TU Delft

Summary

AE4240 - Advanced Aircraft Design I

January 12, 2015

Contents
1 Design objectives
1.1 Conceptual Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2 Preliminary design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2
2
3

2 Boundary layer effects and subsonic

2.1 Subsonic cruise drag . . . . . .
2.2 Friction drag . . . . . . . . . . .
2.3 Drag prediction . . . . . . . . .

4
4
5
5

cruise drag
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3 Relation between geometry and pressure distribution

4 Interference effects and area ruling
4.1 Area rule . . . . . . . . . . . . .
4.2 Interference drag . . . . . . . . .
4.2.1 Wing-body interference
4.2.2 Empennage interference
4.2.3 Nacelle interference . .

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

5 Pressure distribution about airfoils

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

6
6
7
8
8
8
8

6 Reynolds effects
10
6.1 Supercritical airfoils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
6.2 Reynold number effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
7 Low
7.1
7.2
7.3

speed and high speed

Maximum lift . . . . .
thickness and camber
Buffet onset . . . . . .

stall
11
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

8 Airfoil with high-lift devices

13
8.1 mutual interference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
8.2 stall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
8.3 design for stall characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
9 Swept wing concept
16
9.1 Wing design requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
9.2 Wing sweep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
10 Tip stall and aeroelastic effects
18
10.1 first generation swept wing aircraft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
10.2 Aeroelastic effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
10.3 forward sweep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
11 Root and Tip effects
19
11.1 Winglets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
12 Examples of modern wing design

21

13 Control surface design

22
13.1 Spoilers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
14 Horizontal tail design
24
14.1 Horizontal tail design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
15 Vertical tail design
26
15.1 vertical fins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
15.2 Rudder design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
15.3 Control surface design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
16 Propeller slipstream effects
29
16.1 lateral stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
17 Engine intakes, exhausts, and reversers
31
17.1 Engine exhausts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
17.2 Thrust reversers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
18 Stall characteristics

33

19 Take-off performance
35
19.1 Chapter 44 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
20 Flight beyond buffet onset and MMO
37
20.1 Chapter 29 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

This is a summary including both the chapters from the book and the lecture slides. The summary is
structured according to the lectures, thus section 1 refers to lecture 1 and the chapters treated during this
lecture, section 2 refers to lecture 2 and the chapters treated during this lecture, etc. In each section the
chapters treated will be mentioned.

Design objectives

(chapter 1-6, 43)

There are two steps which have to be taken into account before the design phase of an aircraft can be started.
Step one is to make a specification of the market and its operational requirements, and the economic situation.
Step two is finding a balance between the aircrafts performance and the capabilities and complexity (costs),
which determines the main requirements and main objective of the aircraft.
For example; for a transport aircraft the main objective is: To transport a payload A over a distance B between
airports of category C against minimum costs (i.e. at an optimum speed D). And some driving parameters
can be: Lift/drag ratio, CLmax , weight, safety, Stability and control, CS-25, etc.
When designing an aircraft the certification regulations (which describe the minimum requirements with
which a design must comply in order to convince airworthiness authorities that an acceptable safety standard
is achieved) and design rules (which are goals and relevant regulations incorporated due to a design
organisations previous experience, they are not necessarily achieved completely) have to be taken into
account.
The design process itself is structured into three main phases:
Conceptual design (weeks to months)
Definition of the performance goals
evaluation of possible competing concepts
generation of many possible concepts
selection of a baseline design
Preliminary design (months to years)
Refined sizing of the baseline design concept
Parametric studies
Global design frozen with the possibility to change only few details
Detail design (years)
Detailed design of the whole vehicle down to each single detail
Accurate evaluation of performances
Fine tuning of the design
Release of drawings for production

1.1

Conceptual Design

When the top-level requirements are determined the conceptual design can be started. The conceptual
design consists of the Class I and the Class II design. For the class I design first the aerodynamic shape is
determined using three design methods:

 Empirical method Handbooks which are a collection of graphs and equations - give a relation between
elementary parameters of the geometry of the aircraft and the desired characteristics of the aircraft.
Analytical method - The aerodynamic characteristics of the detailed geometry are obtained by physical
insight (pressure distributions).
Computational fluid dynamics - obtain the intended characteristics of the aircraft by directly determining the required detailed aerodynamic shape through the use of fluid dynamics and the associated
pressure distribution.
During the conceptual design a balance has to be found between required volumes, weight distribution,
main dimensions, and engine performance without being too optimistic about those parts, since being too
optimistic can cause high costs because the design has to be changed. Therefore, the designer should have
detailed knowledge about (almost) every aspect of the design.
Usually a thrust to weight versus wing loading diagram is made which determines the design point. In this
diagram the take-off, climb, cruise, descent, landing, and stall conditions are taken into account. From this
design point to find an external shape that optimizes the lift and drag characteristics attention has to be
paid to the tailplane, fin, and control surfaces to get satisfactory flight handling characteristics.
The conceptual design finally gives requirements on geometry (volume and configuration), weight, aerodynamics, handling characteristics, and cost (lift cycle and operating) which have to be used in the preliminary
design.

1.2

Preliminary design

During the preliminary design the aircraft geometry is determined. This leads to an aerodynamic design
loop. The aerodynamic design loop consists of the airplane characteristics, which determine the external
shape which determines the aerodynamic characteristics which again determines the airplane characteristics.
Taken into account are the aerodynamic design goals, which are to minimize drag and postpone separation.
The aircraft geometry is dependent on:
Internal geometry constraints
Required volume (cargo, seats, legroom, headroom, in-flight entertainment)
Shape of the fuselage (usefulness of the volume)
Access to the available volume
Internal geometry constraints related to structural consideration
thickness distribution along the span of the wing and tail surface torsion boxes
Thickness of flaps
Thickness of engine struts and pylons
External geometry w.r.t. aerodynamic considerations
External geometry w.r.t. Manufacturability and cost control
The external geometry should lead to optimum aircraft performance, satisfactory flight handling characteristics
while fulfilling the requirements on the internal geometry and producibility.
The crucial question in aerodynamic design is: What is the proper design (=target) pressure distribution?
In order to postpone flow separation and minimize drag the following goals should be kept in mind:
for components that do not have to produce resultant forces, local super velocities should be minimised

 for components that do need to produce resultant forces (wing, rudder), the pressure distribution at the
relevant flight conditions should be optimised such that the momentum loss in the boundary layer and
behind the shock wave is minimal.
for components that must tolerate a large variation in local flow direction, leading-edge shapes and
design pressure distributions must be found, which cope with this variation.

Boundary layer effects and subsonic cruise drag

(chapter 7, 8, 9, and 40)

Flow in real life displays viscosity (friction)
The flow sticks to the surface
there is a friction acting on the surface (wall-shear stress)
Development of a boundary layer along the airfoil surface
Due to the boundary layer, skin-friction drag and pressure drag have to be taken into account. Also, due to
the boundary layer, flow transition (from laminar to turbulent) and flow separation is occurring.
While the boundary layer increases, the slope of the velocity distribution becomes more shallow. This is
directly related to the shear stress, which ties to the friction coefficient, and hence the friction coefficient
decreases over the slope of a flat plate.
The shape factor is the factor necessary to scale from a flat plate to a certain body. The shape factor (H) is
dependent on both the displacement thickness ( ) and the momentum thickness ():

U =

y=0
Z y=

U U =
H=

y=

y=0

(U u)dy

(1)

u(U u)dy

(2)
(3)

Flow separation occurs when the slope of the velocity distribution at the wall is zero and becomes negative.
This is often due to a fast increasing pressure distribution.
Boundary layer transition depends on the Reynolds number, pressure gradient and receptivity, where a
laminar boundary layer separates earlier than a turbulent boundary layer because a turbulent boundary
layer has extra momentum near the wall, which makes it withstand an unfavourable pressure gradient
without separating.

2.1

Subsonic cruise drag

(chapter 40)
Cruise drag is a major driver to fuel consumption and determines the range and economics for the operator.
When the cruise drag is predicted too low the required range cannot be reached, but if the cruise drag is
predicted too high there is no advantage relative to the aircraft in service.
Each component produces drag, this drag consists of:
Zero lift drag

 Friction drag - The friction drag of a smooth flat plate with the same length as the component is
determined. The shape factor then determines the ratio between this drag and the actual drag of
the component.
Contributors to the friction drag are:
Wetted area
Surface roughness/imperfections
Shape (local supervelocities and pressure drag components)
Pressure drag
Excrescences
Interference drag
Lift dependent drag - consists of form drag and induced drag. Where the form drag consists of friction
drag and pressure drag.
Compressibility drag
Trim drag

2.2

Friction drag

Friction drag is the largest drag component in climb/cruise. However, the friction coefficient decreases with
Reynolds number, therefore larger aircraft have lower friction drag coefficients. This is only up to a certain
Reynolds number, after this critical Reynolds number the friction drag stays the same.
This is because: with increasing Reynolds number the boundary layer decreases. At this point their additional
drag is friction drag, which stays the same, and therefore, since the boundary layer decreases, the drag
decreases. When the boundary layer becomes thin enough that the grains of the material begin to protrude
from the boundary layer, a wake is produced, after which this wake will increase while the boundary layer
decreases, keeping the total drag measured constant. The critical Reynolds number depends on the relative
equivalent sand-roughness height (ks /l). Since the cruise Reynolds number is often higher than the critical
Reynolds number, reducing the grain size leads to reduced drag.
The equivalent skin friction drag is determined by the wetted area (often called parasite drag or equivalent
flat plate area).

2.3

Drag prediction

The drag of an aircraft can be predicted using the following methods:

Form factor and flat plate analysis (for zero-lift drag)
Trim drag calculated from moment distribution
Bookkeeping gives thrust/zero-lift drag/induced drag
CFD methods do not always give better results than classical methods. When wind-tunnel testing is an
option to determine shape factors which then dont have to be estimated.

