ALBERT

Description: The compressible dynamic stall flowfield over a NACA 0012 airfoil transiently pitching from 0 to 60 deg at a constant rate under compressible flow conditions has been studied using real-time interferometry. A quantitative description of the overall flowfield, including the finer details of dynamic stall vortex formation, growth, and the concomitant changes in the airfoil pressure distribution, has been provided by analyzing the interferograms. For Mach numbers above 0.4, small multiple shocks appear near the leading edge and are present through the initial stages of dynamic stall. Dynamic stall was found to occur coincidentally with the bursting of the separation bubble over the airfoil. Compressibility was found to confine the dynamic stall vortical structure closer to the airfoil surface. The measurements show that the peak suction pressure coefficient drops with increasing freestream Mach number, and also it lags the steady flow values at any given angle of attack. As the dynamic stall vortex is shed, an anti-clockwise vortex is induced near the trailing edge, which actively interacts with the post-stall flow.

Description: The effect of the porous leading edge of an airfoil on the blade-vortex interaction noise, which dominates the far-field acoustic spectrum of the helicopter, is investigated. The thin-layer Navier-Stokes equations are solved with a high-order upwind-biased scheme and a multizonal grid system. The Baldwin-Lomax turbulence model is modified for considering transpiration on the surface. The amplitudes of the propagating acoustic wave in the near field are calculated directly from the computation. The porosity effect on the surface is modeled in two ways: (1) imposition of prescribed transpiration velocity distribution and (2) calculation of transpiration velocity distribution by Darcy's law. Results show leading-edge transpiration can suppress pressure fluctuations at the leading edge during blade-vortex interaction and consequently reduce the amplitude of propagating noise by 30% at a maximum in the near field.

Description: A method has been developed for calculating the viscous flow about airfoils with and without deflected flaps at -90 deg incidence. This method provides for the solution of the unsteady incompressible Navier-Stokes equations by means of an implicit technique. The solution is calculated on a body-fitted computational mesh using a staggered-grid method. The vorticity is defined at the node points, and the velocity components are defined at the mesh-cell sides. The staggered-grid orientation provides for accurate representation of vorticity at the node points and the continuity equation at the mesh-cell centers. The method provides for the noniterative solution of the flowfield and satisfies the continuity equation to machine zero at each time step. The method is evaluated in terms of its stability to predict two-dimensional flow about an airfoil at -90-deg incidence for varying Reynolds number and laminar/turbulent models. The variations of the average loading and surface pressure distribution due to flap deflection, Reynolds number, and laminar or turbulent flow are presented and compared with experimental results. The comparisom indicate that the calculated drag and drag reduction caused by flap deflection and the calculated average surface pressure are in excellent agreement with the measured results at a similar Reynolds number.

Description: Rotor noise prediction codes predict the thickness and loading noise produced by a helicopter rotor, given the blade motion, rotor operating conditions, and fluctuating force distribution over the blade surface. However, the criticality of these various inputs, and their respective effects on the predicted acoustic field, have never been fully addressed. This paper examines the importance of these inputs, and the sensitivity of the acoustic predicitions to a variation of each parameter. The effects of collective and cyclic pitch, as well as coning and cyclic flapping, are presented. Blade loading inputs are examined to determine the necessary spatial and temporal resolution, as well as the importance of the chordwise distribution. The acoustic predictions show regions in the acoustic field where significant errors occur when simplified blade motions or blade loadings are used. An assessment of the variation in the predicted acoustic field is balanced by a consideration of Central Processing Unit (CPU) time necessary for the various approximations.

Description: The U.S. National Aeronautics and Space Administration (NASA) Balloon Program has been highly successful since recovering from the catastrophic balloon failure problems of the early to mid 1980s. Balloons have continued to perform at unprecedented success rates. The comprehensive research and development (R&D) effort has continued with advances being made across the spectrum of balloon related disciplines. The long duration balloon project will be transitioning from a development effort to an operational capability this year. Recently, emphasis has been placed on the development and implementation of new support systems and facilities. A new permanent launch facility at Fort Sumner, New Mexico has been established. New ground station support equipment is being implemented, and a new heavy load launch vehicle is scheduled to be implemented in 1992. The progress, status and future plans for these and other aspects of the NASA program will be presented.

Description: The catastrophic balloon failure during the first half of the 1980's identified the need for a comprehensive and continuing balloon research and development (R&D) commitment by NASA. Technical understanding was lacking in many of the disciplines and processes associated with scientific ballooning. A comprehensive balloon R&D plan was developed in 1986 and implemented in 1987. The objectives were to develop the understanding of balloon system performance, limitations, and failure mechanisms. The program consisted of five major technical areas: structures, performance and analysis, materials, chemistry and processing, and quality control. Research activitites have been conducted at NASA/Goddard Space Flight Center (GSFC)-Wallops Flight Facility (WFF), other NASA centers and government facilities, universities, and the balloon manufacturers. Several new and increased capabilities and resources have resulted from this activity. The findings, capabilities, and plan of the balloon R&D program are presented.

Description: Caps have been used to structurally reinforce scientific research balloons since the late 1950's. The scientific research balloons used by the National Aeronautics and Space Administration (NASA) use internal caps. A NASA cap placement specification does not exist since no empirical information exisits concerning cap placement. To develop a cap placement specification, NASA has completed two in-hangar inflation tests comparing the structural contributions of internal caps and external caps. The tests used small scale test balloons designed to develop the highest possible stresses within the constraints of the hangar and balloon materials. An externally capped test balloon and an internally capped test balloon were designed, built, inflated and simulated to determine the structural contributions and benefits of each. The results of the tests and simulations are presented.

Description: The purpose of this Note is to present results from an analytic/experimental study that investigated the potential for passively changing blade twist through the use of extension-twist coupling. A set of composite model rotor blades was manufactured from existing blade molds for a low-twist metal helicopter rotor blade, with a view toward establishing a preliminary proof concept for extension-twist-coupled rotor blades. Data were obtained in hover for both a ballasted and unballasted blade configuration in sea-level atmospheric conditions. Test data were compared with results obtained from a geometrically nonlinear analysis of a detailed finite element model of the rotor blade developed in MSC/NASTRAN.

Description: The paper considers the compressible Rayleigh equation as a model for the Mach wave emission mechanism associated with high-temperature supersonic jets. Solutions to the compressible Rayleigh equation reveal the existence of several families of supersonically convecting instability waves. These waves directly radiate noise to the jet far field. The predicted noise characteristics are compared to previously acquired experimental data for an axisymmetric Mach 2 fully pressure balanced jet operating over a range of jet total temperatures from ambient to 1370 K. The results of this comparison show that the first-order supersonic instability wave and the Kelvin-Hemlhlotz first-, second-, and third-order modes have directional radiation characteristics that are in agreement with observed data. The assumption of equal initial amplitudes for all of the waves leads to the conclusion that the flapping mode of instability dominates the noise radiatio process of supersonic jets. At a jet temperature of 1370 K, supersonic instability waves are predicted to dominate the noise radiated at high frequency at narrow angles to the jet axis.

Description: The objective of the present work is to study the mixing characteristics of a linear array of supersonic rectangular jets under conditions of screech synchronization. The screech synchronization at a fully expanded jet Mach number of 1.61 is achieved by a precise adjustment of the internozzle spacing. To our knowledge, such an experiment on the resonant mixing of screech synchronized multiple rectangular jets has not been reported before. The results are compared with the case where the screech was suppressed in the multijet configuration.

Description: The objective of the present investigation is to assess the effect of the spatial order of accuracy used for the evaluation of the inviscid fluxes on the resolution of higher order quantitites, such as velocity gradients. The viscous terms are computed as second-order accurate with central difference formulas, even though for the explicit part of the algorithm higher order approximations may be used. A viscous/inviscid method is used, and the outer part of the flowfield is computed with the inviscid flow equations. The viscous boundary-layer type flow region close to the body surface is computed with an algebraic eddy viscosity model. Results obtained with the conservative and nonconservative formulations and the viscous/inviscid approach are compared with available experimental data. The effect of grid refinement on the accuracy of the solution is also presented.

Description: ARC2D is a computational fluid dynamics program developed at the NASA Ames Research Center specifically for airfoil computations. The program uses implicit finite-difference techniques to solve two-dimensional Euler equations and thin layer Navier-Stokes equations. It is based on the Beam and Warming implicit approximate factorization algorithm in generalized coordinates. The methods are either time accurate or accelerated non-time accurate steady state schemes. The evolution of the solution through time is physically realistic; good solution accuracy is dependent on mesh spacing and boundary conditions. The mathematical development of ARC2D begins with the strong conservation law form of the two-dimensional Navier-Stokes equations in Cartesian coordinates, which admits shock capturing. The Navier-Stokes equations can be transformed from Cartesian coordinates to generalized curvilinear coordinates in a manner that permits one computational code to serve a wide variety of physical geometries and grid systems. ARC2D includes an algebraic mixing length model to approximate the effect of turbulence. In cases of high Reynolds number viscous flows, thin layer approximation can be applied. ARC2D allows for a variety of solutions to stability boundaries, such as those encountered in flows with shocks. The user has considerable flexibility in assigning geometry and developing grid patterns, as well as in assigning boundary conditions. However, the ARC2D model is most appropriate for attached and mildly separated boundary layers; no attempt is made to model wake regions and widely separated flows. The techniques have been successfully used for a variety of inviscid and viscous flowfield calculations. The Cray version of ARC2D is written in FORTRAN 77 for use on Cray series computers and requires approximately 5Mb memory. The program is fully vectorized. The tape includes variations for the COS and UNICOS operating systems. Also included is a sample routine for CONVEX computers to emulate Cray system time calls, which should be easy to modify for other machines as well. The standard distribution media for this version is a 9-track 1600 BPI ASCII Card Image format magnetic tape. The Cray version was developed in 1987. The IBM ES/3090 version is an IBM port of the Cray version. It is written in IBM VS FORTRAN and has the capability of executing in both vector and parallel modes on the MVS/XA operating system and in vector mode on the VM/XA operating system. Various options of the IBM VS FORTRAN compiler provide new features for the ES/3090 version, including 64-bit arithmetic and up to 2 GB of virtual addressability. The IBM ES/3090 version is available only as a 9-track, 1600 BPI IBM IEBCOPY format magnetic tape. The IBM ES/3090 version was developed in 1989. The DEC RISC ULTRIX version is a DEC port of the Cray version. It is written in FORTRAN 77 for RISC-based Digital Equipment platforms. The memory requirement is approximately 7Mb of main memory. It is available in UNIX tar format on TK50 tape cartridge. The port to DEC RISC ULTRIX was done in 1990. COS and UNICOS are trademarks and Cray is a registered trademark of Cray Research, Inc. IBM, ES/3090, VS FORTRAN, MVS/XA, and VM/XA are registered trademarks of International Business Machines. DEC and ULTRIX are registered trademarks of Digital Equipment Corporation.

Description: Panel method computer programs are software tools of moderate cost used for solving a wide range of engineering problems. The panel code PMARC_12 (Panel Method Ames Research Center, version 12) can compute the potential flow field around complex three-dimensional bodies such as complete aircraft models. PMARC_12 is a well-documented, highly structured code with an open architecture that facilitates modifications and the addition of new features. Adjustable arrays are used throughout the code, with dimensioning controlled by a set of parameter statements contained in an include file; thus, the size of the code (i.e. the number of panels that it can handle) can be changed very quickly. This allows the user to tailor PMARC_12 to specific problems and computer hardware constraints. In addition, PMARC_12 can be configured (through one of the parameter statements in the include file) so that the code's iterative matrix solver is run entirely in RAM, rather than reading a large matrix from disk at each iteration. This significantly increases the execution speed of the code, but it requires a large amount of RAM memory. PMARC_12 contains several advanced features, including internal flow modeling, a time-stepping wake model for simulating either steady or unsteady (including oscillatory) motions, a Trefftz plane induced drag computation, off-body and on-body streamline computations, and computation of boundary layer parameters using a two-dimensional integral boundary layer method along surface streamlines. In a panel method, the surface of the body over which the flow field is to be computed is represented by a set of panels. Singularities are distributed on the panels to perturb the flow field around the body surfaces. PMARC_12 uses constant strength source and doublet distributions over each panel, thus making it a low order panel method. Higher order panel methods allow the singularity strength to vary linearly or quadratically across each panel. Experience has shown that low order panel methods can provide nearly the same accuracy as higher order methods over a wide range of cases with significantly reduced computation times; hence, the low order formulation was adopted for PMARC_12. The flow problem is solved by modeling the body as a closed surface dividing space into two regions: the region external to the surface in which an unknown velocity potential exists representing the flow field of interest, and the region internal to the surface in which a known velocity potential (representing a fictitious flow) is prescribed as a boundary condition. Both velocity potentials are required to satisfy Laplace's equation. A surface integral equation for the unknown potential external to the surface can be written by applying Green's Theorem to the external region. Using the internal potential and zero flow through the surface as boundary conditions, the unknown potential external to the surface can be solved for. When the internal flow option, which allows the analysis of closed ducts, wind tunnels, and similar internal flow problems, is selected, the geometry is modeled such that the flow field of interest is inside the geometry and the fictitious flow is outside the geometry. Items such as wings, struts, or aircraft models can be included in the internal flow problem. The time-stepping wake model gives PMARC_12 the ability to model both steady and unsteady flow problems. The wake is convected downstream from the wake-separation line by the local velocity field. With each time step, a new row of wake panels is added to the wake at the wake-separation line. Time stepping can start from time t=0 (no initial wake) or from time t=t0 (an initial wake is specified). A wide range of motions can be prescribed, including constant rates of translation, constant rate of rotation about an arbitrary axis, oscillatory translation, and oscillatory rotation about any of the three coordinate axes. Investigators interested in a visual representation of the phenomenon they are studying with PMARC_12 may want to consider obtaining the program GVS (ARC-13361), the General Visualization System. GVS is a Silicon Graphics IRIS program which was created for the purpose of supporting the scientific visualization needs of PMARC_12. GVS is available separately from COSMIC. PMARC_12 is written in standard FORTRAN 77, with the exception of the NAMELIST extension used for input. This makes the code fairly machine independent. A compiler which supports the NAMELIST extension is required. The amount of free disk space and RAM memory required for PMARC_12 will vary depending on how the code is dimensioned using the parameter statements in the include file. The recommended minimum requirements are 20Mb of free disk space and 4Mb of RAM. PMARC_12 has been successfully implemented on a Macintosh II running System 6.0.7 or 7.0 (using MPW/Language Systems Fortran 3.0), a Sun SLC running SunOS 4.1.1, an HP 720 running HP-UX 8.07, an SGI IRIS running IRIX 4.0 (it will not run under IRIX 3.x.x without modifications), an IBM RS/6000 running AIX, a DECstation 3100 running ULTRIX, and a CRAY-YMP running UNICOS 6.0 or later. Due to its memory requirements, this program does not readily lend itself to implementation on MS-DOS based machines. The standard distribution medium for PMARC_12 is a set of three 3.5 inch 800K Macintosh format diskettes and one 3.5 inch 1.44Mb Macintosh format diskette which contains an electronic copy of the documentation in MS Word 5.0 format for the Macintosh. Alternate distribution media and formats are available upon request, but these will not include the electronic version of the document. No executables are included on the distribution media. This program is an update to PMARC version 11, which was released in 1989. PMARC_12 was released in 1993. It is available only for use by United States citizens.

