Module 8 Aerodynamic
Module 8 Aerodynamic
fJ
o Module 8
lJ
o
iJ
fJ
fJ
o
E
n
l . .!l
o
;[]
o
U
fJ
U
o
., ., .
U 2nd Edition
U
ExclusivelY fro,. ·vww.airtechbooks. com
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
r'
I'
~
Preface
u.' Thank you for purchasing the Total Training Support Integrated Training System, We are
sure you will need no other reference material to pass your EASA Part-66 exam in this Module,
These notes have been written by instructors of EASA Part-66 courses, specifically for
practitioners of varying experience within the aircraft maintenance industry, and especially those
who are self-studying to pass the EASA Part-66 exams. They are specifically designed to meet
the EASA Part-66 syllabus and to answer the questions being asked by the UK CAA in their
examinations.
U The EASA Part-66 syllabus for each sub-section is printed at the beginning of each of the
chapters in these course notes and is used as the "Learning Objectives".
n,
'oJ We suggest that you take each chapter in-turn, read the text of the chapter a couple of times, if
only to familiarise yourself with the location of the information contained within. Then, using
r'I
, your club66pro.co.uk membership, attempt the questions within the respective sub-section,
I• and continually refer back to these notes to read-up on the underpinning knowledge required to
'-'
answer the respective question, and any similar question that you may encounter on your real
u Part-66 examination. Studying this way, with the help of the question practice and their
explanations, you will be able to master the subject piece-by-piece, and become proficient in the
subject matter, as well as proficient in answering the CAA style EASA part-66 multiple choice
[] questions.
u
I , We regularly have a review of our training notes, and in order to improve the quality of the
I '
U notes, and of the service we provide with our Integrated Training System, we would appreciate
your feedback, whether positive or negative.
So, if you discover within these course notes, any errors or typos, or any subject which is not
c'
particularly well, or adequately explained, please tell us, using the 'contact-us' feedback page of
L. the club66pro.co.uk website. We will be sure to review your feedback and incorporate any
changes necessary. We look forward to hearing from you.
Finally, we appreciate that self-study students are usually also self-financing. We work very
hard to cut the cost of our Integrated Training System to the bare minimum that we can provide,
and in making your training resources as cost efficient as we can, using, for example, mono
printing, but providing the diagrams which would be better provided in colour, on the
club66pro.co.uk website. In order to do this, we request that you respect our copyright policy,
I··-~
and refrain from copying, scanning or reprinting these course notes in any way, even for sharing
i I
I , with friends and colleagues. Our survival as a service provider depends on it, and copyright
u
abuse only devalues the service and products available to yourself and your colleagues in the
future, and makes them more expensive too.
c Module 8 Preface
TIS Integrated Training System
© Copyright 2009
Integrated Training System nI
Designed in association with the i
club66pro.co.uk question practice aid
I J
Intentionally Blank
,"I,
L 1
N
I I
•j
~
, I
',- __ J
II I
" J
i
:
~_ .1
:]
i, i
ii
TIS Integrated Training System
Module 8 Preface n
© Copyright 2009 ,nj
I !
\
'--'
[
Module 8 Chapters
i'
U 1, Physics of the Atmosphere
r,
2, Aerodynamics
3. Theory of Flight
I!
~
4. Flight Stability and Dynamics
r'
I,
I !
L..,..:
r',I
!
L
il
U
I':
1 '
'--"
I I
L
[1
v'
Ii
U
iii
Module 8 Preface
TIS Integrated Training System
r'!
~
© Coovriaht 2009
Integrated Training System
Designed in association with th,e .
club66pro.cQ,uk question practice aid
r,
lj
n
LJ
n
III
I
~.J
n
IJ
Intentionally Blank :I
I I
~_J
il
! I
U
n
1
iJ
1
:1
;j
,i "I
n
,1
iv Module 8 Preface
TIS Integrated Training System
© Copyright 2009
n
I i
: J
II '
i 1
I ,
U
Basic Aerodynamics
,---,
,,
I .!
,
,~
n
I ,
U Module 8.1 Physics of the Atmosphere 1-1
Use and/or disclosure is TTS Integrated Training System
govemed by the statement
on page 2 of this chapter. © CopvriQht 2010
i i
\~ ~
stored in a retrieval system or transmitted in any form by any other means whatsoever: i.e.
photocopy, electronic, mechanical recording or otherwise without the prior written permission of
Total Training Support Ltd. i
· ,I
i
Basic knowledge for categories A, B1 and B2 are indicated by the allocation of knowledge levels indicators (1,2 or
C""'
3) against each applicable subject. Category C applicants must meet either the category B1 or the category B2
· I
basic knowledge levels. .J
The knowledge level indicators are defined as follows:
LEVEL 1
• A familiarisation with the principal elements of the subject.
Objectives: ,",
• The applicant should be familiar with the basic elements of the subject. i
~_
I
J
• The applicant should be able to give a simple description of the whole subject, using common words and
examples.
• The applicant should be able to use typical terms. 1· ,
,J
LEVEL 2
• A general knowledge of the theoretical and practical aspects of the subject.
• An ability to apply that knowledge.
Objectives:
• The applicant should be able to understand the theoretical fundamentals of the subject.
• The applicant should be able to give a general description of the subject using, as appropriate, typical
examples.
• The applicant should be able to use mathematical formulae in conjunction with physical laws describing the
subject.
'l
•1
• The applicant should be able to read and understand sketches, drawings and schematics describing the
subject.
• The applicant should be able to apply his knowledge in a practical manner using detailed procedures.
LEVEL 3
'I
• A detailed knowledge of the theoretical and practical aspects of the subject.
·) I,
• A capacity to combine and apply the separate elements of knowledge in a logical and comprehensive
manner.
Objectives:
• The applicant should know the theory of the subject and interrelationships with other subjects.
• The applicant should be able to give a detailed description of the subject using theoretical fundamentals
and specific examples.
• The applicant should understand and be able to use mathematical formulae related to the subject.
• The applicant should be able to read, understand and prepare sketches, simple drawings and schematics
describing the subject.
• The applicant should be able to apply his knowledge in a practical manner using manufacturer's
instructions.
• The applicant should be able to interpret results from various sources and measurements and apply
corrective action where appropriate.
n
I. ,}
r-i
U Table of Contents
c ~m~~ffi
Pressure
Performance Ceilings
The Gas Laws
10
16
17
8
r"
, I
L:
n
u
r, ' ,
I'!
:I
\. J
Intentionally Blank
-
,
~j
I
"I
.J
'l
,j
.....
, ,
;, j
(""--:
I i
Integrated Training System
I ' Designed in association with the
'-' ctub66pro.co.uk question practice aid
o
c
u
n
L Module 8.1 Physics of the Atmosphere 1-5
Use andfor disclosure is
governed by the statement TIS Integrated Training System
) i on page 2 of this chapter. © CODvrioht 2010
Integrated Training System :-"'I
I '
Designed in association with tile ,i ,I
club66pro.co.uk question practice aid
,..,
f I
\ .. J
r";
I I
'.J
,1
LJ
Intentionally Blank
n
n
lJ
['I
L
Module 8.1 Physics of the Atmosphere
('
U
The Atmosphere
,n
, ,
W
'
The gaseous envelope surrounding the Earth is called the atmosphere. There is no defined
upper limit to the atmosphere, but most aviation activity takes place within the first 60,000 ft and
therefore we need not study above that.
Gas Composition
The gases found in the atmosphere are in the following proportions (by volume):
Nitrogen 78%
Oxygen 21%
Other gases 1 % (e.g. argon, carbon dioxide, water vapour)
Oxygen is essential for the sustenance of life and the combustion of materials. In the context of
aviation, oxygen is required for the combustion of fuel, a deficiency of this gas resulting in
incomplete burning and reduced engine efficiency.
o Water vapour is present in the atmosphere in varying proportions, and is responsible for the
weather around the earth, which in turn affects aircraft operations and performance. Additionally
the presence of water vapour may cause icing of the airframe or engine which may impair an
f1u aircraft's performance.
(a) The Troposphere - Temperature decreases with an increase in height. In this region
nearly all significant weather occurs.
(b) The Tropopause - The upper limit of the troposphere where temperature stops decreasing
[' with an increase of height. The tropopause is therefore the upper limit of significant
weather, the first point of lowest temperature, and additionally it is the region for maximum
wind strengths.
The height of the tropopause varies with latitude, season of the year, and prevailing
weather conditions with the result that it is usually higher in low latitudes, in summer and in
fine weather.
u Equator
45° NIS
Poles
16-47km
10-12km
7% - 9 km
53,000-57,000 ft
33,000-39,000 ft
25,000-29,000 ft
r~
t !
~' Module 8.1 Physics of the Atmosphere 1-7
Use andlor disclosure is
governed by the statement TTS Integrated Training System
on page 2 of this chapter. © Copyright 2010
'. I
Integrated Training System
Designed in association with the
club66pro,co.uk question practice aid
n
u
(c) The Stratosphere - From the tropopause to approximately 50 km above mean sea level,
and is characterized by the temperature being steady or increasing with height
n
(d) The Mesosphere - From 50 km to 80 km. The temperature generally decreases with
U
height. ~
•l 'I
(c) The Thermosphere or Ionosphere - Temperature increases with height.
,...,
, I,
,
; J
Temperature
'l
: I
(a) Units C. J
The temperature scales most commonly used are Celsius (also known as Centigrade),
Fahrenheit and Kelvin (also known as Absolute). 1
,J
The first two scales are based on the melting point of ice, being O°C and 32°F respectively, and
the boiling point of water, being 100°C or 212°F
Heat is a form of transfer of energy, and is related to the random movement of molecules in a
substance. If heat is reduced, the molecules become less active. The minimum temperature to i ,
which a substance can be reduced is approximately minus 273°C and this is known as Absolute :, j
Zero, or 0 K. Correspondingly, the melting point of ice is equivalent to 273 K and the boiling
point of water to 373 K.
11.
"
To convert from one temperature scale to another, the following formulae may be used:
9
F= -C+32
5
5
C = -(F - 32)
9 ,
r i
" J
(b) Temperature Variation in the Troposphere
At ground level, in general, the temperature increases with a decrease of latitude.
With increasing altitude, the conductive and convective effects from the earth are reduced so
that temperature will usually decrease with height up to the tropopause.
Equator -80°C
45°N/S -56°C
Poles -45°C
:i
" ,}
1-8 Module 8.1 Physics of the Atmosphere
Use andfor disclosure is
TIS Integrated Training System gOllerned by the statement
© Copyright 2010 on page 2 of this chapter. ,Ii
: !
" J
Integrated Training System
Designed in association with the
club66pro,co.uk question practice aid
There is, therefore, a reversal of temperatures with latitude in comparison to those found at
ground level. This is partly because the tropopause is higher at the equator and the temperature
decrease is effective over a greater height.
27,OOOft -45'C
TROPOPAUSE
r'u
n
I ,
LJ
A representative value of 2°C/1 000 ft is a typical value for the troposphere, and this figure is
used as the reference for the Jet Standard.
The International Standard Atmosphere (ISA) uses the comparable value of 1.98°C/1 000 ft.
n
u For meteorological purposes, differentiation between dry (that is, not saturated) and saturated
adiabatic lapse rates is made, and the values of 3°C/1 000 ft and 1SC/1 000 ft respectively are
used. The difference of lapse rate for saturated air is caused by the release of latent heat during
condensation, thus reducing the temperature change.
Firstly, considering airframe performance, a reduction of density (p) reduces lift (L). This may be
counteracted by increasing the true airspeed (V) to achieve the required amount of lift (L):
~,
I'
,
~
I
where: =
C L coefficient of lift
and S = surface area
The dynamic pressure is gained at the expense of an increased take-off run, cruising TAS or
1w" landing run according to the stage of flight.
n
'. J
On the credit side, drag (D) reduces with increase of temperature:
D = CD % pV2 S I
: J
, J
A piston engine's performance is related to the temperature of the air being drawn into the
cylinder head. The higher the temperature, the lower the density and weight of fuel/air mixture
that can be burnt in the combustion chamber. The power output of the engine therefore falls
with increase of temperature. tl
, I
'- J
For a propulsion system, piston or jet:
Definition
Pressure is the force exerted on a unit area, Le.:
---
'; I
:, J
P ressure = Force =-
Mass x Acceleration
-------
Area Area
In the atmosphere, pressure is caused by the mass of the gaseous molecules acting under the
:l
force of gravity on a given area. As all molecules act under gravity then the pressure can also
be considered to be the weight of a column of air on a unit area.
,,1,
Unit Area
'"',j'
'.
11
~ 1
n
,-_..
, I
(' Units
! r
L! The metric units of pressure are dynes per square centimetre, where the dyne is the force
(' , required to accelerate 1 gram by 1 centimetre per second.
I i
LJ The Systeme international (SI) units of pressure are Newtons per square metre, where the
Newton is the force required to accelerate 1 kilogram by 1 metre per second, The Newton is
equal to 105 dynes.
,--, Although largely obsolete, the Imperial system of units is still encountered, and pressure is
I '
I i expressed in pounds per square inch.
LJ
r In meteorology the unit of pressure is the millibar (mb), which is equivalent to 1000 dynes per
: "
square centimetre.
U
Before the introduction of the millibar, meteorological pressure was measured in terms of the
length of a column of mercury in a barometer that the weight of the atmosphere could support.
Vacuum
(l
II '
~,
n
I I
LJ
Length 01 Column
of Mercury
Air Proportional to
Pressure Pressure
iI l
U
n
[I
LJ
Figure 1.3 - Principle of the Mercury Barometer
r',
\ i
L!
Also, it should now be apparent that the rate of pressure decrease with height is not constant. In
the first 10,000 ft. the pressure falls at a rate of approximately 1 mb per 30 ft but between
:J
30,000 ft and 40,000 ft the pressure decrease is closer to 1 mb per 88 ft.
Pressure Altitude
The altitude at which a given pressure occurs in the International Standard Atmosphere is rl
I I
called the pressure altitude. L)
If, for example, the pressure at the top of Mount Everest were determined as 300.9 mb, then the r'"
pressure altitude would be 30,000 ft. Assuming the same mean sea level conditions, and two ! I
l . .J
columns of air of the same height, but differing temperatures, then the cold air would have a
greater mass than the warm air due to the density difference. The pressure of the atmosphere,
however, is caused by the mass of overlying molecules on a unit area. The pressure above the
column of warm air is therefore higher than that above cold air. Because a higher pressure is
found at a lower level, then the pressure altitude above warm air is lower than the pressure
altitude above cold air. Alternatively it can be expressed that the true altitude of an aircraft is
more than that indicated (assuming the correct mean sea level pressure has been set on the
subscale) above warm air, and less than that indicated above cold air.
{I
,I
\ ,
,,'"""I,
ti
n
! I
1-12 Module 8.1 Physics of the Atmosphere \ 1
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter.
Ii Integrated Training System
Designed in association with the
U club66pro.co.uk question practice aid
n
I!
L' Pressure Corresponding To 700mb
and
15001t A Pressure Altitude Of 10,OOOlt
1500ft
~=
r:
!
- -- m --- - _ 1.00mb
- - - - - ----- - - -
665mb
U
o 1013mb 1013mb
c
r,
! '
W
c ,.-'-,
i f
I f
'---'
r'
f '.
n
I •
n
:, i
Density
Definition
Density is the mass per unit volume of a substance, at a specified temperature and pressure,
,.....,
i
" J
, Mass
Density = - - -
Volume
Units
Density is expressed in grams, or kilograms per cubic metre for metric or SI units, respectively, ,1
The Imperial units are pounds per cubic feet. Factors affecting density when considering a gas ':, J
are:
n
u
Pressure
Density =
Gas constant x Absolute temperature
For a given temperature, therefore, an increase of pressure increases density, or, at a given
pressure, a decrease in temperature increases density,
Air at lower levels in the atmosphere is compressed by the mass of the air above it. With
increasing altitude, the overlying mass reduces and air can now expand, resulting in a further
reduction of pressure,
With increasing altitude the temperature also decreases, but at a rate lower than the pressure.
Density, therefore, decreases with height.
We have already seen that density at sea level tends to be higher at the Poles than at the
Equator. However, at 26,000 ft, the density value is similar at all latitudes.
r-:
I ,
U
Variation of Density with Humidity
rl The total pressure of the atmosphere is equal to the sum of the individual pressures of the
u gases. The pressure of moist air is less than that for dry air, and so humidity decreases the total
pressure, From the gas equation, it can be seen that the reduction in pressure results in a lower
f':
u
density. The greater the humidity, the lower the density.
Density Altitude
This is defined as the altitude in the International Standard Atmosphere at which a given density
is found,
There are, however, additional effects of density performance. Above about 300 kt TAS, air
,,rl'
becomes significantly compressed, and locally increases the density, At much higher speeds
U
'
this may give a marked increase in drag, and when increasing altitude, this can offset the
otherwise reducing drag value.
n
I ,
U A similar compressibility effect increases drag on a propeller blade, reducing its efficiency,
particularly at higher altitudes.
A jet engine's performance, however, is enhanced by this compressibility effect as mass flow is
improved.
ri
\ : Air Density and the Human Body
L..i
The reduced density of air with increasing altitude means that in a given volume of air breathed
in, the oxygen content has decreased. Above 10,000 ft this reduction leads to hypoxia, its
effects ranging from lack of judgment to sleepiness or collapse, according to height.
At night, the reduced intake of oxygen impairs night vision at altitudes of 4,000 ft and above.
To counter these problems, aircraft operating above 10,000 ft must have an enriched oxygen
i! supply, either in conjunction with a pressurized cabin, or through face-masks. At night, ideally,
U oxygen should be available from ground level upwards.
Performance Ceilings
Service Ceiling
This is defined as the altitude at which the rate of climb of an aircraft falls to a specified figure,
usually 100 ft. per minute. ~
: I
" J
Absolute Ceiling
The absolute ceiling is the altitude at which the rate of climb of an aircraft falls to zero.
Piston-Engined Aircraft
For such aircraft operating under 26,000 ft. the improved atmospheric density found in winter in
high latitudes will give the highest ceiling. 'l
I. I
'I i"
',)
n
, !
,l
1-16
1
, ,,
,
Module 8.1 Physics of the Atmosphere
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter.
l
C' Integrated Training System
! ! Designed in association with the
I~
club66pro.co.uk question practice aid
Boyle's Law
The volume (V) of a given mass of gas at constant temperature is inversely proportional to
pressure (P):
PV =constant
This can be expressed in the form:
o Charles' Law
The volume of a given mass of gas at constant pressure, increases by _1_ of its volume at O°C
273
for every 1°C rise in temperature:
V
ti "T= constant
U
The alternative expression below is also useful:
V, = V2
r: T, T2
LJ
Combined Boyle's and Charles' Law Equation
~1
I . The results of both laws may be combined in one equation, expressing the behaviour of a gas
U under varying conditions of pressure, volume and temperature:
P, V, P2 V 2
--=- -
T, T2
n
( i
I'---':
""""1
) i
I .
U
In order to provide a datum for aircraft performance comparison, and instrument calibration, this I• • 1
assumed set of conditions is used. Whilst representative, these conditions do not necessarily
reflect actual conditions in the atmosphere. The values used are listed below: rl
, I
. }
,-..,
'" I
I, J
n
:J
( \
i~
1-18 Module 8.1 Physics of the Atmosphere .
i ,I
Use andlor disclosure is
TTS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter.
C
I~-: Integrated Training System
I'-' i Designed in association with the
club66pro.co.uk question practice aid
r'l
I : International Standard Atmosphere (ISA) Data
l.J ------------------------------------------------------ ----------------
Altitude Temperature Kinematic Speed
rn ft °c Pressure Density Viscosity Viscosity of
C Ratio Ratio Ratio Ratio Sound
------------------------------------------------ -----------------------
0 0 15.2 1.0000 1.0000 1.0000 1.0000 340.3
rLJ ,
i
152
304
500
1000
14.2
13.2
0.9821
0.9644
0.9855
0.9711
0.9973
0.9947
1.0121
1. 0243
339.7
339.1
457 1500 12.2 0.9470 0.9568 0.9920 1. 0367 338.5
n
-'
609
762
914
1066
2000
2500
3000
3500
11.2
10.2
9.3
8.3
0.9298
0.9129
0.8962
0.8798
0.9428
0.9289
0.9151
0.9015
0.9893
0.9866
0.9839
0.9812
1.0493
1.0622
1. 0752
1.0884
338.0
337.4
336.8
336.2
r-') 1219 4000 7.3 0.8637 0.8881 0.9785 1.1018 335.6
Iu i 1371 4500 6.3 0.8477 0.8748 0.9758 1.1155 335.0
1524 5000 5.3 0.8320 0.8617 0.9731 1.1293 334.4
----------------------------------------------------------------------
Ii
l-i
1676 5500 4.3 0.8166 0.8487 0.9704 1.1434 333.8
1828 6000 3.3 0.8014 0.8359 0.9677 1.1577 333.2
1981 6500 2.3 0.7864 0.8232 0.9649 1.1722 332.6
Ii 2133 7000
2286 7500
1.3 0.7716
0.7571
0.8106
0.7983
0.9622
0.9595
1.1870
1.2020
332.0
L.J 2438 8000
0.3
-0.6 0.7428 0.7860 0.9567 1.2172
331. 4
330.8
2590 8500 -1. 6 0.7287 0.7739 0.9540 1. 2327 330.2
ri 2743 9000 -2.6 0.7148 0.7620 0.9512 1. 2484 329.6
L 2895 9500 -3.6 0.7012 0.7501 0.9485 1. 2644 329.0
3048 10000 -4.6 0.6877 0.7385 0.9457 1. 2807 328.4
I" ' ----------------------------------------------------------------------
I ! 3200 10500 -5.6 0.6745 0.7269 0.9430 1.2972 327.8
U 3352 11000 ~6.6 0.6614 0.7155 0.9402 1.3140 327.2
3505 11500 -7.6 0.6486 0.7043 0.9374 1. 3310 326.6
r,
3657 12000 -8.6 0.6360 0.6932 0.9347 1. 3484 326.0
U 3810
3962
12500
13000
-9.6
~10.6
0.6236
0.6113
0.6822
0.6713
0.9319
0.9291
1.3660
1. 384 0
325.4
324.7
4114 13500 ~11.5 0.5993 0.6606 0.9263 1.4022 324.1
r'U
, I 4267
4419
14000
14500
~12.5
-13.5
0.5875
0.5758
0.6500
0.6396
0.9235
0.9207
1.4207
1. 4396
323.5
322.9
4572 15000 -14.5 0.5643 0.6292 0.9179 1.4588 322.3
r~i ----------------------------------------------------------------------
LJ 4724
4876
15500
16000
-15.5
-16.5
0.5531
0.5420
0.6190
0.6090
0.9151
0.9123
1.4783
1.4981
321.7
321. 0
r--'\ 5029 16500 -17 .5 0.5311 0.5990 0.9094 1.5183 320.4
I : 5181 17000 -18.5 0.5203 0.5892 0.9066 1.5388 319.8
I !
L! 5334 17500 -19.5 0.5098 0.5795 0.9038 1. 5596 319.2
5486 18000 -20.5 0.4994 0.5699 0.9009 1.5809 318.5
5638 18500 -21.5 0.4892 0.5604 0.8981 1.6025 317.9
i~. 1. 6244 317.3
5791 19000 -22.4 0.4791 0.5511 0.8953
L.J
5943 19500 -23.4 0.4693 0.5419 0.8924 1. 6468 316.7
6096 20000 -24.4 0.4595 0.5328 0.8895 1.6696 316.0
n
\ i'
----------------------------------------------------------------------
U 6248 20500 -25.4 0.4500 0.5238 0.8867 1.6927 315.4
6400 21000 -26.4 0.4406 0.5150 0.8838 1. 7163 314.8
r-'
I
6553 21500 -27.4 0.4314 0.5062 0.8809 1. 7403
1. 7647
314.1
313.5
\ i 6705 22000 -28.4 0.4223 0.4976 0.8781
c....!
6858 22500 -29.4 0.4134 0.4891 0.8752 1. 78 95 312.9
7010 23000 -30.4 0.4046 0.4806 0.8723 1.8148 312.2
7162 23500 -31.4 0.3960 0.4723 0.8694 1. 8406 311. 6
iLJ• 7315 24000 -32.3 0.3876 0.4642 0.8665 1.8668 311. 0
7467 24500 -33.3 0.3793 0.4561 0.8636 1. 8935 310.3
r·'l
[
'-'
: Module 8.1 Physics of the Atmosphere 1~19
Use and/or disclosure is
govemed by the statement TIS Integrated Training System
r- on page 2 of this chapter. © CODvriaht 201 0
:. J
(,
I !.
Altitude Temperature Kinematic Speed
w
rn ft °c Pressure Density Viscosity Viscosity of
Ratio Ratio Ratio Ratio Sound
-----------------------------------------------------------------------
18592 61000 -56.3 0.06746 0.08972 0.7947 8.8572 295.1
18897 62000 -56.3 0.06430 0.08551 0.7947 9.2932 295.1
19202 63000 -56.3 0.06128 0.08150 0.7947 9.7508 295.1
19507 64000 -56.3 0.05841 0.07768 0.7947 10.231 295.1
19812 65000 -56.3 0.05566 0.07403 0.7947 10.735 295.1
o
r:
I ,
U
f":
, ,
\ .
U
rl
I i
U Module 8.1 Physics of the Atmosphere 1-21
Use and/or disclosure is
govemed by the statement TIS Integrated Training System
on page 2 of this chapler. © CODvriQht 201 0
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
'I
, I
"J
n
"'--
G
;, .,l
(i
!I
'j
n
'I
'. J
Intentionally Blank
"
jJ
n
:
, ,\
n
~ j
n
LJ
:l
\ I
~
, I
, ,I
\
l
" )
:1
1-22 Module 8.1 Physics of the Atmosphere " J
Use and/or disclosure is
TIS Integrated Training System govemed by the statement
on page 2 of this chapter.
© Copyright 201 0
u
c Integrated Training System
Designed in association with the
club66pro.cQ,uk question practice aid
r~
U Humidity
Some water in the form of invisible vapour is intermixed with the air throughout the atmosphere.
It is the condensation of this vapour which gives rise to most weather phenomena: clouds, rain,
snow, dew, frost and fog_ There is a limit to how much water vapour the air can hold and this
-l limit varies with temperature. When the air contains the maximum amount of vapour possible for
I, a particular temperature, the air is said to be saturated. Warm air can hold more vapour than
LJ
cold air. In general the air is not saturated, containing only a fraction of the possible water
I:
u
vapour.
The amount of vapour in the air can be measured in a number of ways. The humidity of a
packet of air is usually denoted by the mass of vapour contained within it, or the pressure that
the water vapour exerts. This is the absolute humidity of air. Relative humidity is measured
by comparing the actual mass of vapour in the air to the mass of vapour in saturated air at the
same temperature. For example, air at 100G contains 9.4 g/m 3 (grams per cubic metre) of water
vapour when saturated. If air at this temperature contains only 4.7 g/m 3 of water vapour, then
U of the thermometer. The difference between wet and dry bulb temperatures is used to calculate
the various measures of humidity.
'-1
Ii ;.
W Definitions
Absolute humidity: The mass of water vapour in a given volume of air (i.e., density of water
l'
u vapour in a given parcel), usually expressed in grams per cubic meter
Actual vapour pressure: The partial pressure exerted by the water vapour present in a parcel.
Ii
LJ Water in a gaseous state (i.e. water vapour) exerts a pressure just like the atmospheric air.
Vapour pressure is also measured in Millibars.
Condensation: The phase change of a gas to a liquid. In the atmosphere, the change of water
vapour to liquid water.
o Dewpoint: the temperature air would have to be cooled to in order for saturation to occur. The
dewpoint temperature assumes there is no change in air pressure or moisture content of the air.
[1
LJ
Dry bulb temperature: The actual air temperature. See wet bulb temperature below.
Mixing ratio: The mass of water vapour in a parcel divided by the mass of the dry air in the
parcel (not including water vapour). il
Relative humidity: The amount of water vapour actually in the air divided by the amount of ,..'">
water vapour the air can hold. Relative humidity is expressed as a percentage and can be lj
computed in a variety of ways. One way is to divide the actual vapour pressure by the saturation
vapour pressure and then multiply by 100 to convert to a percent.
Saturation of air: The condition under which the amount of water vapour in the air is the
n
maximum possible at the existing temperature and pressure. Condensation or sublimation will n
begin if the temperature falls or water vapour is added to the air. u
Saturation vapour pressure: The maximum partial pressure that water vapour molecules
would exert if the air were saturated with vapour at a given temperature. Saturation vapour
pressure is direCtly proportional to the temperature.
Specific humidity: The mass of water vapour in a parcel divided by the total mass of the air in n
l)
Wet bulb temperature: The lowest temperature that can be obtained by evaporating water into
the air at constant pressure. The name comes from the technique of putting a wet cloth over the
bulb of a mercury thermometer and then blowing air over the cloth until the water evaporates.
n
I j
Since evaporation takes up heat, the thermometer will cool to a lower temperature than a
thermometer with a dry bulb at the same time and place. Wet bulb temperatures can be used
along with the dry bulb temperature to calculate dew point or relative humidity.
,l )
,.
n
,n \
U
ni I
1-24 Module 8.1 Physics of the Atmosphere \. J
( 1
,
" 1,
LJ
I'1,
U
Basic knowledge for categories A, B1 and B2 are indicated by the allocation of knowledge levels indicators (1, 2 or
3) against each applicable subject. Category C applicants must meet either the category B1 or the category B2
basic knowledge levels.
The knowledge level indicators are defined as follows:
n
LEVEL 1
• A familiarisation with the principal elements of the subject.
Objectives:
• The applicant should be familiar with the basic elements of the subject.
• The applicant should be able to give a simple description of the whole subject, using common words and
examples.
• The applicant should be able to use typical terms. "
,
IJ
I
LEVEL 2
• A general knowledge of the theoretical and practical aspects of the subject.
• An ability to apply that knowledge.
Objectives:
n
n
, I
• The applicant should be able to understand the theoretical fundamentals of the subject.
• The applicant should be able to give a general description of the subject using, as appropriate, typical U
examples.
• The applicant should be able to use mathematical formulae in conjunction with physical laws describing the
subject.
• The applicant should be able to read and understand sketches, drawings and schematics describing the
subject.
• The applicant should be able to apply his knowledge in a practical manner using detailed procedures.
LEVEL 3
• A detailed knowledge of the theoretical and practical aspects of the subject.
• A capacity to combine and apply the separate elements of knowledge in a logical and comprehensive
manner.
Objectives:
• The applicant should know the theory of the subject and interrelationships with other subjects.
• The applicant should be able to give a detailed description of the subject using theoretical fundamentals
and specific examples. n
• The applicant should understand and be able to use mathematical formulae related to the subject. u
• The applicant should be able to read, understand and prepare sketches, simple drawings and schematics
describing the subject.