Relation between geometry and pressure distribution

(chapter 7, 10, and 11)

If no (lift) forces should be generated by an aircraft part, shock waves, separation and friction drag should
be reduced or eliminated completely.
If (lift) forces have to be generated, the pressure distribution has to be optimized and the overall characteristics
of the aircraft have to be determined (CL , CL CM , etc.).
Cp and V are linked by the following equation:
Cp = 2

V
V

(4)

V 2
dp
=
r
dn

(5)

This is related to geometry by the following equation:

For a concave wall:

For a convex wall

dp
dn

dp
dn

< 0, Cp > 0, and V < 0.

> 0, Cp < 0, and V > 0.

However, these relations do not hold at the leading and the trailing edge. Besides, the effect of compressibility
has not been accounted for.
For the flow around transonic airfoils, the following relation holds (however, this relation does not take
entropy increase due to shock-waves into account):

2
Mlocal
=

2
1

1+
1+

1 2
2 M

1
2
2 M cp

1

1

(6)

Away from the stagnation areas a direct relationship exists between surface curvature and pressure coefficient.
The stronger the (convex) curvature, the higher the (positive) supervelocities and the (negative) Cp -values.
Hence, local shape changes affect the pressure coefficient and the effect on the pressure coefficient depends
on the Mach number of the flow. Gradual changes in shape result in a smoother pressure distribution
reducing friction drag and/or eliminating separation. Finally, the Mach number has an effect on the region
of influence of a body, the pressure coefficient over the body, the sensitivity of the forces with respect to the
inflow angle, the drag of the body, the effectiveness of control surfaces, and the handling characteristics of
the aircraft.

Interference effects and area ruling

(chapter 12)

4.1

Area rule

In order to reduce the drag of an aircraft the area rule is used. The area rule states that two airplanes with
the same longitudinal cross-sectional area distribution have the same wave drag, independent of how the
area is distributed laterally (i.e. in the fuselage or in the wing). Furthermore, to avoid the formation of strong
shock waves, this total area distribution must be smooth. As a result, aircraft have to be carefully arranged
such that at the location of the wing, the fuselage is narrowed or waisted, such that the total area doesnt
change much. Similar but less pronounced fuselage waisting is used at the location of a bubble canopy and
perhaps the tail surfaces.
Also, according to the area rule the wave drag produced by a slender body of arbitrary dimensions in
supersonic flow may be modelled by an equivalent body that is axi-symmetric if:
6

 The body ends with an axi-symmetric portion

The body ends in a point
The body ends in a cylindrical portion parallel to the free stream
A related concept is the Sears-Haack body, which is the shape with the minimum wave drag for a given length
and a given volume. However, the Sears-Haack body shape is derived starting with the Prandtl-Glauert
equation, which governs small disturbance supersonic flows, but this equation is not valid for transonic flows
where the area rule applies. So although the Sears-Haack body shape, being smooth, will have favourable
wave drag properties according to the area rule, it is not the theoretically optimum.
Most jet airliners have a cruising speed between Mach 0.8 and 0.85. For aircraft operating in the transonic
regime (Mach 0.8-1.0), wave drag can be minimized by having a cross-sectional area which changes smoothly
along the length of the aircraft. This is known as the area rule, and is the operating principle behind the
design of anti-shock bodies. Reducing wave drag improves fuel economy.
Wave drag is only dependent on cross-sectional area distribution. It is independent of the actual shape of
components. And it is dependent on the second derivative of area; hence small changes in area gradient
yield low wave drag and large changes in area gradient yield high wave drag. However, shortcomings are
that no lift is considered, the flow is assumed irrotational (no shockwaves are present) and no viscous effects
are considered.
An anti-shock body, (also known as Whitcomb body or Kuchemann carrot) is a pod positioned on the
leading edge or trailing edge of an aircrafts aerodynamic surfaces to reduce wave drag at transonic speeds
(Mach 0.81.0). They promote isentropic compression of the supersonic flow and postpone shock-induced
separation

4.2

Interference drag

In order to optimize an aircraft body the interference between parts has to be minimized. When having a
locally convex curvature the pressure decreases while a locally concave curvature increases the pressure.
This also means that higher local convex curvatures lead to higher supervelocities.
On aircraft parts which are not intended to generate aerodynamic forces local supervelocities should be
minimized:
Front fuselage including cockpit canopies
Centre and rear fuselage sections
Engine struts and pylons
Fins in cruise flight
Tailplane-fin fairings
Wing-body interference
interference issues are:
Every curvature generates supervelocities
Multiple surfaces close together (nacelle, pylon, wing/body/VT)
Summation of supervelocities on each component
High supervelocities can cause interference drag
In high-subsonic conditions interference drag is more pronounced than in low subsonic conditions
Using CFD to change the shape of components, the local supervelocities can be minimized.

4.2.1

Wing-body interference

When the wing and the body are considered together, supervelocities of the individual components are added.
The lift over the wing is increased due to the presence of the fuselage. Due to the presence of supervelocities
the lift distribution over the wing is altered.
4.2.2

Empennage interference

The horizontal tail produces a down-force during cruise to balance the aircraft. The vertical tail has to
minimize drag during cruise. Interference between those two can lead to lower pressure on the top surface,
reduced down force and a nose-down pitching moment.
A solution of this can be to allow the flow to expand by waisting the fuselage (hence local area ruling).
A way to reduce T-tail interference is to add a fairing between the horizontal tail and the vertical tail.
4.2.3

Nacelle interference

Constraints are:
Attached inflow in all conditions
Structural rigidity of pylon and nacelle
Sufficient space for structure and systems
Inboard supervelocities of the nacelle are always higher than outboard supervelocities.
Nacelle-wing interference can be reduced by placing the nacelle at a certain position from the wing. By
using CFD techniques it has become possible to place the nacelle closer to the wing, even to previously
thought to be unacceptable positions.

Pressure distribution about airfoils

(chapter 15)
The characteristics of a pressure distribution are:
A stagnation point at or near the leading edge (Cp = 1)
The height and location of the maximum supervelocity (Cpmin )
The ratio between Cpmin and Cp
The pressure gradient

dp
dx

behind Cpmin

The trailing edge pressure (for inviscid flow CpT E 1)

To determine the overall characteristics of the pressure distribution over an airfoil section, the characteristics
of the boundary layer have to be taken into account ( , , and cf ).
The velocity gradient is linked to the friction coefficient, a lower velocity gradient means a lower friction
coefficient.
When comparing pressure distribution over airfoils, the following can be noted:
At Cl = 0 ( = 0) Cpmin is proportional to the sections relative thickness

 With increasing , the leading edge suction peak increases much faster and is followed by a much
stronger adverse pressure gradient for the airfoil section with the small leading edge radius than the
section with the large leading edge radius.
On the thick section, boundary layer effects are much stronger than on the thin section, and flow
separation is approached much quicker
Increasing Reynolds number causes boundary layer effects to decrease.
Lift on an airfoil next to each other can be increased in two ways:
1. Increasing camber by control surface deflection - the pressure distribution over that control surface
will change (this is more effective on thin airfoil sections because at thick sections this can lead to
trailing edge flow separation).
2. Increasing the angle of attack
In order to avoid drag creep, but have an as thick as possible airfoil, a sonic rooftop shape can be selected.
This is characterized by a constant Cp = Cp from the leading edge up to a certain point.
When the freestream Mach number increases the shock moves downstream.
A stream through a nozzle has the following characteristics:
If At A the flow is chocked and sonic at the throat
Increasing back pressure reduces Me to subsonic
Shock wave initially at the exit
Large Mach number in front of shock wave
Large pressure jump over shock wave
Increasing back pressure reduces Me even more
Shock position moves closer towards the throat
Mach number in front of shock reduces
Pressure jump across shock reduces
Lower M in front of shock means the shock is pulled forward
Higher M in front of shock means the shock pushed back
The position of the shock is dominated by:
Local area ratio
Ratio between back pressure and inlet pressure of stagnant air
The result is a relation between geometry, Mach number in front of the shock, pressure jump and Mach
number behind the shock.
The NACA 6-series is developed for pressure distributions which are favourable for developing a laminar
boundary layer. However, as soon as the speed of sound is surpassed locally this laminar boundary layer
has unfavourable characteristics. This is because the supervelocities reach a maximum near mid-chord. A
small increase in Mach number then produces a sharp increase in supersonic local velocity, resulting in a
strong shock wave and flow separation at the foot of the shock.
Therefore the NACA 4-series was modified such that in sub-critical flow:
conditions the highest supervelocities were concentrated near the leading-edge (suction peak)
followed by an area with almost constant pressure
rapid deceleration behind the pressure peak

 Behind the leading edge curvature distribution has to be such that in transonic conditions the interactions between expansion waves, the sonic line and the reflected compression waves occur behind the
suction peak, leading to a weak shockwave at the end of the supersonic region.

