0 0 ⎥⎦

where the linear relationship between the sideslip angle Δβs and Δvs is

Δvs

Δβs = . (7.12)

U es

space form as

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ Δp s ⎥ ⎢ Δps ⎥ ⎡ Δξ ⎤ ⎢ Δp s ⎥ ⎢ Δps ⎥ ⎡ Δξ ⎤

⎢ ⎥ = A⎢ ⎥ + B ⎢ ⎥ or as ⎢ ⎥ = A⎢ ⎥ + B ⎢ ⎥ , (7.13)

⎢ Δrs ⎥ ⎢ Δrs ⎥ ⎢⎣ Δζ ⎥⎦ ⎢ Δrs ⎥ ⎢ Δrs ⎥ ⎢⎣ Δζ ⎥⎦

⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎢ Δφ ⎥ ⎢ Δφs ⎥ ⎢ ⎥ ⎢ Δφs ⎥

⎣ s⎦ ⎣ ⎦ ⎣ Δφs ⎦ ⎣ ⎦

free ebooks ==> www.ebook777.com

Aircraft Dynamic Response 269

that is, as

⎡ Δvs ⎤ ⎡ Δβs ⎤

⎢ ⎥ ⎢ ⎥

⎢ Δps ⎥ ⎢ Δps ⎥ ⎡ Δξ ⎤

x = Ax + Bu with x = ⎢ ⎥ or x = ⎢ ⎥ and u = ⎢ ⎥ . (7.14)

⎢ Δrs ⎥ ⎢ Δrs ⎥ ⎣ Δζ ⎦

⎢ ⎥ ⎢ ⎥

⎢ Δφs ⎥ ⎢ Δφs ⎥

⎣ ⎦ ⎣ ⎦

state of equilibrium in a uniform uncontrolled flight with constant forward

speed and their principal features were discussed in Section 6.2.1. These

modes were the short period, the phugoid, the roll subsidence, the Dutch roll

and the spiral modes. For completeness, we observe that th