Evolution of Forward-Swept
Wing Configurations |
|
Research on the
swept wing as a drag-reducing mechanism for high
subsonic and transonic speeds during the late 1930’s
and early 1940’s resulted in some of the
first conventional aft-swept wing aircraft during
World War II. At that time, it was also recognized
that forward-swept wings (FSW) could produce the
same beneficial effect for performance. Furthermore,
the FSW also promised improved low-speed controllability.
Stalls were expected to start at the wing root
rather than the tip (in contrast to aft-swept wings),
thereby maintaining the effectiveness of outboard
ailerons and their contributions to roll control
at low speeds. The onset of shock waves at high
speeds was also expected to begin at the wing root,
which again maintains aileron effectiveness at
high speeds. With lateral control effectiveness
assured across the operational envelope, there
would be no need for drag-producing leading-edge
high-lift devices. Finally, the general layout
of the FSW resulted in a more aft location of the
wing spar carry-through structure in the fuselage,
which results in more fuselage internal volume.
During 1942, the German Junkers Design Bureau
initiated studies of an FSW bomber designated the
Ju-287. First flown in 1944, the Ju-287 exhibited
several problems, the most serious of which was
a tendency to increase g-loading during
a turn without control inputs from the pilot. The
analysis of the problem by Junkers revealed that
the cause was wing structural deformation from
the aerodynamic loads on the forward-facing wingtip
panels. At high speeds, the deformation was predicted
to become very severe, exceed structural limits,
and result in wing failure. This potentially catastrophic
phenomenon was referred to as aeroelastic divergence.
Further analysis indicated that the structural
modifications required to avoid the divergence
problem for the aluminum wing of the Ju-287 would
result in excessive weight and unacceptable performance
penalties. Other interest in FSW configurations
during World War II came from the American Cornelius
Aircraft Company, which worked on several configurations,
including the XFG-1, a piloted towable glider used
to transport fuel.
Researchers at Langley also investigated FSW
as part of a program to develop variable-sweep
wings. In one of these investigations, an existing
wind-tunnel model of the Bell X-1 was equipped
with a variable-sweep wing, and tests were conducted
for a FSW version of the aircraft in the Langley
300-MPH 7- by 10-Foot Tunnel.
Following World War II, the only significant
FSW aircraft built was the German Hansajet business
jet, which was designed by the same chief engineer
who designed the Ju-287. The aircraft never enjoyed
a large market.
The German Junkers Ju-287 forward-swept-wing
bomber.
The problem of aeroelastic divergence stood squarely
in the progress of FSW options for relatively high-speed
aircraft, and the challenge of providing sufficient
rigidity versus weight turned many designers away
from the concept.
In the 1970’s, two activities coupled to
stimulate interest in FSW configurations. First,
Grumman became interested in conducting aerodynamic
research to determine methods to improve its Highly
Maneuverable Aircraft Technology (HiMAT) configuration,
(which had lost in the design competition to Rockwell)
including revolutionary wing configurations. The
second activity was the remarkable advocacy and
influence of Dr. Norris J. Krone, Jr., a retired
Air Force pilot who had written his doctoral thesis
on eliminating aeroelastic divergence of FSW configurations
by using advanced tailored composites for structural
rigidity. Krone subsequently became a manager at
the Defense Advanced Research Projects Agency (DARPA),
and Krone’s discussions with the Grumman
managers led to a resurgence of interest in an
examination of the FSW concept. During 1977, DARPA
released a request for proposals (RFP) for a highly
advanced technology demonstrator that would integrate
advanced aerodynamics (with emphasis on the FSW)
and advanced flight controls. Responses were received
from Grumman, Rockwell, and General Dynamics.
On December 22, 1981, DARPA announced that Grumman
had been selected to develop the new technology
demonstrator, to be known as the X-29. |
High-Angle-of-Attack
Technology |
|
Initial information
exchanges between DARPA and Langley on independent
high-angle-of-attack evaluations of the competing
FSW configurations occurred in early 1980, when
Krone visited Joseph R. Chambers and his staff
at the 30- by 60-Foot (Full-Scale) Tunnel. Langley
agreed to provide support in this area as requested
for all three competing industry teams. General
Dynamics proposed an FSW version of the F-16 as
a candidate design for the DARPA competition. Exploratory
static wind-tunnel data had already been generated
by cooperative tests led by Langley researcher
Sue B. Grafton in the Langley 12-Foot Low-Speed
Tunnel in 1978. Final tests of the General Dynamics
design occurred in the Full-Scale Tunnel in April
1980. Rockwell’s FSW configuration underwent
preliminary static and dynamic tests in the Full-Scale
Tunnel in March 1981. Grumman’s FSW configuration
was tested in the same tunnel during three entries
beginning in November 1980. As a result of these
tests, DARPA was provided with a timely, independent
assessment of the high-angle-of-attack characteristics
of all three competing designs. Langley gained
considerable experience with the unique aerodynamic,
stability, and control characteristics of FSW configurations
at high angles of attack.