Reynolds effects

(Chapter 15 and 18)

6.1

Supercritical airfoils

Characteristics of supercritical airfoils are:

Small drag penalty at design Mach number
High lift coefficient
Sensitive to Reynolds number and Mach variations
For supercritical airfoils first the upper surface is designed after which the lower surface is designed to
modify spanwise lift and moment distribution.
For supercritical airfoils the following is valid:
The leading edge can be modified to reduce drag creep and postpone the divergence drag. This is
due to the partial isentropic recompression on the upper surface behind the leading-edge suction peak
and the weaker shock wave terminating the area of supersonic flow. Hereby a small radius means less
drag creep and a large radius more drag creep.
The cusp at the trailing edge allows off the surface pressure recovery, hence a less steep adverse
pressure gradient. If a sharp or a blunt trailing edge is chosen depends on the design Mach number,
for high Mach numbers the drag will be lower for a blunt trailing edge, and for low Mach numbers
the drag will be lower for a sharp trailing edge.
The drag divergence Mach number is dependent on the thickness ratio of the airfoil. High thicknesses
mean lower drag divergence Mach numbers.
The sonic rooftop prevents drag creep and postpones divergence drag. The design is such that Cp = Cp
from leading edge to around 0.3-0.6 x/c. This keeps the maximum local velocity around M = 1 and
thus no mixed subsonic/supersonic flow will occur.
Relations between airfoil shape, pressure distribution, lift coefficient, and design Mach number are:
When adding thickness to the airfoil at constant camber, higher local velocities will occur due to the
stronger surface curvature. For a given lift coefficient this will lead to a lower design or drag divergence
Mach number, or for an increase in design Mach number this will lead to a thinner section.
adding thickness only to the lower part of the airfoil results in an increase in supervelocities on the
lower surface. This may only marginally decrease the lift coefficient in the design condition but may
appreciably lower the drag divergence Mach number at lower lift coefficients
Decreasing the aft loading by thickening the rear part of the airfoil results in a lower lift coefficient at
the design Mach number. A positive effect his modification leads to a less negative zero-lift pitching
moment coefficient.

10

6.2

Reynold number effects

For supercritical airfoil sections operating near their design point the upper surface has a region of supersonic
flow coupled with an almost constant static pressure. The Reynolds effects are much stronger than for subsonic
flow, this has two causes:
1. The boundary layer may be laminar up to the shock wave at low Reynolds numbers. The shock wave
terminating the supersonic region will then be a lambda shock wave which the boundary layer can
negotiate much easier than a straight shock wave associated with a turbulent boundary layer. With
increasing Reynolds number transition will occur progressively further ahead of the shock wave.
2. Increasing the Reynolds number when the boundary layer is turbulent will decrease the displacement
thickness and therefore effectively increase the flow curvature near the section surface. This will
produce a more aft shock wave position at constant Mach number and angle of attack. This again may
cause an increase in shock wave strength and even a second shock wave.
The effects of the Reynolds number increase on an airfoil with turbulent boundary layer are:
Aft moving shock wave, leading to an increase in lift coefficient at constant angle of attack due to the
larger area of supersonic flow
Before flow separation occurs the linear part of the pitching moment versus lift coefficient curve increases with increasing Reynolds number due to the rear movement of the shock, leading to a more
negative pitching moment coefficient at high Reynolds numbers
The drag rise number increases
Buffet is postponed
There are two types of separation: Shock induced separation bubble (low dependence on Reynolds number),
and rear separation (strong dependence on Reynolds number).
At high Mach numbers the Reynolds number has an increasing effect on the pressure distribution of the
airfoil. Also the lift and moment coefficients rise with increasing Mach number.
Increasing Reynold number decreases the displacement thickness of the turbulent boundary layer, leading
to:
effective increase in curvature
Higher supervelocities
More aft position of the shockwave
May cause increase in shock strength or second shock

Low speed and high speed stall

(Chapter 17 and 19)

When the flow separates a plateau is visible in the pressure distribution.
There are three types of stall
1. Leading edge stall - characteristics are:
Abrupt stall
Flow separation over the entire airfoil
Present on airfoils with moderate leading-edge radii
Leading edge radius has significant effect on maximum lift coefficient
11

2. Trailing edge stall - characteristics are:

Gradual stall
Flow separation moves forward with angle of attack
occurs on airfoils with large leading edge radius and strong upper surface curvature
3. Thin-airfoil stall - characteristics are:
Gradual stall at low angle of attack
Occurs of airfoils with a sharp leading edge or low Reynolds numbers
Development of a laminar separation bubble - this lowers the suction peak and thus gives a
lower adverse pressure gradient.
From the pressure distribution it can be concluded if the flow is separated or not. If the flow is separated the
pressure coefficient at the trailing edge is negative. The point of separation is when the pressure distribution
shows a plateau (or a straight distribution curve).

7.1

Maximum lift

Up until a Reynolds number of 10 million an increase in Reynolds number for a constant Mach number
means a thinner relative boundary layer, which tolerates larger unfavourable pressure gradients, increasing
the maximum lift.
The Reynolds number also decreases the minimum pressure coefficient until the critical pressure coefficient,
after this value shock waves will occur and the minimum pressure coefficient will increase again. This is
because of:
Larger Rec increases tolerance to strong adverse pressure gradient behind pressure peak
Airfoil can achieve higher angles of attack without leading edge stall
The result of this is that higher lift coefficients can be achieved
Increasing R ec moves the trailing edge separation point upstream
The result is that the lift coefficient decreases
Up until the Mach number where the flow at the leading edge reaches a local Mach number of 1 the maximum
lift coefficient and minimum pressure coefficient are independent of Mach number.
Hence, for Mloc < 1 the Reynolds number is limiting for the maximum lift coefficient, for Mloc 1 the Mach
number is limiting for the maximum lift coefficient.

7.2

thickness and camber

Increasing the leading edge radius leads to decreased overspeeds, which reduces the adverse pressure
gradient and thus postpones leading-edge separation.
When an airfoil section has a sharp leading edge with a small leading edge radius the flow has, at large
angles of attack, to deal with a strong curvature leading to high velocities. When the leading-edge radius
is increased without altering the upper surface curvature distribution the peak velocities can, at the same
angle of attack, be lowered, due to less severe curvature, without affecting the high-speed characteristics.
Consequently the angle of attack can be increased to higher values than the basic section before the suction
peak collapses, leading to a higher maximum lift coefficient.
The shape of the lower surface section has no effect on the maximum lift coefficient.

12

7.3

Buffet onset

Buffet is a form of airframe vibration caused by pressure fluctuations in separated flow felt by the occupants
in the cockpit and cabin. Buffet can appear in different forms:
Low speed buffet due to flow separation close to stall
Buffet due to lift dumper or speed-brake extensions
Buffet due to local flow separation
High speed buffet due to flow separation caused by shock waves
High speed buffet occurs when a separation bubble, which starts at the foot of the shock wave and gradually
increases rearwards, reaches the trailing edge or when the boundary layer separates near the trailing edge
before the shock wave becomes strong enough to cause separation at its foot.
Trailing edge pressure divergence is used in wind tunnel tests as an indication of buffet onset on the real
aircraft.
High speed buffet develops gradually, in particular with increasing angle of attack. If this flow regime is
penetrated too far it may result in structural damage.
There is a margin between normal flight conditions and buffet onset both with regard to speed and to angle
of attack because:
Manoeuvre in cruise flight
Deal with disturbances due to turbulence, be it in speed or in normal load factor
Deal with upsets due to aircraft system failures
Buffet onset occurs when when the local Mach number in front of the shockwave reaches a given value
(Mloc = 1.35 1.50).
The buffet determines:
Maximum CL at certain M
Maximum M at certain CL
ceiling at which the aircraft can fly at a certain M

Airfoil with high-lift devices

(chapter 25)
The high lift devices are installed to:
Increases the maximum lift coefficient for landing with acceptable drag penalty
Reduces the stall speed of the aircraft
Allows for shorter landing length
Increases the maximum lift coefficient with acceptable drag penalty during take-off
Reduces the minimum unstick speed
Reduces the take-off field length
Reduce the pitch angle at low speeds (for trailing-edge devices)
The result is that the wing sizing is done based on:

13

 Cruise performance
Climb performance
Field performance
Volume considerations
Three effects determine the increase in maximum lift by the use of high-lift devices:
increase in camber
Higher camber produces more lift but also more drag (hence variable camber = high lift devices)
More camber leads to more circulation at the trailing edge and thus in more lift at constant angle
of attack. (where the angle of attack is the angle between the chord line of the basic section and
the direction of the undisturbed flow)
Increase in effective chord/effective wing area
Mutual interaction effect
This is for a continuous camber line, if the various components are decoupled and moved a small distance
away from each other, the lift at the leading and trailing edge of each part will be zero. However, the
mutual interaction of the vortices will remain and the total chordwise lift distribution over the compound
airfoil section will be similar that on a single (highly curved) section.
The insertion of slats gives the following characteristics:
Suppress the pressure peaks on the proceeding components through mutual interference.
Create a new boundary layer on each component postponing stall
Generate additional drag

8.1

mutual interference

Each lifting airfoil can be represented by a vortex.

The main body (centre wing) experiences downwash of the preceding vortex (slat) and upwash of the succeeding vortices (flaps), this leads to a decrease of lift and suction peak on the front part and increase of lift
on the rear part. Hence the slat, which experiences only upwash, has a high increase in lift. And thus the
last flap component has very little lift (however, the effect on the preceding components is large, increasing
the overall lift).
The upwash creates a velocity vector on the succeeding vortex. Due to the vertical component of this velocity
vector the velocities on the lower side of the airfoil are reduced and the lift is increased. Due to this upwash
effect on the lower side the stagnation point shifts either aft to the lower side of the airfoil, resulting in a
higher pressure peak, higher supervelocities on the upper side, and lower supervelocities on the lower side.
Or aft on the upper side, resulting in a lower pressure peak, lower supervelocities on the upper side, and
higher supervelocities on the lower side.
The mutual interaction of the vortices of the individual components ensures that the total chordwise lift
distribution over the compound airfoil section is similar to that on a single, highly curved section. The gaps
between the individual components postpone separation to higher angles of attack and are therefore highly
desirable.
By separating each component, every component will form its own boundary layer resulting in wake. This
new boundary layer postpones separation. The wake of the elements grow if:
1 dCp
0.007
>
1 Cp dx

14

(7)

In practice this usually means that no wake instabilities occur before the boundary layer of the underlying
component separates.
Due to earlier separation merged boundary layers should be avoided.
The combined effects on each component of being placed in a flow field with upwash or downwash, the
smaller chord length and the dumping velocity make that compound airfoil sections can reach much higher
CLmax than single airfoil sections.
The highest CLmax is reached when the wake of the preceding component and the boundary layer of the
succeeding component do not merge. However, if merging occurs the lift-drag ratio improves.

