Swept wing

A swept wing is a wing planform favored for high subsonic jet speeds first investigated
in Germany from 1935 onwards until the end of the Second World War. Since the introduction
of the MiG-15 and North American F-86 which demonstrated a decisive superiority over the
slower first generation of straight-wing jet fighters during the Korean War, swept wings have
become almost universal on all but the slowest jets (such as the A-10). Compared with straight
wings common to propeller-powered aircraft, they have a "swept" wing root to wingtip direction
angled beyond (usually aftward) the spanwise axis. This has the effect of delaying the drag rise
caused by fluid compressibility near the speed of sound as swept wing fighters such as the F-86
were among the first to be able to exceed the speed of sound in a slight dive, and later in level
flight.
Unusual variants of this design feature are forward sweep, variable sweep wings and pivoting
wings. Swept wings as a means of reducing wave drag were first used on jet fighter aircraft. The
four-engine propeller-driven Tu-95 aircraft also has swept wings.
The angle of sweep which characterizes a swept wing is conventionally measured along the 25%
chord line. If the 25% chord line varies in sweep angle, the leading edge is used; if that varies,
the sweep is expressed in sections (e.g., 25 degrees from 0 to 50% span, 15 degrees from 50% to
wingtip). Angle of sweep equals 1/2[180 deg - (nose angle)].
If you're historically inclined, you may be interested to know that the first swept-wing airplane
appeared in 1907. It was built by J. W. Dunne, a gifted Irishman who also had some interesting
ideas about the nature of Time. It had the distinction of being stable in all axes -- uncapsizeable,
in the boat-influenced language of the era. Wing sweep played a part in its stability; it provided a
powerful dihedral effect, tending to roll the airplane out of a sideslip and consequently to keep it
right side up without pilot attention, like a model glider.
Like many aspects of wing shape that designers tried out during the first two or three decades of
powered flight, marked sweep -- the Dunne biplanes had about 30 degrees -- was eventually
discarded. By World War II, the "ideal" wing shape was straight, moderately tapered, with an
aspect ratio of between five and nine, and rounded tips.
At the end of the war, Operation Paperclip, a massive paper chase through abandoned
laboratories that was intended to harvest Germany's technical and scientific secrets, uncovered
documents regarding the application of wing sweep to high-speed aircraft. The information
arrived just in time to allow American companies like North American and Boeing to scrap
conventional wings and adopt swept ones for airplanes like the F-86 Sabre and the B-47
Stratojet. The swept wing proved manifestly superior for airplanes flying faster than about 70
percent of the speed of sound.
Revolutions often turn out to have been annouced a decade or two in advance by people or
events that were promptly forgotten; hence the expression "before his time." Wing sweep had
been proposed in 1935 at an aeronautics conference in Rome by Adolph Busemann, a brilliant
character who pops up here and there, Zelig-like, in the history of aerodynamics before settling
down as a professor at the University of Colorado at Boulder, where he died in 1986. It was a
remark of Busemann's, too, in 1951 that led Richard Whitcomb, one of NASA's most productive
thinkers, to the idea of the "transonic area rule," which has been a staple of aerodynamics ever
since.
Since the postwar period, we have come to associate swept wings with fast airplanes. They just
look right -- as though they were being pulled backward by the sheer force of the wind. Through
magical thinking, marketing departments have found in swept vertical tails a solution to the
problem of making slow airplanes appear fast.
The real function of wing sweep is somewhat more complex -- a bit mystifying, perhaps, but not
magical.
Below around 400 knots, the drag of a typical clean airplane increases in the expected way, as a
function of speed alone. As speed continues to increase, however, shock waves begin to form,
notably on wings and canopies. Shock waves are sudden discontinuities in the temperature and
pressure of air, and they behave somewhat like large spoilers, greatly increasing the size of the
airplane's disturbed wake. The result is an extremely rapid increase in drag. The speed at which
shock waves begin to appear is called the "critical Mach number." It is not an absolute barrier to
further acceleration, but no airplane of the immediate postwar period -- engine power was still
quite limited -- could achieve a top speed much higher than its critical Mach number, because the
drag increase was as steep as a cliff.
Efforts to "break the sound barrier" emphasized wing design, because the wing generated the
most extensive shock waves. The first airplane to do the trick, the Bell X-1, had a wing that was,
for the time, unusually thin. Its thinness delayed the formation of shocks and the consequent drag
rise. An airfoil whose thickness is one-sixth of its chord might have a critical Mach number of
.72, while the same profile, but with a thickness one-tenth of the chord, would have a critical
Mach number of .8.

Subsonic and transonic behavior


In the transonic the swept wing also sweeps the shock which is at the top rear of the wing. Only the
velocity component perpendicular to the shock is affected.
As an aircraft enters the transonic speeds just below the speed of sound, an effect known as wave
drag starts to appear. Using conservation of momentum principles in the direction normal to
surface curvature, airflow accelerates around curved surfaces, and near the speed of sound the
acceleration can cause the airflow to reach supersonic speeds. When this occurs, an oblique
shock wave is generated at the point where the flow slows down back to subsonic speed. Since
this occurs on curved areas, they are normally associated with the upper surfaces of the wing, the
cockpit canopy, and the nose cone of the aircraft, areas with the highest local curvature.
Shock waves require energy to form. This energy is taken out of the aircraft, which has to supply
extra thrust to make up for this energy loss. Thus the shocks are seen as a form of drag. Since the
shocks form when the local air velocity reaches supersonic speeds over various features of the
aircraft, there is a certain "critical mach" speed (or drag divergence mach number) where this
effect becomes noticeable. This is normally when the shocks start generating over the wing,
which on most aircraft is the largest continually curved surface, and therefore the largest
contributor to this effect.
One of the simplest and best explanations of how the swept wing works was offered by Robert T.
Jones: "Suppose a cylindrical wing (constant chord, incidence, etc.) is placed in an airstream at
an angle of yaw - ie., it is swept back. Now, even if the local speed of the air on the upper surface
of the wing becomes supersonic, a shock wave cannot form there because it would have to be a
sweptback shock - swept at the same angle as the wing - ie., it would be an oblique shock. Such
an oblique shock cannot form until the velocity component normal to it becomes supersonic."
[1]

