Institute for Flight Systems and Automatic Control

Flight Mechanics II
Flight Dynamics

by
Prof. Dr.-Ing. U. Klingauf

1. English Edition
Summer Semester 2022
Contents III

Contents
List of abbreviations V

1 Introduction 1
1.1 Flying / Handling Qualities . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Flight Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Aircraft Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2 Flight Motion with Six Degrees of Freedom 11

2.1 Preliminary Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 Choice of Coordination Systems . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Finding the Equations of Motion . . . . . . . . . . . . . . . . . . . . . . . . 13
2.4 External Forces and Moments . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.4.1 Weight FG , MG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.4.2 Thrust FT , MT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.4.3 Aerodynamic Forces/Moments FA , MA . . . . . . . . . . . . . . . . . 17
2.5 Nonlinear Aircraft Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.6 Transformation of the Force Equations into the Aerodynamic System . . . . 18
2.7 Decoupling in Longitudinal and Lateral Motion . . . . . . . . . . . . . . . . 21
2.7.1 Longitudinal Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.7.2 Lateral Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

3 Static Stability 25
3.1 Stationary Longitudinal Motion . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.1.1 Lift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.1.2 Pitching Moment due to Angle of Attack (Cmα ) . . . . . . . . . . . . 27
3.1.3 Neutral Point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.1.4 Determination of the required Tail Volume . . . . . . . . . . . . . . . 34
3.1.5 Moment Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.1.6 Fixed and free Elevator . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.1.7 Static Stability during pull-up maneuver . . . . . . . . . . . . . . . . 41
3.1.8 Elevator Hand Forces with direct (aerodynamic) control . . . . . . . 43
3.1.9 Trimming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.1.10 Control Aids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.1.11 Summary of the Longitudinal Motion Derivatives . . . . . . . . . . . 51
3.2 Stationary Lateral Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.2.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.2.2 Yawing Motion (Single Degree of Freedom) . . . . . . . . . . . . . . . 55
3.2.2.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.2.2.2 Side Force Derivatives . . . . . . . . . . . . . . . . . . . . . 55
IV Contents

3.2.2.3 Yaw Moment Derivatives . . . . . . . . . . . . . . . . . . . 56

3.2.2.4 Rudder Trim . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.2.2.5 Rudder Foot Force . . . . . . . . . . . . . . . . . . . . . . . 61
3.2.3 Rolling Motion (Single Degree of Freedom) . . . . . . . . . . . . . . . 62
3.2.3.1 General Definitions . . . . . . . . . . . . . . . . . . . . . . . 62
3.2.3.2 Side Force Derivatives . . . . . . . . . . . . . . . . . . . . . 63
3.2.3.3 Roll Moment Derivatives . . . . . . . . . . . . . . . . . . . . 64
3.2.3.4 Stationary Rolling . . . . . . . . . . . . . . . . . . . . . . . 67
3.2.3.5 Aileron Hand Force . . . . . . . . . . . . . . . . . . . . . . . 69
3.2.4 Cross-Couplings between Yawing And Rolling Motion . . . . . . . . . 70
3.2.4.1 Roll Moment due to Sideslip Angle . . . . . . . . . . . . . . 70
3.2.4.2 Roll Moment due to Yaw Rate . . . . . . . . . . . . . . . . 71
3.2.4.3 Yaw Moment due to Roll Rate . . . . . . . . . . . . . . . . 72
3.2.4.4 Yaw Moment due to Aileron Deflection . . . . . . . . . . . . 74
3.2.5 Summary of the Lateral Motion Derivatives . . . . . . . . . . . . . . 77
3.2.6 Stationary Sideslip Flight . . . . . . . . . . . . . . . . . . . . . . . . 77
3.2.7 One Engine Out Flight . . . . . . . . . . . . . . . . . . . . . . . . . . 78
3.2.8 Stationary Turning Flight . . . . . . . . . . . . . . . . . . . . . . . . 80

4 Dynamic Stability 85
4.1 Stability Axis System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
4.2 Linearization Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
4.2.1 Linearization of Differential Equations . . . . . . . . . . . . . . . . . 87
4.2.2 Linearization of Trigonometric Functions . . . . . . . . . . . . . . . . 88
4.3 Dynamic Longitudinal Stability . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.3.1 Linearization of the Equations of Motion . . . . . . . . . . . . . . . . 89
4.3.2 Longitudinal Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.3.3 Poles of the Longitudinal Motion . . . . . . . . . . . . . . . . . . . . 97
4.3.4 Requirements on Handling Qualities . . . . . . . . . . . . . . . . . . 97
4.3.5 Example of the derivatives of the longitudinal motion of the A300 . . 99
4.4 Dynamic Lateral Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.4.1 Linearization of the Equations of Motion . . . . . . . . . . . . . . . . 100
4.4.2 Lateral Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
4.4.3 Poles of the Lateral Motion . . . . . . . . . . . . . . . . . . . . . . . 106
4.4.4 Requirements on Handling Qualities . . . . . . . . . . . . . . . . . . 107
Contents V

List of abbreviations
EASA European Aviation Safety Agency
 1

1 Introduction

The lecture Flight Mechanics II deals with the dynamic behavior of aircraft. Although an
aircraft can be viewed as a single mass object with three translational degrees of freedom
for the evaluation of its performance (cf. Flight Mechanics I) the evaluation of its dynamic
behavior requires all spatial equations of motion. During this lecture the aircraft will be
perceived as a rigid body so that its spatial motion is described with six degrees of freedom
(three translational, three rotational). The order is further increased if elastic degrees of
freedom are included in the model (e.g., elastic wing bending). These effects are studied in
aero elasticity, which is not a part of this lecture.
The following sections will deal with essential terminology in flight dynamics. Subsequently
in Chapter 2 the non-linear equations of motion of an aircraft will be established. In order to
analyze these equations with the methods of system theory and control systems engineering,
the equations will be linearized. Analyzing symmetrical aircraft (all operational aircraft,
with minor exceptions, are considered symmetrical in their xb -zb -plane) allows to describe the
equations of motion in longitudinal and lateral motion with three degrees of freedom. This
reduces the operational capacity required to establish the equations of motion. Continuing
on this basis, in Chapter 3, the conditions for a stationary flight (moment equilibrium) are
derived. This is also denoted as stationary stability.
In Chapter 4, the dynamic behavior of an aircraft, commonly known as its flying qualities (or
handling qualities), is analyzed. The dynamic behavior includes the control and disturbance
behavior of the aircraft. In modern aircraft, the dynamic behavior is increasingly controlled
by flight controllers. In an integrated design process, the aircraft design is significantly
influenced by the advancements and possibilities of system control. This leads to the
point where aircraft are deliberately designed unstable to improve flight performance (e.g.,
improved agility, reduced fuel consumption), using flight controllers to impose dynamic
stability ("control configured vehicle").

1.1 Flying / Handling Qualities

Besides achieving the required flight performances (cf. Flight Mechanics I), every new aircraft
design requires a very careful advanced calculation of the static and dynamic behavior in
the entire flight envelope in order to verify that the handling quality requirements are met.
In this context, the term flying qualities covers the overall behavior of the aircraft in the
presence of external disturbances (e.g. wind gusts) and in maneuvers initiated by the pilot.
While it is relatively easy to evaluate flight performance by specifying lengths, times, velocities
and accelerations, the numerical evaluation and objective assessment of flying qualities
2 1 Introduction

always causes great difficulties. For a long time, aircraft were designed using standard
values gained from experiences of similar projects. Afterwards, the fine-tuning was done
by test pilots. It was not uncommon, that major design changes were necessary during the
flight tests (such as new stabilizers and rudders, fuselage extensions, etc.).
As computational methods became more sophisticated, specifications were established for
determining of the most important dynamic quantities, which differ for civil and military
applications. These are listed in the specifications for flying and handling qualities, which are
binding for every new aircraft design today (e.g., European Aviation Safety Agency (EASA)
Certification Specifications CS-23/25, in the military field formerly MIL-F-8785, currently
MIL-STD-1797).
From a control point of view, the flying qualities include the natural behavior, the control/input
behavior and the disturbance response of the aircraft. The natural behavior refers to the
frequency and dampening of the natural modes, which are determined by the position
of the eigenvalues (poles). The control behavior describes the response of the aircraft
to control inputs from the pilot (command/control transfer function). The translational
(orbital) degrees of freedom have a minor role, since the responses are slow and even a
slight instability can be controlled by the pilot easily. In contrast, stability in the rotational
degrees of freedom (roll, pitch, yaw) is very important. Natural frequency and dampening
must be in a range that represents a good compromise between stability and agility and is
perceived as comfortable by the pilot. For pilot evaluation, the Cooper-Harper rating scale
(see Figure 1.1) is used. Finally, the disturbance response is determined by the reaction
of the aircraft to external stimuli (disturbances), e.g., wind gusts, and is described by the
disturbance transfer function.

Figure 1.1: The Cooper-Harper Rating Scale

1.2 Stability 3

The Cooper-Harper rating scale can be used for rating the handling qualities of an aircraft.
After the flight test, the pilot starts completing the Cooper-Harper rating scale questionnaire
in the lower left corner. As the pilot answers the questions, they move across the questionnaire
until they receive a "Pilot Rating". The rating scale ranges from a value of "1" to "10".
"1" implies that the aircraft has good or excellent handling qualities, while a value of "10"
means that the rated aircraft is not controllable. The rating scale is divided into four groups,
which provide different measures for improving the handling qualities.

1.2 Stability

An aircraft is stable when it returns to its initial state after a disturbance without pilot
action. The initial state is generally considered to be a stationary flight state, which is
characterized by the fact that all variables (states), representing the flight motion, do not
change over time. Examples of stationary flight conditions are unaccelerated horizontal flight
or unaccelerated climb.
An essential prerequisite for stable behavior is the automatically appearance of a restoring
moment after a disturbance, returning the aircraft to its initial position. The moment must
therefore have the characteristics of a return spring with positive spring stiffness in order
to create a stable equilibrium position. This property is called static stability in flight
mechanics.
The static analysis is not sufficient to evaluate the dynamic process of returning to the
equilibrium point, which is for a stable system the decreasing of amplitude over time or the
dampening. For this purpose, the dynamic stability of the flight motion in the presence of
small perturbations around a stationary initial state must be investigated. The dynamic
stability can only exist, if there are restoring moments for each perturbation, meaning the
static stability also exists. This corresponds to the control engineering definition of stability.
Thus, a system is dynamically stable if the poles have a negative real part. Dynamic stability
is of crucial importance for the pilot’s ability to control the aircraft. The desired handling
qualities are determined by the requirement to be able to transfer the aircraft as safely and
quickly as possible from one state of equilibrium to the other.
The following terms are used for the stability of the aircraft motion around the three axes:
• Lateral Stability: the stability around the longitudinal axis xb (Roll)
• Longitudinal Stability: the stability around the lateral axis yb (Pitch)
• Directional Stability : the stability around the vertical axis zb (Yaw)

1.3 Flight Control

Essentially, a distinction between primary and secondary flight control is made. The primary
flight control includes all safety-critical control elements. The secondary flight control
includes other control devices that are required for certain phases of flight, e.g., for takeoff
4 1 Introduction

and landing.

Figure 1.2: Control surfaces of an A320 [Brockhaus]

The primary flight control consists of the different rudders (control surfaces) on the wings and
the empennage (tail). When deflecting the ailerons, elevator or rudder (ξ, η, ζ), moments (L,
M , N ) about the longitudinal, lateral and vertical axis (xb , yb , zb ) are created to change the
attitude (Φ, Θ, Ψ) of the aircraft. Figure 1.3 depicts the relationship of rudder deflections
and created moments. Secondary flight control includes the high-lift system with slats and
flaps, spoilers and nose wheel control.

t Path
Fligh
Horizon

Pitch

Sideslip

Yaw
Lateral
Axis

Aileron
Longitudinal Axis

Rudder

Aileron

Roll Lateral Axis

Figure 1.3: Correlation of rudder deflections and created moments

1.3 Flight Control 5

Using a mechanical flight control system, ailerons and elevator are operated by a control
stick or a yoke, the rudder is operated by the pedals (cf. see Figure 1.4).

FH
- ξL
Stick
+ FP
+!
Left
Aileron
Tailplane Pedals

Vertical Right
Stabilizer Aileron

+ ξR

Fig. 1.1.1 +η

Figure 1.4: Correlation of pilot input and control surfaces

Further control elements are the throttle lever, which can be used to set thrust for each
engine, the trim wheel, trim lever or trim switch, which can be used to adjust the setting
angle of the horizontal tail for the purpose of longitudinal trim. Alternatively, a trim tab
attached to the elevator can be adjusted. Trimming from the view of a pilot means, that
the trim devices are used to set the control input forces (hand or foot force) to zero in a
steady-state flight condition. This is particularly relevant when larger input forces are to be
applied by the pilot.
In the latter case the dimensioning of the input forces (hand forces for longitudinal and
lateral control as well as the foot forces for directional control) is of essential importance.
The input forces must always be in the correct ratio to the corresponding control effects,
i.e., the flight movements following a control actuation, in order to provide the pilot with
feedback on the control input and the flight condition.
A distinction is made between "aerodynamic controls", in which the pilot must directly
apply the forces to overcome the aerodynamic control surface hinge moments, and "servo
controls", in which the pilot does not receive feedback of the control surface hinge moments
but rather actuates hydraulic or electric actuators through his control movements. Servo
controls became necessary as the size and speed of modern aircraft increased. To give the
pilot a sense of steering, "artificial" steering forces are often generated by separate systems.
In all cases, careful tuning of input forces in the various axes is necessary.
In modern "fly-by-wire" flight control systems, there is often no longer a direct link between
the pilot’s control inputs and the rudder deflections. Instead, the pilot inputs represent set
values, e.g., for a set pitch rate. The flight controller then converts the control inputs into
corresponding control surface deflections. In some cases, the controller uses several control
surfaces simultaneously to achieve the set value. With this concept, the approach of an
artificial control feeling can no longer be implemented without further systems. The aircraft
of the Airbus family therefore have passive sidesticks whose actuating forces are realized via
6 1 Introduction

(nonlinear) spring characteristics.

When controlling the longitudinal motion, the trim point can be changed by operating the
elevator or, as a special form, by shifting the aircraft’s center of gravity. This procedure
was first used by Otto Lilienthal in his glider. Among others, this procedure is used e.g.
in supersonic aircraft. By pumping the fuel in different tanks (aftward), the aircraft’s
longitudinal stability is reduced, thus also the occurring control drag. In vertical takeoff
aircraft, the natural shifting of the center of gravity when fuel is flown out is compensated
by pumping the residual fuel in different tanks. The shift of the center of gravity in flight
is generally unavoidable (e.g., passenger movement, load airdrop) and should be within the
allowable tolerance. This is an important consideration when designing the horizontal tail
size.
Another form of control used in the longitudinal and vertical axes is thrust vector control.
On the one hand this can increase agility, as very high moments can be applied by even a
small rotation of the thrust vector. On the other hand the thrust vector can also be used
to trim in the longitudinal axis. Thrust vector control is primarily used on combat aircraft
(e.g., Lockheed Martin F-22 Raptor, Sukhoi Su-27 "Flanker"). In vertical takeoff aircraft
control moments in the hover phase are either generated by control nozzles or applied by
thrust modulation of the main engines. Control nozzles have the advantage of very small
response times, but represent a relatively large permanent thrust loss. The use of modulated
engine thrust for control purposes has the advantage that relatively large control moments
are available, but it requires arrangements with engines decentralized to the center of gravity.

1.4 Aircraft Configurations

Depending on the arrangement and presence of the vertical tail, a distinction is made between
basic aircraft configurations. The classic and still most frequently used configuration is the
so-called kite configuration, in which the horizontal tail is located behind the wing. In the
canard configuration, on the other hand, the horizontal stabilizer is located in front of the
wing. The first aircraft in canard configuration was the Wright Brothers’ flyer. In addition,
there are also various designs without a horizontal tail. In this case, the airfoil must be
designed by a specific profile or configuration so that it is able to fly inherently stable (e.g.,
flying wing with reflexed airfoil profiles, delta configuration, blended wing body).
Figure 1.5 and 1.6 depict typical kite configurations. In this concept the functions of the
wings (generating lift) and the empennage (generating moments about the lateral and vertical
axes) are clearly separated. Due to the large lever arm, the horizontal and vertical tail
can be comparatively small. Moreover, the wing lies in the undisturbed inflow and can
therefore operate efficiently. A disadvantage is that the horizontal tail usually has to generate
downforce to achieve sufficient static stability. Furthermore, a pitching-up moment can only
be generated by additional downforce at the horizontal tail, which reduces the overall lift.
When initiating a climb, the aircraft therefore initially loses altitude (all-pass behavior).
1.4 Aircraft Configurations 7

Figure 1.5: Motorsegler Diamond HK 36 Super Dimona

Figure 1.6: Airbus A320

These disadvantages are avoided in the canard configuration (Figure 1.7, Figure 1.8). To
maintain longitudinal stability, the canard (horizontal stabilizer) usually generates lift. To
initiate a climb, additional lift is generated by increasing the effective angle of attack of the
canard through rotation, causing the aircraft to pitch up and climb immediately. The major
disadvantage of the canard is the disturbed inflow at the wing behind the canard, which
leads to increased drag. In supersonic aircraft delta wings are often combined with canards,
although the canards are used only as auxiliary control surfaces at large angles of attack
(Tu-144) or to control unstable aircraft designs (EF2000).
8 1 Introduction

Figure 1.7: Gyroflug Speed Canard

Figure 1.8: Eurofighter EF2000 Typhoon

Flying wing aircraft do not require a horizontal tail or stabilizer. In this case, the drag of
the horizontal tail is eliminated. However, special measures are required to achieve static
stability. This can be achieved, for example, by sweeping the wing or reflexed airfoils profiles.
Nevertheless, the degree of stability is generally lower than with a kite configuration. Another
disadvantage is the short lever arm of the control surfaces to the center of gravity, which
is why relatively high forces are required to generate the control moments. Examples of
flying wing aircraft are the Horten gliders (Figure 1.9) and the blended wing body concepts
(Figure 1.10), which are currently being investigated by NASA. This is expected to result in
greater economic efficiency for future commercial aircraft.

Figure 1.9: Horten IX V-1

1.4 Aircraft Configurations 9

Figure 1.10: NASA X-38

 11

2 Flight Motion with Six Degrees of

Freedom

2.1 Preliminary Conditions

In the context of this lecture the following assumptions and conditions are made for establishing
and simplifying the equations of motion:

• The Earth is viewed as an inertial system (planar and at rest).

• The aircraft is considered as a rigid body without elastic degrees of freedom. This
is permissible as long as the frequencies of the elastic eigenmodes are significantly
larger than those of the rigid body motion. Assumed that the rudder deflections are
independent of the motion ("fixed rudders"), the aircraft represents a system with six
degrees of freedom.
• The aircraft is assumed to be symmetrical regarding the xb -zb -plane. Thus, the cross-
products of inertia Ixy and Iyz vanish if they are given in the body-fixed coordinate
system.
• Aerodynamic moments and gyroscopic moments of the engines are neglected. However,
the moment caused by the distance between the installation axis of the engine and
the center of gravity of the aircraft (eccentric thrust installation) is still considered.
Neglecting gyroscopic moments can lead to significant errors in small, high-thrust
aircraft with high rotational speeds (e.g., fighter aircraft) or in large rotating masses
(large, heavy propellers) due to the gyroscopic effect. Likewise, air flow through the
propeller plane at an angle to its plane of rotation leads to a shift of the thrust vector
towards the "leading" propeller blade (similar to a helicopter). This effect, also called
"p-factor", is particularly pronounced in light propeller aircraft at low-speed flight at
high angles of attack (initial climb after takeoff).
• The aircraft mass and mass distribution are independent of time. Thus, for example,
fuel consumption during a maneuver is neglected.
• All forces acting on the aircraft are combined into resultants and related to the center
of gravity.
• The air forces behave proportionally to the aerodynamic influencing variables (e.g.,
angle of attack), dynamic effects are neglected (e.g., force build-up over time).
• The mutual influence of the flow around the airframe and the engine is neglected.
12 2 Flight Motion with Six Degrees of Freedom

• The wind influence is neglected.