8.2

stall

Types of high-lift device separation:

Trailing edge of the flap at small angle of attack
Trailing edge separation just in front of the flap
Leading edge separation - abrupt stall.
Flow separation when slats are added:
stall on the slat
Trailing edge separation on the main wing
leading edge of the main wing

8.3

design for stall characteristics

The most effective high-lift configuration is one where separation is reached on all components simultaneously
over the full span of the wing. On an aircraft this may produce unacceptable flying characteristics.
In order to find the highest maximum lift, the largest flap setting should be selected with or without only
limited flow separation on the flap at low angles of attack. The slat however should be set at such an
angle that trailing edge separation occurs near the trailing edge on the main component, just prior to flow
separation on the slat.
At high Reynolds numbers on modern airfoil sections with effective high-lift devices this limiting peak Mach
number occurs at low free-stream Mach numbers, which lie in the range of Mach numbers where on actual
aircraft the maximum lift coefficient is determined. The maximum lift coefficient becomes then a function of
aircraft weight and flight altitude.
Reynolds and Mach effects:
High Reynolds numbers allow high CPmin values
A high CPmin implies high supervelocities over the leading edge
Combined with high take-off and landing speeds this may lead to local supersonic flow
As a general rule Mloc < 1.58, which limits CLmax
Minimum air pressure may then be less than 30% of ambient
This result can be properly translated to 3D wings.

15

Swept wing concept

(Chapter 20)

9.1

Wing design requirements

Wing design requirements are:

At cruise Mach number and cruise altitude
Design lift coefficient at top of climb
Lift-to-Drag ratio at design lift coefficient
Pitching moment coefficient at design lift coefficient
Pitching moment distribution with span (to limit structural warping)
At low speed conditions
CLmax for all aircraft configurations (clean, take-off, landing)
Stalling characteristics for all aircraft configurations over the complete c.g. range (pitch and roll)
Maximum lift-to-drag ratio (to meet climb requirement)
Lift curve slope and maximum angle of attack in ground effect (for rotation requirement)
Stability and control (everywhere in the flight envelope)
Pitching moment due to angle of attack
rolling moment due to sideslip
Rolling moment due to aileron deflection/spoiler deflection
Around the boundaries of cruise flight conditions
Margin (CL and M) between design point and buffet onset boundary
Acceptable stability and control (pitch and roll) between buffet onset and maximum buffet penetration boundary (CL and M)
Acceptable stability and control between MMO and MD .
Structural constraints
Wing weight
Tank volume
Landing gear volume
Space for systems: kinematic systems for wing movables, anti-icing systems, fuel systems, hydraulic system, electric system
Lightning strike protection
Fixation points for: engines
The prime characteristics of wing design are:
1. Aerodynamic: Mdesign and CLdesign .
2. Geometric: Aspect ratio (A), taper ratio (), twist distribution (), sweep angle (), and airfoil section
in the outboard wing.

16

9.2

Wing sweep

At the start of aviation, swept wings were not in use, mainly due to not large enough wind tunnels to do
correct tests. The first swept wing was implemented not to reduce drag, but to account for a miscalculation
in the centre of gravity position.
The main reason for applying sweep is to increase the drag-divergence Mach number. Wing sweep also
affects other aerodynamic parameters such as lift curve slope.
Swept-wing aircraft will have a lower maximum lift coefficient and will therefore have to pay more attention
to low speed performance and often require high-lift devices.
Wing sweep is introduced because it decreases the effective curvature, which leads to lower overspeeds,
which makes that strong shock waves occur at higher Mach numbers, which postpones the drag divergence
to higher Mach numbers.
Wing sweep influences the flow over the wing in the following ways:
Near the wing root the pressure distribution deviates increasingly from the prediction according to
simple sweep theory
The increasing supervelocities over the aft part of the wing root region may lead to the formation of
shock waves and flow separation.
These may then spread further outboard and lead to a sharp drag rise much earlier than would be
expected from simple sweep theory
At M = 0.95, sweep theory looses meaning.
At low speed the sweep lowers maximum section lift coefficient of the outboard wing/tip, this leads to a
reduction of the wing maximum lift coefficient, therefore effective high-lift devices are required for landing
and take-off.
From the relation between the effective and the streamwise lift coefficients it can be concluded that the lift
coefficient based on the free-stream velocity will be lower than the section CL if wing sweep is applied. This
means that high-speed aircraft with highly swept wings require more attention to their performance in the
low-speed regime than aircraft with straight wings because their maximum lift coefficient is lower.
As a consequence a swept-wing aircraft will often require powerful high-lift devices in order to show satisfactory take-off and landing performance for a given wing area.
Flow from infinity will initially move outboard at the stagnation point. Following the contour along the
top surface, the flow will experience an acceleration inboard which reduces along the chord, this bends the
streamlines inboard. This is due to an increase in effective velocity on the upper side.
On the lower side the effective velocity decreases, this curves the streamlines outboard. For a wing with
forward sweep the situation is reversed.
Inside the boundary layer the kinetic energy of the air particles is reduced towards the surface, but the
pressure gradient remains the same. The boundary layer air will therefore tend to move the lower pressure
regions, = perpendicular to the isobars. On an aft swept wing this will force the boundary layer outboard.
Hence the boundary layer on the inboard wing is thinner, and on the outboard wing thicker than on the
comparable straight wing. This suggests unfavourable aerodynamic characteristics because flow near the tip
would separate earlier than expected based on the airfoil section characteristics.
At the wing the pressures are higher over the forward part of the section and lower over the rear part. The
reverse occurs at the wing tip, this effects becomes stronger with increasing Mach number.

17

10

Tip stall and aeroelastic effects

(chapter 20 and 21)

Swept wings sometimes have fences in order to:
Align the streamlines in high-speed conditions
Prevent tip stall by locally renewing the boundary layer
Postpone stall through mutual beneficial interference
The oldest device for preventing tip stall is a vertical plate fitted on the wing upper surface in a streamwise
direction, thus forming a physical barrier for the boundary-layer crossflow, the full-chord fence.
A fence placed on the wing provides a physical barrier for the boundary layer crossflow and creates an
increase in supervelocities inboard and a decrease outboard of the fence thus altering the shape of the
isobars and causing early flow separation inboard of the fence.
high supervelocities can lead to premature separation inboard of the fence. Optimizing the shape and
spanwise position of a fence is done experimentally. No sufficiently accurate computer models exist.
The lift coefficient at which the pitch-up tendency starts (and thus the maximum usable lift coefficient) is
increased considerably over the complete Mach number range investigated.
A disadvantage of wing fences is the increase in drag. Although their surface area may be small they may
produce as much as 20 drag counts. For this reason on modern aircraft wing fences are only applied when at
a late stage in the development or during flight testing stalling characteristics are found to be unsatisfactory.
Instead of fences also shark or dog teeth (which are local leading edge extensions), saw cuts, or leading
edge boundary layer fences can be placed. These create a streamwise vortex at high angles of attack such
that the boundary-layer cross flow on the inboard wing is swept inboard.
In order to tailor the stalling characteristics the leading edge radius is increased without changing the upper
surface, this as not to change the high-speed characteristics.
The effects of wing sweep are:
Increases drag divergence Mach number
Reduces maximum lift coefficient
Reduces lift-curve slope
Shifts aerodynamic centre and centre of pressure aft
The result of this is:
Fancy high-lift devices required to increase maximum lift coefficient
Geometrical modifications to prevent tip stall required

10.1

first generation swept wing aircraft

The first swept wings were designed along the same principles as straight wings, this led to unsatisfactory
stalling characteristics, and pitch-up tendencies both at high and low speeds due to tip stall.

10.2

Aeroelastic effects

In a swept wing is the main load-carrying component the torsion box consisting of front and rear spar, ribs
and upper and lower skin panels. In flight the lift forces will, partly counteracted by the weight of the wing

18

itself, the fuel, and the engines make the wing flex upward.
The torsional deformation about the aeroelastic axis is limited.
When ailerons are deflected the extra lift applies behind the aeroelastic axis of the torsion box producing a
torsion moment. A downward aileron deflection leads to a leading-edge-down torsion moment twisting the
torsion box an producing a decrease in local angle of attack. The resulting change in lift on the outer wing
counteracts the lift due to aileron deflection lowering the aileron efficiency. At high dynamic pressures the
structural deformation may lead to aileron reversal.

10.3

forward sweep

The advantages of forward sweep are:

Not prone to tip stall
No asymmetric wing drop
Aileron control up to high angle of attack
Higher sweep angle for shock wave for given c/4
Either less geometric sweep is desired for the same shock sweep
or for the same geometric sweep there is less wave drag
Possibility of NLF
Reduction of leading edge sweep which leads to a reduced attachment line instability.
The disadvantages are:
Highly swept trailing edge: reduces effectiveness of high-lift devices
Could reduce divergence speed (or have a heavier wing)
Reduced stability in Dutch roll mode
Root stall can cause rapid loss in lift (abrupt stall) and pitch-up
In order to postpone root stall a close coupled canard can be sued to decrease the effective angle of attack
near the root.

11

Root and Tip effects

(chapter 22 and 40)

On a swept wing with constant airfoil section and near-constant wing-fuselage junction, in subsonic flow the
isobars tend to curve rearward at the wing root and forwards at the wing tip.
When a shockwave occurs, this occurs further rearwards at the root than at the tip, which results in lower
drag rise Mach number than would be expected.
For an infinite 45 swept wing the isobars curve across the plane of symmetry perpendicular, causing a
decrease in supervelocities over the forward part of the centerwing and an increase over the rear part. These
changes lead to an unfavourable pressure gradient, increasing the possibility of boundary layer separation.
Tip shape for swept wings are disadvantageous from a drag point of view because suction forces on the
outer wing are decreased. High-speed characteristics are however improved by delaying shockwaves and
increasing the drag-rise number.