• The applicant should be able to apply his knowledge in a practical manner using manufacturer's
instructions.
• The applicant should be able to interpret results from various sources and measurements and apply
corrective action where appropriate. n
! •
lJ
Table of Contents
[1
U
Module 8.2 Aerodynamics _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 7
fi
u
Airflow 7
Atmospheric Pressure 7
Streamline Flow 8
Flow Continuity 8
Bernoulli's Theorem 10
r -, Airspeed 11
I
I
II
U Aerodynamic and Geometric Definitions 13
o Chord line
Mean Camber line
Thickness/Chord Ratio and Fineness Ratio
13
13
13
Angle of Attack 13
Wash Out 13
Wash In 14
u Wing Area
Mean Chord (Geometric)
Taper Ratio
14
14
14
o Aspect Ratio
Wing Loading
Sweep Angle
14
14
14
Dihedral 14
Anhedral (or Cathedral) 15
Axes and Flight Controls (Primary Controls) 15
D~ ~
Introduction 19
f' Profile Drag 19
J Induced Drag 24
Total Drag 27
u"
1 i Wave Drag 30
~ n
o Introduction
Pressure distribution
Pressure gradients
33
34
35
Lift Equation 36
Lift/Drag Ratio 37
Movement ofthe Centre of Pressure 38
n
U Spanwise Distribution of Pressure 39
n
L J
;~
i, J
n
n
n
: I
'" .J
n
i, JI
n
lj
rl
U
[J
o
o
o
[1
u Module 8.2 Aerodynamics 2-5
Use and/or disclosure is
governed by the statement TIS Integrated Training System
on oaoe 2 of this chaater. © CODvriaht 2009
Integrated Training System
Designed in association with tile
club66pro.co.uk question practice aid
n
•I
lJ
n
.~
t... )
!~
'. )
Intentionally Blank
'I
l. J
n
, I
~.J
' II
: J
c I
n
LJ
n
'II
L J
Airflow
o Atmospheric Pressure
o In the previous chapter it was shown that the atmosphere exerts pressure at all times. This type
of pressure, which exerts a force on all bodies, is called static pressure and acts equally in all
directions. When air is in motion, however, it possesses an additional energy (kinetic energy)
due to the fact that it is moving, and the faster it moves the more kinetic energy it has. If moving
air is now brought to rest against some object, the kinetic energy is turned into pressure energy.
This pressure on the surface of the body which causes the moving air to stop is called dynamic
I"l pressure. The value of dynamic pressure depends on tie density of the air and its speed and
lJ may be expressed as:
This is an important equation which affects all aerodynamic studies. Any object in still air will
c experience static pressure in all directions but an object which is moving, or is placed in a
moving airstream will experience an additional pressure due to the moving air being brought to
rest.
o
r~
LJ
o t t
Figure 2.1 - Still air and the additional pressure created in moving air
U If the speed of the moving air is comparatively slow, say 100 kts, the dynamic pressure exerted
by it is quite small in relation to the static pressure at sea level. In fact the dynamic pressure will
only amount to less than 2% of the static pressure. If, however, the speed is increased to, say
450 kts, the dynamic pressure rises considerably, to about 30% of the static pressure. It is
important to note that at low speeds the density of the air is not significantly affected by these
changes in pressure and the air can be considered as an incompressible fluid. At high speeds,
however, say in excess of 300 kts, this assumption can no longer be made and the changes in
density due to compressibility become significant.
o
o Use and/or disclosure is
governed by the statement
Module 8.2 Aerodynamics 2-7
TIS Integrated Training System
on page 2 of this chapter. © Copyright 2009
l j
Streamline Flow 1
It is useful to illustrate the path followed by air when it passes around fixed objects and the
idiom used is that of streamlines. A streamline is the path traced out by a single particle of
airflow such that this particle does not cross the path of any other. This can be illustrated by
n
dropping dye into a stream of water and watching the visible path of the dye when it moves with
the water.
i]
o
n
l J
U
Flow Continuity
When water flows down a tube the principle of continuity of flow applies and the mass flow in
n
i
l ..
the tube is the same at any point along its length. This rule applies even if the tube is not of
constant diameter and this is clearly shown in the Figure 2.3. The mass flow at A, Band C is
the same so if the density of the water is 'p', the cross sectional area of the tube 'a' and the
n
speed of the water is 'v' then:
:]
Mass flow =p a v
The continuity theorem states that the mass flow at any point A = the mass flow at point B = the
mass flow at point C.
n
0 E 0 II
I I
l
n
J
t
A
t
B •
C
Figure 2.3 - Airflow through a duct of increasing cross sectional area
I.]
n
o
2-8
TIS Integrated Training System
Module 8.2 Aerodynamics
Use andfor disclosure is
govemed by the statement
n
© Copyright 2010 on page 2 of this chapter.
[1
I ,
L.J
Integrated Training System
Designed in association with the
club66pro.co.uk. question practice aid
c Venturi effect
In a venturi tube, such as that shown in Figure 2.4, that is a tube that has constriction in it, the
o previous rule still applies; mass flow is always a constant even if the tube is not of constant
diameter If, therefore, the pressure is measured at points 1 and 2 in the venturi it can be said:
u
c ..
.. -
'--- ....
- ..
c
o
•
+
Point 1
,...---
+
Point 2
---- ----
Considering the fluid as incompressible then as the cross sectional area a1 is considerably
c bigger than the cross sectional area a2, the speed V1 must be less than the speed V2. In other
words, as the flow passes through the venturi the speed of the fluid increases. This can often be
seen when watching the flow of a river through the arch of a bridge observing how the water
speeds up as it flows through the arch or constriction. The streamlines associated with this flow
are shown below. It can be seen that the streamlines draw closer together as they pass through
the venturi.
Moving away from the concept of the tube to that of an aircraft wing, as illustrated in Figure 2.5,
it can be seen that due to the curvature of the wing on its upper surface a venturi has been
o created between the upper surface and the undisturbed air some distance above it. The
streamlines will be similar to those in the venturi and, of course, the flow of the air will be
increased in speed as it passes through he venturi.
o
o
u
-t
Point 1 pott-2~-----------------------=
o Figure 2.5 - Airflow around an aerofoil section
Bernoulli's Theorem n
During the last century Bernoulli put forward his theorem stating that the total pressure (i.e.
static + dynamic) in a fluid is constant if no work is done by it or on it. n
Total pressure H = S + % pV2 = constant.
Referring back to Figure 2.5 and looking at the point ahead of the wing marked 1 we can find
n
the total pressure at this point:
n
Similarly the total pressure at the point marked 2 can also be expressed as:
However, Bernoulli's theorem states that the total pressure in a fluid is constant, therefore these
two expressions must equal each other.
Therefore:
Considering the density to be a constant factor and knowing the speed at point 1 is less than
the speed at point 2, it follows that the pressure at point 1 must be higher than the pressure at
n
"- ..J
point 2. To put it differently there is a reduction in pressure over the upper surface of the wing
as a result of Bernoulli's Theorem. It is this reduction in pressure over the upper surface of the
wing of an aircraft that creates lift and is the reason an aircraft can fly. n
Stagnation
Referring to Figure 2.6, note the flow of air around an object. Notice how the air divides - some o
flows over the top of the wing and some below it and right in the centre, at the leading edge of
the wing, the air is brought completely to rest at point A. This point is called the stagnation point
and it is where the full dynamic pressure plus whatever static pressure is effective at the time
o
n
will be felt.
~~~-------~~~==~~
A
n
:n
---- ~-----
Figure 2.6 - The stagnation point
i ..
c Airspeed
The principle of the stagnation point is used in the measurement of airspeed. Refer to Figure
2.7. Air is directed from a pitot tube facing into the airflow to a flexible diaphragm in the airspeed
indicator. This flexible diaphragm, in the form of a capsule, in fact is a stagnation point and will
o feel the full effect of dynamic pressure. Static pressure is fed to both sides of the capsule so that
it cancels out. The resultant movement of the diaphragm can be taken by a suitable linkage to a
dial, this indicating airspeed.
o It should be noted that the airspeed indicator is in fact a dynamic pressure indicator but is
calibrated suitably in knots. As it measures dynamic pressure directly it is extremely useful
c when flying the aircraft as most aerodynamic functions of the aircraft are directly related to
dynamic pressure. For instance, the stalling speed of an aircraft is always measured in
indicated airspeed and remains, for the same weight, pretty well a constant figure regardless of
r1 altitude.
L No mention has been made yet of compressibility and in fact this should be taken into account.
The airspeed indicator reading (corrected for instrument and position errors), when corrected for
D compressibility at all speeds is called equivalent air speed (EAS).
r1
U
o i~ Static
u
lJ
fl
L.J
D
o Use and/or disclosure is
governed by the statement
Module 8.2 Aerodynamics 2-11
TIS Integrated Training System
on page 2 of this chapter. © Copyright 2009
.J
Integrated Training System
Designed in association with the
c1ub66pro.co.uk question practice aid
n
l._.l
Speed Definitions
(a) Indicated Airspeed (lAS) - The dynamic pressure of air against a vehicle, or indicated
n
airspeed, is equal to Y, pV2, where p density, and V =true airspeed. An airspeed
indicator, calibrated to ISA mean sea level conditions records the dynamic pressure as a i.J
speed. If, for example, the indicated reading were 200 kts, then it means that the
dynamic pressure is the same as it would be at a true air speed of 200 kts at standard n
conditions at mean sea level. LJ
r·...,
(b) Rectified Airspeed (RAS) - The indicated airspeed, corrected for instrument and
position errors (IE and PEl.
lJ
(c) Equivalent Airspeed (EAS) - The rectified airspeed corrected for compressibility (C). It
should be noted that compressibility is always a subtracted quantity.
It is the airspeed at sea level which represents the same dynamic pressure as that flying
lJ
"
at the true airspeed (TAS) at altitude. It is useful for predicting aircraft handling,
aerodynamic loads, stalling etc.
Note that instrument error and position error are sometimes neglected and thus EAS is
n
(d)
then considered to be the same as lAS, and the two terms are used interchangeably.
(e) Calibrated Airspeed (CAS) - Some airspeed indicators are corrected for mean sea level
compressibility. Calibrated airspeed is the value of this reading, corrected for instrument
and position errors.
(f) Mach Number (Mn) - Mach number is the ratio of TAS to the local speed of sound l
,.
(LSS). I,
, I
I. I
II
2-12
TTS Integrated Training System
Module 8.2 Aerodynamics
Use and/or disclosure is
govemed by the statement
n
© Copyright 201 0 on page 2 of this chapter. n
IJ
[1
o Chord line
o The chord line of an aerofoil is the straight line joining the leading edge to the trailing edge. It is
normally used as a reference line when measuring the angular position of the wing related to
the airflow.
o
Leading Edge
[] Trailing Edge
rl
U -" - Line
- Camber - - - - --
Chord Line
[] Max Thickness
Chord Length
11
,
c.J
!
Figure 2.8 - A typical aerofoil
[J
Mean Camber line
o A line which joins the leading edge to the trailing edge such that it is equidistant from the upper
surface and lower surface of the aerofoil. If it is curved the aerofoil is described as cambered.
[]
Angle of Attack
r1 The angle of attack is the angle between the chord line of the wing and the direction of the
LJ relative airflow.
c
[1u Module 8.2 Aerodynamics 2-13
Use and/or disclosure is
governed by the statement TTS Integrated Training System
r--~I
, I on page 2 of this chapter. © Copyright 2009
Integrated Training System
Designed in association with the
club66pro.co uk question practice aid
Wash In n
U
An increase in angle of incidence from root to tip, caused by an intentional twist in the wing. It is
rarely incorporated because it decreases the stability of the wing.
n
,r .
Root chord
Aspect Ratio
The ratio of the wing span to the mean chord, or alternatively span 2 to wing area.
Wing Loading
The weight of the aircraft divided by the wing area.
Sweep Angle
The angle between the lateral axis and the Y. chord line (may be referred to as the leading 1
LJ
edge).
Dihedral
The upward inclination of the wing to the plane through the lateral axis.
:n,
c J
I
U
~--~ - - - -
Dihedral Angle o
Figure 2.10 - Dihedral angle n
u
o
_ ~Anhedral Angle
--~~----
- --
Figure 2.11 - Anhedral angle
D (a) Elevators
The elevator is attached to the trailing edge of the tailplane and controls the pitching moment
[J about the lateral axis. A backward movement of the control column moves the elevator up and
causes the aircraft nose to pitch up.
o
u
u
o
u
o
n
U
Figure 2.12 - The Elevator controls rotation about the Lateral Axis (pitching) - Longitudinal
Control
Figure 2.13 - The Ailerons control rotation about the Longitudinal Axis (roll) - Lateral Control D
(c) Rudder n
The rudder is attached to the rear edge of the fin and causes the aircraft to yaw about the
normal axis. Movement of the right rudder pedal forward moves the rudder to the right causing
the aircraft to yaw to the right about the normal axis.
n
i..
o
n
L J
n
IJ
n
iJ
n
,J
I .
Figure 2.14 - The Rudder controls rotation about the Normal Axis (yaw) - Directional Control
o
ruI
c
n Stabilatot'
rl
u
Figure 2.15 - The Stabilator
u (b) Spoilers
May be used instead of or in addition 10 ailerons. When the spoiler is operated it causes a loss
of lift on the side it is raised, thus causing a roll to that side. Movement of the control column to
the right causes the right spoiler to rise but the left spoiler to remain retracted.
sPOiler~
C ~
r:
L
Figure 2.16 - The Spoilers
,r" '
L.J
,r-':
U
[1
LJ
n
u
1
LJ
,...,
I !
U
o
n
Intentionally Blank n
u
n
~J
n
1.. ...
o
n
U
,..,,
,
I I
lJ
n
u
il
2-18 Module 8.2 Aerodynamics
lJ
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter.
n
II
u
r--'"
I '
Integrated Training System
LJ Designed in association with the
club66pro.co.uk question practice aid
o Drag
Introduction
It is convenient to study the subject of drag under two distinct headings:
Ii
LJ • Profile Drag or 'Zero Lift Drag'
• Induced Drag or 'Lift Dependent Drag'
o Profile Drag
Ii
I . Profile Drag is named Parasite Drag in the USA, where it means the same. Elsewhere, the
L-'
term Parasite Drag is reserved to denote the drag on an helicopter fuselage, in order to
differentiate it from the Profile Drag on the rotating blades. However, in this latter differentiation
of the use of the terms, their cause is identical, and is described below.
Succeeding layers adjacent to the surface will, through the same viscous action, be subject to
retardation, but to a lesser degree with increasing distance (albeit a very small one) from the
surface. A point is therefore reached where the airflow will be unaffected, and its velocity will be
that of the 'free stream' airflow.
This layer of air from the surface where there is zero velocity, to the point where there is no
retardation, is referred to as the 'Boundary Layer' and is normally defined as the region in
which the velocity of flow is less than 99% of the free stream value.
The boundary layer exists in two forms: (a) Laminar Flow, and (b) Turbulent Flow.
Physical laws dictate that at some point along a surface which is subject to a moving airstream,
'1 the flow will change from laminar to turbulent. This point is of importance in the study of drag,
I'-' i the significant feature being that the drag is greater in the turbulent layer than in the laminar.
The main variables which dictate the change from the laminar state to the turbulent are:
[J (a) Velocity of flow, (b) Viscosity of the fluid, or air, (c) Size of the object.
'1
U
u Use and/or disclosure is
Module 8.2 Aerodynamics 2-19
TIS Integrated Training System
governed by the statement
n on page 2 of this chapter. © Copyright 2009
i
,,, J
I
Generally speaking, the transition point (that point at which the laminar flow turns turbulent) for
an aerofoil section will be at the point of maximum section depth where the velocity of flow is n
greatest (refer to Venturi effect). As can be clearly seen, it pays to maintain laminar boundary
layer flow as long as possible over an aerofoil section in order to reduce drag, and to keep the
surface as smooth as possible.
[l
One method of ensuring a greater percentage of laminar flow is to maintain an increasing depth
of section as far back from the leading edge as possible, thereby locating the point of maximum
velocity farther aft. This results in a wing section known as a laminar flow wing; a description
which is, of course, only partially true;
n
lJ
Figure 2.17 indicates non-laminar and laminar sections.
n
IJ
c c -
n
t
Conventional Section
t
laminar Flow Section
(i)
(ii)
The depth of the laminar layer typically given as 0.07in.
The depth of the turbulent layer typically given as 0.7in.
n
(iii) The velocity gradients of the two layers being different leads to the greater shearing
or friction effect occurring in the turbulent layer.
n
LJ
Distance from
Surface 0.7i'n
n
Distance from
Surface .071n
Transition
Point
I
o
I
I 1 - - -....-1 Velocity Profile
I
Turbulent
..... ':'. ···_···7······ .
Laminar Sub-Layer
o
Figure 2.18 - The Boundary Layer in the transition from laminar to turbulent flow
o
2-20 Module 8.2 Aerodynamics
Use and/or disclosure is
n
I.. J
If we regard a flat plate at right angles to an airflow as being an extreme case, the kinetic
o energy of the airstream is largely brought to rest and converted to pressure energy: Figure 2.19
also shows the point'S' which is referred to as the stagnation point. The pressure brought to
rest, is referred to as the 'Dynamic Pressure'; it is of considerable importance, and is the
o pressure experienced by any object when a moving airstream is brought to rest: it is quite
distinct from static pressure. The formula for dynamic pressure is % pV2 where p air density =
o and V is velocity.
It may be seen from Figure 2.19, that the flow behind the plate is composed of vortices, and
since these have low pressure in the centre, we now have high pressure in front and low behind
rlLJ the plate: this results in a drag force in the direction of the moving airstream.
- - - - ·s· .:)-.;
'-'7 .
"!J.;>
-
c :;::
o (c)
Figure 2.19 - Turbulent wakes behind a flat plate and a sphere
c and this of course, is wasteful. By substituting a cylindrical section for the plate, we produce a
less abrupt change in the path which the airflow is trying to follow. In this case, fewer vortices
are generated: there is less difference in pressure from the front to the rear of the shape, and a
degree of 'streamlining' has been achieved.
Taken a step further, reference to Figure 2.20 will show a more streamlined shape as in a
,r, . symmetrical aerofoil section. This shape allows the airflow a much more gradual passage from
U the front of the section to the rear than in the case of the cylinder. The end result therefore, of
streamlining, is to produce much less vortex generation, reduced turbulence, and greatly
f1 reduced drag.
I:
~
: !
I i
LJ
n
U Use andfor disclosure is
Module 8.2 Aerodynamics 2-21
governed by the statement
TIS Integrated Training System
on page 2 of this chapter, © Copyright 2009
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
Although by careful streamlining considerable reductions in the form drag are achieved there is
a limit to extension of the method. The 'fineness ratio' of an aerofoil section is a measure of its
o
streamlining, and is defined as the chord to thickness length ratio. Figure 2.21 shows a section ;]
of conventional ratio, but if this ratio is too great, the resulting very thin section leads to
structural difficulties.
•
Length •
::==~IThickness
Figure 2.21 - Length and thickness of an aerofoil
The importance of the unit of dynamic pressure has already been emphasized and the above
n
basic formula is now modified by its inclusion. ifu
The new formula therefore, becomes:
r,
I .
I !
Drag = Co % pV2 S
LJ
rl.
where CD is the 'coefficient of drag', and S the wing area. The unit of dynamic pressure % pV2,
I
is very often written simply as 'q' because of its frequent use.
U
As a point of interest, the value of CD in the basic formula is about 0.6 for a flat plate. Other
n
L'
values of the drag coefficient that are of interest are:
•
I
I
I
I
I
I
I
I
I
I
I
.,
T"
I I
L.!
r1
w
o lAS -------_
c
fl
I :
~
r!
u
r'- ,
I ,
Induced Drag
Introduction
Pressure distribution over the upper and lower
surfaces of the wing was examined in the
previous chapter. As high pressure exists
underneath the wing and low pressure on top
of the wing, the one place where these
pressures will attempt to equalize is around
the wing tip. The high pressure underneath the
wing moves upwards towards the low i"l
pressure on the upper surface and in doing so U
assumes a rotary motion. This rotary motion
spirals back from the wing tip, moving in an I
anticlockwise direction from the right-hand eJ
wing tip as viewed from behind and in a
clockwise direction from the left-hand wing tip.
Energy is required to produce this rotational
vortex from each wing tip and this energy can
come only from thrust. The vortices therefore
create drag and this drag is called induced drag.
,
J
, , , ,
.,- -...... , ,
\ I I , --- ,
t I
6 :-'1
' ,,
\
....
, \ !
I., \, I "..• l .. ,
,,
•
•"
.... _-_ .. ,, , ,,
,,
t t t
;
, , ..... _....
,
I
,
, I I
;
,
;
,
-- - ,
,
n
Figure 2.24 - Wing tip vortices i\,_.J!
n
2-24 Module 8.2 Aerodynamics
lJ
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 201 0 on page 2 of this chapter.
i'
. I
LJ
[,1 Integrated Training System
Iw : Designed in association with the
club66pro.co.uk question practice aid
Downwash
The effect of the vortex is to deflect the air downwards as it passes over the trailing edge of the
wing, in other words producing downwash. As the maximum strength of this movement is close
o to the vortex, as one moves from the wing tip towards the fuselage the downwash steadily
decreases. Thus for a given strength of vortex, the larger the wing span the less will be the
effect of thisdownwash velocity.
I'L'
The angular deflection of the airflow will depend on the speed. For a given downwash velocity
the deflection angle will be greater at low speeds than at high speeds, as shown in Figure 2.25.
11 v, V2
....., e1 ~~~-----~~e-2------------D-o-w-n-w-as~h-
" Downwash
U --
r~
. I
.......... .......... Angle
..........
,
.....
--__
--------.....
Angle
U
Figure 2.25 - Downwash angles affected by airspeed
c The total reaction force of a wing is at right angles, not to the initial direction of the airflow, but to
the resultant between the original direction and the final direction. It will be readily seen that the
more the final flow is deflected downwards - in other words the bigger the downwash - the
more the total reaction is tilted rearwards, and this is clearly illustrated in Figure 2.26, The actual
usable lift in level flight has to be perpendicular. This leaves a small rearward component of the
total reaction force and this is induced drag.
From Figure 2.26 it will be seen that the larger the lift component the bigger will be the rearward
component Di, induced drag. Induced drag is in fact proportional to lift-squared.
Lift ~~ Drag
rl --------
,I '
~
'
-------- ----
Initial Flow - - ......
o Final Flow
(Downwash)
I'
I I
Figure 2.26 - Induced drag increases with lift exponentially (square law)
u
Span Effect n
d
The wing span of the aircraft has a marked effect on the amount of induced drag. The strength
of the vortex diminishes from the wing tip towards the fuselage and therefore the downwash
created by it also diminishes. For a given strength of tip vortex, therefore, the longer the wing
span the lower will be the average downwash and the lower the induced drag. For a given
amount of lift, the longer span and short tip chord produces a weaker vortex than a wing with a
short span and long tip chord and will therefore give less induced drag. In other words, the
higher the aspect ratio, the lower the induced drag.
Summary of Effects
The amount of induced drag created by a wing depends upon the amount of downwash and we
saw from previously that the slower the speed the bigger the angular deflection downwards of 11
the air. This therefore means that the induced drag is largest at low Speeds and decreases as
the speed increases. We can say from this that induced drag varies inversely as the square of
U
the speed or, induced drag is proportional to 1 2
speed o
To summarize the effects of induced drag, then:
Several deductions can be made from this summary. It becomes apparent that gliders and
sailplanes having very long, narrow wings - wings with a high aspect ratio - and normally flying
at very low speeds when induced drag is at its highest, therefore benefit from high aspect ratio
wings to reduce this drag to a minimum. Conversely, large jet transport aircraft do not usually
have high aspect ratio wings and, in addition, are usually of very high weight. From this it is
apparent that at low speed they will have very high induced drag.
n
Above 4° it decreases.
• At zero angle of attack a cambered aerofoil produces some lift and some drag.
• At zero angle of attack a symmetrical wing produces no lift but some drag.
• On a wing in flight, 2/3 of the lift is produced by the upper surface and the remainder by
the lower surface.
n
2-26 Module 8.2 Aerodynamics
:J
Use and/or disclosure is
TIS Integrated Training System governed by the statement ,.....,
© Copyright 2010 on page 2 of this chapter. , !
l j
n
I'
Integrated Training System
I I. Designed in association with the
Ll club66pro.co.uk question practice aid
Total Drag
,-I
I
Introduction
U
In preceding sections it has been shown that the aircraft is subjected to two types of drag,
profile drag and induced drag. Profile drag increases with speed and is proportional to the
square of the speed, and induced drag decreases with speed and is inversely proportional to
the square of the speed. These two curves of profile and induced drag are shown against speed
in Figure 2.27. The two curves can he amalgamated to give the total drag curve of the aircraft.
The lowest point on this total drag curve gives the speed at which the total drag is a minimum.
This speed is called the minimum drag speed, Vmd. For a constant weight and in straight and
level flight the Vmd will be a constant indicated airspeed for all altitudes. It would be reasonable
to assume that one would be better off flying the aircraft at Vmd because the drag is least at this
speed. In practice however, aircraft are not normally operated at this speed because the overall
efficiency, especially that of the engine, may be better at a higher speed.
n I
Iu i
o
(, I .
o
n
i :
U
rl
L
Speed ..
Figure 2.27 - Drag curves
o
I'II
U
It is of some importance when handling an aircraft to know if the speed is stable. By this it is n
lJ
meant that if for some reason the speed increases, perhaps due to temporary turbulence, does
the speed tend to decay back to its original value or not? The answer to this question can be
found by examining the total drag curve which is shown in Figure 2.28.
'1
~J
~
a:
T, n,"
a
'l, .
I ,. j
I
I
Vmd
SPEED
=
Consider an aircraft flying at speed X. In straight and level flight thrust drag so the thrust
required is indicated by the horizontal line T 1 . If for some reason the speed increases from X to
Y, the thrust remaining unchanged, the drag now exceeds tie thrust so the speed will drop back
to its original value at X. If, on the other hand, the speed decays to point Z, thrust is now greater
than drag and the speed will automatically return to its original value at X. It should be noted
that the two speeds quoted here are above Vmd. On the other side of the curve with the aircraft
flying at speed A the thrust level is now T2. If the speed now reduces to B the drag becomes
higher than the thrust and the speed will continue to decay. If, conversely, the speed increases
to point C, the thrust becomes higher than the drag and the speed will continue to increase. []
These second examples are speeds below Vmd. This simple illustration makes clear why at
speeds higher than V md the speed tends to be stable and at speeds below V md the speed is not
stable.
rr
I I
,J
This speed instability below V md is most marked on jet transport aircraft. The great weight of
some of the larger types of such aircraft today produces very high induced drag values and
makes handling on the approach somewhat difficult. Considerable anticipation is required to
either increase or decrease speed, the whole thing being aggravated to the slow response of jet n
engines. It would be fair to say that this type of aircraft requires more precise handling on the U
approach than the piston-engined aircraft and is less forgiving of imprecise handling.
In the section on induced drag it was shown that if weight is increased then induced drag also
increases. The minimum drag speed occurs at the point where the curve for profile drag crosses
n ....,
the curve for induced drag. As was seen from the first diagram, at this speed the value of the
i, \~
n
u
2-28 Module 8.2 Aerodynamics
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter.
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
I'I
I
induced drag is the same as that of profile drag. In other words profile drag equals induced drag
L.! and total drag is double the value of either one.
Figure 2.29 illustrates the fact that an increase in aircraft weight will raise the speed at which
Vmd occurs.
r~
I
L:
n
I I
"-"
u
o
fl, "
u
•
,•
I
o Vmdl
SPEED
Vmd'l
~
- - - - - - . . ,..
o It was shown in the previous chapter that the effect of an increase in aspect ratio is to decrease
induced drag. From this it follows that aircraft with high aspect ratios will have a lower V md than
aircraft with low aspect ratios.
Ii
w
F"
, I
I:
!.-J
n
I :
Li
, ,
I-~,
, I
L!
o
o Use and/or disclosure is
governed by the statement
Module 8.2 Aerodynamics 2-29
TIS Integrated Training System
,I on page 2 of this chapter. © Copvrioht 2009
'. i
L I
Wave Drag n
It has been shown that drag is the same at any altitude for a given lAS but an aircraft climbing at
this constant lAS has a steadily increasing Mach number. When this Mach number reaches a
certain value the drag starts to increase because of compressibility effects. This drag is known
n
, ,
as wave drag and its effect on the total drag curve is illustrated in Figure 2.30.
Il
I~ J
:l
.,
t .,
c .J
J
.'""'I
I. J
n
:1
SPEED --..)0
..... U
Figure 2.30 - Total Drag curves - with and without compressibility effect (Wave drag) n
l j
~
i I
n,
.
....,
lJ
n Induced Drag is directly related to lift because as the angle of attack is increased the induced
U drag will increase. Due to the greater pressure difference between the upper and lower surfaces
of the wing, the tip vortex (the basic origin of induced drag) will become intensified. For a given
speed the greater the angle of attack, the greater the induced drag.
It is important to realize that although induced drag increases with increased lift when increasing
Ir'I angle of attack, the increase in lift will always be much greater than drag up to and including the
I '.
U stalling angle.
ri
, I Remember the stalling angle is the angle above which a given aerofoil will stall.
I
~
I
Induced Drag is influenced by the aspect ratio of the wing, the higher the aspect ratio for a given
Induced drag is always greatest towards the wing tip where the tip vortex is generated by air
rl
u
flowing from the underside of the wing to the upper surface, where it then flows aft and down
behind the wing and tends to converge behind the aircraft. It can also be said the flow on the
upper surface of the wing tends to flow aft and towards the wing root, and on the undersurface
of the wing, aft and towards the wing tip, as a direct result of the influence of the vortex at the
wing tip. The flow towards the root and tip on the upper and lower surfaces cause numerous
vortices to form at the trailing edge of the wing.
n
I I
I I
~
The greater the chord length at the wing tip the more intense the tip vortex becomes and so the
greater the induced drag. Hence a high aspect ratio wing with a long span and a short chord will
produce less induced drag than a low aspect ratio wing with a short span and a long chord.
It is also important to note that a cambered aerofoil, even at zero angle of attack will produce
some lift and some drag. Even at some negative angles of attack a cambered aerofoil will
produce some lift and drag. But remember, a symmetrical aerofoil at zero angle of attack will
produce no lift but some drag.
n
~1
rl
, J
;J
, I
l J
n
n
!I
:" • .J
,:i,
C..J
n
L.J
Lift
[l
t.J
Introduction
It has been shown that if a streamlined body is placed in a moving airstream it produces drag, a
force in the direction of the airflow. It should be noted that the streamlined body we were
examining was symmetrical in shape. This drag force was the total force produced by the
streamlined body. If we now incline the streamlined body at a small angle to the airflow the total
force is now no longer in the direction of the airflow and this is illustrated in Figure 2.31. The
total force can now be resolved into two forces, drag, and the one at right angles to it, lift.
