You are on page 1of 678

free ebooks ==> www.ebook777.

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

Boca Raton London New York

CRC Press is an imprint of the


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

© 2015 by Taylor & Francis Group, LLC


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

No claim to original U.S. Government works


Version Date: 20140707

International Standard Book Number-13: 978-1-4665-7336-9 (eBook - PDF)

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

and the CRC Press Web site at


http://www.crcpress.com
free ebooks ==> www.ebook777.com

To my teachers, Hans Wagner, Horst Leipholz,

Holt Ashley, Art Bryson and Geoff Hancock

www.ebook777.com
free ebooks ==> www.ebook777.com
free ebooks ==> www.ebook777.com

Contents

List of Acronyms................................................................................................. xvii


Preface.................................................................................................................... xix
Author.................................................................................................................. xxiii

1 Introduction to Flight Vehicles..................................................................... 1


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 Basic Principles Governing Aerodynamic Flows................................... 33


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 Mechanics of Equilibrium Flight............................................................... 67


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

3.5 Stability in the Vicinity of the Minimum Drag Speed...................77


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 Aircraft Non-Linear Dynamics: Equations of Motion......................... 103


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

5 Small Perturbations and the Linearised, Decoupled Equations


of Motion....................................................................................................... 145
5.1 Introduction........................................................................................ 145
5.2 Small Perturbations and Linearisations......................................... 145

www.ebook777.com
free ebooks ==> www.ebook777.com
x Contents

5.3 Linearising the Aerodynamic Forces and Moments: Stability


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 Longitudinal and Lateral Linear Stability and Control...................... 185


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

6.6.2 Perturbation Analysis of Longitudinal Trimmed


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

7 Aircraft Dynamic Response: Numerical Simulation


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 Aircraft Flight Control............................................................................... 333


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

8.6 Application to the Design of Stability Augmentation


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

Chapter Highlights....................................................................................... 474


Exercises......................................................................................................... 475
Answers to Selected Exercises....................................................................484
References...................................................................................................... 485

9 Piloted Simulation and Pilot Modelling................................................ 487


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 Flight Dynamics of Elastic Aircraft......................................................... 529


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

10.4.8 Transformation to Centre of Mass Coordinates........... 550


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

In the last decade, we have seen a phenomenal increase in air travel to


­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

MATLAB® is a registered trademark of The MathWorks, Inc. For product


information, please contact:

The MathWorks, Inc.


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

modelling of damage in laminated composite plates, non-linear flutter anal-


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.

1.2  Components of an Aeroplane


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.

1.2.3  Tail Surfaces or Empennage


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.

1.2.4  Landing Gear


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  Basic Principles of Flight


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.

1.3.2  Drag and Its Reduction


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.

1.3.3  Aerodynamically Conforming Shapes: Streamlining


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.

1.3.4  Stability and Balance


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.

1.4  Flying Control Surfaces: Elevator, Ailerons and Rudder


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

Engines Underwing leading


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.

1.4.1  Flaps, High-Lift and Flow Control Devices


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:

1. Short chord/short span passive devices


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).

1.4.2  Introducing Boundary Layers


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).

to a random loading on the wing that results in the so-called phenomenon of


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

Thus, as a consequence of these openings or slots, separation now occurs at


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.

actuated by one of several hydraulic channels to provide for redundancy and


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.

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


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.

1.6  Modes of Flight


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

attitude affect the forces acting on an aeroplane. Flight at constant velocity


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.

1.6.1  Static and In-Flight Stability Margins


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

1.7  Power Plant


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.

1.7.1  Propeller-Driven Aircraft


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.

1.7.2  Jet Propulsion


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

turbine which is employed as a normal power unit to convert as much as


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.

1.8  Avionics, Instrumentation and Systems


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:

1. Stand-alone standard avionics equipment for communication, navi-


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.

A detailed description of each of these classes of systems or even their


generic features is well beyond the scope of this introductory section.
free ebooks ==> www.ebook777.com
Introduction to Flight Vehicles 21

1.9  Geometry of Aerofoils and Wings


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.

1.9.1  Aerofoil Geometry


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.

1.9.2  Chord Line


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).

1.9.4  Leading and Trailing Edges


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

1.9.5  Specifying Aerofoils


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,

The first digit: Tells us that the aerofoil is a 6-series aerofoil


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

1.9.6  Equations Defining Mean Camber Line


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

In practice, it is customary to employ more than one curve to model the


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.

1.9.7  Aerofoil Thickness Distributions


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)

where α positive angle of slope at chordwise position x determined by


­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⎦

1.9.8  Wing Geometry


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 ⎟⎠

Notes: c, chord; s, semi-span; b, span = 2s; η = y/s.


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 )

Mean aerodynamic chord c=


( 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⎠

Typically, a wing of arbitrary planform shape is divided into J trape-


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 ).

What is the aspect ratio of the planform?


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

1.3 Consider a trapezoidal planform.


(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

where 0 ≤ n ≤ 1 is the chord fraction (e.g. 0 for the leading edge,


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 ⎟ .

Hence, show that

⎛ ⎛ 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

What is the aspect ratio of the planform?


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.

Answers to Selected Exercises

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.

2.2  Continuity Principle


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

2.2.1  Streamlines and Stream Tubes


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.

2.3  Bernoulli’s Principle


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.

2.4  Laminar Flows and Boundary Layers


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].

2.5  Turbulent Flows


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.

2.6  Aerodynamics of Aerofoils and Wings


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:

1. Flow effects such as compressibility and viscosity


2. Aerofoil and wing shape and geometry parameters
3. Size and scale effects
4. Orientation of the body relative to the flow

An understanding of the flow past an aerofoil is essential for a holistic under-


standing of the aerodynamics of aerofoils and wings.

www.ebook777.com
free ebooks ==> www.ebook777.com
36 Flight Dynamics, Simulation, and Control

2.6.1  Flow around an Aerofoil


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.

2.6.2  Mach Number and Subsonic and Supersonic Flows


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  Properties of Air in the Atmosphere


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.

2.7.2  Air Density


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

reduces to 3% of its corresponding sea level value. At an altitude of 220 km,


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.

2.7.5  Effects of Pressure and Temperature


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

Thus, the density ρ is given by

N
ρ= (2.2)
v

and the universal gas law may also be expressed as

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

2.7.7  Bulk Modulus of Elasticity


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

The square of the velocity of sound is given by

K
a2 = . (2.6)
ρ

Assuming that the fluid is barotropic, the pressure–density relationship is


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)

2.7.8  Temperature Variations with Altitude: The Lapse Rate


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.

2.8 International 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:

Pressure, p0 = 1.01325 × 105 N/m2 Density, ρ 0 = 1.22505 kg/m3


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:

1. For altitudes in the range 0 ≤ h (in km) ≤ 11 km (troposphere),

Tropospheric lapse rate, α = 6.5 ( Lk ,1 = 6.5 K/km ) , (2.8a)


Temperature variation with altitude h, T1 = T0 − α × h, (2.8b)


⎛α ⎞
Temperature ratio, Tratio = 1 − ⎜ ⎟ × h, (2.8c)
⎝ T0 ⎠

g
Pressure ratio exponent , n = = 5.256, (2.8d)
α×R

Density ratio = Tratio (n − 1), (2.8e)


Pressure ratio = Tratio n, (2.8f)

a1
Ratio of the speed of sound, = Tratio ( 0.5 ) , (2.8g)
a0

Density at height, h = 1.22505 × Density ratio. (2.8h)


2. For altitudes in the range 11 ≤ h (in km) ≤ 20 km (stratosphere), which


is a constant temperature region,

Temperature, T2 †= −56.5°C = 216.65 K, (2.9a)


α = 0.0 ( Stratospheric lapse rate ) , (2.9b)


Over this entire altitude range, ratio of the speed of sound,

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

7. For altitudes in the range 110 ≤ h (in km) ≤ 120 km,


T = T9 − Lk ,9 ( h − ha ) (2.14)

where
T9 = 240 K
Lk,9 = 12.0 K/km
ha = 110.0 km

8. For altitudes in the range, 120 ≤ h (in km) ≤ 150 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

For altitudes above 86 km, ρ = p/RT and p is given by the following


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

2.9  Generation of Lift and Drag


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 ⎠

Thus, the lift is given by

c c
⎛ qu + ql ⎞
L=
∫ ( pl − pu ) dx = ρ∫ ( qu − ql ) ⎜ U +
0 0
⎝ 2 ⎟ dx. (2.20)

In the limit when the U ≫ (qu + ql)/2,

c c

L=
∫ ( p − p ) dx = ρU ∫ ( q − q ) dx = ρUΓ. (2.21)
0
l u
0
u l

This expression may also be written as


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

This is the Kutta–Joukowski theorem that relates circulation, and therefore


vorticity, around an aerofoil to the lift.
free ebooks ==> www.ebook777.com
Basic Principles Governing Aerodynamic Flows 47

2.10  Aerodynamic Forces and Moments


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

where the circulation, Γ, is defined to be positive in the clockwise direction.


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.

one must necessarily consider a subdomain enclosing the body to evaluate


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

It is possible, in principle, to predict the time-dependent forces and


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

L = N cos α + T sin α = ρU c γ . (2.27)


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

L = K γρU 2 c sin α (2.29)


where it is assumed that γ = K γU sin α .


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

L = πρU 2 c sin α. (2.31)


Thus, the simplified analysis presented earlier is valid provided Kγ = π and


the non-dimensional lift force per unit span may be expressed as

L
CL = = 2Γˆ = 2π sin α. (2.32)
1
ρU c
2

2.10.1  Aerodynamic Coefficients


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

An important feature of the lift coefficient, CL, is that it may be analytically


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)

while the actual experimentally determined value is usually much lower.


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.

2.10.2  Aerofoil Drag


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

Equivalent airspeed (EAS) α


(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.

2.10.3  Aircraft Lift Equation and Lift Curve Slope


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)

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

A typical experimentally determined plot of CL versus α is shown, for wings


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

CLT = a1αT + a2η + a3β


(2.41)

where CLT is also defined as

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

where i0 is the incidence angle for zero lift.

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)

Aircraft angle of attack = α


x-axis

U0

Horizontal
(b) γ = Climb angle

iw = wing setting 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.

Thus, αT is the tail plane angle of attack is expressed as

⎛ ∂ε ⎞
αT = ( α w − i0 ) ⎜ 1 − ⎟ + iT + i0 − iw . (2.46)
⎝ ∂α ⎠

The tail plane angle of attack may then be expressed as

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

2.10.4  Centre of Pressure


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.

2.10.5  Aerodynamic Centre


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.