19

The amplitude of the pressure distribution can be scaled by the thickness. The thickness should be increased
to increase the amplitude of the pressure distribution.
Hence, in order to improve the velocity distribution due to thickness at the root, the thickest point should
move forward and the overall thickness should be increased.
On a swept forward wing there is boundary layer inflow towards the root, which causes the wing root to
stall first and produce a pitch-up. Especially with the T-tail situation may occur where the aircraft cannot
recover from stall.
Near a straight sided fuselage wall the flow over a wing is similar to the flow over the centre part of the
wing alone, for swept-back wings.
Vortices on a swept-back wing will show similar patterns as isobars near the wing root, but not identical.
Vortices tend to curve backward and across the plane of symmetry perpendicular. This causes a decrease in
lift at the wing root.
Taper and sweepback increase the loading of a high transonic cruise optimized wing in the outboard part,
this compromises the design for high maximum lift.
improving the velocity distribution due to lift at the root;
Lift adds velocity distribution to thickness-induced velocity distribution
Sweep concentrates lift near trailing edge
The solutions are:
Reduce camber of root profile to shift pressure peak more forward and also reduces lift.
Increase root incidence to increase lift coefficient to match desired pressure distribution on top surface
In practice combine thickness, camber and incidence measures to have an acceptable isobar pattern on the
top surface at the design condition (Cl and M) The result could be too much lift inboard for elliptical lift
distribution.
The effect of wing taper and sweep on spanwise lift distribution:
The upper surface pressure distribution is constant over the span
The required reduction of lift on the inboard wing is achieved by increasing the root section thickness
via the lower airfoil contour
This has a significant benefit in wing weight, stiffness and fuel volume.
Minimizing root effects:
Difficult to design a wing with straight isobars between fuselage and tip
Geometric sweep affects isobaric pattern
Thickness affects isobaric pattern
Circulation distribution affects isobaric pattern
Near the root various design changes should be made
increase in thickness
thickest point moves forward
Increase in incidence angle
Decrease in camber (or even negative camber)

20

Rounded tip sweeps isobars aft: more form drag near the tip; less wave drag at high speeds.
To obtain a similar velocity distribution at the root and mid semi span the forward part of the root section
must be thickened and the rear part made thinner. The opposite applies to the tip.
By giving a negative camber line the loss of lift in swept wings can be compensated
A constant percentage isobars can be achieved by increasing supervelocities on the lower surface of the
inboard wing.

11.1

Winglets

End slats increase lift curve slope and decrease induced drag.
Tip vortex due to pressure difference on upper and lower wing. Induced angle of attack is produced, causes
induced drag.
The classic winglet design involves:
leading edge behind thickest point to prevent adverse interference effects
Winglet has higher camber than wingtip
Winglet has higher critical Mach number than wing tip
winglet loading same as wing loading
Elliptical lift distribution preserved
The shape of winglets:
Large enough radius of inner curve to prevent interference drag
Drag decrease dependent on toe in toe out angle with respect to interference
Result: only drag decrease at particular lift coefficients
While an equivalent increase in wingspan would be more effective than a winglet of the same length, the
bending force becomes a greater factor. A three-foot winglet has the same bending force as a one-foot
increase in span, yet gives the same performance gain as a two-foot wing span increase.
Wingtip/winglets are influenced by:
Root bending moment (depends on control surface deflection and roll manoeuvres)
Flutter considerations
Stability and control considerations
therefore: winglets could be problematic as add-on, wing should therefore be designed with winglets.

12

Examples of modern wing design

(chapter 23 and 24)

In order to get straight isobars over most of the wing upper surface, the following modifications can be made:
Furthermore the following has to be taken into account:
Mloc < 1.2 to prevent drag creep
If Mloc = 1.35 1.45: separation and buffet

21

mod. no.
1

Modification
Increase the thickness of the forward part of
the root section. Decrease the thickness of
the rear part of the root section.
Increase the thickness-chord ratio of the root
section

Decrease the positive camber of apply negative camber on the root section.

Increase the incidence of the root section.

Modify the wing lower surface along the

span (mostly on the inner wing).
Modify the lower surface velocity distribution
on the root section regarding front and rear
loading.
Modify the leading-edge region on the outer
wing.

Reason
To obtain similar chordwise upper surface velocity distributions due to thickness along the
span.
To obtain identical chordwise upper surface
velocity distributions due to thickness along
the span.
To adapt the pattern of the chordwise uppersurface velocity distribution due to lift to that
of the basic airfoil section.
To obtain identical chordwise upper surface
velocity distributions along the span.
To obtain the desired spanwise distribution
of the local lift coefficient.
To minimise the wing pitching moment.

To obtain satisfactory stalling characteristics.

Pressure gradient should never cause trailing edge separation - this limits the maximum amount of aft
loading
Shock wave position should not be too much aft
The bottom side of the wing can be modified to get the target lift distribution
High aft loading outboard results in highly negative Cm -Increased front loading near root can compensate this
Increase in nose radius to postpone tip stall
Low sweep trailing edge fro flap effectiveness

13

Control surface design

(chapter 33 and 34)

Control surfaces perform three functions:
1. Provide a means to achieve equilibrium (either trim or at non-zero control force)
2. Allow for manoeuvring in pitch roll and yaw in all flight conditions (cross wind landing)
3. counteract gusts for flight path tracking
Requirements are:
High deflection without separation
Linear behaviour over the entire range of deflections
The linearity between lift force and control surface angle decreases with increasing control-surface-chordto-airfoil-chord ratio.
On a swept airfoil the root of the rear spar is a heavily loaded structural element. To maximise the spar
height may therefore seem attractive. The resulting increase in the section tail angle will however decrease
the aerodynamic effectiveness of the rudder. Therefore a compromise has to be made.

22

When designing an aileron the following has to be taken into account:

Effect increases linearly with speed
Linear effect of dynamic pressure
Increasing the Mach number results in compressibility effects
Beyond Mmo shock waves in combination with flow separation
Aeroelastic deformation reduces aileron effectiveness
The deformation is lessened at higher altitudes
Vortex generators on the wing can be an aerodynamic fix if the aileron effectiveness decreases due to
separation or boundary layer: They can prevent:
Leading edge separation
Separation at the foot of the shock wave
Separation at the hinge line
The type of aerodynamic balance on control surfaces has a large effect on the linearity of the hinge moment
coefficient versus angle of attack and versus control surface deflection.
Dorsal fins:
Increase the maximum sideslip angle which can be reached before control is lost
Increase the sideslip angle where a significant rudder pedal force lightening and eventually rudder
lock occurs

13.1

Spoilers

Spoilers have three functions:

Roll control
Speed brakes in flight
Lift dumpers during decelerating ground runs in landings or aborted take-offs.
Wing mounted control surfaces spoil the flow over the wing, which reduces lift and increases drag.
Spoilers have three aerodynamic properties:
decrease lift
increase drag
change pitching moment
However, for each of the three given functions only one of the three aerodynamic properties are necessary,
thus:
For roll control through spoiler deflection on one wing half also the overall lift of the aircraft is affected
With symmetric deflection of spoiler panels as speed brakes also the wing lift is affected and the
longitudinal behaviour of the aircraft.
Spoiler deflection in the lift dumper function also produces a change in the pitching moment and hence
may either increase or decrease the nose wheel load.
These unintended effects require often a complicated mixing of the various functions and highly non-linear
relations between control wheel force and individual panel deflection angles.

23

 At touch down major portion of weight carried by wing

This leads to low braking forces
Lift spoiling allows high wheel loads and thereby higher deceleration.
For spoilers there are closed-shroud and open-shroud systems:
Open shroud spoilers show strong non-linear behaviour
The control response is dependent on angle of attack and flap angle
Closed shroud spoilers aim to decouple the air-brake function from the other spoiler functions and from
the lift generation
Most current transport aircraft accept the disadvantages of open shroud spoilers because of reduced
hardware complexity
Some design considerations of spoilers:
Strong interaction between spoiler functions
Forces and moments are non-linear with:
Spoiler deflection
Flap deflection
Angle of attack
Mach number
Dynamic pressure
System redundancy requires multiple spoiler panels
Complex mixing schedules are required

14

Horizontal tail design

(chapter 30 and 31)

Effect of planform parameters:
Aspect ratio, A
High A high
Low A low

dCL
d ,

dCL
d ,

low stall

high stall

Little effect on CLmax

Taper ratio,
Little effect on

dCL
d

Effect on CLmax depends on c/4

High le yields vortex lift and high CLmax
Low le and high results in tip stall due to low R etip and low CLmax
Sweep, c/4
high c/4 low

dCL
d ,

high stall

L
low c/4 high dC
d , low stall

24

 Effect on CLmax depends on formation of leading edge vortex

14.1

Horizontal tail design

Tail surfaces perform three functions

1. They provide static and dynamic stability
2. They enable aircraft control
3. They provide a state of equilibrium in each flight condition
An important driver of the horizontal tail design is: The ability to maintain a state-of-equilibrium, particular
in extreme flight conditions.
Design requirements are:
1. They shall provide a sufficiently large contribution to static and dynamic longitudinal, directional, and
(lateral) stability.
Requires: high aspect ratio, minimum sweep (for high aspect ratio) and high

dCLh
de Sh .

2. They shall provide sufficient control capability.

Requires: high aspect ratio, minimum sweep (for high aspect ratio) and high

dCLh
dh Sh .

