Lecture Notes

EG-296
Flight Mechanics

fundamentals of aircraft performance,

stability and control
recommended literature:

• John D. Anderson, Jr., Introduction to Flight, 5th edition, McGraw-Hill, 2005

• Thomas R. Yechout et al, Introduction to Aircraft Flight Mechanics, American

Institute of Aeronautics and Astronautics (AIAA), 2003

• Barnes W. McCormick, Aerodynamics, Aeronautics and Flight Mechanics, 2nd

edition, John Wiley & Sons, 1995

• [Link]

ii
Contents

1 Basic Aerodynamics 1
1.1 Physical Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 The Standard Atmosphere . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Mach Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 Reynolds Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5 Boundary Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.6 Bernoulli Equation for Incompressible Flow . . . . . . . . . . . . . . . 8
1.7 Pressure Distribution and Flow Separation . . . . . . . . . . . . . . . 9
1.8 Airfoil Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.9 High Lift Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.10 Finite Wings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.11 Aircraft Aerodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . 31
1.12 Remarks on Trans-/Supersonic Flight . . . . . . . . . . . . . . . . . . 39

2 Aircraft Performance 43
2.1 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.2 Brief Introduction to Aeroplane Propulsion . . . . . . . . . . . . . . . 46
2.3 Straight, Level and Unaccelerated Flight . . . . . . . . . . . . . . . . 55
2.4 Climbing Flight and Flight Envelope . . . . . . . . . . . . . . . . . . 63
2.5 Gliding Flight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
2.6 Take-Off and Landing . . . . . . . . . . . . . . . . . . . . . . . . . . 71
2.7 Range and Endurance . . . . . . . . . . . . . . . . . . . . . . . . . . 77
2.8 Turning Flight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
2.9 V–n Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
2.10 Energy Method for Accelerated Flight . . . . . . . . . . . . . . . . . 89

iii
3 Static Stability and Control 93
3.1 Static and Dynamic Stability . . . . . . . . . . . . . . . . . . . . . . 93
3.2 Aircraft Rotations and Control Surfaces . . . . . . . . . . . . . . . . 95
3.3 Longitudinal Static Stability . . . . . . . . . . . . . . . . . . . . . . . 97
3.4 Longitudinal Control . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
3.5 Stick-Free Longitudinal Static Stability . . . . . . . . . . . . . . . . . 113
3.6 Directional Static Stability and Control . . . . . . . . . . . . . . . . . 118
3.7 Lateral Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
3.8 Coupling Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

iv
Chapter 1

Basic Aerodynamics

1.1 Physical Units

Basic physical units:

1 in [inch] = 0.0254 m [metre]

1 ft [foot] = 0.3048 m
length 1 km [kilometre] = 1, 000 m
1 mi [mile] = 1, 609 m
1 nm [nautical mile] = 1, 852 m

mass 1 lb [pound] = 0.453592 kg [kilogram]

1 t [ton] = 1, 000 kg

time 1 h [hour] = 3, 600 s [seconds]

1
2 Basic Aerodynamics – wgd

Derived physical units:

1 N [Newton] = 1 kg m s−2
force 1 kp [kilopond] = 9.80665 kg m s−2 = 1 kg · g
1 kN [kilonewton] = 1, 000 kg m s−2

1 Pa [Pascal] = 1 N m−2 = 1 kg m−1 s−2

stress, 1 kPa [kilo Pascal] = 1, 000 Pa
pressure
1 MPa [mega Pascal] = 1, 000, 000 Pa
1 bar = 100, 000 Pa = 10 N cm−2

energy,
work 1 J [Joule] = 1 N m = 1 kg m2 s−2

1 W [Watt] = 1 J s−1 = 1 kg m2 s−3

1 kW [kilo Watt] = 1, 000 W
power 1 MW [mega Watt] = 1, 000, 000 W
1 bhp [brake horse power] = 745.7 W
1 PS [Pferdestärke] = 735.5 W = 75 kp m s−1

The quantity g = 9.80665 m s−2 denotes the gravitational acceleration of the earth.

For calculations involving angles we will need the relation

360 deg = 2 π rad .

wgd – Basic Aerodynamics 3

1.2 The Standard Atmosphere

Generally, in the atmosphere of the earth, the pressure, the temperature and the
density of the air are functions of altitude. The variations of these quantities with
time and location are negligible and mostly due to weather conditions.
The altitude of an aircraft is typically determined by measuring the free stream
air pressure and subsequent calculation of the altitude. In order to ensure that
the pilots of any aeroplanes at the same time in the same location have consistent
altimeters the international community introduced the notion of the standard at-
mosphere, which relates the pressure, the temperature and the density of the air
uniquely to the altitude and ignores local variations.
The standard atmosphere is composed of layers with constant temperature gra-
dients, i. e. d T / d h = const. in each layer:

71.5 K
troposphere : 0 − 11 km : T = 288.2 K − h (1.1)
11 km
stratosphere : 11 km − 20 km : T = 216.7 K (1.2)
K
20 km − : T = 216.7 K + h (1.3)
km
The equation for the perfect gas

p=ρRT with R specific gas constant, Rair = 287 J (kg K)−1 , (1.4)

the hydrostatic equation

dp
= −ρ g with gravity acceleration g = 9.81 m s−2 (1.5)
dh
and the boundary condition p = p0 = 101, 325 N m−2 at h = 0 suffice to determine
the pressure p. Using (1.1) and (1.4) in (1.5) and solving the differential equation
for p renders
!
g 11 km
71.5 h R 71.5 K
p = p0 1 − . (1.6)
288.2 11 km
Similarly, (1.2), (1.4) and (1.5) give
!
g h − 11 km
p = p11 exp − , (1.7)
R 216.7 K
4 Basic Aerodynamics – wgd

where p11 = 22, 621 N m−2 . The density ρ may then be calculated from (1.4). At
sea-level, the air density is obtained as ρ0 = 1.225 kg m−3 .
Tables of the standard atmosphere are provided in most text books on air flight.

We also note the dependency of the speed of sound a on the temperature, i. e.

q
a= γRT with γ ratio of specific heats, γair ≈ 1.4 . (1.8)

T [K]
200 220 240 260 280 300
30

25
T
stratosphere
20
h [km]

isothermal
15 region
ρ

10

p
5 troposphere

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4

ρ [kg m−3 ], p 105 [N m−2 ]

at sea-level at 8000 m
density ρ0 = 1.225 kg m−3 0.526 kg m−3
pressure p0 = 101, 325 N m−2 35, 651 N m−2
speed of sound 340 m s−1 = 1, 225 km h−1 308 m s−1 = 1, 109 km h−1
wgd – Basic Aerodynamics 5

1.3 Mach Number

The dimensionless Mach number is defined in every point of the fluid flow field as
v
Ma = , (1.9)
a
where v is the local flow velocity. In order to characterise the flight of an airplane
we introduce the freestream Mach number
v∞
Ma∞ = , (1.10)
a
where v∞ corresponds to the true airspeed or free stream velocity. Importantly,
for Ma∞ < 0.3, the air flow can be assumed to be incompressible, whereas, for
Ma∞ > 1.0, we expect the occurrence of pressure shock waves.

Characterisation of flow:

0 0.3 0.8 1.2 5.0 Ma∞

hypersonic
supersonic
transsonic
subsonic
subsonic (incompressible)

In transsonic flow, we typically have local areas of the flow field where the flow is
subsonic and others where the flow is supersonic.
6 Basic Aerodynamics – wgd

1.4 Reynolds Number

The dimensionless Reynolds number is defined as
ρ v∞ h
Re = , (1.11)
µ
where h is the characteristic length scale of the problem under consideration and µ
is the viscosity of the fluid medium. The Reynolds number describes the ratio of
the inertia forces to the viscous forces, both of which act on the fluid particles.
If the Reynolds number is small, then the fluid flow is typically laminar, i.e. the
streamlines are smooth and regular.
If the Reynolds number is large, then the fluid flow is likely to be turbulent in at
least some areas of the flow field, i.e. the streamlines break up and the fluid particles
move in a random irregular fashion. Turbulent fluid flow is always unsteady. In the
presence of turbulence, we are interested in average velocities and average properties
of the flow.
The critical Reynolds number Recrit , which is associated with the transition from
laminar to turbulent flow is extremely problem dependent. For the flow through a
circular pipe it is approximately 2300.
 wgd – Basic Aerodynamics 7

1.5 Boundary Layer

The flow field around a submerged body can be divided into the boundary layer and
the region outside the boundary layer. The boundary layer is a thin region adjacent
to the surface of the solid body, where the fluid flow is dominated by viscous forces
(friction). Outside the boundary layer, the effect of the fluid viscosity is negligible
(potential flow).
In a boundary layer we often observe transition from laminar to turbulent flow.
Laminar boundary layers are thinner than turbulent ones. Skin friction is smaller
in laminar boundary layers than in turbulent ones.

transition
laminar turbulent
v∞
average velocity
exact velocity

1111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
boundary layer thickness flow separation
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111111
8 Basic Aerodynamics – wgd

1.6 Bernoulli Equation for Incompressible Flow

Consider a stream tube of infinitesimal thickness in steady inviscid fluid flow:

v v + dv
streamline dA dA + d2A fixed control volume

in out
(p + dp)(dA + d2A)
p dA ds

By definition mass transport along the tube is constant: ṁ = ρ dA v = const.

Conservation of momentum in flow direction renders

0 = p dA − (p + dp)(dA + d2A) + (ρ dA v) v − (ρ dA v) (v + dv).

Neglect of higher order terms gives the Euler equation

0 = dp + ρ v dv . (1.12)

Using ρ = const. and integrating along a streamline between points A and B we

obtain for incompressible fluid flow
Z Z
B B 1
0 = dp + ρ v dv = pB − pA + ρ (vB2 − vA2 ) .
A A 2
The Bernoulli equation for incompressible flow follows as
1 2 1
pA + ρ vA = pB + ρ vB2 . (1.13)
2 2
wgd – Basic Aerodynamics 9

1.7 Pressure Distribution and Flow Separation

Consider the incompressible fluid flow around a sphere and assume attached flow.
The stream tubes passing over the surface of the sphere are longer and narrower
than the others. Those fluid particles, which travel from point A to B and then on
to C are first accelerated and then decelerated, i.e. v∞ = vA = vC < vB . It then
follows from (1.13), that these particles encounter first a drop and then an increase
of pressure, i.e. patmos = pA = pC > pB . At the same time, they have to work
against skin friction on the surface of the solid sphere. Between point B and C, both
the adverse pressure gradient and the skin friction slow down the particles.

Therefore, the attached flow pattern is only physically possible if the adverse
pressure gradient is sufficiently small. Otherwise, the particles are decelerated and
finally move in the opposite direction, which causes flow separation. Flow separation
creates a wake region dominated by vortices.

attached flow separated flow with vortices

A B C

resulting pressure drag

pressure distribution

The integral of the pressure over the surface of the solid sphere renders the force
which is exerted by the fluid flow on the submerged sphere. This force is denoted as
the pressure drag and must not be confused with the skin friction. For attached flow,
the pressure drag is zero. However, flow separation typically results in a pressure
drag which is significantly larger than skin friction drag. This is due to the failure
of the flow to recover the high pressure at the rear of the body.
 10 Basic Aerodynamics – wgd

The flow in turbulent boundary layers possesses more kinematic energy than
laminar flow, and therefore has a greater potential to overcome adverse pressure
gradients. Thus, turbulent boundary layers are often created deliberately (dimples
on golf ball) in order to repress separation and consequently reduce the pressure
drag (despite the increase in friction drag the overall drag reduction is significant).

laminar boundary layer turbulent boundary layer

separation point separation point
 wgd – Basic Aerodynamics 11

1.8 Airfoil Fundamentals

Geometry and Terminology

mean camber
chord line
max. camber
thickness trailing edge

angle of attack α

freestream
velocity v∞ chord c
leading edge

The angle of attack α is defined as the angle between the freestream velocity and
the chord line. The latter corresponds to the straight line connecting the leading
and the trailing edges.

The NACA four digit code describes the airfoil geometry:

NACA 2 4 12
maximum thickness in percent of chord
location of maximum camber in tenth of chord
maximum camber in percent of chord
National Advisory Committee for Aeronautics (precursor to NASA)

Note that, besides this classical family of airfoils, there are several other series.
Modern airfoils, such as the supercritical airfoil, are typically designed with com-
plex computer analysis tools for specific requirements and feature very individual
geometries.
12 Basic Aerodynamics – wgd

Generation of Lift and Origin of Drag

We consider the airflow past an airfoil. For simplicity, we assume incompressibility
of the air (Ma∞ < 0.3).

B
A

C D
stagnation point
flow separation

The streamlines passing over the top of the airfoil are longer, the space between
them is narrower, than at the bottom. The air particles flowing over the airfoil
evidently need to accelerate more than those passing underneath, i. e.

v A = v C = v ∞ < vD < vB . (1.14)

It then follows from (1.13) that

pA = pC = patmos > pD > pB . (1.15)

In the stagnation point, we have v = 0 and thus p > patmos . We obtain a qualitative
pressure distribution as displayed below.
pressure
wgd – Basic Aerodynamics 13

The integration of the pressure renders the lift force L (perpendicular to v∞ ), the
pressure drag force Dp (parallel to v∞ ) and the moment M . The moment M depends
on the choice of point A.

Dp A

For subsonic flows, the total drag force acting on the airfoil includes the pressure
drag Dp and the skin friction drag Df due to the air viscosity, i. e.

D = D p + Df . (1.16)

The effect of skin friction in lift direction is negligible. For transsonic and supersonic
flows, another drag component is observed, which arises from the formation of shock
waves, see Section 1.12.
For a given airfoil geometry, lift and drag forces depend on the angle of attack
α and the velocity v∞ . Note that there exists a point A, such that M is the same
for all angles of attack α. This point is called the aerodynamic centre of the airfoil
and is of crucial importance in the design of aeroplanes.

Note also that lift can be generated with bodies of very primitive geometries as
shown below.
14 Basic Aerodynamics – wgd

Aerodynamic Coefficients
For the characterisation of airfoil geometries, we introduce coefficients which are
independent of the airfoil size and the velocity v∞

L
lift: Cl = (1.17)
v∞ q∞ c b
c

D
drag: Cd = (1.18)
q∞ c b
α b

M
moment: Cm = (1.19)
q∞ c 2 b

The dynamic pressure q∞ in the stagnation point is obtained from (1.13)

1 2 1 2
patmos + 0 + ρ v = patmos + q∞ + 0 ⇒ q∞ = ρv (1.20)
2 ∞ 2 ∞
The aerodynamic coefficients Cl , Cd , Cm depend primarily on the angle of attack α.
In the range of very large velocities, they also depend on the Reynolds number Re
and the Mach number Ma.

For the design and study of airfoils, it is useful to introduce the drag coefficients
Cd,f and Cd,p , which relate to friction drag and pressure drag, respectively, i. e.