One limiting factor in swept wing design is the so-called "middle effect". If a swept wing is
continuous - an oblique swept wing, the pressure iso-bars will be swept at a continuous angle
from tip to tip. However, if the left and right halves are swept back equally, as is common
practice, the pressure iso-bars on the left wing in theory will meet the pressure iso-bars of the
right wing on the centerline at a large angle. As the iso-bars cannot meet in such a fashion, they
will tend to curve on each side as the near the centerline, so that the iso-bars cross the centerline
at right angles to the centerline. This causes an "unsweeping" of the iso-bars in the wing root
region. To combat this unsweeping, German aerodynamicist Dietrich Kchemann proposed and
had tested a local indentation of the fuselage above and below the wing root. This proved to not
be very effective.
[2]
During the development of the Douglas DC-8 airliner, uncambered airfoils
were used in the wing root area to combat the unsweeping.
[3][4]
Similarly, a decambered wing
root glove was added to the Boeing 707 wing to create the Boeing 720.
[5]

Swept wings for the transonic range
-
Tu-95 propeller-driven bomber with swept wings, cruise speed 710 km/h
-
KC-10 Extender, cruise speed: 908 km/h
-
HFB-320 Hansa Jet with forward swept wings, cruise speed: 825 km/h
-
Planform view of the swept wing and tailplane of a Virgin Atlantic Airbus A340-600
takeoff. The undercarriage are still retracting
Supersonic behavior


At supersonic speeds an oblique shock exists in front of the leading edge of the wing. The velocity
component perpendicular to the shock is different upstream and downstream of the shock. The velocity
component parallel to the shock is the same on both sides of the shock.


The Convair F-106 Delta Dart is optimized for supersonic flight and has a highly swept delta wing.


Grumman F-14 Tomcat, an example of a variable-geometry aircraft, shown in the high-sweepback
configuration that is the optimum for supersonic speeds. The wings lie behind the shock cone generated
in supersonic flight.
Airflow at supersonic speeds generates lift through the formation of shock waves, as opposed to
the patterns of airflow over and under the wing. These shock waves, as in the transonic case,
generate large amounts of drag. One of these shock waves is created by the leading edge of the
wing, but contributes little to the lift. In order to minimize the strength of this shock it needs to
remain "attached" to the front of the wing, which demands a very sharp leading edge. To better
shape the shocks that will contribute to lift, the rest of an ideal supersonic airfoil is roughly
diamond-shaped in cross-section. For low-speed lift these same airfoils are very inefficient,
leading to poor handling and very high landing speeds.
[6]

One way to avoid the need for a dedicated supersonic wing is to use a highly swept subsonic
design. Airflow behind the shock waves of a moving body are reduced to subsonic speeds. This
effect is used within the intakes of engines meant to operate in the supersonic, as jet engines are
generally incapable of ingesting supersonic air directly. This can also be used to reduce the speed
of the air as seen by the wing, using the shocks generated by the nose of the aircraft. As long as
the wing lies behind the cone-shaped shock wave, it will "see" subsonic airflow and work as
normal. The angle needed to lie behind the cone increases with increasing speed, at Mach 1.3 the
angle is about 45 degrees, at Mach 2.0 it is 60 degrees.
[7]
For instance, at Mach 1.3 the angle of
the Mach cone formed off the body of the aircraft will be at about sin = 1/M ( is the sweep
angle of the Mach cone)
[8]

Generally it is not possible to arrange the wing so it will lie entirely outside the supersonic
airflow and still have good subsonic performance. Some aircraft, like the English Electric
Lightning or Convair F-106 Delta Dart are tuned entirely for high-speed flight and feature highly
swept planforms without regard to the low-speed problems this creates. In other cases the use of
variable geometry wings, as on the Grumman F-14 Tomcat, allows an aircraft to move the wing
to keep it at the most efficient angle regardless of speed, although the cost in complexity and
weight makes this a rare feature.
Most high-speed aircraft have a wing that spends at least some of its time in the supersonic
airflow. But since the shock cone moves towards the fuselage with increased speed (that is, the
cone becomes narrower), the portion of the wing in the supersonic flow also changes with speed.
Since these wings are swept, as the shock cone moves inward, the lift vector moves forward as
the outer, rearward portions of the wing are generating less lift. This results in powerful pitching
moments and their associated required trim changes.
Disadvantages


Spanwise flow of the boundary layer
When a swept wing travels at high speed, the airflow has little time to react and simply flows
over the wing almost straight from front to back. At lower speeds the air does have time to react,
and is pushed spanwise by the angled leading edge, towards the wing tip. At the wing root, by
the fuselage, this has little noticeable effect, but as one moves towards the wingtip the airflow is
pushed spanwise not only by the leading edge, but the spanwise moving air beside it. At the tip
the airflow is moving along the wing instead of over it, a problem known as spanwise flow.
The lift from a wing is generated by the airflow over it from front to rear. With increasing span-
wise flow the boundary layers on the surface of the wing have longer to travel, and so are thicker
and more susceptible to transition to turbulence or flow separation, also the effective aspect ratio
of the wing is less and so air "leaks" around the wing tips reducing their effectiveness. The
spanwise flow on swept wings produces airflow that moves the stagnation point on the leading
edge of any individual wing segment further beneath the leading edge, increasing effective angle
of attack of wing segments relative to its neighbouring forward segment. The result is that wing
segments farther towards the rear operate at increasingly higher angles of attack promoting early
stall of those segments. This promotes tip stall on back swept wings, as the tips are most
rearward, while delaying tip stall for forward swept wings, where the tips are forward. With both
forward and back swept wings, the rear of the wing will stall first. This creates a nose-up
pressure on the aircraft. If this is not corrected by the pilot it causes the plane to pitch up, leading
to more of the wing stalling, leading to more pitch up, and so on. This problem came to be
known as the Sabre dance in reference to the number of North American F-100 Super Sabres
that crashed on landing as a result.
The solution to this problem took on many forms. One was the addition of a fin known as a wing
fence on the upper surface of the wing to redirect the flow to the rear (see the MiG-15 as an
example.) Another closely related design was addition of a dogtooth notch to the leading edge
(Avro Arrow). Other designs took a more radical approach, including the Republic XF-91
Thunderceptor's wing that grew wider towards the tip to provide more lift at the tip. The Handley
Page Victor had a planform based on a crescent compound sweep or scimitar wing that had
substantial sweep-back near the wing root where the wing was thickest, and progressively
reducing sweep along the span as the wing thickness reduced towards the tip.
Modern solutions to the problem no longer require "custom" designs such as these. The addition
of leading edge slats and large compound flaps to the wings has largely resolved the issue. On
fighter designs, the addition of leading edge extensions, included for high maneuverability, also
serve to add lift during landing and reduce the problem.
The swept wing also has several more problems. One is that for any given length of wing, the
actual span from tip-to-tip is shorter than the same wing that is not swept. Low speed drag is
strongly correlated with the aspect ratio, the span compared to chord, so a swept wing always has
more drag at lower speeds. Another concern is the torque applied by the wing to the fuselage, as
much of the wing's lift lies behind the point where the wing root connects to the plane. Finally,
while it is fairly easy to run the main spars of the wing right through the fuselage in a straight
wing design to use a single continuous piece of metal, this is not possible on the swept wing
because the spars will meet at an angle.
Forward sweep
Main article: Forward-swept wing