•,,,,.
Lift
r~. Resultant
\'-'! I
o ,,,
,
r'
I
L.J
!
.. I
I
I
I
..
...
(0
,, ,I
u Angle of
Attask
The diagram may give the impression that the lift and drag forces are approximately equal, but it
~
\ ,
I , has only been drawn this way for the sake of clarity. An aerofoil section in fact, produces lift
'-
many times greater than the value of drag it also produces. Bernoulli's theorem indicated that
[' there will be a reduction in pressure over the upper surface of the wing; this reduction provides
I, ''
~
approximately two thirds of the lift produced by a wing. The general pressure distribution over
the surfaces of a wing at a small angle of attack is illustrated in Figure 2.32 .
i. "i
LJ
11
n
I i
L.!
n
Stagnation
Point
.'.
A B c E •D
Figure 2,32 (a) - Pressure distribution around an aerofoil at a low angle of attack
,ij
Pressure distribution :, J
The upper surface of the wing produces a considerable reduction in pressure but the lower
surfaces produce a mixture of increase and decrease in pressure as well. The detail of Figure
2,32 shows that at the leading edge of the wing, point A, the full pressure is felt, this being the
stagnation point As the air moves over the upper surface of the wing, towards station B, it is
o
approaching an area of lower pressure and at station B the pressure is just atmospheric or
static, Past station B the pressure steadily reduces until it reaches its minimum value at C as n
t .)
indicated by the longest vector, and after C as the air moves towards the trailing edge of the
wing the pressure, although below static pressure, is now gradually increasing, The fact that the
air travelling from C towards 0 at the trailing edge is now moving against an adverse pressure
gradient is of considerable importance when we come to discuss stalling, On the under-surface
of the wing at point A the pressure was above static, in fact the full dynamic pressure was felt
there and to some extent an increase in pressure is felt on the under-surface of tie wing up to
about point E Thereafter the wing under-surface produces a small venturi of its own which
n
;)
gives a reduction in pressure, and in order to limit this reduction the under-surface of the wing is
i \
given considerably less curvature than the upper. \...,j
The pressure distribution as shown in Figure 2,32 (a), is for a comparatively small angle of Ii
, I
i., .J
attack, say about 4°C, Changes in the angle of attack of the aerofoil produce very considerable
changes in the pressure distribution and Figure 2,32 (b) illustrates the pressure pattern at a high
angle of attack, say about 12°.
ni I
l . .J
n
-
.i
I)
r'u:
II
LJ
\1
I .
I I
-....1
"
. I
\~1
Figure 2,32 (b) - Pressure distribution around an aerofoil at a high angle of attack
r"
I l
U
Pressure gradients
r;u The most obvious difference between Figure 2.32 (b) and (a) is the change of shape of the
below static pressure on top of the wing. The main feature of this new shape is that the point of
minimum pressure is very much nearer the leading edge of the wing than it was before. This
means that the air traveling from C to the trailing edge of the wing has to deal with a very much
rl
l..J longer and larger adverse pressure gradient. The only means available to the air to travel
against this adverse pressure gradient is its own kinetic energy - its energy of motion, and if
that adverse pressure gradient proves to be too great for the kinetic energy of the air, the flow
will in fact break away from the wing. On the undersurface of the wing the effect of the increase
in pressure is enhanced, thus providing more lift and the small amount of negative pressure
r:
\ r
towards the trailing edge has been reduced. The overall effect of the increase in the angle of
'--" attack is to increase lift but this process can only be carried out to a certain point and when this
point is reached, the wing stalls. The relationship between the angle of attack and lift is
n
ILJ: illustrated below. It can be seen that there is a steady increase in lift as the angle of attack
increases and then a sudden decrease at the stalling angle which occurs at about 16°.
o
i"'
U
[1
~ Module 8.2 Aerodynamics 2-35
Use and/or disclosure is
governed by the statement TIS Integrated Training System
on page 2 of this chapter. © Copyright 2009
I I
L .\
n
lJ
The size of a wing will obviously affect the amount of lift produced and this must therefore be
added to the equation:
The shape of a wing will also influence the amount of lift that can be generated and this
n
, I
\ ,
produces a factor, dependent upon the cross-sectional area of the wing, called the coefficient of
lift, CL. As demonstrated with angle of attack this will have an influence upon the amount of lift tiI
, I
generated. '- .J
The coefficient of lift is in fact derivative of the wing shape and its angle of attack. The full lift
equation can therefore be written:
The shape of the lift curve for any wing will be more or less the same but it should be noted that
the higher the camber of the wing the greater the lift it will develop. This is illustrated in Figure
2.34 where a cambered section is compared to a symmetrical section. A point of interest is that
although the cambered section still generates lift at a zero angle of attack the symmetrical
section does not. T'l,
, J
n
I '
I
L. .J
n
,"""'
i t
I~ .J
Angle of
Attack
lj
'i
Figure 2.34 - Lift curves for cambered and symmetrical aerofoils I ':
U
o Lift/Drag Ratio
The total resultant force derived from airflow over a wing can be resolved into two forces, lift and
o drag. The whole object of the exercise is of course to produce lift and in an ideal situation would
be done without incurring drag. Unfortunately, this is never possible but it is of great importance
to know the ratio between lift and drag so that the aircraft can be designed to provide the
11 maximum amount of lift for the minimum amount of drag. Lift and drag vary with the angle of
I '.
J
LJ
i attack and the variations of these two are shown in Figure 2.35 and 2.36.
r~
i
U
!
t
Angle of _ _ _ ___
Attack
o 18'
0.32 ,
I Angl&s,of
Usual
Flight·
I -..!
I
f'
,I !'
'-'
0.28'
;
I
0.24
c CD
0.20
I "~
~aq
U)~,
.::;'1
'
0.16
I ;':'i I
o ;I
0.12
V
! / , I
/r
I
0.08
II I
: iI
0.0'
----r : ,I
a - .i ,
fi
u
-5' o s· 10 15• 20'
Angle of Attack
If these two figures are combined mathematically they produce a curve as shown in Figure 2.37. n
It can be seen that there is a steady increase in the lift/drag ratio, which is what is desirable, u
until an angle of attack of about 4°. Thereafter the situation deteriorates as the lift/drag ratio
lessens until, at an angle of attack of around 15°, it tails right off, this being the stalling angle.
The highest point on this curve where we are getting the largest amount of lift for the smallest
amount of drag, occurs at about 4 0 and this is therefore the optimum angle of attack. Obviously,
the combination of most lift for least drag is the most efficient and why aircraft are usually flown
;,I i
o 5' 18'
;! Usual I i I
- - : ,Angles ,of Flig~t-, : '!I
25 ,
l J
r--
" ~I •
~ 1 iii '-.-~----'--,- - - " T - ' - - - j
'" .-. I '
,"
~
10 .!.! I I I
:t:: I i 'I
.-= I W,
LJ
'II-~:
:J 5
I
'-~~'-'.- - . ~ ~':" "'- :--:.'-. -
{
="0>' I
I I !!! c, :
0
l t
'5'
I
I : ; ~
'I
, .
;
. ;
. lJ
-5- o 5' 10' 15' 20' 25'
Angle of Attack n
,, J'
,i I,
Figure 2.38 - Centre of Pressure of an aerofoil \. J
r;
, I
r1
u Spanwise Distribution of Pressure
{'
The amount of lift produced by the upper surface of the wing will gradually decrease from root to
I I tip. This means that although the pressure on top of the wing is all below static pressure, it is
LJ
much lower near the root than it is near the tip. On the underside of the wing the reverse applies
,\
I "
and the pressure near the root is much higher than it is near the tip. Looked at in plan view, this
will cause the air flowing over the upper surface of the wing to be deflected inwards and the air
U flowing over the underside of the wing to be deflected outwards. This is illustrated in Figure
2.39.
I
I
/
n
U
r~
! !
W
o impart a downward velocity to the airflow. This downward movement of the air as it passes over
the trailing edge of the wing is called downwash.
rl
LJ
fl
i :
w Module 8.2 Aerodynamics 2-39
Use and/or disclosure is
:l
1
\J
n
'i
. I
'...J
Intentionally Blank
n
u
it
U
n
; I
\,J
n
:1
n
., I )
2-40 Module 8.2 Aerodynamics
Use andfor disclosure is
TIS Integrated Training System
© Copyright 2010
govemed by the statement
on page 2 of this chapter. n
, I
• J
n
I I
: )
~
ri
U Stall
rl Introduction
U
It has already been shown that the lift produced by a wing steadily increases as the angle of
[', attack is increased, but only up to a certain point. Past this angle of attack the lift decreases
IW ! rapidly and the angle at which this occurs is known as the stalling angle.
The Cause
The cause of the stall is the inability of the air to travel over the surface of the wing against the
adverse pressure gradient behind the point of minimum pressure. Figure 2.40 illustrates the
r --. pressure distribution over the upper surface of the wing at a small angle of attack, say about 4°.
i ! The minimum pressure point is at B, and the air travels from A to B without difficulty as it is
LJ
moving from high to low pressure. However, from B to C it is being forced to travel from low to
high pressure, that is, against an adverse pressure gradient. This poses no problems at low
1-\ angles of attack because the kinetic energy of the air is adequate to take it to the trailing edge.
[',
I '
LJ
A
nu
Figure 2.40 - Point of minimum pressure
o
n
u
n
I '
, !
U
nI
tI ,
n
u
AL--_ _
il
n
lJ
As angle of attack is increased however, the minimum pressure point moves forward and the
distance B to C increases until at the stalling angle, it covers most of the wing. This is illustrated
in Figure 2.41. When the angle of attack reaches a certain value the air runs out of kinetic
energy and breaks away from the surface of the wing in a random manner. Lift decreases
sharply and drag increases considerably.
Alleviation
Various design features can be incorporated in the wing which will assist in ensuring that the
root of the wing stalls before the tip. These are:
n
LI
• The wing may be twisted so that the tip is at a smaller angle of incidence than the root,
which will ensure that the root reaches its stalling angle before the tip. This is known as
'Wash-out' .
n
• The cross-section of the wing tip may be given a higher camber than the root, which will
give it a higher coefficient of lift.
n
• A stall-inducer may be fitted to the wing root as illustrated in Figure 2.42. These strips
reduce the effective camber of the root. This reduces its coefficient of lift and will cause it
to stall before the tip.
n
n
. )I
i
rJ
l
2-42
TIS Integrated Training System
Module 8.2 Aerodynamics
Use andfor disclosure is
n
governed by the statement
on page 2 of this chapter.
n
© Copyright 2010
i. i
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
c
o \
o
o
(,
Altitude Effect
fi In straight and level flight at the stall, for a given wing area, cross-section and weight, the lift is
I :
\...J of fixed value. This is a most fortunate occurrence when one considers the lift equation:
As lift at the stall is a fixed value and angle of attack, wing area and coefficient of lift are also
constant, the total value of Yo pv2 must also be constant. Yo pV2 is dynamic pressure shown on
the airspeed indicator and it is for this reason that for a given weight an aircraft will always stall
at the same indicated airspeed regardless of height.
Weight Effect
Any change of weight will require a different value of lift for straight and level flight, an increase
in weight requiring an increase in lift. At the stalling angle in level flight, the greater the weight
the more the lift required and, therefore, the higher the stalling speed. A useful rule of thumb in
n this context is that the percentage increase in stalling speed is half the percentage increase in
I :
I ! weight. Thus:
\....J
For example:
Further details of calculating staling speeds are given later in this chapter.
:l i.- .J
Effect of Shape
A wing does not normally stall over its entire length simultaneously. The stall begins at one part n
'.J
I '
of the wing and then spreads. The main factor governing where the stall begins is the shape of
the wing, and will be dealt with in a later section. It is plainly undesirable that a wing stalls from
its tip first as this can lead to control difficulties. Any tendency to drop a wing at the stall may
well lead to spinning. Further advantages of having a wing stall from its root rather than tip first
are that aileron control can be maintained up to the point of full stall and the separated airflow
from the wing root will cause buffet over tie tail which serves to act as a stall warning
When the angle of attack increases to high values the upward inclination of the thrust line n
provides a vertical component which acts in concert with the lift to support the aircraft's weight. ;J
L.
The slipstream from the propeller increases the speed of the air flowing over the wing, thus
delaying the stall. Caution should be exercised in power-on stalls as their effect may result in a
tip stall on a wing which normally stalls from the root. n
r
~J
n
: I
I•. J
Ii
U Centre of Gravity Position Effect
The stalling speed will be affected by the position of the centre of gravity. If the centre of gravity
is forward of the centre of pressure a down-load is required from the horizontal stabilizer. The
effect of this is that the lift is supporting not only the weight through the centre of gravity but also
the down-load on the tail, therefore the lift will have to be higher and in turn the stalling speed
will be higher. The nearer that the centre of gravity approaches to the centre of pressure, the
less will be the down-load and the stalling speed will consequently be reduced.
,rl,
LJ
,f"I
LJ
o
Tall trim
FwdCG limit
c=[:~=~~,-:" "~'" C ,
occ
1l ~" CD'" u.~ '_:r_"-"-'-0-'
_____
..
Tall trim
n
U Operational CG
r'
I i
U
Cl
I '
u
n
LJ Module 8.2 Aerodynamics 2-45
Use and/or disclosure is TIS Integrated Training System
governed by the statement
on page 2 of this chapter. © Copyright 2009
'. J
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid n
Icing Effect
The effect of ice formation on a wing is to corrupt the camber of the wing and so considerably to
reduce the coefficient of lift. This can be brought about by extremely thin layers of ice - even .....,
hoar frost - and the utmost care must be taken to de-ice the wings of an aircraft prior to takeoff i I
t .J
if there is any suggestion that ice may be present on the wings. The drastic effect of ice in
reducing the coefficient of lift and, as a result, causing the stalling speed to be much higher than
normal is illustrated in Figure 2.44. n
LJ
'l
iI ,I
n
, J
'I
lJ
.....,
!u I
ANGLE OF ATTACK II
'.J
Figure 2.44 - Stall angle with and without icing
n
Stall Warning Devices LJ
It is not normal to have an angle of attack indicator on the flight deck; it is usual instead to have n
some form of stall warning alarm operated by a switch which is sensitive to angle of attack. The :, ,\
warning may take the following forms:
Spinning
Following a stall involving a wing drop, a spin may develop. Referring to Figure 2.45, the wing
which drops increases its effective angle of attack due to having acquired a downward velocity.
This increase in angle of attack causes a further decrease in lift and an increase in drag. The
o
up-going wing, however, experiences a decrease in angle of attack and an increase in lift. As
the lift has been reduced on the downgoing wing it will continue to drop and any attempt to raise
it by the use of ailerons merely aggravates the situation because it will increase the angle of
n
attack still further. At the same time the increase in drag on the downgoing wing, coupled with a
decrease in drag on the up-going wing, will produce a yawing moment towards the dropped
wing. From this it can be seen that the aircraft will roll and yaw towards the dropped wing, and
n
this motion may be self-sustaining. If it is self-sustaining, the motion is described as a spin. il , J
2-46
TIS Integrated Training System
Module 8.2 Aerodynamics
Use anellor disclosure is
governed by the statement
n
© Copyright 2010 on page 2 of this chapter.
r'l
I j
'-'
,11 ,
LJ
o
YAW
1
ii
I i
U INCREASED V
= INCREASED L
o DROPPING WING
~
BECOMES STALLED
ct OR MORE STALLED
200
ct -L ...
-18'
RISING WING
BECOMES LESS
ct~ STALLED
j1 16'
+~~
u
II
u
r~
o
n
U
n
1 :
I '
c...i Module 8.2 Aerodynamics 2-47
Use andfor disclosure is TTS Integrated Training System
governed by the statement
© Copyright 2009
f1 on page 2 of this chapter.
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid n
The Deep Stall
Conventional recovery from a stall is by easing the stick forward to lower the nose and then
o
applying power. However, some aircraft of current design will enter into what is known as a
n
deep stall, or a super-stall, from which normal recovery is not possible. Broadly speaking, these
lJ
aircraft have swept back wings, high speed wing sections and a high T-tail.
n
lJ
The airflow following a stall in a conventional aircraft is illustrated in Figure 2.46. It can be seen
that although the air has broken away in a random manner from the upper surface of the wing,
the horizontal stabilizer and the elevators are still in undisturbed air. The result of this is that the
horizontal stabilizer will produce a sharp nose down pitch which may be assisted by application
of elevator.
o
-.--.---------.. . . -
. - 2~~~~~:::~
--_
o.
This can be contrasted with the state of affairs when an aircraft with a high T-tail is stalled. This U
"
time the separated air from the wings, following the stall, entirely covers the horizontal stabilizer
:)
and elevators, virtually reducing their effectiveness to nil. In the case of aircraft with sweep back I I
L.J
on the wings, the wing itself may develop a nose up pitching moment after the stall. This is due
to the tendency of a swept wing to stall at the tip and so cause the centre of pressure to move
forwards. The situation is often aggravated because the aircraft has now acquired a vertical
downward velocity which will progressively increase the angle of attack way beyond the stalling
angle. Finally, this type of aircraft is often equipped with rear-mounted engines and the effect of
turbulent air entering the engine intakes may be to cause them to flame out, causing a complete
,i I
l . ..J
loss of power.
Obviously an aircraft with these characteristics cannot be permitted to stall. When such an n
LJ
aircraft is first built, it is equipped with a tail-mounted parachute for use in test flying to thing the
nose down in the event of it entering a super-stall. For general airline operation, aircraft of this
type are fitted with equipment called a stick pusher. This is actuated by an angle of attack
n
, I
U
sensor on the fuselage (usually de-iced) which senses that the angle of attack is approaching
the stall. Signals are then sent to an electro-hydraulic system, the rams of which physically
push the control stick forward, thus preventing the aircraft from entering the stall.
r
I
It stands to reason that when the C l is maximum, V must be a minimum value (low speed, high
angle of attack).
'----1
c This minimum value of velocity is, therefore, the stalling speed (Vs), when the CL is at maximum
value.
,r--'l,
I !
Therefore: (iii) Lift = Weight = CL(max) % pVs S
2
LJ
So, rearranging the formula, it becomes:
Weight
Therefore (v) = (Cl % P S transposed)
CLImax) tps
r", Normal stall speed
IU'
Weight
=
o (vi) Vs
CLImax)
11
u 1. Weight
o 2. Load Factor
3. Wing Area
4. Change in CL(max)
that if the weight increases, the division thereof by CL(max) S results in an increased stalling
nJ
I.
speed (Vs).
2. Load Factor
Any manoeuvre that requires an increase in total lift without a corresponding increase in wing
area, must increase the effective total weight acting on the aerofoils.
This apparent weight increase is known as a load factor, which is defined as the ratio of the
n
load acting on the aircraft during the manoeuvre to the loading acting on the aircraft in straight
and level flight.
Load Factor =
Total Lift
Aircraft Weight
=
Total Weight
Actual Weight
n
u
As demonstrated in the previous paragraph, any increase in weight results in a higher stalling
speed. This new stalling speed may be calculated from the following formula: r;
LJ
New Vs =Old Vs x ~ Load Factor
Where increased wing area is obtained by the use of Fowler flaps, the division of a given weight
r;
by an increased value of (S) results in a lower value of V. ,L._I
4. Change in CL(max)
The use of flaps increases the CL of that wing. Once again, the division of a given weight by a n
larger value of CL results in a lower stalling speed. This is the advantage of the use of flap IJ
during the landing manoeuvre because it permits the original value of lift to be retained at a
lower speed. It is particularly useful in the lowering of the approach speed.
weight) of the wings and allows a much lower stalling speed to be attained. The slipstream at
high power settings provides an extra boost to the stagnating airflow over the aerofoil and thus
controls the boundary layer.
nJ
2-50 Module 8.2 Aerodynamics
I•.
n
TIS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter.
[ : Integrated Training System
U Designed in association with the
club66pro.co.uk question practice aid
Vertical component of
thru st (300 I b)
l0001b
[1
LJ
.In I, Figure 2.47 - Vertical component of thrust
LJ
Wing Tip Stalling
[-i An aircraft wing is designed to stall progressively from the root section to the tips.
LJ
The reasons for this are as follows:
o • Washout:
This is a progressive reduction of wing incidence from the root to the tip. This results in
the wing root reaching the critical angle of attack before the tip.
• Root spoilers:
This method employs a triangular-section strip fixed to the leading edge of the wing near
o the root. At high angles of attack, the airflow is obstructed in following the contour of the
leading edge and this results in a breakdown of the airflow whereby an early stall is
induced at the wing root.
II
LJ
,...,,
! ,
Note: Tapering the aerofoil from root to tip gradually reduces the CL towards the tips; this in
itself reduces the high rolling moment which would occur if one tip stalled before the other.
2.0 1
n
1.S ..... ... LJ
r--
1.6
V ~": ~~ ~o
l1
1.4
/ ~~
J..!!"
:;V
_l if
1.2 ~ I
/'
... ~,
,
1.0 I
:f
.~
-..;.
V
<II
0
iii
i
~-8
.
/ I
~
0
2 ii
V 0"
Ii=
8
:1:
-'-6
/ :;
.c
·i
0
~
In
'0
/
.~
;§
E ;§
-4 "
.~
c:
V III
::;:
E"
'x
2 I '"
:E
o
/ ,
-4 0' 4' 8• 12 • 16 20' 24' 28'
'Angle of Attack
'l
~J
n
lJ
2-52
TIS Integrated Training System
Module 8.2 Aerodynamics
Use andfor disclosure is
govemed by the statement
n
© Copyright 2010 on page 2 of this chapter.
1
i. )
o
Integrated Training System
o Designed in association with the
club66pro.co.uk question practice aid
r, As discussed under wing tip vortices, the net direction of the airflow is altered.
ILJ\
1.2
1.0' ______________ _
Max CL
o 0.8·
11L.! CL
~
.S>
0.6 (r "
o -;:.
~~ "
9.'"
o 0.4
o
o .4 " 0" 4" S" 12" i6"
o Angle of Attack
o Conversely, aircraft with low aspect ratio wings (broad tips) induce a large amount of downwash
which alters the net direction of the airflow significantly. These effects are illustrated in Figure
n 2.50.
LJ
Because of this altered airflow, low aspect ratio wings have significantly higher stalling angles
fl than do wings of high aspect ratio.
LJ
o
o Use andfor disclosure is
govemed by the statement
Module 8.2 Aerodynamics 2-53
TIS Integrated Training System
© Copyright 2009
il on page 2 of this chapter.
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
Note: This explains why rectangular wings usually stall from the root to the tip. The effective
stalling angle of attack is reduced at the tips due to the presence of the wing tip vortex resulting n
-
in the net direction of the airflow being altered, as shown in Figure 2.51.
n
.-- --p,:"--
Effective Stalling Angle of Attack Reduced
,:!.'(~.~:!:--
,- ••\\30\ '..~
Stalling Angle Higher
~es" .
Non-Effective Stalling
n
Angle of Attack et Direction of the R.A.F
n
Figure 2.50 - Effect of induced downwash on angle of attack
o
an\ I\./I..f. n
f\esU I\
Swept wings therefore, have higher stalling angles than those of unswept wings. n
n
1l J
r-;
:I
2-54 Module 8.2 Aerodynamics " .J
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter. 'I
i )
, J
u
Integrated Training System
o Designed in association with the
club66pro.co.uk question practice aid
u
o
o Elliptical Wing Rectangular Wing
i:
u
o
o Moderate Taper Wing High Taper Wing
o
u
o Pointed Tip Wing Sweepback Wing
n
u Figure 2.52 - Effect of wing planform on stall propagation
o
rlLi
o
o
u
o
o Use and/or disclosure is
govemed by the statement
Module B.2 Aerodynamics 2-55
TIS Integrated Training System
on page 2 of this chapter. © Copyright 2009
IJ
Integrated Training System
Designed in association with, the
club66pro.co.uk question practice aid
Flap Effect o
With each successive increase of flap, the characteristics of the aerofoil are changed, i.e. the
chord line assumes a steeper inclination, being the straight line from leading edge to trailing
edge. The critical stalling angle (about 15 degrees) is therefore reached with little or no
D
inclination of the longitudinal axis of the aircraft (i.e. aircraft in straight and level attitude). Any
further increase in flap setting in this attitude would result in the critical stalling angle of attack
being exceeded. To prevent this, the aircraft would have to be placed in a nose down attitude,
n
thereby reducing the critical angle of attack to within limits (about 15 degrees). rI I
i J
Straight and Level
.' No Flap LJD :!" 4·
....(
n
~«( R.A.F.
Flap Down
Aircraft Stalled
Lower Nose Attitude
Reduced.
Aircraft Unsta"ed.
n
LJ
~AEEEE ~,
,oc :;;. RAF '
. '('«
Figure 2.53 - The effect on chord line and hence angle of attack of flap deployment
Thus, the effect of flap reduces the stalling angle although the critical angle of attack remains
n
about 15 degrees.
D
Note: The stalling angle, or level flight stalling angle, is increased when leading edge flaps are
employed. n
lJ
Further reference to wing planforms and their stall characteristics are discussed in Chapter 8.3.
n
n
o
'0,
I,
n
2-56 Module 8.2 Aerodynamics
Use andfor disclosure is
n
n
TIS Integrated Training Syslem governed by the statement
© Copyright 2010 on page 2 of this chapter.
II I
u
Integrated Training System
Designed in association with the
club66pro,co.uk question practice aid
u
u TTS Integ rated
o
u Training System
o
c Module 8
11
U
[] Basic Aerodynamics
o 8.3 Theory of Flight
o
r:I
I
~I
lJ
o
[J
o Use andfor disclosure is
governed by the statement
on page 2 of this chapter.
Module 8.3 Theory of Flight 3-1
TTS Integrated Training System
© Copyright 2010
I I
I I
ilI J
Copyright Notice
© Copyright. All worldwide rights reserved. No part of this publication may be reproduced,
stored in a retrieval system or transmitted in any form by any other means whatsoever: i.e.
ilL,
photocopy, electronic, mechanical recording or otherwise without the prior written permission of
Total Training Support Ltd.
n
Knowledge Levels - Category A, 81, 82 and C Aircraft n
Maintenance Licence U
Basic knowledge for categories A, B1 and B2 are indicated by the allocation of knowledge levels indicators (1,2 or
3) against each applicable subject. Category C applicants must meet either the category B1 or the category B2 n
basic knowledge levels.
The knowledge level indicators are defined as follows:
II
LEVEL 1
n
LJ
• A familiarisation with the principal elements of the subject.
Objectives:
• The applicant should be familiar with the basic elements of the subject.
• The applicant should be able to give a simple description of the whole subject, using common words and
examples.
• The applicant should be able to use typical terms. .J
'-
LEVEL 2
• A general knowledge of the theoretical and practical aspects of the subject.
• An ability to apply that knowledge.
Objectives:
•
• The applicant should be able to understand the theoretical fundamentals of the subject.
The applicant should be able to give a general description of the subject using, as appropriate, typical
examples.
n
• The applicant should be able to use mathematical formulae in conjunction with physical laws describing the
subject.
• The applicant should be able to read and understand sketches, drawings and schematics describing the
subject.
n
• The applicant should be able to apply his knowledge in a practical manner using detailed procedures. rJ
I
l
I
LEVEL 3
• A detailed knowledge of the theoretical and practical aspects of the subject.
n
• A capacity to combine and apply the separate elements of knowledge in a logical and comprehensive
U
manner.
Objectives:
• The applicant should know the theory of the subject and interrelationships with other subjects.
n
LJ
• The applicant should be able to give a detailed description of the subject using theoretical fundamentals
and specific examples.
• The applicant should understand and be able to use mathematical formulae related to the subject.
• The applicant should be able to read, understand and prepare sketches, simple drawings and schematics
n
n
describing the subject.
• The applicant should be able to apply his knowledge in a practical manner using manufacturer's
instructions. L.
• The applicant should be able to interpret results from various sources and measurements and apply
corrective action where appropriate.
n
3-2
TIS Integrated Training System
Module 8.3 Theory of Flight
Use andlor disclosure is
governed by the statement
o
© Copyright 201 0 on page 2 of this chapter.
n
l.., j
" I
o Stall Patterns
Sweepback
Alleviating the Tip Stall
24
25
32
r'
I I
Pitch-up 33
L Forward Sweep 36
Delta Wings 37
Polymorphic Aircraft 40
Canard Design 42
Summary 42
u Flight Controls
Introduction
45
45
Aerodynamic Balancing 46
rl Power Operated Control 48
L.J
Mass Balance 49
3-Axis Control 50
u Tabs
Flight Forces
55
59
o Introd uction
Pitching Moments
Climbing, Gliding and Turning
59
59
61
[1 Turning and Manoeuvres: Points to Note 70
U Basic Manoeuvres 81
Lift 81
[J Stalling Angle 83
Stalling Speed 83
Multi-Engined Aircraft 83
[J V-Speeds 88
n
n
L..J
n
:-l
l J
n
n
[]
o
o
3-4
TIS Integrated Training System
Module 8.3 Theory of Flight
Use and/or disclosure is
govemed by the statement
n
© Copyright 201 0 on page 2 of this chapter.
n
c 1 Integrated Training System
L Designed in 'association with the
c!ub66pro.co.uk question practice aid
Ti
LJ
o
D
n
LJ
o
D
C
r"
LJ Use and/or disclosure is
Module 8.3 Theory of Flight 3-5
govemed by the statement TIS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
1 I
L .J
11J
l
'1
1 1
I ..J
n
lJ
Intentionally Blank
o
n
n
;j
...,
i I
l J
n
n
3-6 Module 8.3 Theory of Flight lJ
Use and/or disclosure is
TTS Integrated Training System govemed by the statement
© Copyright 2010 on page 2 of this chapter.
" !
[1
J
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
rI
L.J Module 8.3 Theory of Flight
Lift Augmentation
r1
U Introduction
High lift devices are incorporated on aircraft wings to reduce the distance required for take-off
and landing. The distance used by an aircraft either to take off or land depends on the speed
involved, and this speed in its turn is related to the stalling speed of the aircraft. An aircraft
cannot approach to land at a speed below its stalling speed, therefore the higher the stalling
r:u speed the longer the distance required to complete the landing run. The same applies for take-
off, the aircraft not being able to leave the ground until it has achieved flying speed, i.e. above
the stalling speed, therefore the lower the stalling speed under these circumstances the less
distance is required. All high lift devices produce the same effect, which is to increase the
coefficient of lift of the wing. The methods used for increasing the CL are:
lJ • Flaps
• Slats
I'I
I
• Boundary layer control
'-'
A flap is a hinged aerofoil section which can be mechanically lowered either from the trailing
11 edge or the leading edge of a wing. The effect of lowering a flap is to increase the overall
U camber of the wing and thus increasing the coefficient of lift. Some types of flap also increase
the wing area, thus augmenting the additional camber and producing even more lift.
rc.
