|
Background
The single technology area
that had the most significant impact within the
discipline of aerodynamics for civil aircraft of
the 1990s was the explosive growth and versatility
of advanced computer-based methods in computational
fluid dynamics. Following years of pioneering efforts
within NASA, industry, DOD, other agencies, and
universities, the powerful computational capabilities
of rapidly evolving modern computers and computational
fluid dynamics (CFD) methodologies have provided
civil aircraft designers with unprecedented flexibility
to assess the impact of configuration variables
and conduct fundamental studies of fluid phenomena.
The utilization of CFD methods
within the civil sector has permitted detailed
studies that have dramatically changed the aircraft
design process. In addition to providing relatively
rapid and detailed understanding of aerodynamic
characteristics, CFD has also provided guidance
that has significantly reduced the number of experimental
tests, wind-tunnel hours, and models required to
develop modern aircraft. Virtually every aspect
of aircraft design now includes analyses using
CFD to evaluate and assess both viscous and inviscid
aerodynamic effects on airfoil and wing design,
high-lift performance, component interference,
propulsion-airframe integration, and cruise performance.
Major challenges that have
been, and continue to be, faced by researchers
in the CFD community revolve around accurate modeling
of critical flow physics, rapid and effective modeling
of the aircraft components under study, solutions
to nonlinear equations with millions of degrees
of freedom in a timely and cost-effective manner,
and establishing the validity and feasibility of
the computational approaches. In this area, NASA
aerodynamicists at Langley, Ames, Glenn, and Dryden
work closely with their peers in other organizations
to mutually provide this Nation with leading-edge
design and analysis tools. Many aircraft in the
current civil fleet have been impacted by these
capabilities.
Langley Research and Development
Activities
Langley and the other NASA
aeronautics Centers (Ames, Glenn, and Dryden) have
been at the forefront of research and development
for CFD codes and the maturation and validation
of integrated computational methodology. The relationship
of the NASA Centers with the civil aircraft industry
has been especially significant because NASA researchers
have been able to supply computational codes and
technology used as building blocks within the industry
for further development of proprietary codes and
design tools. At Langley, significant CFD contributions
to aircraft of the 1990s resulted directly from
a Center management commitment to computational
excellence and the innovation and world-class expertise
of its staff.
In the 1960s and 1970s, management
at Langley became increasingly aware of the significant
value of CFD to the analysis and design of future
aircraft. They observed emerging computational
efforts in areas ranging from relatively simple
nonviscous codes such as “panel” methods
to tremendously complex approaches required to
provide viscous flow solutions to the Navier-Stokes
equations for fluid flow and made a decision that
Langley should be a leader in developing and disseminating
CFD codes to industry for the analysis of aircraft
characteristics across the speed range from subsonic
to supersonic flight.
Langley’s Director for
Aeronautics, Robert E. Bower, made a major commitment
to establish and nurture a major new CFD organization
in the 1970s. It was during this time that lead
researchers such as Percy J. “Bud”
Bobbitt, Jerry C. South, Jr., Richard W. Barnwell,
Ajay Kumar, Douglas L. Dwoyer, and others established
the vision, hiring policies, investments, and organizational
dedication that permitted Langley’s emerging
CFD organizations to flourish. Investments in leading-edge
computers and other hardware provided the staff
with the tools required to advance the state of
the art. A key leadership action came from Roy
Harris, who collocated the CFD researchers together
in a common work area so that they could communicate
their methods, progress, and ideas. Thus, Langley
avoided a splintered and disparate activity and
rapidly nurtured a world-class capability.
Even a modest discussion of
the key individuals, technical advances in CFD,
and broad contributions of Langley’s historical
CFD program is far beyond the scope and intent
of this document. Thus, the reader is referred
to the extensive NASA literature for more details
and a broader perspective on all aspects of the
Langley CFD effort.
Applications by the Civil Aircraft
Industry
As might be expected, every
major civil aircraft manufacturer has an aerodynamics
group that is well versed in the latest technology
for both experimental and computational approaches
to aircraft design and development. Typically,
each company develops and utilizes its own highly
protected and proprietary methods and codes for
the computational design process. Fundamental research
conducted by NASA on computational methods is,
of course, evaluated by the industry and inserted
into proprietary toolboxes if cost-effective or
technically required. NASA extensively tests and
uses the codes, along with wind tunnels, flight
tests, and other analytical approaches, to carry
out its own programs. In the interest of achieving
effective technology transfer and dissemination
of results, NASA contracts with and/or works cooperatively
with industry and academia to further accelerate
CFD technology.
The civil industry is especially
critical of evolving computational methods from
the perspective of cost, timeliness, and validity.