Description: This program determines the supersonic flowfield surrounding three-dimensional wing-body configurations of a delta wing. It was designed to provide the numerical computation of three dimensional inviscid, flowfields of either perfect or real gases about supersonic or hypersonic airplanes. The governing equations in conservation law form are solved by a finite difference method using a second order noncentered algorithm between the body and the outermost shock wave, which is treated as a sharp discontinuity. Secondary shocks which form between these boundaries are captured automatically. The flowfield between the body and outermost shock is treated in a shock capturing fashion and therefore allows for the correct formation of secondary internal shocks . The program operates in batch mode, is in CDC update format, has been implemented on the CDC 7600, and requires more than 140K (octal) word locations.

Description: This theoretical aerodynamics program, TAD, was developed to predict the aerodynamic characteristics of vehicles with sounding rocket configurations. These slender, axisymmetric finned vehicle configurations have a wide range of aeronautical applications from rockets to high speed armament. Over a given range of Mach numbers, TAD will compute the normal force coefficient derivative, the center-of-pressure, the roll forcing moment coefficient derivative, the roll damping moment coefficient derivative, and the pitch damping moment coefficient derivative of a sounding rocket configured vehicle. The vehicle may consist of a sharp pointed nose of cone or tangent ogive shape, up to nine other body divisions of conical shoulder, conical boattail, or circular cylinder shape, and fins of trapezoid planform shape with constant cross section and either three or four fins per fin set. The characteristics computed by TAD have been shown to be accurate to within ten percent of experimental data in the supersonic region. The TAD program calculates the characteristics of separate portions of the vehicle, calculates the interference between separate portions of the vehicle, and then combines the results to form a total vehicle solution. Also, TAD can be used to calculate the characteristics of the body or fins separately as an aid in the design process. Input to the TAD program consists of simple descriptions of the body and fin geometries and the Mach range of interest. Output includes the aerodynamic characteristics of the total vehicle, or user-selected portions, at specified points over the mach range. The TAD program is written in FORTRAN IV for batch execution and has been implemented on an IBM 360 computer with a central memory requirement of approximately 123K of 8 bit bytes. The TAD program was originally developed in 1967 and last updated in 1972.

Description: This program, which is called 'AOFA', determines the complete viscous and inviscid flow around a body of revolution at a given angle of attack and traveling at supersonic speeds. The viscous calculations from this program agree with experimental values for surface and pitot pressures and with surface heating rates. At high speeds, lee-side flows are important because the local heating is difficult to correlate and because the shed vortices can interact with vehicle components such as a canopy or a vertical tail. This program should find application in the design analysis of any high speed vehicle. Lee-side flows are difficult to calculate because thin-boundary-layer theory is not applicable and the concept of matching inviscid and viscous flow is questionable. This program uses the parabolic approximation to the compressible Navier-Stokes equations and solves for the complete inviscid and viscous regions of flow, including the pressure. The parabolic approximation results from the assumption that the stress derivatives in the streamwise direction are small in comparison with derivatives in the normal and circumferential directions. This assumption permits the equation to be solved by an implicit finite difference marching technique which proceeds downstream from the initial data point, provided the inviscid portion of flow is supersonic. The viscous cross-flow separation is also determined as part of the solution. To use this method it is necessary to first determine an initial data point in a region where the inviscid portion of the flow is supersonic. Input to this program consists of two parts. Problem description is conveyed to the program by namelist input. Initial data is acquired by the program as formatted data. Because of the large amount of run time this program can consume the program includes a restart capability. Output is in printed format and magnetic tape for further processing. This program is written in FORTRAN IV and has been implemented on a CDC 7600 with a central memory requirement of approximately 35K (octal) of 60 bit words.

Description: The Comprehensive Analytical Model of Rotorcraft Aerodynamics, CAMRAD, program is designed to calculate rotor performance, loads, and noise; helicopter vibration and gust response; flight dynamics and handling qualities; and system aeroelastic stability. The analysis is a consistent combination of structural, inertial, and aerodynamic models applicable to a wide range of problems and a wide class of vehicles. The CAMRAD analysis can be applied to articulated, hingeless, gimballed, and teetering rotors with an arbitrary number of blades. The rotor degrees of freedom included are blade/flap bending, rigid pitch and elastic torsion, and optionally gimbal or teeter motion. General two-rotor aircrafts can be modeled. Single main-rotor and tandem helicopter and sideby-side tilting proprotor aircraft configurations can be considered. The case of a rotor or helicopter in a wind tunnel can also be modeled. The aircraft degrees of freedom included are the six rigid body motion, elastic airframe motions, and the rotor/engine speed perturbations. CAMRAD calculates the load and motion of helicopters and airframes in two stages. First the trim solution is obtained; then the flutter, flight dynamics, and/or transient behavior can be calculated. The trim operating conditions considered include level flight, steady climb or descent, and steady turns. The analysis of the rotor includes nonlinear inertial and aerodynamic models, applicable to large blade angles and a high inflow ratio, The rotor aerodynamic model is based on two-dimensional steady airfoil characteristics with corrections for three-dimensional and unsteady flow effects, including a dynamic stall model. In the flutter analysis, the matrices are constructed that describe the linear differential equations of motion, and the equations are analyzed. In the flight dynamics analysis, the stability derivatives are calculated and the matrices are constructed that describe the linear differential equations of motion. These equations are analyzed. In the transient analysis, the rigid body equations of motion are numerically integrated, for a prescribed transient gust or control input. The CAMRAD program product is available by license for a period of ten years to domestic U.S. licensees. The licensed program product includes the CAMRAD source code, command procedures, sample applications, and one set of supporting documentation. Copies of the documentation may be purchased separately at the price indicated below. CAMRAD is written in FORTRAN 77 for the DEC VAX under VMS 4.6 with a recommended core memory of 4.04 megabytes. The DISSPLA package is necessary for graphical output. CAMRAD was developed in 1980.

Description: ARC2D is a computational fluid dynamics program developed at the NASA Ames Research Center specifically for airfoil computations. The program uses implicit finite-difference techniques to solve two-dimensional Euler equations and thin layer Navier-Stokes equations. It is based on the Beam and Warming implicit approximate factorization algorithm in generalized coordinates. The methods are either time accurate or accelerated non-time accurate steady state schemes. The evolution of the solution through time is physically realistic; good solution accuracy is dependent on mesh spacing and boundary conditions. The mathematical development of ARC2D begins with the strong conservation law form of the two-dimensional Navier-Stokes equations in Cartesian coordinates, which admits shock capturing. The Navier-Stokes equations can be transformed from Cartesian coordinates to generalized curvilinear coordinates in a manner that permits one computational code to serve a wide variety of physical geometries and grid systems. ARC2D includes an algebraic mixing length model to approximate the effect of turbulence. In cases of high Reynolds number viscous flows, thin layer approximation can be applied. ARC2D allows for a variety of solutions to stability boundaries, such as those encountered in flows with shocks. The user has considerable flexibility in assigning geometry and developing grid patterns, as well as in assigning boundary conditions. However, the ARC2D model is most appropriate for attached and mildly separated boundary layers; no attempt is made to model wake regions and widely separated flows. The techniques have been successfully used for a variety of inviscid and viscous flowfield calculations. The Cray version of ARC2D is written in FORTRAN 77 for use on Cray series computers and requires approximately 5Mb memory. The program is fully vectorized. The tape includes variations for the COS and UNICOS operating systems. Also included is a sample routine for CONVEX computers to emulate Cray system time calls, which should be easy to modify for other machines as well. The standard distribution media for this version is a 9-track 1600 BPI ASCII Card Image format magnetic tape. The Cray version was developed in 1987. The IBM ES/3090 version is an IBM port of the Cray version. It is written in IBM VS FORTRAN and has the capability of executing in both vector and parallel modes on the MVS/XA operating system and in vector mode on the VM/XA operating system. Various options of the IBM VS FORTRAN compiler provide new features for the ES/3090 version, including 64-bit arithmetic and up to 2 GB of virtual addressability. The IBM ES/3090 version is available only as a 9-track, 1600 BPI IBM IEBCOPY format magnetic tape. The IBM ES/3090 version was developed in 1989. The DEC RISC ULTRIX version is a DEC port of the Cray version. It is written in FORTRAN 77 for RISC-based Digital Equipment platforms. The memory requirement is approximately 7Mb of main memory. It is available in UNIX tar format on TK50 tape cartridge. The port to DEC RISC ULTRIX was done in 1990. COS and UNICOS are trademarks and Cray is a registered trademark of Cray Research, Inc. IBM, ES/3090, VS FORTRAN, MVS/XA, and VM/XA are registered trademarks of International Business Machines. DEC and ULTRIX are registered trademarks of Digital Equipment Corporation.

Description: Panel methods are moderate cost tools for solving a wide range of engineering problems. PMARC (Panel Method Ames Research Center) is a potential flow panel code that numerically predicts flow fields around complex three-dimensional geometries. PMARC's predecessor was a panel code named VSAERO which was developed for NASA by Analytical Methods, Inc. PMARC is a new program with many additional subroutines and a well-documented code suitable for powered-lift aerodynamic predictions. The program's open architecture facilitates modifications or additions of new features. Another improvement is the adjustable size code which allows for an optimum match between the computer hardware available to the user and the size of the problem being solved. PMARC can be resized (the maximum number of panels can be changed) in a matter of minutes. Several other state-of-the-art PMARC features include internal flow modeling for ducts and wind tunnel test sections, simple jet plume modeling essential for the analysis and design of powered-lift aircraft, and a time-stepping wake model which allows the study of both steady and unsteady motions. PMARC is a low-order panel method, which means the singularities are distributed with constant strength over each panel. In many cases low-order methods can provide nearly the same accuracy as higher order methods (where the singularities are allowed to vary linearly or quadratically over each panel). Low-order methods have the advantage of a shorter computation time and do not require exact matching between panels. The flow problem is solved by assuming that the body is at rest in a moving flow field. The body is modeled as a closed surface which divides space into two regions -- one region contains the flow field of interest and the other contains a fictitious flow. External flow problems, such as a wing in a uniform stream, have the external region as the flow field of interest and the internal flow as the fictitious flow. This arrangement is reversed for internal flow problems where the internal region contains the flow field of interest and the external flow field is fictitious. In either case it is assumed that the velocity potentials in both regions satisfy Laplace's equation. PMARC has extensive geometry modeling capabilities for handling complex, three-dimensional surfaces. As with all panel methods, the geometry must be modeled by a set of panels. For convenience, the geometry is usually subdivided into several pieces and modeled with sets of panels called patches. A patch may be folded over on itself so that opposing sides of the patch form a common line. For example, wings are normally modeled with a folded patch to form the trailing edge of the wing. PMARC also has the capability to automatically generate a closing tip patch. In the case of a wing, a tip patch could be generated to close off the wing's third side. PMARC has a simple jet model for simulating a jet plume in a crossflow. The jet plume shape, trajectory, and entrainment velocities are computed using the Adler/Baron jet in crossflow code. This information is then passed back to PMARC. The wake model in PMARC is a time-stepping wake model. The wake is convected downstream from the wake separation line by the local velocity flowfield. With each time step, a new row of wake panels is added to the wake at the wake separation line. PMARC also allows an initial wake to be specified if desired, or, as a third option, no wakes need be modeled. The effective presentation of results for aerodynamics problems requires the generation of report-quality graphics. PMAPP (ARC-12751), the Panel Method Aerodynamic Plotting Program, (Sterling Software), was written for scientists at NASA's Ames Research Center to plot the aerodynamic analysis results (flow data) from PMARC. PMAPP is an interactive, color-capable graphics program for the DEC VAX or MicroVAX running VMS. It was designed to work with a variety of terminal types and hardcopy devices. PMAPP is available separately from COSMIC. PMARC was written in standard FORTRAN77 using adjustable size arrays throughout the code. Redimensioning PMARC will change the amount of disk space and memory the code requires to be able to run; however, due to its memory requirements, this program does not readily lend itself to implementation on MS-DOS based machines. The program was implemented on an Apple Macintosh (using 2.5 MB of memory) and tested on a VAX/VMS computer. The program is available on a 3.5 inch Macintosh format diskette (standard media) or in VAX BACKUP format on TK50 tape cartridge or 9-track magnetic tape. PMARC was developed in 1989.

Description: The Transonic Airfoil analysis computer code, TAIR, was developed to employ a fast, fully implicit algorithm to solve the conservative full-potential equation for the steady transonic flow field about an arbitrary airfoil immersed in a subsonic free stream. The full-potential formulation is considered exact under the assumptions of irrotational, isentropic, and inviscid flow. These assumptions are valid for a wide range of practical transonic flows typical of modern aircraft cruise conditions. The primary features of TAIR include: a new fully implicit iteration scheme which is typically many times faster than classical successive line overrelaxation algorithms; a new, reliable artifical density spatial differencing scheme treating the conservative form of the full-potential equation; and a numerical mapping procedure capable of generating curvilinear, body-fitted finite-difference grids about arbitrary airfoil geometries. Three aspects emphasized during the development of the TAIR code were reliability, simplicity, and speed. The reliability of TAIR comes from two sources: the new algorithm employed and the implementation of effective convergence monitoring logic. TAIR achieves ease of use by employing a "default mode" that greatly simplifies code operation, especially by inexperienced users, and many useful options including: several airfoil-geometry input options, flexible user controls over program output, and a multiple solution capability. The speed of the TAIR code is attributed to the new algorithm and the manner in which it has been implemented. Input to the TAIR program consists of airfoil coordinates, aerodynamic and flow-field convergence parameters, and geometric and grid convergence parameters. The airfoil coordinates for many airfoil shapes can be generated in TAIR from just a few input parameters. Most of the other input parameters have default values which allow the user to run an analysis in the default mode by specifing only a few input parameters. Output from TAIR may include aerodynamic coefficients, the airfoil surface solution, convergence histories, and printer plots of Mach number and density contour maps. The TAIR program is written in FORTRAN IV for batch execution and has been implemented on a CDC 7600 computer with a central memory requirement of approximately 155K (octal) of 60 bit words. The TAIR program was developed in 1981.

Description: The TAWFIVE program calculates transonic flow over a transport-type wing and fuselage. Although more complex Euler and Navier-Stokes methods are available, TAWFIVE combines a multi-grid acceleration technique in the iterative solution of the potential equation with the use of integral-form boundary-layer equations to provide a computationally efficient and sufficiently accurate design tool. TAWFIVE simplifies the solution process by breaking the problem into a loosely coupled set of modified equations. The inviscid method, using standard inviscid equations (nonlinear full potential), is valid in the "outer" region away from the wing, whereas the boundary-layer equations are valid in the thin region near the solid surface of the wing. The two types of equations are coupled by a technique of modifying surface boundary conditions for the inviscid equations. This interaction process starts with a solution of the outer flow field. Pressures are computed at the wing surface and are used to calculate the boundary layer. The boundary-layer and wake properties are then computed using a three-dimensional integral method, and the computed displacement thickness is added to the surface of the "hard" geometry. This new displaced wing surface is then regridded and the inviscid flowfield is recomputed. New values of the inviscid pressures are then used by the boundary-layer method to predict a new displacement thickness distribution. An under-relaxed update of the previously predicted displacement thickness is then made to obtain a new displacement thickness correction that is added to the "hard" geometry. These global iterations are continued until suitable convergence is obtained. Input to TAWFIVE is limited to geometric definition of the configuration, free-stream flow quantities, and iteration control parameters. The geometric input consists of the definition of a series of airfoil sections to define the wing and a series of fuselage cross sections to model the fuselage. High-aspect-ratio wings are modeled more accurately than low-aspect-ratio wings since no special provisions are made to accurately model the wing-fuselage juncture or the wingtip region. The user can specify the solution either in terms of lift or in terms of angle of attack. TAWFIVE can produce tabular output and input files for PLOT3D (COSMIC program number ARC-12779). TAWFIVE is written in FORTRAN 77 for CRAY series computers running UNICOS. The main memory requirement is 2.7Mb for execution. This program is available on a 9-track 1600 BPI UNIX tar format magnetic tape. TAWFIVE was under development from 1979 to 1989 and first released by COSMIC in 1991. CRAY and UNICOS are registered trademarks of Cray Research, Inc.