• Ground effects are neglected.

Thus, the rigid body motion is caused solely by the following forces and moments, which by
definition act on the center of gravity:

• Aerodynamic forces and moments

• Thrust forces and moments
• Mass forces and moments (weight, inertia)

2.2 Choice of Coordination Systems1

The equations of motion can be set up in different coordinate systems depending on the
purpose. For the description of the flight path the geodetic coordinate system is depicted
as an earth-fixed system. However, this has the disadvantage that the inertia tensor of
the aircraft is a time-variable and thus the transformation into the earth-fixed system
becomes complicated. Moreover, the aerodynamic quantities have to be transformed from
the aerodynamic to the earth-fixed system.
The latter transformation is omitted when the aerodynamic coordinate system is chosen.
However, the weight, the thrust as well as the inertia tensor have to be transformed, albeit
over the comparatively small angles α and β.
The establishment of the equations of motion becomes simpler with the choice of an body-
fixed coordinate system. This is preferably aligned with the principal axes so that the
cross-products of inertia would vanish. In this case the inertia tensor is a constant diagonal
matrix, the thrust is fixed to the aircraft, so that in this case a transformation is not necessary.
Only the aerodynamic quantities have to be transformed into the principal axis system by
relatively small angles. Therefore, for the establishment of the 6-degree-of-freedom model
the aircraft-fixed principal axis system is then chosen. The quantities without index are
related to the principal axis system, only the principal moments of inertia are given the
index "i" for the sake of clarity.
As an alternative, the so-called stability axis system can be used, which corresponds to
the aerodynamic system in the reference state αT r , βT r = 0. Compared to the body-fixed
system, this system is merely rotated by the fixed angle αT r and is particularly suitable for
the analysis of dynamic stability. It is therefore used in chapter 4 and defined in more detail.
If an body-fixed coordinate system is chosen as the reference frame, moments can occur on
the aircraft on the longitudinal, lateral and vertical axes, see Figure 2.1. The origin of the
coordinate system is located in the aircraft’s center of gravity, however, its position is viable
to change (payload, fuel consumption). The rotations about these three axes are called
rolling, pitching and yawing.
1
also see chapter Beschreibung der flugmechanischen Achsensysteme in the lecture script of Flugmechanik
I
2.3 Finding the Equations of Motion 13

Figure 2.1: The motions around the coordinate axes [Brockhaus]

2.3 Finding the Equations of Motion

The equations of motion for a rigid body, with respect to a rotating coordinate system, are
given by Newton’s second law of motion: the principle of linear momentum

  
dV
F=m +Ω×V (2.1)
dt Ω=0

and the principle of angular momentum

 
dB
M= + Ω×B (2.2)
dt Ω=0

dV V
= W x
dt

V
yn

xn

zn

Figure 2.2: Derivation of a vector in a rotating coordinate system [Brockhaus]

14 2 Flight Motion with Six Degrees of Freedom

With the velocity vector

 
u
V= v  (2.3)
w

and the rotational velocity vector

 
p
Ω= q  (2.4)
r

follows the acceleration

 
  u̇
dV
= v̇ 
 (2.5)
dt Ω=0 ẇ

and the Coriolis term:

 
i j k qw − rv
Ω × V = p q r =  ru − pw  . (2.6)
u v w pv − qu

The angular momentum vector of the aircraft, neglecting the rotating engine parts, is

  
Ixi 0 0 p
(B)ΩE =0 = 0 Iyi 0
   q . (2.7)
0 0 Izi r

This equates to

 
  Ixi ṗ
dB
= Iyi q̇  (2.8)
dt
Izi ṙ

and

 
(Izi − Iyi ) qr
Ω × B = (Ixi − Izi ) pr  . (2.9)
(Iyi − Ixi ) pq

If one considers the mass distribution similar to Figure 2.3 as a distribution of point masses
then the moments of inertia can be stated as follows:
2.4 External Forces and Moments 15

x m1 m3
2 2 y
r3
r1 m2
2
r2

m2 m3 m1
2 2 2
z
Figure 2.3: Model of mass distribution on the aircraft ("dumbbell model")

m = m1 + m2 + m3
Ixi = m2 r22 + m3 r32
Iyi = m1 r12 + m3 r32
Izi = m1 r12 + m2 r22

2.4 External Forces and Moments

As external forces we assume weight, thrust and aerodynamic forces. They are each to be
transformed into the principal axis system.

2.4.1 Weight FG , MG

Assuming that the principal axis system is rotated by the angle ι with respect to the body-
fixed system (design axes, see Figure 2.4), the weight FG is transformed into the principal
axis system by the Euler angles and ι:

 
− sin Θ cos ι − cos Θ cos Φ sin ι
FG =  cos Θ · sin Φ  · mg (2.10)
− sin Θ sin ι + cos Θ cos Φ cos ι

Since the weight acts at the center of gravity the resulting moment is MG = 0.
16 2 Flight Motion with Six Degrees of Freedom

y = yb = ye

x
ze
xb
zb
z

xa
xe

Figure 2.4: Coordinate systems: principal axes, body-fixed, aerodynamic and experimental
system)

2.4.2 Thrust FT , MT

FFT
σσ
xx

ιι SP
CG
xxbf rrFT
αα

xexe

Figure 2.5: Lever arm and rotation of the thrust vector

According to Figure 2.5, the thrust force FT has the lever arm rT to the center of gravity
and is rotated by the angel σ with respect to the body-fixed longitudinal x-axis. The
transformation into the principle axis system thus results in the following forces and moments
for a symmetrical thrust installation:

 
cos (σ − ι)
FT =  0  · FT , (2.11)
− sin (σ − ι)
 
0
MT = rT  · FT .
 (2.12)
0
2.5 Nonlinear Aircraft Model 17

2.4.3 Aerodynamic Forces/Moments FA , MA

The aerodynamic forces and moments are transformed from the experimental system into
the principal axis system as follows:

  
cos (α + ι) 0 − sin (α + ι) −De
FA =  0 1 0   Ye  , (2.13)
sin (α + ι) 0 cos (α + ι) −Le
  
cos (α + ι) 0 − sin (α + ι) le
MA =  0 1 0  me  . (2.14)
sin (α + ι) 0 cos (α + ι) ne

2.5 Nonlinear Aircraft Model

Using the equations 2.10 to 2.14 in compliance with

ρ
K = CK V 2 S (K = De , Ye , Le )
2
and
ρ
Le = Cle V 2 Ss,
2
ρ
Me = Cme V 2 Slµ ,
2
ρ 2
Ne = Cne V Ss
2

the equations for the forces (2.1) and moments (2.2) are obtained:

     
u̇ rv − qw − sin Θ cos ι − cos Θ cos Φ sin ι
m  v̇  = m pw − ru + mg  cos Θ sin Φ 
ẇ qu − pv − sin Θ sin ι + cos Θ cos Φ cos ι
   
cos (σ − ι) −CDe cos (α + ι) + CLe sin (α + ι)
+F  0  + ρ V 2S  C Ye  (2.15)
2
− sin (σ − ι) −CDe sin (α + ι) − CLe cos (α + ι)

       
Ixx ṗ (Iyy − Izz ) qr 0 Cle cos (α + ι) − Cne sin (α + ι)
 Iyy q̇  = (Izz − Ixx ) pr  + F rT  + ρ V 2 Ss  lµ
C  (2.16)
2 s me
Izz ṙ (Ixx − Iyy ) pq 0 Cle sin (α + ι) + Cne cos (α + ι)

To completely describe the system (2.15 and 2.16), the equations for the change of the
attitude angles (kinematics) and for the motion of the center of gravity of the aircraft in the
geodetic system (navigation) are still required.
18 2 Flight Motion with Six Degrees of Freedom

If Ψ, Θ, Φ are the Euler angles relative to the body-fixed system xb , yb , zb (cf. DIN 9300-1),
then the following applies

      
Φ̇ 1 sin Φ tan Θ cos Φ tan Θ cos ι 0 sin ι p
Θ̇ = 0 cos Φ − sin Φ  ·  0 1 0  q  (2.17)
sin Φ cos Φ
Ψ̇ 0 cos Θ cos Θ
− sin ι 0 cos ι r
and
   
x˙g cos Θ cos Ψ sin Θ sin Φ cos Ψ + cos Φ sin Ψ sin Θ cos Φ cos Ψ − sin Φ sin Ψ
 y˙g  =  cos Θ sin Ψ sin Θ sin Φ sin Ψ − cos Φ cos Ψ sin Θ cos Φ sin Ψ − sin Φ cos Ψ
z˙g − sin Θ cos Θ sin Φ cos Θ cos Φ
 
cos α cos β
·  sin β  V (2.18)
sin α cos β

The equations 2.15 to 2.18 represent a coupled system of 12 nonlinear ordinary differential
equations of first order. Since the equations 2.15 and 2.16 are independent of the azimuth
angle Ψ, these two equations can be decoupled from the overall system together with the
first two equations of 2.17, creating a system of 8 differential equations. If the solution of
this system is known, Ψ (last equation in 2.17) and the center-of-mass trajectory (2.18) can
be calculated by simple integration.
If the control input variables (throttle position δT , elevator deflection angle η, aileron deflection
angle ξ and rudder deflection angle ζ) are combined to the input vector u,

uT = δT η ξ ζ (2.19)
 

and the dependence of the aerodynamic forces and moments as well as the thrust on these
input variables is modeled, we get the aircraft model in state space representation

ẋ = f (x, u) (2.20)

which is based on equations 2.15 to 2.18 with the state vector

xT = u v w p q r Φ Θ Ψ xg yg zg . (2.21)
 

This system of equations can be used as basis for a numerical flight simulation.

2.6 Transformation of the Force Equations into the

Aerodynamic System

It is often convenient to represent the force equations in 2.15, which have been derived for the
(aircraft-fixed) principal axis system, in the aerodynamic system. Instead of the velocities
2.6 Transformation of the Force Equations into the Aerodynamic System 19

u, v, w, the states V , α, β are now used.

The velocity vector is in aerodynamic coordinates:

 
V
V = 0
 (2.22)
0

The rotational velocities can be calculated by transforming the sum of the rotational velocities
in the principal axis system and the relative rotation of the aerodynamic system to the
principal axis system. When transforming the body-fixed rotational velocities into the
aerodynamic system, the relative rotational velocities between the two coordinate systems
must be must be taken into account. This means for the relative rotational velocities in the
body-fixed system:

Ωba
b = −α̇jb + β̇ke

Figure 2.6: Transformation from body-fixed (b) to aerodynamic (a) coordinate system.

The rotational velocity α̇ points in the y-direction of the body-fixed system (unit vector jf
), the rotational velocity β̇ points in the z-direction of the experimental coordinate system
(unit vector ke ). The negative sign of α̇ results from the fact that an increase of α from a
body-fixed point of view leads to a negative rotation of the aerodynamic system. Expressed
in body-fixed coordinates this results for the relative rotational velocity:
20 2 Flight Motion with Six Degrees of Freedom

   
0 0
ba
Ωb = −α̇
  + Tbe · 0 

0 b β̇ e

By means of the general transformation matrix

 
cos α 0 − sin α
Tbe =  0 1 0 
sin α 0 cos α

follows
 
−β̇ sin α
Ωba
b =
 −α̇ 
β̇ cos α b

This results in the total rotational velocity Ωa , taking into account the superimposed rotation
angles from the principal axis:

   
cos (α + ι) cos β sin β sin (α + ι) cos β p − β̇ sin (α + ι)
Ω = − cos (α + ι) sin β cos β − sin (α + ι) sin β  ·  q − α̇  (2.23)
− sin (α + ι) 0 cos (α + ι) r + β̇ cos (α + ι)

For the Coriolis term, we get:

 
0
Ω×V = −p sin (α + ι) + r cos (α + ι) + β̇ ·V (2.24)
p cos (α + ι) sin β − q cos β + α̇ cos β + r sin (α + ι) sin β

With the forces

   
cos α cos β sin β sin α cos β − sin Θ
FG = − cos α sin β cos β − sin α sin β  ·  sin Φ cos Θ  · mg, (2.25)
− sin α 0 cos α cos Φ cos Θ
 
cos (α + σ) cos β
FT = − cos (α + σ) sin β  · F , (2.26)
− sin (α + σ)
     
−D cos β sin β 0 −De
FA =  Y  = − sin β cos β 0 ·  Ye  (2.27)
−L 0 0 1 −Le

we finally obtain the force equations 2.15 in the aerodynamic system:

2.7 Decoupling in Longitudinal and Lateral Motion 21

   
V̇ 0
m  β̇V  = mV  p sin (α + ι) − r cos (α + ι) 
α̇V cos β −p cos (α + ι) sin β + q cos β − r sin (α + ι) sin β
 
− cos α cos β sin Θ + sin β sin Φ cos Θ + sin α cos β cos Φ cos Θ
+ mg  cos α sin β sin Θ + cos β sin Φ cos Θ − sin α sin β cos Φ cos Θ 
sin α sin Θ + cos α cos Φ cos Θ
   
cos (α + σ) cos β −cDe cos β + cYe sin β
ρ
+ F − cos (α + σ) sin β  + V 2 S  cDe sin β + cYe cos β  . (2.28)
2
− sin (α + σ) −cLe

2.7 Decoupling in Longitudinal and Lateral Motion

2.7.1 Longitudinal Motion

The previously established nonlinear equations of motion can be simplified considerably,

when a sideslip-free flight without a bank angle is considered (β = 0, Φ = 0). This case
is very interesting because it appears quite frequently under normal flight conditions and it
allows a decoupling of the equations of motion for the longitudinal and lateral motion. This
simplifies the analytical study of flight dynamics.
Decoupling means, that the system of equations breaks down into two independent parts.
One part describes the longitudinal motion (pitching motion and translation in the x-z
plane), the other part describes the lateral motion (rolling and yawing motion as well as
flight with a sideslip angle).
For the case β = 0, Φ = 0 and the simplification due to

sin α cos Θ − cos α sin Θ = − sin (Θ − α)

respectively
cos α cos Θ + sin α sin Θ = cos (Θ − α)

the force equations in aerodynamic coordinates (2.28) results as follows:

     
V̇ 0 − sin (Θ − α)
m β̇V  = mV p sin (α + ι) − r cos (α + ι) + mg  0 
α̇V q cos (Θ − α)
   
cos (α + σ) −c
ρ 2  De 
+F  0  + V S c Ye . (2.29)
2
− sin (α + σ) −cLe

In this case, the difference between the inclination angle Θ and the angle of attack α is equal
to the air-path inclination (climb/descent angle) γ:
22 2 Flight Motion with Six Degrees of Freedom

γ|Φ=β=0 = Θ − α (2.30)

It shows that the force equations of the longitudinal motion (see first and third equation in
Equation 2.15) are independent of the quantities of the lateral motion. The same is true in
reverse for the force equation of lateral motion (second equation in Equation 2.15), if the
angle of attack αT r is considered constant.
The moment equations Equation 2.16 can also be easily decoupled. For p = r = 0, the
pitching moment equation (second equation in 2.16) becomes independent of the roll and
yaw moment equations (first and third equations in 2.16).
The longitudinal motion can then be described by the four state variables

xTlong = V (2.31)
 
α q Θ .

In the lateral motion, on the other hand, the four state variables are

xTlat = β p r Φ (2.32)
 

The two systems of equations together describe the dynamic behavior of the aircraft. They
are independent of azimuth angle Ψ as well as the aircraft’s position xg , yg , and altitude zg
(unless there is a significant change in altitude that affects air density ρ).
The simplified equations of motion are then :

   2 
V̇ −g sin (Θ − α) + mF
cos (α + σ) − ρV2mS cDe
 α̇  q + Vg cos (Θ − α) − mV
F
sin (α + σ) − ρV S
c 
2m Le 
 = 2 (2.33)
 q̇   F rT
+ ρV2IyiSlµ cme
 
Iyi

Θ̇ q

It should be noted that V and α are defined in the aerodynamic system (cf. Equation 2.29),
while q is related to the aircraft-fixed principal axes and Θ to the body-fixed design axes (cf.
Equation 2.16 and Equation 2.17).

2.7.2 Lateral Motion

To decouple the equation of motion of the lateral motion, the variables of the longitudinal
motion cannot be set to zero. Rather, a stationary reference state α = αT r , Θ = Θ0 , V =
V0 , F = F0 , q = 0 is assumed. The force equation in y-direction is then:
2.7 Decoupling in Longitudinal and Lateral Motion 23

β̇ = p · sin (αT r + ι) − r · cos (αT r + ι)

g
+ · (cos αT r sin β sin Θ0 + cos β sin Φ cos Θ0 − sin αT r sin β cos Φ cos Θ0 )
V0
F0 ρV0 S
− · sin β cos (αT r + σ) + (CDe · sin β + CYe · cos β) (2.34)
mV0 2m

Assuming small bank angles Φ and due to

sin Θ0 cos αT r − cos Θ0 sin αT r = sin (Θ0 − αT r )

the equation is simplified to

β̇ = p · sin (αT r + ι) − r · cos (αT r + ι)

g g
+ sin β sin (Θ0 − αT r ) + Φ cos β cos Θ0
V0 V0
F0 ρV0 S
− · sin β cos (αT r + σ) + (CDe · sin β + CYe · cos β) (2.35)
mV0 2m

From the force equation in x-direction in Equation 2.33, it follows for the steady state V̇ = 0:

F0 g ρV0 S
cos (αT r + σ) − sin (Θ0 − αT r ) − CDe = 0 (2.36)
mV0 V0 2m

Thus Equation 2.35 is simplified to:

g ρV0 S
β̇ = p · sin (αT r + ι) − r · cos (αT r + ι) + Φ cos β cos Θ0 + CYe cos β (2.37)
V0 2m

The simplified equations of motions for the lateral motion then result to:

  p · sin (α + ι) − r · cos (α + ι) + g Φ cos β cos Θ + ρV0 S C · cos β 

β̇ Tr Tr V0 0 2m Ye
ρV02 Ss
 ṗ   (Cle · cos(αT r + ι) − Cne · sin(αT r + ι)) 
 = 2Ixi (2.38)

 ṙ   ρV02 Ss 

2Izi
(Cle · sin(αT r + ι) − Cne · cos(αT r + ι)) 
Φ̇ p · cos ι + r · sin ι + tan Θ0 (−p · sin ι + r · cos ι)

As in the case of longitudinal motion, the quantities in Equation 2.38 for describing lateral
motion are defined in different coordinate systems. Here, the sideslip angle β is defined in
the aerodynamic system, while p and r are related to the aircraft-fixed principle axes and Φ
to the body-fixed design axes.
 25

3 Static Stability

3.1 Stationary Longitudinal Motion

Longitudinal motion is defined as a symmetrical flight motion in the vertical plane. This
means that the airspeed vector moves within the plane of aircraft symmetry and the bank
angle Φ and the sideslip angle β are equal to zero.
The aircraft is in stationary equilibrium when the sums of the forces and moments are
equal to zero. Besides the stationary horizontal flight (load factor n = 1), the following
considerations also include the quasi-stationary pull-up maneuver, for which n > 1 is valid
and for which the sums of the forces and moments can be set constant for small time interval.
Hereby it is assumed that the forces in the direction of the path are balanced .
Now Z = 0 and M = 0 remain as equilibrium conditions, where M represents the
P P

moment about the lateral axis (pitching moment).

3.1.1 Lift

The lift of an aircraft consists of the contributions of the different aircraft parts such as the
wing, fuselage and empennage (tail). It can be shown that the lift contribution of the fuselage
part in front of the wing is approximately equal to the lift contribution of the imaginary wing
section extended in the fuselage area, so that the total wing lift (including the wing section
located in the fuselage) represents the lift of the wing-fuselage arrangement, i.e., LW = LW F .
According to Figure 3.1 in the aerodynamic coordinate system, the forces orthogonal to the
inflow direction (in the direction of the negative zA -axis) are:
26 3 Static Stability

LAWF
FR

εH
D
WW LHH
A
α�WF V
FR

ichdtuirnegction αW� -F-�αDW DHH

W
Fliggr ht
Flu VH
Xa
�αW
D
η

Za

Figure 3.1: Forces on the wing and tail

Wing + Fuselage: LW F = CLW F · S · q = (CLα )W F · αW · S · q

Horizontal tail: LH = LH · cos αD − WH · sin αD ≈ LH

The dynamic pressure is defined as q = ρ2 V 2 .