Following the award of DARPA’s X-29 contract
to Grumman, Langley’s support in the area
of high-angle-of-attack technology expanded to
include additional dynamic force tests and free-flight
model tests in the Full-Scale Tunnel, spin and
spin recovery tests in the Langley 20-Foot Vertical
Spin Tunnel, control system development studies
and piloted assessments of high-angle-of-attack
behavior in the Langley Differential Maneuvering
Simulator (DMS), and assessments of spin-entry
and post-stall motions using a radio-controlled
drop model.
One of the most important, unexpected results
of the X-29 high-angle-of-attack study came during
preliminary static and dynamic tests of a 0.16-scale
free-flight model in the Full-Scale Tunnel. The
X-29 government and industry team (and the entire
technical community) had expected the X-29 to be
heavily damped in roll at high angles of attack
as a result of the tendency of the FSW to maintain
attached airflow at the wingtips during stall.
However, when Sue Grafton conducted the first dynamic
force tests to measure aerodynamic damping in roll,
the results indicated that the X-29 configuration
would exhibit very unstable values of roll damping
at angles of attack above about 25 deg. This result
came as a complete surprise, and additional tests
were quickly planned to confirm the suspected impact
of the unstable damping. Grafton conducted a special
“free-to-roll” test, in which the X-29
model was mounted to a sting assembly that contained
a roll bearing which provided a 360-deg roll capability.
The test technique evaluated the tendency of the
model to display unsatisfactory roll characteristics
at high angles of attack. At low angles of attack,
the model was very stable, with no tendency to
oscillate or diverge (in agreement with the results
of the dynamic force test). When the angle of attack
of the model was increased to about 25 deg, however,
the model suddenly exhibited large amplitude wing
rocking motions of a periodic nature. The wing
rock was a nonlinear phenomenon, in that the model
motions were self-initiating and built up to a
limited amplitude independent of the magnitude
of the initial disturbance.
Sue Grafton with the F-16 forward-swept-wing
model.
The early identification of the wing rock led
to more tests, wherein it was determined that the
wings of the X-29 were indeed working as predicted.
That is, the airflow remained attached at the wingtips.
However, vortical flow shed from the long, pointed
forebody of the X-29 was found to be interacting
on the upper fuselage and inner wing and causing
the unstable damping, which was so large that it
overwhelmed the stabilizing influence of the attached
flow at the wingtips. Interestingly, the X-29 incorporated
the forward fuselage of the Northrop F-5A, which
is also known to exhibit wing rock at low speeds
and high angles of attack because of the same phenomenon.
With the cause of the instability identified, the
X-29 flight control system could be modified to
increase the level of the artificial roll damping
provided by feedback to the flaperons. Fortunately,
the flaperons of the X-29 retained their effectiveness
because of the favorable flow patterns of the FSW
at high angles of attack. Estimates indicated that
sufficient damping could be provided by the flight
control system via the roll damper, which utilized
the flaperons.
Free-flight tests of the X-29 model were first
conducted in the Full-Scale Tunnel in January 1982.
A special challenge faced the Langley team, since
the X-29 airframe was designed for a very high
level of aerodynamic instability in pitch (relaxed
static stability) with a highly responsive, redundant
flight control system that provided stability augmentation.
The X-29 would be unflyable without the stabilizing
inputs of the stability augmentation system. No
other aircraft (and no other free-flight model)
had ever flown with such a high level of relaxed
stability. The X-29 incorporated a level of relaxed
longitudinal stability (-32 percent at low speeds)
that was an order of magnitude more unstable than
the F-16. Under the leadership of Luat T. Nguyen,
the staff programmed the X-29 control laws into
the Langley computer that was used to replicate
full-scale controls for the model flight tests.
With a vane on the model nose boom providing information
on angle of attack, the control system of the model
performed flawlessly during the entire test program.
The X-29 model became the first flying vehicle
with such a level of relaxed stability.
X-29 free-flight model undergoing
tests at high angles of attack in the Langley Full-Scale
Tunnel.
During the free-flight tests led by Daniel G.