2.10.6  Pitching Moment Equation


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,

Mcg = Mwb + Lwb ( xcg − x ac ) − LT lT , (2.52)


where lT is the horizontal distance of the tail plane AC to the aircraft CG


(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

Dividing the pitching moment equation by ρU 02SW c 2,

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

and it follows that

CMcg = C M0 +
( xcg − xac ) C − CLT
ST lt
(2.55)
L
c SW c

where lt = lw + lT, lw is the horizontal distance of the wing AC to the aircraft


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 ⎠

Unless otherwise stated, it is assumed that Sref = SW = S.

www.ebook777.com
free ebooks ==> www.ebook777.com
60 Flight Dynamics, Simulation, and Control

We may further assume that

∂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 ⎠

where hac represents the non-dimensional position of the wing’s AC hn repre-


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

CMcg = C M0 + CL ( hcg − hn ) − VT ( a1ηT + a2η + a3β ) (2.64)


where

ST T
CL = CLwb + CL . (2.65)
SW

The aforementioned equation is normally known as the stick fixed pitching


moment equation and plays a key role in the static stability, stick fixed trim
and control of the aircraft.

2.10.7  Elevator Hinge Moment Coefficient


The elevator hinge moment coefficient may be expressed in a manner similar
to the tail plane lift coefficient as

CH = b1αT + b2η + b3β (2.66)



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

Show that the centre of lift is given by

4s
yCL = .

Hence, show that for a trapezoidal planform,


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

For an elliptic wing, it may be assumed that

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

(i) Determine the lift curve slope of the wing.


(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.

Answers to Selected Exercises


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

2.7 (i) α0 = 0°, CL = 0.09163, CH = 0.002356, L = 50.511 N, H = 0.01558 N m. (ii) η = 5°.


(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

L = mg cos γ , T − D = mg sin γ , (3.2)


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

L = mg cos ( − γ ) , D − T = mg sin ( − γ ) . (3.3)



In the case of gliding flight, T = 0 and the equilibrium equations are

L = mg cos ( − γ ) , D = mg sin ( − γ ) . (3.4)



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)

mg = Ytrim sin ( φ ) + ∑ L cos ( φ


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

2. Turn coordination and banking: When an aeroplane is making a hor-


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)

where mUΩ is the d’Alembert force (so-called centrifugal force)


­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

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

3.2  Speeds of Equilibrium Flight


Assuming the lift is given in terms of the aerodynamic lift coefficient as

1
L= ρU 02CLS, (3.11)
2

the condition for vertical equilibrium, L = mg, yields

2mg
U0 ≈ . (3.12)
ρ C LS

Thus, the stalling speed of the aircraft is given by

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

Hence, it follows that

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

Hence, it follows that the equivalent sea level gliding speed is

2mg cos ( − γ ) 2mg C L CD


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

where h = U sin γ. In general, this equation must be integrated to obtain the


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)

the steady rate of climb is given by

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)

Hence, eliminating U0 and cos γ,


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,

ηP0 Te 0 2mg CD 2mg


+ = , (3.23)
mg mg ρCLS CL ρCLS

which allows one to estimate the minimum power required to climb.


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  Basic Aircraft Performance


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

CD Minimum Minimum Maximum


Power Drag Speed/Drag
speed speed ratio speed

Total drag

Equivalent airspeed (EAS)

FIGURE 3.4
Definition of various optimum flight speeds.

which may be expressed as

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/CL is minimum when


CD 0
K− = 0. (3.29)
CL2
free ebooks ==> www.ebook777.com
Mechanics of Equilibrium Flight 75

Equation 3.29 is satisfied when

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

The EAS corresponding to minimum power is

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

The EAS corresponding to minimum power is

2L U min drag
U min power = = 1/ 4 . (3.39)
ρ0 ( CL )max drag /speed S ⎛1⎞
⎜ ⎟
⎝3⎠

The estimation of these speeds provides an idea of the range of speeds at


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.

3.4  Conditions for Minimum Drag


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

First, it is observed that the differential of the dynamic pressure q with


respect to the total velocity U is

dq
= ρU . (3.41)
dU
free ebooks ==> www.ebook777.com
Mechanics of Equilibrium Flight 77

Then, differentiating the drag with respect to the dynamic pressure q,

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

3.5  Stability in the Vicinity of the Minimum Drag Speed


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.

3.6  Range and Endurance Estimation


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

depend on its aerodynamics (primarily, its drag polar), the characteristics of


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

where tsfccruise is the thrust-specific fuel consumption during cruise and


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)

which may also be expressed as

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 ⎠

For a propeller-driven aircraft, the endurance is expressed as

ηpr ( L D )cruise 2ρSCL* ⎛ 1 1 ⎞


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

CMcg = CM0 + CL ( hcg − hn ) − VT ( a1ηT + a2η + a3β ) , (3.54)


where
a1 ⎛ ∂ε ⎞
hn ≡ hac − VT ⎜1− ⎟ , (3.55)
a⎝ ∂α ⎠

the elevator angle is

CM0 − CMcg ( hcg − hn ) a η + a3β


η= + CL − 1 T . (3.56)
a2VT a2VT a2

Applying the condition for trim,

C M0 + CL ( hcg − hn )
a1ηT + a2η + a3β = . (3.57)
VT

Solving for the elevator angle, in the trimmed state,

st -fixed CM0 a1ηT + a3β ( hcg − hn )


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,

b2CM0 ⎛ b2 ( hcg − hac ) 1 ⎛ b a ⎞⎛ ∂ε ⎞ ⎞


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

the required hinge moment to maintain the aircraft in trim is

b2CM0 ⎛ b2 ( hcg − hac ) b1 ⎛ ∂ε ⎞ ⎞


CHsttrim
-fixed
= + b1ηT + b3β + ⎜ + ⎜1− ⎟ ⎟ CL , (3.62)
a2VT ⎜
⎝ a2VT a⎝ α ⎠ ⎟⎠
∂α

which may be rearranged and expressed as

b2CM0 b2 ( hcg − hʹn )


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

b2η + b3β = −b1αT , (3.65b)



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

CLT = a1αT + a2η + a3β, (3.68)


and eliminating the elevator angle again as before gives

(C ) T
L
stick -free
= a1αT + a3β. (3.69)

3.8  Stability of Equilibrium Flight


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

natural tendency to diverge from the trimmed equilibrium state. Stability of


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

generate an additional aerodynamic moment at this point, the weight of the


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.

3.9  Longitudinal Static Stability


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

CMcg = CM0 + CL ( hcg − hn ) − VT ( a1ηT + a2η + a3β ) , (3.72)


∂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

3.9.1  Neutral Point (Stick-Fixed)


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.

3.9.2  Neutral Point (Stick-Free)


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 ⎠ ⎠

The condition for stability, ∂C Mcg ∂CL < 0, now reduces to

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 ⎠

where from first of Equations 3.67b,

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.

3.10.1  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

Relative air velocities


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

where lT = lW + lT. Since the rotational acceleration is set to equal to ng’s, it


follows that

U 02
ng = , (3.77)
R

and the angular velocity of the aircraft, q, is given by

U0
q= , (3.78)
R
and we obtain

nglT
ΔαT = . (3.79)
U 02

The condition for force equilibrium is now modified as

Lpo − mg = nmg (3.80)


where Lpo is the total lift in a pull-out manoeuvre. Hence, it follows that

Lpo = ( n + 1) mg. (3.81)


It follows that

CLpo = ( n + 1) Cw = ( n + 1) CL , (3.82)

where CLpo is the coefficient of lift in a pull-out manoeuvre.

3.10.2  Manoeuvre Margin: Stick-Fixed


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

CMcg = CM0 + CL ( hcg − hn ) − VT ( a1ηT + a2η + a3β ) . (3.83)


www.ebook777.com
free ebooks ==> www.ebook777.com
88 Flight Dynamics, Simulation, and Control

In the case of the pull-out manoeuvre, CL is replaced by CLpo, ηT by ηT + ΔαT


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

and it follows that

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

and is defined as the manoeuvre margin (stick-fixed). The manoeuvre margin is


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.

the CG is at the manoeuvre point, ∂η/∂n = 0. When the aircraft is executing


an extreme manoeuvre, it should remain at least as stable as in steady flight.
Thus, the manoeuvre point must be behind the NP.

3.10.3  Manoeuvre Margin: Stick-Free


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.

3.11  Lateral Stability and Stability Criteria


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)

By a similar argument, a disturbing rolling moment also produces an inde-


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

3.12  Experimental Determination of Aircraft Stability Margins


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

CMcg = CM0 + CL ( hcg − hn ) − VT ( a1ηT + a2η + a3β ) . (3.94)


Considering the case of stick-fixed trim,

CMcg = CM0 + CL ( hcg − hn ) − VT ( a1ηT + a2η + a3β ) = 0. (3.95)



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

Hence, for a positive stability margin,

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

and for instability,

d st -fixed H
ηtrim = − n > 0. (3.99)
dCL a2VT

Given this information, one way of finding the aircraft’s NP stick-fixed is to


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.

3.13  Summary of Equilibrium- and Stability-Related Equations


1. Pitching moment equation

CMcg = CM0 + CL
( xcm − xac ) − V C .
T LT
c

2. Tail plane angle of attack

CLwb ⎛ ∂ε ⎞ CL ⎛ ∂ε ⎞
αT = ⎜1− ⎟ + ηT ≈ ⎜1− ⎟ + ηT .
a ⎝ ∂α ⎠ a ⎝ ∂α ⎠

3. Tail plane lift coefficient equation

CLT = a1αT + a2η + a3β.


4. Hinge moment equation

b1 ⎛ ∂ε ⎞
CH = b1αT + b2η + b3β, CH = CL ⎜1− ⎟ + b1ηT + b2η + b3β.
a⎝ ∂α ⎠

5. Alternate forms of pitching moment equation

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

CMcg = CM0 + CL ( hcg − hn ) − VT ( a1ηT + a2η + a3β ) ,


where

a1 ⎛ ∂ε ⎞
hn ≡ hac + VT ⎜1− ⎟.
a⎝ ∂α ⎠

6. Elevator angle in terms of pitching moment

CM0 − CMcg ( hcg − hn ) a η + a3β


η= + CL − 1 T .
a2VT a2VT a2

7. Alternate forms of hinge moment equation

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

8. Pitching moment–hinge moment relation


a2
CMcg = CM0 − VT CH + CL ( hcg − hʹn ) − VT ( a1ηT + a3β ) .
b2

9. Trim
Stick-fixed: CMcg = 0.

st - fixed CM0 a1ηT + a3β ( hcg − hn )


ηtrim = − + CL ,
a2VT a2 a2VT

b2CM0 b2 ( hcg − hʹn )


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 ⎠

b2C M0 b2 ( hcg − hʹn )


= + CL + b1ηT + b3β,
a2VT a2VT

st -free b1 ⎛ ∂ε ⎞ b1 b3
ηtrim = −CL ⎜1− ⎟ − ηT − β,
ab2 ⎝ ∂α ⎠ b2 b2

st -free b2CM0 b2 ( hcg − hʹn ) b1


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.

10. Conditions for stability: The NPs


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

12. Manoeuvre margin


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

Geometry (aerofoil section, area, sweep, dihedral angle, etc.) and


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

and the endurance,

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

(ii) For propeller-driven engines, it is customary to use the power-specific


fuel consumption rate, psfc. In this case, the general expressions for
the range and endurance are

W0 −W fuel
ηpr dW
R=− ,
W0
∫ psfcT

and for the endurance,

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 ⎠

(iii) At the start of cruise segment at a height of 12,000 m, a turboprop


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 .