Assuming small values for the wing downwash at the horizontal tail αD , the horizontal tail
lift is:

∂αH
LH = CLH SH q H = (CLα )H (αW − αD + ϵH + · η)SH · q H
| {z } ∂η
=αH

The total lift of the aircraft is obtained by superimposing the lift of the wing-fuselage
combination and the horizontal tail:

L = CL · Sq = LW F + LH , or in coefficients
SH q H
CL = CLW F + CLH (3.1)
S q

If the lift is represented with αD = αD0 + ∂α

∂α
D
· αW and an effective setting angle (or angle of
incidence) of the horizontal tail is introduced with ϵ∗H = ϵH + ∂α H
∂η
· η and assuming a linear
dependence of the lift on the angle of attack, we obtain from Equation 3.1:

SH q H ∂αD SH q H ∗
CL = (CLα )W F αW + (CLα )H αW (1 − ) + (CLα )H (ϵ − αW0 ) (3.2)
S q ∂α S q H

The slope of the lift of the aircraft CLα = ∂CL

∂αW
is thus:
3.1 Stationary Longitudinal Motion 27

 
(CLα )H SH q H ∂αD
CLα = (CLα )W F 1+ (1 − ) . (3.3)
(CLα )W F Sq ∂α

As a result of Equation 3.2 and Equation 3.3, the wing’s angle of attack at which the total
lift disappears is:

(CLα )H SH q H ∗
(αW )CL =0 = − (ϵ − αD0 ). (3.4)
CLα S q H

CAL
C
"*H > 0
!WD0 = 0
0

"*H = 0

α
!WF
- (α!WF)
CLA = 0
α!

Figure 3.2: Slope of the lift with respect to the angle of attack

Counting the aircraft’s angle of attack from the angle at which CL = 0 applies, this yields
to α = αW − (αW )CL =0 according to Figure 3.2, or using Equation 3.4 this yields to:

(CLα )H SH q H ∗
α = αW + (ϵ − αD0 ) (3.5)
CLα S q H

3.1.2 Pitching Moment due to Angle of Attack (Cmα )

We can find the equation for the pitching moment of the wing according to the designations
in Figure 3.3 and using the definition M = Cm Slµ q, with the mean aerodynamic chord length
lµ ,
Z b
1 +2 2
lµ = l dy
S − 2b
28 3 Static Stability

LAWF
FR

S.P.

zF +M
D
WWF LH
A
α
!WF V
zH
DHH
W
αW!F--!
αDW MH0
xLAWFF
R

xxcSg

rH

Figure 3.3: Levers arms for moment equilibrium

The pitching moment of the wing results to:

MW = M0W + (LW cos αW + DF sin αW )(xcg − xLW ) + (LW sin αW − DF cos αW )zW . (3.6)

Since small angles αW are considered in general and the drag is relatively small compared
to the lift, we can calculate for Equation 3.6 with the approximation CmW = Sql
mW
µ
:

xcg − xLW
CmW = Cm0W + (CLα )W αW (αW small!) (3.7)
lµ

The contribution of the fuselage is already considered in the lift. The zero moment contribution
of the fuselage should always be combined with the corresponding contribution of the wing.
The fuselage causes a forward movement of the attack point of the lift force, so that
xLW F < xLW applies.
In the following equations, the wing and fuselage are always considered as one unit with the
index "WF":

Cm0W F = Cm0W + Cm0R

xcg − xLW F
CmW F = Cm0W F + (CLα )W F αW . (3.8)
lµ

The contribution of the horizontal tail with the effective distance of the horizontal tail lift
to the center of gravity of the airplane rH is :
3.1 Stationary Longitudinal Motion 29

" #
≈0
(
((
MH = −rH LH cos(αW − αD ) + (
W( sin(α
H( ((W − αD )
((

 
(
+zH LH(
sin(α
((W (−
((αD ) − 
(
W cos(α
H W − αD ) + MH0
≈0 ≈0

Again, small angles are assumed. If the drag contribution of the horizontal tail as well as
the influence of the elevation of the horizontal tail are neglected, and MH0 = 0 (symmetrical
horizontal tail profile), we get the following equation using CmH = Sl mH
µq
:

LH rH SH rH q H
Cm H ∼
=− = −(CLα )H (αW − αD + ϵ∗H ). (3.9)
Slµ q Slµ q

SH rH q H
Abbreviated notaion: VH = "Horizontal-tail volume ratio"
Slµ q

The moment coefficient of the total aircraft is calculated using Equation 3.8 and Equation 3.9:

Cm = CmW F + CmH or
 
xcg − xLW F ∂αD
Cm = Cm0W F + (CLα )W F αW − (CLα )H VH αW (1 − ) + ϵH − αD0 . (3.10)
∗
lµ ∂α

Figure 3.4 illustrates the contributions of the wing-fuselage combination and horizontal tail
to the total pitching moment of the aircraft.
The slope of the pitching moment Cmα = ∂Cm
∂αW
is calculated using Equation 3.10:

xcg − xLW F ∂αD

Cmα = (CLα )W F − (CLα )H VH (1 − ). (3.11)
lµ ∂α

The pitching moment coefficient cm0 of the whole aircraft is defined as the zero pitching
moment coefficient, for which the total lift is equal to zero (hereby applies: αW = (αW )CL =0 ).
If the relation αW = (αW )CL =0 from Equation 3.4 is inserted into Equation 3.10 then the
following equation applies:

(CLα )H SH q H ∗
Cm0 = (Cm0 )W F − (CLα )H VH (ϵ∗H − αD0 ) + (CLα )W F (ϵH − αD0 )
CLα Sq
  (3.12)
xcg − xLW F (CLα )H ∂αD
− + VH · (1 − ) .
lµ (CLα )W F ∂α

While the wing-fuselage neutral point xLW F and the location of the horizontal tail neutral
point are fixed points on the aircraft in the context of linearization, rH is dependent on the
30 3 Static Stability

Lα D0

D
Lα WF

CG LWF
Lα

Figure 3.4: Pitching moment contributions

position of center of gravity xcg according to Figure 3.3. To achieve a total pitching moment
equalling to zero and independent of the position of center of gravity, the horizontal tail
distance is redefined as the distance to the previously mentioned fixed points (Figure 3.5).
Using

rH∗
=rH + xcg − xLW F resp.
∗
SH q H xcg − xLW F SH q H rH
∗
VH = VH + = (cf. Figure 3.5) (3.13)
Sq lµ Sqlµ

we receive for the zero pitching moment coefficient cm0 :

3.1 Stationary Longitudinal Motion 31

SH q H xcg − xLW F ∗
Cm0 = (Cm0 )W F − (CLα )H VH∗ (ϵ∗H − αD0 ) + (CLα )H (ϵH − αD0 )
Sq lµ
(CLα )H SH q H xcg − xLW F ∗
− (CLα )W F · (ϵH − αD0 )
CLα Sq lµ
(CLα )H 2 SH q H ∗ ∂αD (CLα )H 2 SH q H 2 ∗
(ϵH − αD0 )VH∗ (1 − )− ( ) (ϵH − αD0 )
CLα Sq ∂α CLα Sq
xcg − xLW F ∂αD
· (1 − )
lµ ∂α
n ∂αD (CLα )H SH q H
= (Cm0 )W F − (ϵ∗H − αD0 ) · (CLα )H VH∗ − (CLα )H VH∗ (1 − )
∂α CLα Sq
 
xcg−
SH q H   xLW F

SH q H xcg−
xLW F

− (CLα )H  + (CLα )H 
 Sq lµ  Sq lµ
 
(CLα )W F + (CLα )H SHSqqH (1 − ∂αD
∂α
) o
·
CLα
| {z }
=CLα

CG
S.P.

xxAFR
LWF rH*
xxcg
S rH

Figure 3.5: Visual definition of rH

∗

The pitching moment coefficient independent of xcg is:

h i
(CLα )H SH q H
Cm0 = (Cm0 )W F − (CLα )H VH∗ 1 − C Lα Sq
(1 − ∂αD
∂α
) (ϵ∗H − αD0 ) (3.14)

 
∗ (CLα )H SH q H ∂αD
Cm0 = (Cm0 )W F − (CLα )H (ϵ∗H ∗
− αD0 ) VH − VH (1 − ) (3.15)
CLα Sq ∂α
32 3 Static Stability

If the new horizontal tail distance is also inserted into Equation 3.11 then the following
equation applies:

 
xcg − xLW F (CLα )H ∗ ∂αD (CLα )H SH q H xcg − xLW F ∂αD
Cmα = (CLα )W F − V (1 − )+ · (1 − )
lµ (CLα )W F H ∂α (CLα )W F Sq lµ ∂α
 
(CLα )H SH q H ∂αD xcg − xLW F ∂αD
= (CLα )W F 1 + (1 − ) − (CLα )H VH∗ (1 − ).
(CLα )W F Sq ∂α lµ ∂α
| {z }
(cf. Equation 3.3)=CLα

Therefore:

h i
xcg −xLW F (CLα )H ∗
Cmα = +CLα lµ
− CLα
VH (1 − ∂αD
∂α
) . (3.16)

3.1.3 Neutral Point

The position of the center of gravity where Cmα = 0 is referred to as the neutral point
(xcg )Cmα =0 = xnp .
According to Equation 3.16 the following definition is valid:

xnp − xLW F (CLα )H ∗ ∂αD

= VH (1 − ). (3.17)
lµ CLα ∂α

For the change of the moment by the lift coefficient in a linearized consideration, we receive
from Equation 3.16 and Equation 3.17:

∂Cm Cmα xcg − xLW F xnp − xLW F

= = − or
∂CL CLα lµ lµ
∂Cm xnp − xcg ∆xnp
=− =− (3.18)
∂CL lµ lµ

∆xnp
where lµ
is often used as a measurement for the angle of attack stability (static margin).

To satisfy static longitudinal stability (or angle of attack stability), ∂C

∂CL
m
must be negative,
i.e., an increase in lift must result in a negative pitching moment (nose-down) by itself
without pilot action.
To satisfy this condition, ∆xnp must be positive according to Equation 3.18, i.e., the neutral
point must be located behind the center of gravity. The definition of the point xnp by
(xcg )Cmα =0 means physically that the change in lift ∆CL = CLα ∆α caused by a change in
angle of attack ∆α about this point does not result in a change of moment. This consideration
is valid for the whole linear range of Cm (α). It can also be applied in the case of nonlinear
dependence of Cm (α), if only sufficiently small sections are considered.
3.1 Stationary Longitudinal Motion 33

A further simplification of the Equations 3.14 and 3.16 is possible by introducing a new
horizontal tail distance, rHnp , related to the neutral point (see Figure 3.6).
It then applies:

∗
rHnp = rH − (xnp − xLW F ) resp.
SH q H xnp − xLW F SH q H rHnp
VHnp = VH∗ − ( )= . (3.19)
Sq lµ Sqlµ

rH*
rH

LAWF
FR CG
S.P. NP
N.P.

xxLWF
AFR

xxcg
S

xxnp
n
rH
n

Figure 3.6: Visual definition of rHnp

By inserting the Equation 3.19 in Equation 3.15

h (CLα )H ∂αD SH q H i
Cm0 = (Cm0 )W F − (CLα )H (ϵ∗H − αD0 ) VH∗ − VH∗ (1 − ) , (3.20)
CLα ∂α Sq
| {z }
xnp −xL
from Equation 3.17: WF
lµ

follows

Cm0 = (Cm0 )W F − (CLα )H (ϵ∗H − αD0 )VHnp . (3.21)

Accordingly from Equation 3.18:

Cmα = −CLα xnpl−x

µ
cg
. (3.22)

The total pitching moment of the aircraft is then

34 3 Static Stability

Cm = Cm0 + Cmα α
∂Cm
Cm = Cm0 + CL . (3.23)
∂CL

3.1.4 Determination of the required Tail Volume

For the design of the horizontal tail for naturally stable aircraft the following two criteria
must be fulfilled:

• Stability Limit: The most aft center of gravity position is determined by the requirement
for a minimum degree of static stability
• Controllability Limit: The most forward center of gravity position is determined by
the control of the pitching moments in the worst case scenarios (e.g., landing approach
with extended flaps or nose wheel lift-off conditions)

Stability Limit

The stability limit is reached, if the center of gravity is located at the most aft position
without loosing static longitudinal stability. If the center of gravity is located exactly at the
neutral point, the aircraft is neutrally stable.
Using Equation 3.11 it applies for xcg = xnp :

xnp − xLW F ∂αD q SH rH

Cmα = 0 = (CLα )W F − (1 − )(CLα )H H .
lµ ∂α qSlµ

If we are using the following definition instead of the lever arm dependent on the center of
gravity position

∗
rH = rH − (xcg − xLW F )

and also considering xcg = xnp , then the following horizontal tail surface area ratio (horizontal-
tail size) required in terms of stability is given:

(CLα )W F xnp −xLW F

SH (CLα )H ∗
rH
= D qH
· xnp −xLW F . (3.24)
S (1 − ∂α
∂α
) q 1− ∗
rH

Controllability Limit

The horizontal-tail size must ensure that sufficient control moments are available throughout
the flight envelope to generate moment equilibrium
P
M = 0.
3.1 Stationary Longitudinal Motion 35

The moment equilibrium from Equation 3.10 can be written as

xcg − xLW F q SH rH
Cm = (Cm0 )W F + CLW F · − CLH H = 0.
lµ q S lµ

Replacing rH again by the center of gravity independent quantity rH

∗
and resolving to SH
S
,
it follows

m0 W F (C )
SH CLW F q xcg − xLW F + lµ CLW F
= ∗
.
S Cm =0 CLH q H rH − (xcg − xLW F )

To generate control moments, the horizontal tail can provide both positive and negative lift,
operating within the following range:

(CLH )min ≤ CLH ≤ (CLH )max .

The extreme value (CLH )min is decisive for the pitch control of the most forward center of
gravity position. The critical case is when the high-lift configuration is activated (extended
slats and flaps). Then, the attainable CLW F value is the largest and the zero pitching moment
of the wing-fuselage combination reaches the largest negative value counteracting against the
control pitching moment of the horizontal tail. Thus, the horizontal tail size for the most
forward center of gravity position is:

xSV −xLW F lµ (Cm0 )W F

SH (CLW F )max q ∗
rH
+ ∗ (C
rH LW F )max
=+ xSV −xLW F (3.25)
S Cm =0 (CLH )min q H 1− ∗
rH

An evaluation of the Equation 3.24 and Equation 3.25 is depicted in Figure 3.7. It can be
seen that the requirement for a certain usable center of gravity area of the size ∆x ∗
rH
cg
defines
the minimum size of the horizontal tail. In the case considered above, the controllability
limit was defined for the landing approach. Other cases are conceivable, such as maneuvering
flight or the rotation while takeoff (nose wheel lift-off), which usually result in different limit
curves than those shown above. The most unfavorable case is always decisive for determining
the horizontal tail size.
Numerical Values:

Stability Limit: Controllability Limit:

(CLα )W F lµ (Cm0 )W F
(CLα )H
= 1, 4 ∗ (C
rH LW F )max
= −0, 1

∂αD (CLW F )max

∂α
= 0, 25 (CLH )min
= −2, 0
36 3 Static Stability

Figure 3.7: Usable center of gravity range

3.1.5 Moment Equilibrium

For an aircraft with aerodynamic control the required elevator deflection angle at which
the pitching moments about the aircraft center of gravity are balanced is obtained from
Equation 3.23 as:

Cm = (Cm0 )η=0 + (Cmα )η · α∗ + (Cmη )α · ηCm =0 = 0 (3.26)

The fulfillment of the lift condition requires

G
CL = (CLα )η · α∗ + (CLη )α · ηCm =0 = with α∗ = α|η=0 (3.27)
Sq

From Equation 3.27 follows, with (Cmα )η = Cmα , (Cmη )α = Cmη , (CLα )η = CLα and (CLη )α =
CLη :
3.1 Stationary Longitudinal Motion 37

CL CLη SH q H ∂αH SH q H
α∗ = − ηC =0 with CLη = (CLη )H = (CLα )H .
CLα CLα m Sq ∂η Sq

Implemented in Equation 3.26 results in

CL CLη
(Cm0 )η=0 + Cmα · + ηCm =0 · [Cmη − Cmα · ]=0
CLα CLα

and as a result the required elevator deflection angle:

(Cm0 )η=0 · CLα + Cmα · CL

ηCm =0 = − . (3.28)
Cmη · CLα − Cmα · CLη

For aircrafts with empennage at the tail, the lift contribution of the horizontal tail is typically
Cmη
very small, so CLη ≪ Cm CLα can be substituted. Equation 3.28 is then rearranged using
α

CL = 2G
SρV 2
and Cmα
CLα
= − xnpl−x
µ
cg
to:

(Cm0 )η=0 xnp − xcg 2G

ηCm =0 ∼ =− + . (3.29)
Cm η lµ Cmη SρV 2
 
The derivative Cmη = ∂C m
∂η
is calculated from the following equation mH = −(CLη )H ·
α
ηSH · rH · q H = Cmη · η · S · lµ · q to

∂αH rH
Cmη = −(CLη )H VH = −(CLα )H VH = −CLη
∂η lµ

In Figure 3.8 the dependability ηCm =0 = f (V ) (3.29) is evaluated for a given aircraft as
an example. It can be seen that increasing the speed, the elevator must be deflected in a
positive direction, which means "pushing" the control stick.

!Cm=0
xxnnp==xxscg
2°
(Cm0)h=0 +!
”

C m! Stabilizer Rud er
Elevator
“

0° V [m/s]
50 100 150
”

Δxnnp/ /l#lµ==
"x
-2°
$H = 0
“

0,05
0,10 xn>>xxcgs
xnp

0,15
-4°

Figure 3.8: Elevator deflection angle in relation to the stationary velocity

38 3 Static Stability

The change in velocity can also be achieved while the stick is held fixed (i.e. fixed elevator
deflection angle) by a movement of the center of gravity, which can be represented by a
change in the static margin δxlµnp with unchanged aerodynamic quantities.

The curves ηCm =0 = f (V ) from Figure 3.8 are valid for constant derivatives, assumed to be
independent of the Mach number. In the case of compressible flows, stronger changes in the
derivatives may occur when higher Mach numbers are reached, resulting in strong deviations
of the curves in Figure 3.8. Such deviations can result from local exceeding of the critical
Mach number (M a = 1) at the wing, wing-fuselage transition, or empennage. Occasionally,
interference phenomena also have a major influence, for example occurring between the wing
and engine nacelles.

hc
m=0
A Change in
compressible range

B V

M= 0,8

Figure 3.9: Change in required elevator angle in compressible flow

Figure 3.9 depicts the deviation of the elevator angle curve ηCm =0 as an example. When
crossing point A, the aircraft will automatically increase its airspeed without pilot action
(i.e. η = const.) in case of a disturbances in the longitudinal direction (e.g. caused by
a longitudinal gust), while simultaneously reducing the air-path inclination angle. This
behavior is unstable with respect to the velocity. This effect, also referred to as tuck-under,
must not occur in civil aircraft within the permitted speed range.
Moment equilibrium can also be achieved by setting the horizontal tail’s horizontal stabilizer.
All larger aircraft have devices to change the stabilizer angle (setting angle) usually by means
of electric actuators. The required stabilizer angle results analogously to Equation 3.29 to:

(Cm0 )η=const., ϵH =0 xnp − xcg 2G

(ϵH )Cm =0 = − + (3.30)
CmϵH lµ CmϵH SρV 2

Figure 3.10 depicts the curve (ϵH )Cm =0 = f (V, η). Stabilizer and elevator of the horizontal
tail complement each other. If a moment equilibrium should be achieved for a given velocity,
different combinations of the setting angle ϵH and the elevator angle η are possible. In the
previous example ( Figure 3.10) the following possibilities yield for V = 150 m/s:
η −2◦ −1◦ 0◦ +1◦ +2◦
ϵH +1, 75◦ +1, 2◦ +0, 65◦ +0, 1◦ −0, 45◦
Thus, instead of establishing the moment equilibrium by setting the elevator angle using the
control stick, it is also possible to achieve moment equilibrium by adjusting the setting angle
of the stabilizer. This process is also called trimming. The associated actuator is the trim
wheel.
3.1 Stationary Longitudinal Motion 39

2°
h = -2
(eH)Cm=0
+ eH
-1
1° +h
0

+1
0° V [m/s]
50 100 150

+2

-1°

-2°

Figure 3.10: Trimming by changing the setting angle εH of the horizontal stabilizer and the
elevator deflection angle η

3.1.6 Fixed and free Elevator

For the definition of the neutral point (subsection 3.1.3) it was assumed, that the pilot holds
the stick in its fixed position (fixed elevator ). When the pilot releases the stick, the elevator
then would rotate about its hinge by the influence of the aerodynamic forces until the the
moment about the hinge becomes zero (free elevator ).
If it is assumed that the elevator has a weight balance, so that no mass forces are effective
at the elevator, then the moment about the elevator hinge (elevator hinge moment Mh,e ) is

∂Mh,e ∂Mh,e ∂Mh,e

Mh,e = · αH + ·η · ηa = 0, (3.31)
∂αH ∂η ∂ηa

with Mh,e = Ch,e · Se le q H .