Df Dp
Cd,f = , Cd,p = . (1.21)
q∞ c b q∞ c b

Thus, with (1.16), we have for subsonic flow

Cd = Cd,f + Cd,p . (1.22)

wgd – Basic Aerodynamics 15

Airfoil Lift Curve

By measuring the lift force L for different angles of attack α (wind tunnel tests), we
can obtain the characteristic lift curve of a specific airfoil.

increasing Re
Cl
cambered airfoil

symmetric airfoil
Cl α

1
α
αL=0
αstall

For reasonable angles of attack, we observe a linear relation between Cl and α. For
symmetric airfoils (zero camber), we clearly have zero lift at α = 0. The derivative
of Cl with respect to α is denoted as Clα .

A good approximation of Cl (α) may be written as

Cl (α) = Cl α · (α − αL=0 ) . (1.23)

16 Basic Aerodynamics – wgd

The sudden drop of the lift coefficient at αstall is due to flow separation before the
trailing edge. This destroys the desired pressure distribution over the airfoil surface
and may lead to the complete loss of lift: We say “the airfoil stalls”.
An increase of the Reynolds number extends the turbulent boundary layer and
thereby represses undesired flow separation. This typically leads to larger Cl max .

“stall”
flow separation
wgd – Basic Aerodynamics 17

Airfoil Drag Polar

Similarly to the lift, we can measure the drag force for different angles of attack in a
wind tunnel experiment. The graph, which displays the drag coefficient Cd against
the lift coefficient Cl , is known as the drag polar of the airfoil.

symmetric airfoil

Cd

increasing Re

cambered airfoil

Cl

For interpretation, recall the linear relation between Cl and α. The pressure drag
or “separation drag” Cd,p is closely related to the airfoil thickness “seen” by the
coming flow and denoted by h in the diagram below. Therefore, Cd,p becomes larger
as α (and Cl ) increase. The skin friction drag Cd,f is fairly independent of α and
dominates the drag associated with small α (small Cl ).
Higher Reynolds numbers suppress flow separation and allow to achieve larger
lift coefficients with less pressure drag despite the increase in skin friction drag.

h
α
18 Basic Aerodynamics – wgd

Drag and Lift Coefficients of NACA 2412

wgd – Basic Aerodynamics 19
20 Basic Aerodynamics – wgd

1.9 High Lift Systems

For certain manoeuvres, such as take-off and landing, very high lift systems are de-
sirable. However, we have seen in the drag polar, that high lift is typically associated
with large drag forces, which need to be overcome by engine thrust. Therefore, in
order not to jeopardise the cruising performance of the aeroplane, we desire high
lift systems that can be activated when needed, e. g. to allow short ground roll at
take-off. For that purpose, a variety of flaps have been developed.
wgd – Basic Aerodynamics 21

For most flap types, the increase of the lift coefficient is due to the increased camber
of the airfoil and the increased angle of attack with respect to the new chord line c′ .
The flap deflection is typically denoted as δf .

max camber
(max camber)’

α′ α c
v∞
c′
δf

plain flaps:
cf
• The optimum flap chord ratio is approximately = 0.25.
c
• The optimum flap angle is approximately 60 deg.

• The maximum achievable increment in Clmax is approximately 0.9.

split flaps:
cf
• The optimum flap chord ratio varies with the airfoil thickness: 0.3 < < 0.4.
c
• The optimum flap angle is approximately 70 deg.

• The maximum achievable increment in Clmax is approximately 0.9.

slotted flaps:
cf
• The optimum flap chord ratio is approximately = 0.3.
c
• The optimum flap angle is approximately 40 deg for single slots and 70 deg
for double-slotted flaps.

• The maximum achievable increment in Cl max is approximately 1.5 for single

slots and 1.9 for double-slotted flaps.
22 Basic Aerodynamics – wgd

powered systems:

• Directed air jets blown through nozzles near the trailing edge can act as “vir-
tual” flaps.

• Distributed suction over the wing surface may be used to delay flow separation.

• Powered high lift systems have the capability to provide values of Cl max large
enough to achieve extremely short takeoff ground roll of the aircraft.

Cl δf = 40 deg

δf = 20 deg

δf = 0

The effect of standard flaps on the slope Cl α of the lift curve is typically negligible.
Introducing the flap effectiveness factor τ , which can be determined from wind
tunnel tests, the lift coefficient of an airfoil with a flap may be written as

Cl = Clα (α − αL=0 + τ δf ) . (1.24)

The dependency of the flap effectiveness on the flap deflection δf is beyond the scope
of this course.
 wgd – Basic Aerodynamics 23

(1) airfoil only

(2) plain flap
(3) split flap
(4) leading edge slat
(5) single-slotted flap
Cl max
(6) double-slotted flap
(7) double-slotted flap
combined with
leading edge slat
(8) like (7) with
boundary layer suction
 24 Basic Aerodynamics – wgd

1.10 Finite Wings

Geometry and Terminology

leading edge

ΛLE

area S Λc/2 mid-chord line

tip chord ct root chord cr

span b

trailing edge

b2
aspect ratio AR = (1.25)
S
ct
taper ratio λ = (1.26)
cr

The angles ΛLE and Λc/2 or Λc/4 are known as sweep angles.

A wing with a large span b and small chord lengths has a large aspect ratio AR.
The imaginary wing with AR → ∞ is known as the “infinite wing”.

A rectangular wing “has no taper” (λ = 1); a delta wing has a taper ratio λ = 0.

Lift, drag and moment coefficients for the finite wing are denoted as CL , CD , CM to
avoid confusion with airfoil data Cl , Cd , Cm .
wgd – Basic Aerodynamics 25

Wing Tip Vortices and Induced Angle of Attack

A wing generates lift by creating a pressure difference between the upper and lower
surfaces. Wing tip vortices are generated as the high-pressure air on the lower wing
surface seeks the relatively lower pressure on the upper surface. These vortices
induce a downward component of velocity called downwash w.

low pressure

high pressure
front view of wing wing tip vortex

v∞

downwash w
wing tip vortices

chor
d line
αef f
α

αi
v∞
w vlocal

The velocity vlocal “seen” by the wing consists of the superposition of the velocity
v∞ and the downwash w. It follows that the effective angle of attack αeff is related
to α and to the induced angle of attack αi by

αeff = α − αi . (1.27)
26 Basic Aerodynamics – wgd

Induced Drag
According to (1.27), the effective angle of attack αeff between the velocity vlocal and
the chord line of the wing section is smaller than α. Relating the direction of the
lift force to the effective freestream velocity vlocal corresponds to a rearward rotation
of the lift force by αi . However, we would like to continue to define the directions
of lift and drag forces L and D with respect to the original freestream velocity v∞ .
Thus, we see that the tilted lift force L′ contributes a certain component of drag.
This drag is known as the induced drag Di .

D1
R1
Di
D ′

The vectors L1 , D1 , R1 represent,

respectively, lift, drag and total
L1
aerodynamic force of the infinite
L′
wing (no downwash).

αi L
R

chor
d line

αef f
α D

αi

v∞
w vlocal

The angle αi is typically very small. Generally, we also have L >> D. Thus,

L′ = L , Di = αi L . (1.28)
wgd – Basic Aerodynamics 27

Under certain assumptions, such as incompressible flow, it can be derived that

CL
αi = , (1.29)
π e AR

where e is the geometry dependent span efficiency factor. Typically, 0.9 < e < 1;
for an elliptic wing e = 1. Note that αi is obtained in radians.

Equation (1.29) reflects the following correlations:

• A large lift coefficient CL implies a large pressure difference between lower and
upper wing surfaces. This causes a strong downwash, which, turn produces a
large αi .

• The smaller the aspect ratio AR, i. e. the shorter the wing, the more dom-
inant are the wing vortices. This causes a strong downwash and leads to a
large αi .

Di
Defining the induced drag coefficient as CD,i = (1.30)
q∞ S

and using (1.28), (1.29) and (1.17) for the finite wing, we obtain
CL L CL2
CD,i = = . (1.31)
π e AR q∞ S π e AR

It follows for the total drag CD of a finite wing at subsonic speeds that

CL2
CD = Cd + . (1.32)
π e AR

The airfoil section drag or profile drag Cd accounts for skin friction and pressure
drag and is given by (1.22). Recall the dependency of Cd on the lift coefficient Cl
of the airfoil. However, the variation of Cd is typically significantly smaller than the
induced drag coefficient CD,i .

The most effective remedy for induced drag is the design of large aspect ratio wings.
Also the addition of vertical surfaces at the wing tips (“winglets”) reduces the
strength of the wing tip vortices and thereby decreases the induced drag.

Page 29 shows a qualitative representation of the drag polar of a finite wing.

28 Basic Aerodynamics – wgd

Lift of the Finite Wing

In Equation (1.28) we have L = L′ . Furthermore, we can write

L = q∞ S CL (α) , L′ = q∞ S Cl (αeff ) . (1.33)

Thus, we obtain
CL (α) = Cl (αeff ) . (1.34)

Using (1.23), (1.27) and (1.29), we can deduce

CL (α) = Cl (α − αi )

= Cl α · (α − αi − αL=0 )
!
CL (α)
= Clα · α − − αL=0 . (1.35)
π e AR

Rearranging (1.35) gives

Cl α
CL (α) = CLα · (α − αL=0 ) with CLα = . (1.36)
Cl α
1+
π e AR

For α = αL=0 no lift is generated, thus there are no wing tip vortices and conse-
quently, αL=0 is identical for the finite wing and the airfoil (“infinite wing”). From
(1.36) we understand that CLα < Clα .

It follows that CL (α) < Cl (α) , which is consistent with the original starting point
of the reduced effective angle of attack.
wgd – Basic Aerodynamics 29

Drag Polar and Lift Curve

decreasing AR
CD

finite wings

infinite wing

CL

infinite wing

Cl finite wing

CL

Clα
CLα

1 1

α
αL=0
30 Basic Aerodynamics – wgd

Estimation of CL and CD for the Finite Wing

Given the lift curve and the drag polar of the NACA 2412 airfoil on pages 18 and
19 as well as the wing geometry properties AR = 10 and e = 0.95, estimate the lift
and drag coefficients CL and CD of the corresponding finite wing for the angle of
attack α = 4 deg with Re < 3 · 104 .

Solution:

From the lift curve of the NACA 2412 airfoil we deduce

αL=0 = −2 deg and Cl α = 0.1 deg−1 .

From (1.36) we get

0.1 deg−1
CLα = = 0.839 deg−1
0.1 180
π
1+
π 0.95 10
and
CL = 0.839 (4 − (−2)) = 0.503 .
With (1.34)
Cl (αeff ) = CL (α) = 0.503 .

Thus, we get from the drag polar of the NACA 2412 airfoil that

Cd (Cl = 0.503) = 0.0065 .

Finally, we use (1.32) to obtain

0.5032
CD,i = = 0.00849
π 0.95 10

CD = 0.0065 + 0.00849 = 0.0150 .

Note that even for this moderate angle of attack the induced drag coefficient CD,i is
larger than the profile drag coefficient Cd .
wgd – Basic Aerodynamics 31

1.11 Aircraft Aerodynamics

Preliminaries
The entire aeroplane consists of the wings, the tail, the fuselage and possibly other
components such as engine nacelles, stores, weapons.

S
We introduce the wing reference area S as illustrated.

The coefficients CL and CD for the total aeroplane must not be confused with those
for the wing. Let L and D denote the lift and drag forces acting on the entire
aeroplane, then
L D
CL = , CD = . (1.37)
q∞ S q∞ S

Using (1.20), (1.37) and (1.37), we can solve for the freestream velocity to obtain
s
2L
v∞ = . (1.38)
ρ S CL

The slowest speed an aeroplane can fly in straight, level, unaccelerated flight is called
the stall speed vstall . In straight, level flight, we have L = W , where W is the weight
of the aeroplane. The minimum of v∞ is achieved with maximum CL . Therefore,
s
2W
vstall = . (1.39)
ρ S CL max
32 Basic Aerodynamics – wgd

Aircraft Drag Polar

The drag polar of an aeroplane can be obtained from wind-tunnel experiments
or flight tests. It can generally be approximated quite accurately by a quadratic
parabola. Therefore, there exists a factor ẽ, such that the drag polar for an aircraft
can be represented as

(CL − CL min(drag) )2
CD = CD min + . (1.40)
π ẽ AR

Equation (1.40) accounts for the fact that CD

the minimum drag CDmin of a well designed
aeroplane is associated with a small positive
lift CL min(drag) > 0; i. e. zero lift is usually
associated with a nose-down orientation of CD min
the aircraft, which does not render minimum
CL
drag.
CL min(drag)

A convenient approximation of the aircraft

drag polar (1.40), sufficiently accurate for
most calculations, is given by CD

CL2
CD = CD0 + , (1.41)
π e AR
CD 0
where e denotes the Oswald efficiency factor
and must not be confused with the span ef- CL
ficiency factor. Always e ≤ 1 and for many
aircraft, e ≈ 0.8.
wgd – Basic Aerodynamics 33

In Equation (1.41), the constant CD 0 is called zero lift drag coefficient or parasite
drag and includes zero lift pressure drag, skin friction and interference drag. In-
terference drag is generated when more than one body (e.g. stores on a wing) are
placed in the same flow field. The coefficient CD 0 is a very important aerodynamic
characteristic of an aeroplane (see diagram on page 36).

The second term in (1.41) represents the drag due to lift. Thus, it includes the
induced drag on all lifting surfaces (mainly wings and horizontal tail) and the incre-
ment of the pressure drag when lift is generated. It is denoted as aircraft induced
drag CD,i , but must not be confused with the “pure” induced drag of the wing in
Equation (1.31)
CL2
CD,i = . (1.42)
π e AR
34 Basic Aerodynamics – wgd

Total Aircraft Drag Force

We are now in a position to express the total drag force acting on an aeroplane in
straight level unaccelerated flight as a function of the freestream velocity v∞ . Using
(1.37) and (1.41), we can deduce
!
CL2 L2
D = CD q∞ S = CD 0 + q∞ S = CD 0 S q∞ + . (1.43)
π e AR π e AR S q∞

With (1.20) and L = W from straight level unaccelerated flight, it follows that

!
2W2
 
1 2 1
D = CD 0 ρ S v∞ + 2
. (1.44)
2 π e AR ρ S v∞
| {z } | {z }
parasite drag induced drag

Equation (1.44) describes how changes in altitude (ρ), load (W ) and configuration
(CD0 , S, e, AR) affect an aircraft’s total drag.

Note that different points on the D-v∞ diagram are associated with different angles
of attack. As v∞ increases, α decreases.

D stall limit

total drag

parasite drag

induced drag

vstall v∞
wgd – Basic Aerodynamics 35

At the drag minimum we have

dD 4W2 1
= CD0 ρ S v∞ − 3
= 0. (1.45)
d v∞ π e AR ρ S v∞

Solving for CD0 and eliminating v∞ with (1.20), (1.37) and L = W gives

4W2 1 W2 1 CL2
CD 0 = = = . (1.46)
π e AR (ρ S)2 v∞
4 π e AR S 2 q∞
2 π e AR

Recalling (1.42) it follows that in the drag minimum the induced drag is equal to
the parasite drag
D ⇒ M IN → CD 0 = CD,i . (1.47)

With L = W we may write for straight, level and unaccelerated flight

D CD W
D = W = W = . (1.48)
L CL CL /CD

Thus, the drag is inverse proportional to the lift over drag ratio. Consequently, the
minimum of drag must coincide with the maximum of CL /CD , which explains the
fundamental importance of (CL /CD )max (see diagram on page 36). To verify this,
we express the ratio CL /CD as a function of CL , i. e.