LET L-13 two-seat glider showing forward swept wing


Grumman X-29 experimental aircraft, an extreme example of a forward swept wing
Sweeping a wing forward has approximately the same effect as rearward in terms of drag
reduction, but has other advantages in terms of low-speed handling where tip stall problems
simply go away. In this case the low-speed air flows towards the fuselage, which acts as a very
large wing fence. Additionally, wings are generally larger at the root anyway, which allows them
to have better low-speed lift.
However, this arrangement also has serious stability problems. The rearmost section of the wing
will stall first causing a pitch-up moment pushing the aircraft further into stall similar to a swept
back wing design. Thus swept-forward wings are unstable in a fashion similar to the low-speed
problems of a conventional swept wing. However unlike swept back wings, the tips on a forward
swept design will stall last, maintaining roll control.
Forward-swept wings can also experience dangerous flexing effects compared to aft-swept wings
that can negate the tip stall advantage if the wing is not sufficiently stiff. In aft-swept designs,
when the airplane maneuvers at high load factor the wing loading and geometry twists the wing
in such a way as to create washout (tip twists leading edge down). This reduces the angle of
attack at the tip, thus reducing the bending moment on the wing, as well as somewhat reducing
the chance of tip stall.
[9]
However, the same effect on forward-swept wings produces a wash-in
effect which increases the angle of attack promoting tip stall.
Small amounts of sweep do not cause serious problems, and had been used on a variety of
aircraft to move the spar into a convenient location, as on the Junkers Ju 287 or HFB-320 Hansa
Jet. But larger sweep suitable for high-speed aircraft, like fighters, was generally impossible until
the introduction of fly by wire systems that could react quickly enough to damp out these
instabilities. The Grumman X-29 was an experimental technology demonstration project
designed to test the forward swept wing for enhanced maneuverability in 1984. The Su-47
Berkut is another notable example using this technology. However no highly swept-forward
design has entered production.
History


A Burgess-Dunne aircraft showing the high angle of sweep.
The first aircraft with swept wings were those designed by the British designer J.W.Dunne in the
first decade of the 20th century. Dunne successfully employed severely swept wings in his
tailless aircraft as a means of creating positive longitudinal static stability.
[10]
Historically, many
low-speed aircraft have had swept wings in order to avoid problems with their center of gravity,
to move the wing spar into a more convenient location, or to improve the sideways view from the
pilot's position. For instance, the Douglas DC-3 had a slight sweep to the leading edge of its
wing. The wing sweep in low-speed aircraft was not intended to help with transonic
performance, and although most have a small amount of wing sweep they are rarely described as
swept wing aircraft. The Curtiss XP-55 was the first American swept wing airplane, although it
was not considered successful.
[citation needed]
The swept wing had appeared before World War I,
conceived as a means of permitting the design of safe, stable, and tailless flying wings. It
imposed self-damping inherent stability upon the flying wing, and, as a result, many flying
wing gliders and some powered aircraft appeared in the interwar years.
[11]

Introduction
The idea of using swept wings to reduce high-speed drag was first developed in Germany in the
1930s. At a Volta Conference meeting in 1935 in Italy, Dr. Adolf Busemann suggested the use of
swept wings for supersonic flight. He noted that the airspeed over the wing was dominated by
the normal component of the airflow, not the freestream velocity, so by setting the wing at an
angle the forward velocity at which the shock waves would form would be higher (the same had
been noted by Max Munk in 1924, although not in the context of high-speed flight).
[12]
Albert
Betz immediately suggested the same effect would be equally useful in the transonic.
[13]
After
the presentation the host of the meeting, Arturo Crocco, jokingly sketched "Busemann's airplane
of the future" on the back of a menu while they all dined. Crocco's sketched showed a classic
1950's fighter design, with swept wings and tail surfaces, although he also sketched a swept
propeller powering it.
[12]

Hubert Ludewieg of the High-Speed Aerodynamics Branch at the AVA Gttingen in 1939
conducted the first wind tunnel tests to investigate Busemann's theory.
[2]
Two wings, one with no
sweep, and one with 45 degrees of sweep were tested at Mach numbers of 0.7 and 0.9 in the 11 x
13 cm wind tunnel. The results of these tests confirmed the drag reduction offered by swept
wings at transonic speeds.
[2]
The results of the tests were communicated to Albert Betz who then
passed them on to Willy Messerschmitt in December 1939. The tests were expanded in 1940 to
include wings with 15, 30 and -45 degrees of sweep and Mach numbers as high as 1.21.
[2]