Description: This computer program is designed to calculate the flow fields in two-dimensional and three-dimensional axisymmetric supersonic inlets. The method of characteristics is used to compute arrays of points in the flow field. At each point the total pressure, local Mach number, local flow angle, and static pressure are calculated. This program can be used to design and analyze supersonic inlets by determining the surface compression rates and throat flow properties. The program employs the method of characteristics for a perfect gas. The basic equation used in the program is the compatibility equation which relates the change in stream angle to the change in entropy and the change in velocity. In order to facilitate the computation, the flow field behind the bow shock wave is broken into regions bounded by shock waves. In each region successive rays are computed from a surface to a shock wave until the shock wave intersects a surface or falls outside the cowl lip. As soon as the intersection occurs a new region is started and the previous region continued only in the area in which it is needed, thus eliminating unnecessary calculations. The maximum number of regions possible in the program is ten, which allows for the simultaneous calculations of up to nine shock waves. Input to this program consists of surface contours, free-stream Mach number, and various calculation control parameters. Output consists of printed and/or plotted results. For plotted results an SC-4020 or similar plotting device is required. This program is written in FORTRAN IV to be executed in the batch mode and has been implemented on a CDC 7600 with a central memory requirement of approximately 27k (octal) of 60 bit words.

Description: This program was developed to predict turbine stage performance taking into account the effects of complex passage geometries. The method uses a quasi-3D inviscid-flow analysis iteratively coupled to calculated losses so that changes in losses result in changes in the flow distribution. In this manner the effects of both the geometry on the flow distribution and the flow distribution on losses are accounted for. The flow may be subsonic or shock-free transonic. The blade row may be fixed or rotating, and the blades may be twisted and leaned. This program has been applied to axial and radial turbines, and is helpful in the analysis of mixed flow machines. This program is a combination of the flow analysis programs MERIDL and TSONIC coupled to the boundary layer program BLAYER. The subsonic flow solution is obtained by a finite difference, stream function analysis. Transonic blade-to-blade solutions are obtained using information from the finite difference, stream function solution with a reduced flow factor. Upstream and downstream flow variables may vary from hub to shroud and provision is made to correct for loss of stagnation pressure. Boundary layer analyses are made to determine profile and end-wall friction losses. Empirical loss models are used to account for incidence, secondary flow, disc windage, and clearance losses. The total losses are then used to calculate stator, rotor, and stage efficiency. This program is written in FORTRAN IV for batch execution and has been implemented on an IBM 370/3033 under TSS with a central memory requirement of approximately 4.5 Megs of 8 bit bytes. This program was developed in 1985.

Description: Turbomachinery components are often connected by ducts, which are usually annular. The configurations and aerodynamic characteristics of these ducts are crucial to the optimum performance of the turbomachinery blade rows. The ANDUCT computer program was developed to calculate the velocity distribution along an arbitrary line between the inner and outer walls of an annular duct with axisymmetric swirling flow. Although other programs are available for duct analysis, the use of the velocity gradient method makes the ANDUCT program fast and convenient while requiring only modest computer resources. A fast and easy method of analyzing the flow through a duct with axisymmetric flow is the velocity gradient method, also known as the stream filament or streamline curvature method. This method has been used extensively for blade passages but has not been widely used for ducts, except for the radial equilibrium equation. In ANDUCT, a velocity gradient equation derived from the momentum equation is used to determine the velocity variation along an arbitrary straight line between the inner and outer wall of an annular duct. The velocity gradient equation is used with an assumed variation of meridional streamline curvature. Upstream flow conditions may vary between the inner and outer walls, and an assumed total pressure distribution may be specified. ANDUCT works best for well-guided passages and where the curvature of the walls is small as compared to the width of the passage. The ANDUCT program is written in FORTRAN IV for batch execution and has been implemented on an IBM 370 series computer with a central memory requirement of approximately 60K of 8 bit bytes. The ANDUCT program was developed in 1982.

Description: The Panel Code for Planar Cascades was developed as an aid for the designer of turbomachinery blade rows. The effective design of turbomachinery blade rows relies on the use of computer codes to model the flow on blade-to-blade surfaces. Most of the currently used codes model the flow as inviscid, irrotational, and compressible with solutions being obtained by finite difference or finite element numerical techniques. While these codes can yield very accurate solutions, they usually require an experienced user to manipulate input data and control parameters. Also, they often limit a designer in the types of blade geometries, cascade configurations, and flow conditions that can be considered. The Panel Code for Planar Cascades accelerates the design process and gives the designer more freedom in developing blade shapes by offering a simple blade-to-blade flow code. Panel, or integral equation, solution techniques have been used for several years by external aerodynamicists who have developed and refined them into a primary design tool of the aircraft industry. The Panel Code for Planar Cascades adapts these same techniques to provide a versatile, stable, and efficient calculation scheme for internal flow. The code calculates the compressible, inviscid, irrotational flow through a planar cascade of arbitrary blade shapes. Since the panel solution technique is for incompressible flow, a compressibility correction is introduced to account for compressible flow effects. The analysis is limited to flow conditions in the subsonic and shock-free transonic range. Input to the code consists of inlet flow conditions, blade geometry data, and simple control parameters. Output includes flow parameters at selected control points. This program is written in FORTRAN IV for batch execution and has been implemented on an IBM 370 series computer with a central memory requirement of approximately 590K of 8 bit bytes. This program was developed in 1982.

Description: An exact, full-potential-equation model for the steady, irrotational, homoentropic, and homoenergetic flow of a compressible, inviscid fluid through a two-dimensional planar cascade together with its appropriate boundary conditions has been derived. The CAS2D computer program numerically solves an artificially time-dependent form of the actual full-potential-equation, providing a nonrotating blade-to-blade, steady, potential transonic cascade flow analysis code. Comparisons of results with test data and theoretical solutions indicate very good agreement. In CAS2D, the governing equation is discretized by using type-dependent, rotated finite differencing and the finite area technique. The flow field is discretized by providing a boundary-fitted, nonuniform computational mesh. This mesh is generated by using a sequence of conformal mapping, nonorthogonal coordinate stretching, and local, isoparametric, bilinear mapping functions. The discretized form of the full-potential equation is solved iteratively by using successive line over relaxation. Possible isentropic shocks are captured by the explicit addition of an artificial viscosity in a conservative form. In addition, a four-level, consecutive, mesh refinement feature makes CAS2D a reliable and fast algorithm for the analysis of transonic, two-dimensional cascade flows. The results from CAS2D are not directly applicable to three-dimensional, potential, rotating flows through a cascade of blades because CAS2D does not consider the effects of the Coriolis force that would be present in the three-dimensional case. This program is written in FORTRAN IV for batch execution and has been implemented on an IBM 370 series computer with a central memory requirement of approximately 200K of 8 bit bytes. The CAS2D program was developed in 1980.

Description: A computer program, QSONIC, has been developed for calculating the full potential, transonic quasi-three-dimensional flow through a rotating turbomachinery blade row. The need for lighter, more efficient turbomachinery components has led to the consideration of machines with fewer stages, each with blades capable of higher speeds and higher loading. As speeds increase, the numerical problems inherent in the transonic regime have to be resolved. These problems include the calculation of imbedded shock discontinuities and the dual nature of the governing equations, which are elliptic in the subcritical flow regions but become hyperbolic for supersonic zones. QSONIC provides the flow analyst with a fast and reliable means of obtaining the transonic potential flow distribution on a blade-to-blade stream surface of a stationary or rotating turbomachine blade row. QSONIC combines several promising transonic analysis techniques. The full potential equation in conservative form is discretized at each point on a body-fitted period mesh. A mass balance is calculated through the finite volume surrounding each point. Each local volume is corrected in the third dimension for any change in stream-tube thickness along the stream tube. The nonlinear equations for all volumes are of mixed type (elliptic or hyperbolic) depending on the local Mach number. The final result is a block-tridiagonal matrix formulation involving potential corrections at each grid point as the unknowns. The residual of each system of equations is solved along each grid line. At points where the Mach number exceeds unity, the density at the forward (sweeping) edge of the volume is replaced by an artificial density. This method calculates the flow field about a cascade of arbitrary two-dimensional airfoils. Three-dimensional flow is approximated in a turbomachinery blade row by correcting for stream-tube convergence and radius change in the through flow direction. Several significant assumptions were made in developing the QSONIC program, including: (1) the flow is inviscid and adiabatic, (2) the flow relative to the blade is steady, (3) the fluid is a perfect gas with constant specific heat, (4) the flow is isentropic and any discontinuities (shocks) are weak enough to be approximated as isentropic jumps, (5) there is no velocity component normal to the stream surface, and (6) the flow relative to a fixed frame in space (absolute velocity) is completely irrotational. These assumptions place some limitations on the application of QSONIC. Sharp leading edges at high incidence and high-Mach-number turbine blade trailing edges with substantial deviation will both cause large velocity peaks on the blade. In addition, the program may have difficulty converging if the passage is nearly choked. Input to QSONIC consists of case control parameters, a geometry description, upstream boundary conditions, and a rotor description. Output includes solution scheme parameters and flow field parameters. A data file is also output which contains data on the solution mesh, surface Mach numbers, surface static pressures, isomachs, and the velocity vector field. This data may be used for further processing or for plotting. The QSONIC is written in FORTRAN IV for batch execution and has been implemented on an IBM 370 series computer with a central memory requirement of approximately 500K of 8 bit bytes. QSONIC was developed in 1982.

Description: This computer program, WIND, was developed to numerically solve the exact, full-potential equation for three-dimensional, steady, inviscid flow through an isolated wind turbine rotor. The program automatically generates a three-dimensional, boundary-conforming grid and iteratively solves the full-potential equation while fully accounting for both the rotating and Coriolis effects. WIND is capable of numerically analyzing the flow field about a given blade shape of the horizontal-axis type wind turbine. The rotor hub is assumed representable by a doubly infinite circular cylinder. An arbitrary number of blades may be attached to the hub and these blades may have arbitrary spanwise distributions of taper and of the twist, sweep, and dihedral angles. An arbitrary number of different airfoil section shapes may be used along the span as long as the spanwise variation of all the geometeric parameters is reasonably smooth. The numerical techniques employed in WIND involve rotated, type-dependent finite differencing, a finite volume method, artificial viscosity in conservative form, and a successive overrelaxation combined with the sequential grid refinement procedure to accelerate the iterative convergence rate. Consequently, WIND is cabable of accurately analyzing incompressible and compressible flows, including those that are locally transonic and terminated by weak shocks. Along with the three-dimensional results, WIND provides the results of the two-dimensional calculations to aid the user in locating areas of possible improvement in the aerodynamic design of the blade. Output from WIND includes the chordwise distribution of the coefficient of pressure, the Mach number, the density, and the relative velocity components at spanwise stations along the blade. In addition, the results specify local values of the lift coefficient and the tangent and axial aerodynamic force components. These are also given in integrated form expressing the total torque and the total axial force acting on the shaft. WIND can also be used to analyze the flow around isolated aircraft propellers and helicopter rotors in hover as long as the relative oncoming flow is subsonic. The WIND program is written in FORTRAN IV for batch execution and has been implemented on an IBM 370 series computer with a central memory requirement of approximately 253K of 8 bit bytes. WIND was developed in 1980.

Description: This computer program calculates the flow field in the supersonic portion of a mixed-compression aircraft inlet at non-zero angle of attack. This approach is based on the method of characteristics for steady three-dimensional flow. The results of this program agree with those produced by the two-dimensional method of characteristics when axisymmetric flow fields are calculated. Except in regions of high viscous interaction and boundary layer removal, the results agree well with experimental data obtained for threedimensional flow fields. The flow field in a variety of axisymmetric mixed compression inlets can be calculated using this program. The bow shock wave and the internal shock wave system are calculated using a discrete shock wave fitting procedure. The internal flow field can be calculated either with or without the discrete fitting of the internal shock wave system. The influence of molecular transport can be included in the calculation of the external flow about the forebody and in the calculation of the internal flow when internal shock waves are not discretely fitted. The viscous and thermal diffussion effects are included by treating them as correction terms in the method of characteristics procedure. Dynamic viscosity is represented by Sutherland's law and thermal conductivity is represented as a quadratic function of temperature. The thermodynamic model used is that of a thermally and calorically perfect gas. The program assumes that the cowl lip is contained in a constant plane and that the centerbody contour and cowl contour are smooth and have continuous first partial derivatives. This program cannot calculate subsonic flow, the external flow field if the bow shock wave does not exist entirely around the forebody, or the internal flow field if the bow flow field is injected into the annulus. Input to the program consists of parameters to control execution, to define the geometry, and the vehicle orientation. Output consists of a list of parameters used, solution planes, and a description of the shock waves. This program is written in FORTRAN IV for batch execution and has been implemented on a CDC 6000 series machine with a central memory requirement of 110K (octal) of 60 bit words when it is overlayed. This flow analysis program was developed in 1978.

Description: This computer program was developed for calculating the subsonic or transonic flow on the hub-shroud mid-channel stream surface of a single blade row of a turbomachine. The design and analysis of blades for compressors and turbines ideally requires methods for analyzing unsteady, three-dimensional, turbulent viscous flow through a turbomachine. Since an exact solution is impossible at present, solutions on two-dimensional surfaces are calculated to obtain a quasi-three dimensional solution. When three-dimensional effects are important, significant information can be obtained from a solution on a cross-sectional surface of the passage normal to the flow. With this program, a solution to the equations of flow on the meridional surface can be carried out. This solution is chosen when the turbomachine under consideration has significant variation in flow properties in the hubshroud direction, especially when input is needed for use in blade-to-blade calculations. The program can also perform flow calculations for annular ducts without blades. This program should prove very useful in the design and analysis of any turbomachine. This program calculates a solution for two-dimensional, adiabatic shockfree flow. The flow must be essentially subsonic, but there may be local areas of supersonic flow. To obtain the solution, this program uses both the finite difference and the quasi-orthogonal (velocity gradient) methods combined in a way that takes maximum advantage of both. The finite-difference method solves a finite-difference equation along the meridional stream surface in a very efficient manner but is limited to subsonic velocities. This approach must be used in cases where the blade aspect ratios are above one, cases where the passage is curved, and cases with low hub-tip-ratio blades. The quasi-orthogonal method solves the velocity gradient equation on the meridional surface and is used if it is necessary to extend the range of solutions into the transonic regime. In general the blade row may be fixed or rotating and the blades may be twisted and leaned. The flow may be axial, radial, or mixed. The upstream and downstream flow conditions can vary from hub to shroud with provisions made for an approximate correction for loss of stagnation pressure. Also, viscous forces are neglected along solution mesh lines running from hub to tip. The capabilities of this program include handling of nonaxial flows without restriction, annular ducts without blades, and specified streamwise loss distributions. This program is written in FORTRAN IV for batch execution and has been implemented on an IBM 360 computer with a central memory requirement of approximately 700K of 8 bit bytes. This core requirement can be reduced depending on the size of the problem and the desired solution accuracy. This program was developed in 1977.