In addition to the elevator, the horizontal tail usually has an auxiliary trim tab, as shown
in Figure 3.11.
40 3 Static Stability

le a
Mh,e

αH
elevator hinge le

lH

Figure 3.11: horizontal tail with auxiliary elevator for hand force reduction

Let this auxiliary trim tab be adjustable with respect to the elevator. The elevator angle
ηe for which the elevator hinge moment becomes zero for a given angle of attack αH and
auxiliary trim tab angle ηa is called float angle. According to Equation 3.31 it is

∂me ∂me
∂αH
· αH + ∂ηa
· ηa Ceα · αH + Ceηa · αH + Ceηa · ηa
ηe = − ∂me
=− (3.32)
∂η
Ceη

The free elevator aligns with the airflow direction, i.e., it rotates similar to a weathercock.
This results in a reduction of the horizontal stabilizer’s lift.

∂LH ∂LH ∂LH

LHf ree = · αH + ·η+ · ηa .
∂αH ∂η ∂ηa

If we neglect the typically small lift component of the trim tab, then after introducing the
dimensionless coefficients and η = ηe from Equation 3.32 it is

Ceα · αH + Ceηa · ηa
CLHf ree = (CLα )H · αH − (CLη )H (3.33)
Ceη

For elevators without trim tab deflections Equation 3.33 simplifies correspondingly and the
ratio of the lift slope with free elevator to the lift slope with fixed elevator (η = 0) is

(CLα )Hf ree (CLη )H Ceα ∂αH Ceα

=1− =1− = δf ree (3.34)
(CLα )H (CLα )H Ceη ∂η Ceη

Thus all essential equations of the sections 3.1.2 to 3.1.3 can be adapted to the case with free
elevator by introducing the factor δf ree corresponding to Equation 3.34. Therefore, instead
of Equation 3.3 the following equation is valid:

(CLα )H SH q H ∂αD
(CLα )l = (CLα )W F [1 + δl (1 − )],
(CLα )W F Sq ∂α

and instead of Equation 3.16:

3.1 Stationary Longitudinal Motion 41

xcg − xLW F (CLα )H ∗ ∂αD

(Cmα )f ree = (CLα )f ree [ − δf ree VH (1 − )]
lµ (CLα )f ree ∂α

CLα
and with VHnpl = (CLα )f ree
· VHnp Equation 3.21 changes to:

(Cm0 )f ree = (Cm0 )W F − δf ree (CLα )H VHnpl (ϵH − αD0 ).

The condition (Cmα )f ree = 0 results with xcg = xnpf ree to:

xnpf ree xL (CLα )H ∗ ∂αD

= W F + δf ree VH (1 − ),
lµ lµ (CLα )f ree ∂α

xnp
whereas lfµree is defined as the neutral point with free elevator. This point is positioned
before the neutral point with fixed elevator. Thus, releasing the elevator has a destabilizing
effect on the static longitudinal stability.

3.1.7 Static Stability during pull-up maneuver

The term pull-up refers to a flight maneuver in which an increase in angle of attack is forced
by pulling the control stick at the end of a descending flight (flight with a negative air-
path inclination angle γ). Hereby the lift increases, thus, a positively directed acceleration
component acting perpendicular to the flight path appears. This leads in a curvature of the
flight path.
The increase in lift-induced drag causes a reduction in airspeed. The following additional
moments and forces appear during pull-up, which are to be superimposed on the values of
horizontal flight:

∂M ∂M ∂M
∆M = · ∆α + · ∆η + ·q (3.35)
∂α ∂η ∂q
 
∂L ∂L ∂L
∆L = · ∆α + · ∆η + ·q (3.36)
∂α ∂η ∂q

The contribution of the additional drag will be, as previously done, neglected in the following
equations. Introducing again dimensionless coefficients and denote

∂Cm rH
Cm q = lµ
= −(CLα )H · VH , q = pitch rate (3.37)
∂(q · ) lµ
V

and with Equation 3.36 the required change in elevator deflection (thus ∆Cm = 0) for the
pull-up maneuver is obtained:
42 3 Static Stability

Cmα · ∆α + Cmq lVµ q

∆η = − . (3.38)
Cmη

If an additional load factor is introduced

∆L ∆CL bz
∆nz = = =
G (CL )nz =1 g

thus follows for Equation 3.36

∆nz CLη
∆α = (CL )nz =1 − · ∆η (3.39)
CLα CLα

Using q = ∆nz g
V
and inserting Equation 3.39 into Equation 3.38, after some calculations the
result is

Cmα (CL )nz =1 + Cmq CLα lVµ2g

∆η = − · ∆nz (3.40)
Cmη CLα − Cmα CLη

Cmη
For aircrafts with the empennage at the tail the approximation CLη ≪ C
Cmα Lα
can be
used. Using Cmα from Equation 3.22 and µ = 2G
ρgSlµ
(µ = normalized mass) the result is:

 
∆η ∼ (CL )nz =1 xnp − xcg Cmq
= − (3.41)
∆nz Cm η lµ µ

The center of gravity position xcg where δη

δnz
= 0 is denoted as maneuver point with fixed
elevator xmp . Thus

 
xcg xmp xnp Cmq
= = − ,
lµ ∆η
=0 lµ lµ µ
∆nz

which leads to

∆η ∼ (CL )nz =1 xmp − xcg

= . (3.42)
∆nz Cm η lµ

For cases where the horizontal tail lift CLη cannot be neglected (e.g., tail-less aircraft),
instead of Equation 3.42 the exact equation must be taken:

∆η (CL )nz =1 xmp − xcg

= xnp −xcg (3.43)
∆nz Cmη + lµ CLη lµ
3.1 Stationary Longitudinal Motion 43

xnp
If the elevator is not fixed, instead of the neutral point with fixed elevator lµ
, the neutral
xnpf ree
point with free elevator lµ
must be used, which is defined as

xmpf ree xnpf ree Cmqf ree

= − . (3.44)
lµ lµ µ

Analogous to the neutral point, the maneuver point with free elevator is located in front of
the maneuver point with fixed elevator. This again means a destabilization of the longitudinal
motion.

3.1.8 Elevator Hand Forces with direct (aerodynamic) control

From the air forces acting on the elevator, the required manual hand force on the control
stick is obtained with the aid of the principal of linear momentum. It states that the sum
of all virtual work due to external forces and moments on the elevator-control system δWi
vanishes:

X
δWi = 0.
i

Hand/Stick paths and hand forces are to be counted positively, if the control stick is moved
clockwise around its bearing axis. With this sign definition, the sum of all virtual work due
to external forces and moments is (cf. Figure 3.12)

X
δWi = FH · δsH + Mh,e · δη = 0,
i

which must disappear for a small (virtual)

 movement δsH of the control stick
 and  small
deflections of the elevator δη = ∆s
∆η
H
· δsH (transmission ratio, index t: ∆s
∆η
H
< 0).
t t
Thus follows

 
∆η
FH + Mh,e =0
∆sH t

or with the elevator hinge moment Mh,e = Ch,e · Se le q H :

 
∆η
FH = − Ce Se le q H . (3.45)
∆sH t

The elevator hinge moment coefficient Ch,e is a function of the horizontal tail’s angle of
attack αH = αW − αW + ϵH and the elevator angle η and, if applicable, the auxiliary elevator
angle ηa :

Ch,e = Ch,eα αH + Ch,eη η + Ch,eηa ηa (3.46)

44 3 Static Stability

+sH Control stick hand force FH

Control stick path sH
( Δη/ΔsH) t <0 +Mh,e
+FH

ηa -η
εH
αH
Horizontal Stabilizer
Control stick

Figure 3.12: Prefix definition for elevator control

Stability Hand Force

Stability Hand Force is designated as the hand force, which is required in horizontal flight
to change the flight velocity.
With the elevator angle, required to achieve moment equilibrium according to Equation 3.28,
and the relation

∂αD CL
αH ∼
= (1 − ) (with ϵH = αW0 = 0) (3.47)
∂α CLα

we obtain from Equation 3.46 with ηa = 0:

   
Ce η Cm α Ce α Cm η ∂αD
Ce = − (Cm0 )η=0 + − (1 − ) CL
Cm η CLα Ceη CLα ∂α

Replacing Cmα by Equation 3.16, and further introducing Cmη = −(CLα )H VH ∂α

∂η
H
, and
replacing VH by VH from Equation 3.13, this leads to
∗

 
Ceα ∂αH (CLα )H ∂αD xcg − xLW F (CLα )H ∗ ∂αD
VH (1 − )+ − VH (1 − )
Ceη ∂η CLα ∂α lµ CLα ∂α
"
Ceα ∂αH (CLα )H ∗ ∂αD xcg − xLW F
= VH (1 − )+ −
Ceη ∂η CLα ∂α lµ
#
Ceα ∂αH (CLα )H SH q H ∂αD xcg − xLW F (CLα )H ∗ ∂αD
(1 − ) − V (1 − ) .
Ceη CLα CLα S q ∂α lµ CLα H ∂α

Due to δf ree from Equation 3.34 the expression in parenthesis rearranges to

3.1 Stationary Longitudinal Motion 45

"
Ce ∂αH (CLa lpha )H ∗ ∂αD xcg − xLW F
= − (1 − α ) VH (1 − )+ −
Ceη ∂η CL α ∂α lµ
| {z }
δf ree
#
(CLα )H SH q H ∂αD xcg − xLW F
(1 − δf ree ) · (1 − )
CLα S q ∂α lµ

After derivation in subsection 3.1.6 follows:

(CLα )H ∂αD xnpf ree − xLW F

· δf ree VH∗ (1 − )=
(CLα )f ree ∂α lµ

xcg −xAF R (CLα )f ree

and the last terms of the square brackets can be combined to lµ CLα
, we finally
obtain

xcg − xnpf ree (CLα )f ree

 
Ceη
Ce = − (Cm0 )η=0 + CL , (3.48)
Cmη lµ CLα

(CLα )f ree
where for an aircraft with empennage at the tail C Lα
= 1 can be set approximately.

Introducing CL = Sq G
and inserting the mentioned equation for Ch,e in Equation 3.45, the
result for the stability/trim force (Cm = 0) is:

ρ 2 G xcg − xnpf ree (CLα )f ree

   
∆η Ceη q H
FH = − Se le −(Cm0 )η=0 V − (3.49)
∆sH t Cmη q 2 S lµ CLα
| {z }
K1

Figure 3.13 illustrates the dependence FH = f (V ) from Equation 3.49.

Slope of the hand force

V0 (trim point Ch,e =

Figure 3.13: Stability Hand Force

46 3 Static Stability

The intersection of the hand force curve (Figure 3.13) with the y-axis is called the trim point.
The corresponding speed is the trim speed, which will be reached, if the pilot releases the
elevator (free elevator). According to Equation 3.49 the stability hand force depends on the
xcg −xnpf ree
static margin with free elevator lµ
. As can be seen in Figure 3.13 the trim speed
can be influenced by changing the zero pitching moment (e.g. by adjusting the horizontal
stabilizer) as well as by changing the static margin with free elevator (e.g. due to a center
of gravity shift) (cf. subsection 3.1.9).
For the trim speed VFH=0 = V0 we obtain from Equation 3.49:

s
2G xnpf ree − xcg (CLα )f ree
V0 = . (3.50)
ρS(Cm0 )η=0 lµ CLα

The slope of the hand force is

dFH
= −K1 · (Cm0 )η=0 · ρ · V (3.51)
dV

2G xnpf ree −xcg (CLα )f ree 1

which leads to the following equation for the trim point using (Cm0 )η=0 ρV0 = S lµ CLα VT
:

G xnpf ree − xcg 1 (CLα )f ree

 
dF
= −2K1 . (3.52)
dV V0 S lµ V 0 CL α

It can be seen that the slope of the hand force increases with Se le ∼ L3 , that it is proportional
to the wing loading and static margin with free elevator, and it is inversely proportional to
the trim speed. It is not depending on the air density and flight altitude. For higher flight
velocities the dependence of the aerodynamic derivatives and the neutral point position
xnpf ree − xcg on the Mach number must be considered:

h i
(CLα )f ree

dFH
 
∆η

Ceη (M ) G (xnpf ree − xcg )(M ) CLα
(M ) q
H
=2 Se le .
dV V0 ∆sH t Cmη (M ) S lµ amT r q

Pull-Up Hand Force

In order to fly a pull-up maneuver with a certain radius rA or a certain load factor, an
additional elevator angle and therefore an additional hand force is required. For the calculation
we start, again, with the hinge moment coefficient Ch in general. Since the pull-up maneuver
is a pitching motion about the lateral axis, the hinge moment refers to the elevator, thus
Ch = Ch,e applies for this section.

∆ChL = Chα ∆αH + Chη ∆η. (3.53)

In this ∆η can be taken directly from Equation 3.40 or 3.41. For the determination of ∆αH
3.1 Stationary Longitudinal Motion 47

it has to be considered that due to the curved flight path, an additional angle of attack
αHdyn = q·rVH = rrHA at the horizontal tail appears (cf. Figure 3.14):

∂αD q · rH
∆αH = ∆α(1 − )+ (3.54)
∂α V
where ∆α can be taken from Equation 3.39 (neglecting the second lift contribution caused
by the horizontal tail).

rH
!Hdyn =
rA

rA

rH

Tr a
jec
tor
y

Figure 3.14: Additional angle of attack during a pull-up maneuver

With the pitch rate q = g

V
∆nz then the horizontal tail’s angle of attack is

" #
1 − ∂α D
gr H
∆αH = ∂α
+ CLnz =1 ∆nz . (3.55)
CLα CLnz =1 V 2

If we add grH
CLn =1 V 2
= rH
µlµ
, we obtain from Equation 3.53
z

" ! #
1 − ∂α

Chη Chα D
rH xcg − xnp Cmq
∆ChL = Cm η ∂α
+ − + · CLnz =1 · ∆nz (3.56)
Cm η Chη CLα µlµ lµ µ

To simplify this equation, we introduce the following relationships derived earlier:

∂αH Ch ∂αH
Cmη = −(CLα )H VH ; 1 − α = δf ree ;
∂η Chη ∂η
xcg − xnp xnp − xLW F xL − xcg xL − xcg (CLα )H ∗ ∂αD
− = + WF = WF + VH (1 − );
lµ lµ lµ lµ CLα ∂α

furthermore applies according to subsection 3.1.6:

xnpf ree − xLW F ∂αD (CLα )H

= +VH∗ · δf ree (1 − ) .
lµ ∂α (CLα )f ree
48 3 Static Stability

This results to:

"
∆ChL Chη Chα ∂αH (CLα )H ∗ ∂αD Ch ∂αH (CLα )H SH q H
= − · VH (1 − )+ α ·
CLnz =1 · ∆nz Cmη Chη ∂η CLα ∂α Chη ∂η CLα S q
∂αD xcg − xLW F Ch ∂αH (CLα )H VH rH
· (1 − ) − α −
∂α lµ Chη ∂η µlµ
#
xcg − xLW F (CLα )H ∗ ∂αD Cmq
+ V (1 − )−
lµ CLα H ∂α µ
( " #
Chη (CLα )H ∗ ∂αD xcg − xLW F SH q H ∂αD
= δf ree V (1− )− · CLα − (1 − δf ree )(CLα )H (1 − )
Cmη CLα H ∂α CLα · lµ S q ∂α
| {z }
=(CLα )f ree
)
(CLα )H VH rH Cm q
− (1 − δf ree ) −
µlµ µ
" #
Chη xnpf ree − xLW F (CLα )f ree xcg − xLW F (CLα )f ree VH rH Cmq
= − −(1−δf ree )(CLα )H − .
Cmη lµ CLα lµ CLα µlµ µ

Thus, the additional hinge moment coefficient for a pull-up maneuver is:

Chη xnpf ree − xcg (CLα )f ree

 
VH rH Cm q
∆ChL = − (1 − δf ree )(CLα )H − CL · ∆nz . (3.57)
Cmη lµ CLα µlµ µ

Inserted into Equation 3.45 and using (Cmq )f ree = −δf ree (CLα )H · VH rH
lµ
and Equation 3.37
the result for the pull-up hand force is

"  #
G xnpf ree − xcg (CLα )f ree

Cmq
∆FHpull−up = K1 − · ∆nz , (3.58)
S lµ CLα µ f ree
 
∆η Ch q
with K1 = − Se le η H
∆sH t Cmη q

∆FHpull−up is proportional to the load factor during pull-up, to the wing loading and, assuming
velocity-independent derivatives, independent of the airspeed.

3.1.9 Trimming

In subsection 3.1.8 the trim speed was defined as the airspeed in a stationary flight condition
with free elevator (FH = 0).
This trim speed can be changed in various ways: Either directly by operating a trim element
like the trim wheel, or by operating the control stick using hand force and then using the
trim element to eliminate the hand force until it becomes zero.
Trim components that can be considered are:
3.1 Stationary Longitudinal Motion 49

A Trim tab, which is hinged to the elevator and does not mechanically affect the elevator
when actuated. Adjusting the trim tab changes the float angle of the elevator and thus
the horizontal tail’s lift. This results in a change in trim speed.
B Adjustable horizontal tail A separate actuator (usually electromechanical) can be used
to change the setting angle of the entire tail stabilizer, which also leads to a change in
trim speed.
C Every change in the center of gravity position causes a change in trim speed. This
trim option, which was common in the first airplanes and is also common today among
hang gliders, is not used in conventional airplanes.

Trim Tab

Trim tabs are located at the trailing edge of the elevator and the rudder. They are rotated
by a drive lever that intersects the elevator in the elevator axis, allowing the trim tabs to be
operated without mechanical interference with the elevator.

top view
elevator
horizontal
stabilizer
br
trim tab

A bt B

fuselage

cross section A-B Mh,e

+ ηt -η

elevator hinge

Figure 3.15: Trim tab

The moments about the hinges of the elevator and rudder are generally denoted as hinge
moments Mh . Therefore the moments about the elevator hinge is Mh,e = Ch,e · Se le · q e with
the elevator’s area Se and depth le . In trimmed condition FH = 0 must be valid:

Ch,e = Ch,eα αH + Ch,eη η + Ch,eηt · ηt = 0

with the deflection angle ηt of the trim tab with respect to the elevator. From this the
floating angle is obtained as

Ch,eα αH + Ch,eηt ηt
ηCh,e =0 = − . (3.59)
Ch,eη
50 3 Static Stability

Neglecting elevator and trim tab lift, the elevator deflection required for moment compensation
accordingly to Equation 3.29 is

(Cm0 )η=0 xnp − xcg CL

ηCm =0 = − + .
Cmη lµ Cm η

By equating both relations, after some intermediate calculation, using the relation δf ree =
C α
1 − ∂α
∂η
H
· Ch,e
h,eη
( Equation 3.34) we get

Ch,eη xnpf ree − xcg (CLα )f ree

 
ηt = (Cm0 )η=0 − CLT r . (3.60)
Ch,eηt · Cmη lµ CLα

Using CLT r = SρV

2G
0
2 the trim tab deflection angle for a given trim speed can be calculated

from Equation 3.60. The trim tab depth is usually 20 to 25 % of the elevator depth. The
span fraction of the trim tab bbrt represents an important quantity for adjusting the trim
effect.