(a) The propeller efficiency is known to be 0.88 and the power-specific


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

3.4 A n American T-37 aircraft has a drag coefficient CD = 0.02 + 0.057CL2 ,


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

where 2s is the wing span and c is the chord length.


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.

Answers to Selected Exercises


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.

4.2  Aircraft Dynamics


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.

4.3  Aircraft Motion in a 2D Plane


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:

mUγ = L + T sin ( δ − γ ) − mg cos γ , mU = −D + T cos ( δ − γ ) − mg sin γ , (4.1a)


 = M + Td (4.1b)

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

α = θ − γ , q = θ and h = U sin γ. (4.3)



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

φ = p, Vψ = g tan φ. (4.4)


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)

An important consequence of these equations is the conditions for trim in


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

Two different point mass approximations of aircraft dynamics are generally


in use. These are

1. The non-equilibrium or energy height model


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


mU = L − mg cos γ. (4.9b)
dt

Thus, from the first of the equations of motion,

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

Hence, it follows that

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


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

mΔh = L − mg = L − Le = −ρgACL Δh (4.14)



or

ρgACL ρg 2 ACL
Δh = − Δh = − Δh = −ω2ph Δh. (4.15)
m Le

An alternate simplification also leads to the energy state approximation. In this


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,

mΔh = L cos γ − D sin γ + T sin δ − mg (4.17a)


mx = −D cos γ + T cos δ − L sin γ. (4.17b)


www.ebook777.com
free ebooks ==> www.ebook777.com
108 Flight Dynamics, Simulation, and Control

Assuming level flight, that is, γ = δ = 0 and that T = D,

mΔh = L − mg = L − Le , (4.18a)

mx = 0, (4.18b)

where Le = mg. If we express the lift as

1
L= ρU 2 ACL , (4.19)
2

assuming that the lift coefficient, CL, is constant and that

1
Le = ρU e2 ACL , (4.20)
2

we may write

mΔh = L − mg = L − Le = −ρgACL Δh (4.21)


or

ρgACL ρg 2 ACL
Δh = − Δh = − Δh = −ω2ph Δh, (4.22)
m Le

which is identical to the one obtained earlier.


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

The phugoid is essentially an energy conservation mode. In this mode, the


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

4.4  Moments of Inertia


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

IOP = h 2dm, (4.25)



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.

By applying Pythagoras theorem, it follows that

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

is the position vector of the arbitrary point and

e = e x i + e y j + e z k (4.28)

is a unit vector in the direction of OP. Hence, h2 may also be expressed as

2
h 2 = ( r ⋅ r ) ( e ⋅ e ) − ( r ⋅ e ) = e ⋅ ⎡⎣e ( r ⋅ r ) − r ( r ⋅ e ) ⎤⎦ = e ⋅ ( r × ( e × r ) ) . (4.29)

The moment of inertia may be expressed as

∫ (( 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 ⎥⎦

is the moment of inertia matrix.

4.5  Euler’s Equations and the Dynamics of Rigid Bodies


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)

in terms of the mutually perpendicular unit vectors i, j and k in the three


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 ⎥⎦

where p, q and r are the three body components of angular velocity.


free ebooks ==> www.ebook777.com
Aircraft Non-Linear Dynamics 113

The characterisation of the motion of a rigid body in a noninertial coor-


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

Thus, the Newtonian equations of motion governing the translational


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

The cross product of ω and the velocity vector v are

⎡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 ⎥⎦

The subscript O is dropped from the vector v for brevity.


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)

The absolute angular momentum of the origin of the reference frame is


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

For rotational motion, we have


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.

4.6  Description of the Attitude or Orientation


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

1. The Obxb axis pointing forward (longitudinal axis)


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

1. The Oexe axis pointing north


2. The Oeye axis pointing east
3. The Oeze pointing towards the centre of the Earth

After performing a sequence of rotations of the space-fixed reference axes,


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

⎡cos ψ −sin ψ 0 ⎤ ⎡ cos θ 0 sinθ ⎤ ⎡ 1 0 0 ⎤


⎢ ⎥⎢ ⎥⎢ ⎥
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 ⎥ , (4.53)
⎢⎣ −sθ cθsφ cθcφ ⎥⎦

where cθ = cos θ,  cϕ = cos ϕ,  cψ = cosψ,  sθ = sin θ, sϕ = sin ϕ, and  sψ = sin ψ


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

an orthogonal matrix and so is there product. The inverse transformation is


defined by

⎡ xB ⎤ ⎡ xI ⎤ ⎡ xI ⎤
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ yB ⎥ = TIB × ⎢ y I ⎥ = TBI × ⎢ y I ⎥ , (4.54)
−1

⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢⎣ zB ⎥⎦ ⎢⎣ zI ⎥⎦ ⎢⎣ zI ⎥⎦

where

⎡1 0 0 ⎤ ⎡cos θ 0 − sin θ ⎤ ⎡ cos ψ sin ψ 0⎤


⎢ ⎥⎢ ⎥⎢ ⎥
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)

⎣⎢cψsθcφ + sψsφ sψsθcφ − cψsφ cθcφ⎥⎦

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,

pB = TBI pI and pI = TIBpB . (4.57)



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

It can also be shown that

⎡ 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 ⎥⎦

Having established the transformation relating the body-fixed axes or


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

⎡ φ ⎤ ⎡ 1 sin φ tan θ cos φ tan θ ⎤ ⎡ pB ⎤


⎢ ⎥ ⎢ ⎥ (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

⎡ φ ⎤ ⎡ 1 0 0 ⎤ ⎡ pB ⎤ ⎡0 sin φ tan θ cos φ tan θ ⎤ ⎡ pB ⎤


⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ (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 ⎥⎦

where the second term is the non-linear term.

4.7  Aircraft Equations of Motion


Hence, the Euler equations take the form

⎡ 0 −rB qB ⎤
dhB ⎢ ⎥
+ rB 0 − pB ⎥ hB = MB . (4.65)
dt ⎢
⎢⎣ −qB pB 0 ⎥⎦

When resolved in the body axes, the moment of momentum vector is


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)

In the case of many symmetric bodies, it is acceptable to assume that the


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 ⎥⎦ ⎢⎣ rB ⎥⎦ ⎢⎣ −qB pB 0 ⎥⎦ ⎢⎣ −I xz −I yz I zz ⎥⎦ ⎢⎣ rB ⎥⎦

(4.70)
Solving for p B, q B and rB, 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

rB =
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

Similarly, the Newtonian equations of motion governing the translational


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)

In order to completely define the attitude (orientation), we need to relate 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 ⎥⎥ . (4.74)
⎢ψ ⎥ ⎢0 sin φ cos θ cos φ cos θ ⎥⎦ ⎢⎣ rB ⎥⎦
⎣ ⎦ ⎣

Furthermore, consider any point, P, on the aircraft with coordinates:

P ≡ [x y z ] . (4.75)

The body components of the velocity, vP, at this point are

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

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 ⎥ , (4.77)
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.

4.8  Motion-Induced Aerodynamic Forces and Moments


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

⎡ cos β sin β 0 ⎤ ⎡ cos α 0 sin α ⎤


⎢ ⎥⎢ ⎥
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

It is the same transformation as the transformation from the body to the


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)

and then β and α are defined from the relations

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,

VT =
( )
( 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)

The velocity components may be expressed as

u = VT cos β cos α , v = VT sin β and w = VT sin α cos β. (4.84)


Furthermore, it can be shown that

⎡u⎤ ⎡VT ⎤ ⎡ u ⎤ ⎡ VT ⎤


⎢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

FAB = TBW ( α , β ) FAW and M AB = TBW ( α , β ) M AW , (4.86)



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

4.9 Non-Linear Dynamics of Aircraft Motion


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

⎡ cos βe sin βe 0 ⎤ ⎡ cos α e 0 sin α e ⎤


⎢ ⎥⎢ ⎥
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 cos βe sin βe sin α e cos βe ⎤


⎢ ⎥
= ⎢ − 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

⎡cos α e 0 −sin α e ⎤ ⎡ cos βe −sin βe 0⎤


⎢ ⎥⎢ ⎥
TBW ( α e , βe ) = ⎢ 0 1 0 ⎥ ⎢sin βe cos βe 0⎥
⎢⎣ sin α e 0 cos α e ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦

⎡cos α e cos βe − cos α e sin βe −sin α e ⎤


⎢ ⎥
=⎢ 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

For convenience, we enunciate the complete set of equations governing the


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

Thus, the translational equations of motion and the rotational equa­


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 ⎦ ⎣⎢ rB ⎦⎥ ⎢⎣ −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

⎡ u ⎤ ⎡ 0 −rb qb ⎤ ⎡ u ⎤ ⎡ − sin θ ⎤ ⎡Xe ⎤


⎢ 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⎥ ⎢
⎢⎣ rB ⎥⎦ ⎢⎣ −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

rB =
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

In order to completely define the attitude (orientation), we need to relate


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 ⎥⎦
⎣ ⎦ ⎣

Furthermore, consider any point, P, on the aircraft with coordinates

P ≡ [x y z ] . (4.104)

The body components of the velocity, vP, at this point are

u = U + zqB − yrB , (4.105a)



v = V + xrB − zpB (4.105b)

and

w = W + ypB − xqB . (4.105c)



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

⎡cos ψ −sin ψ 0 ⎤ ⎡ cos θ 0 sin θ ⎤ ⎡ 1 0 0 ⎤


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 ⎥⎦

where cθ = cos θ, cϕ = cos ϕ, cψ = cos ψ, sθ = sin θ, sϕ = sin ϕ, and sψ = sin ψ


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

4.9.1  Equations of Motion in Wind Axis Coordinates, V T, α and β


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

⎡u⎤ ⎡VT ⎤ ⎡cos α cos β ⎤


⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ v ⎥ = TBW ( α , β ) × ⎢ 0 ⎥ = ⎢ sin β ⎥ VT , (4.108)
⎢⎣ w ⎥⎦ ⎢⎣ 0 ⎥⎦ ⎢⎣ sin α cos β ⎥⎦

⎡ u ⎤ ⎡ VT ⎤ ⎡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 ⎥⎥
⎣⎢ ⎦ ⎣ ⎦

where TBW (α, β) is given by

⎡cos α 0 − sin α ⎤ ⎡cos β − sin β 0⎤


⎢ ⎥⎢ ⎥
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

⎡ cos β sin β 0 ⎤ ⎡ cos α 0 sin α ⎤


⎢ ⎥⎢ ⎥
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

The translational equations of motion are

⎡u⎤ ⎡u⎤ ⎡u⎤ ⎡ 0 w −v ⎤ ⎡ pB ⎤


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,

⎡ VT ⎤ ⎡ 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

⎡ VT ⎤ ⎡ 0 0 0 ⎤ ⎡ pB ⎤ ⎡XA ⎤


⎥ ⎢ ⎥⎢ ⎥ 1 ⎢ ⎥
⎢ VTβ ⎥ = −VT ⎢ −sα 0 cα ⎥ ⎢ qB ⎥ + ⎢ YA ⎥
⎢V α cos β ⎥ m
T ⎢⎣cαsβ −cβ sαsβ ⎥⎦ ⎢⎣ rB ⎥⎦ ⎢⎣ ZA ⎥⎦
⎣⎢ ⎦⎥