Murri and Sue Grafton, the model exhibited large
amplitude wing rock near an angle of attack of
25 deg when the roll damper component of the flight
control system was turned off, as had been observed
in the free-to-roll tests. The wing rock became
more severe with increasing angle of attack, and
flights usually resulted in loss of control of
the model near an angle of attack of 30 deg. The
amplitude and frequency of the motions were in
close agreement with the preliminary tests. When
the roll damper was engaged, the motions quickly
damped out and the model displayed satisfactory
characteristics.
The timely identification of the unstable roll
damping of the X-29 configuration was a major contribution
of Langley for the aircraft. If the full-scale
aircraft flight program had begun without the provision
for adequate control system gains and an awareness
of a possible roll-damping problem, the high-angle-of-attack
characteristics of the X-29 could have been unacceptable.
Ultimately, the X-29 did display the wing rock
tendency in flight (although not to the degree
of severity indicated by the model tests). However,
the controls of the full-scale aircraft were even
more effective than those of the free-flight model,
and the control system had been designed with provision
to increase the gain of the roll damper. The two
X-29 aircraft subsequently exhibited entirely satisfactory
characteristics in flight, and the technical community
learned a lesson regarding the integrated aerodynamic
contributions of FSW configurations with long,
pointed fuselages.
In 1983, Luat Nguyen led a piloted assessment
of the X-29 at high angles of attack in the DMS.
The objectives were to study the high-angle-of-attack
flying characteristics of the aircraft and its
susceptibility to departure under maneuvering conditions,
identify and develop control law concepts to provide
good flying qualities and a high level of departure-spin
resistance, assess the effectiveness of the Grumman
flight control laws at high angles of attack and
provide recommendations for modifications, and
provide support for flight-test planning and coordination
with NASA Dryden Flight Research Center.
A priority issue in the DMS studies was the effect
of Reynolds number on aerodynamic characteristics.
Results of high Reynolds number tests in the NASA
Ames 12-Foot Pressure Tunnel indicated that significant
effects of Reynolds number on pitching moments
and yawing moments existed for the X-29 configuration.
Working with flight control team members from Grumman
and Dryden, Nguyen completed an in-depth study
of X-29 handling qualities at high angles of attack
and provided numerous recommendations for the specific
design of the flight control system for high-angle-of-attack
conditions. Control law concepts were identified
for control surface interconnects for optimum roll
coordination, wing rock suppression, and automatic
departure and spin prevention.
|
Spin Tunnel Tests |
|
Tests of the X-29
in the Langley 20-Foot Vertical Spin Tunnel, which
began in 1981 under Raymond D. Whipple, concentrated
on three areas. First, the developed spin and spin
recovery characteristics of the X-29 were determined
for various aircraft loadings and erect and inverted
spins. As previously mentioned, large Reynolds
number effects had been predicted for the X-29
configuration. These effects were very noticeable
at very high angles of attack (near 90 deg), where
the model might exhibit critical spins. To correct
for these effects (which were caused by the unique
forebody shape that was incorporated from the F-5A)
at the low speeds involved in the Spin Tunnel conditions,
the Langley staff employed auxiliary strakes on
the nose of the X-29 spin model. The X-29 model
exhibited two types of spin. One type was flat,
with an angle of attack of about 85 deg, with marginal
to unsatisfactory spin recovery. The second spin
was very oscillatory, with satisfactory to excellent
recovery characteristics.
The second area of interest was determination
of the size of the emergency spin recovery parachute
required for the number 2 aircraft, which would
be used in the high-angle-of-attack flight-test
program. Working with Grumman and Dryden, Whipple
and the X-29 team arrived at a recommended parachute
size, truss structure, and deployment mechanism.
As a result of the outstanding maneuverability
and spin resistance exhibited by the X-29, the
emergency parachute system was never utilized in
flight tests to terminate post-stall maneuvers.
The third area of interest in the Spin Tunnel
test program involved concern over the possibility
of a longitudinal tumbling phenomenon for the X-29.
This concern had risen because of the high level
of inherent longitudinal instability designed into
the X-29. Specifically, Langley researchers expressed
concern over whether the combination of very low
airspeed and very high angle of attack (such as
during recovery from a “zoom climb”
to zero airspeed) might result in the aircraft
pitching over into an end-over-end tumbling that
would result in incapacitating g-levels
for the pilot. The Spin Tunnel staff first addressed
this issue in free-spinning tests wherein the X-29
model was launched tail first, without rotation,
into the vertically rising airstream. The results
of these exploratory tests showed that, without
inputs from the control system, the model would
indeed pitch over and develop a continuous tumbling
motion about its center of gravity. The wild gyrations
quickly caused the spin tunnel model to impact
the walls of the tunnel, so an additional test
technique was developed to study the issue under
more controlled conditions. In these tests, the
model was mounted on a special single-degree-of-freedom
test apparatus that permitted 360-deg free-pitching
motions. The additional studies of tumbling on
this apparatus provided insight to a possible solution
to the problem. The X-29 flight control system
included aft-fuselage strake flaps, which were
intended to be used only as trimming devices. However,
during the tumbling tests it was found that the
strake flaps were extremely efficient in promoting
recovery from the tumble motions. In view of this
result, the control system was modified to permit
use of the strake flaps as control devices. This
approach prevented the X-29 from entering the uncontrollable
tumbling motions.