LJ
Flaps
r'
I 1 There are many different types of flap in common use and some of the more usual ones are
U shown below.
Plain Flap
The plain flap is an aerofoil section merely hinging down from the trailing edge of the wing.
o
Figure 3.1 - Plain Flap
r'
i1 !•
L.J Module 8.3 Theory of Flight 3-7
Use and/or disclosure is
governed by the statement
TTS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
Integrated Training System
Designed in association with the
c!ub66pro.cQ,uk question practice aid
Split Flap
As Illustrated below, the split flap hinges down from the undersurface of the trailing edge. This
has the advantage that the camber of the upper surface of the wing is not disturbed but, at the
o
n
same time the flap will produce a considerable amount of drag.
:J
- n
Figure 3.4 - Fowler Flap
,....,
1J
i
'J
i
Krueger Flap
This is a leading edge flap which increases the leading edge camber, and is illustrated in below.
I)
u
11
L.J Figure 3.5 - Krueger Flap
The increase in camber caused by the lowering of flaps produces more lift from the given wing
section. If we consider straight and level flight, on lowering of flaps the greater lift will enable
r" either the angle of attack to be reduced or the speed to be reduced. Generally speaking, a
I !i
LJ compromise is reached between these two factors and the speed is considerably reduced with
a small reduction in the angle of attack. The effect of lowering flaps is not constant from zero
angle down to their full deployment. A selection of flaps down to approximately 30° will give a
very large increase in lift for a comparatively small penalty in drag, but further lowering, to say
60°, will not produce much increase in lift but will produce a considerable increase in drag.
When considering the distance required for take"off one might first feel that the lowest speed for
takeoff would give the shortest distance, the whole effect being achieved by large flap angle.
r~
I i Unfortunately, as already mentioned, a large flap angle incurs a very high drag penalty which
LJ reduces the acceleration of the aircraft, so you would end up with a rather long distance before
you could reach the un"stick speed. A lower flap angle would give a higher un-stick speed, but,
fl
, . with less drag, better acceleration would be achieved. In practice, a compromise is reached
w
between these two limits and a flap setting of somewhere in the region of 10° to 12° is used for
take-off.
11LJ
The distance required to land depends on the touch-down speed. The lowest speed will be
given by selection of full flap, this giving the lowest stalling speed. In addition the selection of full
n
LJ
flap will produce a very considerable amount of drag which will assist in decelerating the aircraft
on landing.
The effect of the slat is to prolong the lift curve by delaying the stall until a higher angle of
attack. When operating at high angles of attack the slat is generating a high lift coefficient
because of its marked camber. The resultant action, aerodynamically, is to flatten the marked
peak of the low pressure envelope, changing it to one with a more gradual gradient, as may be :I
seen on Figure 3.9. This flattening means that the boundary layer does not undergo the sudden ij
thickening due to negotiating the steep pressure gradient that existed behind the former peak,
so retaining its energy and enabling it to penetrate almost the full chord before separating.
Figure 3.10 also shows that although the pressure distribution is flatter, the area of the low
pressure region is unchanged or even increased. The passage of the boundary layer over the '"l
, I
wing is assisted by the air flowing through the slot (between slat and leading edge) accelerating ,J
through the venturi effect, thus adding to the kinetic energy and so helping it to penetrate
F~
against the adverse pressure gradient.
!J
3-10 light
Use andfor disclosure is
TIS Integrated Training System governed by the statement
© Copyright 201 0 on page 2 of this chapter.
II
I !
L~j
r 1
J
[1
, , ,
" .. -, With slot open
,,
1'-''
,
Without slot
Angle of Attack
Figure 3.8 - Change in stall angle with slot open
2.0
r 1 /....-r..., , ....-,-....
I!
'-l /
I
I ,;,
, I"
I \
/ ·ra/ I'
/ ,$/ I
1.5 q". " . / ~~'
~ , II
~~/ L. I I
r-- " ~/ I
~ 'I I I.
I i / I I
I
u
.I I I
1.0 / I I
(' I I I
U 0
5 I
I I
I
I
I
I I
I I
I I
0.5 I )
I I
c I
I
I
I
I
I
I
I
0
5 10 15 20 25 30
Angle of Attack
n
u
l
\ 1
n
;J
To summarize the foregoing thus far, slats have the effect of: ....,
!
I. ,
I
• Delaying separation until an angle of attack of 25° to 28° is attained, during which time:
• Lift coefficient has increased by approximately 70%. It should be evident that the stalling
speed of a slatted wing is significantly reduced, e.g. if an un-slatted wing stalls at 100 kts,
its fully slatted counterpart would stall at approximately 80 kts. The exact amount of
reduction depends on the length of leading edge covered by the slat, and the chord of
the slat.
n
LJ
Automatic slats are often located at the wing leading edge in front of ailerons in order to
increase the stalling angle by being automatically extended when the aileron moves down.
Normally the action of a down going aileron will reduce the stalling angle of the wing at that
point.
Slat Control
Since slats are of use only at high angles of attack, some method must be used to fair the slats
with the leading edge, thus precluding increased drag at normal flight configuration.
If the slats are small and the drag is negligible they may be fixed, i.e. non-automatic. Large
slats, however, are invariably of the automatic type. They are usually of the mechanical control,
hydraulically actuated kind, their selection being mechanically matched to the selection of flap,
the linkage being such that slats are extended before flap and before the speed reaches that
used for approach and landing. The reverse occurs on takeoff, when slats are fully in only after
flap is up, and at the correct airspeed. ii
, J
I
In the event of malfunction either of flaps or slats, it is usual to be able to 'split' the linkage
between the two, thus isolating the inoperative control, allowing the serviceable unit to operate
normally. On some aircraft the stall sensing unit may be used to extend slats only if the sensor
is activated by approach to the stall angle. Figure 3.11 illustrates a typical slat segment of the
o
kind more common to aircraft with swept back wings. fl
U
3-12 Module 8.3 Theory of Flight
Use anellor disclosure is
TIS Integrated Training System governed by the slatement
© Copyright 2010 on page 2 of this chapter.
Integrated Training System
Designed in association wi til the
club66pro.co.uk question practice aid
[1
u
Slat Closed
SI~t Open
r ''l
I '
I .1,
L../
ri
w
(,
LJ
(1
LJ
r'
\ i
'-'
ri
I
..-.JI
It can be seen that the nature of the boundary layer is a controlling factor in the determination of
surface friction drag, but more important still, the nature of the boundary layer also determines
the maximum lift coefficient, the stalling characteristics of a wing, the value of form drag and, to
some extent, the high speed characteristics of an aircraft. i'"""'J
The boundary layer cannot be eliminated entirely, though some measure of control of it may be
afforded by wing devices, one already dealt with being LE slots, which have the effect of re-
energizing the boundary layer. Others are:
" i
• Boundary layer fences to restrict the boundary layer outflow. They also check the
:J
•
spanwise growth of the separation 'bubble' along the leading edges.
Boundary layer suction; suitably placed suction points draw off the weakened layer so
that a new high energy layer can take its place.
n
• Boundary layer blowing; high velocity air injected into the boundary layer to increase its h, ,
energy. \~ 1
• Vortex generators; these re-energise the boundary layer, and are usually positioned
ahead of control surfaces.
n
3-14 Module 8.3 Theory of Flight
Use and/or disclosure is
i, I ,
TIS Integrated Training System govemed by the statement
© Copyright 2010 on page 2 of this chapter.
(1
I '
'--'
n
, I
Integrated Training System
Designed in association with the
U club66pro.co.uk question practice aid
• LE extension, also known as a 'sawtooth' LE - restricts the outward flow of the boundary
layer.
• LE notch has the same effect as LE extension.
r~
I '
c.....'
n
,
"-.J
'
Slats
[1 Flaps
u
o
o
r"
II
\....J
r'r Figure 3.13 - Typical slat and flap positions on a transport category aircraft
( i
.:...;
L.;
r-"
iI ir
W
n
;]
o.
Slat actuator
n
Slat Shown Fully Extended
4.0·
3.0
~
c
"
'u :0.'> ..,.-";;;- Max Ct.
~ .,,!> s\O ..,./ \ 2.4 at 22'
g 2.0' ~\" .....
-.rf'(\~/""
""
:J "" - - MaxCL
1.6at15'
Angle of Attack
15' 25' 30' n
Figure 3.15 - Effect on stall angles with flaps and slots .1 l j
i
\ J
f, Integrated Training System
U Designed in association with the
club66pro.co.uk question practice aid
---
_-----------
..
---------------
---------------
[I 1.
~
n
-~l~~~~~~~~~:::~~:::~~::----
:}~ ~ ~ ~TI~ ~ ~ ~ -~ ~I-t.~i;~-------- - -.-__-
~_~~~~==~~---
---------------------~----:.::::::=~3;=~z===~=:-::.:~~2:~~~~.
___
----
L.J :::::=:::::~:::=--
:~~~~¥:~~~~~~~~~~~§~~~~l;;~~:.:.:::
..._-----------
_---------- --:::.-::::::.-.. . --
~----
. . . --=-----:...----=--:..------
------_._----- --------_._---------
---
_-_----- ---- ----
--- ----------
......................
..................... -----
....................
...... -------~--
_=_ ____ : : -
L
0- _ _ _ _ _ _ _ _ ____ __ - - - __________
:==::::::::=~=-::-:.~c:_:::::::~;;::".......''''' -_ ---
rl ~~~~f:.~~~~~~~::=:=:=_~_~~~
-----------_ _---- ~~~~2~• •
--
::::--:-::::::::'::-::::':'- "--l-::'::.::I~~B--------J~j:
--_.
---
I
U
I ..
------ ---. --.-.
(-,I
\1
'-I
Figure 3.16 - Effect on airflow of slot
IIu
I'.
u
'-l
\
tv
i1
\J
r1
L
1. ,
, >
,I
; j
,, ,I
1i. J
M
1 i
:J
n
• J
:]
• I
, 1
\. J
I'
U Wing Planforms
II
U
Introduction
fl The previous chapters have considered the basics of lift, drag and stalling, and explained the
u causes of these phenomena. It is however, also necessary to study another important aspect of
the design of wings, i.e. the planform. The planform is the geometrical shape of the wing as
viewed from above; it largely determines the amount of lift and drag obtainable from a given
wing area, and has a marked effect on the value of the stalling angle of attack.
This chapter is concerned mainly with the low-speed effects of various wing planforms.
ri
'-'
ratio of 6. An alternative method of determining aspect ratio is by dividing the span by the mean
chord of the wing. Thus, a span of 50 ft with a mean chord of 5 ft gives an aspect ratio of 10.
From the preceding examples it can be seen that the smaller the area or mean chord in relation
II
L...'
to the span, the higher is the aspect ratio. A rough idea of the performance of a wing can be
obtained from knowledge of the aspect ratio.
U
r'
I I Aspect Ratio and Induced (Vortex) Drag
The origin and formation of trailing edge and wing tip vortices was explained earlier and it was
c shown that induced downwash was the cause of induced drag. The induced drag produced by a
wing is proportional to the lift generated.
The amount of induced drag under a given set of conditions can be found from the formula:
kC
rJ,. 2
o Induced drag =
2
qS
0.28
0.24 o
0.20
CD
0.16
0.12 n
,~
0.08
0.04
i
0
5 10 15 20 25 30 -- ~ iJ
Figure 3.17 - Co variation with angle of attack (00)
The reduced effective angle of attack of very low aspect ratio wings can delay the stall
considerably. Some delta wings have no measurable stalling angle up to 40° or more inclination
to the flight path. At this sort of angle the drag is so high that the flight path is usually inclined
downwards at a steep angle to the horizontal. Apart from a rapid rate of descent, and possible
loss of stability and control, such aircraft may have a shallow attitude to the horizon and this can
be deceptive. The condition is called the super stall or deep stall, although the wing may in fact
be far from a true stall and still be generating appreciable lift.
n
3-20 Module 8.3 Theory of Flight . I
\" )
f:
U Use of High Aspect Ratio
While a high aspect ratio wing will minimize induced drag, long thin wings conversely increase
rl weight and have relatively poor stiffness characteristics. Also the effects of vertical gusts on the
L airframe are aggravated by higher aspect ratio. Broadly it can be said that the lower the cruising
speed of the aircraft, the higher the aspect ratios that can be usefully employed. Aircraft
r 1 configurations which are developed for very high speed flight (especially supersonic flight)
u operate at relatively low lift coefficients and demand great aerodynamic cleanness. This usually
results in the use of low aspect ratio planforms.
fi
U
rl
U
rl u
1.0
0.5
Root Tip
.'1
, i
, I
'-._,J
. , Tip Chord
Taper Ratio, IA) =R Ch d
oot or
!;.I.!iRtical Rectangular
o[p n~ _.1
n
, ,
Sweepback
I
nI
Pointed Tip, A = 0 ,
',--.J
I
F
,.,
E I I
'.1
n, ,
I ,
Figure 3.18 - Lift distributions and stall propagation characteristics of different wing planforms
A planform previously considered is the simple rectangular wing which has a taper ratio of 1. A
characteristic of the rectangular wing is a strong vortex at the tip with local downwash behind
the wing which is high at the tip and low at the root. This large non-uniformity in downwash
causes similar variation in the local induced angles of attack along the span. At the tip, where , rn
high downwash exists, the local induced angle of attack is greater than the average for the
wing. Since the wing angle of attack is composed of the sum of U1 and uo, a large, local U1 n l J
3-22
TIS Integrated Training System
Module 8.3 Theory of Flight
Use andior disclosure is
governed by the statement
n
© Copyright 2010 on page 2 of this chapter.
n
i I
'-. 1
r'
) i
I '
I '
G
reduces the local ao , creating low local lift coefficients at the tip. The reverse is true at the root
of the rectangular wing where low local downwash exists. This situation creates an induced
angle of attack at the root which is less than the average for the wing, and a local section angle
of attack higher than the average for the wing. The result is shown by the graph, which depicts a
local coefficient at the root almost 20% greater than the wing lift coefficient.
i!
LJ
The effect of the rectangular planform may be appreciated by matching a near elliptical lift
distribution with a planform with a constant chord. The chords near the tip develop less lift
pressure than the root and consequently have lower section lift coefficients. The great non-
uniformity of local lift coefficient along the span implies that some sections carry more than their
share of the load while others carry less. Hence, for a given aspect ratio, the rectangular
planform will be less efficient than the elliptical wing. For example, a rectangular wing of A 6 =
would have 16% higher induced angle of attack and 5% higher induced drag than an elliptical
wing of the same aspect ratio.
At the other extreme of taper is the pointed wing which has a taper ratio of zero. The extremely
small area at the pointed tip is not capable of holding the main tip vortex at the tip and a drastic
rc change in downwash distribution results. The pointed wing has greatest downwash at the root
iLJ• and this downwash decreases towards the tip. In the immediate vicinity of the pointed tip an
upwash is encountered which indicates that negative induced angles of attack exist in that area.
r-' The resulting variation of local lift coefficient shows low C1at the root and very high C1 at the tip.
I• I'
I ' The effect may be appreciated by realizing that the wide chords at the root produce low lift
'-'
pressures while the very narrow chords towards the tip are subject to very high lift pressures.
The variation of CII CL throughout the span of the wing of taper ratio = 0 is shown on the graph.
As with the rectangular wing, the non-uniformity of downwash and lift distribution results in the
inefficiency of this planform. For example, a pointed wing of A = 6 would have 17% higher
r',
1...1
induced angle of attack and 13% higher induced drag than an elliptical wing of the same aspect
ratio.
o Between the two extremes of taper will exist planforms of more tolerable efficiency. The
=
variations of CII CL for a wing of taper ratio 0.5 are similar to the lift distribution of the elliptical
wing and the drag due to lift characteristics are nearly identical. A wing of A = 6 and taper ratio
r,-
, .
I I
= 0.5 has only 3% higher a1 and 1% greater C D1 than an elliptical wing of the same aspect ratio.
w
The elliptical wing is the ideal of the subsonic aerodynamic planform since it provides a
r' minimum of induced drag for a given aspect ratio. However, the major objection to the elliptical
LJ planform is the extreme difficulty of mechanical layout and construction. A highly tapered plan-
form is desirable from the standpoint of structural weight and stiffness, and the usual wing
planform may have a taper ratio from 0.45 to 0.20. Since structural considerations are important
in the development of an aeroplane, the tapered planform is a necessity for an efficient
configuration. In order to preserve aerodynamic efficiency, however, the planform is tailored by
rl
, , wing twist and section variation to obtain as near as possible the elliptic lift distribution.
I~I
,
·'0
'
i '
LJ
r
I 'i.
<J Module 8.3 Theory of Flight 3-23
Use andlor disclosure is
governed by the statement TIS Integrated Training System
on page 2 of this chapter. © Copyright 2010
i
]
M
! i
I I
Stall Patterns ;,. J
An additional outcome of planform area distribution is the stall pattern of the wing. The desirable
stall pattern of any wing is one where the stall begins at the root sections first. The advantages
of the root stalling first are that ailerons remain effective at high angles of attack, favourable stall
II
warning results from the buffet on the tailplane and aft portion of the fuselage, and the loss of
downwash behind the root usually provides a stable nose-down moment to the aircraft. Such a :l
l. J
stall pattern is favoured but may, in fact, be difficult to obtain with certain wing configurations.
• The elliptical planform has constant lift coefficients throughout the span from root to tip.
Such a lift distribution means that all sections will reach the stall at essentially the same
wing angle of attack and the stall will both begin and progress uniformly throughout the
span. While the elliptical wing would reach high lift coefficients before an incipient stall,
there would be little advance warning of a complete stall. Also, the ailerons may lack
effectiveness when the wing operates near the stall and lateral control may be difficult.
• The lift distribution of the rectangular wing exhibits low local lift coefficients at the tip,
and high local lift coefficients at the root. Since the wing will initiate the stall in the area of
highest local lift coefficients, the rectangular wing is characterized by a strong root-stall
tendency. This stall pattern is of course, favourable since there is adequate stall warning
buffet, adequate aileron effectiveness, and usually strong stable moment changes on the
aircraft. Because of the great aerodynamic and structural inefficiency of this planform,
however, the rectangular wing finds limited application, chiefly to low cost, low speed,
light planes.
n. ,
• The wing of moderate taper (taper ratio = 0.5) has a lift distribution which is similar to
that of the elliptical wing. Hence the stall pattern is much the same as that of the elliptical
wing.
(1.1.
• The highly tapered wing of taper ratio 0.25 shows the stalling tendency inherent with
high taper. The lift distribution of such a wing has distinct peaks just inboard from the tip.
Since the wing stall is started in the vicinity of the highest local lift coefficient, this plan-
form has a strong 'tip stall' tendency. The initial stall is not started at the exact tip but at
that station inboard from the tip where the highest local lift coefficients prevail.
• The pointed tip wing of taper ratio equal to zero develops extremely high local lift
coefficients at the tip. For all practical purposes the pointed tip will be stalled at any
condition of lift unless extensive tailoring is applied to the wing. Such a planform has no
practical application to an aircraft which is definitely subsonic in performance.
• Sweepback applied to a wing planform alters the lift distribution in a way similar to
decreasing the taper ratio. The full significance of sweepback is discussed in the
following paragraphs.
,-,
, I
~ J
Sweepback
Lift Coefficient
Swept Back
WingA=2
1.6
1.4
No Sweep
1.2 A" 2
1.0
0.8 Tailless Delta
WingA= 2
0.6
fl 0.4
I '
'-' 0.2
4 8 12 16 20 2428 32 36
Angle of Attack
The main reason for the lowering of the CL slope is best explained by examination of the
Figures 3.20 and 3.21. From the first it can be seen that the velocity V can be divided into two
components V 1 parallel to the leading edge which has no effect on the lift, and V 2 normal to the
(; leading edge which does affect the lift and is equal to V cos A. Therefore, all other factors
being equal, the CL of a swept wing is reduced in the ratio of the cosine of the sweep angle.
L
Figure 3.21 shows that an increase in fuselage geometric incidence'" will only produce an
increase in the angle of attack'" cos A in the plane perpendicular to the wing quarter chord line.
Since it has already been said that it is airflow in the latter plane which effects CL, the full incre-
ment of lift expected from the'" change is reduced to that of a '" cos A change.
11
:\. jI
"" 1
"" "---J
V "'
Considering the graph of Figure 3.19, the stall occurs on all three wings at angles of attack
,.....,
considerably greater than those of wings of medium and high aspect ratios. On all aircraft it is
. I
desirable that the landing speed should be close to the lowest possible speed at which the u
aircraft can fly; to achieve this desirable minimum the wing must be at the angle of attack
corresponding to the CLmax
n \., .>
:l
'. .J
rl
I ,
U
Incidence
change lI~
in this plane
fl
I :
'-'
/
Incidence·
change A oc cos ,A
in this plane
Figure 3.21 - The effective incidence change when an aircraft's pitch changes
n
u
n
,
L'
\
r-~I
Figure 3.22 - Three levels of sweep, and corresponding aspect ratio reduction
I i
U On all wings of very low aspect ratio, and particularly on those with a swept-back planform, the
angles of attack giving the highest lift coefficients cannot be used for landing. This is because
f;
. swept-back planforms have some undesirable characteristics near the stall and because the
,LJ ,
'
exaggerated nose-up attitude of the aircraft necessitates, among other things, excessively long
and heavy undercarriages. The maximum angle at which an aircraft can touch down without
n recourse to such measures is about 150 and the angle of attack at touch-down will therefore
u have to be something of this order. The CL corresponding to this angle of attack is lower than
the CLmax for each wing. Compared with the maximum usable lift coefficient available for
11
I , landing aircraft with unswept wings, those of the swept and delta wings are much lower, neces-
u
sitating higher landing speeds for a given wing loading. It is now apparent that, to obtain a
n common minimum landing speed at a stated weight, an unswept wing needs a smaller area
U than either of the swept planforms. The simple swept wing needs a greater area, and so a lower
wing loading, in order that the reduced CL can support the weight at the required speed. The
tailless delta wing needs still more area, and so a still lower wing loading, to land at the required
n
lJ
speed. Figure 3.22 shows typical planforms for the three types of wing under consideration, with
areas adjusted to give the same stalling speed. The much larger area of the delta wing is
evident. J
Effect of Sweepback on Drag
The main reason for employing sweepback as a wing planform is to improve the high speed
!l
\ _.J
characteristics of the wing. Unfortunately this has adverse effects on the amount of drag
produced at the higher range of angles of attack. The induced drag increases approximately in
proportion to _1_. This is because, as already explained, CL is reduced by sweeping the
cosA
wing, and therefore to maintain the same lift the angle of attack has to be increased. This
increases the induced downwash and hence the induced drag.
The practical significance of this high increase in drag is the handling problems it imposes
during an approach to landing. Because of the greater induced drag, the minimum drag speed
is higher than that for a comparable straight wing, and the approach speed is usually less than i
the minimum drag speed. Therefore, if a pilot makes a small adjustment to the aircraft's attitude iJ
by, for example raising the nose slightly, the lift will be increased slightly, but there will be a
large increase in drag, which will result in a rapid fall off in speed, with a large increase in power n,
needed to restore equilibrium. In fact, the stage may be reached where even the use of full i I
,-
power is insufficient to prevent the aircraft from descending rapidly.
n
• I
\ --;
Increased
Profile
Drag'
Drag
r Swept Win!!
Jl
/...........
Straight Wing
----------~ :l
VApp
lAS •
Figure 3.23 - Differences in drag at Approach Speed of a swept and straight wing
n
(VApp)
On some aircraft this problem is overcome by employing high drag devices, such as airbrakes
or drag-chutes, to increase the profile drag. This results in a flatter drag curve with the minimum
drag speed closer to the approach speed. A further advantage is that more power is required on
the approach, which on turbojet aircraft, means better engine response.
;-,
i I
'" J
n
I . Effect of Sweep back on Stalling
U When a wing is swept back, the boundary layer tends to change direction and flow towards the
tips. This outward drift is caused by the boundary layer encountering an adverse pressure
gradient and flowing obliquely to it over the rear of the wing. The pressure distribution on a
swept wing is shown by isobars in Figure 3.24. The velocity of the flow has been shown by two
components, one at right angles and the other parallel to the isobars. Initially, when the
boundary layer flows rearwards from the leading edge it moves towards a favourable pressure
gradient, i.e. towards an area of lower pressure. Once past the lowest pressure however, the
component at right angles to the isobars encounters an adverse pressure gradient and is
reduced. The component parallel to the isobars is unaffected, therefore the result is that the
actual velocity is reduced (as it is over an unswept wing) and also directed outwards towards
c the tips.
Pressure
Gradient
Across Wing
Isobars
o
Ii
I....0:
r,
I I
I \
<-..J
Pooling of
Boundary
r"', Layer at Tip
1 i
I,
U
Figure 3.24 - Isobars on the upper surface of a wing, and resultant airflow spanwise drift
r-j
I :
U The direction of the flow continues to be changed until the component at right angles to the
isobars is reduced to zero, whilst the parallel component, because of friction, is also slightly
C reduced. This results in a 'pool' of slow moving air collecting at the tips.
(1
U
The spanwise drift initiates a tendency towards tip stalling, since it thickens the boundary layer n
over the outer parts of the wing and makes it more susceptible to separation, bringing with it a ~ J
sudden reduction in elmax over the wing tips.
:li
At the same time as the boundary layer is flowing towards the tips, at high angles of attack, the ·, I
airflow is separating along the leading edge. Over the inboard section it re-attaches behind a
short 'separation bubble', but on the outboard section it re-attaches only on the trailing edge or
n
;, .I
fails to attach at all. The separated flow at the tips combines with the normal wing tip vortices to
form a large vortex (the ram's horn vortex). The factors which combine to form this vortex are:
.......
;L JI
,.-,
, ,
i 1
• J
,.....,
J
· .I
Flow
Around
Wing
Tips
n
" .f
These factors are illustrated in Figure 3.25 and the sequence of the vortex development and its
effect on the airflow over the wing is shown in Figure 3.26. From the latter it can be seen that
the ram's horn vortex has its origin on the leading edge, possibly as far inboard as the wing
root.
r;
The effect of the vortex on the air above it (the external flow) is to draw the latter down and
behind the wing, deflecting it towards the fuselage (See Figure 3.26) !)
The spanwise flow of the boundary layer increases as angle of attack is increased. This causes
the vortex closer inboard to become detached from the leading edge (see Figure 3.26). As a
result, outboard ailerons suffer a marked decrease in response with increasing angle of attack.
n
i I
·.
, I
[:
U
n
, I This, in turn, means that comparatively large aileron movements are necessary to manoeuvre
\....J'
the aircraft at low speeds; the aircraft response may be correspondingly sluggish. This effect
may be countered by limiting the inboard encroachment of the vortex as described on page 34,
or by moving the ailerons inboard. Another possible solution is the use of an all-moving wing tip.
,.
\, I.
'--'
1\
I ,
'-'
_.
-
.("j,
\ '
U
o
rl
I ' ~
LJ External Flow at Tip
Ie
I ;
L.J
('1
U
f;
u
<;
1 "
: , Module 8.3 Theory of Flight 3-31
U Use and/or disclosure is
govemed by the statement
TIS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
I I
\. J
Vortex generators
The purpose of these devices is to re-energise the boundary layer by making it more turbulent. n
; I
The increased turbulence results in high-energy air in layers immediately above the seriously ~_J
retarded layer being mixed in and so re-energizing the layer as a whole. In effect, the laminar
flow is 'tripped' to turn it into turbulent flow, effectively forcing the transition point further towards
the leading edge, at critical sections of the wing.
Vortex generators are most commonly fitted ahead of control surfaces to increase their effect by ;")
speeding up and strengthening the boundary layer, thus delaying separation point at such areas lJ
of high camber, albeit at the expense of a slight increase in drag. Vortex generators also
markedly reduce shock-induced boundary layer separation, and reduce the effects of the upper n
surface shockwave. u
Leading Edge Extension
Also known as a 'sawtooth' leading edge, the extended leading edge is a common method used
il ~, J
to avoid the worst effects of tip stalling. The effect of the extension is to cut down the growth of
the main vortex. A further smaller vortex, starting from the tip of the extension, affects a much
smaller proportion of the tip area and in lying across the wing, behind the tip of the extension, it
has the effect of restricting the outward flow of the boundary layer. In this way the severity of the
tip stall is reduced and with it the pitch-up tendency. Further effects of the leading edge ;l ,J
extension are:
• The thickness-chord ratio of the tip area is reduced, with consequent benefits to the
critical Mach number.
• The CP of the extended portion of the wing lies ahead of what would be the CP position if
no extension were fitted. The mean CP position for the whole wing is therefore farther
forward and, when the tip eventually stalls, the forward shift in CP is less marked,
:J
therefore reducing the magnitude of the nose-up movement. ,....,
: I
"I
n
3-32 Module 8.3 Theory of Flight '. )
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter. ,-,
, ,I
c
o Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
r~
Leading Edge Notch
U The notched leading edge has the same effect as the extended leading edge in so far as it
o causes a similar vortex formation thereby reducing the magnitude of the vortex over the tip area
and, with it, the tip stall. Pitch-up tendencies are therefore reduced. The leading edge notch can
be used in conjunction with an extended leading edge, the effect being to intensify the inboard
vortex behind the devices to create a stronger restraining effect on boundary layer outflow. The
choice whether to use either or both of these devices lies with the designer and depends on the
desired flight characteristics of the aircraft.
Pitch-up
Longitudinal Instability
Longitudinal instability results when the angle of attack of a swept wing increases to the point of
tip stall. The instability takes the form of a nose-up pitching moment, called pitch-up, and is a
self-stalling tendency in that the angle of attack continues to increase once the instability has
set in. The aerodynamic causes of pitch-up are detailed in the following paragraphs.
Wing
l
r~ Moment
I '
LJ
o Figure 3.28 shows that the maximum downwash from a swept-back wing in unstalled flight
comes from the tip portions; this is to be expected since the CL is highest over these parts of the
wing. When the wing tips stall, effective lift production is concentrated inboard and the
maximum downwash now operates over the tailplane and increases the tendency to pitch up.
This effect can be reduced by placing the tailplane as low as possible in line with, or below, the
rl wing chord line, so that it lies in a region in which the downwash changes with angle of attack
U are less marked.
G
U
n
,I ''
u Use andfor disclosure is
Module 8.3 Theory of Flight 3-33
TIS Integrated Training System
governed by the statement
on page 2 of this chapter. © Copyright 201 0
tJ
Integrated Training System
Designed in association with tile
club66pro.co.uk question practice aid
rJ
n
Unstalled Tip Stalied \J
\ .
r. I
Ie.)