⎡ cα cβ sβ sαcβ ⎤ ⎧ ⎡ cos σ ⎤ ⎡ − sin θ ⎤ ⎫


1⎢ ⎥⎪ ⎢ ⎥ ⎢ ⎥⎪
+ ⎢ −cαsβ cβ −sαsβ ⎥ ⎨T ⎢ 0 ⎥ + mg ⎢ sin φ coss θ ⎥ ⎬ .
m
⎢⎣ −sα 0 cα ⎥⎦ ⎪⎩ ⎢⎣ − sin σ ⎥⎦ ⎢⎣cos φ cos θ ⎥⎦ ⎪


(4.119)

The equations for the body angular velocity components in body-fixed


­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

rB =
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

⎡ φ ⎤ ⎡ 1 sin φ tan θ cos φ tan θ ⎤ ⎡ pB ⎤


⎢ ⎥ ⎢
⎢ θ ⎥ = ⎢0 cos φ − sin φ ⎥⎥ ⎢⎢ qB ⎥⎥ , (4.122)
⎢ψ ⎥ ⎢0 sin φ cos θ cos φ cos θ ⎥⎦ ⎢⎣ rB ⎥⎦
⎣ ⎦ ⎣

which may also be expressed as

φ = pB + sin φ tan θqB + cos φ tan θrB , (4.123a)


θ = cos φqB − sin φrB , (4.123b)


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

⎡ xi ⎤ ⎡u⎤ ⎡VT ⎤ ⎡cos α cos β ⎤


d ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
yi = TIBv P = TIB ⎢ v ⎥ = TIBTBW ( α , β ) × ⎢ 0 ⎥ = TIB ⎢ sin β ⎥ VT
dt ⎢ ⎥
⎢⎣ zi ⎥⎦ ⎢⎣ w ⎥⎦ ⎢⎣ 0 ⎥⎦ ⎢⎣ sin α cos β ⎥⎦

i.e.

⎡ xi ⎤ ⎡ cψcθ cψsθsφ − sψcφ cψsθcφ + sψsφ ⎤ ⎡cos α cos β ⎤


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

Pitching moment coefficient

⎛ u ⎞ ⎛ c ⎞
Cm = Cm0 + Cmα α + Cmu ⎜ (
⎟ + Cmη η + Cmq q + Cmα α ⎜ ) ⎟ (4.126e)
⎝ VT∞ ⎠ ⎝ 2VT∞ ⎠

Yawing moment coefficient

⎛ 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.

4.9.2  Reduced-Order Modelling: The Short Period Approximations


The short period approximation is obtained by assuming that VT ⊕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

rB =
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

⎡ xi ⎤ ⎡ cψcθ cψsθsφ − sψcφ cψsθcφ + sψsφ ⎤ ⎡cos α cos β ⎤


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

rB =
( I xx N a + I xz La ) . (4.133c)
Δ
free ebooks ==> www.ebook777.com
Aircraft Non-Linear Dynamics 137

The lateral inertial position is given by


⎡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 ) ⎥⎦

4.10  Trimmed Equations of Motion


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

while the rotational equations may be expressed as

⎤ ⎡ 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 ⎦ ⎢⎣ rB ⎥⎦ ⎣ −qB pB 0 ⎥⎦ ⎢⎣ −I xz
s
−I yz
s
I zz ⎦ ⎢⎣ rB ⎥⎦

= M NB + TBW ( α e , βe ) M AS . (4.141)

Now, subtracting the steady-state trim equations, the Newtonian equations


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)

The Euler equations governing the rotational motion are

⎡ 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.

4.10.1  Non-Linear Equations of Perturbed Motion


The non-linear equations of perturbed motion about an equilibrium state are
given by

⎡ u ⎤ ⎡ 0 −rb qb ⎤ ⎡ u ⎤ ⎡ − ( sin θ − sin θe ) ⎤ ⎡ Xe − Xe 0 ⎤


⎢ 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⎥ ⎢
⎢⎣ rB ⎥⎦ ⎢⎣ −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 ⎥⎦ ⎥⎦
⎣⎣ ⎦ ⎣

4.10.2  Linear Equations of Motion


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

⎡ Le − Le 0 ⎤ ⎡ LAβΔβ + LAp p + LAr r + LAξξ + LAζ ζ ⎤


⎢ ⎥ ⎢ ⎥
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 ⎥⎦

The complete set of linear perturbation equations of motion of an aircraft in


steady forward flight in the stability axes are given by the following:

mΔu + mgΔθ = X Au Δu + X Aα Δα + X Aq q + X Aηη + X AτΔτ. (4.146a)


mU 0 Δα − mU 0 q = ZAu Δu + ZAα Δα + ZAq q + ZAηη + ZAτΔτ. (4.146b)


I yy q = M Au Δu + M Aα Δα + M Aq q + M Aα Δα + M Aηη + MTτΔτ. (4.146c)


Δθ = q. (4.146d)

h = U 0 ( Δθ − Δα ) . (4.146e)

free ebooks ==> www.ebook777.com
Aircraft Non-Linear Dynamics 141

mU 0 Δβ + mU 0 r − mgΔφ = YAβΔβ + YAp p + YAr r + YAξξ + YAζ ζ. (4.146f)


I xx p − I xz r = LAβΔβ + LAp p + LAr r + LAξξ + LAζ ζ. (4.146g)


I zz r − I xz p = N AβΔβ + N Ap p + N Ar r + N Aξξ + N Aζ ζ. (4.146h)


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

Stability axes: A wind axis system aligned with the direction of


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)

Parameter Roll Pitch Yaw Remarks


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

4.2 Express the aircraft’s linear longitudinal equations of motion in state-


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.

5.2  Small Perturbations and Linearisations


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)

Hence, the linearised equations of motion are

⎡ m ( Δu + ΔqWe − ΔrVe ) ⎤


⎢ ⎥
⎢ 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 ⎥
⎢ ⎥

Further, the gravitational force perturbation vector is

⎡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 ⎥⎦

In addition, we have the relations

⎡ Δφ ⎤ ⎡ 1 sin φe tan θe cos φe tan θe ⎤ ⎡ Δp ⎤


⎢ ⎥ ⎢ ⎥⎢ ⎥

⎢ 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

Finally, the aforementioned equations must be complemented by perturba-


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

⎡cos ψ −sin ψ 0 ⎤ ⎡ cos θ 0 sin θ ⎤ ⎡ 1 0 0 ⎤


⎢ ⎥⎢ ⎥⎢ ⎥
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)

ΔTIB ( ψ e , θe , φe ) = TIB ( ψ e + Δψ , θe + Δθ, φe + Δφ ) − TIB ( ψ e , θe , φe ) , (5.10)


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:

FAS = TWB ( α e , βe ) TBW ( α , β ) FAW = TBW ( α − α e , β − βe ) FAW (5.11a)


and

M AS = TWB ( α e , βe ) TBW ( α , β ) M AW = TBW ( α − α e , β − βe ) M AW . (5.11b)


www.ebook777.com
free ebooks ==> www.ebook777.com
148 Flight Dynamics, Simulation, and Control

When the small perturbation assumption is invoked, we may write Δα = α − αe,


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,

⎡ 1 −Δβ −Δα ⎤ ⎡ 1 −Δβ −Δα ⎤


⎢ ⎥ ⎢ ⎥
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.

5.3 Linearising the Aerodynamic Forces


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

Transforming to the body axis perturbation translational and angular


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

Now following the small perturbation assumption,


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

The derivatives referred to as stability derivatives in most textbooks are the


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.

5.4  Direct Formulation in the Stability Axis


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

Hence, it follows that

⎡ 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 ⎥⎦

The non-linear translational equations of motion in the stability axes are

⎡ 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

and the rotational equations are

⎡ 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

⎡ us ⎤ ⎡U es + Δus ⎤ ⎡ Δus ⎤ ⎡ Δu ⎤ ⎡ ps ⎤ ⎡ Δps ⎤ ⎡ Δp ⎤


⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ vs ⎥ = ⎢ Δvs ⎥ , ⎢ Δvs ⎥ = TWB ( α e , βe ) ⎢ Δv ⎥ , ⎢ qs ⎥ = ⎢ Δqs ⎥ = TWB ( α e , βe ) ⎢ Δq ⎥ ,
⎢ ws ⎥ ⎢ Δws ⎥ ⎢⎣ Δws ⎥⎦ ⎢⎣ Δw ⎥⎦ ⎢⎣ rs ⎥⎦ ⎢⎣ Δrs ⎥⎦ ⎢⎣ Δr ⎥⎦
⎣ ⎦ ⎣ ⎦
(5.24)

⎡ φs ⎤ ⎡φ⎤ ⎡ φse ⎤ ⎡ φe ⎤ ⎡ Δφs ⎤ ⎡ Δφ ⎤


⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ 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)

The components of the aerodynamic force and moment perturbation vectors


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

The components of the aerodynamic force and moment perturbation


­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

while the contributions of the controls, when activated, are

⎡ ⎤ ⎡   

⎢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 ⎣ ⎦

In these equations, only the contributions to the aerodynamic forces and


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:

FAS = TWB ( α e , βe ) TBW ( α , β ) FAW = TBW ( α − α e , β − βe ) FAW (5.37a)


and

M AS = TWB ( α e , βe ) TBW ( α , β ) M AW = TBW ( α − α e , β − βe ) M AW . (5.37b)


However, considering trimmed flight, FASe = FAWe and M ASe = M AWe. Subtracting
the latter pair of equations from the former pair, it follows that

FAS − FASe = TBW ( α − α e , β − βe ) FAW − FAWe



= FAW − FAWe + ( TBW ( α − α e , β − βe ) − I ) FAW , (5.38a)

M AS − M ASe = TBW ( α − α e , β − βe ) M AW − M AWe

= M AW − M AWe + ( TBW ( α − α e , β − βe ) − I ) M AW (5.38b)



free ebooks ==> www.ebook777.com
Small Perturbations and the Linearised, Decoupled Equations of Motion 157

which may be expressed as

FAS − FASe = TBW ( α − α e , β − βe ) ( FAW − FAWe ) + ( TBW ( α − α e , β − βe ) − I ) FAW ,



(5.39a)

M AS − M ASe = TBW ( α − α e , β − βe ) ( M AW − M AWe ) + ( TBW ( α − α e , β − βe ) − I ) M AW .