Description: A computer program has been developed for the design of supersonic rotor blades where losses are accounted for by correcting the ideal blade geometry for boundary layer displacement thickness. The ideal blade passage is designed by the method of characteristics and is based on establishing vortex flow within the passage. Boundary-layer parameters (displacement and momentum thicknesses) are calculated for the ideal passage, and the final blade geometry is obtained by adding the displacement thicknesses to the ideal nozzle coordinates. The boundary-layer parameters are also used to calculate the aftermixing conditions downstream of the rotor blades assuming the flow mixes to a uniform state. The computer program input consists essentially of the rotor inlet and outlet Mach numbers, upper- and lower-surface Mach numbers, inlet flow angle, specific heat ratio, and total flow conditions. The program gas properties are set up for air. Additional gases require changes to be made to the program. The computer output consists of the corrected rotor blade coordinates, the principal boundary-layer parameters, and the aftermixing conditions. This program is written in FORTRAN IV for batch execution and has been implemented on an IBM 7094. This program was developed in 1971.

Description: This program obtains a transonic flow solution on a blade-to-blade surface between blades of a turbomachine. The flow must be essentially subsonic, but there may be locally supersonic flow. The solution is two-dimensional, isentropic, and shock free. The blades may be fixed or rotating. The flow may be axial, radial, or mixed, and there may be a change in stream-channel thickness in the through-flow direction. A loss in relative stagnation pressure may be accounted for. The program input consists of blade and stream-channel geometry, stagnation flow conditions, inlet and outlet flow angles, and blade-to-blade stream-channel weight flow. The output includes blade surface velocities, velocity magnitude and direction at all interior mesh points in the blade-to-blade passage, and streamline coordinates throughout the passage. The transonic solution is obtained by a combination of a finite-difference, stream-function solution and a velocity-gradient solution. The finite-difference solution at a reduced weight flow provides information needed to obtain a velocity-gradient solution. This program is written in FORTRAN IV for batch execution and has been implemented on the IBM 360 computer with a central memory requirement of approximately 36K of 8 bit bytes. This program was developed in 1969 and last updated in 1979.

Description: This program is a revision of an existing program for blade-to-blade aerodynamic analysis of turbomachine blades and it is a simpler program while consistent with related programs. The analysis is for two-dimensional, subsonic, compressible (or incompressible), nonviscous flow in a circular or straight infinite cascade of blades, which may be fixed or rotating. The flow may be axial, radial, or mixed, and the stream channel thickness may change in the through-flow direction. The program input consists of blade and stream channel geometry, total flow conditions, inlet and outlet flow angles, and blade-to-blade stream channel weight flow. The output includes blade surface velocities, velocity magnitude and direction at all interior mesh points in the blade-to-blade passage, and streamline coordinates throughout the passage. This program was developed on an IBM 7094/7044 DCS.

Description: This computer program gives the blade-to-blade solution of the two-dimensional, subsonic, compressible (or incompressible), nonviscous flow problem for a circular or straight infinite cascade of tandem or slotted turbomachine blades. The blades may be fixed or rotating. The flow may be axial, radial , or mixed. The method of solution is based on the stream function using an iterative solution of nonlinear finite-difference equations. These equations are solved using two major levels of iteration. The inner iteration consists of the solution of simultaneous linear equations by successive over-relaxation, using an estimated optimum over-relaxation factor. The outer iteration then changes the coefficients of the simultaneous equations to correct for compressibility. The program input consists of the basic blade geometry, the meridional stream channel coordinates, fluid stagnation conditions, weight flow and flow split through the slot, and inlet and outlet flow angles. The output includes blade surface velocities, velocity magnitude and direction throughout the passage, and the streamline coordinates.

Description: This FORTRAN IV computer program which incorporates the method of characteristics was written to assist in the design of supersonic inlets. There were two objectives: (1) to study a greater variety of supersonic inlet configurations and (2) to reduce the time required for trial-and-error procedures to arrive at optimum inlet design. The computer program was written with the intention of being able to construct a variety of inlet configurations by interchanging specific subroutines. In this manner, greater flexibility of choice was attained, and the time required to program a specific inlet configuration was greatly reduced. The second objective was accomplished by a reformulation of the boundary value problem for hyperbolic equations. By this reformulation of the boundary data, the engineering design quantities, throat Mach number and flow angle, were introduced as direct input quantities to the computer program. As a consequence of introducing the engineering parameters as input, the computer program will calculate the surface contours required to satisfy the specific throat conditions. Inviscid flow is assumed and the method used to calculate the inlet contour results in minimum distortion to the flow in the throat. This program was developed on an IBM 7094.

Description: This program represents a subsonic aerodynamic method for determining the mean camber surface of trimmed noncoplaner planforms with minimum vortex drag. With this program, multiple surfaces can be designed together to yield a trimmed configuration with minimum induced drag at some specified lift coefficient. The method uses a vortex-lattice and overcomes previous difficulties with chord loading specification. A Trefftz plane analysis is used to determine the optimum span loading for minimum drag. The program then solves for the mean camber surface of the wing associated with this loading. Pitching-moment or root-bending-moment constraints can be employed at the design lift coefficient. Sensitivity studies of vortex-lattice arrangements have been made with this program and comparisons with other theories show generally good agreement. The program is very versatile and has been applied to isolated wings, wing-canard configurations, a tandem wing, and a wing-winglet configuration. The design problem solved with this code is essentially an optimization one. A subsonic vortex-lattice is used to determine the span load distribution(s) on bent lifting line(s) in the Trefftz plane. A Lagrange multiplier technique determines the required loading which is used to calculate the mean camber slopes, which are then integrated to yield the local elevation surface. The problem of determining the necessary circulation matrix is simplified by having the chordwise shape of the bound circulation remain unchanged across each span, though the chordwise shape may vary from one planform to another. The circulation matrix is obtained by calculating the spanwise scaling of the chordwise shapes. A chordwise summation of the lift and pitching-moment is utilized in the Trefftz plane solution on the assumption that the trailing wake does not roll up and that the general configuration has specifiable chord loading shapes. VLMD is written in FORTRAN for IBM PC series and compatible computers running MS-DOS. This program requires 360K of RAM for execution. The Ryan McFarland FORTRAN compiler and PLINK86 are required to recompile the source code; however, a sample executable is provided on the diskette. The standard distribution medium for VLMD is a 5.25 inch 360K MS-DOS format diskette. VLMD was originally developed for use on CDC 6000 series computers in 1976. It was originally ported to the IBM PC in 1986, and, after minor modifications, the IBM PC port was released in 1993.

Description: This code was developed to aid design engineers in the selection and evaluation of aerodynamically efficient wing-canard and wing-horizontal-tail configurations that may employ simple hinged-flap systems. Rapid estimates of the longitudinal aerodynamic characteristics of conceptual airplane lifting surface arrangements are provided. The method is particularly well suited to configurations which, because of high speed flight requirements, must employ thin wings with highly swept leading edges. The code is applicable to wings with either sharp or rounded leading edges. The code provides theoretical pressure distributions over the wing, the canard or horizontal tail, and the deflected flap surfaces as well as estimates of the wing lift, drag, and pitching moments which account for attainable leading edge thrust and leading edge separation vortex forces. The wing planform information is specified by a series of leading edge and trailing edge breakpoints for a right hand wing panel. Up to 21 pairs of coordinates may be used to describe both the leading edge and the trailing edge. The code has been written to accommodate 2000 right hand panel elements, but can easily be modified to accommodate a larger or smaller number of elements depending on the capacity of the target computer platform. The code provides solutions for wing surfaces composed of all possible combinations of leading edge and trailing edge flap settings provided by the original deflection multipliers and by the flap deflection multipliers. Up to 25 pairs of leading edge and trailing edge flap deflection schedules may thus be treated simultaneously. The code also provides for an improved accounting of hinge-line singularities in determination of wing forces and moments. To determine lifting surface perturbation velocity distributions, the code provides for a maximum of 70 iterations. The program is constructed so that successive runs may be made with a given code entry. To make additional runs, it is necessary only to add an identification record and the namelist data that are to be changed from the previous run. This code was originally developed in 1989 in FORTRAN V on a CDC 6000 computer system, and was later ported to an MS-DOS environment. Both versions are available from COSMIC. There are only a few differences between the PC version (LAR-14458) and CDC version (LAR-14178) of AERO2S distributed by COSMIC. The CDC version has one main source code file while the PC version has two files which are easier to edit and compile on a PC. The PC version does not require a FORTRAN compiler which supports NAMELIST because a special INPUT subroutine has been added. The CDC version includes two MODIFY decks which can be used to improve the code and prevent the possibility of some infrequently occurring errors while PC-version users will have to make these code changes manually. The PC version includes an executable which was generated with the Ryan McFarland/FORTRAN compiler and requires 253K RAM and an 80x87 math co-processor. Using this executable, the sample case requires about four hours to execute on an 8MHz AT-class microcomputer with a co-processor. The source code conforms to the FORTRAN 77 standard except that it uses variables longer than six characters. With two minor modifications, the PC version should be portable to any computer with a FORTRAN compiler and sufficient memory. The CDC version of AERO2S is available in CDC NOS Internal format on a 9-track 1600 BPI magnetic tape. The PC version is available on a set of two 5.25 inch 360K MS-DOS format diskettes. IBM AT is a registered trademark of International Business Machines. MS-DOS is a registered trademark of Microsoft Corporation. CDC is a registered trademark of Control Data Corporation. NOS is a trademark of Control Data Corporation.

Description: This program provides a wing design algorithm based on modified linear theory which takes into account the effects of attainable leading-edge thrust. A primary objective of the WINGDES2 approach is the generation of a camber surface as mild as possible to produce drag levels comparable to those attainable with full theoretical leading-edge thrust. WINGDES2 provides both an analysis and a design capability and is applicable to both subsonic and supersonic flow. The optimization can be carried out for designated wing portions such as leading and trailing edge areas for the design of mission-adaptive surfaces, or for an entire planform such as a supersonic transport wing. This program replaces an earlier wing design code, LAR-13315, designated WINGDES. WINGDES2 incorporates modifications to improve numerical accuracy and provides additional capabilities. A means of accounting for the presence of interference pressure fields from airplane components other than the wing and a direct process for selection of flap surfaces to approach the performance levels of the optimized wing surfaces are included. An increased storage capacity allows better numerical representation of those configurations that have small chord leading-edge or trailing-edge design areas. WINGDES2 determines an optimum combination of a series of candidate surfaces rather than the more commonly used candidate loadings. The objective of the design is the recovery of unrealized theoretical leading-edge thrust of the input flat surface by shaping of the design surface to create a distributed thrust and thus minimize drag. The input consists of airfoil section thickness data, leading and trailing edge planform geometry, and operational parameters such as Mach number, Reynolds number, and design lift coefficient. Output includes optimized camber surface ordinates, pressure coefficient distributions, and theoretical aerodynamic characteristics. WINGDES2 is written in FORTRAN V for batch execution and has been implemented on a CDC CYBER computer operating under NOS 2.7.1 with a central memory requirement of approximately 344K (octal) of 60 bit words. This program was developed in 1984, and last updated in 1990. CDC and CYBER are trademarks of Control Data Corporation.

Description: This code was developed to aid design engineers in the selection and evaluation of aerodynamically efficient wing-canard and wing-horizontal-tail configurations that may employ simple hinged-flap systems. Rapid estimates of the longitudinal aerodynamic characteristics of conceptual airplane lifting surface arrangements are provided. The method is particularly well suited to configurations which, because of high speed flight requirements, must employ thin wings with highly swept leading edges. The code is applicable to wings with either sharp or rounded leading edges. The code provides theoretical pressure distributions over the wing, the canard or horizontal tail, and the deflected flap surfaces as well as estimates of the wing lift, drag, and pitching moments which account for attainable leading edge thrust and leading edge separation vortex forces. The wing planform information is specified by a series of leading edge and trailing edge breakpoints for a right hand wing panel. Up to 21 pairs of coordinates may be used to describe both the leading edge and the trailing edge. The code has been written to accommodate 2000 right hand panel elements, but can easily be modified to accommodate a larger or smaller number of elements depending on the capacity of the target computer platform. The code provides solutions for wing surfaces composed of all possible combinations of leading edge and trailing edge flap settings provided by the original deflection multipliers and by the flap deflection multipliers. Up to 25 pairs of leading edge and trailing edge flap deflection schedules may thus be treated simultaneously. The code also provides for an improved accounting of hinge-line singularities in determination of wing forces and moments. To determine lifting surface perturbation velocity distributions, the code provides for a maximum of 70 iterations. The program is constructed so that successive runs may be made with a given code entry. To make additional runs, it is necessary only to add an identification record and the namelist data that are to be changed from the previous run. This code was originally developed in 1989 in FORTRAN V on a CDC 6000 computer system, and was later ported to an MS-DOS environment. Both versions are available from COSMIC. There are only a few differences between the PC version (LAR-14458) and CDC version (LAR-14178) of AERO2S distributed by COSMIC. The CDC version has one main source code file while the PC version has two files which are easier to edit and compile on a PC. The PC version does not require a FORTRAN compiler which supports NAMELIST because a special INPUT subroutine has been added. The CDC version includes two MODIFY decks which can be used to improve the code and prevent the possibility of some infrequently occurring errors while PC-version users will have to make these code changes manually. The PC version includes an executable which was generated with the Ryan McFarland/FORTRAN compiler and requires 253K RAM and an 80x87 math co-processor. Using this executable, the sample case requires about four hours to execute on an 8MHz AT-class microcomputer with a co-processor. The source code conforms to the FORTRAN 77 standard except that it uses variables longer than six characters. With two minor modifications, the PC version should be portable to any computer with a FORTRAN compiler and sufficient memory. The CDC version of AERO2S is available in CDC NOS Internal format on a 9-track 1600 BPI magnetic tape. The PC version is available on a set of two 5.25 inch 360K MS-DOS format diskettes. IBM AT is a registered trademark of International Business Machines. MS-DOS is a registered trademark of Microsoft Corporation. CDC is a registered trademark of Control Data Corporation. NOS is a trademark of Control Data Corporation.

Description: Since its early beginnings, NASA has been actively involved in the design and testing of airfoil sections for a wide variety of applications. Recently a set of programs has been developed to smooth and scale arbitrary airfoil coordinates. The smoothing program, AFSMO, utilizes both least-squares polynomial and least-squares cubic-spline techniques to iteratively smooth the second derivatives of the y-axis airfoil coordinates with respect to a transformed x-axis system which unwraps the airfoil and stretches the nose and trailing-edge regions. The corresponding smooth airfoil coordinates are then determined by solving a tridiagonal matrix of simultaneous cubic-spline equations relating the y-axis coordinates and their corresponding second derivatives. The camber and thickness distribution of the smooth airfoil are also computed. The scaling program, AFSCL, may then be used to scale the thickness distribution generated by the smoothing program to a specified maximum thickness. Once the thickness distribution has been scaled, it is combined with the camber distribution to obtain the final scaled airfoil contour. The airfoil smoothing and scaling programs are written in FORTRAN IV for batch execution and have been implemented on a CDC CYBER 170 series computer with a central memory requirement of approximately 70K (octal) of 60 bit words. Both programs generate plotted output via CALCOMP type plotting calls. These programs were developed in 1983.