Max Downwash
Max Downwash
L
Resultant
Lr :I
Downwash ! \
............. "
~
Nose-up
Pitching Moment
,
I
\
I
Tail Moment
Increased
\
Wing Moment
Decreased
Increased
rail Moment
o
n
Figure 3.28 - Downwash effects on the tailplane during tip-stall
; I , ,
"
Washout due to Flexure
When a swept wing flexes under load, all chordwise points at right angles to the main spar are ;l "
raised to the same degree, unless the wing is specially designed so that this is not so.
Therefore in Figure 3.29, the points A and B rise through the same distance and the points C
and D rise through a same distance but one that is greater than A and B. Therefore Crises
farther than A and there is a consequent loss in incidence at this section. This aeroelastic effect
is termed 'washout due to flexure', and it is obviously greatest at the wing tips.
n
(1
I I
l.J
'I
: I
',.~J
n
, I
3-34 Module 8.3 Theory of Flight 'c_ )
Use and/or disclosure is
TIS Integrated Training System govemed by the statement
© Copyright 2010 on page 2 of this chapter.
Ir"I
LJ
o It is most noticeable during high-G manoeuvres when the loss of lift at the tips and the
consequent forward movement of the centre of pressure causes the aircraft to tighten up in the
manoeuvre. A certain amount of washout due to flexure is acceptable provided the control in
f: pitch is adequate to compensate for it, but it can be avoided by appropriate wing design.
I !
LJ
Pitch-up on Aircraft with Straight Wings
On aircraft with low aspect ratio, short-span wings, pitch-up can be caused by the effect of the
wing tip vortices. As the angle of attack is increased the vortices grow larger until, at or near the
stall, they may be large enough to affect the airflow over the tailplane. As each vortex rotates
ii
, ! inwards towards the fuselage over its upper half, the tailplane incidence is decreased giving rise
'-' to a pitch-up tendency.
Rate of Pitch-up
From the pilot's point of view, pitch-up is recognized when the pull force on the control column
which is being applied to the aircraft near the stall has to be changed to a push force to prevent
the nose from rising further; the more the speed decreases the farther forward must the control
column be moved to restrain the nose-up pitch. Pitch-up in level flight or in any 1G stall is
usually gentle, since the rate at which the stall is spreading is comparatively slow and is usually
accompanied by the normal pre-stall buffeting. When the stall occurs in a manoeuvre incurring
accelerations due to 'G' force, the onset of pitch-up can be violent and sudden, corresponding
to the rate of spread of the stall.
r'
,
L~
I
[}
o
Module 8.3 Theory of Flight 3-35
Use and/or disclosure is
governed by the statement TIS Integrated Training System
on page 2 of this chapter. © Copyright 2010
j'
i
t
I
.J
Forward Sweep
n
u
General
The benefits of wing sweep can be achieved by sweeping the wing backwards or forwards, yet
only in recent years has the forward swept wing (FSW), become a serious alternative to
sweepback. The reason for this lies in the behaviour of wing structures under load. n
,. )
The main advantages lie in the sub/transonic regime. Taking the 70% chordline as the average
position for a shock-wave to form when the critical Mach number is approached, the sweep
angle of this chordline influences wave drag.
n
The FSW can maintain the same chord-line sweep as the swept-back wing (SSW) but due to a l
I ~
geometric advantage, achieves this with less leading edge sweep and enjoys the advantages
accruing from this subsonically. n
~ J
The decision to employ FSW or SSW will depend, inter alia, on the speed regime envisaged for
the design. Due to better lift/drag ratio in the sub-sonic and near transonic speed range - typical
combat air patrol - fuel consumption is improved over the SSW. For a high speed supersonic
interception the higher supersonic drag is a disadvantage.
Wing Flexure
n
i.. "'
Under flexural load the airflow sees a steady increase in effective angle of attack from root to
tip, the opposite effect to aft-sweep. Under g loading, lift will be increased at the tips, leading to
pitch-up as the centre of pressure moves forwards. Additionally, the increased angle of attack at
the tips now leads to increased wing flexure, which itself leads to increased effective angle of
attack at the tips. The result of this aeroelastic divergence is likely to be structural failure of the
wing, so it is not surprising that sweepback was considered to be a better option until
"I.J
I :
comparatively recently. What changed the situation was the development of carbon fibre
technology, which made possible controlled wing twist under load; so allowing the effect
described to be eliminated. n
Vortex Generation
Figure 3.30 shows the difference in ram's horn vortex behaviour. In the swept forward design
the ram's horn vortex develops inwards towards the root, not outwards towards the tips.
n
iJ
There will, of course, still be vortices from the wing tips, but these no longer reinforce and
aggravate the ram's horn vortex, which now lies along the fuselage, or slightly more outboard if
a small section of the wing root is swept back.
n
n
,
lJ
.
rLJ1
,-,
,,, ,,.
U
n
\ !
I '
J
[J
Figure 3.30 - Spanwise drift of forward and rearward swept wings
u Stalling
A swept forward wing will tend to stall at the root first. This stall can be controlled in a number of
ways. Since a conventional tailplane would tend to lie in a vortex, the popular option is to
combine forward sweep with a canard foreplane. Downwash from a carefully placed canard can
delay root stall, and even the vortices from the canard can be used to energise the airflow over
rl
i I inboard sections of the wing, maintaining lift up to higher angles of attack.
L)
The root-stall characteristics give better control at the stall as aileron control is retained, but may
incur a penalty in directional control as the fin and rudder are acting in the chaotic turbulence
from the root separation.
r,
U Delta Wings
Tailless Delta
On aircraft Llsing this type of wing the angle of attack is controlled by movement of the trailing
edge of the wing: an upward movement produces a downward force on the trailing edge and so
i1u increases the angle of attack. When compared with an identical wing which uses a separate
tailplane to control the angle of attack, the tailless delta reveals two main differences:
n'-'
edge. If a given wing/aerofoil combination has a hinged trailing edge for use as an elevator,
then when the trailing edge is moved from one angular position to another, the effective aerofoil
section of the wing has been changed.
nI \
U When such a wing reaches its stalling angle in level flight, the trailing edge elevator must be
raised to impose a downward force on the trailing edge to maintain the wing at the required
[1 angle of attack. The raised trailing edge has two effects: it deflects upwards the airflow passing
L!
fl
I I
u Use andfor disclosure is
Module 8.3 Theory of Flight 3-37
governed by the statement TIS Integrated Training System
on page 2 of this chapter. © Copyright 2010
i. J
~
---__rt c:::--r--- n
Basic Section
[l
Lift n
n
; !
, J
o
Angle of Attack
"l. JI
i
Figure 3.32 - Effect on stalling angle of delta without and with a tailplane !l
\ J
o All else being equal (planform, aspect ratio, area, etc), changes in the amount of camber (by
altering the angular setting of the trailing edge elevator) do not affect the stalling angle
o appreciably. That is, the angle between the chord line and the direction of the airflow remains
constant when at maximum Cl irrespective of the setting of the hinged trailing edge. Figure 3.32
o
r,
illustrates this point and it can be seen that for both the 'tailed' and 'tailless' aircraft the stalling
angle is the same when measured on the foregoing principles.
It is however, normal practice and convention to measure the stalling angle with reference to the
I, I' chord line obtained when the moveable trailing edge is in the neutral position, and not to
u
assume a new chord line with each change in trailing edge movement. When the stalling angle
is measured with reference to the conventional fixed chord line, it can be seen from Figure 3.32
that the angle is greater. The diagram also shows that, because the wing proper is set at a
greater angle at the stall when a trailing edge elevator is used, the fuselage attitude is more
nose-up, giving a more exaggerated attitude at the stall in level flight.
Since it is easier to refer to angle of attack against a fixed chord line, the basic chord line is
always used as the reference datum. This convention is the reason for the apparently greater
stalling angles of tailless delta wings; it is perhaps a more realistic method, as the pilot is
invariably aware of the increased attitude of his aircraft relative to the horizontal, but is not
r: always aware of increases in the angle of attack.
U The Cl Curve
n
<-I
The peak of the curve for the lift coefficient is very flat and shows little variation of Clover a
comparatively wide range of angles. This very mild stalling behaviour enables the delta wing to
be flown at an angle of attack considerably higher than that of the Clmax> possibly with no ill
,-. effects other than the very marked increase in the drag. The fiat peak denotes a gradual stall,
IJ with a consequent gradual loss of lift as the stalling angle is exceeded.
o advantage.
o
o
r',
I
! I
G
• Leading edge flow separation causes CP to be situated nearer mid-chord. Hence there
will be less difference between CP subsonic and CP supersonic than before, and
o
Ii
longitudinal stability is thereby improved. , I
!
'."
• The vortex core is a region of low pressure, therefore an increase in CL may be expected.
On the conventional delta this cannot be utilized as the vortex seldom approaches
anywhere near the wing root and most of its energy appears in the wake behind the
wing, where it produces high induced drag. On the slender delta the low pressure in the 'I
vortex is situated above the wing and can result in an increase in CL of as much as 30% I, J
under favourable conditions.
n
, ,
, I
LJ
Polymorphic Aircraft
General
An aircraft which is designed to fly at supersonic speeds most of the time usually has poor low
speed characteristics which have to be accepted, although various high lift devices are available
for reducing take-off and landing speeds and improving the low speed handling qualities. In
order to achieve the desired high speed performance, the aircraft has thin symmetric wing
sections and highly swept or delta wing planforms; these wings are very inefficient at low n
speeds where unswept wing planforms and cambered wing sections are required. iJ
n
In the case of an aircraft which is required to be operated efficiently at both high and low , I
I ,
speeds, variable wing sweep is a desirable feature to be incorporated in the design. The wings U
can thereby be swept back when the aircraft is being flown at high speeds and swept forward
again when flying at low speeds. Such aircraft are often labeled 'swing wing'.
essential that it should be kept as small as possible. Various design methods are available for
reducing or eliminating the trim changes produced by sweeping the wings.
Wing Translation
The aerodynamic centre can be moved forward again by translating the wing forwards as it is
swept back. This method involves extra weight and structural complications.
'l
U Movement
The aircraft can be designed so that the CG moves rearwards in step with the aerodynamic
o centre by mounting some weight in the form of engines, etc at the wing tips. As, however,
engines would have to swivel to remain aligned with the airflow, additional weight and other
complications result. Another possible method of moving the CG is by transferring fuel to
11
I, suitable trim tanks in the rear fuselage.
u
Leading Edge Fillet and Pivot Position
ri Another solution can be obtained by positioning the pivot point outboard of the fuselage inside a
1 i
U
fixed, leading edge fillet, called a 'glove'. The optimum pivot position for minimum movement of
the CP depends on the wing planform, but it is usually about 20% out along the mid-span.
However, the fixed glove-fairing presents a highly swept portion of the span at low-speed,
minimum-sweep settings. This incurs the undesirable penalties that variable geometry is
designed to overcome. A compromise between sweeping the whole wing and a long glove
giving the minimum CP shift, is usually adopted as indicated in Figure 3.33.
o
f'
, I
U
Pivot Pivot
o
,',
I
U
i
Fuselage Pivot
Glove-Fairing Pivot
Large CP Shift
Small CP Shift
r,
, ;
Figure 3.33 - CP shift in a polymorphic aircraft
I .
U
o
u
Module 8.3 Theory of Flight 3-41
Use andfor disclosure is
governed by the statement
TIS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
,
.J
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
n
Canard Design o
A canard-type configuration is one which has a foreplane located forward of the wing instead of 'I
the more conventional tail plane. On an aircraft with a long slender fuselage with engines
mounted in the tail and a CG position well aft, this layout has the obvious geometric advantage
U
of a longer moment arm. This enables the stability and trim requirements to be satisfied by a
foreplane of smaller area than that of a tailplane. The trim drag problem will also be reduced
because, at high speeds, an up-load will be required on the foreplane to trim the aircraft. There
o
are, however, certain disadvantages with this layout:
Stalling problems
On a 'conventional' tailplane configuration, the wing stalls before the tailplane, and longitudinal
control and stability are maintained at the stall. On a canard layout, if the wing stalls first,
stability is lost, but if the foreplane stalls first then control is lost and the maximum value of CL is
n
reduced. One possible solution is to use a canard surface and a wing trailing edge flap in
combination, with one surface acting as a trimming device, and the other as a control.
Alternatively, an auxiliary horizontal tailplane at the rear may be used for trim and control at low
speed.
Interference Problems
In the same way as the airflow from the wing interferes with the tail unit on the conventional tail n
layout, so the airflow from the foreplane interferes with the flow around the main wing and ( J
vertical fin in a canard layout. This can cause a reduction in lift on the main wing, and can also
result in stability problems. The interference with the vertical fin can cause a marked reduction
in directional static stability at high angles of attack. The stability may be improved by employing
twin vertical fins in place of the single control vertical fin.
Summary
l. )
Planform Considerations
Planform is the geometrical shape of the wing when viewed from above, and it largely
determines the amount of lift and drag obtainable from a given area, it also has a pronounced l
( )
effect on the stalling angle of attack.
Aspect ratio (A) is found by dividing the square of the wing span by the area of the wing:
'II
'. )
2
A = Span or Span
Area Mean Chord
rl In the aerodynamic sense, the elliptical wing is the most efficient planform because the
lJ uniformity both of lift coefficient and of downwash incurs the least induced drag for a given
aspect ratio.
o Any swept-back planform suffers a marked drop in C lmax when compared with an unswept wing
with the same significant parameters. Also the boundary layer tends to change direction and
flow towards the tips.
The spanwise drift of the boundary layer sets up a tendency towards tip stalling on swept wing
aircraft. This may be alleviated by the use of one or more of the following:
[1
I I.
'--' • Boundary layer fences.
• Leading edge slots.
n
• Boundary layer suction.
U • Boundary layer blowing.
o •
•
•
Vortex generators.
Leading edge extension.
Leading edge notch.
n • Longitudinal instability.
I I • Centre of pressure movement.
U
• Change of downwash over the tailplane.
• Washout due to flexure.
A FSW stalls at the root first, prolonging aileron control. The configuration may offer an
When compared with a delta which uses a separate tailplane to control angle of attack, the
o •
•
The Cl max is reduced.
The stalling angle is increased.
o
rl
LJ Use and/or disclosure is
Module 8.3 Theory of Flight 3-43
TIS Integrated Training System
governed by the statement
on page 2 of this chapter. © Copyright 201 0
LJ
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
Advantages
n
• The control surface is ahead of any shocks which may form on the mainplane.
n
• On an aircraft with a long slender fuselage with engines mounted in the tail and the CG
position well aft, the fore-plane has the advantage of a long moment arm.
• The stability and trim requirements can be satisfied with a smaller foreplane area.
• Because up-loads will be required, the trim drag problem is reduced.
n
u
Disadvantages
•
•
If the wing stalls first stability is lost.
If the foreplane stalls first control is lost.
n
• In the same way as the airflow from the wing interferes with the tail unit on the
conventional tail layout, so the airflow from the foreplane interferes with the flow around n
the main wing and vertical fin of the canard configuration.
r,
, I
lJ
Ii
; I
',)
n
n
1". .1
n
"J
n
n
3-44
TIS Integrated Training System
Module 8.3 Theory of Flight
Use and/or disclosure is
governed by the statement
n
© Copyright 2010 on page 2 of this chapter.
o Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
o Flight Controls
n
U Introduction
C The purpose of flight controls is to enable the aircraft to be rotated about its three axes. Control
in pitch is exercised by elevators which move the aircraft about its lateral axis, control in roll by
the ailerons which move the aircraft about its longitudinal axis and control in yaw by the rudder
n
u
which moves the aircraft: about it's normal, or vertical, axis. Controls usually take the form of
hinged aerofoils mounted on the trailing edge of the wing, the horizontal stabilizer or tailplane
and the rudder normally attached to the trailing edge of the fin. When they are moved they alter
the effective camber of the section to which they are attached and therefore alter the amount of
lift being generated. Within reason, controls are positioned as far away as possible from the axis
of rotation about which they are effective, so that they create the largest moment for the least
amount of force.
When a control surface is deflected the forces acting on it try to return it to the neutral position.
The total returning force is the lift force on the control surface multiplied by the distance of the
centre of pressure of the control surface to the hinge. This force is called the hinge moment and
is shown in Figure 3.34.
rl
u
F
o
o , Hinge
1
Moment
rl I
LJ I
I
- ......"1t x '..
I
Hinge Moment. FX
r1
u
Figure 3.34 - Hinge moment of a flight control surface
Obviously this hinge moment has to be opposed by some force if the control is to remain
deflected, and this force is supplied by the pilot through the control column or rudder bar. The
degree of lift force generated by a control surface will depend on the square of the speed, and
as the speed increases it can reach considerable magnitude. Because of this, on any but very
low speed aircraft the amount of control force required will be far too high for easy operation of
the controls without some form of assistance. This assistance is called aerodynamic control
balancing. Various forms of aerodynamic balancing are used, and they all operate on the
principle of either reducing the hinge moment or producing a force which will help to balance the
rl hinge moment by acting in the opposite direction. The various types of aerodynamic balancing
U used on current aircraft are discussed in the following paragraphs:
o
o Use and/or disclosure is
govemed by the statement
on page 2 of this chapter.
Module 8.3 Theory of Flight 3-45
TIS Integrated Training System
© Copyright 2010
L. j
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
Aerodynamic Balancing n
Inset Hinge
This type of aerodynamic balancing is commonly used on modern aircraft and achieves its
reduction of control column loading by positioning the hinge so that part of the control surface
leading edge moves in the opposite direction to the remainder of the control surface. Figure
3.35 shows an example of this type of aerodynamic balance.
l~~\~
n
U
.l '+
_________ Reduced ,,-.,
I I i, ,J
f I
--;x!.- n
I I
Inset Hinge U
Care must be taken in the design of this type of balance to ensure that the centre of pressure is
not too near the hinge line. When a control surface is deflected its centre of pressure moves
forward, and if the margin between the centre of pressure and the hinge line is too small it is "I
t
u
possible that the centre of pressure will move forward of the hinge line and so lead to the
surface overbalancing. iI
:d I
Horn Balance
In this type of system a portion of the control surface itself acts ahead of the hinge line, so n
producing a force in opposition to the hinge moment. Such a balance is shown in Figure 3.36. lJ
'I
LJ
n
Horn Balance
o
Figure 3.36 - Horn Balance n
U
3-46
TTS Integrated Training System
Module 8.3 Theory of Flight
Use andfor disclosure is
governed by the statement
o
© Copyright 2010 on page 2 of this chapter.
Integrated Training System
o Designed in association with the
club66pro.co.uk question practice aid
o Balance Tab
This type of system has the same effect as the horn balance but is produced by attaching a
small aerofoil to the trailing edge of the control surface and is arranged so that when the control
ri surface is moved the balance tab is automatically moved in the opposite direction mechanically.
U
o [ l~_o.--"Ih
o Figure 3.37 - Balance Tab
The balance tab, although quite small, is acting at a considerable distance from the hinge line of
o the main control surface, and therefore produces a considerable assistance in moving it. There
is some loss of overall effect of the control in this system and also a small drag penalty is
incurred. It should be noted that when the main control surface is deflected the tab moves in the
opposite direction, but, the chord line of the tab remains parallel to the chordline of the fixed
surface as shown in Figure 3.37. This balanced motion achieves the required objectives whilst
avoiding any excessive drag penalties.
o
r,
Servo Tab
This type of tab is used on larger subsonic aircraft as an aid to the pilot in overcoming heavy
I I control loads. With this type of system when the control column is moved the tab is moved by
U the control input which then causes the airflow to be deflected which in tum moves the control
surface. The system is shown in Figure 3.38.
rUl
Servo tab
o Control surface up
Control rod
down
o
o
o -- -- --
o
o From pilot's control
Figure 3.38 - Servo Tab
fl
LJ
n
:I
l J
o
Figure 3.39 - Anti-Balance Tab r""\
:I ..I
Internal Balance
Although fairly commonly used this form of aerodynamic balance is not obvious because it is
contained within the contour of the parent control surface. When the control is moved, a
pressure difference is generated between upper and lower surfaces. This difference will try to
n
deflect the beak ahead of the hinge-line on the control producing a partial balancing moment.
The effectiveness is controlled in some cases by venting air pressure above and below the
beak.
Hinge line
n
:... J
Flexible Seal
n
n
I. I
Beak n
Figure 3.40 - Internal Balance rJ
o Control $urfa~
o
o Structure
.-J.- /nteI'ConnI9ctilr19 valve
o
Iu'"'i linkage to pilot's cOl1trol
o Return
PreSSIJre
Return
o Mass Balance
Figure 3.41 - A simple power operated control system
o Problems can arise with oscillatory movements of the control surface caused by variation in the
moment of the control surface centre of gravity about the hinge. This variation can be brought
about through the flexing of the entire structure when a load is applied to it. It is possible under
o some circumstances for these oscillations to be divergent and cause complete failure of the
structure.
o This form of oscillation is called 'Flutter', and as the main factor involved is the moment of the
centre of gravity about the control surface hinge, the possibility of it being reduced by moving
the centre of gravity nearer to the hinge line must be considered. This is usually achieved by
o adding weights to the control surface so positioned that they act in front of the hinge line and
thereby move the control surface centre of gravity to, or just in front of, the hinge line, as shown
in Figure 3.42. On modern aircraft the mass balance weights are normally housed inside the
o
11u
Hinge
o
o AuXiliary weight
Weight of aileron
Figure 3.42 - Typical position of mass balance
Ailerons
control the aircraft about the lateral axis - that is in pitch.
when raised. This creates the additional drag to balance the increased drag of the down-going
aileron. Frise ailerons are not in general use today as, in particular, they are most unsuited to
high speed aircraft. 0
n
u
3-50
TIS Integrated Training System
Module 8.3 Theory of Flight
Use andlor disclosure is
governed by the statement
n
© Copyright 201 0 on page 2 of this chapter.
n
. J
r 1
U
Integrated Training System
o Designed in association with the
club66pro.co.uk question practice aid
o pressure as does the wing, and depending on the vertical distance between the fin centre of
pressure and the centre of gravity on the longitudinal axis of the aircraft, there will be a
tendency for the aircraft to roll to the left in this case.
o A tall fin will clearly produce a greater tendency to roll than a short one, (Figure 3.44 (b».
o The phrase 'tendency to roll' is emphasized, in view of the fact that the roll is normally totally
masked by the extra lift from the faster moving wing. Clearly this masking effect will depend
upon the span of the aircraft; a greater span producing more lift, (Figure 3.44 (c».
[1
~
If, on certain aircraft (e.g. Britannia and Belfast) there is a noticeable roll due to yaw, this can be
eliminated by interlinkage of ailerons and rudder. It is therefore, most unlikely that any pilot
c would ever notice any adverse roll when yawing the aircraft.
(a) (c)
o L 1
o --~
r'
U
o Figure 3.44 - Adverse roll tendency due to rudder deflection
o ailerons, and when an aileron is deflected downwards the resultant force on the aileron passes
through its centre of pressure as illustrated in Figure 3.45.
11u
[1u
'i
\ I
U
'l
U Use andfor disclosure is
governed by the statement
Module 8.3 Theory of Flight 3-51
ITS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
I'1.
t.J
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
Reduction in CL
Due to Wing Twist
Wing ""..,.-
--r---_
--.,,/'
"- --
,..------
Twist
,...-
These days, the idea of the all-moving tail plane has been adopted for use on low speed aircraft
as well. The additional force that it generates enables a smaller horizontal stabilizer to be used
with a consequent reduction in weight and drag.
"
U
n
I ,
U
3-52
TTS Integrated Training System
Module 8.3 Theory of Flight
Use and/or disclosure is
governed by the statement
n
© Copyright 201 0 on page 2 of this chapter,
n
[1
I !
'--'
u Spoilers
Mention was made earlier of a type of control called a spoiler. As shown in Figure 3.46, spoilers
are panels in the upper surfaces of the wing that are hinged at their leading edges and can be
,,-1, opened and shut so that, when open they reduce the amount of lift being generated by the wing.
i I
U
rl
I .
LJ SPOiler.~
o C - ~
o
o Spoilers have three normal uses:
Figure 3.46 - Spoilers
r-IL.J
• When operated through small angles independently on one wing or the other they can be
used to augment aileron control, or even replace it.
D • On landing. After the aircraft has actually contacted the runway, if the spoilers on both
wings are fully opened the lift is completely destroyed; this is called 'lift dumping'. It
enables the aircraft to sit more firmly on the ground and thus reduce the distance
o •
required to stop and it is also of considerable value in crosswind landings.
Lift dumping in high level flight.
o Airbrakes
Modern jet transport aircraft have extremely low drag coefficients in the cruising configuration.
o This makes them somewhat reluctant to slow down from high speeds, and on some aircraft
airbrakes are fitted.
o An airbrake usually consists of a flat section which can be raised into the airflow so that it
creates the maximum amount of drag whilst at the same time not damaging the lift. Airbrakes
are not all that common however, and resort usually has to be made to other means of slowing
c down rapidly when required. One of the most commonly used methods is the lowering of the
main gear to create additional drag, but on some aircraft reverse thrust from the engines may
be selected in the air as well as on the ground.
u combined into one control called an elevon. When the control column is moved backwards or
forwards both surfaces move up and down together, but when the control column is moved
sideways one elevon comes up and the other goes down. Another combination of controls is
o that of elevator and rudder to produce a Vor butterfly tail. This is then called a ruddervator.
o
o Use and/or disclosure is
govemed by the statement
Module 8.3 Theory of Flight 3-53
TIS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
Integrated Training System
Designed in association with the
c!ub66pro.co.uk question practice aid
,:-l,
l J
Single
Rudder
n
Span Tab n
Spoilers
(7 Per Side) Elevator
lJ
'I!=~~~~~;;.(Single Span)
Slats
Stabilizer
n
~= __-------
~--Double
Flap
Flaperon
Slotted
, n
~- Flap
n
n
Figure 3.47 - An example aircraft control surface configuration
Aileron operation is confined to the lower airspeed envelope and at high speed the ailerons are
n
locked out of action and lateral (Roll) control is a function of the spoilers and flaperons. This
later function is to reduce wing flexure at high speed.
Some aircraft employ two sets of ailerons, one outboard for lower speeds and the other inboard
for high speed. In such cases the outboard ailerons are locked out at high speeds.
n
n
n
o
o Tabs
o Introduction
Tabs are small aerofoil sections hinged to the trailing edges of control surfaces. The main
purposes for which they are used are:
11
U
• Trimming
u •
•
Aerodynamic balancing
Servo operation
Trim Tabs
For aircraft in flight to be in equilibrium, the moments about each of the three axes of the aircraft
o must balance. If they do not balance then an additional force must be supplied by deflection of
the controls to keep the aircraft in equilibrium. It is most desirable that continuous control
surface deflection be applied at the control column because of the physical effort involved, and
o to overcome this problem trimming tabs are provided. The action of a trim tab is best
understood by considering the situation with an aircraft which tends to fly nose down
continuously. To correct this the elevator must be deflected upwards and maintained in this
u position.
o Tab Neutral
ii
LJ Fl
o I
I
~D-"
o B cp
o To bring this about, the trim tab attached to the trailing edge of the elevator is deflected
downwards as shown in Figure 3.49. This diagram indicates that the total force exerted
o downwards by the elevator is F1 x d. the distance of the elevator's centre of pressure from its
hinge, A. The trim tab, having been moved in the opposite direction, exerts a force F2 upwards
through its centre of pressure, and although this force is smaller than F1 it has a much longer D
arm from the hinge of the main control, B. therefore its total moment is able to balance out that
of the main control. It should be noted that the action of the trim tab also slightly reduces the
effect of the main control surface. The final force exerted is F1 - F2 and this will necessitate a
slightly larger deflection of the main control surface with a subsequent small increase in drag.
n
Trimming tabs are normally controlled either by trim wheels in the cockpit or, alternatively, by
electrical switches activating motors. These controls are usually arranged so that they act in a
n
natural sense, that is to say with the control wheel moved in the direction in which the aircraft is
required to be trimmed, i.e. aircraft nose high, move the top of control wheel forward towards
n
the nose to bring it down. On some light aircraft fixed trimming tabs may be fitted and they n
consist of small sheets of metal which can be bent permanently to correct known out-of-trim lJ
forces, but they cannot, of course, be adjusted in flight. As with any other aerodynamic control
surface the effect of a trim tab is proportional to the square of the speed. At high speeds very
small trim adjustments will achieve the desired effect whereas at low speeds a considerable
movement may be required.
n
n
~h:~tm-:=::t;::;:::~::iS$ D
_
To Control Column
Fixed Tab.
Fixed Tab.
Adjustable
on the Ground -
To Control Column n
Trim Tab
3-56
TIS Integrated Training System
Module 8.3 Theory of Flight
Use and/or disclosure is
n
n
govemed by the statement
© Copyright 2010 on page 2 of this chapter.
u
Integrated Training System
o Designed in association with the
club66pro.co.uk question practice aid
o
o =-
..~.'...."'.....,
==>:,'"- . . -.... -.--.-:.; .......
............... ...... "
,~
"'- -:~.p", '" ...
o Balance Tab
Figure 3.51 - Balance and Anti-balance tabs
Anti-Balance Tab
o Spring Tab
On some aircraft, aerodynamic balance is not considered necessary at low speeds but is
o required at higher speeds when the aerodynamic loads increase considerably. A type of
balance tab to deal with this situation is called a spring tab. The basis of the system is illustrated
in Figure 3.52. The movement of the control column is transmitted to a lever pivoted on the
o main control surface but not directly operating it. Operation of the surface is through springs,
and at low aerodynamic loads the movement of this pivot arm is transmitted to the main control
surface through the springs, and no alteration in the geometry between the pivot and balance
D tab takes place. When the aerodynamic loads increase, however, transmission of control
column movement, via the pivot arm, to the control surface compresses the spring. This upsets
o the geometry of the system and brings into operation the balance tab on the trailing edge, thus
giving some assistance in moving the control surface.
o Control
Input
Control
Input
o Free
. to Pivot
~--;,-
u The Spring Tab System The Servo-Tab System
When manual controls are used to operate very large control surfaces the loads involved, even
o with the assistance of balance tabs, may be unacceptable. Under these circumstances servo
tabs are used to operate the control surfaces. A servo tab is a small aerofoil section, once again
attached to the trailing edge of the main control surface, which is directly operated by the control
column. In this system there is no direct connection between the control column and the main
lJ control surfaces, the control column only operating the servo tabs. As the system depends
entirely upon aerodynamic forces, any movement of the control column when the aircraft is on
the ground will produce no control surface movement, only the servo tabs will move. This
D system is illustrated in Figure 3.52. Care must be exercised in pre-flight checks for full and free
movement of control surfaces when servo tabs are used.
o
lJ Use and/or disclosure Is
govemed by the statement
Module 8.3 Theory of Flight 3-57
TIS Integrated Training System
on page 2 of thIs chapter. © Copyright 201 0
rl
lJ
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
o
Variable Incidence Tailplane
r
.. J
In the section on flight controls, mention was made of the variable incidence horizontal
stabilizer. In this system the incidence of the entire horizontal stabilizer is changed, as required,
to adjust for longitudinal trim requirements instead of using a conventional trim tab. The
horizontal stabilizer is pivoted about its central point and moved by electro-hydraulic jacks. The
o
change in longitudinal trim which this system can exert, compared to an ordinary trimming tab,
is very considerable, and in view of this precautions have to be taken against the result of a
runaway in the operating system, winding the horizontal stabilizer either fully up or fully down
This is usually achieved by having the left and right halves of the horizontal stabilizer entirely
separate from each other and each being operated by two separated systems. This will obviate
the possibility of both sides of the stabilizer running away together and also produces a multiple
redundancy in the event of partial failure of one of the electro-hydraulic systems. This form of
longitudinal trimming is a necessity on large jet aircraft where the conventional system of
elevator and trim is unable to cope with the large trim changes brought about by, firstly, large
centre of gravity movements; secondly, wide operating speed range and. finally, high lift n
L.J
devices. It will be remembered that as the elevator is not deflected in any way for the purposes
of trimming, the full range of control is always available from it.