CL CL
= , (1.49)
CD CL2
CD0 +
π e AR

and equate the derivative with zero to obtain

2 CL
d (CL /CD ) 1 CL
= − π e AR ! = 0 , (1.50)
d CL CL2 CL2
2
CD 0 + CD 0 +
π e AR π e AR
which can be simplified to give the same result as (1.47), i. e.

L/D ⇒ M AX → CD0 = CD,i . (1.51)

Thus, when the total drag of an aircraft is minimal, the parasite and the induced
drag are equal and the lift over drag ratio is maximal.
36 Basic Aerodynamics – wgd

Historical Developments

17,18

16

16

17,18
wgd – Basic Aerodynamics 37

(1) SPAD XIII (2) Fokker D-VII (3) Curtiss JN-4

(4) Spirit of St. Louis (5) Lockheed Vega

(6) Douglas DC-3 (7) Boeing B-17

(8)Boeing B-29 (9)Mustang P-51 (10) Lockheed P-80

38 Basic Aerodynamics – wgd

(12) Lockheed F-104 (13) McDonnell F-4

(11) North American F-86
Starfighter Phantom II

(15) General Dynamics

(14) Boeing B-52 (16) Cessna 172 Skyhawk
F-111

(17) Boeing B-747 (18) Airbus A-380

wgd – Basic Aerodynamics 39

1.12 Remarks on Trans-/Supersonic Flight

Airfoil Critical Mach Number and Wave Drag
Consider the two dimensional flow around an airfoil. The critical Mach number Macr
is defined as the smallest Mach number Ma∞ , for which there is at least one point
of the flow field where the local Mach number is Ma = 1. Due to the larger velocity
differences in the flow field, the critical Mach number is larger for thick than for
thin airfoils.

The drag divergence Mach number Madrag div of an airfoil is the smallest Mach num-
ber Ma∞ , for which a shock wave is generated in the flow field. The sudden increase
of the drag as Ma∞ is increased beyond Madrag div is denoted as the wave drag. This
is partly due to the dissipative nature of the shock wave itself and partly due to
a significant increase of separation drag caused by the pressure discontinuity along
the shock wave. The shock wave pattern changes as the velocity is increased even
further. Shock waves also significantly affect the generation of lift.

Note that generally Macr < Madrag div < 1 . (1.52)

For Ma∞ > Madrag div , the drag coefficient for an airfoil becomes

Cd = Cd,f + Cd,p + Cd,wave . (1.53)

For the entire aeroplane, the effect of wave drag may be included in (1.41). Thus,
the discussions in the next chapters do not generally require to distinguish subsonic
or supersonic flight.
40 Basic Aerodynamics – wgd

Cd

Cd,f +Cd,p

Ma∞

Macr 1
Madrag div

Ma < 1
v∞

Ma∞ < Macr

Ma >1 Ma < 1
v∞
Macr < Ma∞ < Madrag div

shock wave

separated flow
Ma >1
v∞

Ma∞ > Madrag div

Ma >1
Ma < 1

v∞

Ma∞ > 1

shock wave
wgd – Basic Aerodynamics 41

Sweep Angle
By sweeping the wings of a subsonic aircraft, the Mach numbers Macr and Madrag div
are increased, i. e. the onset of wave drag is delayed to higher velocities.

We typically obtain

Maairfoil
cr
Maairfoil
cr < Maswept
cr
wing
< . (1.54)
cos Λ

v∞
Λ
v∞ cos Λ

However, note that sweeping the wing adversely affects the generation of lift. Due
to the different aerodynamic loading, wing sweep also requires stiffer and stronger
wing structures.

A reliable investigation of the behaviour of swept wings needs to be based on wind

tunnel experiments or on sophisticated computer analysis.
42 Basic Aerodynamics – wgd

Extreme Sweep and Delta Wings for Supersonic Speeds

In supersonic flight, the wave drag is reduced significantly if the sweep angle Λ is
chosen smaller than the Mach cone angle µ. The velocity component normal to the
leading edge is then subsonic. This tends to reduce the intensity and the complexity
of the shock wave pattern for the aircraft.

µ Mach cone

v∞ t
at

delta wing

drag coefficient
CD for a cone

Ma
 Chapter 2

Aircraft Performance

2.1 Preliminaries
Some Geometry and Terminology

th r
ust

airc
raft
refe
renc
e lin
freest e φ L
ream
flight v∞
direct α T
ion
Θ D
γ
horizon

pitch angle Θ
thrust angle φ
angle of attack α
flight path angle γ
engine thrust T

43
44 Aircraft Performance – wgd

Basic Equations of Motion

For simplicity, we regard the aircraft as a point mass m with two degrees of freedom
(horizontal and vertical translation).

Using Newton’s second law “force = mass × acceleration” and performing the sum-
mation of the forces parallel and perpendicular to the flight path, we obtain
d v∞
m = T cos(α + φ) − D − W sin γ (2.1)
dt
2
v∞
m = − T sin(α + φ) − L + W cos γ , (2.2)
r
d v∞ v2
where is the acceleration of the aircraft in flight direction and ∞ is the
dt r
acceleration due to the local curvature of the flight path (centripetal acceleration).

v∞

fli
r gh
tp
at
h
wgd – Aircraft Performance 45

Load Factor
We introduce the load factor n as the ratio of lift force over aeroplane weight, i. e.
L
n = . (2.3)
W
We will see that in different flight manoeuvres the load factor takes different values.

At this stage, it is convenient to rewrite the expression (1.44) such that it holds, at
least approximately, for all flight manoeuvres. To this end, we use (2.3) in (1.43).
1 2
With (1.20), i. e. q∞ = ρ v∞ , we then obtain
2
!
2 (n W )2
 
1 2 1
D = CD 0 ρ S v∞ + 2
. (2.4)
2 π e AR ρ S v∞

We may also express the velocity v∞ in terms of n W rather than L. The expression
(1.38) may then be rewritten as
s
2nW
v∞ = . (2.5)
ρ S CL
46 Aircraft Performance – wgd

2.2 Brief Introduction to Aeroplane Propulsion

Before we enter into the detailed discussion of aircraft performance, we ought to
have a basic understanding of aircraft propulsion systems. Clearly, we can expect
the qualitative and quantitative properties of the aircraft engines to have substantial
impact on the overall performance of the aeroplane. However, the detailed discussion
of aircraft engines is beyond the scope of this course.

The most commonly known types of propulsion systems include

• propeller with piston engine,

• turbojet engine,

• turbofan engine,

• turboprop engine,

• ramjet,

• rocket engine.

On the following pages we present some basic information on the piston propeller
and the gas turbine engines. The ramjet may be seen as a primitive turbojet engine
without any rotating components. The incoming air is compressed solely by the
high freestream velocity. Therefore, ramjets only work at very large speeds and are
rather impractical. Rocket engines were initially mounted on high speed aeroplanes,
but were quickly replaced by the more efficient jet engines.
wgd – Aircraft Performance 47

Propeller with Piston Engine

Until about 1940, the piston engine represented the only practical means of aero-
plane propulsion. It was sufficiently powerful and, at the same time, light enough
to be mounted on an aeroplane. It was continuously further developed, such that
several engine types are now available. The main differences include the different
types of cooling systems and different piston arrangements. However, as shown in
the diagram, the specific fuel consumption SFC is very similar for most engines. It
decreases slightly with the engine power. SFC denotes the weight (or mass!) of fuel
burnt per time unit to generate a certain amount of power. For piston engines, SFC
is nearly independent of the throttle setting, i. e.
− dWfuel /dt
SFC = = const. (2.6)
P
In English units SFC (also known as brake specific fuel consumption BSFC) is
measured in [lb/(bhp h)]. In the SI system, we typically use [N/(kW h)]. With
g = 9.81 m s−2 , we have
lb 0.453592 · 9.81 N N
1 ·g = = 5.97 . (2.7)
bhp h 0.7457 kW h kW h

A valid guess for any piston engine is SFC = 0.5 lb (bhp h)−1 , which corresponds
to approximately
N
SFC = 3 . (2.8)
kW h

reciprocating
piston engine
48 Aircraft Performance – wgd

Mounted to the shaft of the piston engine is the propeller, which consists of a
number of blades. These blades have airfoil-like sections and are generally pitched
significantly. The pitch angle β denotes the angle between the section chord line and
the plane of rotation. The pitch of modern propellers changes with the radius. Many
propellers have variable pitch (pilot-controlled pitch), as opposed to fixed pitch.
The airflow around the propeller blades is extremely complicated, since it involves
induced flow velocities (similar to the finite wing) as well as rotational flow and
substantial interference phenomena. It is therefore evident that the engine power P
can not be transformed into propulsive power available PA without loss. We may
write
PA = η P , (2.9)
where η is the propeller efficiency factor. It can be shown with experiments that η
depends on the propeller geometry (parameter β for propellers with variable pitch)
and on the so-called dimensionless advance ratio J, which is defined as
v∞
J = , (2.10)
nD
where n is the number of rotations per time unit and D is the propeller diameter.
The advance ratio corresponds to the ratio of the velocity with which the propeller
advances through the air over the tangential velocity of the blade tips in the plane
of rotation. The diagram below shows the typical efficiency for a propeller with
variable pitch.
wgd – Aircraft Performance 49

For the discussion of airplane performance we need the dependency of power avail-
able PA on the velocity v∞ . Assuming variable pitch which is always adjusted to the
aeroplane speed v∞ , we obtain the qualitative diagram displayed below. For rough
estimates of the airplane performance, we may approximate the power available di-
agram by the straight line PA = const. as shown. Note the drop of PA as the Mach
number Ma∞ = v∞ /a leaves the incompressibility range.

PA

v∞

For piston propellers, the dependency of power available on altitude is approximated

fairly well by assuming that PA is proportional to the air density, i. e.
ρalt
PA, alt = PA, sea , (2.11)
ρsea
where the subscripts “sea” and “alt” denote quantities at sea level and at some
higher altitude, respectively. We conclude from ρalt /ρsea < 1 that power available
decreases as the altitude is increased.
50 Aircraft Performance – wgd

Gas Turbine Engines

In 1903, the first gas turbine was constructed that was able to produce more power
than needed to run its own components. However, the first substantial working
gas turbine engine was constructed as late as 1937 and the first aeroplane with a
turbojet engine flew in 1939. The first operational jet fighter began service in 1942.
The first jet propelled passenger flight took place in 1952 and may be seen as the
beginning of mass travelling. Today most commercial and military aeroplanes em-
ploy gas turbine engines.
We distinguish the turbojet, the turbofan and the turboprop engines. The most
characteristic feature of a gas turbine is the continuous nature of the thermody-
namic processes in the gas turbine as opposed to the reciprocating engine.

turbojet engine: The three basic components of the turbojet engine are the com-
pressor, the burner and the gas turbine.

1 2 3 4 5

1 inlet diffuser
2 compressor
3 burner
4 turbine
5 exhaust nozzle

The compressor consists of a combination of rotating and stationary blades. In the

burner, the fuel is injected into the compressed air and the mixture is burnt, whereby
heat is generated and the volume of the gas mixture expands significantly. Typically
the ratio of air to fuel by weight is about 60/1. However, only approximately 25% of
wgd – Aircraft Performance 51

the air is used to support combustion. The remainder bypasses the fuel nozzles and
mixes downstream of the burner to cool the hot gases before they enter the turbine.
The mixed air, still very hot (about 1100◦ C), passes through the turbine, which is
composed of rotating and stationary blades. The turbine extracts energy from the
moving gases, which is used to drive the compressor. In fact, nearly 75% of the
combustion energy is required by the compressor. The remaining 25% represent the
kinetic energy of the exhaust, which provides the thrust. The number of rotations
per minute of the turbine wheel is generally in the order of 10,000 rpm. Turbojet
engines can be designed to work at speeds far beyond the sound barrier.
A turbojet engine may be equipped with an afterburner, which relies on injecting
fuel into the extended exhaust nozzle (afterburner duct). Since only 25% or so of
the air is used to support combustion in the burner, there is sufficient oxygen in the
turbine exhaust mixture to support additional burning. The additional heat further
expands the exhaust and thereby increases the exhaust velocity. Afterburning can
more than double the thrust of a gas turbine engine, but at a proportionately greater
increase in fuel consumption.

turboprop engine: This engine may be regarded as a turbojet engine, which is

equipped with a modified gas turbine that extracts nearly all the energy from the
moving gases, leaving only a small residual thrust. This turbine drives not only the
compressor, but also provides the shaft power for the propeller. Alternatively the
compressor and the propeller may be driven by two sequential gas turbines as shown
below.

Due to the fact that turboprops basically represent propellers powered by a gas
engine, their use is limited to speeds far below the sound barrier. However, the de-
velopment of high speed turboprops is ongoing and they are already being employed
for Mach numbers much larger than the incompressibility limit of Ma∞ = 0.3.
52 Aircraft Performance – wgd

Like turbojets, turboprops may be equipped with afterburners. Note that turboshaft
engines, which work similarly to turboprops, are commonly used to drive the rotors
of modern helicopters.

turbofan engine: The turbofan engine may be regarded as a compromise between

the turbojet engine and the turboprop. The turbines drive the compressor and a
fan, but they do not absorb as much energy as in the turboprop, such that the
exhaust jet provides a major part of the thrust. The fan may be regarded as a
propeller with a large number of blades rotating inside a duct, which suppresses
radial airflow. Such a fan is capable of providing thrust at velocities in the whole
range of subsonic speeds. The ratio by weight of the air that passes through the
fan (secondary flow) to the air that passes through the gas engine (primary flow)
is called the bypass ratio BPR. Early turbofan engines had bypass ratios of around
1/1; the latest engines have ratios of up to 10/1. Turbofan engines are generally
quieter and much lighter and consume less fuel than turbojets. Since the 1970ies,
most commercial aircraft have been equipped with large BPR turbofans, allowing
for fuel efficient subsonic flight with most of the thrust being provided by the fan,
whereas a typical configuration for military aircraft is the low BPR turbofan with
afterburners.

high BPR turbofan with short ducts

low BPR turbofan with long ducts

wgd – Aircraft Performance 53

For turbojets and turbofans the fuel consumption is expressed in terms of thrust
specific fuel consumption TSFC. This expression denotes the weight of fuel burnt per
time unit to generate a certain amount of constant thrust. In English units, TSFC
is typically given in terms of mass rather than forces, i. e. [lb/(h lb)], whereas in
SI units we have [N/(h N)]. However, due to mass (or force) cancelling out of the
expression, the dimension of TSFC is always 1/time, and there is no need for unit
conversion. Independently of the throttle setting, the fuel burnt per unit of time
and thrust is nearly constant for turbojets and may be approximated as constant
for turbofans, i. e.
− dWfuel /dt
TSFC = = const. (2.12)
T
The diagram below shows some representative values for TSFC. We note that TSFC
is lower for very strong engines than it is for smaller ones.
54 Aircraft Performance – wgd

For the study of airplane performance, we need to know the dependency of thrust
available TA on the flight velocity (v∞ ) and on the altitude (ρ). At constant altitude,
the thrust available of turbojets and turbofans typically decreases slightly as the
velocity increases. Depending on the air pressure, this decrease may amount to
30% or so. For rough estimates and for the qualitative understanding of the flight
mechanics of a specific aeroplane, it is sufficiently accurate to approximate the thrust
available by TA = const. at a given altitude. As the altitude is changed, TA is
proportional to the change of the air density, i. e.
ρalt
TA, alt = TA, sea , (2.13)
ρsea
where the subscripts “sea” and “alt” denote quantities at sea level and at some
higher altitude, respectively. We conclude from ρalt /ρsea < 1 that thrust available
decreases as the altitude is increased.