3. Control shall be possible with acceptable control forces.

Requires: High aspect ratio of the control surfaces, and F = Ge Ch 21 V 2 Sc cc
4. Shall be able to handle high angles of attack.
Requires: Low aspect ratio of the control surfaces, Sweep is beneficial.
Limiting: High speeds in combination with deflected flaps, icing conditions
5. The tail surfaces shall be able to provide a maximum force sufficiently large to balance the total tail-off
forces and moments so that static equilibrium is achieved in all flight conditions.
Requires: Sufficiently large tail surface, sufficiently high maximum lift coefficient for both a range of
control surface deflections and the effect of ice roughness.
Static equilibrium is to be achieved for: all aerodynamic conditions and all centre of gravity positions.
6. They shall be able to handle high Mach numbers without flow separation.
Requires: Mach number at which extensive flow separation starts above MD for control deflection
necessary to pull up at n = 1.5
Furthermore:
high aspect ratio higher weight, flutter analysis for T-tails.
A few degrees anhedral (negative dihedral) has a very beneficial effect.
High taper could lead to premature tip stall, lower weight.
Sweep makes stall more gradual
CLh depends on:
Flow parameters (Reynolds number, Mach number, Angle of attack)
Shape parameters (planform shape, airfoil shape, control surface deflection, chord ratio, gap/overhang,
tab deflection, horn geometry)

25

 For wings with the same Aspect ratio, increasing the sweep leads to increased stall angle of attack
and increased maximum lift coefficient.
Main types of aero-elastic deformation are:
Wing bending and torsion
Result in:
Re-distribution of spanwise and chordwise loading
Reduction in wing lift curve slope
Reduction in aileron effectiveness
A forward shift of aerodynamic centre position on swept wings
Fuselage bending and torsion
Results in: reduction in stabiliser and fin lift curve slope.
Solution: down-rig the elevator leads to negative zero-lift pitching moment which tends to bend
the rear fuselage and tailplane attachment upwards (increasing longitudinal stability)
Stabiliser or horizontal tail bending and torsion
Fin or vertical tailplane bending and torsion
results in: lower fin lift curve slope, reduced rudder effectiveness
Deformation of movable stabiliser attachment.
Results in: reduction in stabiliser lift curve slope
Effect of ice:
In certain atmospheric conditions ice may accrue on the leading edges of wings and tail surfaces
This leads to a strong reduction of the stall angle of attack and thereby to a much reduced lifting
capability
Ice accretion on tail surfaces may therefore lead to sudden dangerous pitching moment changes
Ice/snow accretion on the ground can lead to significant maximum lift loss, and must therefore be
removed before flight.
In certain flight conditions ice accretion on tail surfaces may lead to potentially catastrophic situations.
Negative tailplane stall combined with elevator lock can lead to unrecoverable pitch-down situations
Ice protection on the leading edges allows to size the tailplane closer to the clean tailplane stall limits
Usually the vertical tailplane is much less critical (highest lift demand on take-off, or landing)

15

Vertical tail design

(chapter 32 and 33)

15.1

vertical fins

Vertical tail surface (fins) are installed to:

1. Provide directional stability in all flight conditions

26

2. Provide yaw control in all flight conditions

3. Ensure safe handling during engine failure conditions
Important design drivers are:
Minimum control speed with OEI, right after take-off:
Relatively low speed
High trust setting
High angle of attack
This leads to: high yaw rate and large side-slip angles when one engine fails.
Maximum cross-wind capability
Low speed
High side-slip angle
Design requirements are:
Shall provide a sufficiently large contribution to static and dynamic stability (directional and lateral)
Requires: High aspect ratio, minimum sweep (for high A), high

dCLv
dv

Sv

Shall provide sufficient directional control capability

Requires: high aspect ratio, minimum sweep (for high A), high

dCLv
dr

Sv .

Directional control shall be possible with acceptable control forces

Requires: high aspect ratio of the control surfaces, F = Ge Ch 21 V 2 Sr cr .
Shall be able to handle high angles of sideslip
Results in: low aspect ratios, large leading edge sweep or dorsal fins
Limiting conditions are: Low speed and high cross wind, OEI conditions right after take-off.
Shall provide sufficiently large forces to balance OEI moments and provide equilibrium
Static equilibrium is to be achieved in all flight conditions (All aerodynamic conditions, all centre of
gravity positions, allowing for control surface deflections, including the effects of ice)
Requires: Sufficiently large tail surface, Measures to ensure a high maximum lift coefficient.
Shall be able to handle high Mach numbers without flow separation.
Requires: Mach number at which extensive flow separation stars is above MD for control deflection
necessary to correct for sideslip.
At a given angle of attack the side force on a vertical tail surface is only dependent on the fin height and
planform is of secondary importance.
Thus for vertical tail design:
High aspect ratio higher weight, Flutter analysis should be considered for T-tails
High taper ratio lower weight, could lead to premature tip stall (can be reduced by sweep)
Dorsal fins:
Generates leading edge vortices which stabilizes the flow and provides a low pressure region over the
main surface
Works identical to strakes at high .

27

high yaw characteristics:

To ensure safe flight characteristics at high yaw angles the vertical tailplane must have a large stalling
angle
This is obtained by a low aspect ratio and sweepback and/or dorsal fin
Racy swept back tails on slow propeller aircraft may well be functional.

15.2

Rudder design

Some handling considerations:

Control forces should be acceptable for reversible flight controls
The force should vary linearly with deflection and side-slip angle
Control forces can be reduced by:
Applying a horn balance
Application of balance tabs
Some structural considerations w.r.t. sweep angle:
Aft spar of Vertical tail highly loaded when rudder deflected
Thicker spar beneficial
Compressibility effects at cruise Mach numbers requires higher sweep of the aft spar
Rudder effectiveness decreases with sweep
Careful trade-off needs to be made to arrive at sweep angle

15.3

Control surface design

Control surfaces perform three functions

1. Control surfaces provide means to achieve a steady state of equilibrium (trim) either at zero or at
non-zero control force
2. Control surfaces allow manoeuvring up to maximum roll, pitch and yaw rates. Also combinations of
manoeuvres should be possible.
3. Control surfaces are used to counteract disturbances such as gusts, both small and large, which might
otherwise cause the aircraft to deviate from its intended flight path
In order to obtain a maximum lift force due to control deflection a deep control surface may seem attractive,
however, increasing the control-surface-chord-to-airfoil-chord ratio decreases the linearity between lift force
and control-surface angle.
On a swept airfoil the root of the spar is a heavily loaded structural element. To maximise the span height
may therefore seem attractive. The resulting increase in the section tail angle will however decrease the
aerodynamic effectiveness of the rudder.
Dorsal fins not only increase the maximum sideslip angle which can be reached before control is lost. But,
as both effects are caused by thickening of the boundary layer and flow separation, dorsal fins also increase
the sideslip angle where a significant rudder pedal force lightening and eventually rudder lock occurs.
For ailerons:
aeroelastic deformation may decrease the ailerons effectiveness

28

 at high lift coefficients large aileron deflections may cause significant flight path disturbances with
large adverse yaw.

16

Propeller slipstream effects

(chapter 36)
Propellers generate thrust by adding momentum to a streamtube that flows through the propeller disc area.
Each propeller blade may be compared to a wing and deflects a streamtube such that a resultant force is
created in the desired direction. The propeller deflects the part of the streamtube behind the blade segment
over the induced angle . This deflection does not cause induced drag but an induced thrust resulting from
the increased impulse in flight direction. The deflection of the flow over the angle is experienced as a
rotation or swirl with angle of the flow behind the propeller. The swirl is an energy loss and should
therefore be minimised. The stream tube behind a propeller in which the velocity of the axial flow is higher
than the undisturbed flow and a rotational velocity is present is called the propeller slipstream.
Components located in the slipstream experience the slipstream as a variation in the oncoming airflow.
Therefore, the oncoming flow is no longer homogeneous with parallel streamlines and so the pressure
distribution may differ considerably from a situation in which these components would be located outside
the slipstream.
In reality there is more complexity:
Radial velocity varies with blade span
Finite blade span causes induced velocities
Multiple blades cause induced velocities
Three power effects affect the longitudinal stability and control of the propeller aircraft
1. Propeller forces:
The propeller thrust will increase when the velocity of the aircraft is reduced. This increase in thrust
may destabilise depending on whether the thrust line is located beneath or above the centre of gravity
of the aircraft. This is especially important at very low speeds, when the propeller thrust is at its
maximum.
2. Increase in wing lift, tail-off pitching moment and downwash
The increase in local wing lift due to the slipstream causes an increase in the downwash behind the
wing. Therefore, the tail will experience a larger downwash, which will reduce the contribution of the
tail to the longitudinal stability of the aircraft.
3. Change in horizontal tail lift due to the increased dynamic pressure in the slipstream
If the horizontal tail is partly or completely immersed in the propeller slipstream it will experience a
higher advantage dynamic pressure.This will result in a larger effective tailplane lift curve slope, and
increases the tailplane contribution to stability.
When the flaps are deployed the slipstream will increase CLmax and increase CL .
An increase in CT has a destabilising effect. Two types of longitudinal stability exists for propeller aircraft:
1. The response to at constant CT
2. The response to due to change in speed (and CT )

29

The downwash gradient at constant CT is:



d
d


CT =constant

d
=
dCL

(

dCL
d


CT =0

(1 + kCT )

(8)