At the time, however, there was no way to power an aircraft to these sorts of speeds, and even
the fastest aircraft of the era were only approaching 400 km/h (249 mph). Large engines at the
front of the aircraft made it difficult to obtain a reasonable fineness ratio, and although wings
could be made thin and broad, doing so made them considerably less strong. The British
Supermarine Spitfire used as thin a wing as possible for lower high-speed drag, but later paid a
high price for it in a number of aerodynamic problems such as control reversal. German design
instead opted for thicker wings, accepting the drag for greater strength and increased internal
space for landing gear, fuel and weapons.
At the time the presentation was largely of academic interest, and soon forgotten. Even notable
attendees including Theodore von Krmn and Eastman Jacobs did not recall the presentation 10
years later when it was re-introduced to them.
[12]
Buseman was in charge of aerodynamics
research at Braunschweig, and in spite of the limited interest he began a research program
studying the concept. By 1939 wind tunnel testing had demonstrated the effect was real, and
practical.
With the introduction of jets in the later half of World War II applying sweep became relevant.
The German jet-powered Messerschmitt Me 262 and rocket-powered Messerschmitt Me 163
suffered from compressibility effects that made them very difficult to control at high speeds. In
addition, the speeds put them into the wave drag regime, and anything that could reduce this drag
would increase the performance of their aircraft, notably the notoriously short flight times
measured in minutes. This resulted in a crash program to introduce new swept wing designs,
both for fighters as well as bombers. The Focke-Wulf Ta 183 was a swept wing fighter design
with a layout very similar to that later used on the MiG-15 that was not produced before war's
end.
A prototype test aircraft, the Messerschmitt Me P.1101, was built to research the tradeoffs of the
design and develop general rules about what angle of sweep to use. None of the fighter or
bomber designs were ready for use by the time the war ended, but the P.1101 was captured by
US forces and returned to the United States, where two additional copies with US built engines
carried on the research as the Bell X-5. The last jet fighter designed by Willy Messerschmitt the
HA-300 had swept wings, Delta Wing in this case.
Technology impact
The Soviet Union was intrigued about the idea of swept wings on aircraft at the end of World
War II in Europe, when their "captured aviation technology" counterparts to the western Allies
spread out across the defeated Third Reich. Artem Mikoyan was asked by the Soviet
government, principally by the government's TsAGI aviation research department, to develop a
test-bed aircraft to research the swept wing ideathe result was the late 1945-flown, unusual
MiG-8 Utka pusher canard layout aircraft, with its rearwards-located wings being swept back for
this type of research. When applied to the jet-powered Mig-15, its maximum speed of
1,075 km/h (668 mph) outclassed the straight-winged American jets and piston-engined fighters
first deployed to Korea.
von Krmn travelled to Germany near the end of the war as part of Operation Paperclip, and
reached Braunschweig on May 7, discovering a number of swept wing models and a mass of
technical data from the wind tunnels. One member of the US team was George S. Schairer, who
was at that time working at the Boeing company. He immediately forwarded a letter to Ben Cohn
at Boeing stating that they needed to investigate the concept. He also told Cohn to distribute the
letter to other companies as well, although only Boeing and North American made immediate
use of it.
In February 1945, NACA engineer Robert T. Jones started looking at highly swept delta wings
and V shapes, and discovered the same effects as Busemann. He finished a detailed report on the
concept in April, but found his work was heavily criticised by other members of NACA Langley,
notably Theodore Theodorsen, who referred to it as "hocus-pocus" and demanded some "real
mathematics".
[12]
However, Jones had already secured some time for free-flight models under the
direction of Robert Gilruth, whose reports were presented at the end of May and showed a
fourfold decrease in drag at high speeds. All of this was compiled into a report published on June
21, 1945, which was sent out to the industry three weeks later.
[14]
Ironically, by this point
Busemann's work had already been passed around.
Boeing was in the midst of designing the Boeing B-47 Stratojet, and the initial Model 424 was a
straight-wing design similar to the B-45, B-46 and B-48 it competed with. A recent design
overhaul completed in June produced the Model 432, another four-engine design with the
engines buried in the fuselage to reduce drag, and long-span wings that gave it an almost glider-
like appearance. By September the Braunschweig data had been worked into the design, which
re-emerged as the Model 448, a larger six-engine design with more robust wings swept at about
35 degrees.
[12]
Another re-work in November moved the engines into strut-mounted pods under
the wings since Boeing was concerned that the uncontained failure of an internal engine could
potentially destroy the aircraft. With the engines mounted away from the wings on struts
equipped with fuse pins, an out-of-balance engine would simply shatter the pins and fall
harmlessly away, sparing the aircraft from destructive vibrations. The resulting B-47 design had
performance rivaling the fastest fighters and trounced the straight-winged competition. Boeing's
winning jet-transport formula of swept wings and engines mounted on pylons under the wings
has since been universally adopted.
In fighters, North American Aviation was in the midst of working on a straight-wing jet-powered
naval fighter then known as the FJ-1. It was submitted it to the Air Force as the F-86. Larry
Green, who could read German, studied the Busemann reports and convinced management to
allow a redesign starting in August 1945.
[12]
A battery of wind tunnel tests followed, and
although little else of the design was changed, including the wing profile (NACA 0009), the
performance of the aircraft was dramatically improved over straight-winged jets. With the
appearance of the Mig-15, the F-86 was rushed into combat and straight-wing jets like the
Lockheed P-80 Shooting Star and Republic F-84 Thunderjet were soon relegated to ground
attack. Some such as the F-84 and Grumman F-9 Cougar were later redesigned with swept wings
from straight-winged aircraft. Later planes such as the North American F-100 Super Sabre would
be designed with swept wings from the start, though additional innovations such as the
afterburner, area-rule and new control surfaces would be necessary to master supersonic flight.
Swept Wings and Effective Dihedral
One facet of tailless aircraft performance has always intrigued us the ability of swept
wing
tailless sailplanes to travel at high speed without exhibiting Dutch roll, yet demonstrate
excellent
spiral stability while thermalling. This differs from what is seen in high performance
conventional
tailed sailplanes.
The designer of a conventional cross-tailed competition F3B machine must very
carefully balance
wing dihedral and vertical stabilizer surface area. There is a tendency to Dutch roll at
high speed,
and opposite aileron must be applied during thermal turns to prevent a spiral dive, even
when the
aircraft is optimized.
Since both vertical stabilizer area and geometric wing dihedral are held constant during
ight,
what is it about swept wings which allows them to violate the rules?
To begin, we need to go over the fundamentals, and so Parts 1 and 2 of this four part
series will be
devoted to explaining effective dihedral itself how it is derived and how it inuences
aircraft
stability.
Pitch, Yaw, and Roll
Diagrams of aircraft in which the three rotational axes are noted can be found in most
aerodynamics textbooks. Our rendition is included here as Figure 1. In simple terms, the
nose of
the aircraft can move up and down through an axis which parallels the wing span (pitch,
Y axis),
and it can move right and left through a vertical axis which passes down through the
fuselage in
the region of the wing (yaw, Z axis). The aircraft can be also be made to rotate around
an axis
which roughly goes through the nose and tail cone (roll, X axis).
Elevator deection changes the camber of the horizontal tail (stabilizer and elevator),
increasing
lift in the direction opposite to elevator deection. By moving the elevator up or down,
the
aircraft tail may be lowered or raised, thus raising or lowering the nose. This is a change
in pitch
(Y axis). The size of the elevator and the distance between the center of gravity (CG)
and the
elevator determine elevator power. The larger the elevator surface area and the larger
the distance
between the CG and the elevator, the more elevator power.
If the elevator is deected upward, the aircraft tail is pushed downward and the nose is
thus
raised. It must be clearly understood that the elevator controls the wing angle of attack,
that is the
angle of the wing to the freestream airow. A larger upward deection of the elevator
will place
the wing at a higher angle of attack. However, when elevator deection is neutralized,
the wing
will return to its normal angle of attack. This is because the horizontal tail acts as a
longitudinal
stabilizer.
The rudder, when deected, pushes the tail either right or left. Rudder deection
changes the
camber of the vertical tail, increasing lift in the direction opposite to rudder deection.
When the
rudder is deected to the right, for example, the tail swings to the left and the nose
swings to the
right, inducing yaw, a rotation around the Z axis.
EffectiveDihedralRevB.FM5
4
Larger deections of the rudder will result in greater yaw angles, but the aircraft will
return to
zero degrees yaw when the rudder is neutralized. The vertical tail thus acts as a
directional
stabilizer.
Aileron deection changes the camber of the wing in the region of the aileron. Upward
deection
reduces lift, and further deection may cause negative lift. As one aileron deects
upward, the
opposite aileron deects downward. Downward deection increases wing lift in the
region of the
aileron. These changes in lift promote the lifting of the wing for which the aileron is
deected
downward, and the depression of the wing for which the aileron is deected upward.
These
coordinated actions induce roll, a rotation around the X axis.
PITCH
ROLL
YAW
FLIGHT
DIRECTION
Y X
Z
+
+
+
u
v
w
r
q
p
CG
Axis Linear
velocity
Angular
displacement
Angular
velocity
X (roll) u
|
p
Y (pitch) v
u
q
Z (yaw) w

