ALBERT

Description: There is a frequent need to measure the frequency stability and phase noise levels of very high performance signal sources that are required for certain spacecraft missions. These measurements need to be done at different locations as the spacecraft subsystems progress through the various stages of development, assembly, test, and integration. Allan Deviation and Phase Noise of high performance sources are generally measured by comparing the unit under test to a reference standard. Five basic requirements are associated with making these kind of measurements: (1) the reference standard performance needs to be equal or better than the unit under test; (2) the measurement system needs to accommodate odd, nonstandard measurement frequencies that can range from 4 MHz to 35 GHz; (3) warm-up frequency drift and aging can corrupt a measurement and must be dealt with; (4) test equipment generated noise must be understood and prevented from limiting the measurements; (5) test equipment noise performance must be verifiable in the field as needed. A portable measurement system that was built by JPL and used in the field is described. The methods of addressing the above requirements are outlined and some measurement noise floor values are given. This test set was recently used to measure state of the art crystal oscillator frequency standards on the TOPEX and MARS OBSERVER spacecraft during several stages of acceptance tests.

Description: A study was conducted to assess the effects of transmitting a precision clock synchronization signal over a commercial multiplexed fiber optic communication system. This study is an evaluation of the distortion and jitter introduced into the signal by this type of transmission system. An analysis comparing signal quality at the multiplexing and demultiplexing ends of the fiber optic communication system shows that the amplitude and phase distortion added to the clock synchronization signal by the transmission system is minimal.

Description: Historically, Maximum Time Interval Error (MTIE) and Maximum Relative Time Interval Error (MRTIE) have been the main measurement techniques used to characterize timing performance in telecommunications networks. Recently, a new measurement technique, Time Variance (TVAR) has gained acceptance in the North American (ANSI) standards body. TVAR was developed in concurrence with NIST to address certain inadequacies in the MTIE approach. The advantages and disadvantages of each of these approaches are described. Real measurement examples are presented to illustrate the critical issues in actual telecommunication applications. Finally, a new MTIE measurement is proposed (ZTIE) that complements TVAR. Together, TVAR and ZTIE provide a very good characterization of network timing.

Description: For proper operation of today's large digital private networks, high quality synchronization must be achieved. A general telecommunication performance objective is to maintain long-term frequency accuracy of ten parts per trillion at all synchronous digital equipment in the network. Many times, however, this is not achieved in private networks. Low quality clocks, errored transmission facilities, and incorrectly designed synchronization plans are often cause for poor performance. It is shown that properly-designed private networks can operate with long term frequency averages between ten parts per trillion to ten parts per million. These performance levels can adversely impact customer applications. The most demanding applications are digital and voice band data, encrypted voice, facsimile, and video. In a typical private network operating at 0.01 parts per million, the user would experience reduced data throughput, dropped encrypted calls, unreadable facsimile pages, or interrupted video transmission dozens of times per day. The major contribution to poor private network synchronization performance is the interaction of Customer Premises Equipment (CPE) clocks and the network facilities used to distribute timing. The performance of typical CPE clocks and facilities, and their impact on customer applications, are discussed. CPE clock performance issues, along with private network architectural constraints, make synchronization planning extremely difficult. Planning is usually costly and requires specialized expertise.

Description: A new method that uses round-trip paths to accurately measure transmission delay for time synchronization is proposed. The performance of the method in Synchronous Digital Hierarchy networks is discussed. The feature of this method is that it separately measures the initial round trip path delay and the variations in round-trip path delay. The delay generated in SDH equipment is determined by measuring the initial round-trip path delay. In an experiment with actual SDH equipment, the error of initial delay measurement was suppressed to 30ns.

Description: A coded time/date information dissemination system (CTD), based on telephone lines and commercial modems, is now in its experimental phase in Italy at IEN. This service, born from a cooperation with other metrological laboratories (TUG, Austria, SNT, Sweden, VSL, The Netherlands), represents an attempt towards an European standardization. Some results of an experimental analysis in which a few modems were tested, both in laboratory conditions and connected to the telephone network, in order to evaluate the timing capability of the system are given. When the system is used in a one-way mode, in many practical cases the modems delay turns out to be the main factor which limits the accuracy, even more than the telephone line delays. If the two-way mode is used, the modems asymmetry, i.e., the delay difference between transmission and reception, is almost always the most important source of uncertainty, provided the link is not including a space segment. Comparing the widely used V.22 modems to the old V.21 ones, the latters turn out to be better both in delay time (30-100 ms V.22, and 7-15 ms V.21) and asymmetry (10-50 micro-s V.22, and 10 ms V.22). Time transfer accuracies of 10 micron-s (same turn) to 100 micro-s (long distance calls) were obtained in two-way mode with commercial V.21 modems.

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: The benefits of using a hypersonic waverider for spacecraft trajectory modification are presented. A waverider is a hypersonic vehicle specifically designed so that the undersurface bow shock is attached to the leading edge, which provides for the highest known lift-to-drag ratios achievable at high Mach number flight. Several viable space missions are suggested which could use such configurations for low-drag aero-assisted maneuvers in planetary atmospheres. It is shown that large changes in the spacecraft velocity vector can be accomplished with acceptably small losses in energy due to drag using a waverider aeroshell. The primary advantage of an aero-assist maneuver is suggested by comparison to a traditional gravity-assist trajectory. Some scaling laws are presented for comparing waveriders designed for different planetary atmospheres, and it is shown that the compositional differences between the terrestrial planets has a minimal impact on waverider design.