The spin parachute and truss
assembly on the X-29 during high-angle-of-attack
tests at Dryden.
X-29 model mounted on single-degree-of-freedom
pitch-tumble test apparatus in the Langley Spin
Tunnel.
|
Drop-Model Tests |
|
In 1987, Langley
upgraded its drop-model test technique and conducted
studies of the high-angle-of-attack and post-stall
characteristics of a 0.22-scale model of the X-29.
The high degree of relaxed stability of the X-29
made high-fidelity simulation of the control system
a mandatory feature of the drop model. Langley
had never flown an unstable model before the X-29
program. The Langley staff, led by Mark A. Croom,
upgraded nearly every element of the drop-model
operation, such as the model control actuators,
transmitters and receivers, data encoders and decoders,
and operational displays including a new cockpit
display and operations monitor. The advanced X-29
control laws developed in the DMS study were programmed
into a ground-based computer for proper simulation
of control effects. This activity was by far the
most challenging drop-model project ever conducted
by Langley to that time.
The results of the drop-model program further
confirmed the need for special control system concepts
for high-angle-of-attack conditions. Without artificial
stability augmentation in roll, the drop model
exhibited the same type of large amplitude wing
rock motions previously displayed by the wind-tunnel
free-flight model. In the case of the drop model,
when the angle of attack was increased to 30 deg
and beyond, the oscillations became so divergent
that the model exhibited uncontrollable 360-deg
rolls that evolved into a roll departure and post-stall
gyrations. When the wing rock suppression system
was engaged, the motions were damped and the model
was controllable to very high angles of attack.
During some flights at extreme angles of attack,
asymmetric yawing moments were encountered that
caused the model to yaw and generate relatively
high rotational rates. When the pilot intentionally
applied prospin control inputs during these conditions,
the model sometimes entered the unrecoverable flat
spin mode exhibited by the spin tunnel model. With
the departure and spin prevention system engaged,
the model was highly resistant to intentional spins
and was extremely maneuverable at high angles of
attack.
Another valuable contribution of the X-29 drop-model
program was a parameter identification effort conducted
by David L. Raney and James G. Batterson of Langley,
in which they analyzed the wing rock motions of
the drop model and extracted values of the critical
aerodynamic parameters that caused the motions.
This complex analysis required the identification
of rapidly changing aerodynamic derivatives over
a range of angle of attack. The results of the
study clearly identified unstable values of aerodynamic
damping in roll to be the cause of the wing rock
motions.
X-29 drop model mounted on launch
apparatus on side of helicopter prior to release
for post-stall tests.
|
Aeroelastic Divergence |
|
The phenomenon of
aeroelastic divergence had dramatically constrained
international interest in the application of conventional
metal FSW concepts. Franklin W. Diederich and Bernard
Budiansky of Langley studied and summarized the
major challenges of the divergence phenomenon in
a NACA Technical Note in 1948 (ref.1). However,
the emergence of composite materials and the aggressive
advocacy of Norris Krone to utilize aeroelastic
tailoring to reduce divergence led to cooperative
studies of FSW technology by the staff of the Langley
16-Foot Transonic Dynamics Tunnel (TDT). Rodney
H. Ricketts, Robert V. Doggett, and Wilmer H. Reed,
Jr. planned and participated in numerous studies
with DARPA, the Air Force, and industry to develop
and verify analytical predictions of the divergence
phenomenon. The studies investigated systematic
generic wings and specific configurations, including
the Rockwell and Grumman FSW designs.
Rodney Ricketts with the Grumman
FSW model during aeroelastic
divergence tests in the Langley 16-Foot Transonic
Dynamics Tunnel.
The analyses and tests of generic wing models
with variations in aspect ratio, airfoils, and
sweep provided invaluable data and methods that
significantly expanded the database, especially
in the transonic regime. Six subcritical response
test techniques were formulated and evaluated at
transonic speeds for accuracy in predicting static
divergence, and two divergence stoppers were developed
and evaluated for use in preventing structural
damage of wind-tunnel models during divergence
tests.