Description: The Supersonic Wing Nonlinear Aerodynamics computer program, LTSTAR, was developed to provide for the estimation of the nonlinear aerodynamic characteristics of a wing at supersonic speeds. This corrected linearized-theory method accounts for nonlinearities in the variation of basic pressure loadings with local surface slopes, predicts the degree of attainment of theoretical leading-edge thrust forces, and provides an estimate of detached leading-edge vortex loadings that result when the theoretical thrust forces are not fully realized. Comparisons of LTSTAR computations with experimental results show significant improvements in detailed wing pressure distributions, particularly for large angles of attack and for regions of the wing where the flow is highly three-dimensional. The program provides generally improved predictions of the wing overall force and moment coefficients. LTSTAR could be useful in design studies aimed at aerodynamic performance optimization and for providing more realistic trade-off information for selection of wing planform geometry and airfoil section parameters. Input to the LTSTAR program includes wing planform data, freestream conditions, wing camber, wing thickness, scaling options, and output options. Output includes pressure coefficients along each chord, section normal and axial force coefficients, and the spanwise distribution of section force coefficients. With the chordwise distributions and section coefficients at each angle of attack, three sets of polars are output. The first set is for linearized theory with and without full leading-edge thrust, the second set includes nonlinear corrections, and the third includes estimates of attainable leading-edge thrust and vortex increments along with the nonlinear corrections. The LTSTAR program is written in FORTRAN IV for batch execution and has been implemented on a CDC 6000 series computer with a central memory requirement of approximately 150K (octal) of 60 bit words. The LTSTAR program was developed in 1980.

Description: The nozzle afterbody is one of the main drag-producing components of an aircraft propulsion system. Thus, considerable effort has been devoted to developing techniques for predicting the afterbody flow field and drag. The RAXJET computer program was developed to predict the transonic, axisymmetric flow over nozzle afterbodies with supersonic jet exhausts and includes the effects of boundary-layer displacement, separation, jet entrainment, and inviscid jet plume blockage. RAXJET iteratively combines the South-Jameson relaxation procedure, the Reshotko-Tucker boundary-layer solution, the Presz separation model, the Dash-Pergament mixing model, and the Dash-Thorpe inviscid plume model into a single, comprehensive model. The approach taken in the RAXJET program requires considerably less computational time than the Navier-Stokes solutions and generally yields results of comparable accuracy. In RAXJET, the viscous-inviscid interaction model is constructed by dividing the afterbody flow field into six separate computational regions: (1) The inviscid external flow solution is based on the relaxation procedure of South and Jameson for solving the exact nonlinear potential flow equation in nonconservative form. (2) The flow field in the inviscid jet exhaust is solved by explicit spatial marching of the conservative finite-difference form of the inviscid flow equations for a uniform composition gas mixture. (3) The properties in the attached boundary-layer region are solved by a modified version of the Reshotko-Tucker integral method for turbulent flows. (4) The analysis of the separated flow region consists of predicting the separation location and calculating the discriminating streamline shape. (5) The jet wake region is determined by either a simple extrapolation model or by an integral method that accounts for entrainment effects. (6) The displacement-thickness distribution arising from entrainment into the jet mixing layer is calculated by the overlaid mixing model. The inviscid external flow solution and inviscid jet exhaust solution provide the necessary flow conditions to calculate the flow in the viscous regions. The viscous and inviscid flow fields are iteratively solved until a final solution is obtained. Input to the RAXJET program consists of body geometry data, free-stream conditions, main logic control parameters, and condition and control parameters for each of the six computational flow regions. Output from RAXJET includes detailed flow results and aerodynamic coefficients. The RAXJET program is written in FORTRAN IV for batch execution and has been implemented on a CDC CYBER 170 series computer with a central memory requirement of approximately 60K(octal) of 60 bit words. The RAXJET program was developed in 1982.

Description: The computer program SALLY was developed to compute the incompressible linear stability characteristics and integrate the amplification rates of boundary layer disturbances on swept and tapered wings. For some wing designs, boundary layer disturbance can significantly alter the wing performance characteristics. This is particularly true for swept and tapered laminar flow control wings which incorporate suction to prevent boundary layer separation. SALLY should prove to be a useful tool in the analysis of these wing performance characteristics. The first step in calculating the disturbance amplification rates is to numerically solve the compressible laminar boundary-layer equation with suction for the swept and tapered wing. A two-point finite-difference method is used to solve the governing continuity, momentum, and energy equations. A similarity transformation is used to remove the wall normal velocity as a boundary condition and place it into the governing equations as a parameter. Thus the awkward nonlinear boundary condition is avoided. The resulting compressible boundary layer data is used by SALLY to compute the incompressible linear stability characteristics. The local disturbance growth is obtained from temporal stability theory and converted into a local growth rate for integration. The direction of the local group velocity is taken as the direction of integration. The amplification rate, or logarithmic disturbance amplitude ratio, is obtained by integration of the local disturbance growth over distance. The amplification rate serves as a measure of the growth of linear disturbances within the boundary layer and can serve as a guide in transition prediction. This program is written in FORTRAN IV and ASSEMBLER for batch execution and has been implemented on a CDC CYBER 70 series computer with a central memory requirement of approximately 67K (octal) of 60 bit words. SALLY was developed in 1979.

Description: The Program for Solving the General-Frequency Unsteady Two-Dimensional Transonic Small-Disturbance Equation, XTRAN2L, is used to calculate time-accurate, finite-difference solutions of the nonlinear, small-disturbance potential equation for two- dimensional transonic flow about airfoils. The code can treat forced harmonic, pulse, or aeroelastic transient type motions. XTRAN2L uses a transonic small-disturbance equation that incorporates a time accurate finite-difference scheme. Airfoil flow tangency boundary conditions are defined to include airfoil contour, chord deformation, nondimensional plunge displacement, pitch, and trailing edge control surface deflection. Forced harmonic motion can be based on: 1) coefficients of harmonics based on information from each quarter period of the last cycle of harmonic motion; or 2) Fourier analyses of the last cycle of motion. Pulse motion (an alternate to forced harmonic motion) in which the airfoil is given a small prescribed pulse in a given mode of motion, and the aerodynamic transients are calculated. An aeroelastic transient capability is available within XTRAN2L, wherein the structural equations of motion are coupled with the aerodynamic solution procedure for simultaneous time-integration. The wake is represented as a slit downstream of the airfoil trailing edge. XTRAN2L includes nonreflecting farfield boundary conditions. XTRAN2L was developed on a CDC CYBER mainframe running under NOS 2.4. It is written in FORTRAN 5 and uses overlays to minimize storage requirements. The program requires 120K of memory in overlayed form. XTRAN2L was developed in 1987.

Description: The NASCRIN program was developed for analyzing two-dimensional flow fields in supersonic combustion ramjet (scramjet) inlets. NASCRIN solves the two-dimensional Euler or Navier-Stokes equations in conservative form by an unsplit, explicit, two-step finite-difference method. A more recent explicit-implicit, two-step scheme has also been incorporated in the code for viscous flow analysis. An algebraic, two-layer eddy-viscosity model is used for the turbulent flow calculations. NASCRIN can analyze both inviscid and viscous flows with no struts, one strut, or multiple struts embedded in the flow field. NASCRIN can be used in a quasi-three-dimensional sense for some scramjet inlets under certain simplifying assumptions. Although developed for supersonic internal flow, NASCRIN may be adapted to a variety of other flow problems. In particular, it should be readily adaptable to subsonic inflow with supersonic outflow, supersonic inflow with subsonic outflow, or fully subsonic flow. The NASCRIN program is available for batch execution on the CDC CYBER 203. The vectorized FORTRAN version was developed in 1983. NASCRIN has a central memory requirement of approximately 300K words for a grid size of about 3,000 points.

Description: The problem of axisymmetric transonic flow is of interest not only because of the practical application to missile and launch vehicle aerodynamics, but also because of its relation to fully three-dimensional flow in terms of the area rule. The RAXBOD computer program was developed for the analysis of steady, inviscid, irrotational, transonic flow over axisymmetric bodies in free air. RAXBOD uses a finite-difference relaxation method to numerically solve the exact formulation of the disturbance velocity potential with exact surface boundary conditions. Agreement with available experimental results has been good in cases where viscous effects and wind-tunnel wall interference are not important. The governing second-order partial differential equation describing the flow potential is replaced by a system of finite difference equations, including Jameson's "rotated" difference scheme at supersonic points. A stretching is applied to both the normal and tangential coordinates such that the infinite physical space is mapped onto a finite computational space. The boundary condition at infinity can be applied directly and there is no need for an asymptotic far-field solution. The system of finite difference equations is solved by a column relaxation method. In order to obtain both rapid convergence and any desired resolution, the relaxation is performed iteratively on successively refined grids. Input to RAXBOD consists of a description of the body geometry, the free stream conditions, and the desired resolution control parameters. Output from RAXBOD includes computed geometric parameters in the normal and tangential directions, iteration history information, drag coefficients, flow field data in the computational plane, and coordinates of the sonic line. This program is written in FORTRAN IV for batch execution and has been implemented on a CDC 6600 computer with an overlayed central memory requirement of approximately 40K (octal) of 60 bit words. Optional plotted output can be generated for the Calcomp plotting system. The RAXBOD program was developed in 1976.

Description: Two separate and distinct theories are incorporated in this computer program to estimate the lift-induced pressures existent on a wing-body combination. These are (1) the second-order shock-expansion theory, which is used to obtain the lifting pressures on the body alone at small angles of attack, and (2) the linear-theory integral equations, which is used to evaluate the lifting pressures induced by the wing. These equations relate the local surface slope at a point on the lifting surface to the pressure differential at the point and the influence of the pressures upstream of the point. The numerical solution of these equations is effected by treating the wing-planform as a composite of elemental rectangles and applying summation techniques to satisfy the necessary integral relations. Most of the input required by this program is involved with the description of the missile planform geometry. The output consists of the computed value of the lifting pressure slope (the differential pressure coefficient per degree angle of attack) for each of the elements in the planform array. A force and moment summary is presented for the configuration under consideration.

Description: A modified strip analysis has been developed for rapidly predicting flutter of finite-span, swept or unswept wings at subsonic to hypersonic speeds. The method employs distributions of aerodynamic parameters which may be evaluated from any suitable linear or nonlinear steady-flow theory or from measured steady-flow load distributions for the underformed wing. The method has been shown to give good flutter results for a broad range of wings at Mach number from 0 to as high as 15.3. The principles of the modified strip analysis may be summarized as follows: Variable section lift-curve slope and aerodynamic center are substituted respectively, for the two-dimensional incompressible-flow values of 2 pi and quarter chord which were employed by Barmby, Cunningham, and Garrick. Spanwise distributions of these steady-flow section aerodynamic parameters, which are pertinent to the desired planform and Mach number, are used. Appropriate values of Mach number-dependent circulation functions are obtained from two-dimensional unsteady compressible-flow theory. Use of the modified strip analysis avoids the necessity of reevaluating a number of loading parameters for each value of reduced frequency, since only the modified circulation functions, and of course the reduced frequency itself, vary with frequency. It is therefore practical to include in the digital computing program a very brief logical subroutine, which automatically selects reduced-frequency values that converge on a flutter solution. The problem of guessing suitable reduced-frequency values is thus eliminated, so that a large number of flutter points can be completely determined in a single brief run on the computing machine. If necessary, it is also practical to perform the calculations manually. Flutter characteristics have been calculated by the modified strip analysis and compared with results of other calculations and with experiments for Mach numbers up to 15.3 and for wings with sweep angles from 0 degrees to 52.5 degrees, aspect ratios from 2.0 to 7.4, taper ratios from 0.2 to 1.0, and center-of-gravity positions between 34% chord and 59% chord. These ranges probably cover the great majority of wings that are of practical interest with the exception of very low-aspect-ratio surfaces such as delta wings and missile fins. This program has been implemented on the IBM 7094.

Description: The SUBAERF2 program was developed to provide for the aerodynamic analysis and design of low speed wing flap systems. SUBAERF2 is based on a linearized theory lifting surface solution. It is particularly well suited to configurations which, because of high speed flight requirements, must employ thin wings with highly swept leading edges. The program is applicable to wings with either sharp or rounded leading edges. This program is a new and improved version of LAR-13116 and LAR-12987, which it replaces. The low speed aerodynamic analysis method used in SUBAERF2 provides estimates of wing performance which include the effects of attainable leading-edge thrust and vortex lift. This basic aerodynamic analysis method has been improved to provide for the convenient, efficient and accurate treatment of simple leading-edge and trailing-edge flap systems. The user inputs flap geometry directly. Solutions can be found for various combinations of leading and trailing edge flap deflections. The program provides for the simultaneous analysis of up to 25 pairs of leading-edge and trailing-edge flap deflection schedules. A revised attainable thrust algorithm improves accuracy at the low Mach numbers sometimes encountered in wind tunnel testing. Also added is a means of estimating the distribution of leading edge separation vortex forces. The revised program has been particularly useful in the subsonic analysis of vehicles designed for supersonic cruise. The SUBAERF2 program is written in FORTRAN V for batch execution and has been implemented on a CDC 175 computer operating under NOS 2.4 with a central memory requirement of approximately 115K (octal) of 60 bit words. This program was originally developed in 1983 and later revised in 1988.

Description: WAVDRAG calculates the supersonic zero-lift wave drag of complex aircraft configurations. The numerical model of an aircraft is used throughout the design process from concept to manufacturing. WAVDRAG incorporates extended geometric input capabilities to permit use of a more accurate mathematical model. With WAVDRAG, the engineer can define aircraft components as fusiform or nonfusiform in terms of non-intersecting contours in any direction or more traditional parallel contours. In addition, laterally asymmetric configurations can be simulated. The calculations in WAVDRAG are based on Whitcomb's area-rule computation of equivalent-bodies, with modifications for supersonic speed. Instead of using a single equivalent-body, WAVDRAG calculates a series of equivalent-bodies, one for each roll angle. The total aircraft configuration wave drag is the integrated average of the equivalent-body wave drags through the full roll range of 360 degrees. WAVDRAG currently accepts up to 30 user-defined components containing a maximum of 50 contours as geometric input. Each contour contains a maximum of 50 points. The Mach number, angle-of-attack, and coordinates of angle-of-attack rotation are also input. The program warns of any fusiform-body line segments having a slope larger than the Mach angle. WAVDRAG calculates total drag and the wave-drag coefficient of the specified aircraft configuration. WAVDRAG is written in FORTRAN 77 for batch execution and has been implemented on a CDC CYBER 170 series computer with a central memory requirement of approximately 63K (octal) of 60 bit words. This program was developed in 1983.

Description: An approximate inverse solution is presented for the nonequilibrium flow in the inviscid shock layer about a vehicle in hypersonic flight. The method is based upon a thin-shock-layer approximation and has the advantage of being applicable to both subsonic and supersonic regions of the shock layer. The relative simplicity of the method makes it ideally suited for programming on a digital computer with a significant reduction in storage capacity and computing time required by other more exact methods. Comparison of nonequilibrium solutions for an air mixture obtained by the present method is made with solutions obtained by two other methods. Additional cases are presented for entry of spherical nose cones into representative Venusian and Matrian atmospheres. A digital computer program written in FORTRAN language is presented that permits an arbitrary gas mixture to be employed in the solution. The effects of vibration, dissociation, recombination, electronic excitation, and ionization are included in the program.