r
Figure 1, rotational axes
EffectiveDihedralRevB.FM5
5
While increased elevator deflection increases the angle of pitch and hence the wing
angle of
attack, and increased rudder deflection increases the angle of yaw, increased aileron
deflection
increases the rate of roll. Holding a specific wing angle of attack or aircraft angle of yaw
is
accomplished by steady continued deflection of the relevant control surface, elevator or
rudder.
Holding a specific angle of bank, however, dictates neutralizing the ailerons. There is no
restoring
moment when aileron deflection is neutralized and so the wings do not return to level.
Instead, the
roll rate reduces to zero and the bank angle is maintained.
The horizontal and vertical stabilizers are thus direct mechanism for returning the
aircraft to its
equilibrium attitude in pitch and yaw, but there is no equivalent mechanism for roll.
Lateral Stability
The above discussion is somewhat simplistic in that roll never takes place by itself.
Rather, roll
always causes some other motion. In straight and level ight, lift equals aircraft mass.
See
Figure 2.
First, and most important, roll causes a sideslip. When there is no pitch input to change
the angle
of attack of the wing, and the airplane is rolled to a specic angle of bank, the vertical
lift
component is less than the mass of the aircraft. A sideward component is at the same
time acting
in the direction of the eventual turn. See Figure 3. The aircraft drops and its velocity
increases so
the vertical lift component will equal the mass. The aircraft accelerates into a circular
path
because of the sideward lift component. See Figure 4.
When this process is initiated the aircraft is banked but the lift is too small, through
lack of
airspeed or lack of C
L
a component of the mass acts along the wingspan and the aircraft
sideslips. It is important to recognize that no component of lift acts along the span, so
the lift does
not cause the sideslip. Once the turn is established, the lift and centrifugal effects
balance the mass,
but initially it is the unbalanced mass which causes the sideslip.
As an example, if the aircraft rolls to the right, the aircraft will sideslip to the right. This is
equivalent to yawing the aircraft to the left, as the air is coming toward the right side of
the
aircraft. Such a sideslip to the right may thus be caused by either an initial roll to the
right, or by
moving the rudder to the left.
If the wing has some dihedral (wing tips higher than wing roots), the sideslip creates a
situation in
which the lower wing, which is moving into the sideslip, meets the oncoming air at a
greater angle
of attack than the higher wing. This generates a restoring force. It should also be noted
that the
lower wing, because it is operating at a greater angle of attack, is generating more drag
than the
higher wing. This creates a yaw to the left, thus reducing the sideslip. The result of all of
these
effects is to both reduce the sideslip and restore the wings to level.
An aircraft is said to have static lateral stability if the right wing rolls upward when there
is a
sideslip to the right. Because dihedral causes this response to a sideslip, lateral stability
is
sometimes referred to as dihedral effect.
Another contributor to lateral stability is the position of the wing on the fuselage. If the
wing is
mounted on the top of the fuselage, any sideslip will change the ow of the air over the
fuselage in
EffectiveDihedralRevB.FM5
6
such a way that the low wing, which is moving into the sideslip, will be operating at a
higher
effective angle of attack. The raised wing, on the other hand, will be operating at a lower
effective
angle of attack. When the wing is located on the top of the fuselage, the wing acts as if
it has some
positive geometric dihedral. This is because of the airow over the top of the fuselage.
See
Figure 5.
If, on the other hand, the wing is mounted on the bottom of the fuselage, the wing
moving into the
sideslip operates at a slightly lower effective angle of attack, while the raised wing
operates at a
slightly higher effective angle of attack. This is due to the airow over the bottom of the
fuselage,
and accounts for the observation that high wing monoplanes have very little dihedral,
some times
none at all, while low wing monoplanes have obviously steep dihedral angles.
LIFT
MASS
Figure 2, mass and lift vectors
while in level ight
RELATIVE WIND STRIKES BOTTOM
OF THIS WING, INCREASING ITS
ANGLE OF ATTACK
RELATIVE WIND STRIKES
SIDE OF VERTICAL
STABILIZER
CHANGING ITS
ANGLE OF ATTACK
RELATIVE WIND
STRIKES TOP OF THIS
WING,
DECREASING
ITS ANGLE
OF ATTACK
ACTUAL LIFT
MASS
Figure 3, initial mass and lift vectors while banked, and effects of relative wind
LIFT
MASS
Figure 4, mass and lift vectors
while in established turn
REQUIRED LIFT
APPARENT LIFT
SIDESLIP
EffectiveDihedralRevB.FM5
7
As the aircraft rolls to the right, the angle of attack of the left wing is immediately
reduced
because of the airow coming from above. The right wing, on the other hand, operates
at a higher
angle of attack because the airow is now coming from the bottom. There is an
accompanying
differential drag effect as well. These effects take place only while the aircraft is actually
rolling,
however, and hence have a damping rather than a corrective action. That is, if the
wings are held
at a constant bank angle there is no vertical velocity on the wings, and no damping
forces are
generated. It should also be noted that if the angle of attack is such that the wing is near
stall, the
downgoing wing may in fact stall, thus reducing lift precipitously and allowing the
already
lowered wing to fall.
Figure 5, effect of fuselage on airow over wing during sideslip
Figure 6, air ow during sideslip
Direction of ight
Air ow
Air ow
for swept wing
EffectiveDihedralRevB.FM5
8
A swept wing, without fuselage or tail, tends toward lateral stability. This is because the
air more
directly meets the wing which is moving into the sideslip, creating more lift and, as
stated
previously for conventional aircraft wings, more drag. Both of these actions result in
restoring
moments. See Figure 6.
A small restoring force in roll is also generated by the vertical tail if it protrudes from the
upper
part of the fuselage. See Figure 3.
Yaw-Roll Coupling
Having said that a roll causes a sideslip and then a yaw, it is imperative to note that the
reverse is
also true a yaw causes a sideslip and then a roll. The coupling of roll, yaw, and
sideslip cannot
be separated.
As an indication of these interactions, place a model aircraft with dihedral on a wire
such that the
wire goes directly from nose to tail through the CG. If the model is canted (rolled)
slightly and
then pushed forcibly down the wire, the model will not right itself. This is because
without
sideslip there is no restoring force, only damping forces, as explained in the previous
section. In
fact, then, it is the interaction of roll, sideslip and yaw which allows the designer to
produce a
stable aircraft.
Aircraft controlled by rudder only (no ailerons) must have substantial dihedral. As the
rudder is
deected, the effective angle of attack of the outboard wing is increased, generating a
rolling
motion into the turn. Without dihedral, rudder deection would simply yaw the aircraft,
and no
roll component would be generated. If the dihedral angle and vertical tail area are
correct, the
aircraft will continue to circle at a constant bank angle once the rudder is returned to the
neutral
position. (See also End Note 1, Aileron drag and adverse yaw.)
We often think adverse yaw is the direct result of aileron deection. Because aileron
deection
changes wing camber, it is easy to see that this differentially affects the drag of the
wings. The
aircraft tends to yaw toward the downward deected aileron, an action opposite to what
is
wanted. Rudder deection must be imparted to counteract this adverse yaw,
substantially
increasing drag during turns.
But there is a second, and more important reason for the appearance of adverse yaw.
Well let Steve
Morris explain...
Adverse yaw is affected by a drag imbalance between the wings while rolling, but
it is not only the drag of aileron deflection that causes this. Imagine the wing is
balanced in a perfect rolling motion (constant roll rate) with the ailerons deflected
and the rolling motion fully developed so that the span loading is identical to before
the maneuver began. This implies the ailerons changed the loading to cause the roll
and the motion of the wing induces angle of attack changes that cancel this roll
torque when a steady-state roll rate is achieved. The downward moving wing sees
a greater angle of attack due to the rolling motion (greatest at the wing tip) and the
upward moving wing sees a lower angle of attack. The local lift vectors are tilted
forward on the downward moving wing and backward on the upward wing by the
rolling motion. This change in lift direction due to the induced velocities of rolling
EffectiveDihedralRevB.FM5
9
produces powerful adverse yaw regardless of profile drag changes. The same
induced velocities tilt the lift vectors on birds when they flap their wings and this is
how they get thrust on the downstroke. Aileron drag is a much smaller part of the
adverse yaw than this induced effect.
Directional Stability and Spiral Instability
In RC sailplanes, if vertical stabilizer area is too large, the pilot must continuously input
rudder
controls to maintain the correct heading. This is because the aircraft will always tend to
turn into
any sideslip, and bank more steeply. If corrective commands are not made, the yaw will
increase
gradually to the point the aircraft is in a spiral an increasing turn of decreasing radius
with the
nose pointing downward and speed rapidly increasing and altitude rapidly decreasing.
See
Figure 7.
A similar effect takes place if the vertical n is too large and there is a slight roll. If, for
example,
the roll is to the right, there is a sideslip to the right. The n, being too large, reacts to
the yaw by
turning the aircraft to the right and increasing the bank angle. With a steeper bank
angle, sideslip
and yaw moment both increase. As in the above paragraph, the aircraft will quickly be in
spiral
toward the ground. In both cases, the aircraft has too much directional stability, making
it prone to
spiral instability. See spiral divergence in Figure 7.
Steve Morris brought another reason for spiral divergence to our attention. He said,
Spiral divergence is also strongly affected by C
lr
, roll torque due to yaw rate,
which is a very strong effect on long-span airplanes which fly slowly. The inboard
wing is flying very slow compared to the outboard wing when turning and this
velocity difference, due to yaw rate, causes the lift across the wing to vary, and the
aircraft rolls into the turn. C
lr
is a big effect on human powered aircraft, hang
gliders, sailplanes, and model sailplanes, especially high aspect ratio slow flyers.
Decreasing vertical stabilizer size decreases directional stability and increases spiral
stability.
Increasing dihedral effect has a similar action. But it should be noted that most full size
aircraft
are at least somewhat prone to spiral divergence. That is, the aircraft has too much
directional
stability, and the pilot is nearly always making very small corrections to prevent a spiral.
Why this
is so will be clear shortly.
Too much dihedral causes its own problems. If dihedral is excessive, crosswind
landings are more
difcult as the aircraft is operating in a sideslip, and lateral gusts may induce large roll
angles and
high roll rates. And there is also an increased tendency toward the notorious Dutch roll,
especially
at high speeds.
EffectiveDihedralRevB.FM5
10
SPIRAL
DIVERGENCE
DUTCH
ROLL
Skidding to right,
roll left
Ever tightening
spiral dive
Sharp gust to right
Inclined lift to left,
skidding to left
Roll r ight, yaw left,
left wing moving forward.
Initial direction of ight
Roll left, yaw right,
right wing moving forward
Inclined lift to r ight,
start skidding right
Yaw left, roll left
Roll left, yaw right
Figure 7, spiral divergence and Dutch roll
EffectiveDihedralRevB.FM5
11
Dutch Roll
A small gust hitting the wing or vertical tail can initiate a rather complex oscillatory
motion
consisting of roll and yaw which are out of phase with each other. First there is a yaw
and roll in
one direction, then recovery with an
overshoot to yaw and roll in the opposite direction. From the rear, the tail cone traces a
circular
arc. See Figure 7 and Figure 8.
While not inherently dangerous, Dutch roll produces a tremendous amount of drag. In
conventional tailed RC sailplanes, Dutch roll most often occurs while
flying at high speed. Dutch roll is caused by too much dihedral and insufficient vertical
stabilizer
area. There is too much spiral stability and insufficient directional stability.
The cure for spiral divergence, reducing vertical stabilizer area and/or increasing
dihedral, thus
makes the aircraft more prone to Dutch roll. The cure for Dutch roll, increasing vertical
stabilizer
size and/or reducing dihedral, makes the aircraft more directionally stable and more
prone to
spiral instability and spiral divergence.
As is usual when designing aircraft, some compromise must be made, and the aircraft
then
designed around what is seen as the best overall performance.
Figure 8, Dutch roll
EffectiveDihedralRevB.FM5
12
The Effects of Taper Ratio and Aspect Ratio on Effective Dihedral
Taper ratio affects effective dihedral. Both the included chart and this short discussion
assume
taper ratio to be defined as tip chord/root chord.
From McCormick, page 544, for a linearly tapered wing, the increment for dihedral angle
is:
where
C
l
= the rate of change of the rolling moment with sideslip angle (simply put, one degree of
effective dihedral is equivalent to a C
l
of -0.0021);
a
= the slope of the wing lift curve slope in radians, adjusted for aspect ratio