Stabilizer Trim

In subsection 3.1.5 it was shown that an adjustable horizontal stabilizer, which can be
set separately by a trim wheel, is suitable for changing the moment equilibrium without
operating the control stick. The elevator angle for a vanishing elevator hinge moment is
according to Equation 3.32 with ηa = 0:

Ch,eα · αH
ηCh,e =0 = − (3.61)
Ch,eη

The moment equilibrium is approximately defined by the following relation (neglecting

(CLα )H SH q H (1 − ∂α
∂α
D
) ≪ CLα Sq):

∂αH
(Cm0 )W F − (CLα )H VH∗ ϵH − (CLα )H · · VH∗ · η+
∂η
 
xcg − xLW F (CLα )H ∗ ∂αD
− V (1 − ) · CL = 0
lµ CLα H ∂α

If we equate the elevator angle from both relationships, the determining equation for the
(C )
vertical stabilizer trim angle is a function of the trim speed and with the approach LCαL f ree ≈
α
1 this leads to

xnpf ree − xcg

 
1 2G
(ϵH )T r = (Cm0 )W F − · . (3.62)
(CLα )H VH∗ lµ SρV0 2
3.1 Stationary Longitudinal Motion 51

3.1.10 Control Aids

Using aerodynamic controls, the control surface hinge moments and thus the hand forces
increase with the third power of the length dimensions, thus a reduction of the hinge moments
is necessary for larger aircraft. Auxiliary rudders are used as aerodynamic measure to reduce
the hinge moment:

a) Path-controlled auxiliary elevator

ηa
η
= const.
max. reduction ≈ 50%
b) Servo tabs (Flettner tabs) with spring force centering
max. reduction ≈ 90%
(i.e. to 10
1
).

+ηa

-η

Drive
Main elevator
Servo tabs

Drive lever without Spring with preload

connection to main
elevator

Figure 3.16: Path-controlled auxiliary elevator (top) and servo tabs (i.e. Flettner tabs)
(bottom)

These measures, especially the spring-force control shown in b), make it possible to achieve
permissibly small input forces even for aircraft up to about 40 to 50 t flying weight.

3.1.11 Summary of the Longitudinal Motion Derivatives

A derivative is the partial derivation of the component of a force or moment or its coefficients
according to one of its influencing variables (definition according to aviation standard).
52 3 Static Stability

Table 3.1: Compilation of longitudinal moment derivatives

Lift Drag Pitching moment

α CLα CDα = 2·k ·CLα ·CL (i.e., Cmα
cf. Equation 3.3 CD = CD0 + kCL 2 ) cf. Equation 3.16
q CLq = ∂Cq·lµ =
L
CDq = ∂C D
q·lµ Cmq = ∂C m
q·lµ
∂ V
∂ V
∂ V

(CLq )H = (CLα )H · VH (CDq )H = (CDα )H · VH (Cmq )H = −(CLα )H VH rlHµ

η CLη = CDη = Cmη =
H SH q H ∂αH SH q H
= (CLα )H · ∂α
∂η S q
= (CDα )H · ∂η S q
= −(CLα )H · ∂αH
∂η
· VH
3.2 Stationary Lateral Motion 53

3.2 Stationary Lateral Motion

3.2.1 Definitions

In lateral motion, the following quantities from the six degrees of freedom are considered:

a) Lateral velocity v or the sideslip angle β,

b) Rotational velocity p or the bank angle Φ,
c) Rotational velocity r or the azimuth Ψ.

x
y
u v
p q

Figure 3.17: Degrees of freedom of longitudinal and lateral motion

Components of
Axis
Translational velocity Rotational velocity
x u (Vx ) p (Ωx )
y v (Vy ) q (Ωy )
z w (Vz ) r (Ωz )

For the variables u and w of the longitudinal motion constant values are considered, thus,
the lateral motion is studied at constant airspeed, constant lift and constant drag.
The rotational motion about the longitudinal axis x is referred to as rolling, the rotational
motion about the lateral axis y as pitching and the rotational motion around the vertical
axis z as yawing. An angular deviation about the x-axis is called banking and an angular
deviation about the y-axis is called slipping.
Depending on which axis system the consideration is based on, the bank angle is denoted
differently (see Figure 3.18). Measured between body-fixed and geodetic system, it is called
bank angle Φ, which is positive when the right wing is below the horizontal plane. Between
aerodynamic and geodetic system it is called air-path bank angle µα .
54 3 Static Stability

Za

Figure 3.18: Body-fixed or aerodynamic vs. geodetic coordinate system

In the lateral motion the angle of attack α is regarded constant. Therefore, the following
relationships between the velocity components, the velocity vector, and the angles α and β
are considered:

v
sin β = , (3.63)
V
w
sin α = . (3.64)
V cos β

x x
xe
u xa
α x a,y α
β
β v
x,z

v
w
y
z
β
y-ye
x-z-Plane = Symmetrical plane α
ya

z
Figure 2.2.3 zo-ze

Figure 3.19: Components of the airspeed

For Equation 3.64, α must be considered as a fixed initial value, which is usually defined for
β = 0.
The sideslip angle β is defined as the angle between the inflow axis xa and the aircraft
symmetry plane x-z (cf. Figure 3.19). Per definition, β is positive for an inflow coming from
starboard (blowing in the right cockpit window).
3.2 Stationary Lateral Motion 55

The relationship between the aerodynamic and the body-fixed system is given by a transformation
consisting of two rotations about the angle of attack α and the subsequent rotation about
the sideslip angle β:
     
xa cos β sin β 0 cos α 0 sin α x
 ya  = − sin β cos β 0  0 1 0  y  (3.65)
za 0 0 1 − sin α 0 cos α z

In the following, aerodynamic forces and moments and the corresponding coefficients and
derivatives are considered in the experimental coordinate system. If not specifically noted,
the angle of attack is assumed to be small, so that cos α = 1 can be considered.

3.2.2 Yawing Motion (Single Degree of Freedom)

3.2.2.1 Definitions

A simplified view of the yawing motion is obtained, if only rotations of the airplane by
the angle Ψ about the zg axis are allowed in horizontal flight (single degree of freedom
consideration). Thus, β and Ψ (or β̇ and Ψ̇) are equal in magnitude to each other (β = −Ψ).
The primary control element of the yawing motion is the rudder, whose deflection angle ζ is
positive when it generates a negative yawing moment (the trailing edge of the rudder must
be moved in the negative y-direction).
The coefficients of the yaw motion are

Y
CY = Side force coefficient,
S·q
N
Cn = Yaw moment coefficient.
S·s·q

The partial derivatives of forces, moments or their coefficients according to their influence
quantities (e.g., Ψ = r, β and ζ) are called derivatives. For the classification the influence
variable which served as the basis for the derivative is added to the coefficient as index.

3.2.2.2 Side Force Derivatives

Side Force due to Sideslip Angle Yβ = CYβ Sq

If an aircraft is flying with a sideslip angle β, a side force is created to which all aircraft
components contribute - albeit in different magnitudes. The local sideslip angle at the
vertical tail is denoted as βV .
The main contributions of the side force are:
56 3 Static Stability

2
Fuselage contribution (Yβ )F = −0, 2VF 3 q (3.66a)
Wing contribution (Yβ )W = −CDp Sq (3.66b)
∂βV
Vertical tail contribution (Yβ )V = (CYβ )V SV q V (3.66c)
∂β

The equations 3.66a to 3.66c are estimates of the contributions of the components, as they
are often made in the preliminary design of aircraft. Due to the definition of the sideslip
angle according DIN9300, all side force derivatives given above have a negative value (e.g.,
(CYβ )V < 0).

Side Force due to Rudder Deflection Yζ = CYζ Sq

A positive rudder deflection ζ creates a positive side force in y-direction. The derivative
with respect to the wing area S is:

∂βV SV q V
CYζ = −(CYβ )V (3.67)
∂ζ S q

Side Force due to Yaw Rate Yr = CYr Sq with Yr = ∂Y

∂ ( r·s
V )

The effective additional sideslip angle at the vertical tail due to a rotational yaw velocity
Ψ̇ = r about the aircraft’s vertical axis is

r · rV
∆βV = −
V0

where rV represents the distance between the neutral point of vertical tail and the aircraft’s
center of gravity. The contribution of the vertical tail in the side force derivative due to a
yaw rate is thus

(CYr )V = −(CYβ )V VV (3.68)

SV ·rV ·q V
with VV = S·s·q
(Vertical-tail volume ratio) and s = half span.

3.2.2.3 Yaw Moment Derivatives

Yaw Moment due to Sideslip Angle / Yaw Stiffness Nβ = Cnβ Ssq

Analogous to the angle of attack stability of the pitching motion, the directional or weathercock
stability is defined as
3.2 Stationary Lateral Motion 57

∂Cn
= Cn β > 0 (3.69)
∂β

The yawing moment of the aircraft is composed of contributions of the fuselage, wing, wing-
fuselage-interference and vertical tail

Nβ = NβF + NβW + NβW +F + NβV . (3.70)

For this approximate consideration, the contributions of the wing and the wing-fuselage-
interference are neglected. The fuselage contribution then remains as the essential quantity
of the wing-fuselage arrangement:

NβF = −2k ∗ qVF = (Cnβ )F Ssq (3.71)

hF0
Here, k ∗ is a constant for the fuselage shape. It is a function of the fuselage slenderness lF
bF0
and the cross-sectional shape hF0
.

If the fuselage is imagined as a cylindrical body with elliptical cross section of constant ratio
of width to height over the entire length of the fuselage, we could find the fuselage volume:

Z le Z lF
π π bF
VF = · bF · hF · dxF = · · hF 2 dxF
4 0 4 hF 0

Inserted into Equation 3.71, the result is

Z lF
π
NβF = −k2 · · q · hF 2 · dxF (3.72)
2 0

bF0 bF0
where k ∗ = −k2 · hF
bF
and k2 can be taken from the given values hF0
and lF
in Figure 3.20.
58 3 Static Stability

1,0
k2 bF 0

0,8 hF 0

0,50
0,6 0,75

1,00
0,4
1,25

1,50
0,2
2,00

hF 0
0,0 0,1 0,2 0,3 0,4 0,5
lF

� bF 0

xF hF 0

lF

Figure 3.20: Factor for calculation of fuselage moment

The vertical tail contribution is

∂βw
NβV = −(CYβ )V · (1 + ) · SV · rV · q V with (3.73)
∂β
SV = vertical tail surface area,
rV = distance between the vertical tail neutral point and the center of gravity,
∂βw
βV = (1 + ) · β effective sideslip angle of the vertical tail,
∂β
(3.74)

where (CYβ )V is always a negative value (e.g., (CYβ )V = −2).

The effective sideslip angle differs from the free sideslip angle through interference effects,
mainly due to asymmetries of the wing-fuselage arrangement. This results in the following
qualitative influence:

βV
High wing configuration: < 1 (destabilizing)
β
(3.75)
βV
Low wing configuration: > 1 (stabilizing)
β

The weathercock stability of the overall aircraft is thus:

3.2 Stationary Lateral Motion 59

2k ∗ · VF ∂βw
Cnβ = − − (CYβ )V · (1 + ) · VV (3.76)
S·s ∂β

·rV ·q V
with VV = SVS·s·q (Vertical-tail volume ratio) and βw
β
the change of the effective sideslip
angle at the vertical tail.

Yaw Moment due to Rudder Deflection Nζ = Cnζ · S · sq

Analogous to subsubsection 3.2.2.2, a positive rudder deflection creates a positive side force
at the vertical tail and thus a negative yaw moment. The derivative is

∂βV
Cnζ = (CYβ )V · · VV (3.77)
∂ζ

S·s2
Yaw Moment due to Yaw Rate / Yaw Damping Nr = Cnr · · q with Nr = ∂N
∂ ( r·s
V )
V0

The main contribution of the yaw damping is generated by the vertical tail. Analogous to
subsubsection 3.2.2.2, the derivative for the vertical tail contribution of the yaw damping is

rV
(Cnr )V = (CYβ )V VV . (3.78)
s

The other aircraft components, especially the fuselage, contribute only a small increase of
about 10% to 20%.
The wing itself yields a contribution as a result of the different relative inflow velocities
of the portside and starboard wing, which vary linearly in the spanwise direction. This
contribution is created by the induced wing yawing moment (Cnr )W 1 (which is depending
on the lift force) and the yawing moment (Cnr )W 2 caused by the airfoil drag CDP .
For a wing with elliptical lift distribution, there is

CL α∞
1 3 1 1 + πΛ
(Cnr )W 1 = −
CL 2 4 πΛ 1 + 2 CLα∞
πΛ
CDP
(Cnr )W 2 =− (Assumptions: CDP = const.). (3.79)
2

In case of rectangular wings (Cnr )W 2 = − 32 CDP .

To give an overview of the magnitude of how the rudder and the wings contribute to the
yaw damping, (Cnr )V and C1L 2 (Cnr )W 1 are represented in Figure 3.21 .
60 3 Static Stability

Thereby, (Cnr )V was calculated in a simplified consideration according to Equation 3.78 as

a function of SSV with VV = SVSsq
rV qV
. In this calculation a constant value of (CYβ )V = −2 was
set.

-0,8
( Cn ) r
rS qS S =1,6
=1
q s -0,10
-0,6 1 (C ) Ellipse CA =5,2
( C ) =-2,0 1,2 2 nr F1 a
Yb S CA
Rectangle
-0,4 -0,06

0,8 -0,04
-0,2 Triangle
0,4 -0,02

0,0 0,1 SS 0,2 4 8 10 12 14

S 0 2 6 L
Figure 3.21: Contributions of vertical tail and wing on the yaw damping

3.2.2.4 Rudder Trim

Considering only the yawing motion (single degree of freedom consideration), a trimmed
flight condition results from the condition
P
N = 0:

N = Nβ · β + Nζ · ζ = 0 (3.80)

β
Vo

βV =β +βW

+ζ

Figure 3.22: Rudder trim

The transmission ratio of rudder deflection angle ζ to sideslip angle β in a trimmed condition
using Equation 3.76 and 3.77 is
3.2 Stationary Lateral Motion 61

2k∗ VF ∂βw
+ (CYβ )V (1 + )VV
 
ζT r Cn S·s ∂β
=− β = (3.81)
β Cn =0 Cnζ (CYβ )V ∂βV
VV
∂ζ

The influence of the free rudder on the directional stability as well as on the ratio ζTβr
corresponding to Equation 3.81 can be found analogously to the longitudinal stability by
introducing

[(CYβ )V ]f ree Cr ∂βV

=1+ β = δVf ree , (3.82)
(CYβ )V Crζ ∂ζ

where

Crβ
ζV = − βV (3.83)
Crζ

represents the floating angle of the rudder.

3.2.2.5 Rudder Foot Force

Analogous to subsection 3.1.9 the rudder hinge moment is donated by Mh,r . With the
deflection path of the pedals sF , the foot force FF is

FF · δsF + Mh,r · δζ = 0

y
M hr FFF F == Foot force
x
z M r ==Rudder hinge
M
ς hrs
FF moment
pedals s =
sF = Foot path
F
sF

Figure 3.23: Rudder control

 
Using the transmission between pedal path and rudder deflection ∆ζ
∆sF
the actuation force
t
for the rudder can be calculated as

   
∆ζ ∆ζ
FF = −Sr · lr · · q V · Ch,r ; >0 (3.84)
∆sF t ∆sF t

Sr and lr represent the area and depth of the rudder, and Ch,r is the coefficient of the rudder
hinge moment, which is related to the local sideslip angle βV of the vertical tail and the
rudder deflection angle ζ by a linear approach:
62 3 Static Stability

Ch,rV = C(h,r)β βV + C(h,r)ζ ζ. (3.85)

The foot force which is required to achieve a sideslip angle β is according to Equation 3.84,
3.85, and using 3.80 is

C(h,r)β βV
   
dFF ∆ζ Cn
= −Sr lr q V C(h,r)ζ − β . (3.86)
dβ ∆sF t C(h,r)ζ β Cnζ

Using the relations 3.76 and 3.77, we also get

δVf ree −1
(z }| { )
C(h,r)ζ C(h,r)β ∂βV 2k ∗ VF
 
dFF ∆ζ βV βV
= −Sr lr q (CYβ )V VV + + (CYβ )V VV
dβ ∆sF t V Cnζ C(h,r)ζ ∂ζ β Ss β
C(h,r)ζ 2k ∗ VF
   
dFF ∆ζ βV
= −Sr lr q (CYβ )V VV δVf ree + . (3.87)
dβ ∆sF t V Cnζ β Ss

3.2.3 Rolling Motion (Single Degree of Freedom)

3.2.3.1 General Definitions

Rolling means a rotational motion of the aircraft about the longitudinal axis. Since there
are no restoring moments due to the bank angle Φ, an airplane inherently has no rolling
stability. (It could only be generated artificially via roll-attitude gyros, which cause a rolling
moment proportional to the bank angle Φ via an actuation of the ailerons).
According to Figure 3.18, the angle Φ is generally defined as the angle between the nodal
line k2 and the body-fixed y-axis. Only in the special case where Θ = 0 (horizontal flight),
Φ represents the inclination of the x-y-plane of the aircraft with respect to the horizontal
plane.
The primary actuators of the rolling motion are the ailerons. Pushing the control stick to
the left or right side causes the port and starboard rudders to deflect in opposite directions
using a linkage system (see Figure 3.24).
The aileron deflection angle ξ is defined as

ξr − ξl
ξ= (3.88)
2

For ξ > 0 the roll moment is therefore L < 0. ξr or ξl are positive, if the aileron’s trailing
edge moves in positive z-direction (downward).
3.2 Stationary Lateral Motion 63

ard
Starbo MrAi
+�
de
Portsi
�r
+�
l
-�
l
Figure 3.24: Roll control scheme

The coefficient of the roll moment is

L
Cl = .
S·s·q

3.2.3.2 Side Force Derivatives

Side Force due to Roll Rate Yp = CYp · S · q with Yp = ∂Y

∂ ( p·s
V )

Wing contribution:
A side force resulting from a stationary rolling motion can only be created if the wing has a
dihedral angle ν. As can be seen in Figure 3.25,

Figure 3.25: Lift distribution when rolling

a side force appears which is proportional to ν. The coefficient can be given for the elliptical
wing with incompressible flow in closed form:
64 3 Static Stability

1 1 ∂CY 4 · CLα∞
C Yp = p·s = −

C
(3.89)
ν ν∂ V 3π(1 + 2 Lα∞ ) πλ

Vertical tail contribution:

As a result of roll rate p, a side force distribution occurs at the vertical tail. For estimation
purposes, only the value for α = 0 is given here:

(CYβ )V hV SV q V
(CYp )V = (3.90)
2 s S q

Side Force due to Aileron Deflection Yξ = CYξ · S · q

Again, a side force due to an aileron deflection can only be created if the wing has a dihedral
angle:

Z b
2 dLξ
Yξ = −2ν · dy.
0 dy

For an elliptical wing, Clξ is obtained according to Equation 3.91 as

1 16
C Yξ = Cl (3.91)
ν 3π ξ

3.2.3.3 Roll Moment Derivatives

S·s2
Roll Moment due to Roll Rate / Roll Damping Lp = Clp · · q with Lp = ∂L
∂ ( p·s
V )
V0

Lp is essentially generated by the antimetric angle of attack distribution on the wing resulting
from the rolling rate (see Figure 3.26). According to the lifting-line theory, the derivative of
roll damping can be obtained as follows for a wing with elliptical lift distribution

1 CLα∞
Cl p = − (3.92)
4 1 + 2 C L α∞
πΛ
3.2 Stationary Lateral Motion 65

Da (y)

p
P S
-b2 V
y
b
S= 2

Figure 3.26: Additional angle of attack when rolling

For CLα∞ = 2π, the following values of Clp are obtained depending on the wing aspect ratio Λ:

Λ= 4 6 8 10 ∞
−Clp = 0,785 0,940 1,045 1,120 1,570

Roll Moment due to Aileron Deflection Lξ = Clξ · S · s · q

Lξ is obtained from the integration of the local lift distribution over the wing span:

Z + 2b
dLξ
Lξ = ydy
− 2b dy

with η = y
b we get
2

 2 Z +1
b l(η)
Lξ = −li q CLξ (η) ηdη
2 −1 li
66 3 Static Stability

A A

Ah

Figure 3.27: Local lift distribution for an aileron deflection

For a wing with elliptical lift distribution, we can obtain

4 πΛ ∂α 3
Clξ = − r (1 − ηA 2 ) 2 (3.93)
3π  πλ
2 ∂ξ
CLα∞
+4+2

with the designations:

1 − ηA : Span extension of the aileron (dimensionless)

∂α
: Flap action of the aileron,
∂ξ
λ : Wing aspect ratio

Figure 3.28 depicts an evaluation of the Equation 3.93 for a flap action ∂α
∂ξ
corresponding to
lAh
lA
= 0, 2 .
3.2 Stationary Lateral Motion 67

Figure 3.28: Aileron rolling moment as a function of referenced aileron span

3.2.3.4 Stationary Rolling

Due to the missing restoring moment in the rolling motion, a quasi-stationary equilibrium
state results when the roll moment created by the ailerons is equal in magnitude to the
opposite directed roll damping moment. This state is called stationary rolling:

∂L ∂L s
L= ·ξ+ ·p· =0
∂ξ ∂p V

or with the derivatives according to Equation 3.92 and Equation 3.93

Cl ξ V
p=− · · ξ. (3.94)
Cl p s

Cl
The expression Clξ can be calculated directly for the wing with elliptical lift distribution
p
using Equation 3.93 and Equation 3.92. With good approximation, in the important range
Cl
of the wing aspect ratio 6 < λ < 10, the numerical value Clξ is independent of λ (see
p
Figure 3.29).