Description: Small areas of high heat transfer and pressure can occur on a vehicle surface due to the influence of an impinging shock on the local flow. A method was needed to determine peak pressure and heating of these areas. This package is a system of computer programs designed to calculate two-dimensional shock interference patterns for six types of interference flows. Results also include properties of the inviscid flow field and the inviscid-viscous interaction at the surface along with peak pressure and peak heating at the impingement point. The six types of interference flow patterns considered are: 1) Type I interference patterns, occurring when two weak shocks of opposite families, BS (bow shock) and IS (impingment shock), intersect when the flow upstream of the impingement point is supersonic, or in the case of a blunt body, takes place well below the sonic point. 2) Type II interference pattern occurs when two shocks of opposite families (bow shock and impinging shock) intersect. Both shocks are weak as in type I, but are of such strength that in order to turn the flow, a Mach reflection must exist in the center of the flow field with an embedded subsonic region occurring between the intersection points (A & B) and the accompanying shear layers. Type II interference occurs on a blunt body when the impinging shock intersects the bow shock near the sonic point. 3) Type III shock interference pattern occurs when a weak impinging shock intersects a strong detached bow shock. On a blunt body the shock intersection occurs near or above the lower sonic point. 4) Type IV interference can occur when the impinging shock intersects a strong bow shock ahead of a subsonic flow region. On a blunt body this shock intersection is located between the lower sonic point and just above the body axis. The impinging shock causes a displacement of the bow shock and the formation of a supersonic jet that is embedded in the subsonic region. A jet bow shock is produced when the jet impinges on the surface, creating a small region with high stagnation heating. 5) Type V interference involves the interaction of two weak shocks of the same family. The interaction produces a shear layer, a supersonic jet, and a transmitted impinging shock. On a blunt body the shock interaction occurs near the upper sonic point. 6) Type VI interference involves the intersection of two weak shocks of the same family, which leads to an entirely supersonic flow field. This type of interference is important because it provides a means for predicting the onset of type V. Peak-heating correlations for laminar and turbulent shock-boundary-layer interactions are included in the programs for types I, II, V, and VI interference patterns. Heating correlations for laminar and turbulent reattaching shear layers obtained from separation studies are included in the program for type III interference. This program is written in FORTRAN IV for batch execution and has been implemented on a CDC 6000 Series computer. This program was developed in 1973.

Description: A computer program has been written to obtain the wave and friction drag of configurations with bodies of revolution and fins. These inviscid flow fields are superimposed and the wave drag of the configuration is obtained by integration of the surface pressures. The friction drag is obtained from the viscous flow field of the body and a flat-plate friction analysis of the fins. The numerical solution of these flow fields, superposition, and integration to obtain total drag have been programmed for high-speed digital computation. A large portion of the input required by the program is involved with the description of the configuration geometry and the specific surface positions where pressures are to be evaluated. In addition to drag forces, an output is available whereby the pressure distributions on the body and fins can be obtained.

Description: This paper is on the control of nonlinear-nonstationary vibration of a frame-stringer structure resulting from high levels of excitaation from a nearby supersonic jet exhaust. The structure exhibits periodic, chaotic, or random behaviors when forced by high-intensity sound from a supersonic jet exhaust with shock loading superimposed on the broadband response. The time history of the pressure, showing the rotation and flapping of the shock structure in the jet column due to large-scale instabilities, indicates that the response is not only nonlinear but also nonstationary. The acoustic pressure radiated by the structure also contains shocks and the formation of harmonics with distance. Control of the structural response is achieved by actively forcing the structure with an actuator at the shock oscillation frequency whose amplitude is locked into a self-control cycle. Results show that the peak power level is reduced by a factor of 63, or 18 dB. As a result, new broadband components emerge with at least four harmonics. At accelerating and decelerating supersonic speeds, the exhaust from the jet induces higher transient loading on the nearby flexible structure due to the occurence of multiple shocks from the jet.

Description: The transitional flight characteristics of a geometrically simplified Short Take-Off Vertical Landing (STOVL) aircraft configuration have been measured in the NASA Ames 7- by 10-Foot Wind Tunnel. The experiment is the first in a sequence of tests designed to provide detailed data for evaluating the capability of computational fluid dynamics methods to predict the important flow parameters for powered lift. The model consists of a 60 deg cropped delta wing platform, blended fuselage and two circular in-line jets that exit perpendicularly from the flat lower surface. The measured flows have a maximum freestream Mach number of 0.2. Model angle of attack is varied between -10 and +20 deg. The flow is ambient temperature in both jet exits and the nozzle pressure ratios are varied between 1 and 3. The data presented includes forces and moments, pressures measured at 281 surface pressure ports and the pressures of the jets. Measurements of the flow are also made in the tunnel test section upstream and downstream of the model and at the jet exits to guide boundary condition selection for the planned computations. Flow visualization and total pressure measurements in the jet plumes provide a description of the three-dimensional jet efflux flowfield.

Description: This test was originally conducted to determine the effects of several empennage and afterbody parameters on the aft-end aerodynamic characteristics of a twin-engine fighter-type configuration. Model variables were as follows: horizontal tail axial location and incidence, vertical tail axial location and configuration (twin-vs single-tail arrangements), tail booms, and nozzle power setting. Jet propulsion was simulated by exhausting high-pressure, cold-flow air from the nozzles. Following a successful test conducted on a single engine nacelle model to validate a CFD code, this model was chosen to be instrumented with pressure taps on the afterbody and nozzles and used as a follow-on test, providing a more complex geometry for the CFD code validation. A more limited test matrix was run to collect the pressure data, employing only the twin-tail configuration and varying only the horizontal and vertical tail locations. Mach number was varied from 0.6 to 1.2. Nozzle pressure ratio was varied from jet-off to 8. Angle-of-attack varied from 0 to 8 deg.

Description: This test was initiated to provide validation data on low aspect ratio wings at transonic speeds. The test was conducted so that the data obtained would be useful in the validation of codes, and all boundary condition data required would be measured as part of the test. During the conduct of the test, the measured quantities were checked for repeatability, and when the data would not repeat, the cause was tracked down and either eliminated or included in the measurement uncertainty. The accuracy of the data was in the end limited by wall imperfections of the wind tunnel in which the test was run.

Description: The data presented in this contribution were obtained in the NASA Langley 16-Foot Transonic Tunnel. Multiple test entries were completed and the results have been completely reported in five NASA reports. The objective of the initial investigation was to determine the effect of empennage (tail) interference on the drag characteristics of an axisymmetric model with a single engine fighter aft-end with convergent divergent nozzles. Two nozzle power settings, dry and maximum afterburning, were investigated. Several empennage arrangements and afterbody modifications were investigated during the initial investigation. Subsequent investigations were used to determine the effects of other model variables including tail incidence, tail span, and nozzle shape. For the final investigation, extensive surface pressure instrumentation was added to the model in order to develop and understanding of the flow interactions associated with afterbody/empennage integration and also to provide data for code validation. Extensive computational analysis has been conducted on the staggered empennage configuration at a Mach number of 0.6 utilizing a three-dimensional Navier Stokes code. Most of the investigations were conducted at Mach numbers from 0.60 to 1.20 and at ratios of jet total pressure to free stream static pressure (nozzle pressure ratio) from 0.1 (jet off) to 8.0. Some angle of attack variation was obtained at jet off conditions.

Description: The purpose of this investigation is to provide a comprehensive data base for the validation of numerical simulations. The initial results of the study (single angle of attack) were presented in ref. 1, where the effects of various parameters and the adequacies of selected turbulence models were discussed. The objective of the present paper is to provide a tabulation of the experimental data. The data were obtained in the two-dimensional, transonic flowfield surrounding a supercritical airfoil. A variety of flows were studied in which the boundary layer at the trailing edge of the model was either attached or separated. Unsteady flows were avoided by controlling the Mach number and angle of attack. Surface pressures were measured on both the model and wind tunnel walls, and the flowfield surrounding the model was documented using a laser Doppler velocimeter (LDV). Although wall interference could not be completely eliminated, its effect was minimized by employing the following techniques. Sidewall boundary layers were reduced by aspiration, and upper and lower walls were contoured to accommodate the flow around the model and the boundary-layer growth on the tunnel walls. A data base with minimal interference from a tunnel with solid walls provides an ideal basis for evaluating the development of codes for the transonic speed range because the codes can include the wall boundary conditions more precisely than interference corrections can be made to the data sets.

Description: The computation of unsteady shock waves, which contribute significantly to noise generation in supersonic jet flows, is investigated. This paper focuses on the difficulties of computing slowly moving shock waves. Numerical error is found to manifest itself principally as a spurious entropy wave. Calculations presented are performed using a third order essentially nonoscillatory scheme. The effect of stencil biasing parameters and of two versions of numerical flux formulas on the magnitude of spurious entropy are investigated. The level of numerical error introduced in the calculation in quantified as a function of shock pressure ratio, shock speed, Courant number, and mesh density. The spurious entropy relative to the entropy jump across a static shock decreases with increasing shock strength and shock velocity relative to the grid, but is insensitive to Courant number. The structure of the spurious entropy wave is affected by the choice of flux formulas and algorithm biasing parameters. The effect of the spurious numerical waves on the calculation of sound amplification by a shock wave is investigated. For this class of problem, the acoustic pressure waves are relatively unaffected by the spurious numerical phenomena.

Description: The structure of the vortical flowfield over delta wings at high angles of attack was investigated. Three-dimensional Navier-Stokes numerical simulations were carried out to predict the complex leeward-side flowfield characteristics, including leading-edge separation, secondary separation, and vortex breakdown. Flows over a 75- and a 63-deg sweep delta wing with sharp leading edges were investigated and compared with available experimental data. The effect of variation of circumferential grid resolution grid resolution in the vicinity of the wing leading edge on the accuracy of the solutions was addressed. Furthermore, the effect of turbulence modeling on the solutions was investigated. The effects of variation of angle of attack on the computed vortical flow structure for the 75-deg sweep delta wing were examined. At moderate angles of attack no vortex breakdown was observed. When a critical angle of attack was reached, bubble-type vortex breakdown was found. With further increase in angle of attack, a change from bubble-type breakdown to spiral-type vortex breakdown was predicted by the numerical solution. The effects of variation of sweep angle and freestream Mach number were addressed with the solutions on a 63-deg sweep delta wing.

Description: Results are presented from an investigation of the structure of energized wakes in the presence of ground or ceiling planes when powered-lift devices are tested at zero and low wind tunnel velocities. Solutions are presented to indicate how the presence of a nearby ground plane causes the energized wake to constrict less than when in free space. In contrast, another set of solutions indicate that the presence of a ceiling plane enhances wake construction so that the wake width is even smaller than when in free space. Hence, a ground plane tends to reduce the chance for interaction between the wake and the energizing elements, while a nearby ceiling plane tends to increase the likelihood for interference.

Description: Despite the extensive experimental and computational data base in the literature on passive porosity, no clear explanation of the governing flow physics exists. It is theorized that the positive porosity concept modifies the external pressure loading by allowing communication between high- and low-pressure regions on the external surface. This study determines the dominant flow phenomena that govern the effectiveness of passive porosity. It aims to assess the contribution of each phenomenon as related to a porous slender axisymmetric forebody. To assess the influence of the mass transfer and pressure equalization phenomena on the effectiveness of passive porosity on slender axisymmetric forebodies, strakes were attached to the 5.0-caliber solid and porous forebodies to force crossflow separation. Longitudinal force and moment data were obtained at a Mach number of 0.1 over an angle-of-attack range of 0 to 55 deg.

Description: A full-scale F/A-18 was tested in the 80 by 120-Foot Wind Tunnel at NASA Ames Research Center to measure the effectiveness of a tangentially blowing slot in generating significant yawing moments while minimizing coupling in the pitch and roll axes. Various slot configurations were tested to determine the optimum configuration. The test was conducted for angles of attack from 25 to 50 deg, angles of sideslip from -15 to +15 deg, and freestream velocities from 67 ft/sec to 168 ft/sec. By altering the forebody vortex flow, yaw control was maintained for angles of attack up to 50 deg. Of particular interest was the result that blowing very close to the radome apex was not as effective as blowing slightly farther aft on the radome, that a 16-inch slot was more efficient, and that yawing moments were generated without inducing significant rolling or pitching moments.

Description: A ray theory approach is used to examine the propagation of sonic booms through a turbulent ground layer, and to make predictions about the received waveform. The rays are not propagated one at a time, as is typical in ray theory; instead, sufficient rays to represent a continuous wave front are propagated together. New rays are interpolated as needed to maintain the continuity of the wave front. In order to predict the received boom signature, the wave front is searched for eigenrays after it has propagated to the receiver. The Comte-Bellot turbulence model is used to generate an instantaneous 'snapshot' of the turbulent field. The transient acoustic wave is assumed to be sufficiently short in duration such that the time-dependacde of the turbulent field may be neglected.

Description: The pressure distribution on a swept wing causes the streamlines at the edge of the boundary layer to be curved. This pressure gradient normal to the external streamline creates a velocity component normal to the external streamline within the boundary layer which is referred to as the crossflow velocity. Because the crossflow velocity profile perpendicular to the wing surface has an inflection point, the profile is unstable. The stationary instability mode takes the form of crossflow vortices. Under these conditions, the boundary layer on the wing is extremely unstable and transition to turbulent flow takes place much closer to the leading edge of the wing than it would on an unswept wing. Higher skin friction drag is associated with turbulent flow, and so better aircraft performance could be obtained if the crossflow could be eliminated One method of controlling crossflow that is being investigated is boundary-layer suction. An extensive airfoil suction experiment in the 8 feet Transonic Pressure Tunnel (TPT) at NASA Langley Research Center will begin late in 1994. Because of the size, complexity, and expense associated with this test, a number of 'risk-reduction' tests are currently being conducted. The 20 x 28 in. Shear Flow Control Tunnel at NASA Langley is being used for some of these tests. Prior to the summer of 1994, a flat plate with a swept leading edge was installed in the 20 x 28 in. tunnel, with a displacement body mounted on the tunnel ceiling that created a pressure distribution on the plate similar to the pressure distribution on a swept wing. The flow over the plate was investigated during the summer of 1994 using a laser Doppler velocimeter (LDV) system. The LDV measurements indicated the possible presence of multiple disturbance modes, a rarely-seen phenomena since, in most tests, one disturbance mode dominates. The possible existence of multiple disturbance modes in the flat plate boundary layer, however, means that the flow in the 20 x 28 in. tunnel is of interest itself, and will be investigated more thoroughly in the future. With a view to these investigations, the boundary layer traverse mechanism in the 20 x 28 in. tunnel was modified to improve its performance, and strain gauges were mounted on the traverse in order to monitor its deflection during a test. Other preliminary work conducted in the 20 x 28 in. tunnel included the use of an infrared camera system. Previous work with this system showed that transition indeed could be detected, but the signal produced by the crossflow vortices was too weak to be detected. It was hoped that spraying the flat plate with naphthalene would augment the heat transfer associated with the crossflow vortices so that they would show up in a IR image; however, experiments showed that this would not work. Another set of tests was conducted in the 20 x 28 in. tunnel to determine the tripping requirements for a set of airfoil-shaped struts that will be used in the 8 feet TPT experiment. Since the Reynolds number associated with these struts is small, a laminar boundary layer would separate early, causing large fluctuations in the flow field. A turbulent boundary layer would remain attached further back, but tripping from laminar to turbulent flow at low Reynolds number is very difficult. However, trip strip configurations were found that should effectively trip the boundary layer at the required conditions. Currently underway is an investigation of the data acquisition requirements for the 8 feet TPT experiment, with the purpose of the finding the minimum amount of data needed to characterize sufficiently the swept-wing boundary layer. This study is being conducted using a numerically generated data set.