Control locks
n
All aircraft require some mechanical means of locking the controls in the neutral position when
on the ground to prevent possible damage from gusts in high winds when parked. On smaller
aircraft these may take the form of wood or metal devices which can be slid over the control
o
surfaces to lock them in position and prevent their movement. On larger aircraft it is common
practice to have internal locking systems, but obviously either type of control lock must be
removed prior to take-off. In this context it should be borne in mind that external locks fitted to a
n
servo tab operating system will permit the movement of the control column on the flight deck
with the locks in place as only the tabs are being operated, therefore, on aircraft with this type of
equipment it is essential to confirm that any external locks fitted have been removed. Most
n
modern aircraft with hydraulically operated controls have integral hydraulic locks.
n
n
l.J
o
n
3-58 Module 8.3 Theory of Flight
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 201 0 on page 2 of this chapter. n
!. J
r I
LJ
rI Integrated Training System
LJ Designed in association with the
ctub66pro.co.uk question practice aid
u Flight Forces
u Introduction
r I The four forces acting in level flight are lift, weight, drag and thrust. The lift acts through the
U centre of pressure. The weight is through the centre of gravity. The drag and thrust act along
lines parallel to the longitudinal axis and this is illustrated in Figure 3.53.
I I
U
LIFT
u L
u !l
THRUST T ... ~----~::::~~:t~~~~-::t-----~ D DRAG
r
u
i ............ t:t7 =-
u W
WEIGHT
Lift = Weight
fI Thrust = Drag
lJ
11 Pitching Moments
LJ The positions of the CP and CG vary throughout flight, and under most conditions are not
coincident, CP varying with angle of attack and CG varying as fuel is used. The result is that the
11 opposing forces (Lift and Weight) set up a couple causing either a nose-up pitch, or a nose-
u down pitch, depending on the relative positions of CP and CG. This is illustrated in the two
diagrams of Figure 3.54.
u
u
u Use and/or disclosure is
governed by the statement
Module 8.3 Theory of Flight 3-59
TIS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
ri
u
Integrated Training System ,...,
Designed in association with the
club66pro.co.uk question practice aid lJ
L
o
o
NOSE DOWN
PITCHING
MOMENT
( DOWNLOAD
rl.I
GNV
o
L
n
o
NOSE UP
PITCHING MOMENT
( [J
n
lJ
11
GNV U
Figure 3.54 - Pitching moments due to the Lift-Weight couple
n
L. J
Ideally, the pitching moments arising from the Thrust and Drag couples should neutralize each
other in level flight, but the ideal is difficult to attain and, as far as possible, the forces are
arranged as in the first diagram. With this arrangement, the TID couple causes a nose-up
moment, and the LIW couple a nose-down moment, the lines of action of each couple, being
n
such that the strength of each couple is equal. If, now, the engine is throttled back, the TID
couple is weakened, and the LIW couple pitches the nose down. o
The tailplane andlor elevator has a stabilizing function in that it supplies the force necessary to
counter any residual pitching moments. If any nose-up or nose-down pitch occurs, the elevator
deflection can be altered to provide an up or down load to trim the aircraft. This is also shown in
Figure 3.54.
If the elevator has to produce a down load balancing force, this effectively increases the aircraft
weight. So to maintain level flight at the same speed, the angle of attack must be increased to
maintain lift. The increase in drag is known as trim drag.
[]
3-60
ITS Integrated Training System
Module 8.3 Theory of Flight
Use and/or disclosure is
governed by the statement
o
n
© Copyright 2010 on page 2 of this chapter.
o
Integrated Training System
Designed in association with the
club6Gpro.co.uk question practice aid
The relationship between Angle of Attack, lAS and Altitude assuming that in level flight lift
equals weight then:
Lift = Weight = % pV2 S CL
=
p density
V=TAS
S = wing area
CL = a constant coefficient of lift
As for a given weight lift will be constant, then the equation must also be constant. The only
variables in the equation are density, TAS and angle of attack. It must be remembered that the
expression % pV2 is dynamic pressure or lAS. In view of this, for a fixed lAS and weight the
angle of attack will be constant for any altitude.
Looking at it from a different point of view, if lAS (% pV2 S) is increased then, to keep the
o equation balanced, the angle of attack must be decreased and vice versa. To summarize:
For optimum aerodynamic efficiency, the maximum amount of lift will be produced for the least
o amount of drag. This, of course, means flying at the maximum lifUdrag ratio which has already
been shown to occur at a fixed angle of attack, usually around 4°. It was also shown that for a
given weight this will represent a fixed indicated airspeed, regardless of height. If however, the
r~
weight decreases due to use of fuel, then it will be necessary to decrease the indicated airspeed
I
u
I to maintain the same angle of attack.
Ii
'-.J Climbing, Gliding and Turning
o Climbing
During a climb, an aircraft gains potential energy by virtue of elevation, achieved by one or a
combination of two means, viz:
o less than that required in straight and level flight, W cos y, and secondly, the thrust has to be
equal and opposite to the sum of drag and weight components along the flight path
=
T D + W sin y. This is shown in Figure 3.55. It is still considered sufficiently correct to assume
=
L D up to about 15° climb angle (cos 15° = 0.9659, i.e. the error is less than 2%).
D
n
L.J Use and/or disclosure is
governed by the statement
Module 8.3 Theory of Flight 3-61
TIS Integrated Training System
~
on page 2 of this chapter. © Copyright 2010
) I
Integrated Training System
Designed in association with the
c!ub66pro.co.uk question practice aid
la)
[J
WSin~ \
\
WEIGHT COMPONENT
o
~ PERPENDICULAR TO
FLIGHT PATH.
~ ANGLE OF
CLIMB
(b)
. Thrust - Drag
Diagram (b) SIn y = -..,...,-,:-:--:-:---"'.
Weight
Weight
Therefore Vc
V
= Thrust - Drag
Weight n
Therefore Vc
= V (Thrust - Drag)
Weight n
= _P_o_w_e_r....:(_av_a_i_la_b_le..:..)_-_P_o_w_e_r_(,-re_q-,-u_ir_e-'..d)
Weight
o
Excess Power
,l =
Weight
LJ
Pt-Pd
or where Pt = Thrust Horsepower
W
Pd =Drag Horsepower
Angle of Climb
From Figure 3.55 (b) above it can be seen that for the maximum angle of climb, where
. Thrust - Drag
SIn y = -...,....,..,.....,.._----=.
Weight
the aircraft should be flown at a speed which gives the maximum difference between Thrust and
i'
=
Drag. Alternatively, if climb angle 0, i.e. level flight, then
L.J
Thrust - Drag = 0
fl
....J
Weight
or
Thrust - Drag = 1
Weight
Ii
w
So, it can be deduced, the factor controlling the angle of climb will be the excess of thrust over
drag.
C Power Available and Power Required
The thrust power curve for a piston engine differs from that of a jet engine, as shown in Figure
D 3.56. The main reason for this is that the thrust of a jet remains virtually constant at a given
altitude, regardless of speed, whereas the piston engine, under the same set of circumstances
C" 1 and for a given BHP, suffers a loss at both ends of its speed range because of reduced
U propeller efficiency.
c
Ilu Use and/or disclosure is
Module 8.3 Theory of Flight 3-63
govemed by the statement TTS Integrated Training System
on page 2 of this chapter. © Copyright 2010
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
The curve depicted in Figure 3.56 can be assumed to apply equally to a piston or a jet propelled
aircraft, i.e. the airframe drag is the same regardless of power and speed. The increase in
power required at the lowest speed is caused by rapidly rising effects of induced drag. n
,. , i.J
/,I' Y
, I
I
POWER AVAILABLE I
I
JET I
I
I
BEST CLIMBING
SPEED ,,
I
,
o
JET
I
I
I
POWER AVAILABLE I
PISTON
,'""?_--- POWER
THRUST
V READ
(DRAG xTAS)
HORSE /'
,/
POWER /
/
./
./"
BEST CLIMBING
SPEED
PISTON
n
U
.-'
'~_.J
!
n
L~
n
Integrated Training System
o Designed in association with the
club66pro.co.uk question practice aid
[: Climb Performance
u The best climbing speed (highest rate of climb), is that at which the excess power is maximum,
so that after some power is used in overcoming drag, the maximum amount of power is
~i
\ : available for climbing. The vertical distance between power available and power required
cJ represents the power available for climbing at that speed. Note, in Figure 3.57, that this speed
for the piston engine is approximately 175 Kt (TAS), and for the jet approximately 400 Kt. In the
latter, there appears to be a fairly wide band of speeds which would still give the same excess
power; in practice the higher speed is used in the interest of engine efficiency. At points X and Y
all available power is being used to overcome drag, therefore these points are the V min and V max
for the particular power setting.
Power
PISTON Reqd. Power Reqd.
M.S.L. 40,000 ft
THRUST
HORSE
POWER
o
o
Module 8.3 Theory of Flight 3-65
Use andfor disclosure is
governed by the statement TTS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
i I
u
Power
Avail. M.S.L. .
,,.'
.....
I
,/'
I
;'
/'
;' 40.000ft.
n
THRUST
HORSE
POWER Power
,
n
Power
Reqd.
M.S.L
Reqd.
40.000 It ..........".
"...... ---
."..
n
]
TAS M.S.L.
100 200 300 400 500 EAS. 40.000 ft
I I i
100 200 300
When the aircraft is placed in a nose down attitude, as in a descent, the component of weight in
the direction of the flight path augments the thrust, the aircraft will accelerate, lift and drag will "I
change, so, in order to achieve a balanced condition with a constant airspeed, thrust must be
reduced. From the foregoing, it may be seen that the controlling factor of the glide angle is the
lJ
lift/drag ratio of the aircraft. An increase in weight will not affect glide angle as all components
will expand by the same proportion, but an increase in weight will increase speed along flight
path.
o
n
n
,i j
L=Wcos ~
D=WSin ~
L TO-TAL
r 1:
\
/ " REACTION
~ I
I
~ :,
I
\
\ HT
H\ WCoSH
V Sin H \
!t
DISTANCE ~~ W Sin H
,[1'
-(a)
'-f
(b) 'W.
o D =Max =W Cos y
n
I ,
w
y = Min =W sin y
Therefore
W Cosy
Max
W Siny
but
n
W Cosy =~
U W Siny D
The best angle of glide therefore depends on maintaining an angle of attack which gives the
best Lift/Drag ratio, or for maximum endurance the aircraft should be flown for minimum drag.
Effect of Wind
Gliding for minimum rate of descent, or for endurance, is unaffected by wind, because the
position at the end of glide is unimportant. But when gliding for range, the target is the point of
arrival, thus the aim is maximum distance over the ground.
il
u Use and/or disclosure is
Module 8.3 Theory of Flight 3-67
govemed by the statement TTS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
\J
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
Gliding for range is achieved as explained previously, i.e. by flying for minimum drag. However,
that applies only in still air conditions. The effect of a headwind will be to decrease ground
distance by approximately the ratio of n
• j
WS
TAS
An increase of airspeed could reduce the time the wind effect would act, and thus improve
ground distance. Similarly, if there were a tailwind, ground distance would be increased, a
reduction of speed would improve the distance, since the wind effect time would be increased.
Effect of Weight
Variation in weight will not affect glide angle, provided speed is adjusted to suit the all-up-weight
(AUW). A simple method of estimating speed changes, to compensate for weight changes (up
to about 20%), is to adjust speed (EAS) by half the percentage change in AUW. e.g. a weight
n
:..J
reduction of 10% would necessitate a decrease in speed of 5%.
:t
Although range is unaffected by weight, glide endurance decreases with weight increase. lJ
Penetration Speed is the optimum gliding speed for any wind speed.
Turning
During a turn weight still acts vertically downwards but a second force, centrifugal force, occurs
from the aircraft traveling along a curved path. This centrifugal force has to be opposed by a
centripetal force which can only be obtained as a resolved part of the lift force. Because the lift
also has to balance the weight in addition to the centripetal force, it is evident that in turn the lift
has to be increased to a greater value than weight. This is illustrated in Figure 3.60.
fl,
,.J
LIFT
....,
Centrifugal Force
CENTRIPETAL
• U
! I
FORCE
i"
'!,L j'
WEIGHT
If the aircraft is banked, with the angle of attack constant, the vertical component of lift will be
too small to balance the weight, thus the aircraft will descend. Therefore, as angle of bank
:J
increases, angle of attack must be increased, the vertical component is then sufficient to II
"I
I...'
3-68
TIS Integrated Training System
Module 8.3 Theory of Flight
Use and/or disclosure is
governed by the statement
o
© Copyright 201 0 on page 2 of this chapter.
o
r1 Integrated Training System
Designed in association with the
LJ club66pro.co.uk question practice aid
o maintain level flight, whilst the horizontal component is sufficient to produce the required
centripetal force.
Effect of Weight
If the lAS in a turn is maintained at a constant figure the increased lift can only be obtained by
an increase in the angle of attack. The increase in lift will, of course, produce more induced drag
which will require an increase in thrust. As the angle of attack has been increased the wing is
nearer to its stalling angle, therefore the stalling speed will be increased. The increase in the
o value of lift is, in fact, equivalent to increasing the aircraft's weight. The amount by which this is
apparently increased is called the load factor or 'n'. For instance, if the weight is apparently
doubled 'n' becomes two and this is called a 2g turn. The increase in stalling speed associated
with the load factor may be calculated from the following formula:
For example an aircraft with a normal stalling speed of 100 kts carrying out a 2g turn would
have its stalling speed increased by 100 x J2 = 140 kts approximately.
In a steady level turn, thrust being ignored, lift is providing both a force to balance weight, and a
rl
centripetal force to turn the aircraft. If the same TAS and angle of bank can be sustained, the
,\ i. turn radius is basically independent of weight or aircraft type.
C-i
U Altitude
With increase in altitude, there is an increase in the minimum radius, mainly due to the
EASfTAS relationship. An additional increase is caused by the reduction in CLmax , because
Mach No is higher at altitude for a given TAS.
r: An increase in altitude will cause the rate of turn to decrease.
I i
LJ
Effect of Thrust
Even in level flight, it can be seen that some aircraft have their thrust line inclined to the
horizontal, thus reducing a component of thrust augmenting lift. In the minimum radius turn, and
maximum rate turns discussed, the aircraft is flown for CL max which is obtained at the critical
angle, the thrust component assists lift, so either less lift is required from the wing, or the turn
can be improved. However, the reduction of thrust with increasing altitude will cause a reduction il
{J
\
in turning performance, in addition to that caused by the EASfTAS relationship and the greater
CL max red uction.
Effect of Flap
n
Lowering of flap produces more lift, also more drag at any given EAS. A smaller radius of turn
may thus be achieved with flap, providing the flap limiting speed is not a critical factor, and the
available power is sufficient to overcome the extra drag.
Although the object on the string is following a curved path of motion, it is continually trying to
obey the first law of motion, i.e. to continue traveling in a straight line ... true or false? True -
should the string be released, centripetal force is removed and the opposite reaction (centrifugal
force) disappears simultaneously. In this instance, the object at once obeys the first law of
motion and flies off in a straight line at a tangent to its previous circular path.
It is important to realize that, without centripetal force, no object whether a car or aircraft can be
made to turn, and the first law of motion applies. n
lJ
Centripetal force during a given turn is directly proportional to the mass of the body, the square
of its speed and is inversely proportional to the radius of the turn. It is calculated from the
formula:
Centripetal force =
W V2
g r
(in Ib) n
Or
mV 2
Centripetal force = - - (in Newtons)
r
n
Where: lJ
c W =
the weightlor m is the mass
V2 = the square of the TAS in feetlsec or m/sec
r = the radius in feet or metres
g = the gravitational force of 9,81 m/s2 (or 32.2 fUsec2)
To calculate the acceleration towards the centre, the following formula applies:
Ti
, '
LJ
Acceleration
r
r"'
, I
LJ
Where V2 = the velocity in feetlsec or metres/sec
r = the radius in feet or metres
Turning
For an aircraft to turn, centripetal force is required, This centripetal force is derived by resolving
the inclined total lift force into two components, namely:
[Iu
• Vertical lift component
• Horizontal lift component
Thus, it is the horizontal lift component that provides the centripetal force required to pull the
n aircraft towards the centre of the turn as it moves along a path of circular motion, (Refer to
U Figure 3,61). However, during a turn, lift has a double role to play. Not only is it resolved into a
horizontal component to provide centripetal force, but also has to provide a lifting force such
[J that the aircraft maintains a constant height during the turn, It will be seen from the Figure 3.61
that any inclination of total lift from the vertical results in a smaller vertical component of lift,
which would then be too small to balance the weight of the aircraft.
fi
U
11
LJ
{
r,,
LJ
f'
I I
W Use and/or disclosure is
Module 8.3 Theory of Flight 3-71
governed by the statement TIS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
TOTAL
LIFT LIFT
.,.-----,
I (I
I
I
I U
I
I
I
I
I
I
I
I
I 8
I
I
I
I 'i
CENTRIPETAL l...ooi;-.---_\:
FORCE ""'<1''''''''':;'-- :\
, J
,..,
;I
u
n
, I
l..)
WEIGHT
Figure 3.61 - Total Lift is the vector sum of Lift and Centripetal Force
Therefore, to prevent the aircraft from descending, the angle of attack must be progressively
increased to maintain a greater total lift. Once this has been accomplished, the vertical
11
component of lift is large enough to maintain level flight, while the horizontal component is large '1
enough to generate the required centripetal force. However, an increase in angle of attack :~ J
results in an increase in drag, which must be balanced by an increase in power if the speed is to
remain constant. n
l J
Steep Turns
A steep turn is classified as a turn having an angle of bank in excess of 45°. Larger bank angles
require a larger angle of attack to produce the required total lift increment. However, the penalty
of large angles of attack is drag. Eventually, the aircraft will reach a speed so low, that any
further increase in angle of attack will result in a stall. At this instant, angle of attack and induced
drag are so high that full power is usually necessary to keep the speed constant. '1
~_, i
,,\
U
II
n
3-72 Module 8.3 Theory of Flight
]
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter.
II
'. )
Integrated Training System
n
Li
Designed in association with the
club66pro.co.uk question practice aid
r-,
I '
I !
LIFT INCREASE REQUIRED
~
---------------
~
; ----------:;~ L
,t
I
,/
,/
...-
I
,
I ...-
c I
I
/
./
o
jl
u
w
Figure 3.62 - Aircraft in a steep turn
c
r:u
n
\
~
:
LJ
n
r,
U
M 'of Flight 3-73
Use and/or disclosure is
govemed by the statement TIS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
LJ
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid n
Wing Loading
Wing loading is the weight of the aircraft divided by the wing area. n
Since L = Wand L = CL Y, pV2 S, then W = CL Y, pV2 S
Thus wing loading (i.e. the weight carried by a wing of given area) affects both the maximum
and minimum stalling speeds.
n
However, modern tendency is to increase the wing loading by decreasing the wing area and
increasing the speed, and to use flap to reduce landing speeds.
n
Load Factor
The load factor of a given aircraft in a given condition of flight is defined as the lift divided by the
n
,~ \
L.J
weight. It is denoted by n.
In straight and level flight, L = W; therefore n = 1. In any manoeuvre where lift is greater or
smaller than weight, L = nW.
n
u
~
: I
In any manoeuvre, the stalling speed is proportional to the square root of the load factor (Vn) U
(Limit load factor).
11500 X 338 2 r,
so: = - - - - -
32.2x2000 · I
'~_ I
Refer to Figure 3.63 for the wing loading calculation. The wing loading which is equal to lift may
,...,
be calculated by Pythagoras' Theorem, where:
•I
,", .. J
TL2 =F2+ L2 l
!J
TL = .J20400 2 + 11500 2
TL =23418 Ib
-,
, I
U
n
3-74 Module 8.3 Theory of Flight !J
Use andlor disclosure is
TIS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter,
n
~ I
· )
r-·"'.
II i.
L.!
However, should one wing stall before the other, the aircraft would tend to roll in or out of the
,-.- .....
turn, due to unequal lift distribution. This roll may in some, cases result in a 'flicking movement'.
)i
u Recovery is initiated by moving the control column forward and, in doing so, decreasing the
angle of attack, thereby unstalling the aircraft. This occurrence is known as a 'high speed' or
'accelerated' stall.
T.L
b
~---------
I
C'l
i i
'-.i
,.,
!••
I
"" 11500
HCL
J
I.CL~
20400
____ _______ ..J-~re'J'/
,r· ....,,
I :
IoJ'
n
I'-.I:
o WI
2
W
tan 0 = -L- - 2
=V
HCL WV /gr gr
---------------------------- n
~
'
I .
I
I
I
I
I
I
I
,,
I
I,
I
,
I
I
,
I
I
,
'li..Jl.2
9 r
o
n
Figure 3.64 - Formula for centripetal force
w
o
A vertical banked turn is impossible because even if Total Lift becomes infinity no vertical
component can be obtained. However, even with a vertical bank there is a limit to the radius of
the turn because (apart from side slipping), the wings must provide all the horizontal force (Le.
ClY, V2S), represented by the formula:
or r = 2W/(ClP S g) (iv)
(v) nI
.
I
" .J'
n
,,
r'
,
I
,
'
U
Furthermore, it is true to say that the load factor varies as the secant (co:ine) of the bank
angle.
c raises the stalling speed. This is true of any turn and the stalling speed may be calculated from
the formula:
From the foregoing example, let us assume the aircraft had a basic stalling speed of 85 knots at
n gross weight. The new stalling speed during the turn is therefore:
U
rn
o New V = old Vs x
= 85 x.J2
n
I: Module 8.3 Theory of Flight 3-77
L.J Use and/or disclosure is
governed by the statement TTS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
During a turn having a bank angle of 45°, the stalling speed is 100 knots. Calculate the
basic stalling speed.
n
Answer: 84 knots
2 Assume an aircraft weighing 11,500 pounds pulling out of a dive. If a force of 4g was
registered, what was the centripetal force and the new stalling speed if the basic stalling
speed was 78 knots?
n,
:
,----
Answer: 46,000 pounds; 156 knots.
Thus, the minimum radius of turn settled by the stalling speed of that aircraft. However, engine
n
power is the final deciding factor in settling the minimum radius.
n
'.J
n
C"">
i "
[.,)
Bank must therefore be held off during climbing turns. The opposite applies for descending
turns, where bank must be held on.
n
LJ
Ii
u ----- -- ----fJ' ......'--...-- ---__
"'- ... - A of A Outer Wing (greatest)
-----__ --... A of A inner wing
11
U
~
---"---"':=::l":_l~ . .
cc
w
Z
z
T
-------..............
'"'-- ............... - ........
... --~:. .
- Q) il Horizontal distance travelled. INNER WING -
~g=
t: !1 ~ ~.----- Horizontal distance travelled OUTER WING
<D·!!!~I-rr-~~~
>"0_
--...::~:-.....---
~
rl
u
t!)
~ ~ _-------:::1:\.~-.:.- -
- - - - ....
AofAOuterwing
n
I \
g _-ll- -----::::::.------- AotA Inner wing (greatest)
r~
u
,fl .
Modu(e 8.3 Theory of Flight 3-79
G' Use and/or disclosure is
govemed by the statement TTS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
Ll
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
n
n
!'l
d
n
U
n
: I
' __ f
Intentionally Blank
U
n
n
3-80 Module 8.3 Theory of Flight
n
'>. )
Use andlor disclosure is
ITS Integrated Training System governed by the statement
© Copyright 2010 on page 2 of this chapter.
[,
!
u
J
n
u Basic Manoeuvres
i:
LJ
Lift
The movement of the elevators UP will produce a down load on the tailplane, resulting in the
n
I ;
longitudinal axis rotating about the lateral axis to increase the angle of attack.
u
The increase in angle of attack will result in the following:
• Centre of Pressure:
Will move forward.
o • Transition Point:
Will move forward.
n
I :
L..J
• Boundary Layer:
Will become thicker.
• Separation Point:
n Will move forward
U • Stagnation Point:
Will move down and aft towards the underside of the wing.
• Induced Drag:
o •
Will increase as the angle of attack and the resultant lift increases.
• Power:
o In order to maintain level flight with increased angle of attack and maintaining
constant airspeed, then power must be increased to balance the increase in drag.
• Power Required:
It may also be said Power Available reduces and Power Required increases.
[,
i ! • Stalling:
L.J Whilst the stalling angle will remain the same, due to the weight increase the
stalling speed increases.
(1
, '
n
Whilst different cambers are used for wing sections to satisfy individual type requirements, a
number of basic principles must be appreciated.
n
Example: High Camber wing at zero angle of attack:
• Left rudder pedal pushed forward causing rudder trailing edge to move to port.
o
---,
•
•
Some part of the leading edge of the rudder will move to starboard.
This action will cause the aircraft to yaw about the normal or vertical axis to port.
i
\ ..,
I
• The airflow over the starboard wing is now at a greater velocity than that over the port
•
wing and so more lift is being generated by the starboard wing than the port wing, the
result being a roll to port.
So a yaw to port will also result in a roll to port.
n
Yaw to Port (Large Fin and Keel Surface)
• Left rudder pedal pushed forward causing rudder trailing edge to move to port and rudder
leading edge to starboard.
• Aircraft yaws to port about the normal or vertical axis.
• Action of rudder on a large fin causing a lift force of high magnitude to cause the fin to
move about the longitudinal axis towards the right in a clockwise rotational movement
when viewed from the rear, hence a roll to starboard.
• So it can be said on an aircraft with a large fin and keel surface when the aircraft is
'1,
,. I
Note: It can therefore be said that an aircraft with a normal or conventional sized fin and keel
surface is spirally stable in that when yawed to port it will readily roll to port and allow a spiral to
n
be executed in a stable manner. nI
I
\. .}
If, however, the aircraft has a large fin and keel surface, when yawed to port it will tend to roll to
starboard and will resist a spiral to port by rolling out of it and so can be said to be spirally
unstable.
o
3-82 Module 8.3 Theory of Flight
Use and/or disclosure is
ITS Integrated Training System governed by the statement
~
© Copyright 201 0 on page 2 of this chapter.
. I
I I
\ )
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
• The angle of attack must be reduced by pushing the control column forward.
• The reduction in angle of attack will result in the Centre of Pressure moving aft.
• The induced drag reducing.
• The Transition Point and Separation Point moving aft.
o
.r-,
•
•
The
The
Stagnation Point moving forward and up towards the leading edge.
Boundary Layer becoming thinner.
I '
Note: Induced drag will reduce as the square of the speed. Profile drag will increase as the
U square of the speed.
11
u Stalling Angle
It must be noted that for a given wing shape the stalling angle will remain the same regardless
Iiu of speed, weight, altitude or any other factor and can generally be regarded as being in the
order of 14° to 15°.
,
c·~
:
I"
~
Stalling Speed
Unlike the stalling angle the stalling speed is a variable quantity.
r.
I :
U
• The stalling speed will be increased if the C of P is forward of the neutral point.
• The stalling speed will be reduced if the C of P is aft of the neutral point.
• The stalling speed will be reduced if a power-on approach is made with a propeller-driven
aircraft due to an increased thrust component from the propeller and the airflow tending
Under normal conditions of flight, thrust is provided in equal proportions to provide Total Thrust
which is opposed to total drag, the two forces acting through the aircraft centreline. (Figure
3.66.)
Consider that the right-hand engine fails. Immediately, Total Thrust moves from the aircraft
r1 centreline to the thrust line of the left engine. Furthermore, the right hand propeller not only
L ceases to produce thrust but generates a considerable amount of drag until the propeller is
o feathered! With Total Thrust moving to the left and Total drag moving to the right, the opposing
forces cause a yaw towards the failed engine. (Figure 3.66 and 3.67)
11
Iwi
rl
u Use andfor disclosure is
Module 8.3 Theory of Flight 3-83
governed by the statement TTS Integrated Training System
n! on page 2 of this chapter. © Copyright 2010
,
I I
" .>
n
\. J
The yaw produces a roll in the same direction (like further effects of rudder), and the aircraft
nose will follow the down-going right wing tip into a spiral dive. It has a similar effect to putting a
boot full of right rudder in and then leaving the aircraft to sort itself out without any help from
other controls. ;1
rJ
n
.'1
l. j
'1
• I
'- j
:l
3-84 Module 8.3 Theory of Flight ·" )
Use andler disclosure is
ITS Integrated Training System governed by the statement
© Copyright 201 0 on page 2 of this chapter.
r'
I '
w
The rudder, however, like any other flying control, is only as effective as the airflow over it and
herein lies a problem. If you let the speed drop too low the rudder will lose its effectiveness and
will be incapable of combating the yawing force of the live engine, aided and abetted by drag
from the failed engine.
The minimum speed at which it is possible to maintain direction on one engine (known as
minimum control speed), cannot be quoted as a single figure for any particular aircraft as it
o varies according to circumstances. The following are the primary factors that affect it:
Altitude:
Since more power means more asymmetric thrust (and therefore yawing action) it follows that
minimum control speed will be at its highest at full throttle altitude where maximum power can
be developed.
rr
Iw ; Load:
A fully loaded aircraft must, speed for speed; fly at a higher angle of attack than when nearly
empty. A higher angle of attack means more drag and that in turn demands more power. So,
"
i '
U
back to square one; more power, more yaw, more yaw and, in consequence, a higher minimum
control speed.
Drag:
This takes us back to Point 2. More drag means more power means more yaw, etc. Drag is
mentioned here as a separate item to draw attention to the fact that flying with cooling flaps
!1
open and the landing gear extended will demand more power from the live engine - and will
U therefore mean an increase in minimum control speed.
Flaps:
Use of flap spoils the lift/drag ratio even though some flaps give very little drag increase until
(' after the first 10 to 15 degrees of depression. As a guideline it is probably best to regard the
LJ flaps as coming under the heading of -drag' and leave them up, unless the aircraft manual
specifically advises otherwise.
Windmilling:
While some of the early light twins had fixed-pitch propellers, these days constant
n
I , speed/feathering types are universal.