= the taper ratio (no taper = 1, sharp tip = 0.0);
I
= the wing dihedral angle in radians, small angles only.
The relationship of taper ratio and effective dihedral is graphically illustrated in the
differences
between Graphs 1 and 2. Note that effective dihedral is always somewhat larger than
geometric
dihedral. For the purposes of these graphs, the slope of the lift curve
a
was adjusted for different
aspect ratios. Given the same taper ratio, note that effective dihedral increases as
aspect ratio
increases.
As the taper ratio gets larger (approaches and exceeds unity), effective dihedral gets
larger. This is
because effective dihedral is influenced by the position of the wing centroid. As the
centroid
moves outboard with larger taper ratio, effective dihedral increases. The area added at
the tip
increases the dihedral effect strongly because of its large moment arm.
Swept Wings and Effective Dihedral The Basics
In an ideal world, an aircraft would have adjustable vertical stabilizer size and adjustable
dihedral.
Dihedral would be reduced and vertical stabilizer area increased during high speed
straight line
flight. While thermalling, dihedral would be increased and vertical stabilizer area
reduced.
Imagine for a moment that dihedral could be automatically adjusted in direct inverse
proportion to
speed. That is, higher speed would reduce dihedral, while slow speed would increase
dihedral. If
dihedral could be controlled appropriately, adjustment of vertical stabilizer size would
become
relatively unimportant.
We usually think of dihedral in geometric terms. That is, we actually see the dihedral
when we
view the aircraft from the front or rear. We block up the wing tip a certain distance
during
construction, or see the dihedral angle noted on outline plans in magazines.
C
l
|
a
6
--
1
2
+
1 +
\.
| | I
=
a
0.105
AR
AR
+ 2
-----------------