This equation says that, when flying at constant airspeed and power setting, the higher the power setting is
chosen and so the thrust, the more the contribution of the horizontal tailplane to the longitudinal stability of
the aircraft decreases. This applies particular when the tailplane is outside the slipstream.
On propeller driven aircraft two types of longitudinal stability at constant power setting exist:
1. The response of the aircraft to a change in angle of attack at constant speed such as in a turn or
pull-up manoeuvre
2. The response of the aircraft to a change in angle of attack due to a change in speed in rectilinear
flight.
negative tailplane stall:
High power, large flap deflection and low speed causes large downwash angles
Very large negative angles of attack at the horizontal tail
Might lead to flow separation
Possible solutions are:
Modify horizontal tail leading edge with negative camber
Change stabilizer incidence with flap setting
Change to T-tail configuration to increase the distance to the vortex sheet and reduce the negative
angle of attack
Slipstream effects:
Propeller forces
Increase in thrust may destabilizing depending on whether the thrust line is located beneath or
above the centre of gravity of the aircraft. This is especially important at very low speeds when
the propeller thrust is at its maximum.
The propeller normal forces (Cnp ) will become quite substantial if the propeller is at a high
effective angle of attack. Since the point of application is usually ahead of the centre-of-gravity,
these too will often have a destabilizing effect
Increase in wing lift, tail-off pitching moment and downwash
The increase in local wing lift due to the slipstream causes an increase in the downwash behind
the wing. Therefore, the tail (also a T-tail) will experience a larger downwash, which will reduce
the contribution of the tail to the longitudinal stability of the aircraft
Usually the increase in wing lift, particularly when the flaps are deflected will increase the tail-off
pitching moment in a nose-down sense
Change in horizontal tail lift due to the increased dynamic pressure in the slipstream
If the horizontal tail is partly or completely immersed in the propeller slipstream, it will experience
a higher average dynamic pressure. This will result in a larger effective tailplane lift curve slope,
and increase the tailplane contribution to stability.
Whether the tail is actually in the slipstream, depends on angle of attack, flap deflection and
power setting.

30

 Oncoming flow at the horizontal tail differs from undisturbed flow due to
The wind and engine nacelle wake, characterized by a lower average dynamic pressure than
the undisturbed flow. A low-set stabilizer may, with the wing flaps retracted or at a take-off
setting, exhibit a diminishing effectiveness at high angle-of-attack if the stabilizer is immersed
in the wing wake
The propeller slipstream, characterized by swirl and an increased average dynamic pressure
relative to the outer flow and by an inflow of the outer flow on the slipstream boundary. The
latter may further complicate the tailplane contribution to stability and control
The downwash, which is a direct consequence of the wing producing lift and is primarily tied to
the lift coefficient with or without the presence of propeller slipstream
Propeller slipstream does not only affect the flow over the horizontal tail but also over the vertical tail and
therefore the directional stability and control of the aircraft. This effect is due to crossflow which is induced
through asymmetric lift distribution which again is caused by propeller swirl.
When the outboard engine fails the cross flow due to the inboard-up engine increases which results in a
moment imbalance. This induces a yaw rate and side slip and thus requires opposite rudder deflection.
Critical situation for directional equilibrium is with flaps down and the outboard-up engine fails. This creates
a large cross flow which is independent of tail size. A large vertical tail will then cause a larger yawing
moment at = 0. Therefore, a balance must be such between moment from operative engine and moment
generated by sideslip and rudder deflection.

16.1

lateral stability

power application will cause an asymmetric lift distribution and a rolling moment. When > 0 the centre
of lift shifts laterally, which causes the vertical tail to generate a force and therefore an opposite rolling
moment occurs. For high power this can cause lateral instability.

17

Engine intakes, exhausts, and reversers

(chapter 37, 38 and 39)

Function of engine intake:
Decelerate flow to approximately M = 0.6
Raise the pressure in the flow in high-speed conditions
Avoid shocks or separated flow
Minimize loss in total energy of the flow (maximize total pressure recovery)
The engine intake must deliver large amounts of air to the engine over a large envelope of operating
conditions, such as Mach 0 to 0.9 an angle of attack from -5 to +20 degrees and large mass flow variations.
Mass flow and velocity determine streamtube diameter
Fan RPS is a compromise between turbine and fan
Fan tips may be supersonic
High speed at the fan increases helical tip Mach number
Cruise design requirements
Minimizing drag

31

 As low as possible supervelocities

Mass flow ratio A /AHL as near as unity as other design requirements permit, such that the
stagnation point is close to the highlight.
The supervelocities on the outside of the intake will be critical which determines the intake area
The cross section of the engine flow streamtube will be independent to the inlet shape
Mass flow varies greatly with engine setting, altitude, and airspeed.
The intake area should be tailored closely to the required mass flow
The external cowling should prevent strong shock waves from occurring
Intake design requirements at low speeds:
MFR > 1.0
Highest supervelocities occur near the throat area
Velocity distribution over the throat area is not uniform (highest velocities occur near the wall)
If MT H > 0.8 large decrease in total pressure recovery efficiency
Maximum MT H is critical at low speed because the required engine mass flow is largest at take-off
and initial climb
Average throat Mach number should be limited to MT H < 0.8 to prevent strong shockwaves
In order to maximise the load-carrying capability of a transport aircraft after an engine has failed every effort
should be made to minimise the extra drag resulting from this flight condition.
With engines on the wing there are then two additional drag sources and one possible drag source:
1. In order to balance the resulting yawing moment a large rudder deflection is required leading to extra
drag from the vertical tailplane (asymmetric drag)
2. The failed engine will produce windmilling or blocked rotor drag. This is the internal drag of the
engine itself
3. The resulting low mass flow ratio on the failed engine will produce higher supervelocities on the
intake exterior surfaces and increased friction drag and possible some wave drag. In combination with
the high angle of attack this may lead to flow separation, with additional drag, either on the lower
inside or on the upper outside of the engine nacelle. The sum of these drag contributions there termed
forebody drag. When the basic external drag at mass flow ratios close to one is added the total drag
is sometimes called cowl drag.
To minimise shock wave strength in cruise and to prevent flow separation on an engine intake at take-off
the intake is as far as possible aligned with the local flow. As intakes on wing-engines operate in the
upwash in front of the wing the intake can operate at high angles of attack. At high angles of attack the
stagnation point at the top has moved far inboard. This creates a leading edge suction peak comparable to
the suction peak on an airfoil. On the other hand at the maximum thrust setting the operative engines will
at low speeds operate at a high mass flow. this will move the stagnation point on the lower outside further
aft with increasing angle of attack leading to high local velocities on the lower inboard nacelle.
In order to align engine intakes with the local flow in cruise and to maximise the intake efficiency at full
thrust and minimize the added drag when an engine fails at high angles of attack at low speed engine
intakes are usually drooped 3 to 5 degrees relative to the engine centre line.

32

17.1

Engine exhausts

Compared to basic one-dimensional jet engine theory the real gas flow through and over engine exhausts
shows a number of differences:
Both internal and external flow have boundary layers
Because of the boundary layer and stream wise wall curvature the internal flow is not one-dimensional.
The axial velocity varies over each cross-section
Despite the application of nozzle guide vanes or stators to reduce swirl, a certain degree of rotation
will be present in the exhaust. This also leads to the internal flow not being one dimensional
Because of the difference in velocity between external and engine exhaust flow or, on a by-pass engine,
between the fan and the core flow mixing occurs on the flow boundaries. This leads to an exchange of
energy between external and internal flow and between fan and core flow.
To account for the above effects the following equations are used:
Vactual
Videal
Factual
Vactual =
actual Videal
m
actual
m
CD =
ideal
m
Factual
Factual
=
= CV CD
CT =
ideal Videal
m
Fideal
CV =

(9)
(10)
(11)
(12)

A high fan pressure ratio indicates high fan exhaust flow velocities. The fan flow is directed towards the
core exhaust and flow suppression occurs on the core flow. this reduces the mass flow from the core and
consequently the discharge coefficient deteriorates. This situation exists particular during take-off and initial
climb.

17.2

Thrust reversers

The main function of thrust reversers is to decrease the ground roll distance, either after landing or in an
aborted take-off. Thrust reversers are also used for taxiing backwards from the gate on reversers thrust.
There are two basic types of thrust reversers in use:
1. Bucket- or target-type reversers
2. Cascade thrust reversers.
In the design of thrust reversers three design goals can be recognized:
1. The maximization of reverse thrust (usually not more than 50% reverse thrust is realised)
2. The minimization of the risk of ingestion of the hot exhaust gas and foreign objects
3. The minimization of adverse effects on stability and control

18

Stall characteristics

(chapter 26)
According to the CS-25 regulation:

CLmax V

min

< (1.06)2 CLmax 1G

33

(13)

where:
For certification purposes both values should be determined from test data obtained at the most forward
C.G. position
Certification speed margins for 1-G and Vmin are different to obtain comparable take-off performance
In order to determine the maximum lift coefficient of a full-scale aircraft, three different definitions of CLmax
are used:
1. The maximum lift coefficient of the aircraft is steady, rectilinear flight:
CLmax 1G =

nW
1
2
2 V SW

(14)

This value of the maximum lift coefficient is usually considered at zero thrust (TC = 0).
2. The maximum lift coefficient of the aircraft without tailplane in steady flight, CLmax T O . This value cannot
be determined directly from flight tests but must be determined from CLmax 1G by subtracting the tail
contribution.
3. The maximum lift coefficient based on the minimum speed measured in a stall manoeuvre.
CLmax V

W
min

1
2
2 V SW

(15)