(5.39b)

Thus, the wind axis forces and moments may be expressed as

FAW = TWB ( α − α e , β − βe ) ( FAS − FASe ) + TWB ( α − α e , β − βe ) FAWe (5.40a)


and

M AW = TWB ( α − α e , β − βe ) ( M AS − M ASe ) + TWB ( α − α e , β − βe ) M AWe . (5.40b)


But

⎡cos α e 0 − sin α e ⎤ ⎡cos βe − sin βe 0⎤


⎢ ⎥⎢ ⎥
TBW ( α e , βe ) = ⎢ 0 1 0 ⎥ ⎢ sin βe cos βe 0 ⎥ , (5.41)
⎢⎣ sin α e 0 cos α e ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦

and since TBW(α e, βe) TWB(α e, βe) = I,

⎡ cos βe sin βe 0 ⎤ ⎡ cos α e 0 sin α e ⎤


⎢ ⎥⎢ ⎥
TWB ( α e , βe ) = ⎢ − sin βe cos βe 0⎥ ⎢ 0 1 0 ⎥ (5.42)
⎢⎣ 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

Hence, assuming small perturbations,

⎡ 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 ⎥⎦

The corresponding inverse relationship, assuming small perturbations, is

⎡ 1 −Δβ −Δα ⎤ ⎡ 1 −Δβ −Δα ⎤


⎢ ⎥ ⎢ ⎥
FAS − FASe = ⎢ Δβ 1 0 ⎥ ( FAW − FAWe ) + ⎢ Δβ 1 0 ⎥ FAWe ,
 ⎢⎣ Δα 0 1 ⎥⎦ ⎢⎣ Δα 0 1 ⎥⎦

(5.46a)

⎡ 1 −Δβ −Δα ⎤ ⎡ 1 −Δβ −Δα ⎤


⎢ ⎥ ⎢ ⎥
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  Decoupled Equations of Motion


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

It is customary to write these equations as

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 ⎥⎦

⎡ cos α e siin α e ⎤ ⎡ −cos θe ⎤


+ 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

⎡cos α e 0 − sin α e ⎤ ⎡cos βe − sin βe 0⎤


⎢ ⎥⎢ ⎥
TBW ( α e , βe ) = ⎢ 0 1 0 ⎥ ⎢ sin βe cos βe 0⎥
⎢⎣ sin α e 0 cos α e ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦

⎡ cos α e cos βe − cos α e sin βe − sin α e ⎤


⎢ ⎥
= ⎢ sin βe cos βe 0 ⎥ (5.51)
⎢⎣ sin α e cos βe − sin α e sin βe cos α e ⎥⎦

Δw = ⎣⎡cos βe sin α e Δus − sin βe sin α e Δvs + cos α e Δws ⎤⎦ . (5.52)


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.

5.5.2 Case II: Motion in the Lateral Direction,


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

where the components of the aerodynamic moment perturbation vector in the


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 ⎦
⎣ ⎦

5.6 Decoupled Equations of Motion in terms of the Stability


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

Hence, the longitudinal equations of motion (with the effects of aerodynamic


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 ⎥⎦

⎡ cos α e sin α e ⎤ ⎡ −cos θe ⎤


+ 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

Δθ s = Δqs , (5.57a)


Δh ≈ Uγ = U ( Δθ − Δα ) ≈ U e ( Δθ − Δα ) , (5.57b)

Δθs = Δθ, (5.57c)

Δw Δw
Δα ≈ ≈ , (5.57d)
U +V
2
e e
2 Ue

where h is measured positive upwards.


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

Δφ s = Δps , (5.59a)


Δψ s = Δrs (5.59b)


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

Δφ = ⎣⎡cos βe cos α e Δφs − sin βe cos α e Δθs − sin α e Δψ s ⎤⎦ . (5.61)


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 ⎥

When βe = 0, these relations reduce to

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

It is important to observe a few important points:

1. All trimmed quantities are evaluated relative to the body axes.


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.

5.7  Addition of Aerodynamic Controls and Throttle


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

Thus, the longitudinal equations of motion with the effects of aerodynamic


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 ⎥⎦

⎡ cos α e sin α e ⎤ ⎡ −cos e ⎤


+ 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)

Δh ≈ U e ( Δθs − Δα s ) . (5.68b)



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

⎡ mΔu s ⎤ ⎡ ⎡ X NB ⎤ ⎡ X NBe ⎤ ⎤ ⎡ −cos θe ⎤


⎢ ⎥ = ⎢⎢ ⎥ − ⎢ 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

Δθ s = Δqs , (5.73a)


Δh = U es ( Δθ − Δα ) . (5.73b)

The lateral equations of motion are

( )
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

+ mg cos θe cos φe Δφ (5.74a)


⎡ 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)

Δψ s = Δrs . (5.75b)


www.ebook777.com
free ebooks ==> www.ebook777.com
168 Flight Dynamics, Simulation, and Control

Dividing through by m and I yy


s
, the longitudinal equations may be written as

⎡ Δu s ⎤ 1 ⎡ ⎡ X NB ⎤ ⎡ X NBe ⎤ ⎤ ⎡ −cos θe ⎤


⎢ 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

Since the last equation is independent of the first four, it is convenient to


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

Δh = −Δws + U es Δθs . (5.79b)


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

The equations may be written as a set of first-order equations which have


been called the concise form by Cook [1] and are given by

⎡1 0 − xq 0 ⎤ ⎡ Δu s ⎤ ⎡ xu xw xq xθ ⎤ ⎡ Δus ⎤


⎢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)

Similarly, the lateral equations are

1
Δv s + ΔrsU es = ( YNB − YNBe )
m
+ ΔvsYv + ΔpsYp + ΔrsYr + Δp sYp + ΔrsYr + ΔξYξ + ΔζYζ

+ g cos θe cos φe Δφ (5.83a)


⎡ I xz
s

⎢ 1 − s ⎥
I xx ⎥ ⎡ Δp s ⎤ ⎡ I xx ⎤ ⎡ ⎡ LNB ⎤ ⎡ LNBe ⎤ ⎤
−1
s
⎢ =⎢ ⎥−⎢
⎢ ⎥ s ⎥ ⎢⎢ ⎥⎥
⎢ I xz
s ⎥ Δrs ⎦ ⎣ I zz ⎦ ⎢⎣ ⎣ N NB ⎦ ⎣ N NBe ⎦ ⎥⎦
 ⎢− s 1 ⎥⎣
⎣ I zz ⎦
⎡ Δvs Lv + Δps Lp + Δrs Lr + Δp s Lp + Δrs Lr + Δξ Lξ + Δζ Lζ ⎤
+⎢ ⎥ (5.83b)
⎣ Δvs N v + Δps N p + Δrs N r + Δp s N p + Δrs N r + ΔξN ξ + ΔζN ζ ⎦
free ebooks ==> www.ebook777.com
Small Perturbations and the Linearised, Decoupled Equations of Motion 171

with

Δφ s = Δps , Δψ s = Δrs (5.84)


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 ⎥⎦

⎡ Yv Yp Yr − U es g cos φe cos θe 0 ⎤ ⎡ Δvs ⎤ ⎡ Y Y ⎤


⎢ ⎥⎢ ⎥ ⎢
⎢ 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

As in the longitudinal case, the last equation in the aforementioned set is


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 ⎥⎦

⎡ Yv Yp Yr − U es − g cos φe cos θe ⎤ ⎡ Δvs ⎤ ⎡ Y Y ⎤


⎢ ⎥⎢ ⎥ ⎢
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

Δψ s = Δrs . (5.86b)


Again as in the longitudinal case, in most real situations, the derivatives


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 ⎥ ⎢ Δrrs ⎥
⎢ ⎥⎢ ⎥
⎣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’.

5.8  Non-Dimensional Longitudinal and Lateral Dynamics


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

Aerodynamics and estimation


Stability derivatives

Dimensional Non- Aerodynamic


dimensional coefficients
Reduce to
state-space form

State-space
derivatives

Divide by
‘inertial coupling’ matrix
Concise
derivatives

Dynamics and controls


(real time and distance)

FIGURE 5.1
Taxonomy of stability derivatives.

the dimensional equations of motion to non-dimensional form is illustrated


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

It then follows that

∂ 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

To non-dimensionalise the lateral equations and introduce the dimension-


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

are the non-dimensional moment of inertia parameters.


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 ⎦

Considering the longitudinal equations of motion and eliminating the states


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

The relationships between the nine longitudinal dimensional stability


        
­ 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

Hence, rearranging the terms,

 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 .

5.9 Simplified State-Space Equations of


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

may be expressed in terms of the aircraft’s mass, moments of inertia, uniform


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

Similarly, the corresponding lateral equations, referred to body-fixed axes, are

⎡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

5.10 Simplified Concise Equations of


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

Similarly, the concise lateral equations, referred to body-fixed axes, are

⎡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

5.3 The perturbation equations of longitudinal motion of an aircraft, in


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 ⎦

Reduce these equations to concise form given by

⎡ 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 ⎦

Hence, or otherwise, determine the relationships between the two sets of


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

Reduce these equations to concise form given by

⎡ Δv s ⎤ ⎡ yv yp yr yφ ⎤ ⎡ Δvs ⎤ ⎡ yξ yζ ⎤
⎢ Δp ⎥ ⎢
⎢ s ⎥ = ⎢ lv lp lr 0 ⎥⎥ ⎢⎢ Δps ⎥⎥ ⎢⎢ lξ lζ ⎥⎥ ⎡ Δξ ⎤
+ ⎢ ⎥.
⎢ Δrs ⎥ ⎢ nv np nr 0 ⎥ ⎢ Δrs ⎥ ⎢ nξ nζ ⎥ ⎢ Δζ ⎥⎦
⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥⎣
⎢⎣ Δφ s ⎥⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφs ⎦ ⎣ 0 0⎦

Hence, or otherwise, determine the relationships between the two sets of


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.

6.2  Dynamic and Static Stability


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.

6.2.1  Longitudinal Stability Analysis


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)

and it follows that the following set of homogeneous equations

⎡ ⎡ 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 ⎦
⎣ ⎣

must be satisfied. From the theory of homogeneous simultaneous equa-


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

Ignoring the q derivatives in the determinant equation without any loss of


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

Evaluating this determinantal equation, we obtain a quartic polynomial


equation in λ, known as the characteristic equation, which has the form

a4λ 4 + a3λ 3 + a2λ 2 + a1λ + a0 = 0. (6.6)

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 (6.7)


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)

Equivalently, the characteristic equation may be written as


( ) ( )
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

In the case of steady flight,

d T T
⎡u w q θ ⎤⎦ = ⎡⎣0 0 0 0 ⎤⎦ (6.11)
dt ⎣

The determinantal condition for the existence of a non-zero solution


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 ⎣

may be obtained by substituting for each of the λ values, λ = λ k± , k = 1 and


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

CASE STUDY 6.1:  LONGITUDINAL DYNAMICS


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:

X w = Zw = X q = X q = Zq = Zq = Mq = 0 and θe = 0. (6.18)


We also assume that the controls are fixed.