For example, the development of a computational
code that requires excessive computer hardware
resources, unacceptable costs, unattainable computational
speeds, or excessive program run times is not highly
valued by industry. Therefore, the acknowledged
utilization of Langley-developed computational
methods by the civil industry is an impressive
endorsement of the value and critical nature of
the technology from the user’s perspective.
Many companies regard detailed information on aerodynamic
design—especially wing design—as the
heart of proprietary advantage and are very reluctant
to share knowledge regarding specific tools or
even the extent of their human and monetary resource
commitments in this critical area. Nonetheless,
several Langley contributions and key individuals
have been cited for particularly valuable contributions
to the industry in the design of specific aircraft
for the 1990s.
Unstructured grid representation
of Boeing 747-400 in USM3D code application.
One of the most important
contributions of Langley to CFD technology involves
the development of the Direct Iterative Surface
Curvature (DISC) code by Richard L. Campbell and
Leigh A. Smith of Langley. Known as an inverse
design method, DISC is an automated design method
that iteratively computes modifications to a wing
to meet a specified target pressure distribution.
By inputting the target pressure distribution,
the designer is able to, in effect, work the problem
from the desired aerodynamic characteristics back
to the generating geometry. DISC has been incorporated
into virtually all design codes used by industry.
Examples of the use of the DISC code by industry
include applications by Gulfstream during the development
of the Gulfstream V advanced business jet and by
Cessna in the design of the Citation X.
Other Langley codes used by
the civil aircraft industry include advanced methods
that provide relatively rapid solutions to the
complex equations that govern the physics of fluid
flows. Two of the most widely used methods are
known as the TLNS3D code and the USM3D code. The
TLNS3D code is used to solve the Navier-Stokes
equations by using multiple grid blocks to accommodate
complex aircraft geometries and a multigrid scheme
to accelerate the solution. The code was developed
by Veer N. Vatsa and others at Langley for use
on vector computers and has been used for sophisticated
aerodynamic computations by Boeing in the development
of the 777 and the 737, and by Gulfstream in the
development of the V jet. This code was the first
Navier-Stokes solver to demonstrate improved speed
and accuracy over the full potential codes with
interacted boundary layers, which was the industry
standard. The USM3D code is an innovative, efficient
solver of the Euler and Navier-Stokes equations
by using an approach known as unstructured gridding.
In this approach, developed by Neal T. Frink and
others of Langley, tetrahedral cells are used in
the representation and solution for flow about
the aircraft. Cessna applied USM3D in the design
of the Citation X.
Cooperative Studies
At the Paris Air Show in June
1985, McDonnell Douglas announced a follow-on to
the wide-body DC-10 transport, to be known as the
MD-11. This stretched version of the DC-10 incorporated
a number of advanced technologies, including redesigned
airfoil sections with more trailing-edge camber,
winglets, and an advanced horizontal tail that
included an integral trim tank. First flight of
the MD-11 occurred on January 10, 1990, and service
entry began later that year. Commercial customers
for the new aircraft numbered over 200, including
Federal Express, which accepted its first all-cargo
MD-11F in June 1991.
Initial flight tests of the
MD-11 indicated an unacceptable range shortfall
of over 400 nmi. McDonnell Douglas initiated a
modification program for the MD-11 known as the
Performance Improvement Program (PIP), which included
focused efforts to improve the aircraft’s
weight, fuel capacity, engine performance, and
aerodynamics. Cumulative improvements from modifications
identified by the PIP from 1990 to 1995 recovered
and subsequently extended the range for the aircraft.
A cooperative Langley and
McDonnell Douglas redesign effort helped the MD-11
reach its performance goals.
During the PIP activities,
McDonnell Douglas representatives approached William
P. Henderson and James M. Luckring of Langley for
assistance in implementing Langley’s advanced
CFD methods for analysis of an engine-pylon flow
separation problem that had been observed in flight
tests. The flow separation region occurred on the
outboard side of the juncture between the engine
pylon and the lower surface of the wing. The challenges
faced by McDonnell Douglas in solving the problem
were significant. The opportunity for flight-test
evaluations of proposed aircraft modifications
was constrained by the availability of the flight
test aircraft, and as a result, all analyses, design,
and fabrication had to be completed within 3 months.
Previous wind-tunnel tests had not identified the
problem prior to flight because of scale effects
involving laminar, rather than turbulent, separation
characteristics. McDonnell Douglas recognized that
advanced CFD methods in existence at Langley provided
a potential mechanism (perhaps the only one) for
pylon redesign efforts within the time constraints.