Description: As part of the subsonic transport high-lift program, flight experiments are being conducted using NASA Langley's B737-100 to measure the flow characteristics of the multi-element high-lift system at full-scale high-Reynolds-number conditions. The instrumentation consists of hot-film anemometers to measure boundary-layer states, an infra-red camera to detect transition from laminar to turbulent flow, Preston tubes to measure wall shear stress, boundary-layer rakes to measure off-surface velocity profiles, and pressure orifices to measure surface pressure distributions. The initial phase of this research project was recently concluded with two flights on July 14. This phase consisted of a total of twenty flights over a period of about ten weeks. In the coming months the data obtained in this initial set of flight experiments will be analyzed and the results will be used to finalize the instrumentation layout for the next set of flight experiments scheduled for Winter and Spring of 1995. The main goal of these upcoming flights will be: (1) to measure more detailed surface pressure distributions across the wing for a range of flight conditions and flap settings; (2) to visualize the surface flows across the multi-element wing at high-lift conditions using fluorescent mini tufts; and (3) to measure in more detail the changes in boundary-layer state on the various flap elements as a result of changes in flight condition and flap deflection. These flight measured results are being correlated with experimental data measured in ground-based facilities as well as with computational data calculated with methods based on the Navier-Stokes equations or a reduced set of these equations. Also these results provide insight into the extent of laminar flow that exists on actual multi-element lifting surfaces at full-scale high-life conditions. Preliminary results indicate that depending on the deflection angle, the slat and flap elements have significant regions of laminar flow over a wide range of angles of attack. Boundary-layer transition mechanisms that were observed include attachment-line contamination on the slat and inflectional instability on the slat and fore flap. Also, the results agree fairly well with the predictions reported in a paper presented at last year's AIAA Fluid Dynamics Conference. The fact that extended regions of laminar flow are shown to exist on the various elements of the high-lift system raises the question what the effect is of loss of laminar flow as a result of insect contamiantion, rain or ice accumulation on high-life performance.

Description: Existing aerodynamic design methods have generally concentrated on the optimization of airfoil or wing shapes to produce a minimum drag while satisfying some basic constraints such as lift, pitching moment, or thickness. Since the minimization of drag almost always precludes the existence of separated flow, the evaluation and validation of these design methods for their robustness and accuracy when separated flow is present has not been aggressively pursued. However, two new applications for these design tools may be expected to include separated flow and the issues of aerodynamic design with this feature must be addressed. The first application of the aerodynamic design tools is the design of airfoils or wings to provide an optimal performance over a wide range of flight conditions (multipoint design). While the definition of 'optimal performance' in the multipoint setting is currently being hashed out, it is recognized that given a wide enough range of flight conditions, it will not be possible to ensure a minimum drag constraint at all conditions, and in fact some amount of separated flow (presumably small) may have to be allowed at the more demanding flight conditions. Thus a multipoint design method must be tolerant of the existence of separated flow and may include some controls upon its extent. The second application is in the design of wings with extended high speed buffet boundaries of their flight envelopes. Buffet occurs on a wing when regions of flow separation have grown to the extent that their time varying pressures induce possible destructive effects upon the wing structure or adversely effect either the aircraft controllability or the passenger comfort. A conservative approach to the expansion of the buffet flight boundary is to simply expand the flight envelope of nonseparated flow under the assumption that buffet will also thus be alleviated. However, having the ability to design a wing with separated flow and thus to control the location, extent, and severity of the separated flow regions may allow aircraft manufacturers to gain an advantage in the early design stages of an aircraft, when configuration changes are relatively inexpensive to make. Continuing the work begun last year, an airfoil design package has been modified to provide some control over the existence and extent of flow separation. This package consists of a 2-D Navier-Stokes flow solver which is coupled to the CDISC (constrained direct/iterative surface curvature) design method. The first modification is a prediction method for determining whether separation is likely based solely upon a given pressure distribution. If separation is predicted but is undesirable, the new routines will modify the pressure distribution to alleviate the problem. This new pressure distribution is then used in the design method to generate a new aerodynamic shape. Since separation may be acceptable in some cases, particularly if the separation does not extend to the trailing edge, another added logic estimates the extent of separation based upon a correlation with calculated separated flow cases. If the flow behind a shock induced separation is not predicted to reattach before the trailing edge, the logic weakens the shock strength and otherwise alters the pressure distribution in order to promote reattachment. This later addition is as yet unreliable due to secondary separation effects, but additional work is being pursued to improve the method.

Description: Historically, large-eddy simulations (LES) have been restricted to simple geometries where spectral or finite difference methods have dominated due to their efficient use of structured grids. Structured grids, however, not only difficulty representing complex domains and adapting to complicated flow features, but also are rather inefficient for simulating flows at high Reynolds numbers. The lack of efficiency stems from the need to resolve the viscous sublayer, which requires very fine resolution in all three directions near the wall. Structured grids make use of a stretching to reduce the normal grid spacing but must carry the fine resolution in the streamwise and spanwise directions throughout the domain. The unnecessarily fine grid for much of the domain leads to disturbingly high grid estimates. Chapman (1979), and later Moin & Jimenez (1993), pointed out that, in order to advance the technology to airfoils at flight Reynolds numbers, structured grids must be abandoned in lieu of what are known as nested or unstructured grids. The finite element method can efficiently solve the Navier-Stokes equations on unstructured grids. Although the CPU cost per time step per element is somewhat higher than structured grid methods, this effect is more than offset by the reduction in the number of elements. The use of unstructured grids, coupled with the advances in dynamic subgrid-scale modeling such as those made by Germano et al. (1991) and Ghosal et al. (1994), make LES of an airfoil tractable. We have chosen the NACA 4412 airfoil at maximum lift as the first simulation since this flow has not been successfully simulated with the Reynolds-averaged Navier-Stokes equations.

Description: This paper presents view-graphs and notes on sonic boom aging. Topic covered include sonic boom propagation, George's minimized F-function, final minimum shock boom, amplitude and age parameters, off-track aging, scaling flight test experiments, the potential for thin shocks, and results of a Boomfile flight test that showed significant waveform distortion.

Description: A model experiment was designed and built to simulate the propagation of sonic booms through atmospheric turbulence. The setup of the model experiment is described briefly. Measurements of the N waves after they propagated across the turbulent velocity field reveal the same waveform distortion and change in rise time as for sonic booms. The data from the model experiment is used to test sonic boom models. Some models yield predictions for the waveform distortion, while others give estimates of the rise time of the sonic booms. A new theoretical model for the propagation of plane N waves through a turbulent medium is described.

Description: The improved simulation of sonic boom propagation through the real atmosphere requires greater understanding of how the transient acoustic pulses popularly termed sonic booms are affected by humidity and turbulence. A realistic atmosphere is invariably somewhat turbulent, and may be characterized by an ambient fluid velocity v and sound speed c that vary from point to point. The absolute humidity will also vary from point to point, although possibly not as irregularly. What is ideally desired is a relatively simple scheme for predicting the probable spreads in key sonic boom signature parameters. Such parameters could be peak amplitudes, rise times, or gross quantities obtainable by signal processing that correlate well with annoyance or damage potential. The practical desire for the prediction scheme is that it require a relatively small amount of knowledge, possibly of a statistical nature, concerning the atmosphere along, the propagation path from the aircraft to the ground. The impact of such a scheme, if developed, implemented, and verified, would be that it would give the persons who make planning decisions a tool for assessing the magnitude of environmental problems that might result from any given overflight or sequence of overflights. The technical approach that has been followed by the author and some of his colleagues is to formulate a hierarchy of simple approximate models based on fundamental physical principles and then to test these models against existing data. For propagation of sonic booms and of other types of acoustic pulses in nonturbulent model atmospheres, there exists a basic overall theoretical model that has evolved as an outgrowth of geometrical acoustics. This theoretical model depicts the sound as propagating within ray tubes in a manner analogous to sound in a waveguide of slowly varying cross-section. Propagation along the ray tube is quasi-one-dimensional, and a wave equation for unidirectional wave propagation is used. A nonlinear term is added to this equation to account for nonlinear steepening, and the formulation has been carried through to allow for spatially varying sound speed, ambient density, and ambient wind velocities. The model intrinsically neglects diffraction, so it cannot take into account what has previously been mentioned in the literature as possibly important mechanisms for turbulence-related distortion. The model as originally developed could predict an idealized N-waveform which often agrees with data in terms of peak amplitude and overall positive phase duration. It is possible, moreover, to develop simple methods based on the physics of relaxation processes for incorporating molecular relaxation into the quasi-one-dimensional model of nonlinear propagation along ray tubes.

Description: A full-scale helicopter rotor test was conducted in the NASA Ames 80- by 120-Foot Wind Tunnel with a four-bladed S-76 rotor system. This wind tunnel test generated a unique and extensive data base covering a wide range of rotor shaft angles-of-attack and rotor thrust conditions from 0 to 100 knots. Three configurations were tested: (1) empty tunnel; (2) test stand body (fuselage) and support system; and (3) fuselage and support system with rotor installed. Empty tunnel wall pressure data are evaluated as a function of tunnel speed to understand the baseline characteristics. Aerodynamic interaction effects between the fuselage and the walls of the tunnel are investigated by comparing wall, ceiling, and floor pressures for various tunnel velocities and fuselage angles-of-attack. Aerodynamic interaction effects between the rotor and the walls of the tunnel are also investigated by comparing wall, ceiling, and floor pressures for various rotor shaft angles, rotor thrust conditions, and tunnel velocities. Empty tunnel wall pressure data show good repeatability and are not affected by tunnel speed. In addition, the tunnel wall pressure profiles are not affected by the presence of the fuselage apart from a pressure shift. Results do not indicate that the tunnel wall pressure profiles are affected by the presence of the rotor. Significant changes in the wall, ceiling, and floor pressure profiles occur with changing tunnel speeds for constant rotor thrust and shaft angle conditions. Significant changes were also observed when varying rotor thrust or rotor shaft angle-of-attack. Other results indicate that dynamic rotor loads and blade motion are influenced by the presence of the tunnel walls at very low tunnel velocity and, together with the wall pressure data, provide a good indication of flow breakdown.

Description: Wind tunnel evaluations of two-dimensional high-lift airfoils and of vehicles operating in ground effect near the tunnel floor require special test facilities and procedures. These are needed to avoid errors caused by proximity to the walls and interference from the wall boundary layers. Pneumatic test techniques and facilities were developed for GTRI aerodynamic research tunnels and calibrated to verify that these wall effects had been removed. The modified facilities were then employed to evaluate the aerodynamic characteristics of blown very-high-lift airfoils and of racing hydroplanes operating in ground effect at various levels above the floor. The pneumatic facilities, techniques and calibrations are discussed and typical aerodynamic data recorded both with and without the test-section blowing systems are presented.

Description: One common method of extrapolating sonic boom waveforms from aircraft to ground is to calculate the nonlinear distortion, and then add a rise time to each shock by a simple empirical rule. One common rule is the '3 over P' rule which calculates the rise time in milliseconds as three divided by the shock amplitude in psf. This rule was compared with the results of ZEPHYRUS, a comprehensive algorithm which calculates sonic boom propagation and extrapolation with the combined effects of nonlinearity, attenuation, dispersion, geometric spreading, and refraction in a stratified atmosphere. It is shown there that the simple empirical rule considerably overestimates the rise time estimate. In addition, the empirical rule does not account for variations in the rise time due to humidity variation or propagation history. It is also demonstrated that the rise time is only an approximate indicator of perceived loudness. Three waveforms with identical characteristics (shock placement, amplitude, and rise time), but with different shock shapes, are shown to give different calculated loudness. This paper is based in part on work performed at the Applied Research Laboratories, the University of Texas at Austin, and supported by NASA Langley.

Description: The purpose of our investigation is to determine the effect of unsteadiness (not associated with turbulence) on rise time. The unsteadiness considered here is due to (1) geometrical spreading, (2) stratification, which includes variation in density, temperature, and relative humidity, and (3) N shaped waveform. A very general Burgers equation, which includes all these effects, is the propagation model for our study. The equation is solved by a new computational algorithm in which all the calculations are done in the time domain. The present paper is a progress report in which some of the factors contributing to unsteadiness are studied, namely geometrical spreading and variation in relative humidity. The work of Pierce and Kang, which motivated our study, is first reviewed. We proceed with a discussion of the Burgers equation model and the algorithm for solving the equation. Some comparison tests to establish the validity of the algorithm are presented. The algorithm is then used to determine the distance required for a steady-state shock, on encountering an abrupt change in relative humidity, to reach a new steady state based on the new humidity. It is found that the transition distance for plane shocks of amplitude 70 Pa is about 4 km when the change in relative humidity is 10 percent. Shocks of amplitude 140 Pa require less distance. The effect of spherical and cylindrical spreading is also considered. We demonstrate that a spreading shock wave never reaches steady state and that its rise time will be less than the equivalent steady state shock. Finally we show that an N wave has a slightly shorter rise time than a step shock of the same amplitude.

Description: The following topics are discussed: (1) X-29 description; Vortex Flow Control (VFC) technology description; (3) X-29 VFC wind tunnel results (forebody only); (4) X-29 VFC wind tunnel results (full configuration yawing moment); (5) X-29 VFC wind tunnel results (full configuration C(sub n) with sideslip); (6) X-29VFC wind tunnel results (full configuration pitching moment); (7) VFC optimized nozzle details; (8) X-29 forebody nozzle configuration; (9) X-29 VFC system stored gas schematic; (10) X-29 VFC system stored gas installation; (11) VFC effectiveness at zero sideslip; (12) VFC effectiveness at 35 AOA with sideslip; (13) 'VFC Roll' at 40 AOA; (14) Effects of VFC on wing rock; (15) Integrated controls C(sub n) prediction; (16) Proposed F-15 with lateral control laws with active VFC; (17) Simulated F-15 roll performance with active VFC; (18) Simulated F-15 spin recovery with active VFC; (19) Test team restructuring; (20) testbed selection; (21) Simulation for risk reduction; (22) Benefits of high pressure system; and (23) Advanced weapon system integration.

Description: The goal of the F-15 Forebody Vortex Control (FVC) program is to develop a production FVC system for the F-15. The system may consist of either a mechanically actuated device such as the strakes developed for the HARV program, or a pneumatic device such as the port blowing system being tested on the X-29. Both types of systems are being evaluated under this program. Background information on the F-15 and a description and overview of forebody vortex controls (FVC) will be presented.

Description: As part of the NASA High-Angle-of-Attack Technology Program (HATP), flight tests are currently being conducted with a multi-axis thrust vectoring system applied to the NASA F-18 High Alpha Research Vehicle (HARV). A follow-on series of flight tests with the NASA F-18 HARV will be focusing on the application of actuated forebody strake controls. These controls are designed to provide increased levels of yaw control at high angles of attack where conventional aerodynamic controls become ineffective. The series of flight tests are collectively referred to as the Actuated Nose Strakes for Enhanced Rolling (ANSER) Flight Experiment. The development of actuated forebody strake controls for the F-18 HARV is discussed and a summary of the ground tests conducted in support of the flight experiment is provided. A summary of the preparations for the flight tests is also provided.