Ricketts led cooperative tests of the proposed
Grumman and Rockwell FSW configurations in the
TDT in 1979. NASA, DARPA, the U.S. Air Force, and
the industry teams participated in the cooperative
tests. The Rockwell model consisted of a semispan
wing mounted to a splitter plate on the tunnel
sidewall. The Grumman model included a representative
fuselage shape. Conclusions from the divergence
tests included dramatic demonstrations that aeroelastic
tailoring was extremely effective in suppressing
divergence for FSW configurations and that new
nonlinear aerodynamic theories were required for
complete analysis of the phenomenon. After the
DARPA X-29 contract was awarded to Grumman, additional
tests were conducted in the TDT in 1983 which demonstrated
the potential coupling of the wing structural modes
with the rigid body pitch mode to create an instability
called body-freedom flutter. Techniques were developed
to analyze these wing-body interactions, and the
wing divergence prediction methods developed at
Langley were used in the flight-test program of
the X-29 at Dryden.
|
Advanced Engine
Nozzlev |
|
In the early 1970’s,
an advanced thrust-vectoring nozzle for V/STOL
aircraft was designed by the General Electric Company
under a contract funded by the Navy. The nozzle
was designed with a deflectable internal hood to
permit large pitch thrust-vector angles for V/STOL
operations, while deflection of the single expansion
ramp could be used for smaller thrust-vector angles
during air-to-air combat. This nozzle, known as
the augmented deflector exhaust nozzle (ADEN),
received considerable test and developmental support
in the technical community during the 1970’s
and early 1980’s. Much of this activity was
advocated by the Department of Defense (DOD), NASA,
the U.S. Air Force, and the U.S. Navy ad hoc interagency
nonaxisymmetric nozzle working group that included
Langley researcher Bobby L. Berrier as the NASA
representative. The Propulsion Aerodynamics Branch
at Langley conducted much of the development work
on the ADEN under the direction of William P. Henderson
and Bobby Berrier. Numerous cooperative ADEN research
programs with DOD and industry were conducted to
optimize nozzle performance and define suitable
propulsion-airframe integration methodologies on
a series of generic and specific (e.g., F-18) fighter
configurations. The first full-scale nonaxisymmetric
nozzle test was conduced at the Glenn Research
Center in 1976 by running an ADEN nozzle attached
to an augmented General Electric YJ101 engine in
an altitude test cell. Thus, by the time of the
DARPA X-29 award to Grumman in 1981, a mature thrust-vectoring
nozzle design was available.
Simple ADEN-like nozzle configuration
(pitch and yaw vectoring)
flown on X-29 free-flight model in the Full-Scale
Tunnel.
During the early phases of the X-29 program,
NASA Headquarters expressed interest in the application
of the ADEN to the canard configured X-29 as an
exploratory flight-test bed, since the canard of
the X-29 could counterbalance the effects of the
deflected ADEN nozzle. The NASA interest in the
X-29 ADEN was to promote technical progress in
vectorable nozzles and ensure that a sufficient
number of candidate nozzles were considered for
further development.
NASA Headquarters requested Langley’s assessment
of the X-29 ADEN configuration for further advocacy
discussions with Grumman and DARPA. Although no
induced lift or aerodynamic wing lift augmentation
would occur for this configuration, the addition
of vectoring was believed by Langley to improve
high-angle-of-attack controllability. In response
to this request from Headquarters, Joseph L. Johnson,
Jr. and his staff at the Full-Scale Tunnel quickly
modified their X-29 free-flight model and conducted
exploratory evaluations of the handling qualities
of the X-29 ADEN at high angles of attack in 1984.
During these tests, a yaw-vectoring side-door capability
(a concept defined and tested by the Propulsion
Aerodynamics Branch in the Langley Jet Exit Test
Facility) was added to a simplified ADEN-type nozzle
to provide both pitch and yaw vectoring, and the
resulting flight tests demonstrated a dramatic
improvement in controllability and agility of the
X-29 at extreme angles of attack. The model could
be flown with precise and effective control to
angles of attack as high as 85 deg. These positive
results were typical of those that had been obtained
with several different aircraft configurations
equipped with other thrust-vectoring nozzles. The
national progress in nonaxisymmetric, thrust-vectoring
nozzles led to interest in other nozzle configurations
and the eventual successful F-15 short takeoff
and landing and maneuver technology demonstrator
(STOL/MTD), F-18 High Alpha Research Vehicle (HARV),
and X-31 flight-test programs.
X-29 ADEN model flying at an
angle of attack of 80 deg in the Langley Full-Scale
Tunnel.
|