Langley and McDonnell Douglas
subsequently formed an analysis team that conducted
a 6-week effort at Langley Research Center using
several computational tools to identify the problem,
assess the effects of geometric changes to the
pylon, and arrive at the new pylon design. Working
with McDonnell Douglas computational experts on-site
at Langley, a team led by Neal T. Frink that included
Richard L. Campbell, Leigh A. Smith, Shahyar Z.
Pirzadeh, and Paresh C. Parikh provided expertise
and analysis during the intensive pylon study.
Critical computational tools used in the effort
included Langley-developed codes such as the VGRID
tetrahedral grid generator (developed by Pirzadeh
of ViGYAN, Inc.), the USM3D computational solver
for unstructured grids (developed by Frink), a
McDonnell Douglas version of Anthony Jameson’s
computational unstructured flow solver known as
AIRPLANE, and the inverse design method known as
DISC (developed by Campbell and Smith).
McDonnell Douglas and Langley
team utilized latest Langley CFD codes to redesign
pylon-wing fairing.
MD-11 pylon fairing retrofitted
to MD-11 fleet. Photograph ”Chris Coduto.
Extensive calculations using
the foregoing CFD tools provided insight into the
complex flow characteristics occurring in the regions
near the intersection of the pylon and wing leading-edge
lower surface. After analyzing pressure distributions,
suction peaks, and the effects of geometric modifications,
the Langley and McDonnell Douglas team identified
a candidate redesign for the pylon fairing that
significantly reduced adverse pressure peaks and
eliminated flow separation at cruise conditions.
Thanks to the guidance and analysis provided by
the CFD effort, a new pylon fairing was designed,
fabricated, and evaluated within the 3-month flight
test window. The flight tests of the new pylon
fairing validated the performance enhancements
predicted by the computational methods. A very
significant drag reduction of about 0.75 percent
was achieved. The fairing was installed in an all-new
MD-11 aircraft and was retrofitted to existing
aircraft. As a final positive result of this classical
cooperative venture, McDonnell Douglas engineers
became fluent in the use of Langley computational
methods, and the methods were subsequently integrated
into the McDonnell Douglas design capabilities.
Cooperative Langley and
Cessna tests of multipoint wing design for Citation
X in National Transonic Facility.
Although less successful than
the McDonnell Douglas MD-11 experience, Langley
researchers also participated in a cooperative
computational-experimental study with Cessna during
the development of the Citation X. In that study,
the objective was to consider a single wing design
that would perform most efficiently at two design
points—a high-speed condition and a slower,
but more efficient, cruise condition. Under the
direction of James M. Luckring and L. Elwood Putnam,
Leigh A. Smith and Raymond E. Mineck led the multipoint
wing design study that utilized Campbell’s
DISC code and other methods to arrive at a multipoint
wing design for the Citation X. The program proceeded
from computational design through testing in the
National Transonic Facility, but the study encountered
unanticipated delays and had no impact on the production
configuration.
Langley’s staff also
supported industry CFD in other ways. For example,
McDonnell Douglas requested that Campbell actively
help at the Long Beach, California, site in the
transfer of his DISC code. Campbell subsequently
spent 3 months at the McDonnell Douglas facility
to develop a version of his code with appropriate
constraints injected by McDonnell Douglas. He also
trained company engineers in the use of his code
and participated in the design of the MD-XX transport.
Although McDonnell Douglas decided not to proceed
with the MD-XX, they commended Campbell’s
contributions to the project and his personal transfer
of critical technology for future utilization.
Close working relationships
among the CFD staffs of the Langley Research Center,
the Ames Research Center, academia, and industry
have resulted in the effective development and
dissemination of extremely valuable codes for the
civil aircraft industry. Industry has incorporated
NASA codes into its design methodology and applied
them to its ongoing product lines. An impressive
perspective of the impact and value of NASA’s
building block technology can be obtained with
an overview of Boeing’s experiences with
wing design for recent transport aircraft. Boeing’s
use of advanced CFD methods has dramatically reduced
the number of candidate wing designs required for
experimental wind-tunnel testing during development
programs.
For example, during the development
of the 757 and 767, Boeing conducted wind-tunnel
testing of over 40 different wing configurations
for each of these aircraft, which represented the
1980 state of the art in wing design. With the
flexibility, guidance, and analysis provided by
CFD, only 18 different wings were tested in the
777 development program using technology of the
1990s. The reduction in wings tested is even more
impressive because the 777 wing is 21 percent thicker
than the wings of the 757 and 767 and has a higher
cruise speed capability. With even more sophisticated
CFD codes now under development, Boeing projects
that future aircraft programs may only require
as few as 5 wings to be tested in a wind tunnel.
|