Influence of The Elastic Torsion of The Wing

The pressure distribution generated by an aileron deflection causes the wing to twist in the
sense that the effective aileron action is reduced as a result of a reduced local angle of attack:

Leff = Lξ · ξ + LαA · ∆αA with ∆αA = −k1 · q · ξ (3.95)

68 3 Static Stability

Clx lA h
C lp lA

0,8 0,30
0,25
0,6 0,20
0,15
0,4
lA
0,2 lAh

0
0 0,2 0,3 0,4 0,5 0,6 �A S
1- �A S
Clξ
Figure 3.29: Relations of the quotient Clp
to the spanwise extension of the aileron

The effective derivative for the roll moment due to aileron deflection is

(Clξ )eff C l αA
=1− k1 q.
Cl ξ Clξ

Equation 3.94 then can be written as

Cl ξ V ClαA
p=− ξ(1 − k1 q) (3.96)
Cl p s Cl ξ

If we now introduce a limit velocity Vcrit for which p = 0, the following applies

Cl ξ 1
q crit = resp.
C l αA k1
Cl ξ 1
k1 = ;
C l αA q crit

Inserted into Equation 3.96, the result is

  "  2 #
Cl ξ V V
p = − ξ · · Vcrit · · 1− (3.97)
Cl p s Vcrit Vcrit

h i
C
cf. Figure 3.30 with k = − Cllp · s
ξVcrit
ξ
3.2 Stationary Lateral Motion 69

1,0
p K
0,8

0,6
(pK)max=0,384

0,4

0,2
1
3
0
0 0,2 0,4 0,6 0,8 1,0
V
Vkrit

Figure 3.30: Influence of the elastic torsion of the wing

3.2.3.5 Aileron Hand Force

For the control system of the rolling motion (cf. Figure 3.24) it applies

FA · δsA + M(h,A)r · δξr + M(h,A)r · δξl = 0

For simplification, we introduce a fictitious hinge moment Mh,A for the ailerons:

1
Mh,A · δξ = (M(h,A)r · δξr + M(h,A)l · δξl ).
2
 
M(h,A)r is therefore the hinge moment of the right wing’s aileron. dsdξA is the transmission
t
ratio. Thus, the aileron hand force FA is

 
dξ
FA = −2SA lA q · · Ch,A
dsA t

Assuming the relationships as geometrically and aerodynamically linear, the aileron hinge
moment created by the wing’s angle of attack on both wings equals each other (it only causes
a buckling force at the linkage between the ailerons), so that we can obtain

 
dξ
FA = −2SA lA · · q · Crξ ξ (3.98)
dsA t
70 3 Static Stability

3.2.4 Cross-Couplings between Yawing And Rolling Motion

The considerations in subsection 3.2.2 and subsection 3.2.3 were based on the simplifying
assumption that yawing and rolling motion are decoupled. In reality, couplings always occur
between the two motions.
For the stationary or quasi-stationary cases, the couplings are caused exclusively by aerodynamic
influences.
The four most important cross derivatives will be briefly discussed in the following.

3.2.4.1 Roll Moment due to Sideslip Angle

Dihedral contribution:

Figure 3.31: Lift distribution while slipping

Due to a lateral velocity v an additional angle of attack occurs (see Figure 3.31):

left wing right wing

w= −v sin ν +v sin ν
∆α = − Vv sin ν + Vv sin ν
∆α ∼
= −βν +βν

This results in a roll moment due to a sideslip angle

Z b
2 1 dL
Lβ = − νydy = Clβ Ssq
− 2b |∆α| dy

For an unswept elliptical wing, we get

4 πΛ
(Clβ )ν = − r ν (3.99)
3π  2
4 + CLπΛ +2
α∞
3.2 Stationary Lateral Motion 71

Fuselage contribution:

Z0
R
b
Figure 3.32: Change in lift due to fuselage influence

The change in angle of attack due to side flow around the fuselage creates an additional roll
moment (Clβ )F = f ( zR0 ). The fuselage contribution can be expressed by an effective dihedral :

" r #
z0  z 2
0 z0 z0
νeff = 3ηR 2 1− + arcsin − π · ηR (3.100)
R R R R

with ηR = R
s
. As an example, for 2R
b
= 0, 12 we get

z0
R
= +0, 5 +1, 0
νeff = 1, 9◦ 2, 9◦

3.2.4.2 Roll Moment due to Yaw Rate

Wing contribution:
The wing contribution of the roll moment due to a yaw rate results from the different
magnitude of the local air flow along the wingspan (see Figure 3.33).

-∆V = ry
y

ΔV

Figure 2.2.17

Figure 3.33: Local air flow along the wingspan for a yawing wing

The following approximation applies for the elliptical wing:

72 3 Static Stability

C L α∞
" #
CL 1
(Clr )W = 1− πΛ
(3.101)
2 2 1 + 2 C L α∞
πΛ

Vertical tail contribution:

l�
V
l�
V

x 4

xe
� hV
p(e) z�
D-AF V
SR

rV
r(e)

z ze
Figure 3.34: Vertical tail geometry

For the effective sideslip angle at the vertical tail due to r, the following applies

rrV rV
(βeff )V ≈ − = −Ωz
V s

and with the effective lever arm zηV − rV α for the moment

LV (r) = (CYβ )V (βeff )V SV (zµV − rV α)q V =

= (Clr )V Ωz Ssq,
z rV 
ηV
(Clr )V = −(CYβ )V VV − α . (3.102)
s s

3.2.4.3 Yaw Moment due to Roll Rate

Wing contribution:
3.2 Stationary Lateral Motion 73

dDi

Figure 3.35: Lift distribution on the rolling wing

a) The following applies for the yaw moment due to different drag on the left and right
wing
Z b
2 dD (p)
i
NW1 (p) = + ydy, (3.103)
b
−2 dy

b) and for the yaw moment due to the lift components in x-direction

Z + 2b
dL
NW2 (p) = − αx ydy (3.104)
− 2b dy

Using the lifting-line theory, the result for both contributions for a wing with elliptical lift
distribution is

L α∞ C
CL 1 − πΛ
(Cnp )W =− (3.105)
4 1 + 2 CLα∞
πΛ

Vertical tail contribution:

Due to the roll rate p, an effective sideslip angle distribution occurs at the vertical tail

pze ze
βeff (ze ) = − = −Ωx
v s

and thus a yaw moment (see Figure 3.36)

74 3 Static Stability

∆V = -pze

ze α>0
1
α α=0
hSe
p(e) ze hS

N ze
2

hS = hS cos α
Figure 2.2.20
e

Figure 3.36: Sideslip angle distribution on the vertical tail

Z ze2
nV (p) = −q V rV Cy (ze )ls (ze )dze = (Cnp )V Ωx Ssq
ze1

The following approximation is obtained for α = 0 :

(CYβ )V hV
(Cnp )V = − VV (3.106)
2 s

With α increased, (Cnp )V can be negative.

3.2.4.4 Yaw Moment due to Aileron Deflection

Wing contribution:
In addition to the primarily desired roll moment, the aileron also provides a contribution to
the total yaw moment. The reasons for this is that a downward deflected aileron (outer wing
during a turn) causes an increase in lift and drag on the outer wing. Inversely to this, an
upward deflected aileron causes a reduction in lift and drag at the inner wing. This results
in a roll moment promoting the turning, but also in a yaw moment directed against the
direction of the turning. This is due to the asymmetric distribution of the induced drag.
This effect, called "adverse yaw" by pilots, must be compensated with the rudder. The
effect especially occurs on wings with high aspect ratio (e.g., high performance gliders) and
at high lift coefficients (low speed flight).
As a result of the induced drag distribution, the following applies

Z + 2b
dDi
N1 (ξ) = (ξ, CL )ydy. (3.107)
− 2b dy
3.2 Stationary Lateral Motion 75

DWi(y)
i (y)

Figure 3.37: Lift distribution due to aileron deflection

One method to reduce the "adverse yaw" are differential ailerons. Since the increase in
drag of a downward deflected aileron is usually greater in magnitude than the the decrease
of drag of an upward deflected aileron by the same angle, differential ailerons cause the
upward deflected aileron to deflect more than the downward deflected aileron. This results
in a reduction of the yaw moment due to aileron deflection. This is also the reason for the
definition of the aileron deflection ξ, which is often defined as the average difference between
the left and right aileron deflection:

ξr − ξl
ξ= (3.108)
2

Vertical tail contribution:

Especially for aircraft with low wing aspect ratio (e.g., delta wing configurations / fighter
aircraft), the ailerons are close to the xb -axis and have to be deflected strongly to generate
the necessary roll moments. The strong aileron deflections also have a large effect on the
spanwise distribution of lift and induced drag. According to Prandtl’s lifting-line theory, the
fluidized bed and the direction of rotation of the fluidized bed in the wing wake depend on
the gradient of lift in the spanwise direction, there is a lateral inflow on the rudder and thus
a further yaw moment contribution. However, this contribution results in a yawing motion
in the direction of the turn (promoting the turning). This rudder contribution is particularly
appearing on wings with low aspect ratio and on ailerons mounted close to the xb axis.

N2 (ξ) = −(CYβ )V β V SV rV q V ,
βV
Cnξ2 = −(CYβ )V ( )VV (3.109)
ξ
76 3 Static Stability

L (y)

Δv( Γ )

Fluidized bed

Horizontal tail β V=
ΔV
V

Figure 3.38: Induction at the vertical tail

The total yaw moment due to a roll rate results from the sum of wing and rudder contribution.
An exemplary curve of the yaw moment due to aileron deflection coefficient Cnξ for a delta
wing is depicted in Figure 3.39. The turning-promoting rudder contribution dominates at
low lift coefficients while the yaw moment due to aileron deflection coefficient changes its
sign at higher lift values.
CL =0,73

Cn
�
CL =0,49
�η SS
QA

CL =0,245

CL =0

Figure 3.39: Example of the angle of attack relationship of Cnξ (ηA ) for a delta airplane with
vertical tail
3.2 Stationary Lateral Motion 77

3.2.5 Summary of the Lateral Motion Derivatives

Note: The values in brackets are the equation numbers of the corresponding derivatives.
Side force Roll moment Yaw moment
Sideslip angle CYβ Clβ Cnβ
β (3.66a) Wing (3.99) (3.76)
Fuselage (3.100)
Roll rate CYp Clp Cnp
p Wing (3.89) (3.92) Wing (3.105)
Vertical tail (3.90) Vertical tail (3.106)
Yaw rate CYr Clr Cnr
r (3.76) Wing (3.101) (3.78)
Vertical tail (3.102)
Aileron deflection CYξ Clξ Cnξ
ξ (3.91) (3.93) (3.107),
(3.109)
Rudder deflection CYζ Clζ Cnζ
ζ (3.67) (3.77)

3.2.6 Stationary Sideslip Flight

An aircraft must be designed in a way, that stationary sideslip flights with large sideslip
angles are possible in accordance with the certification regulations.
From the equilibrium conditions

X
Y =0
X
L=0
X
N =0

the following three equations are obtained for the aileron and rudder deflection, and the
bank angle as a function of the sideslip angle β:

X
Y =0: − G sin Φ + Yβ β + Yζ ζ + Yξ ξ = 0
X
L = 0 : Lβ β + Lζ ζ + Lξ ξ = 0 (3.110)
X
N = 0 : Nβ β + Nζ ζ + Nξ ξ = 0

or
78 3 Static Stability

CL Φ + CYζ ζ + CYξ ξ = −CYβ β (Φ small: tan Φ ≈ Φ) (3.111)

Clζ ζ + Clξ ξ = −Clβ β (3.112)
Cnζ ζ + Cnξ ξ = −Cnβ β (3.113)

AL

W sinΦ

Φ
ze W

Figure 3.40: Sideslip flight with bank angle

From Equation 3.112 and Equation 3.113, the following applies for ζ and ξ:

Clζ Cnβ − Cnζ Clβ ∼ Clβ

ξ=− β =− β,
Clζ Cnξ − Cnζ Clξ Clξ
(3.114)
Cn Cl − Cl ξ Cn β ∼ Cn β
ζ=− ξ β β =− β
Clζ Cnξ − Cnζ Clξ Cn ζ

and with Equation 3.113 for Φ:

β ζ ξ
Φ=− (CYβ + CYζ + CYξ ),
CL β β
C Cl (3.115)
β nβ
Φ∼
= − (CYβ − CYζ − C Yξ β )
CL Cnζ Clξ

3.2.7 One Engine Out Flight

The failure of the starboard engine is considered (see Figure 3.41). The bracketed sign
applies alternatively to the failure of the portside engine.

Without banking: Φ = 0◦

With the equilibrium of forces and moments

3.2 Stationary Lateral Motion 79

X
Y = 0 : C Yβ β + C Yζ ζ + C Yξ ξ = 0
X
L = 0 : Clβ β + Clζ ζ + Clξ ξ = 0 (3.116)
X − FT · yFT
N = 0 : Cnβ β + Cnζ ζ + Cnξ ξ = (+)
Ssq

the following relations apply for stationary flight:

− FT ·yFT
(+) Ssq (CYζ Clξ − Clζ CYξ ) − C Yζ C l ξ FT · yFT
β= ∼
= (+) · (3.117)
D1 D1 Ssq
+ FT ·yFT − F ·y
CYβ ((−)Cl ξ )(+)Clβ (CYξ TSsqFT ) CYβ Clξ FT · yFT
ζ=
Ssq ∼
= · (3.118)
D1 D1 Ssq
− FT ·yFT
CYβ ((+) Cl ζ Ssq
)
CYβ Clζ − CYζ Clβ FT · yFT
−
ξ= = (+) (3.119)
D1 D1 Ssq
with D1 = CYβ (
Cl
ζ
Cnξ − Cnζ Clξ ) − Clβ (CYζ −  n ) + Cn (CY Cl − 
CYξ C 
ζ β ζ ξ
Cl
ζ
CYξ ) (3.120)

Portside Starboard

yF

+ζ

Figure 3.41: Definitions during single engine flight

Without sideslip: β = 0◦ (Side force compensation through banking)

X
Y = 0 : CL tan Φ + CYζ · ζ + CYξ · ξ = 0
X
L = 0 : Clζ · ζ + Clξ · ξ = 0 (3.121)
X − FT · yFT
N = 0 : Cnζ · ζ + Cnξ · ξ = (+)
Ssq
80 3 Static Stability

The following conditions apply for stationary flight:

− CYζ Clξ − Clζ CYξ FT · yFT − C Yζ C l ξ − 

Cl CYξ FT · yFT
Φ = (+) · = (+) ζ
· (3.122)
D2 Ssq Cl
 ζ
Cnξ − Clξ Cnζ Gs
FT y FT
+ CL Cl ξ FT · yFT + Cl ξ ( G ) · ( s ) · CL
ζ= (−) · = (−) (3.123)
D2 Ssq Cl
 ζ
Cnξ − Clξ Cnζ
FT y FT
− CL Cl ζ FT · yFT − Clζ ( G ) · ( s ) · CL
ξ= (+) · = (+) (3.124)
D2 Ssq Cl
 ζ
Cnξ − Clξ Cnζ
with D2 = CL (
Cl
ζ
Cnξ − Clξ Cnζ ). (3.125)

This case, which must be verified for every multi-engine aircraft, can be decisive for the
dimensioning of the rudder size and the foot forces, thus the aircraft can be controlled in an
one engine out scenario at minimum flight speed (e.g. take-off with engine failure).

3.2.8 Stationary Turning Flight

Right turn: Ω > 0, Φ > 0, Left turn:: Ω < 0, Φ < 0

AL

rT
y

mg

ze

Figure 3.42: Forces during stationary turning flight in the inertial system

Assuming that the resulting air force is in the symmetry plane (Y = 0; β ̸= 0, but very
small, cf. Equation 3.126),the following force equilibrium applies:

X
Y = 0 : mg sin Φ − mV Ω cos Φ = 0
X (3.126)
Z=0: − L + mg cos Φ + mV Ω sin Φ = 0
3.2 Stationary Lateral Motion 81

Solving the equation Y = 0 using V = ΩrT , the result for the bank angle or the turn
P

radius is

V V2 V2
tan Φ = Ω = resp. rT = (3.127)
g rT g g · tan Φ

Using L = G cos Φ(1 + tan2 Φ) from Z = 0 and cos Φ = √1+tan

1
2 Φ , we get the definition of
P

the load factor n = G :

L

p 1
n= 1 + tan2 Φ = (3.128)
cos Φ
or

√
tan Φ = n2 − 1 (3.129)

As an example, numerical values for n, Φ and rT are listed in Table 3.2 for M = 0, 8 and
H = 0 km.