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 ⎦

while the equation for the altitude is

⎛ Δw ⎞
h = U es ⎜ Δθ − s ⎟ . (6.20)
⎝ Ue ⎠

free ebooks ==> www.ebook777.com
Longitudinal and Lateral Linear Stability and Control 191

CASE STUDY 6.1: (continued)  LONGITUDINAL DYNAMICS


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

CASE STUDY 6.1: (continued)  LONGITUDINAL DYNAMICS


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

Lateral State-Space Derivatives


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

CASE STUDY 6.1: (continued)  LONGITUDINAL DYNAMICS


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

The linear dynamics is therefore governed by

⎡1 0 0 0 ⎤ ⎡ Δu ⎤ ⎡ −0.014 −9.81 0.0043 0 ⎤ ⎡ Δu ⎤


⎢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

⎡ Δu ⎤ ⎡ −0.014 −9.81 0.0043 0 ⎤ ⎡ Δu ⎤


⎢ ⎥ ⎢
⎢ 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 characteristic polynomial is

Δ ( λ ) = λ 4 + 2.1643λ 3 + 9.9317 λ 2 + 0.1172λ + 0.0055. (6.23)


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

CASE STUDY 6.2:  BOEING 747 LONGITUDINAL


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

⎡ u ⎤ ⎡ −0.00643 0.0263 0 −32.2 ⎤ ⎡ u ⎤ ⎡ 0 ⎤


⎢ ⎥ ⎢ 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)

while the equation for the altitude is

⎛ w ⎞
h = U e ⎜ θ − ⎟ . (6.25)
⎝ U e ⎠

The corresponding characteristic polynomial is

Δ ( λ ) = λ 4 + 1.2984λ 3 + 1.6822λ 2 + 0.0100λ + 0.0002. (6.26)


At M = 0.65 h = 20,000 ft, the characteristic polynomial is

Δ ( λ ) = λ 4 + 1.1955λ 3 + 1.5960λ 2 + 0.0106λ + 0.00676. (6.27)


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

CASE STUDY 6.3:  LONGITUDINAL


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

⎡1 0 0 0 ⎤ ⎡ Δu ⎤ ⎡ −0.0352 −32.17 0.107 0 ⎤ ⎡ Δu ⎤


⎢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)

Equation 6.28 reduces to

⎡ Δu ⎤ ⎡ −0.0352 −32.17 0.107 0 ⎤ ⎡ Δu ⎤


⎢ ⎥ ⎢
⎢ 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 corresponding characteristic polynomial is

Δ ( λ ) = λ 4 + 5.5122λ 3 + 0.5088λ 2 + 0.1220λ + 0.0038. (6.30)


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

CASE STUDY 6.3: (continued) LONGITUDINAL


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

6.2.2  Lateral Dynamics and Stability


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

must be satisfied. From the theory of homogeneous simultaneous equations, it


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

a4λ 4 + a3λ 3 + a2λ 2 + a1λ + a0 = 0. (6.36)

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)

Equivalently, the characteristic equation may be written as


⎡ 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.

CASE STUDY 6.4:  LATERAL DYNAMICS OF THE DC 8


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 ⎦

⎡ −0.0868 0 −251.22 9.81⎤ ⎡ Δv ⎤


⎢ −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

CASE STUDY 6.4: (continued) LATERAL


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

⎡ Δv ⎤ ⎡ −0.0868 0 −251.22 9.81⎤ ⎡ Δv ⎤


⎢ 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

CASE STUDY 6.5:  BOEING 747 LATERAL DYNAMICS IN


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

⎡β ⎤ ⎡ −0.00558 0.0802 −0.9968 0.0415 ⎤ ⎡β ⎤ ⎡ 0.00729 ⎤


⎢ ⎥ ⎢ 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

Δ ( λ ) = λ 4 + 0.5856λ 3 + 0.9097 λ 2 + 0.5083λ + 0.0037. (6.46)



At M = 0.65, h = 20,000 ft,

Δ ( s ) = s 4 + 1.0999s3 + 1.3175s2 + 1.0594s + 0.01129. (6.47)



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.

CASE STUDY 6.6:  LATERAL DYNAMICS OF A


HIGHLY MANOEUVRABLE FIGHTER AIRCRAFT
The governing lateral coupled equations of motion with controls fixed are

⎡β ⎤ ⎡ −0.746 0.006 −0.999 0.0369 ⎤ ⎡β ⎤


⎢ ⎥ ⎢ 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 characteristic polynomial is

Δ ( λ ) = λ 4 + 1.4949λ 3 + 2.5408λ 2 + 2.8142λ + 0.0112. (6.49)



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

6.3 Modal Description of Aircraft Dynamics


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

Δh = −Δws + U es Δθs . (6.51)

The last equation is independent of all the others and represents the height
integration mode.

6.3.1  Slow–Fast Partitioning of the Longitudinal Dynamics


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

⎡1 0 − xq 0 ⎤ ⎡ Δu s ⎤ ⎡ xu xw xq xθ ⎤ ⎡ Δus ⎤ ⎡ xη xτ ⎤


⎢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

We may write them as

⎡ Δu s ⎤ ⎡ xu xθ ⎤ ⎡ Δus ⎤ ⎡ xw xq ⎤ ⎡ Δws ⎤ ⎡ xη xτ ⎤ ⎡ Δη⎤


⎢  ⎥=⎢ + +
⎢⎣ 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 ⎥⎦

and from the last two equations,

⎡ Δws ⎤ ⎡ zw zq ⎤ ⎡ ⎡ zu zθ ⎤ ⎡ Δus ⎤ ⎡ zη zτ ⎤ ⎡ Δη⎤ ⎤


−1

⎢ Δq ⎥ = − ⎢ m ⎢⎢ + ⎥ . (6.55)
⎣ s⎦ ⎣ w mq ⎥⎦ ⎢⎣ ⎣ mu mθ ⎥⎦ ⎢⎣ Δθs ⎥⎦ ⎢⎣ mη mτ ⎥⎦ ⎢⎣ Δτ ⎥⎦ ⎥⎦

The first two equations then are

⎡ Δu s ⎤ ⎡ xu xθ ⎤ ⎡ Δus ⎤ ⎡ xw x q ⎤ ⎡ zw z q ⎤ ⎡ zu zθ ⎤ ⎡ Δus ⎤


−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 ⎥⎦

which may be expressed compactly 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⎤
+ ⎢⎢ − ⎥ . (6.57)
⎢⎣ ⎣ 0 0 ⎥⎦ ⎢⎣ 0 1 ⎥⎦ ⎢⎣ mw mq ⎥⎦ ⎢⎣ mη mτ ⎥⎦ ⎥ ⎢⎣ Δτ ⎥⎦

These equations represent the equations of motion of the phugoid mode


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

⎡1 0 0 − xq ⎤ ⎡ Δu s ⎤ ⎡ xu xθ xw xq ⎤ ⎡ Δuus ⎤ ⎡ xη xτ ⎤


⎢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

⎡1 − zq ⎤ ⎡ Δw s ⎤ ⎡ zw zq ⎤ ⎡ Δws ⎤ ⎡ zη zτ ⎤ ⎡ Δη⎤


⎢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)

then the short period equations are

⎡ Δw s ⎤ ⎡ zʹw zʹq ⎤ ⎡ Δws ⎤ ⎡ zʹη zʹτ ⎤ ⎡ Δη⎤


⎢ Δq ⎥ = ⎢ mʹ + . (6.62)
⎣ s⎦ ⎣ w mʹq ⎥⎦ ⎢⎣ Δqs ⎥⎦ ⎢⎣ mʹη mʹτ ⎥⎦ ⎢⎣ Δτ ⎥⎦

The aforementioned pair of first-order equations represents the fast subsys-


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

the standard approximate equations for the longitudinal short period


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.

6.3.2  Slow–Fast Partitioning of the Lateral Dynamics


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 ⎥ ⎢ Δrrs ⎥ ⎢ nv np nr nφ ⎥ ⎢ Δrs ⎥ ⎢ nξ nζ ⎥ ⎢⎣ Δζ ⎥⎦
⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥
⎣0 0 0 1 ⎦ ⎢⎣ Δφ s ⎥⎦ ⎣ 0 1 0 0 ⎦ ⎣ Δφs ⎦ ⎣ 0 0⎦
(6.63)

and

Δψ s = Δrs . (6.64)


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 Δrs + ⎡⎣lv lp lr lφ ⎤⎦ ⎢ ⎥ + ⎡lξ lζ ⎤⎦ ⎢ ⎥ = 0. (6.66)
⎢ Δrs ⎥ ⎣ ⎣ Δζ ⎦
⎢ ⎥
⎣ Δφs ⎦

Solving for Δps, it follows that

⎡ Δvs ⎤
l 1 ⎢ ⎥ 1 ⎡ Δξ ⎤
Δps = − r Δrs − ⎡⎣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

Eliminating Δps from the lateral equations, we obtain

⎡ lr ⎤
⎢1 − y r + yp 0⎥
lp ⎥ ⎡ Δvv s ⎤ ⎡ yv
⎢ yr yφ ⎤ ⎡ Δvs ⎤ ⎡yp ⎤ ⎡ Δvs ⎤
⎢ l ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ 1 ⎢ ⎥ ⎢ ⎥
⎢0 1 − nr + r np 0 ⎥ ⎢ Δrs ⎥ = ⎢ 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 ⎥⎦ ⎣ ⎦

which may be simplified as

⎡ lr ⎤
⎢1 − y r + yp 0⎥
lp ⎥ ⎡ Δvv s ⎤ ⎡ ⎡ yv
⎢ yr yφ ⎤ ⎡ yp ⎤ ⎤ ⎡ Δvs ⎤
⎢ l ⎥⎢ ⎥ ⎢⎢ ⎥ 1 ⎢ ⎥ ⎥⎢ ⎥
⎢0 1 − nr + r np 0 ⎥ ⎢ Δrs ⎥ = ⎢ ⎢ 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 ⎥ ⎣ Δrs ⎦ ⎢⎣ ⎣ 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

which may be expressed as

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 = Δrs = 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

and from the first three equations, we obtain

⎡ 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)

The equation for the spiral mode is

−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

Hence, the spiral mode equation reduces to

⎡ 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

when (lvnp − lpnv) > 0. The condition is completely equivalent to the condition


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 ⎥⎦ ⎣ ⎦

From most analysis, it is the fast subsystem that is of greater importance.


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 ⎥⎦ ⎢⎣ Δrs ⎥⎦ ⎢⎣ nv np nr ⎥⎦ ⎢⎣ Δrs ⎥⎦ ⎢⎣ nξ nζ ⎥⎦ ⎣ ⎦

It is convenient to express these as

⎡1 − y r − y p ⎤ ⎡ Δv s ⎤ ⎡ yv yr y p ⎤ ⎡ Δvs ⎤ ⎡ yξ yζ ⎤
⎢ ⎥⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎡ Δξ ⎤
⎢0 −nr 1 − np ⎥ ⎢ Δrs ⎥ = ⎢ nv nr np ⎥ ⎢ Δrs ⎥ + ⎢ nξ nζ ⎥ ⎢ ⎥ . (6.85)
Δζ
⎢0 1 − lr −lp ⎥⎦ ⎢⎣ Δp s ⎦⎥ ⎢⎣ lv lr lp ⎥⎦ ⎢⎣ Δps ⎥⎦ ⎢⎣ lξ lζ ⎥⎦ ⎣ ⎦

The characteristic equation of the system has the form

( Trss + 1) ( s2 + 2ζ Dr ωDr s + ωDr


2
) = 0 (6.86)

free ebooks ==> www.ebook777.com
Longitudinal and Lateral Linear Stability and Control 209

where approximately, Trs = −(1/lp), which is the time constant associated


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.