I :
L!
The drag from a windmilling propeller is very considerable however, and since it will provide a
great deal of 'anti-thrust' (i.e. drag). Minimum control speed will be appreciably higher than
usual until the propeller is feathered. Remember that wind milling drag is asymmetric drag -
and that is Poison.
o
o
Ii
u Module 8.3 Theory of Flight 3-85
Use andfor disclosure is
governed by the statement TIS Integrated Training System
r~
on page 2 of this chapter. © Copyright 201 0
! i
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
Pilot limitations:
Modern aircraft are equipped with adequate trim, so the pilot should not be hampered by the
physical limitation of being unable to apply sufficient rudder. Adequate experience and training
in asymmetric procedures will ensure the ability to operate at low minimum control speeds.
Critical Engine:
When both propellers rotate in the same direction, slipstream and torque effects have a natural
tendency to create yaw. In the case of modern piston-engine aircraft, where the propellers turn
clockwise when seen from the rear, the yaw tendency is to the left.
Failure of an engine means loss of power and that, in turn, induces a drop in speed. To maintain
height the angle of attack must be increased so that the aircraft is flying along in a tail-
down/nose-up attitude.
In the tail-down attitude the propshafts are inclined upwards and the tops of the propeller discs
are therefore tilted backwards. If you think about it, that means the down-going propeller blade
(Le. the one on the right of the disc when seen from astern during clockwise rotation) will have a
n
bigger angle than the up-going blade on the other side. It is a curse well known to pilots taking
off in a tailwheel aircraft, where, until the tail is raised, the propeller shaft is effectively tilted. n
u
As illustrated in Figure 3.68, and keeping in mind that we are maintaining height on reduced
power in a tail-down flight attitude:
Because the down-going blade has an increased angle, more thrust is being generated by the
right half of the propeller disc than by the left. In effect, the centre of thrust for the port engine is
moved towards the aircraft centreline whilst that for the starboard engine is moved away.
The amount of yawing force that can be generated by an engine depends on the amount of
thrust and the moment arm, through which it acts. Since moment arm B is longer than moment
arm A, the starboard engine will clearly exert more yawing force during engine-out flight than
could the port engine.
Consequently, the greater yaw (and therefore the higher minimum control speed) would in this
instance result from the loss of the port engine. In other words, when the propellers rotate
clockwise, the critical engine is on the left.
,...,
.I
u
n
U
J
3-86
TTS Integrated Training System
Module 8.3 Theory of Flight
Use and/or disclosure is
governed by the statement
n
© Copyright 201 0 on page 2 of this chapter,
n
I, ! \
'-'
-~-
~
"I :
p.gaing
blade
I
.
U
\ I
rli
i
c.J
I
Down-going
blade
r-'
I :
L/
f TUrning force TUrning force"
r..
-e..o I D/ engine
Dead engine 1-1_ _~
c ""
o
("
U
r', Figure 3.69 - An exaggerated comparison of forces which shows that from a control standpoint
I
U
!
the port engine is the worse one to lose
o It is not always easy to demonstrate a meaningful difference in minimum control speed between
the port and starboard engines, but much ado has been made of the subject. In any case, some
popular light twins now have handed propellers, the left one turning clockwise and the right one
I' anticlockwise, thus the minimum control speed is the same for both engines.
Ii
c....l
o
Use and/or disclosure is
Module B.3 Theory of Flight 3-87
governed by the statement TIS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
! \
·,
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
V-Speeds
The following is a list of 'V' codes relating to multi-engine aircraft handling:
This is, in fact, minimum control speed with an added safety margin to cater for the following
factors which could apply if an engine fails during or immediately after take-off: ,....,
;I
LJ
• Element of surprise
•
•
Failure of the critical engine (Le. the port)
Landing gear down, flaps in take-off position, propeller wind milling [J
• Pilot of average strength and ability
Provided the aircraft has attained V 2 , it should be possible to maintain direction and height while
things are being sorted out.
n
U Vyse - Speed for best engine-out rate of climb:
This should be marked on the ASI as a blue radial line and is often referred to as the 'blue line'
i" ; speed.
u
V3 - The all engines screen speed:
The speed at which the aeroplane is assumed to pass through the screen height with all
engines operating on take-off.
Vs - Stallspeed:
The speed at which the aircraft exhibits those qualities accepted as defining the stall.
n
Vfe - Maximum speed for flight with flaps extended: , !
~..1
n
• I
U
n
~J
"U
, I
c
n
High Speed Flight
I. Introduction
U
Low speed aerodynamics is based on the assumption that air is incompressible; the attendant
o errors are negligible since at low speeds the amount of compression is negligible. At speeds
approaching that of sound, however, compression and expansion in the vicinity of the aircraft
are sufficiently marked to affect the streamline pattern about the aircraft. At low subsonic
speeds a flow pattern is established about the aircraft, but at high subsonic and supersonic
speeds the flow around a given wing can be controlled, and its behaviour predicted. In the
transonic range where a mixture of subsonic and supersonic flow exists, marked problems of
control and stability arise, necessitating special design features to minimize the effects of
flu compressibility.
r:u Whilst "Supersonic Flight" is not a part of the EASA Part-66 Module-8 syllabus, the effects of
compressibility are experienced by aircraft flying at speeds considerably lower than the speed of
sound (Mach 1). The term High Speed Flight is used to denote aircraft speeds from about
Mach 0.6 and above. Most transport category aircraft cruise at between Mach 0.7 and 0.85.
Definitions
fl
u Speed of Sound - The speed at which a very small pressure disturbance is propagated in a
fluid under certain conditions. Speed of sound is proportional to the absolute temperature (K)
r.
,
\.....:
' and can be calculated from the formula:
o C= ~ =38.95°
o
Ii
I '
U
I"'
I "
J Use and/or disclosure is
Module 8.3 Theory of Flight 3-91
TTS Integrated Training System
governed by the statement
on page 2 of this chapter. © Copyright 2010
,n ,
Integrated Training System
Designed in association witll the
club66pro,co.uk question practice aid n
Mach Number (M) - The ratio of True Airspeed (TAS) to the local speed of sound applicable to ,n,
air temperature. Thus U
Mach No (M) = TAS therefore at sea level temperature 15°C TAS = 529 kts.
LSS
Free Stream Mach No (MFs): - The Mach number of the flow at a point unaffected by the
presence of the aircraft.
Local Mach Number (Md: - When an aerofoil is placed in a subsonic airflow, the flow is n
~._J
accelerated in some places, and slowed down in others. The local Mach number is the speed at
some specified region of flow, and may be greater than, the same as, or lower than MFS.
Critical Mach Number (Merit): - This is the lowest MFS which for a given aerofoil and angle of
attack, gives rise to a ML of 1.0 on the aerofoil. As will be seen, Merit for a wing varies with angle
of attack.
Critical Drag Rise Mach Number: - This relates the Mach number to an appreciable increase
of drag associated with compressibility effects, usually 10 -15% higher than Merit n
Airflow n
lJ
• Subsonic flow when free stream Mach numbers are such that local Mach numbers are
less than M 1.0 at all points.
I
:, J
• Transonic flow, the MFS is high enough to produce ML, some of which are greater than M
1.0. :-)
• Supersonic flow, MFs is such that at all points ML are greater than M 1.0. , I
l ___ J
Speed of Sound
Anything which moves through the air creates pressure waves and, what may not be generally
realized, these waves not only travel out in all directions from the object but they radiate at the
speed of sound. If the object is moving at a speed less than the speed of sound these pressure
waves will be able to move away from the object. When considering aircraft moving at very high
n
n
speed it is possible that the sound wave cannot get away from it, because the aircraft's speed is
close to the radiation speed of the waves. It is this which gives rise to the problems of high U
speed flight.
n
il J
I
Figure 3.70 illustrates the situation of an aircraft flying at less than the speed of sound. If its
starting point is A, then the pressure waves set out in all directions from the aircraft are moving
steadily away and by the time point B is reached they will be well clear of the aircraft. This
r-~.
I ,
Integrated Training System
I . Designed in association with tile
LJ club66pro.co.uk question practice aid
should be contrasted with the situation illustrated in (b) where the aircraft is traveling just at the
Ii
u speed of sound. The pressure waves are also traveling at the speed of sound with the result
that they pile up ahead of the aircraft and form into a pressure wave, also called a shock wave.
ri
I I
u
[J
M=O.5
fi
u
1'1
I I
L.J
w
>
o ~~II'------+------~------·I----~~
o or:(
~
II:
o
Z
o
o
o
u Use and/or disclosure is
governed by the statement
on page 2 of this chapter.
Module 8.3 Theory of Flight 3"93
TTS Integrated Training System
© Copyright 2010
Integrated Training System
Designed in association with the
club66pro,co.uk question practice aid
M =2.0· Mach
o
AFTER 'I
Mach FORE Cone
:I
• >
,...,
, I
\J
n
o
n
, I
{cl Pressure Waves forming Mach Cone with Object lJ
n
, I
~.J
M.2.0
o ....,
: \
'.J
n
(d)
n
governed by the statement
© Copyright 2010 . on page 2 of this chapter.
Integrated Training System
Designed in association with the
club66pro.cQ,uk question practice aid
Shock Waves
When a shock wave is formed the pressure distribution over the wings is materially altered,
o causing considerable alterations in the values of lift and drag and also affecting control
operation. It could be argued that few civil passenger transport aircraft are capable of reaching
the speed of sound, however, the air over the upper surface of the wing is deliberately
r--, accelerated in order to produce lift and even though the aircraft itself may be flying below the
I ' speed of sound, some of the air flowing over the wings may be accelerated to Mach 1.0. When
LJ
the airflow over the upper surfaces of the wing reaches Mach 1.0, the actual speed of the
o aircraft is called the critical Mach Number or Merit. When this point is reached a shock wave
forms over the upper surface of the wing because the pressure waves from the rear of the wing
that are trying to move forward are meeting air traveling at exactly the same speed flowing
backward. This is similar to trying to move along a moving walkway in the wrong direction at the
same speed as the walkway is traveling. The point at which this shock wave usually forms is
just aft of the point of maximum camber of the wing where the acceleration of the air is greatest.
[1 In front of the shock wave the flow is at or higher than Mach 1 whilst behind the flow it is still
I i
subsonic.
w
At the shock wave, the normal laws of physics seem to break down and as the air passes
through the shock wave the pressure increases and the temperature increases. If the speed of
the aircraft is increased still further the region of supersonic flow on top of the wing also
increases and the shock wave will start to move back towards the trailing edge. On the
undersurface the curvature of the wing is usually less than on the upper surface and the shock
wave will form later. However, once having formed, if the actual speed of the aircraft is further
increased, this shock wave will also move rearward and when the actual speed of the aircraft
reaches Mach 1 both shock waves will have migrated to the trailing edge of the wing. At the
same time another shock wave will form close to the leading edge of the wing, this is called the
_1
I . bow wave. If speed is further increased this bow wave will actually touch the leading edge of the
I I
wing and is then termed an 'attached bow wave'. This is illustrated in Figure 3.71 and further
'-'
speed increases will not change the relative positions of these two shock waves, but will just
bend them backwards.
C.P. c.P.
o
2.0
r1,
u
Figure 3.71 - Shock wave development on an aerofoil
Figure 3.72 illustrates the behaviour of the shock waves from a speed below Mach 1 to one well
o in excess of the speed of sound.
r:
U
o Use and/or disclosure is
governed by the statement
on page 2 of this chapter.
Module 8.3 Theory of Flight 3-95
TIS Integrated Training System
© Copyright 201 0
Integrated Training System
Designed in association with the
c1ub66pro.co.uk question practice aid
=
(e) M 2.0
Supersonic
Supersonic Supersonic Row
... o
Supersonic Supersonic
I~
L..J
Wave Drag
For aircraft not designed for transonic and supersonic flight, the formation of these shock waves
II
I I will have a marked effect on lift, drag and also on the general stability of the aircraft and its
U
control. The basic cause of the problems is the separation of the airflow behind the shock wave
o due to the rise in pressure. This causes the boundary layer to separate, reducing the amount of
lift produced by the wing and an increase in drag. This increase in drag is very marked at Mcrit
and produces 'wave drag' - as previously mentioned in the chapter on total drag. If speed can
be increased further against this drag force the shock waves will move towards the trailing
edge, thus reducing the amount of separated air and, in fact, the lift will start to increase again
and the drag decrease. However, this will only occur in aircraft which are designed for transonic
o flight but the total developed lift at higher speeds is less than at subsonic speeds for the same
angle of attack. This is because CL is less for any given wing section and angle of attack at
supersonic speeds. The change in both lift and drag are illustrated in the following graphs. It
o should be noted that this loss of lift which occurs as a result of the shock wave is not dissimilar
in effect to that produced by a low speed stall. For this reason it is sometimes called a 'high
speed stall' and gives rise to the first of the control difficulties encountered in flying an aircraft at
or above Merit.
c t
o
[I
u
[1. M... rlt
U Figure 3.73 - Wave drag
o Wave drag arises from two sources, viz: energy drag and boundary layer separation.
Energy drag stems from the nature of changes occurring as a flow crosses a shock wave.
Energy lost due to temperature rise across the shock wave becomes drag on the aerofoil. The
[1 more oblique the shock way the less energy they absorb, but since they become more
I 1
U extensive laterally and affect more air, energy drag rises as MFs increases.
Boundary layer separation; at certain stages of shock wave movement there is considerable
flow separation. The turbulence represents lost energy and contributes to the drag. As MFS
increases through the transonic range the shock waves move to the trailing edge and
separation decreases, thus drag decreases.
separation of the boundary layer and the formation of the bow shock wave above M 1.0.
Another factor affecting CL is the amount of warning the air gets of the wing's approach.
n
As speed increases compressibility effects increase and the reduced upstream warning
causes flow displacement to start closer to the wing. This effectively increases the angle
of attack, so increasing CL. There is a slight loss of lift due to movement of the stagnation
point forward, but overall there is an increase in CL.
(b) Transonic Variations in CL - In considering this aspect, five significant speeds are
n
selected, A, B, C, 0 and E (previous graph), and are used in Figure 3.74. n
cJ
n
J
n
n
u
3-98
ITS Integrated Training System
Module 8.3 Theory of Flight
Use and/or disclosure is
governed by the statement
o
© Copyright 201 0 on page 2 of this cI1apter.
Integrated Training System
o Designed in association with the
club66pro.co. uk question practice aid
11
u
o ,
n
U I
-_. -- -- ---- - j-._- --. --- ---- --- ---_._- j---------- ---_._--- _
..
o j I
~
I i
o
fl
]
u
]
- S~h.'>ni" - '
I
' j
1_.- --'-.
'----<..._ _---L.._ _ _ _ _ _ _ _ _ L._~-.~
C
Trans!}ni" ----_r' -
.
Sup<:roonic--
!
I
__
(I,. M.ri. ", Mach No_ 1.4
o Figure 3.74 - Lift fluctuations as aircraft flies through the transonic region
At A MFS = 0.75, the flow accelerates rapidly from the stagnation point along both upper
and lower surfaces, giving a sharp drop in pressure, and the wing is above Merit. Over
the top surface of the wing, as yet there is no shock wave, and CL has risen by 60% of its
low speed value for the same angle of attack. Over the bottom surface flow is still
subsonic.
At B MFS = 0.81. With the acceleration to this speed, the shock wave has formed and is
strengthened, and will be approximately 60% chord; there is no shock wave on the
undersurface. Behind the shock wave on the rear part of the wing there is no real change
o in pressure differential between upper and lower surfaces; ahead of it and behind the
40% (approximately) chord, pressure differential has increased considerably due to
supersonic acceleration up to the shock wave. This effectively increases the CL to
[1 roughly double its incompressible value. It also causes the Cp to move rearward to
U approximately 30% chord. Flow under the bottom surface becomes sonic.
o At C MFs = 0.89. A shock wave has formed on the undersurface and moved to the trailing
edge, while the upper shock wave has remained virtually stationary. The reason for the
differing behaviour is the effect each shock wave has on the boundary layer. Such an
o arrangement of shock waves leads to a pressure distribution such that the wing behind
the upper shock wave is producing negative lift, which has to be subtracted from the
positive lift producing area. Lift coefficient has dropped to approximately 30% below its
11w incompressible value and centre of pressure moves forward to approximately 30% chord.
The reason for the slope between Band C is the relative movement between upper and
lower shock waves.
o At 0 MFS = 0.98. The top surface shock wave is forced to the trailing edge, the area of
negative lift is replaced by the orthodox pressure differential. The CL is approximately
o
o Use and/or disclosure is
governed by the statement
on page 2 of this chapter.
Module 8.3 Theory of Flight 3-99
TIS Integrated Training System
© Copyright 201 0
[1
, ,
lJ
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid n
10% above basic value, and the Cp has moved rearward to approximately 45% chord;
this movement of the C p is experienced by all aircraft going through the transonic range. n
At E MFs = 1.4. Above M 1.0 the bow shock wave forms, and at M 1.4 is almost attached
to the leading edge. The whole of the wing is producing lift, and the C p is at
approximately mid-chord position. The CL is reduced to a value of 30% less than its
n
incompressible value due to the stagnation point moving to the most forward point on the
leading edge, and to the loss of pressure energy through the bow shock wave.
n
u
n
:1u
il
LJ
n
U
n
c."
[J
o
n
n
n
Ll
o
o
3-100
TIS Integrated Training System
Module 8.3 Theory of Flight
Use andfor disclosure is
governed by the statement
n
© Copyright 2010 on page 2 of this chapter.
1
, J
c
Integrated Training System
[1
u Designed in association with the
club66pro.co.uk question practice aid
o The shock wave positions for each station considered above are shown in Figure 3.75.
o
[J MFS =0.75
ShockWave
o
o MFS =0.81
o
f:u
MFS '" 0.89
rlu
I,
u
MFs = 0.98
Bow Wave
o
o Figure 3.75 - Shock wave positions at various increasing speeds
o
[1
LJ
Supersonic Fall in CL n
The full explanation of this aspect is beyond the scope of these notes, but suffice it here to point
out that in practice, any decrease in lift between the upper limit of the transonic range and MFs
1.4 would be masked by trim changes resulting from passing through the transonic range. The
n
practical result is that the lift curve slope becomes progressively gentler with an increase in MFs
in the supersonic range. The variations in CL at supersonic speeds depend mainly on attendant
compressibility problems with increase of speed, and adverse pressure gradient increases with
speed increase.
• Modification of airflow over the tailplane. The effect of mainplane shock waves is to
modify the flow over the tailplane which will tend to pitch the aircraft nose down.
.;1
The effects on an aircraft's handling characteristics of nose down pitch are two-fold. U
(i) At some Mach No an aircraft will become unstable with respect to speed,
necessitating a rearward movement of the control column. This particular problem
is dealt with more fully in Mach Trim.
n
(ii) The requirement for a large up deflection of elevator/tailplane reduces the amount
of available control deflection for manoeuvres.
iJ
This nose-down tendency is known as Mach Tuck, or Tuck-under 1u
Supersonic Longitudinal Stability
The rearward movement of the C p in the transonic range continues as the aircraft accelerates n
lJ
into full supersonic flight. Thus all aircraft experience a marked increase in longitudinal stability.
o Directional Stability
The trend towards rear mounted engines, and consequently an aft CG, has meant a decreased
o arm about which the fin can act. Also, the supersonic decrease in CL for a given angle of attack
caused by sideslip means a reduction in fin effectiveness. Subsonically, the fuselage side force
in a sideslip acts in front of the CG and the vertical fin surfaces are able to overcome the
o destabilizing condition. In supersonic flight the fuselage side-force moves forward. As long as
the aircraft is in balanced flight no problem arises, but if the relative airflow is off the longitudinal
axis a destabilizing force at the nose results. This is caused by asymmetry in the strength of the
o
o Weak Shockwave
Small
o Pressure
Rise
Pressure
o Gradient
Force
o
o Figure 3.76 - The effect of the shock wave on the fin
o The nose force illustrated in Figure 3.76 is tending to prevent the nose being tumed into the
relative airflow and is therefore destabilizing. The force increases with speed and has a longer
arm than the fin and rudder. The point of application of the force is difficult to define, but is
o located at that part of the fuselage where the cross-sectional area is increasing.
One answer to this problem is to fit longer fins and increase their numbers, but there is a limit if
o only for wave drag considerations. A better method is the fitting of yaw dampers, which have
already been dealt with.
u
o
o
o Use andfor disclosure is
governed by the statement
on page 2 of this chapter.
Module 8.3 Theory of Flight 3-103
TIS Integrated Training System
© Copyright 201 0
II
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid n
n
Mach Trim IJ
The device which corrects or compensates for longitudinal instability at high Mach numbers is
the Mach Trimmer. As stated previously, at some Mach number an aircraft will become unstable n
with respect to speed; this is potentially dangerous since any inattention on the part of the pilot
U
in allowing a small increase in Mach No will produce a nose down pitch, which will give further
increase in Mach No, in turn leading to even greater nose down pitch. However, the Mach
Trimmer will in fact correct or compensate for the initial increase in speed.
o
The Mach Trimmer is sensitive to Mach number and is programmed to feed into the
elevator/stabilizer a signal which is proportional to Mach number so that stability remains
n
positive. The signal fed into the elevator/stabilizer simply causes their deflection in a direction to
compensate for the trim change.
n
Mach trim operation in normal conditions will not be shown up by the behaviour of the aircraft,
but will usually be indicated by activation of the trim wheel and/or illumination of a monitor light.
Mach trim operation should be checked against Mach number for any significant change in flight
n
condition. u
Key Points
• Fixed Trim tabs are used to correct permanent out-of-trim faults and can only be adjusted
o
•
•
on the ground.
Fixed Trim tabs should only be adjusted by an engineer.
A Balance tab is fitted to assist the pilot in moving the controls in flight.
o
• A Servo tab is activated by movement of the control column which directly moves the tab
which then aerodynamically moves the control surface.
IJ
n
• On some supersonic aircraft longitudinal trim is achieved by moving fuel from one fuel
tank to another. The tanks are positioned fore and aft under the floor and by pumping
fuel from one to the other the C of G is moved.
• A high aspect ratio wing gives improved lift and reduced drag.
n
•
•
Employment of swept wings, or wings with swept leading edges, will delay Mcrit.
For a given wing area at a given angle of attack a swept wing will produce less lift.
lJ
• A swept wing is more prone to tip stall.
• Spanwise movement of airflow over a swept wing may be reduced by:
o
o
Wing Fences.
Leading edge Notches.
nJ
,~
o Saw, or Dogtooth Leading edges.
o Vortex Generators.
o Wing Blowing.
n
lJ
3-104
ITS Integrated Training System
© Copyright 201 0
Module 8.3 Theory of Flight
Use and/or disclosure is
governed by the statement
n
on page 2 of this chapter.
n
IJ
Integrated Training System
Designed in association with the
LJ club66pro.co.uk question practice aid
o
rL.J1
TTS Integrated
Training System
["
LJ
U Module 8
[1
U
o Basic Aerodynamics
Iiu
8.4 Flight Stability and Dynamics
r'
,
l
u
11
L
[":
U
o
o
[1
Copyright Notice n
r:
© Copyright. All worldwide rights reserved. No part of this publication may be reproduced,
stored in a retrieval system or transmitted in any form by any other means whatsoever: i.e.
U
n
photocopy, electronic, mechanical recording or otherwise without the prior written permission of
Total Training Support Ltd.
LEVEL 1
• A familiarisation with the principal elements of the subject.
Objectives: n
•
• The applicant should be familiar with the basic elements of the subject.
The applicant should be able to give a simple description of the whole subject, using common words and
iJ
examples.
• The applicant should be able to use typical terms. n
!J
LEVEL 2
• A general knowledge of the theoretical and practical aspects of the subject. rJ
L
• An ability to apply that knowledge.
Objectives:
• The applicant should be able to understand the theoretical fundamentals of the subject.
• The applicant should be able to give a general description of the subject using, as appropriate, typical
examples.
• The applicant should be able to use mathematical formulae in conjunction with physical laws describing the
subject. n
I. 1
• The applicant should be able to read and understand sketches, drawings and schematics describing the
subject.
• The applicant should be able to apply his knowledge in a practical manner using detailed procedures. n
;. I
"_ •.J
LEVEL 3
• A detailed knowledge of the theoretical and practical aspects of the subject.
n
• A capacity to combine and apply the separate elements of knowledge in a logical and comprehensive
U
manner.
Objectives:
•
• The applicant should know the theory of the subject and interrelationships with other subjects.
The applicant should be able to give a detailed description of the subject using theoretical fundamentals
o
and specific examples. n
• The applicant should understand and be able to use mathematical formulae related to the subject.
• The applicant should be able to read, understand and prepare sketches, simple drawings and schematics
U
describing the subject.
• The applicant should be able to apply his knowledge in a practical manner using manufacturer's
instructions.
• The applicant should be able to interpret results from various sources and measurements and apply
corrective action where appropriate.
n
4-2 Module 8.4 Flight Stability and Dynamics U
Use and/or disclosure is
n
TTS Integrated Training System governed by the statement
© Copyright 201 0 on page 2 of this chapter.
o
Integrated Training System
D Designed in association with the
club66pro,co.uk question practice aid
o Table of Contents
r-
Ii
u
Module 8.4 Flight Stability and Dynamics _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 7
c Static Stability
Directional Stability
Longitudinal Stability
7
8
10
Lateral Stability (Stick Fixed) 18
Dynamic Stability 29
General 29
Periodic Time 30
Damping 31
11 Dynamic Stability of Aircraft 31
lJ Longitudinal Dynamic Stability
Lateral Dynamic Stability
31
33
Summary 35
r1
L
c
o
ri
LJ
[1
u
r1
L.:
o
Ii
LJ
'I
, I
,J
u
Intentionally Blank
n
n
LJ
n
U
n
TIS Integrated Training System governed by the statement
© Copyright 201 0 on page 2 of this chapter.
ru 1
c
roo,
U
fl
LJ
fi
u
r1
LJ
"
U
o
11
L...'
o
r1
LJ
D
o Use and/or disclosure is
govemed by the statement
Module 8.4 Flight Stability and Dynamics 4-5
TIS Integrated Training System
on paQe 2 of this chapter. © Copyright 201 0
Integrated Training System
Designed in association with the
club66pro,co.uk question practice aid
n
o
o
il
o
n
"J
n
LJ
Intentionally Blank
n
n
lJ
n
u
n
;j
, ,
~
lJ
n
o
Integrated Training System
o Designed in association with the
club66pro.co.uk question practice aid
rl ~ Initial Stability
LJ
fl
o New Position
Before considering the response of the aircraft to disturbance it is necessary to resolve the
motion of the aircraft into components about the three body axes passing through the eG.
It is important to realize that the motion involved is angular velocity and the disturbance
[1 assumed s an angular displacement. In the first instance it is helpful to consider these
components separately although, in other than straight and level flight, the motion of the aircraft
is more complex, e.g. in a level turn the aircraft is pitching and yawing.
Directional Stability o
A simple approach both to directional and to longitudinal stability is to consider a simple dart.
The flights or vanes of a dart ensure that the dart is aligned with the flight path. Consider first
the pair of vanes which impart positive directional stability; these may be referred to as the
n
vertical stabilizers. Figure 4.2 shows how a displacement in yaw through an angle ~, resulting in
sideslip, produces a restoring moment and, therefore, positive directional (static) stability. Two
points are worth noting:
.n
(a) The dart rotates about its centre of gravity (CG). o
(b) The momentum of the dart momentarily carries it along the original path, i.e. the ,...,
!I
relative airflow RAF is equal and opposite to the velocity of the dart. r~ J
n
Restoring Sideslip .:)
Moment Angle (~)
Force
n
RAF :I
" . .J
•
c'" CG
(Plan View)
Vertical
Stabilizer
An aerodynamic shape like a fuselage may be unstable. Reference to the Figure 4.3 shows that
this occurs when the centre of pressure (CP) is in front of the CG.
Force
l J
4
,I "
I
I
I
Flight Path U
I
\
\
\ CP
Unstable II
, I
Moment
:J
Figure 4.3 - If the Centre of Pressure is forward of the Centre of Gravity the object is unstable
n
4-8 Module 8.4 Flight Stability and Dynamics
U
Use and/or disclosure is
TIS Integrated Training System govemed by the statement
© Copyright 201 0 on page 2 of this chapter.
l
u
1'1
I :
~
o It is necessary. therefore, to add a vertical stabilizer or fin to produce positive directional stability
and this has the effect of moving the CP behind the CG (Figure 4.4). In general it may be said
that the keel surface of the fuselage ahead of the CG has an unstable influence, while the keel
surface behind the CG has a stable influence. (For simplicity, the rudder is considered to be
'locked'.)
fl
I '
U
Restoring
c Moment
o Flight Path
rl
l.j CP
Ii
, :
~
Vertical
Stabilizer
(Plan View)
r,
I ,
o Unstable
Influence
Stable
Influence
o ISide View)
For a given displacement, and therefore sideslip angle, the degree of positive stability will
depend upon the size of the restoring moment, which is determined mainly by:
r,
U • Design of the vertical stabilizer.
• The moment arm.
Longitudinal Stability
The analogy of the dart can be wed with advantage to introduce the concept of static l
longitudinal stability. In this case the dart is viewed from the side and the horizontal stabilizers LJ
produce a pitching moment (M) tending to reduce the displacement in pitch. On an aircraft, the
tailplane and elevators perform the functions of a horizontal stabilizer and the conclusions
reached will be equally valid. For simplicity, the explanation is limited to stick-fixed static
stability, i.e. elevators locked.
o
Figure 4.5 shows a wing with the CP forward of the CG by the distance x. A nose-up ;l
•.J
displacement will increase the angle of attack, increase the lift (L) by the amount dL and
increase the wing pitching moment by the amount dLx. The result is to worsen the nose-up
displacement: an unstable effect. In the figure at b, the CP is aft of the CG and the wing
n
u
moment resulting from a displacement in pitch will be stabilizing in its effect.
The pitching moment is also affected by the movement of the CP with angle of attack and it
follows, therefore, that the relative positions of the CP and CG determine whether the wings
have a stable or unstable character.
n
LJ
.'""1
I I
,J
,..,
lJ
4-10
TTS Integrated Training System
Module 8.4 Flight Stability and Dynamics
Use andfor disclosure is
governed by the statement
o,..,
© Copyright 201 0 on page 2 of this chapter. , I
lj
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
r:
! '
r-
I I
I '
L!
L
_x__
a Unstable Contribution
r- o
--,
U
L
c b Stable Contribution
Figure 4.5 - Pitching moments with CG aft and CG forward of the Centre of Pressure
o Taking the worst case, therefore, the wing may have an unstable influence and the horizontal
stabilizer must be designed to overcome this.
n
u The simplified diagram of Figure 4.6, illustrates the growth of a system of forces due to
displacement in pitch, in this case an increase in angle of attack. The tail contribution must
overcome the unstable wing (and all other) contributions for positive static longitudinal stability.