Table 3.2: Numerical values for n, Φ and rT for M = 0, 8 and H = 0

n tan2 Φ Φ [◦ ] rT [km]
1 0 0 ∞
2 3 60 4,37
5 24 78,5 1,54
8 63 82 0,955

Compensation of the Longitudinal Moment

For compensation of the longitudinal moment, additional lift ∆L is necessary, which is

created by the additional angle of attack ∆α.
With G = L0 , L0 + ∆L = G · n, ∆L = L0 · n − L0 = (n − 1) · L0 and ∆CL = (n − 1) · CL0 the
following applies for the additional forces in the symmetry plane and the pitching moment:

lµ
CLα ∆α + CLq q · + CLη ∆η = (n − 1)CL0
V
X lµ
M = 0 : Cmα ∆α + Cmq q · + Cmη ∆η = 0 (3.130)
V
resp.

lµ
CLα ∆α + CLη ∆η = (n − 1)CL0 − CLq · q ·
V (3.131)
lµ
Cmα ∆α + Cmη ∆η = −Cmq · q ·
V
82 3 Static Stability

The following is obtained for the additional angle of attack

[(n − 1) · CL0 − CLq · q · lVµ ]Cmη + Cmq · q · lµ

C
V Lη
∆α = (3.132)
CLα · Cmη − Cmα · CLη

When neglecting the lift due to pitching as well as the lift of the elevator, we obtain

CL0
∆α ∼
= (n − 1) · (3.133)
CLα

The additional elevator deflection angle during the turning flight is

−CLα · Cmq · q · lVµ − Cmα [(n − 1) · CL0 − CLq · q · lµ

]
∆η = V
(3.134)
CLα · Cmη − Cmα · CLη

The rotation vector Ω can be represented in the experimental system as (cf. Figure 3.43):

tan Φ
q = Ω · sin Φ = Ω · √
1 + tan2 Φ
r = Ω · cos Φ

r
q
W f

Figure 3.43: Rotation vector in the experimental system

Using 3.127 and 3.128 in this equation, the result is

n2 − 1 g
q= · (3.135)
n V

Further using

1 g · ρ · S · lµ lµ · g
= = ,
µ 2·G CL0 · V 2

the following applies

3.2 Stationary Lateral Motion 83

lµ CL0 (n − 1) · (n + 1)
q =
V µ n

and thus

n+1
Cm α + µn
[CLα Cmq − CLq Cmα ]
∆η = (n − 1)CL0 (3.136)
Cmα CLη − CLα Cmη

Aileron and Rudder during Turning Flight

From the force and moment equilibrium

X s
Y : CYβ β + CYζ ζ + CYξ ξ = −CYr r
V
X s
N : Cnβ β + Cnζ ζ + Cnξ ξ = −Cnr r (3.137)
V
X s
L : Clβ β + Clζ ζ + Clξ ξ = −Clr r
V

the following applies

Ω g tan Φ
D = CYβ (Cnζ Clξ − 
Cl
ζ
Cnξ ) r = Ω cos Φ = √ = √
2
1 + tan Φ V 1 + tan2 Φ
− Cnβ (CYζ Clξ − 
Cl
ζ
CYξ ) and
√
n2 − 1 g
+ Clβ (CYζ Cnξ − Cnζ CYξ ) r= ·
n V

For the sideslip angle

−CYr CYζ CYξ

−Cnr Cnζ Cnξ
√
−Clr Clζ Clξ n2 − 1 g · s
β= · · 2 (3.138)
D n V

and the rudder deflection

CYβ −CYr CYξ

Cnβ −Cnr Cnξ
√
Clβ −Clr Clξ n2 − 1 g · s
ζ= · · 2 (3.139)
D n V

and the aileron deflection

84 3 Static Stability

CYβ CYζ −CYr

Cnβ Cnζ −Cnr
√
Clβ Clζ −Clr n2 − 1 g · s
ξ= · · 2. (3.140)
D n V

For a turning flight without sideslip (β = 0) and neglecting the force equation Y , we
P

obtain from Equation 3.137 for the rudder and aileron deflection:

√
Cnζ Clr − Clξ Cnr n2 − 1 g · s
ξ= · · 2
Cl Cnξ − Cnζ Clξ n V
(3.141)
ζ
√

Cnr Clξ − Clr Cnξ n2 − 1 g · s
ζ= · · 2
Cl
 ζ
Cnξ − Cnζ Clξ n V
 85

4 Dynamic Stability

4.1 Stability Axis System

In the previous chapter, only stationary flight conditions were taken into account. The
analysis was limited to determining the conditions under which equilibrium of forces and
moments is achieved. For example, the required elevator deflection angle for a certain lift
coefficient in horizontal flight, or the aileron and rudder deflection angles for a steady-
state flight with constant sideslip angle were asked. Mathematically, this required solving
algebraic equations or systems of equations. These states are called stationary because they
are independent of time.
In dynamic processes time is an essential variable. If a system is disturbed from a stationary
state, the system’s states vary over time until a new stationary state is reached. The
disturbance can occur inside the system (e.g. by rudder actuation), or externally (by
disturbances, such as gusts). The dynamic behavior of the aircraft is basically described
by the differential equations set up in chapter 2. However, since these are highly nonlinear,
they cannot be used in this form for analytical considerations, such as stability studies.
Thus, a linearization of the equations of motion is necessary.
The linearization is possible in the range of small deviations around a stationary state in
good approximation. The derivatives can then be regarded as approximately constant. The
linearization leads to differential equations or coupled systems of differential equations with
constant coefficients, which can be analyzed with the methods of system theory. For example,
by applying the Laplace transformation on the equations of motion, we obtain quite simple
algebraic equations. The solutions obtained in the frequency domain can be returned to the
time domain by relatively simple transformations using correspondence tables. However, it
is also possible to predict important properties of the solutions in the time domain without
a back transformation by applying the rules of the Laplace transformation (e.g., natural
frequency and damping of oscillatory modes of the system).
To simplify the linearization, the longitudinal and lateral motion of the aircraft are considered
independently. The equations for the study of the dynamic processes are set up in the
stability axis system as already mentioned in section 2.2. The further explanations of the
stability axis system are based on the longitudinal motion as an example. The stability axis
system is then an aircraft-fixed axis system (x, z) whose axes coincide with those of the
aerodynamic system in the stationary (trimmed) initial state (x0 , z0 ).
86 4 Dynamic Stability

In the initial state (index 0)1 , cf. Figure 4.1 , x0 lies in the direction of the aerodynamic
velocity vector V0 , which is tilted by the climb angle γ0 with respect to the earth-fixed
coordinate system in the absence of wind. The attitude relative to the aircraft is given by
the angle αT r between x0 and the zero lift direction of the wing-fuselage combination

2mg cos γ0
CLT r = CLα αT r = (4.1)
ρSV0 2

(cf. Equation 3.2 and Equation 3.10 considering Cm = 0).

CL
Tr
Zero lift direction oferthe
Nullauftriebsrichtungd
Wing-fuselage combination
Flügel-Rumpf-Kombination (CLTr)2

F0 (CLTr)1

V0 �
�Tr
CG
S.P.
x0
�0 �0 �
��Tr�1 ��Tr�2

xg
rF

z0 zg

Figure 4.1: Stationary initial state

If we now consider the axis system x0 , z0 fixed for the stationary (trimmed) initial state
rigidly connected to the airplane, it will be rotated by the angle ∆Θ with respect to the
initial state in case of a deflection due to a disturbance, see Figure 4.2.

1
The initial state is generally marked with the index 0, exceptions are the angle of attack and the lift
coefficient in the trim state, which are marked with the index "‘Tr"’ to avoid confusion with already
defined quantities
4.2 Linearization Approaches 87

xf

xf0
Da+Dg
V a
xa

V0 aTr Dq
xa0 Dg
g q0
g0 q
xg

Zf
Zf0 Za Za0 Zg

Figure 4.2: Disturbed state

According to Figure 4.2, the changes of the individual state variables compared to the steady-
state initial condition are described as follows:

γ = γ0 + ∆γ
Θ = Θ0 + ∆Θ with Θ0 = γ0 + αT r
(4.2)
α = αT r + ∆α with ∆α = ∆Θ − ∆γ
V = V0 + ∆V.

4.2 Linearization Approaches

4.2.1 Linearization of Differential Equations

In a stationary equilibrium state of a nonlinear differential equation ẋ = f (x, u), the changes
of the states, and thus the derivatives, are equal to zero.

ẋ = 0. (4.3)

In this equilibrium point (trim point) the states x and the control variables u are constant
and therefore denoted by the index "0"

f (x0 , u0 ) = 0. (4.4)
88 4 Dynamic Stability

Starting from this trim point, small deviations of the states and control variables are considered:

x = x0 + ∆x, (4.5)
u = u0 + ∆u. (4.6)

The vectorial differential equation ẋ = f (x, u) is now approximated abound the trim point
by a Taylor-series, which is discontinued after the first order to obtain a linear differential
equation:

∂f ∂f
ẋ = f (x0 , u0 ) + · ∆x + · ∆u. (4.7)
|{z} | {z } ∂x x0 ,u0 ∂u x0 ,u0
∆ẋ 0 | {z } | {z }
A B

The result is a linear system of equations in state space representation with the Jacobian
matrices A and B:

∆ẋ = A∆x + B∆u. (4.8)

The system matrix A and the input matrix B are a function of the trim point and can only
be assumed constant for small deviations around the trim point (small ∆x, ∆u). Since only
∆-variables occur in the linear differential equation, the notation ∆ can be dropped. This
leads to a general linear state equation of the form

ẋ = Ax + Bu. (4.9)

4.2.2 Linearization of Trigonometric Functions

Further simplifications can be applied, if the angles between the coordinate systems are
considered to be small. Then the trigonometric functions can be approximated with the
approach

x2 x4
cos x = 1 − + − ··· ≈ 1 (1% error at 8◦ ),
2! 4!

x3 x5
sin x = x − + − ··· ≈ x (1% error at 12◦ )
3! 5!
4.3 Dynamic Longitudinal Stability 89

4.3 Dynamic Longitudinal Stability

4.3.1 Linearization of the Equations of Motion

According to section 2.7 the longitudinal motion can be described by the four state variables
V, α, q, Θ. Equation 2.33 using the assumption β = Φ = 0 yields the following four coupled
differential equations f1 to f4 :

   2   
V̇ −g sin (Θ − α) + FmT cos (α + σ) − ρV2mS CDe f1
g FT ρV S
 α̇  q + cos (Θ − α) − sin (α + σ) − C f2 
  
Le 
 =
 q̇  
V
FT rT
mV
ρV 2 Slµ
2m
= . (4.10)
 Iyi
+ 2Iyi Cme   f3 
Θ̇ q f4

According to section 4.2 the system matrix A of the linear state space representation is
derived by creating the Jacobian matrix ∂x ∂f
x0 ,u0
. The elements of the system matrix are
thus dimensionful derivatives. For better understanding, the first row of the system matrix
∂f1
∂x x0 ,u0
is derived as an example:

∂f1 ρV S cos (αT r + σ) ∂F

= XV + FTV · cos(αT r + σ) = − CDe +
∂V m m ∂V  
∂f1 F0 S ∂CDe S ∂CDe
= Xα = g cos (Θ0 − αT r ) − sin (αT r + σ) − q̄ = q̄ CLe −
∂α m m ∂α m ∂α
∂f1
= Xq = 0
∂q
∂f1
= XΘ = −g cos (Θ0 − αT r ) = −g cos γ0 ≈ −g. (4.11)
∂Θ

The simplification of Xα is obtained by inserting f2 , where α̇ = q = 0 at the trim point. If

the partial derivatives for the remaining functions f2 , f3 and f4 are created the same way,
the linear system of equations of the longitudinal motion in matrix notation is obtained as
follows:

 
XV + FTV · cos(αT r + σ) Xα 0 −g · cos(γ0 )
   
V̇ ∆V
FT
 α̇  (ZV − V0V · sin(αT r + σ)) Zα 1 + Zq − Vg0 · sin(γ0 )  ∆α 
 
 
 q̇  =   · 
 q 

 MV + MFV Mα Mq 0 
Θ̇ 0 0 1 0 ∆Θ

This linearized system only describes the behavior of the aircraft without considering control
inputs. In the longitudinal motion, two primary control inputs are considered, thrust and
elevator deflection angle η.
For modeling the thrust, a control variable δT is introduced. δT is considered to be proportional
to the thrust lever position in the cockpit and also to the engine’s degree of throttling,
therefore it is called throttle position. δT is in the range 0 − 1 so δT = Fmax
F
applies.
90 4 Dynamic Stability

Elevator deflection and throttle position together form the input vector
 
η
u=
δT

The Jacobian matrix ∂u ∂f

x0 ,u0
is called input matrix B, also consisting of dimensionful
derivatives. Finally, the linearized system equation of the longitudinal motion considering
control inputs is obtained as follows:

    
XV + FTV · cos(αT r + σ) Xα 0 −g · cos(γ0 )

V̇ ∆V
F
 α̇  (ZV − VT0V · sin(αT r + σ)) Zα 1 + Zq − Vg0 · sin(γ0 )  ∆α 

 =
 q̇   · 
q 

MV + MFV Mα Mq 0  
Θ̇ 0 0 1 0 ∆Θ
 
Xη FδT · cos(αT r + σ)
Fδ   
 Zη − V0T · sin(αT r + σ) ∆η

+ · (4.12)
Mη MδT · m  ∆δT
0 0

The derivatives FTV and FδT describe the change in thrust as a function of airspeed V and
throttle position δT . To derive FTV we use the thrust approach:

 nρ  nV
ρ V
F = F0 · · . (4.13)
ρ0 V0

If the differential is formed at the operating point / trim point (index "0"), this results in
the following correlation:

 nV −1
∂F ∂F nV V nV
= = F0 = F0 . (4.14)
∂V ∂V ρ=ρ0 V0 V0 V0
V =V0

With FTV = 1 ∂F
m ∂V
and considering the altitude dependence of the thrust, we get

nV
FTV = F0 (ρ). (4.15)
m · V0

Using the throttle position δT the derivative FδT is

1 ∂F Fmax
FδT = = . (4.16)
m ∂δT m

The following list describes the relationship between the dimensionful derivatives used in
Equation 4.12 and the dimensionless derivatives denoted in chapter 3.
4.3 Dynamic Longitudinal Stability 91

ρS
XV = − V0 (2CDT r + V0 CDV )
2m
ρS 2
Xα = V (CLT r − CDα )
2m 0
ρS 2
Xη = V CDη
2m 0
ρSlµ
MV = V0 (2CmT r + V0 CmV )
2Iyi
rT
MFTV = FT
Iyi V
ρSlµ 2
Mα = V Cm α
2Iyi 0
ρSlµ2
Mq = V0 Cmq (4.17)
2Iyi
ρSlµ 2
Mη = V Cm η
2Iyi 0
rT
MδT = Fδ
Iyi T
ρS
ZV =− (2CLT r + V0 CLV )
2m
ρS
Zα =− V0 (CDT r + CLα )
2m
ρS
Zq =− V0 CLq
2m
ρS 2
Zη =− V CL
2m 0 η

4.3.2 Longitudinal Modes

Assuming small angles σ, αT r and γ0 , the linear system of equations of the longitudinal
motion can be further simplified. Furthermore, rT = 0 is assumed, eliminating MFTV and
MδT . Zq can also be neglected. This results in:

  
Xα 0 −g
    
V̇ XV + FTV ∆V Xη F δT  
 α̇   ZV Zα 1 0 ·  ∆α  +  Zη 0  ∆η
(4.18)
  
  =   ·
 q̇   MV Mα Mq 0   q  Mη 0  ∆δT
Θ̇ 0 0 1 0 ∆Θ 0 0

Based on this system of equations, the natural modes of the longitudinal motion can now be
studied. The Laplace transformation of the system equation ẋ = Ax + Bu yields

s · x = Ax + Bu. (4.19)

Rearranging results in:

92 4 Dynamic Stability

(sI − A)x = Bu. (4.20)

Setting the external excitation u to zero, we obtain the characteristic equation of the system.
The solution of the characteristic equation

|sI − A| = 0 (4.21)

provides the eigenvalues (poles) of the system. In the longitudinal motion of a typical
stable aircraft, there are two conjugate complex pairs of poles, which relate to two different
oscillation modes, called natural modes: The short-period mode and the phugoid mode.
These two natural modes will be analyzed in detail in the following.

Short-Period Mode

The short-period mode is a (mostly) well-damped, short-period oscillation. The change in

flight velocity V can be neglected approximately. Essentially, only the angle of attack α
and the inclination angle Θ oscillate, while flight velocity V and air-path inclination γ are
approximately constant. Figure 4.3 depicts the oscillation.

xb

Figure 4.3: Oscillation of the short-period mode

After neglecting the velocity equation in Equation 4.18, the remaining equations for the angle
of attack and the pitch rate do not depend on the inclination angle Θ anymore. Therefore,
the equation for the inclination angle can also be removed. This results in the following
system of equations as an approximation for describing the short-period mode:

       
α̇ Zα 1 ∆α Zη  
= · + · ∆η . (4.22)
q̇ Mα Mq q Mη

The characteristic polynomial of Equation 4.22 is

s2 − s (Zα + Mq ) + Mq · Zα − Mα = 0. (4.23)

For a stable aircraft, the characteristic polynomial yields a conjugate complex pair of poles.
With the definition
4.3 Dynamic Longitudinal Stability 93

[s − (σ + iω)] · [s − (σ − iω)] = (4.24)

s2 − 2σs + σ 2 + ω 2 = (4.25)

s2 + 2Dω0 · s + ω02 =0 (4.26)

the natural frequency ω0 and the damping coefficient (or damping ratio) D, or the damping
exponent σ of the short-period mode can be determined by equating the coefficients of the
characteristic polynomial. The position of the poles in the complex plane is depicted in
Figure 4.4.

σ
D =− (4.27)
ω
√ 0
ω0 = σ 2 + ω 2 (4.28)

iw

s1
X w

j sin j = D
w0 s
s

s2 X

Figure 4.4: Natural frequency and damping as a function of the position of the poles

The natural frequency ω0 of the short-period mode is

(4.29)
p
ω0SP = Mq · Zα − Mα .

Mα (or Cmα ) is the predominant influence on ω0 . Cmα is the decisive influencing quantity for
static stability in longitudinal motion. It is a restoring moment that is proportional to the
deviation of α from the trim point. Compared to a mass-spring-damper system, Mα acts as
a kind of spring stiffness (also called "spring constant of the short-period mode"). Together
with the rotating mass of the aircraft, this results in an oscillating system, which motion is
heavily damped.
For the damping of the short-period mode one obtains
94 4 Dynamic Stability

1
σSP = · (Zα + Mq ) . (4.30)
2
The essential influencing factor for σ is Mq (or Cmq ). The pitch damping Cmq counteracts
the rotational velocity.

Phugoid Mode

The remaining conjugate complex pair of poles describes the phugoid mode of the aircraft, in
which the flight velocity V and the air-path inclination γ oscillate, while the angle of attack
α is approximately constant. The oscillation is depicted in 4.5 for an earth-fixed observer.

� =const
α= Const. Flugbahn
Flight path

Figure 4.5: Phugoid mode from the perspective of an earth-fixed observer.)

From the perspective of an observer flying alongside, the phygoid would look like depicted
in Figure 4.6.

t2 ∆x = 0
∆θ = Δγ = 0
xf Δz = -max

√2

∆x = -max t1
∆θ = Δγ = -max ∆x = +max
Δz = 0 xf ∆θ = Δγ = +max
t3 xf Δz = 0
1

∆x = 0
∆θ = Δγ = 0
Δz = +max xf
t0

Figure 4.6: Phugoid mode from the perspective of an observer flying alongside. [Brockhaus]

An approximation for calculating the phugoid mode results from the following subsystem.
Since the angle of attack is approximately constant, the equation for α in Equation 4.18
4.3 Dynamic Longitudinal Stability 95

can be removed. In addition, since the remaining velocity equation does not depend on the
pitch rate q, the corresponding equation for q can also be removed. However, this would
result in a singularity in the state equation due to θ̇ = 0. Thus, the equation for Θ̇ must
be substituted by an equation, which is determined by the already eliminated equation for
α̇ = 0 = ZV ∆V + Zα ∆α + q + Zη ∆η, assuming that Zα = 0 and q = Θ̇. The result is the
system of equations:

         
V̇ XV + FTV −g ∆V Xη Xδ T ∆η
= · + · . (4.31)
Θ̇ −ZV 0 ∆Θ −Zη 0 ∆δT

The characteristic polynomial here is

s2 − s · (XV + FTV ) − g · ZV = 0. (4.32)

The natural frequency of the phygoid using Equation 4.25 and Equation 4.26 is obtained as
r
ρS
(4.33)
p
ω0P hu = −g · ZV = g CLT r
m

and for the damping one obtains

XV + FTV
σP hu = , (4.34)
2

respectively for the damping coefficient:

σP hu
DP hu = − . (4.35)
ω0P hu

Using the condition for horizontal flight

2mg
CLT r = − , (4.36)
SρV 2 0

and the dimensionless coefficients, the result is:

√ g
ω0P hu = 2 , (4.37)
V0

g CDT r
σP hu = − · , (4.38)
V0 CLT r

1 CDT r
DP hu = √ · . (4.39)
2 CLT r

This result is very interesting, since the natural frequency of the phugoid depends only on
the airspeed, but not on aircraft parameters. The damping is proportional to the lift-to-drag
ratio: The smaller the aircraft’s drag, the smaller the damping.
96 4 Dynamic Stability

The phugoid is an oscillation in which there is a constant exchange of potential and kinetic
energy. This is made clear by the following observation. In undamped oscillation, the total
energy is constant and amounts to

V2
m + mgH = const. (4.40)
2

The lift equation in the undisturbed state leads to

k g
mg = LT r = kV0 2 respectively = 2. (4.41)
m V0

For the motion in longitudinal direction it applies

mḦ = ∆L = k(V 2 − V0 2 ). (4.42)

From the energy equation follows

V 2 − V0 2 = (HT r − H)2g

and therefore

2g 2
Ḧ = (HT r − H). (4.43)
V0 2

d2 (H−HT r )
Because of Ḧ = dt2
the frequency ω of this oscillation is

 2
2 g
ω =2
V0

respectively the period of oscillation is

√ V0
T = 2π
g

which is the same result as in Equation 4.37.