TA

v∞
wgd – Aircraft Performance 55

2.3 Straight, Level and Unaccelerated Flight

Thrust Required
We will see that the study of the simplest of flight situations, i. e. straight, level
and unaccelerated flight, often abbreviated by “SLUF”, allows to draw a variety of
conclusions about the performance of the specific aeroplane under consideration.
d v∞ v2
By definition, we have = ∞ = 0, and γ=0. (2.14)
dt r

Furthermore, we assume that φ + α is small; thus

sin(φ + α) ≈ 0 and cos(φ + α) ≈ 1 . (2.15)

Hence, (2.1) and (2.2) reduce to

T = D, L = W . (2.16)

We conclude that the lift required to sustain SLUF is equal to the weight of the
aeroplane. With (2.3) it follows that the load factor is unity, i. e. n = 1. The
thrust required to sustain SLUF is equal to the drag. Henceforth, we denote the
thrust required as TR .
W
With (2.16) and (1.48) it follows that TR = . (2.17)
CL /CD
The thrust required is proportional to the inverse of the lift over drag ratio. The min-
imum thrust required is associated with the maximum lift over drag ratio. Equation
(2.17) demonstrates the fundamental importance of the aircraft drag polar, i. e. the
ratio CL /CD , for the choice of the aircraft weight and the aircraft engine.

Recalling (2.4) with n = 1, we note that TR depends on the aircraft weight (W ),

the aircraft configuration (CD 0 , S, e, AR), the flight altitude (ρ) and the flight
velocity (v∞ ), i. e.
!
2W2
 
1 2 1
TR = D = CD 0 ρ S v∞ + 2
. (2.18)
2 π e AR ρ S v∞
| {z } | {z }
parasite drag induced drag
56 Aircraft Performance – wgd

Power Required and Power Available

Typically, thrust required and thrust available are compared in terms of power,
“power = force × velocity”, therefore,

PR = TR v∞ and PA = TA v∞ . (2.19)

Note that the minimum of power required PR does not coincide with the minimum
of thrust required TR . It can be shown that the velocity associated with the drag
minimum vmin TR can be constructed from the tangent through the origin as illus-
trated.

With L = W , (1.38) and (2.17) we get from (2.19) that

s s
CD 2W 2 W 3 CD
PR = W = , (2.20)
CL ρ S CL ρ S CL3/2

It follows from (2.20) that the minimum power required is associated with the max-
3/2
imum of CL /CD .

Power required can straightforwardly be expressed as a function of velocity by com-

bining (2.19) and (2.18), i. e.
!
2W2
 
1 3 1
PR = CD0 ρ S v∞ + . (2.21)
2 π e AR ρ S v∞
wgd – Aircraft Performance 57

jet-propelled airplane
PA

PR

v∞

vstall vmin PR vmax

vmin vmin TR

Note that flight at different velocities v∞ with vmin ≤ v∞ ≤ vmax is associated with
different angles of attack and different throttle settings. In order to reduce the flight
velocity the pilot generally has to increase the angle of attack and adjust the throttle
setting such that the power provided by the engine equals PR .
58 Aircraft Performance – wgd

Three Important Velocities

q
First we rearrange (1.42) to give CL = π e AR CD,i . (2.22)

With TR = D , we conclude from (1.47), (1.51) and (2.22) that, for the minimum
of thrust required,



 v∞ = vmin TR





 CD,i = CD0


TR ⇒ M IN → (2.23)
 CL


 = (CL /CD )max



 CD q


 CL = π e AR CD 0 .

Using (2.21), we can derive for the minimum of power required that

dPR d(TR v∞ ) 3 2 2W2 1

= = CD0 ρ S v∞ − 2
= 0. (2.24)
dv∞ dv∞ 2 π e AR ρ S v∞

It follows that
!2
2L 1 CL2
3 CD 0 = 2
= = CD,i . (2.25)
S ρ v∞ π e AR π e AR

Recalling (2.20) and (2.22) we may summarise






v∞ = vmin PR







 CD,i = 3 CD 0
PR ⇒ M IN → 3/2 (2.26)

 CL 
3/2



 = CL /CD




CD q
max


 CL = 3 π e AR CD 0 .
wgd – Aircraft Performance 59

1/2
Furthermore, we introduce the velocity v̂∞ , which renders a maximum of (CL /CD ) .
It is convenient to introduce v̂∞ at this stage, even though we will not need it before
the discussion of range and endurance in Section 2.7. With (1.41) the expression
1/2
(CL /CD ) may be written as a function of CL , i. e.
1/2 √
CL CL
(CL ) = . (2.27)
CD CL2
CD 0 +
π e AR
At the maximum we have

1/2
 √
d CL /CD 1 CL 2 CL
= √ − 2 = 0. (2.28)
dCL 2 CL (CD 0 + CD,i ) (CD 0 + CD,i ) π e AR

Resubstituting (1.42) into (2.28), we find CD 0 = 3 CD,i . (2.29)

Thus, using (2.22), we can state that


1


 CD,i = CD0



 3

 1/2
for flight at v∞ = v̂∞ → CL 
1/2

(2.30)
= CL /CD



 CD r
max


 π

 CL = e AR CD 0 .
3

L
The velocities vmin TR , vmin PR and v̂∞ can then be found by using n = = 1 and
W
the expression for CL given in (2.23)4 , (2.26)4 , (2.30)3 , respectively, in Equation (2.5).

We obtain vmin PR < vmin TR < v̂∞ . (2.31)

A graphical interpretation of the velocities is shown on the next page.

60 Aircraft Performance – wgd

TR

tangent to graph
through origin

v∞

PR

tangent to graph
through origin

v∞

vmin PR v̂∞
vmin TR
wgd – Aircraft Performance 61

Effect of Altitude on Power Diagrams

Let the subscripts “sea” and “alt” denote quantities at sea level or at a higher
altitude, respectively. With (1.38) we can then relate the flight velocities valt and
vsea as follows
s s s s
2W ρsea 2W ρsea
valt = = = vsea . (2.32)
ρalt S CL ρalt ρsea S CL ρalt

Similarly, equation (2.20) renders

s s s s
2 W 3 CD ρsea 2 W 3 CD ρsea
PR, alt = = = PR, sea . (2.33)
ρalt S CL3/2 ρalt ρsea S CL3/2 ρalt

Thus, the graph for PR at a certain altitude

q
is obtained from the graph at sea level
by scaling both axis with the factor ρsea /ρalt > 1 .

For power available of the propeller-driven aeroplane, we recall from (2.11) that
ρalt
PA, alt = PA, sea . (2.34)
ρsea

For the jet-propelled aeroplane a similar relation as given by (2.13) holds for thrust
available. Combining (2.13) and (2.19) renders (2.34). Thus, power available at a
specific altitude is generally obtained by scaling down power available at sea level
with ρalt /ρsea < 1 .

The qualitative diagrams on the next page illustrate that vmax changes significantly
with altitude. The interval [vmin , vmax ] becomes smaller as the altitude increases.
Note however, that a higher altitude means larger velocities vmin PR and vmin TR .
62 Aircraft Performance – wgd

jet-propelled aeroplane

PA,sea
PR,alt

PR,sea
PA,alt

v∞
vmin,sea vmin,alt vmax,alt vmax,sea

propeller-driven aeroplane

P PA,sea

PR,alt
PR,sea

PA,alt

v∞
vmin,sea vmin,alt vmax,alt vmax,sea
wgd – Aircraft Performance 63

2.4 Climbing Flight and Flight Envelope

We assume that φ + α is small; thus sin(φ + α) ≈ 0 and cos(φ + α) ≈ 1 . For
straight, unaccelerated climbing flight, (2.1) and (2.2) then reduce to

T = D + W sin γ , L = W cos γ . (2.35)

From (2.35)2 , we obtain for the load factor

L
n = = cos γ ≤ 1 . (2.36)
W
The rate of climb ROC is related to v∞ by

ROC = vclimb = v∞ sin γ . (2.37)

With (2.35), it follows that

T −D
ROC = v∞ . (2.38)
W
We recall that v∞ D = v∞ TR = PR . By replacing T v∞ with power available PA
we obtain the maximum rate of climb for a specific velocity v∞ and altitude, i. e.
PA − PR
ROC = . (2.39)
W
The term PA − PR is identified as the excess power. The maximum climb rate is
calculated as
(PA − PR )max
ROCmax = . (2.40)
W

Remark: It follows from (2.36) and (2.4) with PR = v∞ TR = v∞ D that PR depends

on γ. With ROC = v∞ sin γ and PR (v∞ , γ) , the Equations (2.38) – (2.40)
are nonlinear in terms of v∞ and γ. For flight path angles γ < 20◦ , we may use
n = cos γ ≈ 1 and ignore the dependency of PR on γ. This allows the straightforward
evaluation of γ for a given v∞ . 2
 64 Aircraft Performance – wgd

Hodograph for Climbing Flight

(PA −PR )max

v∞

hodograph
ROC

ROC

ROCmax

v∞
γmax
v∞ γ vhoriz
vmax ROC

The maximum excess power (PA − PR )max and the associated freestream velocity
v∞ = vmax ROC can be obtained graphically by shifting the PA (v∞ ) curve as illus-
trated.

The hodograph is obtained from the ROC(v∞ ) diagram by exploiting

vhoriz = v∞ cos γ . (2.41)
Note that the maximum climb angle γmax is not associated with ROCmax and that
the hodograph changes with altitude.

At low velocities v∞ , a propeller-driven airplane possesses a larger proportion of

excess power PA − PR than a jet-propelled aircraft. Consequently, ROC at low v∞
and the maximum climb angle γmax are typically larger for propeller-driven airplanes.
wgd – Aircraft Performance 65

How high can an airplane fly?

The effect of altitude on power required and power available is described by (2.33)
and (2.34), respectively. If h2 q> h1 denote two different altitudes, such that
ρ2 < ρ1 , then PR is scaled with ρ1 /ρ2 > 1 , whereas PA is scaled with ρ2 /ρ1 < 1 .
Thus, the excess power PA − PR decreases as the altitude increases.

Consequently, there is an altitude with PA − PR = 0 for exactly one velocity v∞ , as

shown in the diagram below. There is no potential for further climb. This altitude
is called the absolute ceiling of the aircraft. At the absolute ceiling, straight, level
and unaccelerated flight can only be sustained for one specific velocity v∞ .

P
PA, sea PR, sea

(PA −PR )sea

PR, abs ceil

PA, abs ceil

v∞

vabs ceil
66 Aircraft Performance – wgd

Service Ceiling
The so-called service ceiling for jet-propelled aeroplanes is defined by the altitude
which allows for a maximum climb rate ROCmax = 100 ft/min = 0.508 m/s. At this
altitude the velocities vmin and vmax still provide a margin for different flight ma-
noeuvres. Thus, it is, for example, possible to avoid collision with another aeroplane
by climbing to a higher altitude. The service ceiling is of substantial importance to
jet-propelled aeroplanes since the fuel efficiency increases with altitude as we will
show in Section 2.7.

altitude h
absolute ceiling

service ceiling

ROC

100 ft/min
 wgd – Aircraft Performance 67

Time to Climb
For the calculation of the time ∆t required to climb from altitude h1 to altitude h2 ,
we need to express the rate of climb ROC as a function of altitude h. The graph
of ROC(h) is displayed qualitatively on the previous page. It may be approximated
with sufficient accuracy by one or more liner sections. We then have
h − h1
ROC(h) = ROC1 − (ROC1 − ROC2 ) , (2.42)
h2 − h1
where ROC1 and ROC2 denote the rates of climb at altitudes h1 and h2 , respectively.

The time required to climb from altitude h1 to altitude h2 may then be obtained as
Z Z h2
h2 dh dh
∆t = = a , (2.43)
h1 ROC(h) h1 b + h

h2 − h1 ROC1 h2 − ROC2 h1
where a = , b = . (2.44)
ROC2 − ROC1 ROC2 − ROC1

The integration renders

!
b + h2
∆t = a ln(b + h2 ) − a ln(b + h1 ) = a ln , (2.45)
b + h1

which becomes  
h2 − h1 ROC2
∆t = ln . (2.46)
ROC2 − ROC1 ROC1

h h
absolute ceiling

service ceiling

h2

h1

ROC t

ROC2 ROC1
∆t
68 Aircraft Performance – wgd

Flight Envelope
The flight envelope is an important diagram for any aeroplane. It shows the flight
altitude displayed against the freestream velocity v∞ or Mach number Ma∞ . We
obtain the flight envelope by evaluating vmin and vmax from the power required and
power available diagrams (see e. g. page 57) at different altitudes. Note that vmin
may be determined by the stall phenomena.

altitude h

stall

absolute ceiling

flight
envelope

v∞
wgd – Aircraft Performance 69

2.5 Gliding Flight

D
L

horizon
Θ γ

α tion
ight direc W
fl

v∞ vsink
H

γ
111111111111111111111111111111111111111111111111111111111111111
000000000000000000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000000000000000000
111111111111111111111111111111111111111111111111111111111111111
R

Assuming straight, unaccelerated flight with T = 0, and taking into account the
orientation of γ, the equations (2.1) and (2.2) reduce to

L − W cos γ = 0 , D − W sin γ = 0 . (2.47)

D L
It follows that = sin γ , n = = cos γ ≤ 1 , (2.48)
W W
sin γ D 1
and thus, tan γ = = = . (2.49)
cos γ L CL / CD

Defining H and R as illustrated, it follows from (2.49) that

1 R L CL
= = = . (2.50)
tan γ H D CD

Thus, in order to minimise the glide angle γ and maximise the glide range R, the
airplane should be flown at L/D = (L/D)max = (CL /CD )max . We recall from (2.23),
that this is associated with the freestream velocity v∞ = vmin TR = vmin D . In the
context of sailplanes, the expression 1/tan γ is also known as the glide number.
70 Aircraft Performance – wgd

With (2.19) and (2.49) it follows that

D PR
vsink = v∞ sin γ = v∞ = . (2.51)
W W
Thus, the sink rate vsink is minimised by flying at the minimum of PR , i. e. with
velocity vmin PR . Note that (2.26)2 –(2.26)4 only hold for SLUF or small γ.
For some average air density (ignoring the dependency of PR on altitude), the max-
imum glide endurance can be obtained from the minimum sink rate as
H HW
Eglide, max = = . (2.52)
vsink, min PR min

For small angles γ (sailplane, gliders), an approximation of the gliding hodograph

may be obtained by evaluating (2.51) for a range of velocities v∞ and by using
vhoriz = v∞ cos γ ≈ v∞ . However, large angles γ require to resolve the dependency
of PR on γ in (2.51) and also in (2.52). The exact hodograph may be obtained
from evaluating vsink and vhoriz for a range of lift coefficients CL . It is important to
understand that the minimum glide angle (maximum range) is not associated with
the minimum sink rate.

vhoriz
γmin
vsink, min γ

v∞

vsink
 wgd – Aircraft Performance 71

2.6 Take-Off and Landing

1111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111
accelerated ground roll rotation climb to 35 ft
0000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111
(increase of α, γ)
0000000000000000000000000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111111111111111111111111

D T

1111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111
R
0000000000000000000000000000000000000000000
1111111111111111111111111111111111111111111
W A
B

During ground roll, we have for the angle of attack and the flight path angle that
α = γ = 0. Furthermore, the thrust angle φ is assumed to be small. With Newton’s
second law and summation of the forces parallel to the runway it follows that
d v∞
m = T −D−R, (2.53)
dt
where m = W/g is the mass of the aeroplane.
Rolling friction R is proportional to the normal force A+B = W −L acting between
the tires and the ground, i. e.