Description: This research will be exploring the prospect of employing bluntness, known to suppress the tendency toward asymmetry on slender forebodies, jointly with pneumatic manipulation as a system of forebody asymmetry control. The influences of jet location and direction, blowing rate, relative noise bluntness, angle of attack, and state of flow separation feeding the vortices (laminar vs. turbulent) will be evaluated.

Description: The results from research on forebody vortex control on both the F/A-18 and the F-16 aircraft will be shown. Several methods of forebody vortex control, including mechanical and pneumatic schemes, will be discussed. The wind tunnel data includes both static and rotary balance data for forebody vortex control. Time lags between activation or deactivation of the pneumatic control and when the aircraft experiences the resultant forces are also discussed. The static (non-rotating) forces and pressures are then compared to similar configurations tested in the NASA Langley and DTRC Wind Tunnel, the NASA Ames 80'x120' Wind Tunnel, and in flight on the High Angle of Attack Research Vehicle (HARV).

Description: It has been shown experimentally that forebody flow control devices provide a significant increase in yaw control for fighter aircraft at high angle-of-attack. This study presents comparisons of the various experimental and computational results for tangential slot blowing on the F/A-18 configuration. Experimental results are from the full-scale and 6 percent-scale model test and computational solutions are from both isolated forebody and and full aircraft configurations. The emphasis is on identifying trends in the variation of yawing moment with blowing-slot exit conditions. None of the traditional parameters (mass flow ratio, blowing momentum coefficient, velocity ratio) succeeded in collapsing all of the results into a common curve. Several factors may effect the agreement between the 6 percent- and full-scale results including Reynolds number effects, sensitivity of boundary layer transition from laminar to turbulent flow, and poor geometric fidelity, particularly of the blowing slot. The disagreement between the full-scale and computed yawing moments may be due to a mismatch in the slot exit conditions for the same mass flow ratio or aircraft configuration modeling. The general behavior of slot blowing on the 6 percent-scale and computational models is correct, but neither matches the full-scale results.

Description: An overview of the computational effort to analyze forebody tangential slot blowing is presented. Tangential slot blowing generates side force and yawing moment which may be used to control an aircraft flying at high-angle-of-attack. Two different geometries are used in the analysis: (1) The High Alpha Research Vehicle; and (2) a generic chined forebody. Computations using the isolated F/A-18 forebody are obtained at full-scale wind tunnel test conditions for direct comparison with available experimental data. The effects of over- and under-blowing on force and moment production are analyzed. Time-accurate solutions using the isolated forebody are obtained to study the force onset timelag of tangential slot blowing. Computations using the generic chined forebody are obtained at experimental wind tunnel conditions, and the results compared with available experimental data. This computational analysis compliments the experimental results and provides a detailed understanding of the effects of tangential slot blowing on the flow field about simple and complex geometries.

Description: This presentation will highlight results from the latest Navy evaluation of the HARV (March 1994) and focus primarily on the impressions from a piloting standpoint of the tactical utility of thrust vectoring. Issue to be addressed will be mission suitability of high AOA flight, visual and motion feedback cues associated with operating at high AOA, and the adaptability of a pilot to effectively use the increased control power provided by the thrust vectoring system.

Description: The proper management of energy becomes a complex task in fighter aircraft which have high angle of attack (AOA) capability. Maneuvers at high AOA are accompanied by high bleed rates (velocity decrease), a characteristic that is usually undesirable in a typical combat arena. Eidetics has developed under NASA SBIR Phase 1 and NAVAIR SBIR Phase 2 contracts a system which allows a pilot to more easily and effectively manage the trade-off of energy (airspeed or altitude) for turn rate while not imposing hard limits on the high AOA nose pointing capability that can be so important in certain air combat maneuver situations. This has been accomplished by incorporating a two-stage angle of attack limiter into the flight control laws. The first stage sets a limit on AOA to achieve a limit on the maximum bleed rate (selectable) by limiting AOA to values which are dependent on the aircraft attitude and dynamic pressure (or flight path, velocity, and altitude). The second stage sets an AOA limit near the AOA for C(sub l max). One of the principal benefits of such a system is that it enables a low-experience pilot to become much more proficient at managing his energy. The Phase 2 simulation work is complete, and an exploratory flight test on the F-18 HARV is planned for the Fall of 1994 to demonstrate/validate the concept.

Description: For the Multi-Axis Thrust Vectoring (MATV) program, a new control law was developed using multi-axis thrust vectoring to augment the aircraft's aerodynamic control power to provide maneuverability above the normal F-16 angle of attack limit. The control law architecture was developed using Lockheed Fort Worth's offline and piloted simulation capabilities. The final flight control laws were used in flight test to demonstrate tactical benefits gained by using thrust vectoring in air-to-air combat. Differences between the simulator aerodynamics data base and the actual aircraft aerodynamics led to significantly different lateral-directional flying qualities during the flight test program than those identified during piloted simulation. A 'dial-a-gain' flight test control law update was performed in the middle of the flight test program. This approach allowed for inflight optimization of the aircraft's flying qualities. While this approach is not preferred over updating the simulator aerodynamic data base and then updating the control laws, the final selected gain set did provide adequate lateral-directional flying qualities over the MATV flight envelope. The resulting handling qualities and the departure resistance of the aircraft allowed the 422nd_squadron pilots to focus entirely on evaluating the aircraft's tactical utility.

Description: This study was designed to investigate flying qualities requirements of alternate pitch command systems for fighter aircraft at high angle of attack. Flying qualities design guidelines have already been developed for angle of attack command systems at 30, 45, and 60 degrees angle of attack, so this research fills a similar need for rate command systems. Flying qualities tasks that require post-stall maneuvering were tested during piloted simulations in the McDonnell Douglas Aerospace Manned Air Combat Simulation facility. A generic fighter aircraft model was used to test angle of attack rate and pitch rate command systems for longitudinal gross acquisition and tracking tasks at high angle of attack. A wide range of longitudinal dynamic variations were tested at 30, 45, and 60 degrees angle of attack. Pilot comments, Cooper-Harper ratings, and pilot induced oscillation ratings were taken from five pilots from NASA, USN, CAF, and McDonnell Douglas Aerospace. This data was used to form longitudinal design guidelines for rate command systems at high angle of attack. These criteria provide control law design guidance for fighter aircraft at high angle of attack, low speed flight conditions. Additional time history analyses were conducted using the longitudinal gross acquisition data to look at potential agility measures of merit and correlate agility usage to flying qualities boundaries. This paper presents an overview of this research.

Description: This viewgraph presentation describes the general methodology used to apply Honywell's Multi-Application Control (MACH) and the specific application to the F-18 High Angle-of-Attack Research Vehicle (HARV) including piloted simulation handling qualities evaluation. The general steps include insertion of modeling data for geometry and mass properties, aerodynamics, propulsion data and assumptions, requirements and specifications, e.g. definition of control variables, handling qualities, stability margins and statements for bandwidth, control power, priorities, position and rate limits. The specific steps include choice of independent variables for least squares fits to aerodynamic and propulsion data, modifications to the management of the controls with regard to integrator windup and actuation limiting and priorities, e.g. pitch priority over roll, and command limiting to prevent departures and/or undesirable inertial coupling or inability to recover to a stable trim condition. The HARV control problem is characterized by significant nonlinearities and multivariable interactions in the low speed, high angle-of-attack, high angular rate flight regime. Systematic approaches to the control of vehicle motions modeled with coupled nonlinear equations of motion have been developed. This paper will discuss the dynamic inversion approach which explicity accounts for nonlinearities in the control design. Multiple control effectors (including aerodynamic control surfaces and thrust vectoring control) and sensors are used to control the motions of the vehicles in several degrees-of-freedom. Several maneuvers will be used to illustrate performance of MACH in the high angle-of-attack flight regime. Analytical methods for assessing the robust performance of the multivariable control system in the presence of math modeling uncertainty, disturbances, and commands have reached a high level of maturity. The structured singular value (mu) frequency response methodology is presented as a method for analyzing robust performance and the mu-synthesis method will be presented as a method for synthesizing a robust control system. The paper concludes with the author's expectations regarding future applications of robust nonlinear multivariable controls.

Description: A review is presented in viewgraph format of an ongoing NASA/U.S. Navy study to determine control power requirements at high angles of attack for the next generation high-performance aircraft. This paper focuses on recent flight test activities using the NASA High Alpha Research Vehicle (HARV), which are intended to validate results of previous ground-based simulation studies. The purpose of this study is discussed, and the overall program structure, approach, and objectives are described. Results from two areas of investigation are presented: (1) nose-down control power requirements and (2) lateral-directional control power requirements. Selected results which illustrate issues and challenges that are being addressed in the study are discussed including test methodology, comparisons between simulation and flight, and general lessons learned.

Description: The results of the envelope expansion phase of the F-16 Multi-Axis Thrust Vectoring (MATV) program are presented in viewgraph format. The objectives and test approach are presented followed by results of testing with the initial control law configuration. The revised flight control laws are discussed followed by test results with the revised control laws. Additional testing added to the program, nose chines, parameter identification maneuvers, and the extended range angle of attack cones are briefly discussed.

Description: The design goals for the X-31 flight control system were: (1) level 1 handling qualities during post-stall maneuvering (30 to 70 degrees angle-of-attack); (2) thrust vectoring to enhance performance across the flight envelope; and (3) adequate pitch-down authority at high angle-of-attack. Additional performance goals are discussed. A description of the flight control system is presented, highlighting flight control system features in the pitch and roll axes and X-31 thrust vectoring characteristics. The high angle-of-attack envelope clearance approach will be described, including a brief explanation of analysis techniques and tools. Also, problems encountered during envelope expansion will be discussed. This presentation emphasizes control system solutions to problems encountered in envelope expansion. An essentially 'care free' envelope was cleared for the close-in-combat demonstrator phase. High angle-of-attack flying qualities maneuvers are currently being flown and evaluated. These results are compared with pilot opinions expressed during the close-in-combat program and with results obtained from the F-18 HARV for identical maneuvers. The status and preliminary results of these tests are discussed.

Description: Technical and nontechnical lessons learned from the X-31 aircraft program are described in this viewgraph presentation. The tactical utility of high angle of attack flight and thrust vector control is discussed.

Description: The primary objective of the quasi-tailless flight demonstration is to demonstrate the feasibility of using thrust vectoring for directional control of an unstable aircraft. By using this low-cost, low-risk approach it is possible to get information about required thrust vector control power and deflection rates from an inflight experiment as well as insight in low-power thrust vectoring issues. The quasi-tailless flight demonstration series with the X-31 began in March 1994. The demonstration flight condition was Mach 1.2 at 37,500 feet. A series of basic flying quality maneuvers, doublets, bank to bank rolls, and wind-up-turns have been performed with a simulated 100% vertical tail reduction. Flight test and supporting simulation demonstrated that the quasi-tailless approach is effective in representing the reduced stability of tailless configurations. The flights also demonstrated that thrust vectoring could be effectively used to stabilize a directionally unstable configuration and provide control power for maneuver coordination.

Description: An innovative inlet total-pressure distortion measurement rake has been designed and developed for the F/A-18 A/B/C/D aircraft inlet. The design was conceived by NASA and General Electric Aircraft Engines (Evendale, Ohio). This rake has been flight qualified and flown in the F-18 High Alpha Research Vehicle (HARV) at NASA Dryden Flight Research Center. The rake's eight-legged, one-piece wagon wheel design was developed at a reduced cost and offers reduced installation time compared with traditional designs. The rake features 40 dual measurement ports for both low- and high-frequency pressure measurements with the high-frequency transducer mounted at the port. The high-frequency transducer offers direct absolute pressure measurements from low frequency to the highest frequency of interest, thereby allowing the rake to be used during highly dynamic aircraft maneuvers. Outstanding structural characteristics are inherent to the design through its construction and use of lightweight materials.

Description: The F-18 High Alpha Research Vehicle has proven to be a useful research tool with many unique capabilities. Many of these capabilities are to assist in characterizing flight at high angles of attack, while some provide significant research in their own right. Of these, the thrust vectoring system, the unique ability to rapidly reprogram flight controls, the reprogrammable mission computer, and a reprogrammable onboard excitation system have allowed an increased utility and versatility of the research being conducted. Because of this multifaceted approach to research in the high angle of attack regime, the capabilities of the F-18 High Alpha Research Vehicle were designed to cover as many high alpha technology bases as the program would allow. These areas include aerodynamics, controls, handling qualities, and propulsion.

Description: This presentation will concentrate on a series of low-speed wind tunnel tests conducted on a 2.5 percent subscale F-18 model and a 2 percent subscale X-31 model. The model's control surfaces were unaugmented; and for the most part, were deflected at a constant angle throughout the tests. The tests consisted mostly of free-to-roll experiments conducted with the use of an air-bearing, surface pressure measurements, off-surface flow visualization, and force-balance tests. Where possible the results of the subscale tests have been compared to flight test data, or to other wind tunnel data taken at higher Reynolds numbers.

Description: This paper shall discuss the evaluation of the original Dryden X-31 aerodynamic math model, processes involved in the justification and creation of the modified data base, and comparison time history results of the model response with flight test.

Description: The U.S./German experimental aircraft X-31A was designed and constructed to demonstrate enhanced fighter maneuverability. Post-stall maneuvering is enabled by applying new technologies such as high angle of attack aerodynamics and flight control system integrated thrust vectoring. Two demonstrator aircraft have been built by the main contractors, Rockwell International and Deutsche Aerospace (formerly MBB). Flight testing started in October 1990 and before the end of 1992 both aircraft had accomplished a significant number of flights covering the entire AoA regime from about -5 to 70 deg. Throughout the envelope expansion, DLR Institute of Flight Mechanics conducted parameter identification (PID) to determine the aerodynamic parameters of the aircraft from flight test data and to compare the results to the predictions from the aerodynamic dataset (ADS). The application of system identification to high AoA / post-stall flight data raises some major problems, which are discussed in this paper. Results from both longitudinal and lateral-directional motion will be presented.

Description: The current research is aimed at developing and extending numerical methods to accurately predict the high Reynolds number flow about the NASA F-18 HARV at large angles of attack. The resulting codes are validated by comparison of the numerical results with in-flight aerodynamic measurements and flow visualization obtained on the HARV. Further, computations have been used to provide an analysis and numerical optimization of a pneumatic slot blowing concept, and a mechanical strake concept, for use as potential forebody flow control devices in improving high-alpha maneuverability.

Description: At high angles of attack, vortical flows play a crucial role in the maintenance of lift for fighter aircraft. However, under certain conditions, the vortical flow can have an adverse effect on the aircraft. On the F/A-18 the LEX vortex can impinge on the tail; at high angles of attack the unsteady flow from vortex bursting can cause structural fatigue on the vertical tails. At high angles of attack, the flow field can be quite complex, and a computational analysis challenging. A full configuration analysis with viscous structured grids can be computationally expensive and the task of generating the requisite grids can be quite difficult. To mitigate these difficulties and provide a medium-fidelity analysis tool, a hybrid structured/unstructured approach was adopted for this work. In this analysis, the formation and roll-up of the LEX vortex is computed with a structured Navier-Stokes solver, and the resulting vortex propagated downstream with an unstructured Euler solver.