=
EffectiveDihedralRevB.FM5
13
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0.0
0 10 20 30
0 10 20 30
5
15
20
25
30 ASPECT RATIO
EFFECTIVE DIHEDRAL, DEGREES
DIHEDRAL, IN DEGREES
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0.0
10
TAPER RATIO = 0.6
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0.0
0 10 20 30
0 10 20 30
5
15
20
25
30 ASPECT RATIO
EFFECTIVE DIHEDRAL, DEGREES DIHEDRAL, IN DEGREES
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0.0
10
TAPER RATIO = 1.0
Graph 1
Graph 2
EffectiveDihedralRevB.FM5
14
But there is also a dihedral effect resulting from certain planforms, contours and from
the interactions
of various components. Weve already noted, for example, that a high wing location
provides
sufficient dihedral effect to mandate use of only a small additional positive geometric
dihedral angle.
A low wing location normally dictates a significant amount of positive geometric
dihedral, as the
position of the wing gives a large amount of negative effective dihedral. See Figure 9.
Wing tip
contour has a smaller, but still significant, effect. See Figure 10.
Wing sweep also provides some amount of dihedral effect. Sweep back gives positive
dihedral effect,
while sweep forward gives a negative dihedral effect.
high wing: equal to between three and
eight degrees of positive dihedral,
C
l
= -0.0006 to -00168
low wing: equal to between three and
eight degrees of negative dihedral,
C
l
= +0.0006 to +0.00168
mid wing: equal to approximately
zero degrees of dihedral,
C
l
= 0.0
Figure 9, wing position and effective dihedral
equivalent to approximately one degree
of positive geometric dihedral,
C
l
= -0.00021
equivalent to approximately zero degrees
of geometric dihedral,
C
l
= 0.0
equivalent to approximately one degree
of negative geometric dihedral,
C
l
= +0.00021
Figure 10, wing tip shape and effective dihedral
EffectiveDihedralRevB.FM5
15
From McCormick, page 547, we find that for a linearly tapered wing, the increment due
to sweep
angle is
:
where
C
l
= increment due to sweep angle, and C
l
-0.00021 is equivalent to one degree of positive
effective dihedral;