R = µR (W − L) , (2.54)

where typically µR ≈ 0.02 for take-off on a smooth runway, or µR ≈ 0.4 for landing
ground roll with the brakes deployed.

With (1.20), (1.37), (1.41), (2.53) and (2.54) the equations which the govern ground
roll of an aeroplane are
72 Aircraft Performance – wgd

dv
m = T − D − µR (W − L) (2.55)
dt
1 2
L = ρ v S CL (2.56)
2 ∞
!
1 2 CL2
D = ρ v∞ S CD 0 + φ . (2.57)
2 π e AR

Due to the vicinity of the wings to the ground, the wing tip vortices and consequently
the downwash and induced drag do not develop to the same extent as at a higher
altitude. This phenomena is termed the ground effect and is accounted for by the
dimensionless factor φ. A good approximation is

(16 h/b)2
φ = < 1, (2.58)
1 + (16 h/b)2

where h is the height of the wings above the ground.

In order to calculate the ground roll distances required for take-off and landing, we
first consider
dv dv ds dv 1 d(v 2 )
= = v = . (2.59)
dt ds dt ds 2 ds

Defining the force F by F = T − D − µR (W − L) , (2.60)

2
dv∞ m d(v∞ )
Equation (2.55) can be rewritten as F = m = , (2.61)
dt 2 ds
m 2
which renders ds = d(v∞ ). (2.62)
2F

From integration we obtain the ground roll distance required to accelerate or decel-
erate from v1 to v2 as
Z v2
2 m 2
s = 2 d(v∞ ). (2.63)
v1 2 F
wgd – Aircraft Performance 73

Take-Off Ground Roll

The smallest velocity v∞ , with which flight can be sustained, is vstall . There-
fore, the ground roll distance required for acceleration before lift-off is minimal for
vLO = vstall . The quantity vLO denotes the lift-off velocity and the stalling velocity
vstall is given by (1.39). However, in order to provide a safety margin, it is common
practice to define s
2W
vLO = 1.2 vstall = 1.2 (2.64)
ρ S CLmax

We note that care must be taken to determine the value of CL = CLmax to be used in
(2.56), (2.57) and in (2.64). On the one hand, the angle of attack is restricted by the
horizontal orientation of the aeroplane during ground roll, which decreases CLmax ,
on the other hand, high lift systems such as flaps may be used to increase CLmax .

Exact Solution: The thrust T usually varies with the velocity v∞ . However,
particularly for jet-propelled aircraft, it is sufficiently accurate to assume T = const.
for constant altitude.
Thus, with (2.55) – (2.58) we obtain the lift-off ground roll distance sLO from (2.63)
by the following manipulations
Z 2 2 2
vLO d(v∞ ) 1 a vLO +b
sLO = 2 +b
= ln , (2.65)
0 a v∞ a b

where
!
φ CL2 ρS g 2g
a = µR CL − CD0 − and b = (T − µR W ) . (2.66)
π e AR W W

If the thrust varies significantly as v∞ increases, then a numerical integration tech-

nique may be required.
74 Aircraft Performance – wgd

Approximation 1: A commonly employed approximation for sLO is based on the

assumption of constant acceleration with an average force Fav = const. Typically,
we choose Fav = F (v∞ = 0.7 vLO ) . With (2.63) and (2.64) it then follows that

1.44 W 2
sLO ≈ . (2.67)
g ρ S CL max [T − D − µR (W − L)]0.7vLO

Approximation 2: If T is much larger than D + R , then (2.67) becomes

1.44 W 2
sLO ≈ . (2.68)
g ρ S CLmax T
Despite this being a very rough approximation, it demonstrates that sLO is pro-
portional to W 2 , rather than just W . Furthermore, if we recall from (2.19) and
(2.34) that thrust T is proportional to air density ρ, we find with (2.68) that the
lift-off ground roll sLO is inverse proportional to ρ2 . Consequently, airports which
are located at higher altitudes require longer runways.
The dependency of sLO on the wing reference area S and the maximum lift coefficient
CL max is evident.
wgd – Aircraft Performance 75

Landing Ground Roll

The typical touch-down velocity of an aeroplane is
s
2W
vTD = 1.3 vstall = 1.3 , (2.69)
ρ S CLmax

which provides an appropriate safety margin. Generally, the forces during landing
ground roll are described by (2.55) – (2.58). However, at touchdown, the thrust T
has usually been reduced to zero or it has been reversed. The rolling friction now
includes the effect of brakes such that µR ≈ 0.4, and the aerodynamic coefficients
CD 0 and CL account for the application of spoilers or drag chutes.

Exact Solution: Assuming T = const. ≤ 0 , the landing ground roll distance is

obtained from (2.63) as
Z 2 2
0 d(v∞ ) 1 a vTD +b
sL = 2
= ln , (2.70)
2
vTD a v∞ + b a b

where the quantities a and b are defined as in (2.66).

Approximation: An approximation for sL is obtained analogously to (2.67) as

1.69 W 2
sL ≈ . (2.71)
g ρ S CL max [−T + D + µR (W − L)]0.7vTD

Recall that for landing T ≤ 0 .

76 Aircraft Performance – wgd

Variation of Forces During Ground Roll

take-off ground roll

T
forces

D+R
D
R

0 ground roll SLO

landing ground roll

L
forces

D+R
R

0 ground roll SL
wgd – Aircraft Performance 77

2.7 Range and Endurance

The maximum distance an aeroplane can traverse on a tank of fuel is denoted as
range R. The maximum time which an aeroplane can stay airborne on a tank of
fuel is denoted as endurance E.

Clearly, R and E are related to the fuel consumption of the aeroplane. According
to Section 2.2, fuel consumption for the propeller-driven aeroplane is given in terms
of fuel burnt per power and time (specific fuel consumption SFC = const.), whereas
for the jet-propelled aeroplane fuel consumption is given in terms of fuel burnt per
thrust and time (thrust specific fuel consumption TSFC = const.). Therefore, we
have to develop separate sets or formula for range and endurance for the two types
of aeroplanes.

In order to account for the changes of aeroplane weight as fuel is being burnt,

 Wf current weight of fuel,


we define W0 weight of aeroplane with full fuel load,



W1 weight of aeroplane without fuel.

The current total weight of the aeroplane may then be expressed as

W = W1 + Wf . (2.72)
78 Aircraft Performance – wgd

Propeller-Driven Aeroplane
From (2.6) we recall that

− dWf / dt
SFC = ≈ const. (2.73)
P
Furthermore, we assume a constant propeller efficiency η and use (2.9) to write

− η dWf / dt
SFC = ≈ const. (2.74)
PA
Rearranging gives
η
dt = − dWf . (2.75)
SFC PA
We obtain for the endurance
Z Z
E W1 η
E = dt = − dWf (2.76)
0 W0 SFC PA
It follows that maximum endurance E is achieved if the airplane is always flown
at vmin PR with minimum power required PA = PR, min . Recalling (2.26), this is
3/2
associated with (CL /CD )max .

Assuming straight level unaccelerated flight with a constant angle of attack, i. e.

L CL
PA = TA v∞ = TR v∞ = D v∞ , W =L, = = const. (2.77)
D CD
s
2W
we recall (2.5) for n = 1 , i. e. v∞ = (2.78)
ρ S CL

and obtain from (2.76)

 s 
Z W0  Z 3/2
η 1 L η CL
W0 ρS 3
E= dW =  W − 2  dW , (2.79)
W1 SFC D v∞ W W1 SFC CD 2

which renders the Breguet formula for the endurance of a propeller-driven airplane
flown at constant altitude, i. e. ρ = const., as
3/2  
η CL q − 21 − 21
E = 2 ρ S W1 − W0 . (2.80)
SFC CD
wgd – Aircraft Performance 79

In order to calculate the range, we manipulate (2.75) and use (2.19) to obtain

ds η η η
ds = − dWf = − dWf = − dWf . (2.81)
dt SFC PA SFC PA / v∞ SFC TA

Integration renders the range as

Z Z
R W0 η
R = ds = dW . (2.82)
0 W1 SFC TA
Thus, the range is maximised if the airplane is always flown at vmin TR with minimum
thrust required TA = TR, min . According to (2.23), this corresponds to (CL /CD )max .
TA CL
For SLUF, we have = 1 . Multiplication of the integrand in (2.82) with
W CD
this term gives
Z W0 η CL 1 η CL W0
R = dW = ln . (2.83)
W1 SFC CD W SFC CD W1

Note that the endurance E as given by (2.80) for the propeller-driven aeroplane
depends on the altitude (ρ), whereas maximum range R as given by (2.83) can be
achieved at any altitude. The endurance is a maximum at sea-level, where ρ is large.
80 Aircraft Performance – wgd

Jet-Propelled Aeroplane
1
With T = TA we obtain from (2.12) dt = − dWf . (2.84)
TSFC TA
Integration renders the endurance as
Z Z
E W1 1
E = dt = − dWf . (2.85)
0 W0 TSFC TA
It follows that maximum endurance is achieved if the airplane is always flown at
vmin TR with minimum thrust required TA = TR, min . Recalling (2.23), this is associ-
ated with (CL /CD )max .

Based on the assumptions of straight level unaccelerated flight and a constant angle
TA CL
of attack we may multiply (2.85) with = 1 to obtain the Breguet formula
W CD
for the endurance of a jet-propelled aeroplane, i. e.
Z W0 1 CL dW 1 CL W0
E = = ln . (2.86)
W1 TSFC CD W TSFC CD W1

TA CL
Manipulating (2.84) and assuming SLUF with (2.78) and = 1 we may write
W CD

ds / dt v∞ CL dWf
ds = − dWf = −
TSFC TA TSFC CD W
s 1/2
2 1 CL dW
= − √ f . (2.87)
ρ S TSFC CD W
Integration renders the range as
Z Z s 1/2
R W1 2 1 CL dW
R = ds = − √ f . (2.88)
0 W0 ρ S TSFC CD W

Thus, for ρ = const. (constant altitude), the range is maximised if the airplane is
1/2
always flown at the velocity v̂∞ that renders maximum CL /CD (see (2.30)). This
velocity can be obtained graphically from the thrust required over velocity diagram
by means of the tangent through the origin (analogously to vmin TR in the power-
required diagram).
wgd – Aircraft Performance 81

1/2
Based on the assumption of SLUF, i. e. CL /CD = const. , we can derive the
Breguet formula for the range of a jet-propelled airplane from (2.88) as
s 1/2 q 
1 8 CL q
R = W0 − W1 , (2.89)
TSFC ρ S CD

For the jet-propelled airplane the endurance as given by (2.86) is independent of the
altitude, whereas the range as given by (2.89) increases with altitude. Maximum
range can therefore be achieved when the airplane is flown near the absolute ceiling.
That is the reason for the introduction of the “service ceiling” as described on
page 66.

It is important to know that, in practice, for both propeller-driven and jet-propelled

aeroplanes, range and endurance depend significantly on the weather conditions,
the pilot skills, the climb to the cruising altitude etc. However, the Breguet formula
(2.80), (2.83), (2.86) and (2.89) provide acceptable approximations and valuable
insight into how range and endurance can be maximised.
82 Aircraft Performance – wgd

Range and Payload

The specifications of larger aeroplanes typically include the range–payload curve.
This is a graph which, for a particular mission profile, presents the effect of trading
payload for fuel on the range of an airplane.

The specific design of any aeroplane renders a certain maximum take-off weight,
which must not be exceeded by the total weight in order to ensure that the aero-
plane is fully operational. Consequently, if the payload is very heavy, the fuel tanks
can only be filled partially and the range is reduced significantly.

The range–payload diagram may also reflect structural limitations, i. e. the maxi-
mum load which can be supported by the airframe, and the limitation arising from
the capacity of the fuel tanks.
payload

structural limit

fuel capacity

maximim take-off weight

range R
wgd – Aircraft Performance 83

2.8 Turning Flight

Level Turn
L
θ
R
θ FR

W v∞

If the pilot increases the bank angle to θ > 0, then the lift force is tilted as illustrated.
Assuming that v∞ and L are controlled such that v∞ = const. and

L cos θ = W , (2.90)

then the aircraft performs a circular level turn. The load factor (2.3) associated with
a level turn is
L 1
n = = ≥ 1. (2.91)
W cos θ
The component of lift directed towards the centre of the turn may be written as
√ √
FR = L2 − W 2 = W n2 − 1 . (2.92)
2
Using Newton’s second law and the radial acceleration v∞ /R gives
2
v∞
FR = m . (2.93)
R
Combining (2.92), (2.93) and W = m g renders
v2
R = √ ∞ . (2.94)
g n2 − 1
The angular velocity or turn rate is obtained as
√
v∞ g n2 − 1
ω = = . (2.95)
R v∞
Recall from (2.4) that the induced drag becomes larger as n increases. Thus, the
thrust required to sustain a level turn is larger than that for straight level unaccel-
erated flight.
84 Aircraft Performance – wgd

Pull-Up and Pull-Down

R W L

L R

If the lift of an airplane, which is initially in straight, level flight, is suddenly in-
creased, the airplane will perform a pull-up manoeuvre. We obtain

FR = L − W = W (n − 1) . (2.96)

Similarly to (2.94) and (2.95) it follows that

2
v∞
R = (2.97)
g (n − 1)

and
g (n − 1)
ω = . (2.98)
v∞

If an airplane, which is initially in level flight, suddenly rolls into an inverted position,
it will perform a pull-down manoeuvre. Analogously to (2.96) – (2.98), we obtain

FR = L + W = W (n + 1) , (2.99)
2
v∞
R = (2.100)
g (n + 1)
and
g (n + 1)
ω = . (2.101)
v∞
wgd – Aircraft Performance 85

On the Manoeuvrability of an Aeroplane

For large n, we may argue that n ≈ n + 1 . The turn radii and turn rates for the
different manoeuvres as given by (2.94), (2.95), (2.97), (2.98) and (2.100), (2.101)
may then be approximated, respectively, by
2
v∞ gn
R = and ω = . (2.102)
gn v∞

Using (2.5), this can be written as

s
2W ρ S CL n
R = and ω = g . (2.103)
ρ S CL g 2W

Hence, high manoeuvre performance (small minimum R, large maximum ω) is as-

sociated with large CLmax and nmax . Depending on the velocity v∞ , the maximum
load factor nmax is determined by the maximum lift coefficient CL max or by the
structural strength of the aircraft as described in the next Section.