CASE STUDY 6.7:  LONGITUDINAL DYNAMICS


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

⎡1 0 0 0 ⎤ ⎡ Δu ⎤ ⎡ −0.014 −9.81 0.0043 0 ⎤ ⎡ Δu ⎤


⎢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)

Setting Δu = Δθ = 0, we have the approximation to the short period


dynamics which is

⎡ 1 0 ⎤ ⎡ Δw ⎤ ⎡ −0.806 251.22 ⎤ ⎡ Δw ⎤


⎢0.001673 = (6.88)
⎣ 1 ⎥⎦ ⎢⎣ q B ⎥⎦ ⎣⎢ −0.03642 −0.924 ⎥⎦ ⎢⎣ qB ⎥⎦

or

⎡ Δw ⎤ ⎡ −0.806 251.22 ⎤ ⎡ Δw ⎤


⎢ q ⎥ = ⎢ −0.03507 . (6.89)
⎣ B⎦ ⎣ −1.3443 ⎥⎦ ⎢⎣ qB ⎥⎦

Hence, the characteristic polynomial is

Δ ( λ ) = λ 2 + 2.1503λ + 9.546. (6.90)


The roots of the characteristic polynomial are −1.0752 ± i2.8966 corre-


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

CASE STUDY 6.7: (continued)  LONGITUDINAL DYNAMICS


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,

⎡ Δu ⎤ ⎡ −0.014 −9.81⎤ ⎡ Δu ⎤ ⎡0.0043 0 ⎤ ⎡ Δw ⎤


⎢ ⎥ = ⎢ + . (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

CASE STUDY 6.8:  LATERAL DYNAMICS OF THE DC 8


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 ⎦

⎡ −0.0868 0 −251.22 9.81⎤ ⎡ Δv ⎤


⎢ −0.0176 −1.18 0.336 0 ⎥⎥ ⎢⎢ Δp ⎥⎥
=⎢ . (6.98)
⎢ 0.00864 −0.01294 −0.23 0 ⎥ ⎢ Δr ⎥
⎢ ⎥⎢ ⎥
⎣ 0 1 0 0 ⎦ ⎣ Δφ ⎦

Multiplying by the inverse of the inertia coupling matrix,

⎡ Δv ⎤ ⎡ −0.0868 0 −251.22 9.81⎤ ⎡ Δv ⎤


⎢ ⎥ ⎢ 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 ⎦

The approximation for the roll subsidence mode follows by setting


Δv = Δr = Δϕ = 0, and we obtain

Δp = −1.1802Δp. (6.100)


The associated time constant is Trs = 1/1.1802 = 0.8473 s. The values


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

CASE STUDY 6.8: (continued) LATERAL


DYNAMICS OF THE DC 8
Hence,

0.333Δr − 0.0175Δv
Δp = = 0.282Δr − 0.0148Δv. (6.102)
1.1802

Eliminating Δp from the remaining three equations,

⎡ 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 ⎥⎦

Hence, the characteristic equation is

Δ ( λ ) = λ 2 + 0.3205λ + 2.231 = 0. (6.106)


The roots of the characteristic polynomial are −0.1603 ± i1.485 corre-


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

CASE STUDY 6.8: (continued) LATERAL


DYNAMICS OF THE DC 8
To establish the approximation for the spiral mode, we assume that
the first three equations are satisfied instantaneously. Hence,

⎡ Δv ⎤ ⎡ −0.0868 0 −251.22 ⎤ ⎡ Δv ⎤ ⎡9.81⎤ ⎡0 ⎤


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)

The associated time constant is Ts = 1/0.0043 = 231.90 s. The values


obtained in Case Study 6.4 are −0.004 for the pole and 249.63 s for the
time constant.

6.3.3  Summary of Longitudinal and Lateral Modal Equations


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.

6.3.3.1  Phugoid or Long Period


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 ⎥⎦

6.3.3.2  Short Period


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)

6.3.3.3  Third Oscillatory Mode


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.

6.3.3.4  Roll Subsidence


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 ⎦

6.3.3.5  Dutch Roll


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.

6.4  Aircraft Lift and Drag Estimation


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

In this expression, e is Oswald’s efficiency factor which is equal to unity


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

elliptic spanwise load distribution. For other load distributions, it is pos-


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)

For a finite wing, it can be shown that

CL = a∞ ( α − α 0 − α in ) = a∞ ( α − α 0 − win U e ) . (6.119)

Substituting for αin and solving for CL,

⎛ ⎞
⎜ a∞ ⎟
CL = ⎜ ⎟ ( α − α 0 ) . (6.120)
a
⎜ 1+ ∞ ⎟
⎝ πeAR ⎠

Thus, CL can be expressed as CL = a(α−α 0) and

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

From the Kutta–Joukowski theorem, lift per unit span is L = ρUeΓ(y1).


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

Laitone [3] presents another generalisation which is

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.

6.4.1  Fuselage Lift and Moment Coefficients


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

Upon integration by parts,

xf xf

M f = ρU α S f dx = ρU αVf , Vf =
∫ S dx. (6.132)
2 2
e

0
e
0
f

The contribution to the aircraft pitching moment is

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.

6.4.2  Wing–Tail Interference Effects


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


(
= 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.

6.4.3  Estimating the Wing’s Maximum Lift Coefficient


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.

6.4.4  Drag Estimation


It can be shown that induced drag is a component of the 3D lift in the drag
direction:

Di = L sin α in or CDi = CL sin α in . (6.137)



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

and k = Oswald’s inverse efficiency factor, given by the empirical relation

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

When FQ = 1, CD0 = Cf Swet/Sref . The skin friction drag is estimated by inte-


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 ⎤
⎣ ⎦ ⎣ ⎦

while in the case of compressible flow, it may be approximated by

( 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 ⎠ ⎪⎭

which are summarised in Table 6.9.


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

Swet = 2Sw or Swet = 3Sw . (6.146)


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

6.5  Estimating the Longitudinal Aerodynamic Derivatives


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

(i) Force equilibrium, x-axis: T0 = (∑ D) 0

(ii) Force equilibrium, z-axis: mg = ( ∑ L )


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

Further, the perturbation moment equation is given by

1
M= ρV 2Sw cCm (6.148)
2

which is normally evaluated at the aerodynamic centre.


The lift coefficients are then expressed as the sum of three components:

CL = CLW + CLF + CLαT αT , (6.149a)

www.ebook777.com
free ebooks ==> www.ebook777.com
226 Flight Dynamics, Simulation, and Control

the drag coefficient as (see Section 6.4.4 on drag estimation)

CD = CDW + CD0 + CDT + CDOther (6.149b)


and the moment coefficient at the aircraft aerodynamic centre as

⎛ ∂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

Retaining only the linear terms and since ∂V/∂u ≈ 1,

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

Similarly, for the corresponding Z derivative, we have

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

Further, compressibility effects may also be included in this formulation. It


may be noted that

∂ ∂ M∞2 ∂
V ≈ M∞ =− . (6.154)
∂u ∂M∞ 1 − M∞2 ∂ 1 − M∞2

The derivative of the drag coefficient is

∂CD ∂C 2CL ∂C
V = V D0 + V L . (6.155)
∂u ∂u πeAR ∂u

Thus, for a 2D aerofoil from the Prandtl–Glauert similarity rule,

∂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

In the case of a propeller-driven constant power engine,

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

The thrust coefficient may be evaluated at equilibrium conditions since

CT = CT 0 = CL sin ( α E ) − CD cos ( α E ) . (6.160)


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

To find the derivative of the steady pitching moment coefficient in forward


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

where rT is the pitching moment arm of the thrust vector.


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

Additionally, to find the pitch and pitch rate derivatives, it is important to


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

Now, the contribution of the tail plane to the force


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

* Some textbooks define q = qc /2V .

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

This is another extremely important stability derivative that provides damp-


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 ⎟
⎝ ⎠

Introducing the non-dimensional time variable, τ = ( 2V c ) t, and recognis-


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

and it follows from Equation 6.178 and 6.177 that

2
⎛ ∂ε ⎞ ∂ε ⎛ 2lt 1 ⎛ 2l ⎞ ⎞
αT = ( α w − i0 ) ⎜ 1 − ⎟ + iT + i0 − iw + ⎜ αʹ − ⎜ t ⎟ αʹʹ ⎟ . (6.179)
⎝ ∂α ⎠ ∂α ⎜⎝ c 2⎝ c ⎠ ⎟

Considering the horizontal tail contribution to the pitching moment


coefficient,

dCm ⎞ ⎛ dCL ⎞ lt ST
( Cm )tail = ⎛⎜ ⎟tail αT = − ⎜ ⎟ αT , (6.180)
⎝ d α ⎠ part ⎝ dα ⎠tail c Sw
part

and it follows from Equations 6.179 and 6.180 that

dCm dCm ∂αT l S ∂ε 2lt ∂ε


= = Cmαʹ = − ( CLα )tail t T × = 2CMac , q . (6.181)
dαʹ dαT ∂αʹ c Sw ∂αT c ∂α

Thus, it follows from Equation 6.175 that

1 ∂Mac 1 ∂Cm 1 ∂ε 1 ∂Mac ∂ε


= = Cmαʹ = CMac , q = . (6.182)
1 2 w 1
2 2 ρV Sw ∂q ∂α
∂  ∂αʹ ∂α
ρ c Sw 2

2 2

Other acceleration-related derivatives may be obtained in a similar way and


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

The 12 primary longitudinal stability derivatives of the forward and plung-


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

6.6  Estimating the Lateral Aerodynamic Derivatives


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

Similar to the concept of induced downwash at the horizontal tail, we may


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 ⎠

Hence, it follows that


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 rolling moment derivative may then be easily estimated.


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

Integrating over the height of the vertical tail,


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

Integrating over the height of the vertical tail,

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)

and the increment in the lift force on the strip is

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

6.6.1  Perturbation Analysis of Trimmed Flight


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.