4.3 Dynamic Longitudinal Stability 97

4.3.3 Poles of the Longitudinal Motion

The approximate considerations used in this section each disregard one degree of freedom
of the overall longitudinal motion and therefore decouple the equations of motion into two
independent systems of equations. Figure 4.7 shows a comparison of the numerical exactly
determined poles of the longitudinal motion with the poles of the approximate solutions for
a typical aircraft configuration (only the upper half plane is depicted).

1,5

V=
V 20
= 5m/ sm/s
205 iω
[1/s]
short-period
Anstellwinkel-
mode
bewegung
160 m/s
160m/ s
1,0

120120m
m/s/ s

0,5
78 m/s
78m/ s

Exact solution
ExakteL ösung
Näherung
Approximation
V=78m
V=78 / s
m/s
phugoid
Phygoide 205 m/s
mode 205m/ s
0
-1,5 -1,0 -0,5 0
σ [1/s]

Figure 4.7: Poles of the short-period mode and the phugoid mode in the complex plane

4.3.4 Requirements on Handling Qualities

The flight qualities have already been explained in section 1.1. Figure 4.8 depicts the results
of an evaluation (pilot survey) on the handling qualities using Cooper-Harper rating for the
short-period mode. The left subpicture shows the ranges that pilots described as good to
unacceptable in simulated target tracking tasks. The right subpicture depicts the normalized
step responses of the corresponding transfer functions.
98 4 Dynamic Stability

ω0 f 2
Bad
5
1/s Hz
4 0.64 Acceptable
6 2
q 5
3 5 3 1
2 0.32 3
Unacceptable
3 4
1 4
0 sec
1.25 t 2.5
0 -1.5
0.2 0.5 1.0 3.0 D

Figure 4.8: Result of an evaluation on the short-period mode by pilot survey [Brockhaus]

The following can be derived from Figure 4.8:

• Area 2: The aircraft responds over-sensitively to gusts. The pilot can no longer follow
the fast changes of the control variable and is therefore in danger of further stimulating
the oscillation, called pilot-induced oscillation - PIO.
• Area 3: The initial response of the aircraft is to weak (compared to area 1), so the
pilot tends to oversteer.
• Area 4: The aircraft responds too slowly for the required flight maneuvers.
• Area 5: The damping is too low, the pilot has difficulty stabilizing the aircraft (PIO)
and the response to gusts is too strong.

Handling qualities are also valued in some regulations or specifications. The specifications
MIL and EASA CS mostly represent empirical values. Such values may result, for example,
from flight test results shown in Figure 4.8. For the damping of the short-period mode and
the phygoid mode, the values are defined in Table 4.1.

Table 4.1: Damping requirements for phugoid mode and short-period mode
Damping Source
Phugoid > 0, 04 MIL-F 8785
Short-Period Mode 0,35 – 1,3 EASA CS 23/25
0,5 –0,7 MIL-F 1797

In the MIL specification different levels for the handling qualities are defined from level
1 (best) to level 3 (worst, aircraft can be controlled safely but with high pilot workload).
According to MIL-STD-8785 Level 3, an unstable phugoid is allowed if the time doubling
the amplitude is at least 55 seconds. According to MIL-F 9490, D > 0, 3 must generally be
valid for all control systems.
4.3 Dynamic Longitudinal Stability 99

4.3.5 Example of the derivatives of the longitudinal motion of the

A300

Geometric parameters, mass

Aerodynamic parameters
L L L
D D

Reference flight configuration (index "r" left out)

L D

Substitute parameters

Figure 4.9: Technical data of the A300 [Brockhaus]

Figure 4.10: The A300 [Brockhaus]

100 4 Dynamic Stability

Table 4.2: The eigenvalues of the A300 [Brockhaus]

Holding Approach
Phygoid Mode Short-Period Mode Phygoid Mode Short-Period Mode
σ[1/s] -0,0038 -0,835 -0,019 -0,588
ω[1/s] 0,0936 0,8323 0,152 0,820
ω0 [1/s] 0,0937 1,1789 0,153 1,010
εD [◦] 2,3 45,1 7,2 35,6
D[−] 0,0409 0,7083 0,125 0,583
T [s] 67,1 7,5 41,4 7,66

4.4 Dynamic Lateral Stability

4.4.1 Linearization of the Equations of Motion

Following section 2.7, the lateral motion can be described by the four state variables β, p, r, Φ.
Equation 2.38 provides the following coupled nonlinear equations of motion f5 to f8 for the
lateral motion:

   g ρV0 S
  
β̇ p · sin (αTr + ι) − r · cos (α Tr + ι) + V0
Φ cos β cos Θ0 + 2m
C Ye · cos β f5
q̄·Ss
 ṗ  
 =

Ixi
(Cle · cos(αT r + ι) − Cne · sin(αT r + ι)) 
 f6 
q̄·Ss = 
 ṙ  
Izi
(C le · sin(α Tr + ι) + C ne · cos(αTr + ι))  f 7

Φ̇ p · cos ι + r · sin ι + tan Θ0 · (−p sin ι + r cos ι) f8
(4.44)
Assuming small angles αT r , ι, Θ0 , β, the equations can be simplified to:

     
β̇ −r + Vg0 Φ + ρV2m 0S
CY e f5
q̄·Ss
 ṗ   C le  f6 
  
 = Ixi
 =  . (4.45)
 ṙ  
 + q̄·Ss Cn
Izi e  f7
Φ̇ p f8

Also for the lateral motion, the system matrix A of the linear state space representation
according to section 4.2 is obtained by forming the Jacobian matrix ∂x ∂f
x0 ,u0
. The primary
control inputs are the rudder deflection ζ and aileron deflection ξ. The input matrix B is
therefore the Jacobian matrix ∂u
∂f
x0 ,u0
. For better understanding, the first row of the system
matrix ∂f5
∂x x0 ,u0
is derived as an example:
4.4 Dynamic Lateral Stability 101

∂f5 ρV0 S
= Yβ = C Yβ (4.46)
∂β 2m
∂f5 ρV0 S
= Yp = C Yp (4.47)
∂p 2m
∂f5 ρV0 S
= Yr − 1 = C Yr − 1 (4.48)
∂r 2m
∂f5 g
= YΦ = (4.49)
∂Φ V0

Thus, a linear system of equations of the lateral motion in state-space representation is

obtained:

g
Yr − 1 · cos Θ0
  
Yβ Yp
    
β̇ V0 ∆β Yξ Yζ  
 ṗ   Lβ Lp Lr 0   p   Lξ Lζ  ∆ξ
 =
 ṙ  Nβ
·
  r  + Nξ Nζ  · ∆ζ
   (4.50)
Np Nr 0
cos γ0 sin γ0
Φ̇ 0 cos Θ0 cos Θ0
0 ∆Φ 0 0

The relationship between the dimensionful derivatives Y , L, N to the dimensionless derivatives

CY , Cl , Cn is listed as follows:

ρSV0 q̄Ss q̄Ss

Yβ = C Yβ Lβ = Cl Nβ = − Cnβ
2m Ix β Iz
ρSV0 ρSs ρSV0 s2
Yp = C Yp Lp = Cl Np =− Cnp
2m 2Ix p 2Iz
ρSV0 ρSs ρSV0 s2
Yr = C Yr Lr = Cl Nr =− Cnr
2m 2Ix r 2Iz
ρSV0 q̄Ss q̄Ss
Yζ = C Yζ Lζ = Cl Nζ = Cnζ
2m Ix ζ Iz
ρSV0 q̄Ss q̄Ss
Yξ = C Yξ Lξ = Cl Nξ = Cnξ (4.51)
2m Ix ξ Iz
102 4 Dynamic Stability

4.4.2 Lateral Modes

Like in the longitudinal motion, the lateral motion can be decoupled into several natural
modes. Since the different modes differ in their characteristics, it is approximately possible
to study the modes separately. The overall motion is composed of a fast aperiodic rolling
mode with a small roll time constant TR , a very slow (sometimes) unstable spiral mode,
and the Dutch Roll mode whose natural frequency is of the order T1R . In the following, the
different modes are analyzed in more detail.

Dutch Roll Mode / Lateral Oscillation

The Dutch Roll mode is essentially an oscillation of the sideslip angle β, which is coupled
by the cross-couplings with a slighter rolling oscillation. The zeros of the characteristic
polynomial usually yield a stable conjugate complex pair of poles sDR1,2 = σDR ± iω0DR .
Figure 4.11 depicts the oscillation sequence of the Dutch Roll mode.

- Lift vector rotated to the left, thus

lateral motion to the left, ...

- vertical tail inflow from the right side, thus

a positive yaw moment appears
- consequence: yawing to the right
(still rolling left wing down)

- Due to yaw rate, right wing moves foreward,

thus lift on left wing is increased
- Consequence: rolling left wing downward

- Lift vector rotated to the right

- Consequence: lateral motion to the right
(still yawing to the left)

- vertical tail inflow from the left side, thus

a negative yaw moment appears
- consequence: yawing to the left
(still rolling right wing down)

- Due to yaw rate, left wing moves foreward,

thus lift on left wing is increased
- Consequence: rolling right wing downward
roll moment

- Lift vector rotated to the left

- Consequence: lateral motion to the left
(still yawing to the right)
side force

- Right wing more lift, thus more induced drag

- Inflow at the VT from the right
- Consequence: rolling left, yawing right

- Lateral motion to the right

- Inflow at the vertical tail (VT) from the right,
thus yawing to the left
crosswind

Figure 4.11: Dutch Roll Mode [Huenecke]

4.4 Dynamic Lateral Stability 103

Assuming a small rolling motion, the differential equations for p and Φ can be removed in
Equation 4.50. Likewise, the side force due to yaw rate Yr can be neglected, so Yr = Cyr = 0.
The system of equations then simplifies to:

       
β̇ Yβ −1 ∆β Y
+ ζ · ∆ζ (4.52)
 
= ·
ṙ Nβ Nr r Nζ

The characteristic polynomial of the yaw motion is given by

s2 − s (Yβ + Nr ) + (Nβ + Nr Yβ ) (4.53)

Furthermore, the damping σ and the natural frequency ω0 is obtained as

2σDR = − (Yβ + Nr ) (4.54)

(4.55)
p
ω0DR = Nβ + Nr Yβ

Approximation of the Spiral and Rolling Mode

Studying the motion of the spiral mode and rolling mode, it is assumed that these two modes
are not depending on the sideslip angle β. The roll mode is only an aperiodic motion around
the body-fixed xb longitudinal axis of the aircraft. The aperiodic spiral mode consists of a
yaw motion coupled with a slow roll motion.

Spiral Mode

The spiral mode depends heavily on the aircraft type or configuration and can be stable or
slightly unstable. The pole is located near the origin on the complex plane. Depending on
the stability (or instability) of the spiral mode, the aircraft performs one of the trajectories
depicted in Figure 4.12.
104 4 Dynamic Stability

Controlled aircraft

Stable Spiral
Radius increases steadily
Aircraft gradually returns to normal attitude, loss of altitude

Disturbance

Superposition
of a Dutch-roll

nt )
nsta 00km
e co of 1
e tim nce
larg ad is t a
ver
in, o
5m Instable Spiral Neutral Spiral
up to 1
( radius decreases steadily Roll angle remains
Tailspin (good to fly, without rudder support)

projection of the flight path

forms a spiral shape

Figure 4.12: Trajectories of the spiral mode depending on the stability [Huenecke]

Assuming a small and constant sideslip angle (β̇ = 0, β = 0), an approximation for the
rolling mode and spiral mode is obtained in Equation 4.56. With the first equation leading
to g · Φ + V0 · r = 0, the system of equations for the lateral motion simplifies to:

g 
0 −1
     
0 0 V0 ∆β Yξ Yζ  
 ṗ   Lβ Lp Lr 0 ·  p  +  Lξ Lζ  · ∆ξ . (4.56)
   
 =
 ṙ  Nβ Np Nr 0   r  Nξ Nζ  ∆ζ
Φ̇ 0 1 0 0 ∆Φ 0 0

The characteristic polynomial

    
g Lβ g
2
s − s Lp − Np − + (Nr Lβ − Nβ Lr ) = 0 (4.57)
V0 Nβ V0 Nβ

yields two poles, one for the roll mode and one for the spiral mode, both usually located on
the real axis of the complex plane. If the real part of the spiral mode’s pole is positive (right
half plane), the aircraft will perform a turning flight with decreasing radius and increasing
bank angle, leading into a tightening downward spiral (so-called spiral dive).
The Hurwitz stability criterion is used to identify the system stability: All coefficients of the
Hurwitz polynomial must exist and must be greater than zero. For example, in Equation 4.57
all coefficients exist, thus the first condition is complied. The second condition leads to:
4.4 Dynamic Lateral Stability 105

g
(Nr Lβ − Nβ Lr ) > 0 (4.58)
V0 Nβ

Since the yaw moment due to sideslip Nβ usually has a positive sign, it applies:

Nr Lβ > Nβ Lr (4.59)

The yaw damping Nr is negative, the roll moment due to yaw rate Lr is typically positive,
and the roll moment due to sideslip Lβ is typically negative. Therefore, the magnitude of
the roll moment due to sideslip Lβ must be great enough to comply with the condition of
Equation 4.59 and thus ensure the stability of the spiral mode. This can be achieved by an
appropriate dihedral angle of the wings (see Section 3.2.4.1).
The influence of the most important coefficients of the spiral mode and the Dutch Roll mode
(Cnβ and Clβ ) on the stability of given aircrafts is depicted in Figure 4.13 as an example.

Spiral Mode limit (E=0)

0,5
Cnβ ausgeführteF
given aircraft l ugzeuge
0,4
Unstable

0,3
stabil
Stable
Dutch Roll limit (R=0)
0,2 R=0,Ta umel grenze

0,1
instabil
Unstable

0
0 0,1 0,2 0,3 0,4
-Clβ

Figure 4.13: Stability areas for spiral mode and Dutch Roll mode

Rolling Mode

The essential state variables of the rolling mode are the roll rate p and the bank angle Φ.
The mode is induced by an aileron deflection ξ. The roll motion can therefore be described
in a good approximation by the following one-degree-of-freedom consideration:
       
ṗ Lp 0 p L
+ ξ · ∆ξ (4.60)
 
= ·
Φ̇ 1 0 ∆Φ 0

The resulting transient behavior of the roll rate due to aileron deflection corresponds to a
first order lag element with time constant TR = − L1p (P T1 behavior). The transient response
due to a step aileron deflection is depicted in Figure 4.14.
106 4 Dynamic Stability

L�
Lp
TR =RoRoll
l lzeitime
tkonstconstant
ant e

1
TR= Lp t

Figure 4.14: Roll time constant

4.4.3 Poles of the Lateral Motion

Figure 4.15 shows a comparison of the numerical exactly determined poles of the lateral
motion with the poles of the approximate solutions. In this example, an aircraft with delta
wing configuration was analyzed.

iω [1/s]

V=205
V=205m/ sm/s 4

160
160mm/s
/ s
3

Roll-Gier-
Dutch Roll Exact solution
ExakteL ösung
Schwingung
Mode
(S (s ) 120m/ s Approximation
Näherung
120 m/s
(s3,4) 3,4 3,4
2

78 m/s
78m / s
1

Rollbewegung(s)
Rolling Mode
1 (s1) Spiralbewegung(s)
Spiral Mode (s2) 2
V=78 (V=78
(V=78-2-05m
205
/ s)m/s)
V=78m/ sm/s

-2 -1 0 1
σ [1/s]
V=205
V=205mm/s
/s

Figure 4.15: Poles of the lateral motion in the complex plane

4.4 Dynamic Lateral Stability 107

4.4.4 Requirements on Handling Qualities

The already mentioned specifications (subsection 4.3.4) also define parameters for the lateral
motion of the aircraft. The pilot controls the lateral motion essentially by the bank angle Φ.
The main control surfaces are the ailerons, while the rudder is used induce the turning flight,
and to counter the appearing sideslip angles. According to [Brockhaus], the most important
evaluation criteria of the handling qualities of the lateral motion are:

• Roll response to aileron inputs, roll time constant, ratio of roll rate to input force;
• Contribution of the Dutch Roll mode in the roll motion;
• Stability of the spiral mode, possible coupling of rolling mode and spiral mode;
• Lateral acceleration at the pilot seat when entering and exiting turns;
• Roll control in crosswind.

Dutch Roll Mode

For the Dutch Roll mode the MIL-F 8785-C specification sets requirements for the damping
coefficient D and natural frequency ω0 as listed in Table 4.3:

Table 4.3: Requirements for the Dutch Roll mode according to MIL-F 8785-C
Handling Quality Level Aircraft Class ω0min Dmin
Level 1 I (light, agile a/c) 1,0 0,19
III (heavy, non-agile a/c) 0,4 0,19
Level 2 all 0,4 0,02
Level 3 all 0,4 0

Spiral Mode

If the stability of the lateral motion increases (Lβ and Lr ), the aircraft also becomes more
directionally stable and thus requires relatively large control surface deflections to initiate a
turn. Therefore, in order to to ensure good maneuverability, often a slightly unstable spiral
mode is allowed. The limitation is defined by the doubling time of the amplitude td , which
is listed in the following (for Level 1) depending on the flight phase:

• td ≥ 12s Category A and C (Takeoff, Approach, Landing, Combat Fight, ect.),

• td ≥ 20s Category B (Cruise, Climb, Descend, ect.).
108 4 Dynamic Stability

Rolling Mode

The decisive parameters for characterizing the handling qualities in the rolling motion are:

• Roll time constant TR

• Time to reach a certain bank angle
• Roll control sensitivity

The roll time constant essentially influences the roll response and thus the dynamic flight
behavior about the roll axis. It is specified in MIL-STD-1797 for Level 1 as:

• TR < 1, 0s for Aircraft Class I (small, agile a/c) and IV (highly agile fighter)
• TR < 1, 4s for Aircraft Class II (medium a/c) and III (heavy, non-agil a/c).
Bibliography 109

Bibliography

[Yechout] Yechout, T.R., Introduction to aircraft flight mechanics: performance,

static stability, dynamic stability, and classical feedback control, Reston,
2003.
[ICAO] International Civil Aviation Organization, Manual of the ICAO Standard
Atmosphere extendet to 32km, Montreal,1964.
[DIN5450] Norm - Atmosphäre DIN 5450, Blatt1, 1968.
[Huenecke] Huenecke, K., Flugtriebwerke - Ihre Technik und Funktion, Stuttgart,
1987.
[Etkin] Etkin, B. / Reid L.D., Dynamics of Flight - Stability and Control, New
York, 1996.
[Stevens] Stevens, B.L., / Lewis, F.L., Aircraft Control and Simulation, New Jersey,
2003.
[Hafer01] Bruening, G. / Hafer, X. / Sachs, G., Flugleistungen, Berlin, 1993.
[Hafer02] Hafer, X., Script zur Vorlesung "Flugmechanik I", Darmstadt.
[Hafer03] Hafer, X., Script zur Vorlesung "Flugmechanik II", Darmstadt.
[Brockhaus] Brockhaus, R., Alles, W., und Luckner, R., Flugregelung, Berlin, 2011.
[Klingauf01] Klingauf, U., Script zur Vorlesung "Grundlagen der Regelungstechnik",
Darmstadt, 2011.
[Klingauf02] Klingauf, U., Script zur Vorlesung "Flugmechanik I - Flugleistungen",
Darmstadt, 2011.