According to (2.103), high manoeuvre performance is also associated with a low

wing loading W/S . However, the wing loading is typically determined by factors
other than manoeuvring, such as payload, range and maximum velocity. As a result
wing loadings for light general aviation aircraft are relatively low, but those for high
performance military aircraft are relatively large.

airplane W/S [ kg m−2 ]

Wright Flyer (1903) 7
Spitfire (1936) 137
sailplane DG-1000 35
Cessna Skyhawk 69
Boeing 747 350 - 700
Airbus A-380 300 - 700
Eurofighter Typhoon 311
General Dynamics F-16 430
86 Aircraft Performance – wgd

2.9 V–n Diagram

In the V − n diagram, we display the maximum admissible value of the load factor
nmax against the freestream velocity v∞ .

From (2.5) it follows for the stall limit that

2
ρ S v∞ CLmax
nmax = . (2.104)
2W

The load factor nmax is also subject to structural limitations of the aeroplane. A
major limitation arises from the strength of the joints between light lift producing
parts (wings) and the bulky heavy parts (fuselage) of the airplane.

The velocity v∞ is also subject to structural limitations. These may be due to

aeroelastic instabilities, temperature effects, excessive dynamic pressure, etc.

nmax manoeuvre point

n1

stall limit
flight envelope

v∞
v∞
∗

− n2

structural limit

Recalling (2.103), we identify the manoeuvre point where CL and n are maximal
and render the highest manoeuvre performance possible. The velocity v∞
∗
is called
the corner velocity.

The area with all admissible pairs (n, v∞ ) is denoted as the flight envelope.
wgd – Aircraft Performance 87

Clearly, the V − n diagram is an interface between the disciplines of flight mechanics

and airframe design. It represents many of the load cases which the airframe has
to withstand. The values for n1 and n2 as defined in the diagram are specified by
the airworthiness authorities for particular aircraft. Typical load factors laid down
in the British Civil Airworthiness Requirements (BCAR) are shown in the table.

aeroplane type

normal semi-aerobatic aerobatic

n1 24, 000 lb 4.5 6.0

2.1 +
W + 10, 000 lb

n2 1.0 1.8 3.0

For military aeroplanes, the value of n1 is of the order of 10 and larger.

88 Aircraft Performance – wgd

Effects of Weight and Altitude on the V–n Diagram

nmax

increasing W v∞

nmax
increasing altitude,
decreasing ρ v∞

The effects of W and ρ on the stall limit are evident from (2.104).

Often, the structural limit arises from a maximum admissible lift force Lmax = const.
or maximum admissible difference force (L − W )max = const., beyond which the
airframe fails. It then follows from L = n W that the structural limit nmax decreases
as W increases.

The maximum admissible velocity typically increases as ρ decreases. This may, for
example, be due to the decrease of the dynamic pressure.
wgd – Aircraft Performance 89

2.10 Energy Method for Accelerated Flight

Modern high performance aeroplanes are capable of highly accelerated climbs. Such
manoeuvres can be analysed with the energy method.

First, we consider the specific total aircraft energy He , which is the sum of the poten-
tial and the kinetic energies of the aeroplane divided by the aeroplane weight, i. e.
 
1 2 1
He = mgh + m v∞ . (2.105)
2 W
With W = m g , we may write
2
v∞
He = h + . (2.106)
2g
We can then draw isolines of He in the altitude over velocity diagram, which are
independent of the specific aeroplane under consideration, i. e. all aeroplanes flying
at altitude h with velocity v∞ have the same energy level He .

Next, we recall (2.1) whereby we assume that φ + α is small, i. e.

d v∞
m = T − D − W sin γ . (2.107)
dt
Rearranging with W = m g gives
v∞ T − v∞ D v∞ dv∞
= v∞ sin γ + . (2.108)
W g dt
We recognise the term on the left hand side as the ratio of the excess power over the
aeroplane weight. This ratio is known as the specific excess power Ps . Furthermore,
we recall from (2.37) that we have for the rate of climb ROC
dh
ROC = = v∞ sin γ . (2.109)
dt
We may then write (2.108) as
dh v∞ dv∞
Ps = + . (2.110)
dt g dt
From the comparison of (2.106) and (2.110), we note that the specific excess power
equals the time derivative of the energy level, i. e.
dHe
Ps = . (2.111)
dt
90 Aircraft Performance – wgd

Ps Diagram
The conclusion of (2.111) is that an aeroplane can change its energy state by the
application of excess power. It is therefore essential to know how much excess
power is available to the aeroplane, when it is flying with velocity v∞ at altitude
h. For this purpose, we draw contour lines of Ps in the h–v∞ system. The contour
line with Ps = 0 describes the flight envelope similar to the diagram on page 68.
In the same diagram, we may plot contour lines of the energy level he . Typical
Ps diagrams for a subsonic and a supersonic aeroplane are displayed below. Note
that, due to Equation (2.4) for drag (thrust required), the contour lines for Ps de-
pend on the aeroplane configuration and on the load factor. The irregular shape of
the Ps contour lines for the supersonic aeroplane is due to the effect of the maximum
drag associated with Mach numbers near Ma∞ = 1 (compare Section 1.12).

zooms and dives: The pilot does not need to apply any excess power in order to
change from one state (v1 , h1 ) on a contour line of He to another state (v2 , h2 ) on the
same contour line. Such “zooms” (h2 > h1 ) and “dives” (h2 < h1 ) basically mean
trading kinetic energy for potential energy or vice versa. Note, however, that these
manoeuvres affect the specific excess power available to the pilot, since the thrust
required and thrust available change with the velocity and altitude.

minimum time to climb: The Ps diagrams may be used to determine the min-
imum time required to climb to the maximum energy height. In fact, for fighter
aircraft, it is advantageous to have a higher energy level than the adversary. Rear-
ranging (2.111) gives
dHe
dt = , (2.112)
Ps
which we integrate to obtain
Z Z
∆t He, 2 1
∆t = dt = dHe . (2.113)
0 He, 1 Ps
Thus, ∆t is minimised by using Ps, max on each energy level. This is represented in
the h–v∞ system by the trajectory that connects the maxima of Ps on each energy
level. In the diagram for the supersonic aeroplane, note the characteristic constant-
energy dive to accelerate through the drag-divergence region near Ma∞ = 1 .
wgd – Aircraft Performance 91

subsonic aeroplane

supersonic aeroplane
92 Aircraft Performance – wgd

construction of Ps contour lines

Chapter 3

Static Stability and Control

3.1 Static and Dynamic Stability

The analysis of the static stability of an aeroplane is concerned with the forces or
moments acting on the aeroplane, which arise from disturbances such as wind gusts
or temporary deflections of the control surfaces. The direction or orientation of these
forces and moments should be such that the aeroplane is returned (at least initially)
towards stable flight. The specific aeroplane configuration under consideration is
then said to be “statically” stable. For unstable configurations, disturbances cause
moments and forces which take the aeroplane further away from stable flight and,
in the worst case, cause the loss of control.

It is important to understand that static stability is necessary, but not sufficient for
stable flight. Only the additional consideration of dynamic stability ensures that a
specific aeroplane configuration is really stable. Dynamic stability is concerned with
the motion of the aeroplane following a disturbance. An aeroplane may be statically
stable and dynamically unstable as illustrated in the diagrams below.

We distinguish three different states of static stability. A configuration may be

stable, i. e. small disturbances cause forces which return the system into stable
equilibrium. The state of the configuration may be neutral, i. e. disturbances do
not cause any forces or moments, or it may be unstable, i. e. disturbances cause
forces which remove the system further away from equilibrium.

93
94 Static Stability and Control – wgd

111111111111111111111111
000000000000000000000000
000000000000000000000000
111111111111111111111111 11111111111111111111
00000000000000000000
000000000000000000000000
111111111111111111111111
000000000000000000000000
111111111111111111111111 00000000000000000000
11111111111111111111
000000000000000000000000
111111111111111111111111 00000000000000000000
11111111111111111111
000000000000000000000000
111111111111111111111111 11111111111111111111
00000000000000000000 00000000000000000000
11111111111111111111
000000000000000000000000
111111111111111111111111 00000000000000000000
11111111111111111111 00000000000000000000
11111111111111111111
000000000000000000000000
111111111111111111111111 00000000000000000000
11111111111111111111 00000000000000000000
11111111111111111111
000000000000000000000000
111111111111111111111111 00000000000000000000
11111111111111111111 00000000000000000000
11111111111111111111
00000000000000000000
11111111111111111111
000000000000000000000000
111111111111111111111111 00000000000000000000
11111111111111111111 00000000000000000000
11111111111111111111
000000000000000000000000
111111111111111111111111 00000000000000000000
11111111111111111111
stable neutral unstable

aperiodic periodic

statically stable
dynamically stable
converging

displacement

time

statically stable
dynamically stable
diverging

statically unstable
dynamically unstable statically stable
dynamically unstable
wgd – Static Stability and Control 95

3.2 Aircraft Rotations and Control Surfaces

x
U roll
L, P
CG

V pitch
M, Q
W y
N, R
yaw
z
The rotations of the aeroplane with respect to the axis of a coordinate system that
moves with the aeroplane as shown are known as pitch, roll and yaw. The associ-
ated angular velocities are denoted by Q, P and R, respectively. These rotational
velocities are due to moments M , L and N as shown. The translational velocities
along the axis are denoted by U , V and W .

aileron
vertical stabiliser rudder
flap

elevator

horizontal stabiliser

Flaps are high-lift systems (compare Section 1.9), whereas ailerons, elevators and
rudder are the conventional control surfaces.
96 Static Stability and Control – wgd

By deflecting the control surfaces during flight the pilot can generate the moments
L, M or N .
wgd – Static Stability and Control 97

3.3 Longitudinal Static Stability

Longitudinal stability is concerned with the relation between the pitching moment
M and the corresponding angular displacements. We define the pitching moment
with respect to the centre of gravity CG of the aeroplane as

M = q∞ S c̄ CM , (3.1)

where c̄ denotes the mean chord line.

Furthermore, for convenience, we introduce the absolute angle of attack αa as

αa = α − αL=0 . (3.2)

CL CL

α αa

αL=0

By means of wind tunnel tests, or by an analysis along the lines of the following
pages, the pitching moment coefficient CM can be represented as a function of the
absolute angle of attack αa . In the range of small αa , this function is typically linear
and shows one angle αa for which CM disappears. By definition, an increase of αa
corresponds to a nose-up rotation of the aeroplane with respect to the freestream
velocity. For a stable configuration, that should generate a nose-down pitching mo-
ment, which returns the aeroplane to the undisturbed flight. Therefore, we conclude
that a requirement for positive longitudinal stability is that the derivative dCM /dαa
be negative, i. e.
dCM
for longitudinal stability: < 0. (3.3)
dαa
Steady and stable flight clearly requires CM = 0 . We say that for CM = 0 the
aeroplane is trimmed. The associated absolute angle of attack αe is known as the
angle of trim. With (3.3), it follows from CM = 0 for some positive αe that

for longitudinal stability: CM (αa = 0) > 0 . (3.4)

98 Static Stability and Control – wgd

CM
stable

nose-up point of trim

pitching
moment αa
nose-down
pitching
moment
dCM
< 0
nose-down nose-up dαa
displacement αe displacement

CM

dCM
> 0
dαa

αa

unstable

On the following pages, we derive an expression for CM in terms of parameters which

describe the configuration of the aeroplane. First, we will account for the contribu-
tion of the wings and the fuselage to the pitching moment, and second, we will study
the contribution of the horizontal tail. The objective is to translate the conditions
(3.3) and (3.4) for longitudinal stability into practical design requirements.
 wgd – Static Stability and Control 99

Contribution of Wings and Body to the Pitching Moment

Lwb

zero lift line Mac,wb z

Dwb

αwb

v∞
hac,wb c̄
h c̄
c̄ CG ACwb

The “body” of the aeroplane includes the fuselage and the engine nacelles. We
then introduce the aerodynamic centre of wing & body ACwb . Analogously to the
aerodynamic centre of the airfoil as discussed in Section 1.8, this is the point with
respect to which the aerodynamic moment Mac,wb is independent of the angle of
attack. The angle αwb denotes the absolute angle of attack for the wing & body
combination. The quantities h and hac,wb are dimensionless parameters and c̄ denotes
the mean zero-lift chord also known as the aerodynamic chord. Note that h defines
the location of the centre of gravity CG of the entire aeroplane.

The pitching moment of the wing & body combination with respect to the centre of
gravity may then be written as

Mcg,wb = Mac,wb + c̄ Lwb cos αwb (h − hac,wb ) + z Lwb sin αwb

− z Dwb cos αwb + c̄ Dwb sin αwb (h − hac,wb ) , (3.5)

which we may divide by S q∞ c̄ to obtain

z
CMcg,wb = CMac,wb + (h − hac,wb ) CLwb cos αwb + CLwb sin αwb
c̄
z
− CDwb cos αwb + (h − hac,wb ) CDwb sin αwb . (3.6)
c̄
100 Static Stability and Control – wgd

Generally, the ratio z/c̄ is negligible and the angle of attack is small. With
z
≈0, sin αwb ≈ 0 , cos αwb ≈ 1 , (3.7)
c̄
we may rewrite (3.6) as

CMcg,wb = CMac,wb + (h − hac,wb ) CLwb . (3.8)

dCLwb
Using CLwb = CLwb α αwb , with CLwb α = , (3.9)
dαwb

we obtain
CMcg,wb = CMac,wb + (h − hac,wb ) CLwb α αwb . (3.10)

The location of the aerodynamic centre as defined by hac,wb and the parameters
CMac,wb and CLwb are usually determined by wind tunnel tests. Note that, for a wing
with positive camber, the moment coefficient CMac,wb is generally negative.
wgd – Static Stability and Control 101

Contribution of the Tail to the Pitching Moment

ht c̄ Lt
zero-lift line (tail)
h c̄
lt
αwb −ε
zt
αt Dt
CG

ACwb
it zero-lift line (w
ing & body)
v∞
ε
v′

Generally, the aerodynamic moment Mac,t of the tail plane is negligible due to zero
camber or due to the relatively small size of the tail. The angle it between the
zero-lift lines of the wing & body combination and the horizontal tail is known as
the tail incidence angle. The angle ε accounts for the downwash caused by the wing
tip vortices. Due to the location of the tail plane behind the wings, this effect may
be very pronounced and must not be neglected. The contribution of the tail to the
moment Mcg,t with respect to the centre of gravity is then obtained as

Mcg,t = − lt Lt cos(αwb − ε) − zt Lt sin(αwb − ε)

− lt Dt sin(αwb − ε) + zt Dt cos(αwb − ε) . (3.11)

Division by q∞ S c̄ gives
lt St zt St
CMcg,t = − CLt cos(αwb − ε) − CLt sin(αwb − ε)
c̄ S c̄ S
lt St zt St
− CDt sin(αwb − ε) + CDt cos(αwb − ε) . (3.12)
c̄ S c̄ S
102 Static Stability and Control – wgd

zt
With ≈ 0, sin(αwb − ε) ≈ 0 , cos(αwb − ε) ≈ 1 , (3.13)
c̄
lt St
it follows that CMcg,t = − CLt . (3.14)
c̄ S

The factor multiplying CLt in (3.14), which quantifies the effect of the tail, is known
as the horizontal tail volume VH , i. e.
lt St
CMcg,t = − VH CLt , and VH = . (3.15)
c̄ S
It is important to understand that VH is not a purely geometrical parameter, but
it depends on the location of the centre of gravity h. For the sake of clarity, we
therefore rewrite (3.14) as

St
CMcg,t = − (ht − h) CLt . (3.16)
S
The angle of attack αt may be expressed as

αt = αwb − ε − it . (3.17)

Due to the linear relations between angle of attack and lift coefficient (1.23) and
between lift coefficient and induced angle of attack (1.29), we can assume a linear
variation of ε with the absolute angle of attack αwb , i. e.