= the taper ratio;
C
L
= the wing lift coefcient;
A
= the wing sweep angle measured at half chord.
Theres a tricky part to this. Note the inclusion of the coefficient of lift, C
L
, in this equation.
Weve several times seen a rule of thumb stating three degrees of sweep is
equivalent to one
degree of geometric dihedral. Contrary to that popular notion, swept wings demonstrate
a
continuously variable dihedral effect. The amount of effective dihedral is directly
dependent upon
both the sweep angle,
A
, and the coefficient of lift, C
L
. These relationships are shown in Graph 3
(taper ratio = 0.6) and Graph 4 (taper ratio = 1.0).
Several points depicted in this Graph are worth noting:
First, if C
L
is zero, effective dihedral is zero, regardless of the sweep angle.
Second, when the wing is swept back, the higher the C
L
value, the greater the effective dihedral.
This is an important consideration, as it relates directly to swept wing performance
characteristics
across the aircraft speed range. This is because velocity is directly related to C
L
. As velocity
decreases, C
L
must increase.
Effective dihedral thus decreases at higher speeds, when C
L
values are low, as when flying
between thermals or racing, and increases at lower speeds and higher C
L
values, as while
thermalling. This is exactly what we want!
Effective dihedral, as related to sweep, is directly proportional to the wing sweep angle,
as
measured at the half chord line. Some amount of sweep, usually nomore than 25
degrees, produces
an amount of effective dihedral which benefit s performance across the speed range.
Severe sweep
angles, however, adversely affect spanwise flow and wing efficiency, and make
launching via
winch more difficult.
C
l
|
1 + 2
3(1 + )

C
L
A
= tan
EffectiveDihedralRevB.FM5
16
Since lateral stability (dihedral effect) increases with both the sweep angle and the
coefficient of
lift, there will be variations in aileron effectiveness depending on the flight regime.
During those
periods when a swept wing is generating a very large coefficient of lift, roll response
may be below
acceptable levels. It is therefore imperative to have large ailerons which can produce
sufficient
power at minimal deflection angles.
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
0.0
0 10 20 30
0 10 20 30
5 10 15 18 20 25 30 SWEEP ANGLE,
A
, DEGREES
EFFECTIVE DIHEDRAL, DEGREES
C
L
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
0.0
TAPER RATIO = 0.6
Graph 3
EffectiveDihedralRevB.FM5
17
Winglets
One final factor also influences effective dihedral winglets. Upright winglets, as
normally seen
on swept wing thermal duration tailless gliders, produce a substantial amount of
effective
dihedral. Why this is so is examined graphically in Figure 11.
Graph 4
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
0.0
0 10 20 30
0 10 20 30
5 10 15 18 20 25 30 SWEEP ANGLE,
A
, DEGREES
EFFECTIVE DIHEDRAL, DEGREES
C
L
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
0.0
TAPER RATIO = 1.0
+
+
+
+
+
+
+






Figure 11, yaw induced roll moment.
Flight direction is toward the viewer
EffectiveDihedralRevB.FM5
18
If winglets are mounted on a plank planform wing, and the wing is then yawed, the
forward
winglet produces some amount of lift toward the wing. The trailing winglet will produce
lift away
from the wing. The side of the winglet which is facing away from the oncoming flow
therefore
has an area of reduced pressure. Adjacent areas of the wing are affected as well. The
gross result
is a rolling moment which is directly related to the amount of yaw. This effect is
maintained when
the wing is swept.
From Nickel and Wohlfahrt, page 108, the skid-roll moment for a wing with winglets is
the same
as that of a conventional wing with the equivalent dihedral angle, EDA (p. 108)
:
where
EDA = equivalent dihedral angle;
h
W
= the height of the winglet;
s = b/2, the semi-span.
From this formula it can be seen that taller winglets (those with greater span) produce
greater
amounts of effective dihedral when the wing span is held constant.
As wing sweep increases, winglet size can be reduced because of the lengthened lever
arm. As
reducing the height of the winglet also reduces effective dihedral, there is a trade-off of
sorts to be
reconciled.
Computing Total Effective Dihedral
It is important to realize that the fundamental concept to be understood with regard to
lateral
stability is the relationship between rolling moment and sideslip. The symbol C
l
is defined as
roll due to sideslip; a C
l
value of -0.00021 is roughly equivalent to one degree of positive
dihedral, as mentioned previously.
The total effective dihedral of an aircraft is the sum of all of the effective dihedral
increments
generated by the planform and its various components. Wing taper ratio, geometric
dihedral and
tip shape, winglets, position of the wing in relation to any fuselage, wing sweep, and
coefficient
of lift must all be taken into consideration. Flaps with swept hinge lines accentuate the
dihedral
effect even further, sometimes making controlled flight impossible.
A vertical tail, if any, also contributes to effective dihedral, as any sideslip generates a
rolling
moment because of side forces on the fin and rudder.
Swept wings, particularly those which use winglets, may suffer from excessive effective
dihedral.
Because effective dihedral is directly proportional to the coefficient of lift, roll authority
may be
lost while thermalling. In extreme cases, severe Dutch roll occurs at low speeds, as
during
landing.
EDA
20h
W
s
= -------------
EffectiveDihedralRevB.FM5
19
Wing twist also affects the dihedral of swept wings. Excessive washout can unload the
tip so
much that the dihedral effect is small or even reversed at high speeds! and may
produce a
wing that cant fly well inverted.
Counteracting Excess Effective Dihedral
Luckily, negative geometric dihedral can be used to inhibit the strong dihedral effect
brought on
by a combination of high C
L
values, severe sweep angles, and/or other factors. But computing
how much anhedral is necessary for controllability across all flight regimes does pose
some
difficulties; there is a limit to how much anhedral (negative dihedral) can be used.
Effective dihedral may be so small at low coefficients of lift that too much geometric
anhedral
may cause the aircraft to invert at high speed. Or the aircraft may be susceptible to
static
directional divergence if too much anhedral is used and the angle of attack becomes too
great.
Countering excess effective dihedral may also be accomplished by mounting the
winglets such
that they hang below the wing. This produces a reversed yaw-roll effect, but places the
winglets in
a position to be easily damaged.
So-called diffuser tips are sometimes used in lieu of straight anhedral or downward
projecting
winglets. Some sources denote this geometry with the term Stromburg wing. See
Figure 12.