CLmax is reached because at some part on the wing upper surface the flow in the boundary layer can no longer
negotiate the large adverse pressure gradients and separates. This results in a change in the distribution of
forces over the wing and a consequent change in downwash characteristics. As a consequence, the aircraft
develops apart from the lift loss a tendency towards abrupt attitude changes.
Medium-speed aircraft with straight or moderately-swept wings and thick leading edges may, even without
leading-edge devices, reach maximum lift coefficients of around CLmax 1G = 3.0.
The stalling characteristics of an aircraft determine the response of the full-scale aircraft to the occurrence
of significant flow separation on the wing. Satisfactory stalling characteristics means that the response to
the flow separation should be either favourable (such as a strong nose-down pitch tendency which restores
a condition of attached wing flow) or should be easily controlled by the pilot. Under no circumstances shall
flow separation cause an abrupt pitch-up or yaw movement which may lead to auto rotation in a spin.
Some general rules to prevent undesirable stalling characteristics are:
Flow separation should start on the inboard wing so that spoilers and ailerons for roll control remain
effective in the stall. This can for example be achieved by making sure that peak Cpmin -values at the
leading edge (either on the wing or on the slat) occur on the inboard wing.
When the stall occurs the tailplane and elevator shall not lose effectiveness due to becoming immersed
in the wake of the separated flow on the wing.
For the spanwise distribution of the peak suction coefficient on the slat as a measure of the local lift coefficient
for various angles of attack:
The low Cpmin near the root is due to the low local lift coefficient as a consequence of the strong
tapering of the inboard wing
Maximum lift is reached at = 20.5 .
Above this angle Cpmin decreases on the inboard wing.
On a T-tail the flow over the horizontal tail plane is initially largely undisturbed after the flow over the
wing separates. Flow separation on the wing decreases the wing lift, so that the vertical balance of forces
changes, making the aircraft lose altitude and increase the angle of attack. A strong nose-down pitching

34

moment at large angles of attack is required to compensate the tendency to pitch-up at an increasing rate.
But if this large-enough nose-down pitching moment is not produced the angle of attack will keep increasing
until the tail plane becomes immersed in the wake of the wing. If the tailplane is immersed in the wake of
the separated flow of the main wing, the aircraft will loose its longitudinal stability. It will remain unstable
and pitch-up until a new equilibrium is found at a very high angle of attack. This may result in a locked-in
stall or deep stall, from which recovery may be extremely difficult. Next to deep stall the elevator will lose
its effectiveness as well. Also, most of the aileron-effectiveness will be eliminated.

19

Take-off performance

(chapter 27, 44)

Take-off performance
Certification tests are performed to determine the performance and flight handling characteristics of
the aircraft in order to guarantee a satisfactory level of safety in its commercial day-to-day operation.
A large part of the regulations defines the flight test required to be performed in order to obtain
the basic data from which the take-off limitations can be calculated for inclusion in the official flight
manual.
Engine failure during take-off has for may years been regarded as one of the most serious causes of accidents:
During take-off the engine operates at its most critical condition
If engine failure occurs the pilot has little time and space to react adequately
The use of high-lift devices during take-off and landing increases the maximum lift coefficient and thus
decreases take-off and landing field length, approach speed and decreases the wing area as required for
optimum cruise performance.
At high fields a high L/D is necessary because both the climbout and climb gradient depend on the L/D
ratio:
s
W T D
sair =
(16)
SCL W
sin =

T
D

W
L

(17)

In the last years, due to limited maximum fuel volume, the interest of aircraft manufacturers has shifted from
sophisticated high lift devices in order to minimise the wing area, to simpler systems with more emphasis on
cost, reliability and maintainability.
Besides lift, high lift devices also increase the drag.
Because CFD has only limited applicability for high lift devices theory can be used fro three-dimensional
configurations for certain particular applications:
Analysing the spanwise lift distribution and the induced drag.
Analysing chordwise pressure distributions and for smoothing irregularities in the pressure distribution.
The effects on flaps on the lift/drag polar is that field attitude might require different flap settings in order
to comply with climb gradient requirements or maximum ground run.
The effect of slats is that they increase the L/D for a given flap setting and required CL .
Take-off performance;
Best one-engine-out climb requires a high L/D and high CL
35

 Increase one, decrease the other

Induced drag is the largest drag component
High aspect ratio is beneficial
The boundary layer and some partial flow separation on the flap upper surface has a de-cambering effect on
the flap causing the effective flap angle to be smaller than the nominal flap angle. If this effective flap angle
is included in calculations based on lifting surface theory, good agreement between theory and experiment
can be obtained in particular with small flap angles. At higher flap angles the agreement may be less good,
particularly at low angles of attack. This is due to the rear fuselage being placed in a strong down wash
resulting in vortices or flow separation on the lower fuselage and, with rear-engine configurations, on the
lower surface or nacelles and pylons. For higher angles of attack the rear fuselage is better aligned with
the local flow direction.
Non-planar lifting surface theory, combined with empirical data can be used to estimate lift-drag ratios for
take-off and landing configurations when no wind tunnel data is available.
Large flap deflections in combination with low result in bad predictions:
Downwash makes large angles with fuselage
Highly 3D flow in vortices
prediction versus wind tunnel tests:
Prediction with empirical correction is fairly good
No maximum lift coefficient can be predicted
L/D can be estimated
How to determine L/D:
Calculate induced drag
Calculate friction drag
Calculate trim drag
Calculate form (pressure) drag due to lift at each section
To find the trim drag the tail-off pitching moment coefficient curves are calculated. These are compared with
wind tunnel data for a centre of gravity position.
When the slope of the linearised CL2 versus CD curve for the clean aircraft configuration is known, the average
increase in the Oswald factor allow the estimation of the drag polar for speeds between 1.2Vs and 1.4Vs for
every flap angle.
When no special measures are taken, boundary layer effects are not incorporated. This leads in the calculations to higher supervelocities than in real flow. One way of reducing this shortcoming is to compare theory
and experiment not at the same angle-of-attack but at the same lift coefficient. This leads to a satisfactory
comparison of the chordwise pressure distribution and the spanwise distribution of the local lift coefficient
particularly when also aero-elastic deformation is taken into account.

19.1

Chapter 44

Engine failout during take-off has for many years been regarded as one of the most serious causes of
accidents because:
1. During take-off the engine operates at its most critical condition
2. If an engine failout occurs the pilot has literally little time and space to react adequately

36

In order to maximise the operational flexibility transport aircraft are certified for a range of take-off flap
settings. For each flap setting a complete free-flight and field performance test programme has to be
executed.
In order to find the necessary runway length three tests are conducted:
1. Take-off with all engines operating, the necessary runway length is increased with 1.15 to account for
non-ideal conditions.
2. Runway length while having a simulated engine failure just before V1 , and still be able to fly over 35
feet high obstacles
3. Runway length while having a simulated engine failure just before V1 , but the take-off is aborted, and
the aircraft has to be able to come to a safe standstill
The largest of the three runway lengths necessary is the required runway length for that particular aircraft.
The minimum unstick speed, VMU , is the minimum speed at which the aircraft is able to take-off within the
certified take-off distance for the atmospheric and aircraft conditions under consideration. This speed is
determined by three parameters:
1. Wing stall in ground effect with the main wheels still on the ground
2. Geometric limitations because the maximum usable CL in ground effect is determined by the rear
fuselage striking the runway
3. Through a limited elevator capacity (the aircraft rotates so late that due to the aircrafts speed and
acceleration the aircraft leaves the ground before the maximum lift coefficient in ground effect is
reached)

20

Flight beyond buffet onset and MMO

The buffet onset boundary

Just as the low speed stall the buffet onset boundary forms a limitation of the aircraft flight envelope
Low-speed stall: trailing edge flow separation of the leading edge suction peak
Buffet onset boundary: separation starts at the foot of the shock wave and creates oscillating pressure
distribution
Airworthiness regulations require that in normal operation of civil transport aircraft the buffet onset boundary
is not surpassed. Buffet may not intrude the normal operating envelope of the aircraft. Buffet can thus only
appear during a pull-up or turn manoeuvre.
The certification regulations require the lift coefficient in operational cruise conditions to be limited such
that a load factor of n = 1.3 can be reached without encountering buffet. This limits the maximum design
lift coefficient.
Buffet and safety:
The airworthiness requirements demand that if the buffet regime is penetrated this should be in fully
controlled flight
Flow separation on a swept wing should therefore not start too far outboard, to prevent strong roll or
pitch-up tendencies
For certification the buffet regime is deliberately penetrated in flight to demonstrate acceptable flight
handling characteristics
The maximum buffet penetration boundary or maximum demonstrated lift boundary has to be recorded

37

The buffet onset boundary is not a uniquely defined physical phenomenon. Flow separation leads to vibration
and it is the latter which is recorded either by a human or by an accelerometer. This means that the intensity
of the buffet as perceived by a human or recorded by a test equipment depends on their position relative to
the nodes and loops of the vibrating structure.
In the wind tunnel the buffet onset boundary may be determined from a number of characteristics
1. Breaks in CL , CM or Cx .
2. Trailing-edge pressure divergence on the outboard wing
3. Divergence of dynamic wing root strain gauge recordings
The onset of buffet depends on the Reynolds number when the boundary layer is involved.

20.1

Chapter 29

Flight characteristics between MMO and MD :

Civil transport aircraft normally do not exceed the Maximum Operating Mach Number (MMO )
Requirement to comply with requirements on flying characteristics above MMO
Failures or severe atmospheric upsets cause the aircraft to pitch down and to exceed MMO in a dive.
Therefore the civil airworthiness requirements require that the aircraft demonstrates acceptable flight
characteristics up to the design Dive Mach Number MD (MD = MMO + 0.05 to 0.09).
It is common practice to design the wing such that the design Mach number Mdesign = MMO 0.03
to 0.05.
When a jet transport aircraft is flying at or near its design condition and the flight speed is increased the
pressure distribution on the wing changes rapidly. Initially the shock wave moves rearwards over a part of
the wing and the pattern of the wing upper surface pressure distribution changes. When the speed is further
increased the boundary layer will separate behind the shock wave and the suction forces over the front part
of the wing will decrease further. On the lower surface the supervelocities increase in a regular pattern
until, in particular at low lift coefficients, locally sonic speed is reached behind which point in the flow will
further accelerate until a shock wave appears. This different development of upper and lower surface pressure
distribution causes, in particular on swept wings, that the development of the tail-off pitching moment at
different angles of attack or lift coefficients with a varying Mach number may show large variations for
different wing geometries.

38