ε = εα αwb , (3.18)

where εα = dε/dα can be determined by means of wind tunnel tests. The lift
coefficient CLt can then be expressed as
 
CLt = CLt α αwb = CLt α (1 − εα ) αwb − it , (3.19)

and (3.16) becomes

St  
CMcg,t = − (ht − h) CLt α (1 − εα ) αwb − it . (3.20)
S
Note that due to the orientation of a positive tail incidence it the term CLt α it cor-
responds to a tail lift force which is directed downwards, whereas the term CLt α αwb
with αwb > 0 is associated with a tail lift force that is directed upwards.
wgd – Static Stability and Control 103

The Total Pitching Moment

The total pitching moment coefficient CMcg is obtained from adding up the contri-
butions of the wing & body and the tail.

CMcg = CMcg,wb + CMcg,t . (3.21)

The expressions (3.10) and (3.20) for CMcg,wb and CMcg,t , respectively, involve the ab-
solute angle of attack of the wing & body combination αwb . Based on the negligible
influence of the tail on the orientation of the zero-lift line of the entire aeroplane, we
may replace αwb by the absolute angle of attack of the entire aeroplane αa . Hence-
forth, for the sake of notational brevity, we also omit the subscript “cg”. Using
(3.10), (3.20) and αwb = αa we may rewrite (3.21) as

St  
CM = CMac,wb + (h − hac,wb ) CLwb α αa − (ht − h) CLt α (1 − εα ) αa − it ,
S
or shorter,
CM = CM 0 + CM α αa , (3.22)
with
St
CM 0 = CMac,wb + (ht − h) CLt α it (3.23)
S
St
CM α = (h − hac,wb ) CLwb α − (ht − h) CLt α (1 − εα ) . (3.24)
S

The Equations (3.22) – (3.24) express the pitching moment coefficient in terms of
the location of the centre of gravity and in terms of some aerodynamic coefficients.

We now recall the requirement (3.4) for longitudinal stability and consider an aero-
plane subjected to airflow at an absolute angle of attack equal to zero. The pitching
moment coefficient then reduces to CM 0 . The condition (3.4) then requires that
CM 0 > 0. For positively cambered wings we have CMac,wb < 0. Thus, it follows
from (3.23) that we can ensure that CM 0 > 0 by providing a sufficiently large pos-
itive tail incidence it (and tail volume). It is important to understand that this
corresponds to a tail lift force which is directed downwards.

We conclude from (3.24) that the requirement (3.3), i. e. CM α < 0 , can be satisfied
simply by providing sufficient tail volume.
104 Static Stability and Control – wgd

Total Lift, Neutral Point and Static Margin

As a preliminary, we begin this subsection by considering the total lift of the aero-
plane. Adding the contributions of wing & body and tail, we may write

   
St St
L = qS CLwb + CLt = qS CLwb α αa + CLt α αt
S S
  
St St
= qS CLwb α + CLt α (1 − εα ) αa − CLt α it , (3.25)
S S

where we have used (3.17) to eliminate αt . The total lift coefficient may then be
expressed as
CL = CL 0 + CL α αa (3.26)

St St
with CL0 = − CLt α it , CLα = CLwb α + CLt α (1 − εα ) . (3.27)
S S

Analogously to the airfoil or the wing & body combination, there exists a location
with respect to which the total pitching moment of the aeroplane is independent of
the angle of attack. This point corresponds to the aerodynamic centre of the entire
aeroplane and it is known as the neutral point NP. Its location is denoted by the
parameter hn , such that hn c̄ is the distance between the leading edge of the wing
and the neutral point. In (3.22) – (3.24), the parameter h denotes the location of
the point of reference for the pitching moment. Therefore, hn can be calculated by
equating CM α in (3.24) to zero and by replacing h with hn . We obtain
St
0 = (hn − hac,wb ) CLwb α − (ht − hn ) CLt α (1 − εα ) , (3.28)
S
which renders
St
hac,wb CLwb α + ht CLt α (1 − εα )
hn = S . (3.29)
St
CLwb α + CLt α (1 − εα )
S

With (3.27)2 we can rewrite (3.29) as

 
1 St
hn = hac,wb CLwb α + ht CLt α (1 − εα ) . (3.30)
CL α S
wgd – Static Stability and Control 105

In some textbooks, the difference between the lift slopes CL α and CLwb α is neglected
(the effect of the tail is small). If we also ignore the dependency of the tail volume
VH on h, i. e. h << ht and VH ≈ ht St /S (see (3.15) and diagram on page 101),
then we may rewrite (3.30) as
CLt α
hn = hac,wb + VH (1 − εα ) . (3.31)
CL α

Based on the expression for CL α in (3.27)2 we can also rewrite (3.24) as

 
St
CM α = h CL α − hac,wb CLwb α + ht CLt α (1 − εα ) , (3.32)
S

Using (3.30) to simplify (3.32) renders

CM α = (h − hn ) CLα . (3.33)

This important result states that the slope CM α of the moment curve is equal to the
slope of the lift curve CL α multiplied by the dimensionless distance of the centre of
gravity behind the neutral point. According to (3.3), for static stability, CM α must
be negative. Therefore, for a statically stable aeroplane the centre of gravity must
be located ahead of the neutral point.

The quantity hn − h is known as the static margin SM, which must be positive to
ensure static stability. Adequate static stability is achieved if the static margin is
at least 5%.

hn − h < 0 (unstable)
CM

hn − h = 0 (neutral)

αa

hn − h > 0 (stable)
106 Static Stability and Control – wgd

neutral point NP
trimmed flight (aerodynamic centre
of entire aeroplane)
equilibrium L

M0

hn − h (static margin)
centre of gravity CG W

nose-up perturbation
→ increased lift
→ nose-down moment

M0

nose-down perturbation
→ decreased lift
→ nose-up moment
M0

W
wgd – Static Stability and Control 107

Special Configurations

T-tail configuration: In a classically configured aeroplane, the horizontal tail is

much smaller than the wings and therefore relatively close to the fuselage. Typically,
a significant part of the tail is located inside the boundary layer of the fuselage. Due
to skin friction, the airflow inside this boundary layer is slower than further away
from the fuselage. This effect can be accounted for by introducing the tail efficiency
factor η, i. e.
qt = η q∞ with η ≤ 1 , (3.34)
where qt is the dynamic pressure at the leading edge of the tail surface. For the
T-tail configuration, we have η = 1. In these notes, we assume for simplicity that
always η = 1 .

Canard configuration: It follows from (3.23) that CM 0 is positive only if the tail
generates a lift force which is directed downwards. This may seem impractical with
respect to the total lift, i. e. the lift of the wings has to counterbalance not only the
weight but also the downward lift of the tail. Therefore, on some aeroplanes, the
horizontal stabilisers are located ahead of the wings. Such horizontal stabilisers are
known as Canard surfaces. The Canard surfaces generate upward lift which causes
a positive pitching moment increment equivalent to the moment generated by the
downward lift of a horizontal tail. Thus, the Canard surfaces stabilise the aeroplane
and at the same time contribute to the total aeroplane lift.
A significant disadvantage of a Canard configuration is that the airflow over the
wings is disturbed by the wake of the Canard surfaces. This adversely affects the
generation of lift of the wings.
108 Static Stability and Control – wgd

3.4 Longitudinal Control

Consider an aeroplane in trimmed configuration at straight level and unaccelerated
flight. In order to fly faster, the pilot needs to increase the throttle setting and
to reduce the angle of attack. In order to fly more slowly, he must reduce the
throttle setting and increase the angle of attack. According to (3.22) the pitching
moment depends exclusively on the aeroplane geometry and on the absolute angle
of attack αa . Therefore, the variation of αa will lead to the loss of trim, unless the
aeroplane geometry is changed to achieve trim for the new angle of attack. Two
different strategies are available:

• shifting the centre of gravity.

Clearly this is impractical for complex aeroplanes. However, for simply con-
figured sailplanes or gliders without control surfaces this is a viable strategy.
It is discussed in more detail on the next page.

• adjusting the tail plane configuration.

By adjusting the lift of the horizontal tail plane the total pitching moment may
be reduced to zero such that the aeroplane is trimmed again. We distinguish
different tail plane configurations, such as the all moveable tail, the horizontal
stabiliser-elevator configuration or the stabilator, which are discussed on the
following pages.
wgd – Static Stability and Control 109

Shifting the Centre of Gravity

In order to study the effect of shifting the centre of gravity on the CM – αa diagram,
we first assume
ht − h
h << ht → ≈ 1 → ht − h ≈ ht . (3.35)
ht
It then follows from (3.22) – (3.24) that shifting the centre of gravity forward results
in a smaller angle of trim αe , whereas a backward shift increases αe .

CM

αa

shift forward shift backward

αe
110 Static Stability and Control – wgd

All-Movable Tail
In this configuration, the tail incidence it can be controlled by the pilot. In order to
express the tail incidence associated with trim in terms of the total lift coefficient
CL (which easily relates to the freestream velocity by (2.5)) we note that (3.22) and
(3.26) may be written, respectively, as

CM = CMac,wb + CM i it + CM α αa (3.36)

CL = CL i it + CL α αa , (3.37)
where
St
CM i = (ht − h) CLt α (3.38)
S
St
CLi = − CLt α . (3.39)
S

Using (3.37) to eliminate αa in (3.36), we obtain

CMac,wb CL α + CM α CL
it = − . (3.40)
CL α CM i − CM α CL i

CM

αa

decrease it increase it

αe
wgd – Static Stability and Control 111

Tail with Elevators

δe
zero lift line wing & body

ihs

αt

Recalling the discussion of flaps in Section 1.9, we conclude from (1.24)

CLt = CLt α (αt + τ δe ) , (3.41)

where τ is the effectiveness factor of the elevator flap. Setting the right hand side of
(3.41) to zero, we find the angle αt, L=0 = − τ δe , which renders zero lift and therefore
defines the orientation of the zero lift line of the tail. The sum of αt, L=0 and the
incidence ihs of the fixed horizontal stabiliser corresponds to the tail incidence it
in (3.36) and (3.37), which describes the zero lift line of the total tail, i. e.

it = ihs − τ δe . (3.42)

We may now use (3.42) in (3.36) and (3.37) to replace it . Next, we eliminate αa , set
CM to zero and solve the resulting equation for δe . We obtain
CMac,wb CLα + CM α CL ihs
δe = − + , (3.43)
CLα CM δ − CM α CLδ τ
which defines the elevator deflection, that renders trimmed flight for a given total
lift coefficient CL . The new expressions in (3.43) are defined as

St
CM δ = − τ CM i = − τ (ht − h) CLt α (3.44)
S
St
CLδ = − τ CLi = τ CLt α . (3.45)
S
112 Static Stability and Control – wgd

Stabilator

− ihs

δe

The stabilator configuration may be regarded as a horizontal stabiliser with an

elevator, whereby the incidence ihs of the stabiliser is variable. Furthermore, the
elevator is linked to the fuselage as shown in the diagram. The elevator deflection
is thus dependent on the incidence ihs , i. e.

δe = ke ihs + δ0 , (3.46)

where ke is the linkage factor. Using (3.46) in (3.42), we obtain

it = ihs (1 − τ ke ) − τ δ0 . (3.47)

Following the same procedure as for the fixed stabiliser-elevator configuration, we

use (3.47) in (3.36) and (3.37) to replace it . We eliminate αa , set CM to zero and
solve the resulting equation for ihs . We thus obtain

CMac,wb CL α + CM α CL 1 τ δ0
ihs = − + , (3.48)
CLα CM i − CM α CLi 1 − τ ke 1 − τ ke

which defines the stabiliser incidence, that renders trimmed flight for a given total
lift coefficient CL .

The linkage factor ke is independent of aerodynamics and depends only on the

mechanical linkage of the elevator to the fuselage. It follows from (3.48) that a
negative ke increases the tail effectiveness in the sense of requiring smaller variations
of ihs to trim the aeroplane for a different velocity (angle of attack).
wgd – Static Stability and Control 113

3.5 Stick-Free Longitudinal Static Stability

In large aeroplanes, hydraulic systems are commonly used to hold the moveable con-
trol surfaces in the desired position. However, small light aircraft, older ones as well
as modern ones, are equipped with simple mechanical gear systems, which the pilot
has to operate by means of sticks or wheels. These systems must be designed such
that, for each point within the flight envelope (V − n diagram), the forces and mo-
ments to be applied by the pilot lie within acceptable limits. It is the gear systems
which can give proper “feel” to the pilot. “Elevator-down”, for example, should be
associated with a push-forward of the stick (decrease angle of attack, accelerate),
whereas “elevator-up” should be achieved by a pull-back of the stick (increase angle
of attack, decelerate).