6.6.2  Perturbation Analysis of Longitudinal Trimmed Flight


The principal longitudinal equilibrium flight conditions are

1. Steady level flight


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

The conditions for steady level flight (γ = 0) may be expressed as

1. Force equilibrium, x-axis T cos ( iE ) = mg sin ( θ ) +∑ D cos ( θ) − L sin ( θ)


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)

Mac + ( Mac )tailplane + Lw ( lw + lT ) + Dw ( zDw + zwt ) c + Dp ( hDp + hwt ) c = mglt



(6.219)

b. About aircraft CG

Mac + ( Mac )tailplane + Lw lw + Dw zDw c + Dp zDp c = LT lT (6.220)


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

Sometimes a different moment arm is defined which is lt = lT + lw, and


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

Relative wind Wing


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)

Similarly, the equilibrium conditions for the transition pull up manoeu-


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

longitudinal steady small perturbation state-space equations, including the


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 ⎦

with the additional conditions, Δq = qB = 0. Thus, the equations reduce to

⎡ Xu −g X w ⎤ ⎡ Δu ⎤ ⎡ Xη Xτ ⎤
⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎡ Δη⎤
⎢ Zu 0 Zw ⎥ ⎢ Δθ ⎥ = − ⎢ Zη Zτ ⎥ ⎢ ⎥ . (6.226)
Δτ
⎢⎣ Mu 0 Mw ⎥⎦ ⎢⎣ Δw ⎥⎦ ⎢⎣ Mη Mτ ⎥⎦ ⎣ ⎦

From the second and third equation, we have

⎡ 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 ⎥⎦

while from the first equations, we obtain

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

CASE  STUDY 6.9:  BOEING 747 LONGITUDINAL


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

⎡ u ⎤ ⎡ −0.00643 0.0263 0 −32.2 ⎤ ⎡ u ⎤ ⎡ 0 ⎤


⎢ ⎥ ⎢ 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)

where η and θ are in radians.


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 ⎥⎦

But from the first of the equations of motion,

1 ⎡u⎤
θ= ⎡ −0.00643 0.0263 ⎤⎦ ⎢ ⎥ . (6.233)
32.2 ⎣ ⎣w ⎦

Eliminating the perturbation velocity components,

⎡ −0.0941 −0.624 ⎤ ⎡ 32.7 ⎤


−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

CASE  STUDY 6.9: (continued)  BOEING 747 LONGITUDINAL


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

Hence, η = 0.108° is the required increment in the elevator angle trim


setting.

6.6.3  Perturbation Analysis of Lateral Trimmed Flight


The two main trimmed flight modes in lateral flight correspond to

1. Steady sideslip
2. Turn coordination and banking

Each of these cases will now be considered independently.

6.6.3.1  Control Settings for Steady Sideslip


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 ⎦

with the additional conditions, Δp = 0 and Δr = 0. Thus, the equations reduce to

⎡ Yv g⎤ ⎡ Yξ Yζ ⎤ ⎡0 ⎤
⎢ ⎥ ⎡ Δv ⎤ ⎢ ⎥ ⎡ Δξ ⎤ ⎢ ⎥
⎢ Lv 0 ⎥ ⎢ ⎥ + ⎢ Lξ
Δφ
Lζ ⎥ ⎢ =
Δζ ⎥⎦ ⎢ ⎥
0 . (6.241)
⎢⎣ N v 0 ⎥⎦ ⎣ ⎦ ⎢⎣ N ξ N ζ ⎥⎦ ⎣ ⎢⎣0 ⎥⎦

From the second and third equations,

−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

and from the first equation,

1 T
Δφ = − ⎡Yv Yξ Yζ ⎤⎦ ⎡⎣ Δv Δξ Δζ ⎤⎦ . (6.243)
g⎣

In computing the inverse of the matrix,

−1
⎡ Lξ Lζ ⎤
Δ ra = ⎢ , (6.244)
⎣ Nξ N ζ ⎥⎦

one could ignore the off-diagonal derivatives when these are small.

6.6.3.2  Control Settings for Turn Coordination and Banking


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 ⎥⎦

Again assuming these trim conditions are small perturbations to those


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

with the additional conditions that the increments Δv = 0 and Δp = 0. Thus,


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⎠

In computing the inverse of the matrix


−1
⎡ Lξ Lζ ⎤
Δ ra = ⎢ , (6.244)
⎣ Nξ N ζ ⎥⎦

one could ignore the off-diagonal derivatives when these are small.

CASE STUDY 6.10:  COORDINATED TURNING OF THE DC 8


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

CASE STUDY 6.10: (continued) COORDINATED


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)

Setting all other states, except Δp, to zero,


T
Δp = −1.18Δp + ⎡⎣ −2.11 0.559 ⎤⎦ ⎡⎣ξ ζ ⎤⎦ . (6.255)

Hence, the associated time constant is Trs = 1/1.18 = 0.8474 s.


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

Eliminating Δp from the remaining three equations,

⎡ 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

CASE STUDY 6.10: (continued) COORDINATED


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

Δ ( λ ) = λ 2 + 0.3205λ + 2.231 = 0. (6.261)


The roots of the characteristic polynomial are −0.1603 ± i1.485 corre-


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

⎡ −2.11 0.559 ⎤ ⎡ ξ ⎤ ⎡ 0.336 ⎤


⎢ −0.0519 ⎥ ⎢
−1.168 ⎦ ⎣ζ ⎦⎥ = −⎢ ⎥ Δr , (6.262)
⎣ ⎣ −0.23 ⎦

while from the first equation, we have

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

CASE STUDY 6.10: (continued) COORDINATED


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 ⎥⎦

⎡ −0.4684 −0.2242 ⎤ ⎡ 0.336 ⎤


= −⎢ Δr (6.264)
⎣ 0.0208 −0.8462 ⎥⎦ ⎢⎣ −0.23 ⎥⎦

which simplifies to

⎡ ξ ⎤ ⎡ 0.1057 ⎤ ⎡ 0.00264 ⎤ ⎡ 0.151° ⎤


⎢ζ ⎥ = ⎢ −0.2016 ⎥ Δr = ⎢ −0.00508 ⎥ rad = ⎢ −0.291°⎥ . (6.265)
⎣ ⎦ ⎣ ⎦ ⎣ ⎦ ⎣ ⎦

We may now estimate the bank angle, which is given by

⎡ 251 5.587 ⎤ ⎡ Δr ⎤ ⎡ 251 5.587 ⎤ ⎡ 0.0251 ⎤


Δφ = − ⎢ ⎢ ⎥ = −⎢ ⎢ ⎥ = −0.64 rad
⎣ 9.81 9.81 ⎥⎦ ⎣ ζ ⎦ ⎣ 9.81 9.81 ⎦⎥ ⎣ −0.00508 ⎦

(6.266)

or

Δφ = −0.64 rad = −36.7°. (6.267)


6.6.4  Perturbations of Coupled Trimmed Flight


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.

6.6.5 Simplified Analysis of Complex Manoeuvres:


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

It is assumed that the manoeuvre is completed in two parts, each involving


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

The total change in the heading must be equal to zero. Consequently,

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

Aft-most CG location critical for stability (least stable case). Forward-


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

For complete aeroplane, increase pitch rate derivatives by a


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

Pitch damping derivative (horizontal tail only),


∂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

(iv) Roll rate derivatives


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

⎡ Δu ⎤ ⎡ −0.0082 −25.7084 0 −32.1709 ⎤ ⎡ Δu ⎤ ⎡ −6.8094 ⎤


⎢ 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

(i) Obtain a pair of first-order equations to approximate the short


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

⎡ Δu ⎤ ⎡ −0.0193 −0.0516 0 −16.4543 ⎤ ⎡ Δu ⎤ ⎡ −0.0442 ⎤


⎢ 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 4 + 1.4106λ 3 + 0.8238λ 2 + 0.0380λ + 0.0200


( )(
= 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

(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 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

⎡ Δv ⎤ ⎡ −0.0893 1.2546 1.0311 8.4045 ⎤ ⎡ Δv ⎤


⎢ 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 ⎦

(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 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 ⎦

(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) 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

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

⎡ Δv s ⎤ ⎡ yv yp yr yφ ⎤ ⎡ Δvs ⎤
⎢ Δp ⎥ ⎢
⎢ s ⎥ = ⎢ lv lp lr 0 ⎥⎥ ⎢⎢ Δps ⎥⎥
.
⎢ Δrs ⎥ ⎢ 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

⎡ Δv s ⎤ ⎡ yv yp yr yφ 0 ⎤ ⎡ Δvs ⎤ ⎡ yξ y ηd yca yζ ⎤


⎢ Δp ⎥ ⎢ ⎡ Δξ ⎤
⎢ s ⎥ ⎢ lv lp lr 0 0 ⎥⎥ ⎢⎢ Δps ⎥⎥ ⎢⎢ lξ lηd lca lζ ⎥⎥ ⎢
Δηd ⎥⎥
⎢ Δrs ⎥ = ⎢ 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

They can be expressed in state-space form as x = Ax + Bu , where

⎡ −0.0507 −3.861 0 −32.17 0 0 ⎤ ⎡ 0 ⎤


⎢ −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 ⎥⎦

(i) Obtain a pair of first-order equations to approximate the short


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⎦ ⎥ ⎣

(i) Obtain an approximate equation for the roll subsidence mode.


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

(i.e. constant section lift coefficient), the wing rolling moment due to


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

6.13 An aircraft has the specifications defined in Exercise 3.5. In addi-


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.

Answers to Selected Exercises


⎡ 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].

7.2  Longitudinal and Lateral Modal Equations


The longitudinal small perturbation aircraft equations of motion (EOMs)
were shown in Chapter 5 to be

⎡1 −X w − X q 0 ⎤ ⎡ Δu s ⎤ ⎡ Xu Xw Xq − g cos θe ⎤ ⎡ Δus ⎤


⎢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

The throttle setting is normalised, so the case of Δτ = 1 corresponds maxi-


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

⎡1 0 − xq 0 ⎤ ⎡ Δu s ⎤ ⎡ xu xw xq xθ ⎤ ⎡ Δus ⎤


⎢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⎦

or, in terms of the angle-of-attack perturbation Δαs, as

⎡ 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

It may be expressed in state vector form as

⎡ Δu s ⎤ ⎡ Δus ⎤ ⎡ Δu s ⎤ ⎡ Δus ⎤


⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ 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 ⎥
⎣ ⎦ ⎣ ⎦

The lateral small perturbation aircraft EOMs were shown in Chapter 5 to be

⎡1 −Yp −Yr 0 ⎤ ⎡ Δv s ⎤


⎢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 ⎥ ⎢ Δrs ⎥
⎢ ⎥⎢ ⎥
⎣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 ⎥ ⎢ Δrrs ⎥ ⎢ 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)
+ ⎢ ⎥
⎢ Δrs ⎥ ⎢ 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)
⎢ Δrs ⎥ ⎢ 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

The lateral small perturbation equations may also be expressed in state-


space form as

⎡ Δv s ⎤ ⎡ Δvs ⎤ ⎡ Δβ s ⎤ ⎡ Δβs ⎤


⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ Δp s ⎥ ⎢ Δps ⎥ ⎡ Δξ ⎤ ⎢ Δp s ⎥ ⎢ Δps ⎥ ⎡ Δξ ⎤
⎢ ⎥ = A⎢ ⎥ + B ⎢ ⎥ or as ⎢ ⎥ = A⎢ ⎥ + B ⎢ ⎥ , (7.13)
⎢ Δrs ⎥ ⎢ Δrs ⎥ ⎢⎣ Δζ ⎥⎦ ⎢ Δrs ⎥ ⎢ 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 ⎥
⎣ ⎦ ⎣ ⎦

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 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