The degree of positive stability for a given change in angle of attack depends upon the
'I difference between the wing moment and the tail moment; this difference is called the restoring
L: moment,
fl
w
i.e. (Total Liftt.il)y - (Total Liftwing)X =net pitching moment.
The main factors which affect longitudinal stability are:
o (a) Design of the tailplane.
r
U (b) Position of the CG.
Lwing
/
Wing I
x ___:.....; . - - _ . _ - - y------.,..,,
,i
I
,I
,
i I'
Flight Path j, dllail
n
u
,I
n. I
r I
I~_ J
• Tail Area. The total lift provided by the wing = CL Wing q S and the total lift produced by the
=
tail CL tail q S.
For a given aerofoil of given planform, the CL varies with angle of attack at a constant q (EAS).
Therefore in comparing tail moments with wing moments, it is necessary only to compare the n
, ,
respective area(s) and moment arms (CG position). L..J
I
• Tail Volume. The product of the tailplane area x moment arm is known as the tail
volume. The ratio of the tail volume to the wing volume is the main parameter used by
the designer in determining the longitudinal stability of the aircraft.
• Planform. The slope of the CL curve for a lifting surface is affected by aspect ratio, taper
and sweepback. The planform of the tailplane therefore affects the change in CL with
change in angle of attack caused by a disturbance. 1
~J
For example, the CL increments will be lower on a swept-back tail than on one of
,....,
rectangular planform. i I
U
• Wing Downwash. Where a disturbance in angle of attack results in a change in the
angle of downwash from the wings, the effective angle of attack at the tail is also
c changed. For example, if the aircraft is displaced nose-up and the downwash angle is
increased, then the effective angle of attack on the tailplane is reduced. The total tail lift
will not be as great as it would otherwise have been and so the restoring moment is
reduced. This decrease in stability is compensated for by moving the CG farther forward,
thereby increasing the moment arm.
Position of the CG
The position of the CG may be marginally under the control of the pilot of the aircraft. From the
c previous diagram it can be seen that its position affects the ratio of the tail moment to the wing
moment and therefore the degree of stability. In particular:
Because the position of the CG affects the positive longitudinal stability, it also affects the
handling characteristics in pitch. The aerodynamic pitching moment produced by deflecting the
elevators must override the restoring moment arising from the aircraft's positive stability, i.e. the
"I i stability that opposes manoeuvre. For a given elevator deflection there will be a small response
u
in an aircraft with a forward CG (stable condition) and a large response in an aircraft with an aft
CG (less stable condition).
Neutral Point
ro Every aircraft Flight Manual gives the permitted range of movement of the CG. The forward
I '
LJ position is determined mainly by the degree of maneuverability required in the particular aircraft
type. Of greater importance to the pilot is the aft limit for the CG. If the CG is moved aft. outside
o the permitted limits, a position will eventually be reached where the wing moment (increasing) is
equal to the tail moment (decreasing). In this situation the restoring moment is zero and the
aircraft is therefore neutrally stable. This position of the CG is known as the neutral point. The
n aft limit for the CG, as quoted in the Flight Manual, is safely
U forward of the neutral point. If the loading limits for the aircraft are exceeded, it is possible to
have the CG position on, or aft of, the neutral point. This unsafe situation is aggravated when
t
Nose
up I Trim Point
n
. I
\..,J
Nose
down
0 +---------~~~--------CL
---
n
---~
-- ---- n
* ... ...,
The difference between the two settings is known as longitudinal dihedral, but has no effect on U
the basic stability of the aircraft. Varying the tailplane incidence only shifts the trim point. As the
CL vs. angle of attack curves of the main plane and tailplane may be regarded as straight lines
(up to the stall), the variation in lift per degree alpha change, does not depend on the initial
incidence settings nor on their difference.
o
n
u
Elevator Angle to Trim
If the angle of attack is increased from the trim point, the aircraft's longitudinal stability will
produce a stable, nose-down pitching moment. To maintain the new angle of attack, an equal
and opposite moment, nose-up, will be required from the elevators. When this is achieved, by
raising the elevators, a new trim point is established, i.e. at the higher angle of attack on the ~
mainplane, the tail has been made to produce a greater nose-up moment by altering the · I
effective camber on the tail. The reverse applies when the angle of attack on the mainplane is i.. J
reduced. This does not usually affect the positive longitudinal stability.
Aerodynamic Centre
In text books on stability it is usual to find that the aerodynamicist writes of the 'aerodynamic
centre' (AC) rather than of the centre of pressure. The AC is a point within the aerofoil, and
usually ahead of the CP, about which the pitching moment is independent of angle of attack; it
is a convenient and calculated datum for the mathematical treatment of stability and control.
c airflow and this will reduce the tail moment. Under these conditions, with the tail moment
reduced, the balance between the tail and wing moments is changed and, therefore, the
position for the CG, about which the moments are equal, will be farther forward, because the
less effective tail requires a longer moment arm. That is, the neutral point is farther forward, so
reducing the stick-free CG margin. Since this margin is a measure of the longitudinal stability it
follows that when the elevators are allowed to float free the longitudinal stability is reduced.
r1
U Manoeuvre Stability (Steady Manoeuvres Only)
,._., In the preceding paragraphs the longitudinal static stability was discussed with respect to a
I '
disturbance in incidence from the condition of trimmed level flight. A pilot must also be able to
U hold an aircraft in a manoeuvre and the designer has to provide adequate elevator control
The following paragraphs consider the effects on an aircraft of a disturbance in angle of attack
o and normal acceleration. It should be carefully noted that the initial condition is, as before,
steady level flight.
The difference between static and manoeuvre stability is that manoeuvre stability deals with a
disturbance in angle of attack (a) and load factor (n) occurring at constant speed, whereas static
stability deals with a disturbance in angle of attack at constant load factor (n 1). =
r1
1 ,
~
If an aircraft is trimmed to fly straight and level (the initial condition, Figure 4.8 (a)), and is then
climbed, dived and pulled out of the dive so that at the bottom of the pull-out it is at its original
I"~, trimmed values of speed and height (Figure 4.8 (b)), then the aircraft can be considered as
U having been 'disturbed' from its initial condition in two ways, both contributing to the overall
manoeuvre stability:
o • It now has a greater angle of attack to produce the extra lift required to maintain a curved
flight path (L = nW). This is the same as the static stability contribution discussed earlier.
• It has a nose-up rotation about its CG equal to the rate of rotation about its centre of pull-
out.
rl
,
1
~
;
'
Angle
of Attack
Flight Path
a Level Flight
L=nW
Nose up
n
Increased Rotation
Angle Flight
of Attack Palh .
,~
.J
b Pull-Out
n
Ll
Figure 4.8 - Aircraft in level flight, and in the pull-out from a dive
Because the aircraft is rotating about its own CG, the tailplane can be considered to be moving
downwards relative to the air or, alternatively, the air can be considered to be moving upwards
relative to the tail plane, In either case the effective angle of attack of the tailplane will be
increased (see Figure 4.9); thus the manoeuvre stability is greater than the static stability in
level flig ht. n
t..J
....,
I I
V
LJ
Vertical
Velocity
Increased
Angle of Attack
If the aircraft's longitudinal stability is greater in manoeuvre, the position of the CG which
achieves neutral stability will be farther aft than for the straight and level case. This position of
the CG is called the manoeuvre point (corresponding to the neutral point) and the distance
n
,_._1'
between the CG and the manoeuvre point is called the manoeuvre margin. It will be seen that
for a given position of the CG, the manoeuvre margin is greater than the CG margin.
Effect of Altitude
Consider an aircraft flying at two different heights at the same lAS (Le. the same value of CLl
and apply the same load factor in each case. Since the TAS is higher at altitude, the rate of
pitch of the aircraft decreases
MV 2
(Centripetal force = - - = MO)V, where 0)= rate of rotation)
o r
Figure 4.10 shows the decrease in tailplane angle of attack due to the higher TAS and lower
c rate of pitch. At the same lAS, the higher aircraft has less manoeuvre stability because of the
reduction in the tailplane contribution.
n
u Change in
o Angle of Attack
at High Altitude
o
o
o Change in
Angle of Attack
at Low Altitude
Low Altitude
High Altitude
n
U
c
r'
LJ
The forces arising on the different parts of the aircraft as a result of the sideslip, produces a
rolling moment tending to restore the aircraft to its initial wings-level position. It is seen therefore
that the lateral static stability of an aircraft reacts to the sideslip velocity (v) or a displacement in
yaw (see diagram overleaf). This effect has a considerable influence on the long-term response
(lateral dynamic stability) of the aircraft.
n
LJ
u
"1
i'.,"I
n
u
"1
U
n
d
n
L,I
fl
..J
rl
u
Ii
i i
u Roll Components ........---- f rr Roll to
o Starboard
r
Roll v
Components
Up·Going Wing
ri
u
o Roll
v
Components
Down-Going Wing
U Each different part of the aircraft will contribute towards the overall value of the lateral static
stability and these contributions will be of different magnitude depending on the condition of
flight and the particular configuration of the aircraft. The more important of these contributions
are:
(,
, I • Wing contribution due to:
U
• Dihedral.
o •
• Sweepback.
Wing/fuselage interference.
rl
,, LJ
• Fuselage and fin contribution.
• Undercarriage, flap and power effects.
I~
I I
u
~~
1/ ~ n
!__ J
,/ --......____ Sideslip
' ~ Velocity (v) ", ,
/ U
,/
I
/
w
,....,
;c •I
Dihedral Effect (I
Dihedral effect can be explained in a number of ways but the explanation illustrated below has ,; .,I
the advantage of relating dihedral effect to sideslip angle. In the Figure 4.13 it will be seen that
due to the geometric dihedral, a point nearer the wingtip (A or D) is higher than a point inboard :J
(B or C). A sideslip to starboard will therefore produce the following effects: lj
• Starboard Wing. The relative airflow will cross the wing (from A to B) at an angle equal
to the sideslip angle. Since point A is higher than point B this will produce the same effect
as raising the leading edge and lowering the trailing edge, i.e. increasing the angle of
attack. So long as the aircraft is not flying near the stalling speed, the lift will increase. n
i I
t ..)
• Port Wing. Bya similar argument, the angle of attack on the port wing will reduce aid its
lift decrease. nu
A stable rolling moment is thus produced whenever sideslip is present (i.e. following a
disturbance in yaw). This contribution depends on the dihedral angle and slope of the lift curve.
It will therefore also depend on aspect ratio being increased with an increase in effective chord
n
length. It is also affected by wing taper. This is one of the most important contributions to the
overall stability and for this reason, the lateral static stability is often referred to as the 'dihedral
n
L j
n
LJ
o - ~~ I
:I :I
I
I
BA
I i
Flight:I II
II
I I
I Path ..I I
I I
I I
I I
I
I
I
[ ! Angle
SideSIiP't- I I
," :.
I
u
I
I
I
I
I I
I
I
I
I
I
I
I
I I I
I I ,
I I I
n I
I
I
I
I
I
U f
I
I
I
I
I
I
I
I
I
I
I
I
I
I
C
I I
I I' A
I I
,n "
LJ C Lower than D· A Higher than B·
Decreased Angle Increased Angle
of Attack of Attack
Ii
, ,
w
Figure 4.13 - Effect of dihedral on angle of attack of an aircraft in sideslip
r'
, I
U
Sweepback
Wing sweepback has the effect of producing an additional stabilizing contribution thus
increasing the 'effective' dihedral of the wing (10 0 of sweep has about the same effect as 10 of
dihedral). The following diagram illustrates the principal effects on wing geometry of sideslip.
• Aspect Ratio. On the leading (low) wing the span is increased and the chord decreased,
fl which is an effective increase in aspect ratio. On the trailing (high) wing, the span is
u
decreased and the chord is increased resulting in a reduction in aspect ratio. This again
produces a stable rolling moment because the more efficient (low) wing produces more
lift.
c • Taper Ratio. Another, smaller effect, arises from a tapered wing. An increase in taper
ratio, defined as tip chord, affects the lift coefficient and also produces a small stable
rolling moment in sideslip.
c
rU1
nI
,
Sideslip
Velocity
re.
Forward High : I
Velocity l J
Leading' i o - - - -____~
Sideslip
Trailing I+----::"L-".""---:;";---
rt
lJ
Angle of Attack
n
Effect of Aspect Ratio u
Angle of Attack
Effeet of Sweep.Back iJ
Figure 4.14 - Effect of sweepback on angle of attack of an aircraft in sideslip II
i~j
Variation with speed
The changes in the slope of the lift curve associated with changes in aspect ratio and sweep
result in variations in lift forces of the 'leading' and 'trailing' wings. The contribution of sweep to
n
.J
the lateral (static) stability therefore becomes more important at the higher values of CL, i.e. at
the lower forward speed, because the CL curves are divergent. This is very important because it
means that the 'dihedral effect' varies considerably over the speed range of the aircraft. At high
speeds a lower angle of attack is needed than that for low speeds, therefore the stability at high
rt
speeds is much less than that at low speeds. To reduce the stability to a more reasonable value I ,
at the higher angles of attack, it may be necessary to incorporate some negative dihedral (i.e. ~J
anhedral) on a swept-wing aircraft.
:l,
,
[J
4-22 Module 8.4 Flight Stability and Dynamics
!J,-
Use and/or disclosure is
TIS Integrated Training System governed by the statement
© Copyright 201 0 on page 2 of this chapter.
o Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
r '!
I Handling Considerations
LJ It has been shown that the 'dihedral effect' of sweepback in sideslip produces a strong rolling
moment. This has been referred to somewhat imprecisely as roll with yaw. Two applications of
this effect at low speeds, where it is strongest, are worth considering:
• Crosswind Landings. After an approach with the aircraft heading into a cross-wind from
the right, the pilot must yaw the aircraft to port to align it with the runway prior to
touchdown. This action will induce a sideslip to starboard and the pilot must anticipate
the subsequent roll to port in order to keep the wings level.
• Wing Drop. The greater tendency of a swept-wing aircraft to drop a wing at a high angle
of attack (aggravated by a steep curved approach) may be further increased by a large
r: deflection of corrective aileron. In such cases the dihedral effect of sweep back may be
U utilized by applying rudder to yaw the nose towards the high wing - sideslip to the left,
c roll to the right. It must be said, however, that modern design has reduced the tip-stalling
tendency and improved the effectiveness of ailerons at high incidence and the problem is
not as acute as it might have been in the 'transonic era'.
,1
I :
u
n
LJ
I\
o
r1
I !
U
'1
U Module 8.4 Flight Stability and Dynamics
4-23
Use and/or disclosure is TTS Integrated Training System
governed by the statement
(I on page 2 of this chapter.
© Copyright 2010
i i
U
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
WinglFuselage Interference
• Shielding Effect. Most aircraft will be affected by the shielding effect of the fuselage. In
a sideslip the section of the trailing wing near the root lies in the 'shadow' of the fuselage.
The dynamic pressure over this part of the wing may be less than that over the rest of the
wing and therefore produces less lift. This effect will tend to increase the 'dihedral effect'
II,J
--
and on some aircraft may be quite considerable. i.
n
LJ
Drag n
----
LJ
~~--
a HighWing
o
n I,-_,J
n
Drag <::::::::: n
l ,J
CG
---
--
b low Wing and High Fin
~ n
Dragl~ ""
u
iJ
C Tee-Tail
n
Figure 4.15 - Wing-fuselage interference - shielding effect n.,
4-24
n
~ J
Module 8.4 Flight Stability and Dynamics
Use and/or disclosure is
TIS Integrated Training System
© Copyright 2010
governed by the slalement
on page 2 of this chapter. r:
.J
r'
I :
I '
U
Integrated Training System
Designed in associatlon with the
club66pro.co.uk question practice aid
r'
I :
U • Vertical Location A stronger contribution towards lateral stability arises from the
vertical location of the wings with respect to the fuselage, It is helpful to start by
considering the fuselage to be cylindrical in cross-section. The sideslip velocity will flow
around the fuselage, being deflected upwards across the top and downwards
underneath. Superimposing a wing in this flow has the following effect, illustrated above:
o (i) High Wing. A high-mounted wing will lie in a region of upwash on the
up-stream side of the fuselage tending to increase its overall angle of attack.
Conversely, on the down-stream side of the fuselage the wing is influenced by the
downwash tending to reduce its angle of attack. The difference in lift produced by
ri
each wing will cause a restoring moment to increase with sideslip, This effect has
I \ been demonstrated to be equivalent to 1° - 3° of dihedral.
u
o (ii) Low Wing. The effect of locating the wing on the bottom of the fuselage
is to bring it into a region of downwash on the up-stream side and into upwash on
the down-stream side of the fuselage. The angle of attack of the leading (low)
wing will be decreased and that of the trailing wing increased. This gives rise to an
f':
U
unstable moment equivalent to about 1° - 3° anhedral.
From these facts it can be seen that there is zero effect on lateral stability when the wing is
n
u mounted centrally on the fuselage. The effect is lessened as separation occurs at the
wing/fuselage junction.
n
I I Fuselage/Fin Contributions
LJ
Since the aircraft is side-slipping, there will be a component of drag opposing the sideslip
velocity, If the drag line of the aircraft is above the CG the result will be a restoring moment
tending to raise the low wing. This configuration is therefore a contribution towards positive
lateral stability. Conversely, a drag line below the CG will be an unstable contribution. The
position of the drag line is determined by the geometry of the entire aircraft but the major
I'II contributions, illustrated in the previous diagram, are:
U
• High wing,
• Low Wing and High fin and rudder.
• Tee-tail configuration.
The tee-tail configuration makes the fin more effective as well as contributing its own extra drag.
o
r:
u
n
•.• ,J
I
, I
LJ
a Destabilizing Effect of Slipstream
n
f.+--y_ L
n
n
~_, .J
Increase in Uft
due to Dihedral Effect
1
LJ
n
n
\. J
• Slipstream. Due to sideslip the slipstream behind the propeller or propellers is no longer
symmetrical about the longitudinal axis. The dynamic pressure in the slipstream is higher
than the free stream and covers more of the trailing wing in sideslip. The result is an
unstable moment tending to increase the displacement. This unstable contribution is
worse with flaps down. 1
,,
I ,
• Flaps. Partial-span flaps alter the spanwise distribution of pressure across a wing. The
local increase in lift coefficient near the root has the effect of moving the 'half-span' nJ
~
r",
I : centre of pressure towards the fuselage (in a spanwise sense). The moment arm of the
U wing lift is thus reduced and a given change in C L due to the dihedral effect will produce a
smaller moment. The overall lateral stability is therefore reduced by lowering inboard
rl
I , flaps. The design geometry of the flap itself can be used to control this contribution. In
U particular, a swept-back flap hinge-line will decrease the dihedral effect, whereas a
swept-forward hinge-line will increase it.
rl
LJ
Design Problems
It is desirable that an aircraft should have positive lateral static stability. If, however, the stability
is too large, it can lead to the dynamiC problems listed below, some of which are discussed
o later:
•
•
Lateral oscillatory problems i.e. Dutch roll.
Large aileron control deflections and forces under asymmetric conditions.
n
I : • Large rolling response to rudder deflection requiring aileron movement to counteract the
LJ
possibility of 'autorotation' under certain conditions of flight.
o
i'
u
o
rul
n
,
~
I
c
fl
L!
C-1
U
o
o
riI
I
I I
L..J
n
l J
'1
!, ,I
n
lJ
n
:
~ __ ..J
\
Intentionally Blank :1
lJ
o
n
LJ
n
), J
n
U
4-28
TTS Integrated Training System
Module 8.4 Flight Stability and Dynamics
Use andfor disclosure is
governed by the statement
n
© Copyright 20t 0 on page 2 of this chapter. 1
U
u
Cl Integrated Training System
Designed in association with the
U club66pro.co.uk question practice aid
Dynamic Stability
11
u
General
r-' When an aircraft is disturbed from equilibrium, the resulting motion and corresponding changes
U in the aerodynamic forces and moments acting on the aircraft may be quite complicated. This is
especially true for displacement in yaw which affects the aircraft both in the yawing and the
rl rolling planes.
I i.
'-'
Some of the factors affecting the long-term response of the aircraft are as follows:
o then removed. If the subsequent system of forces and moments tends initially t decrease the
displacement, then that body is said to have positive static stability. It may, however, overshoot
the equilibrium condition and then oscillate about it. The terms for possible forms of motion
r' which describe the dynamic stability of the body are listed below:
U • Amplitude increased - negative stability.
• Amplitude constant - neutral stability.
[1
u • Amplitude 'damped' - positive stability.
• Motion heavily damped; oscillations cease and the motion becomes 'dead-beat' positive
stability.
• Motion diverges - negative dynamic stability.
c Figure 4.17 illustrates these various forms of dynamic stability; in each case shown, the body
has positive static stability.
n
I :
U
o
r1u
n
I I
U
--I~
vv
Initial
~ T'-=======-----
Positive Dynamic Stability
n
Disturbance
.Negative Dynamic Stability
(Dead Beat Convergence/Subsidence)
n
n
U
Neutral Dynamic Stability
Negative Dynamic Stability (Divergence)
o
Positive Dynamic Stability (Damped Phugllidi "
LJ
Figure 4.17 - Forms of dynamic stability '1
u
Dynamic stability is more readily understood by use of the analogy of the 'bowl and ball'
described earlier. For example, when the disturbance is removed the ball returns to the bottom
of the bowl and is said to have static stability. However, the ball will oscillate about a neutral or
equilibrium position and this motion is equivalent to dynamic stability in an aircraft.
n
n
If the oscillations are constant in amplitude and time then a graph of the motion would be as
U
o
shown below. The amplitude shows the extent of the motion, and the periodic time is the time
taken for one complete oscillation. This type of motion is known as simple harmonic motion.
Periodic Time
The time taken for one complete oscillation will depend upon the degree of static stability, Le.
the stronger the static stability, the shorter the periodic time.
n
l J
II
• I
II
n
Figure 4.18 - Dynamic Stability
4-30
TIS Integrated Training System
Module 8.4 Flight Stability and Dynamics
Use and/or disclosure is
governed by the statement
n
© Copyright 2010 on page 2 of this chapter. 'I
,j
Integrated Training System
Designed in association with the
club66pro,co.uk question practice aid
Damping
In the simple analogy given it is assumed that there is no damping in the system; the
oscillations will continue indefinitely and at a constant amplitude. In practice, however, there will
always be some damping if only because the viscosity of the fluid (air) is a damping factor which
is proportional to the speed of mass. Damping can be expressed as the time required (or
11 number of cycles) for the amplitude to decay to one half of its initial value (Damped Phugoid).
U An increase in the damping of the system (e.g. from a more viscous fluid) will cause the
oscillations to die away more rapidly and, eventually, the damping will be such that the
o oscillation ceases. In this case, after the disturbance has been removed, the mass returns
slowly towards equilibrium but does not overshoot it, i.e. the motion is 'dead-beat' (Positive
Dynamic Stability).
o Dynamic stability depends on the particular design of the aircraft and the speed and height at
which it is flying. It is usually assumed that for 'conventional' aircraft the coupling between
longitudinal (pitching) and lateral (including directional) motions can be neglected. This enables
Design Specification
o Oscillatory motions which have a long periodic time are not usually important: even if the motion
is not naturally well damped, the pilot can control the aircraft fairly easily. To ensure satisfactory
handling characteristics however, it is essential that all oscillatory motions with a periodic time of
IlL.J the same order as the pilot's response time are heavily damped. This is because the pilot may
get out of phase with the motion and pilot-induced oscillations (PIO) may develop. The minimum
damping specified is that oscillations may decay to one half of their original amplitude in one
c complete cycle of the motion. Some modern aircraft do not satisfy this requirement and in many
cases it has been necessary to incorporate auto stabilization systems such as pitch dampers or
yaw dampers to improve the basic stability of the aircraft.
o Longitudinal Dynamic Stability
rl
LJ When an aircraft is disturbed in pith from trimmed level flight it usually oscillates about the
original state with variations in the values of speed, height and indicated load factor. If the
aircraft has positive dynamic stability, these oscillations will gradually die away and the aircraft
returns to its initial trimmed flight condition. The oscillatory motion in pitch can be shown to
consist of two separate oscillations of widely differing characteristics: the phugoid and the short-
o period oscillation.
o
o
o
rl
U Use and/or disclosure is
Module 8.4 Flight Stability and Dynamics 4-31
TIS Integrated Training System
govemed by the statement
on page 2 of this chapter. © Copyright 2010
.I
~--->
lDisplacement o
~90id o
n
lJ
Short
Period
Oscillation n
Figure 4.19 - Oscillations of Longitudinal Dynamic instability
Phugoid
This is usually a long period, poorly damped motion involving large variations in the speed and
height of the aircraft but with negligible changes in load factor (n). It can be regarded as a
constant energy motion in which potential energy and kinetic energy are continuously
interchanged. The phugoid oscillation is usually damped, and the degree of damping depends
on the drag characteristics of the aircraft. The modern development towards low-drag design
has resulted in the phugoid oscillation becoming more of a problem.
[]
Short-Period Oscillation
This oscillatory motion is usually heavily damped and involves large changes of load factor with
n
only small changes in speed and height. It can be regarded simply as a pitching oscillation with
one degree of freedom. As stated earlier, the time taken for one complete oscillation will depend
upon the static stability, and in this case it is the periodic time of the short-period oscillation.
o
Stability Factors
The longitudinal dynamic stability of an aircraft, that is, the manner in which it returns to a
n
condition of equilibrium, will depend upon:
n
4-32 Module 8.4 Flight Stability and Dynamics iJ
Use and/or disclosure is
TIS Integrated Training System govemed by the statement
© Copyright 2010 on page 2 of this chapter.
o Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
u Short
Period
c iilli
Oscillation
o RAF
\i
U
II!/!
Figure 4.20 - Pitching Moment Curves and Trim Point
o Lateral Dynamic Stability
When an aircraft in trimmed level flight is disturbed laterally, the resulting motion consists of the
C following components:
o •
•
Rolling Motion. Initially the roll will only change the angle of bank, and will be rapidly
damped.
Spiral Motion. A combination of bank and yaw will result in a gradually tightening spiral
o motion if the aircraft is unstable in this mode. The spiral motion is not usually very
important because, even if it s divergent, the rate of divergence is fairly slow and the plot
can control the motion.
o • Dutch Roll. This is an oscillation involving roll, yaw and sideslip. The periodic time is
usually fairly short and the motion may be weakly damped or even undamped. Because
of these characteristics of the Dutch Roll oscillation, lateral dynamic stability has always
Spiral Stability
o The lateral stability of an aircraft depends on the forces that tend to right the aircraft when a
wing drops. At the lime time however, the keel surface (including the fin) tends to yaw the
aircraft into the airflow, in the direction of the lower wing. Once the yaw is started, the higher
c wing, being on the outside of the turn and traveling slightly faster than the lower, produces more
lift. A rolling moment is thereby set up which opposes, and may be greater than, the correcting
moment of the dihedral, since the roll due to yaw will tend to increase the angle of bank.
[i
U If the total rolling moment is strong enough to overcome the restoring force produced by the
dihedral and damping in yaw effects, the angle of bank will increase and the aircraft will enter a
11
u diving turn of steadily increasing steepness. This is known as spiral instability. A reduction in fin
area, reducing g directional stability and the tendency to yaw into the sideslip result in a smaller
o gain in lift from the raised wing and therefore in greater spiral stability.
This form of instability is not very important. Many high performance aircraft when yawed, either
c by prolonged application of rudder or by asymmetric power, will develop a rapid rolling motion in
r1
U Module 8.4 Flight Stability and Dynamics 4-33
Use and/or disclosure is
governed by the statement
TIS Integrated Training System
on page 2 of this chapter. © Copyright 201 0
Ij
Integrated Training System
Designed in association with the
club66pro.co.uk question practice aid
the direction of the yaw and may quickly enter a steep ;spiral dive; this is due to the interaction
of the directional and lateral stability.
n
I... J
Dutch Roll
Oscillatory instability is more serious than spiral instability and is commonly found to a varying
degree in combinations of high wing loading, sweepback (particularly at low lAS) and high
altitude. Oscillatory instability is characterized by the combined rolling and yawing movement or
n
'wallowing' motion. When an aircraft is disturbed laterally the subsequent motion may be either
U
of the two extremes. The aerodynamic causes of oscillatory instability are complicated and a
simplified explanation of one form of Dutch Roll is as follows: n
LJ
Consider a swept-wing aircraft seen in planform. If the aircraft is yawed, say to starboard, the
port wing generates more lift due to the larger expanse of wing presented to the airflow and the
aircraft accordingly rolls in the direction of yaw. However, in this case the advancing port wing
o
also has more drag because of the larger area exposed to the airflow. The higher drag on the
port wing causes a yaw to port which results in the starboard wing obtaining more lift and
rever3ing the direction of the roll. The final result is an undulating motion in the directional and
n
lateral planes which is known as Dutch Roll. Since the motion is caused by an excessive
restoring force, one method of tempering the effects is to reduce the lateral stability by setting nL.J
the wings at a slight anhedral angle.
The lateral dynamic stability of an aircraft is largely decided by the relative effect of:
n
I. ,I
Too much weathercock stability will lead to spiral instability whereas too much dihedral effect
will lead to Dutch Roll instability.
II
l....J
n
n
t J
o
o
o
n
o
n
4-34 Module 8.4 Flight Stability and Dynamics u
Use andfor disclosure is
TIS Integrated Training System governed by the statement
© Copyright 201 0 on page 2 of this chapter.
:1l I
u
Integrated Training System
[J Designed in association with the
club66pro,co.uk question practice aid
u Summary
c (a)
(b)
(c)
Tail area.
Tail volume.
Planform.
o (d)
(e)
Wing downwash.
Distance from CPtail to CG.
n
I--.J• • Position of CG.
11 Manoeuvre stability is greater than the static stability' in level flight and a greater elevator
LJ deflection is necessary to hold the aircraft in a steady pull-out.
(a) Dihedral.
(b) Sweepback.
• Wing/fuselage interference.
• Fuselage and fin contribution.
• Undercarriage, flap and power effects.
•
•
Linear velocity and mass.
The static stabilities in roll, pitch and yaw.
n
•
•
•
Angular velocities about the three axes
Moments of inertia about the three axes
Aerodynamic damping moments due to roll, pitch and yaw.
n
The longitudinal dynamic stability of an aircraft depends upon: n
•
•
•
Static longitudinal stability.
Aerodynamic pitch damping.
Moments of inertia in pitch.
n
•
•
Angle of pitch.
Rate of pitch n
The lateral dynamic stability of an aircraft is largely decided by the relative effect of:
n
• Dihedral effect.
• Weathercock stability.
rI
~
n
n
'.J
n
lJ
n
n
n
4-36 Module 8.4 Flight Stability and Dynamics
Use and/or disclosure is
TIS Integrated Training System
© Copyright 201 0
governed by the statement
on page 2 of this chapter.
n
, j