− δe

We have seen that the elevator angle of trim or the tail incidence of trim (moveable
tails, stabilators) vary with the flight velocity. A mechanical gear system would
therefore require that the pilot permanently holds on to the stick and exerts a force
to keep it in the right position associated with trim for the desired velocity. This is
clearly very impractical. Therefore, we will, in the following, consider what happens
when the stick is left free, i. e. when the elevator is “floating”.
114 Static Stability and Control – wgd

Tail with Elevators

For the horizontal stabiliser-elevator configuration, the elevator hinge moment H
may be expressed as
1 2
H = ρ v∞ Se c̄e CH , (3.49)
2
where Se and c̄e denote, respectively, the surface area of the elevator and the mean
chord of the elevator. For symmetrical sections, the hinge moment coefficient CH
depends only on the angle of attack and on the elevator deflection and may be
written as
CH = CH α αt + CH δ δe . (3.50)
The coefficients CH α and CH δ correspond to the derivatives of CH and are constant.
If the elevator is floating, then the hinge moment is zero and (3.50) may be solved
for δe to give
CH α
δe = − αt . (3.51)
CH δ
Using (3.51) in (3.41) renders
!
CH α
CL′ t = CLt α 1 − τ αt , (3.52)
CH δ

where CL′ t denotes the stick-free tail lift coefficient. The factor

CH α
Fe = 1 − τ (3.53)
CH δ
is known as the free elevator factor and, for the configuration under consideration,
Fe < 1 . Thus, the floating elevator basically renders a new, smaller slope of the
tail lift curve
CL′ t α = CLt α Fe < CLt α . (3.54)
If we use CL′ t α rather than CLt α and repeat the derivation of the location of the
neutral point along the lines of Section 3.3, then we obtain the location of the
stick-free neutral point h′n as
St
hac,wb CLwb α + ht CLt α Fe (1 − εα )
h′n = S . (3.55)
St
CLwb α + CLt α Fe (1 − εα )
S
wgd – Static Stability and Control 115

Following the simplifying assumptions which lead to (3.31), we may write

CLt α
h′n = hac,wb + Fe VH (1 − εα ) . (3.56)
CLα
We may then formulate the criteria for stick-free longitudinal static stability

CM ′α = (h − h′n ) CL ′α < 0 , (3.57)

where h′n − h is the stick-free static margin. We conclude from (3.56) that, for
Fe < 1, this is smaller than the stick-fixed static margin. Therefore, the longitudinal
stability of an aeroplane with an elevator tail degrades as the stick is freed.
116 Static Stability and Control – wgd

Stabilator and Stick-Free Trim

For the stabilator tail, we introduce a hinge moment coefficient CH similar to the
one for the elevator in (3.50),

CH = CH α αt + CH δ δe . (3.58)

For the calculation of the free elevator factor Fe of the stabilator tail, we have to
work with the absolute angle of attack αa of the entire aeroplane since αt is measured
with respect to ihs and ihs is variable. With αwb ≈ αa and it = ihs the equation
(3.17) for the tail angle of attack can be written as

αt = αa (1 − εα ) − ihs . (3.59)

Also, we recall (3.41) and (3.46), i. e.

CLt = CLt α (αt + τ δe ) (3.60)

δe = ke ihs + δ0 . (3.61)
Setting CH equal to zero and eliminating αt , δe and ihs renders, after some lengthy
manipulation,
! !
(1 − τ ke ) CH α τ CH α − CH δ
CL′ t = CLt α (1 − εα ) 1 − αa + δ0 . (3.62)
CH α − CH δ ke CH α − CH δ ke

Thus, we obtain for the stick-free case

!
∂ CL′ t (1 − τ ke ) CH α
= CLt α (1 − εα ) 1 − . (3.63)
∂ αa CH α − CH δ ke

The comparison with

∂ CLt
= CLt α (1 − εα ) (3.64)
∂ αa

for the stick-fixed case renders the free elevator factor Fe for the stabilator tail as
(1 − τ ke ) CH α
Fe = 1 − . (3.65)
CH α − CH δ ke
wgd – Static Stability and Control 117

It is a straightforward exercise to show that (3.55) – (3.57) also hold for the aeroplane
with a stabilator tail. In most stabilator configurations, we find that ke and CH δ are
negative whereas CH α is always positive. Provided that CH α < CH δ ke , we therefore
get Fe > 1 . Consequently, the longitudinal stability is improved by freeing the stick!

The linkage of the stabilator tail to the fuselage (ke and δ0 ) can be designed such
that the aeroplane, flown with a free stick, always trims itself whenever the flight
velocity is changed. For different locations of the centre of gravity, trim can be
achieved by moving the attachment point of the link to the fuselage.
118 Static Stability and Control – wgd

3.6 Directional Static Stability and Control

β
x

y
v∞

vside

We define the sideslip angle β as illustrated. Furthermore, we introduce the yawing

moment coefficient CN by writing

N = q∞ S b CN . (3.66)

For static stability, a perturbation β must generate a restoring positive yawing

moment N . Therefore, directional static stability requires that
∂ CN
= CN β > 0 . (3.67)
∂β
 wgd – Static Stability and Control 119

N In the following we neglect the contribution

of the wings & body to the yawing moment.
The moment N is then generated by the lift
CG force Lv of the vertical tail and we may write

v∞ N = − lv Lv . (3.68)

−β With
Lv = q∞ Sv CLv (3.69)
lv
and

CLv = CLv α (−β + τ δr ) (3.70)

Lv we obtain
ACv
N = lv q∞ Sv CLv α (β − τ δr ) . (3.71)

The comparison with (3.66) renders

lv Sv
CN = CLv α (β − τ δr ) . (3.72)
δr b S

Note that the first two terms in (3.72) are known as the vertical tail volume
lv Sv
VV = . (3.73)
b S
We may then define the constant coefficients
lv Sv
CN β = CLv α (3.74)
b S
and
lv Sv
CN δ = − CLv α τ . (3.75)
b S

It follows from (3.67) that the desired amount of directional stability can be achieved
by choosing a sufficiently large tail volume VV .
120 Static Stability and Control – wgd

3.7 Lateral Control L

W
z y

v∞
δL

δa = δL + δR

δR

The aeroplane can be rolled, i. e. the bank angle θ can be controlled, by deflecting
the ailerons as illustrated. The aileron deflection δa is defined as the angle between
the left and the right ailerons. The diagram shows positive aileron deflection (left
aileron up, right aileron down).

The rolling moment L (not to be confused with the lift force!) may be expressed as

L = q∞ S b Cl , (3.76)

where Cl (lower case “l” to avoid confusion with lift coefficient) denotes the rolling
moment coefficient. Since the ailerons represent the only direct mechanism to control
roll, we have
Cl = Cl δ δa . (3.77)
Note that for δa = 0 the rolling moment disappears. Thus, a perturbation θ 6= 0
generates neither a restoring nor a diverging rolling moment. The lateral static
stability is neutral (be, however, aware of the dihedral effect discussed on page 127).
kness

wgd – Static Stability and Control 121

Adverse Yaw

v∞
less lift
less drag

more lift
more drag

Due to the effect of the aileron deflections on induced drag, a positive antisymmet-
ric aileron deflection, which is intended to initiate a level left turn, generates an
undesired positive yawing moment forcing the nose to the right. This phenomena is
known as adverse yaw.

Adverse yaw can be reduced by applying asymmetric aileron deflection, such that
the upward deflection on one side is larger than the downward deflection on the other
(typically twice as big). This can be achieved by appropriately designed mechani-
cal gearing systems and reduces the drag differential significantly. The parameter
δa controlled by the pilot corresponds to the angle between the left and the right
aileron and is not affected by the mechanical linkage.
ercen

122 Static Stability and Control – wgd

Aileron Reversal

k, AC

v∞

Consider a symmetric airfoil, supported in the aerodynamic centre by a rotational

spring, with a flap. This experimental set-up is representative of the section of a
wing near the wing tip. The stiffness of the spring corresponds to the torsional
stiffness of the wing. If the spring (the wing) is rigid, then the positive deflection of
the flap will generate lift. In the following we study the situation with k < ∞.

Due to the symmetry of the airfoil the pitching moment coefficient Cm is nonzero
only for δ 6= 0, i. e.
Cm = Cm0 + Cmα α + Cmδ δ (3.78)

with Cm 0 = 0 , Cm α = 0 , Cm δ < 0 . (3.79)

The rotational displacement of the spring is equal to the angle of attack α. The
internal moment of the structure must be in equilibrium with the aerodynamic
loading. Rotational equilibrium formulated with respect to the support point may
be written as
k α = q∞ c2 Cm δ δ . (3.80)

q∞ c2 Cm δ
Solving (3.80) for α gives α = δ. (3.81)
k

Using this in the expression for the lift coefficient (1.24) renders
!
q∞ c2 Cmδ
Cl = Clα (α + τ δ) = Clα + τ δ. (3.82)
k
wgd – Static Stability and Control 123

From (3.82) we conclude that, due to Cm δ < 0, there is a critical dynamic pressure
q∞ , for which the sign of the lift turns from positive to negative. This phenomena
is known as aileron reversal. Due to warping of the wing, a positive deflection of
the flap (or aileron) at high flight velocities causes a reduction of lift rather than an
increase. In order to delay aileron reversal to higher velocities, we need to provide
more structural stiffness to the wing (increase k), which typically leads to a larger
aeroplane weight.
124 Static Stability and Control – wgd

3.8 Coupling Effects

In this section, we briefly discuss some coupling effects between lateral and di-
rectional stability and control. Generally, longitudinal motion in the xz–plane is
independent of lateral and directional motion. The latter are, however, very closely
coupled such that we often speak of lateral-directional stability and control.

Coordinated Turn
A level turn is initiated by aileron deflection which rolls the aeroplane into the desired
bank angle (compare Section 2.8). However, the roll primarily causes sideslip β and
does not directly ensure that the aeroplane turns its nose into the desired direction.
The yaw needed to coordinate the turn is, at least partly, provided by the vertical
stabiliser, i. e. by CN β > 0 (see (3.67)). However, specifically in the presence
of adverse yaw, the vertical stabiliser may not generate sufficient yaw. A certain
rudder deflection is then required to achieve a coordinated turn, such that the x–axis
is always tangential to the flight path. In a coordinated turn the yaw rate and the
bank angle are consistent, such that the gravity experienced by the pilot and the
passengers is always perpendicular to the wings, i. e. they do not feel subjected to
any forces to the right or to the left.
wgd – Static Stability and Control 125

Asymmetric Engine Failure

Consider a twin engine aeroplane with one ye

engine suddenly failing. The pilot has to
trim the aeroplane with only one engine T
and then try to reach the nearest airport.

N
D

Lv
Rotational equilibrium in the xy–plane requires

TR ye + N = 0 . (3.83)

For trimmed controllable flight the sideslip has to vanish (or at least be very small),
such that β = 0 . Recalling TR = q∞ S CD and using (3.71) in (3.83) then gives
Sv
CD ye − lv CLv α τ δr = 0 . (3.84)
S
lv Sv
Solving for δr and using VV = renders
b S
CD ye
δr = , (3.85)
VV CLv α τ b
which is the rudder deflection needed to balance the asymmetric thrust. Note that
(3.85) is important for the design of vertical tail and rudder, since the vertical tail
volume VV must be large enough to keep δr in the linear range, i. e. to avoid stall
at the vertical tail.
126 Static Stability and Control – wgd

Lv
−θ

It remains to balance the sideways lift force generated by the rudder deflection. This
can be done by rolling the aeroplane such that a component of weight is used to
counterbalance the sideways lift Lv . Assuming that θ is small we may write

− W θ = Lv . (3.86)

Using (3.85) and L = W for small θ we obtain

Lv q∞ Sv CLv α τ δr CD ye
θ = − = − = − . (3.87)
L q ∞ S CL CL lv

minimum control speed: A sudden asymmetric engine failure will result in a

sudden yawing moment. This is accompanied by a rolling moment due to the lift
differential on the two wings, which results from the different airspeeds of the wings.
The pilot needs to react quickly with the appropriate rudder and aileron deflections.
This requires the control surfaces to be sufficiently effective at the current flight ve-
locity. Recalling that rudder and aileron lift forces are proportional to q∞ , it follows
that there exists a minimum velocity below which control of the aeroplane can not
be recovered.
Similarly, the equations (3.85) and (3.87) suggest that there is a maximum drag
coefficient CD beyond which the rudder deflection δr and the bank angle θ become
impractical. Recalling that the (induced) drag increases as the velocity decreases,
we conclude that this renders a minimum flight velocity for which asymmetric thrust
can be counterbalanced by rudder deflection and roll.
The minimum velocity for which controlled trimmed flight can be recovered and
maintained is known as the minimum control speed vMC . It represents a critical
design factor. Engine failures at smaller flight velocities usually result in fatal acci-
dents.
wgd – Static Stability and Control 127

Rolling Moment with Sideslip – Dihedral

Generally, it follows from positive sideslip (sideslip to the right) that the lift force
generated by the right wing is larger than the one generated by the left wing and
vice versa. The lift force differential then causes a rolling moment, which rolls the
aeroplane towards the direction opposite to the sideslip. This phenomena is known
as the dihedral effect and may be expressed as

Clβ < 0 . (3.88)

Generally, a value of Cl β slightly smaller than zero is desirable for the handling of
the aeroplane. If Cl β is positive or |Cl β | is too large, then the aeroplane becomes
difficult to fly.

Now consider a positive rolling perturbation, i. e. the aeroplane is suddenly banked

by the angle θ. Consequently, the lift force is tilted to the right and causes positive
sideslip. Due to (3.88) a negative restoring rolling moment is generated. Thus, the
dihedral effect may be seen as providing static stability with respect to perturba-
tions of the bank angle.

The parameter Cl β is determined mainly by the dihedral angle Γ. For each wing, the
sideslip velocity may be decomposed into components parallel and perpendicular
to the wing surface as illustrated. Depending on its orientation, the perpendicular
component displaces the freestream velocity v∞ such that the angle of attack de-
creases or increases. Therefore, we conclude that the larger Γ is chosen, the more
lift is generated on the wing facing the sideslip wind, and consequently the larger is
the resulting rolling moment.

In the presence of wing sweep, the component of the sideslip parallel to the wing
surface has a similar effect as is illustrated below. However, we recall that the pri-
mary design criteria for wing sweep is the drag divergence at high flight velocities.

Finally, the dihedral effect depends significantly on the placement of the wings on
the fuselage. In fact, some aeroplanes with high wings (and/or large wing sweep)
have a negative dihedral angle Γ in order to avoid an excessive dihedral effect.
128 Static Stability and Control – wgd

decrease angle of attack increase angle of attack

∆α = −β Γ ∆α = +β Γ

Γβv∞ βv∞
βv∞ Γβv∞

Γ Γ

βv∞
βv∞

v∞ cos(Λ + β) v∞ v∞ cos(Λ − β)
v∞
> v∞ cos(Λ + β)
Λ

high wing
→ more dihedral

low wing
→ less dihedral
wgd – Static Stability and Control 129

Other Coupling Effects

rolling moment with rudder deflection: In a classically configured aeroplane,

the vertical tail with the rudder is situated at some distance above the centre of
gravity of the aeroplane. Consequently, any horizontal lift force of the vertical tail
and rudder generates a rolling moment with respect to the centre of gravity.

rolling moment with yaw rate: While the aeroplane is yawing in the positive
direction, the left wing experiences a larger freestream velocity than the right wing.
This lift force difference results in a positive rolling moment. Vice versa, a negative
yaw rate produces a negative rolling moment.

yawing moment with roll rate: While the aeroplane is rolling, the freestream
velocities on the two wings are displaced by a component vertical to the wing.
Consequently, the lift force is tilted forward on one wing and backwards on the
other. This results in a yawing moment.

There exist other, more complicated coupling effects, which are beyond the scope of
these notes.