Description: Four different arraying schemes applicable to deep space communications are discussed and analyzed. These include symbol stream combining (SSC), baseband combining (BC), carrier arraying (CA) and full spectrum combining (FSC). Complexity versus performance is traded off throughout the paper and benefits to the reception of existing spacecraft signals are discussed.

Description: The scattering properties of perfectly conducting and resistive strips are predicted for strips which are located on a dielectric slab backed by a perfectly conducting ground plane. The spectral domain Green's function is used to relate the currents and fields on the strip, and the resulting integral equation is solved using the method of moments. Both TE and TM strips are examined using piecewise linear and pulse subdomain basis functions, respectively, to model the current on the strip. Calculated results are compared with results measured at the NASA Langley Research Center.

Description: POMESH is a computer program capable of predicting the performance of reflector antennas. Both far field pattern and gain calculations are performed using the Physical Optics (PO) approximation of the equivalent surface currents. POMESH is primarily intended for relatively small reflectors. It is useful in situations where the surface is described by irregular data that must be interpolated and for cases where the surface derivatives are not known. This method is flexible and robust and also supports near field calculations. Because of the near field computation ability, this computational engine is quite useful for subreflector computations. The program is constructed in a highly modular form so that it may be readily adapted to perform tasks other than the one that is explicitly described here. Since the computationally intensive portions of the algorithm are simple loops, the program can be easily adapted to take advantage of vector processor and parallel architectures. In POMESH the reflector is represented as a piecewise planar surface comprised of triangular regions known as facets. A uniform physical optics (PO) current is assumed to exist on each triangular facet. Then, the PO integral on a facet is approximated by the product of the PO current value at the center and the area of the triangle. In this way, the PO integral over the reflector surface is reduced to a summation of the contribution from each triangular facet. The source horn, or feed, that illuminates the subreflector is approximated by a linear combination of plane patterns. POMESH contains three polarization pattern definitions for the feed; a linear x-polarized element, linear y-polarized element, and a circular polarized element. If a more general feed pattern is required, it is a simple matter to replace the subroutine that implements the pattern definitions. POMESH obtains information necessary to specify the coordinate systems, location of other data files, and parameters of the desired calculation from a user provided data file. A numerical description of the principle plane patterns of the source horn must also be provided. The program is supplied with an analytically defined parabolic reflector surface. However, it is a simple matter to replace it with a user defined reflector surface. Output is given in the form of a data stream to the terminal; a summary of the parameters used in the computation and some sample results in a file; and a data file of the results of the pattern calculations suitable for plotting. POMESH is written in FORTRAN 77 for execution on CRAY series computers running UNICOS. With minor modifications, it has also been successfully implemented on a Sun4 series computer running SunOS, a DEC VAX series computer running VMS, and an IBM PC series computer running OS/2. It requires 2.5Mb of RAM under SunOS 4.1.1, 2.5Mb of RAM under VMS 5-4.3, and 2.5Mb of RAM under OS/2. The OS/2 version requires the Lahey F77L compiler. The standard distribution medium for this program is one 5.25 inch 360K MS-DOS format diskette. It is also available on a .25 inch streaming magnetic tape cartridge in UNIX tar format and a 9-track 1600 BPI magnetic tape in DEC VAX FILES-11 format. POMESH was developed in 1989 and is a copyrighted work with all copyright vested in NASA. CRAY and UNICOS are registered trademarks of Cray Research, Inc. SunOS and Sun4 are trademarks of Sun Microsystems, Inc. DEC, DEC FILES-11, VAX and VMS are trademarks of Digital Equipment Corporation. IBM PC and OS/2 are registered trademarks of International Business Machines, Inc. UNIX is a registered trademark of Bell Laboratories.

Description: The Interactive Digital Signal Processor, IDSP, consists of a set of time series analysis "operators" based on the various algorithms commonly used for digital signal analysis work. The processing of a digital time series to extract information is usually achieved by the application of a number of fairly standard operations. However, it is often desirable to "experiment" with various operations and combinations of operations to explore their effect on the results. IDSP is designed to provide an interactive and easy-to-use system for this type of digital time series analysis. The IDSP operators can be applied in any sensible order (even recursively), and can be applied to single time series or to simultaneous time series. IDSP is being used extensively to process data obtained from scientific instruments onboard spacecraft. It is also an excellent teaching tool for demonstrating the application of time series operators to artificially-generated signals. IDSP currently includes over 43 standard operators. Processing operators provide for Fourier transformation operations, design and application of digital filters, and Eigenvalue analysis. Additional support operators provide for data editing, display of information, graphical output, and batch operation. User-developed operators can be easily interfaced with the system to provide for expansion and experimentation. Each operator application generates one or more output files from an input file. The processing of a file can involve many operators in a complex application. IDSP maintains historical information as an integral part of each file so that the user can display the operator history of the file at any time during an interactive analysis. IDSP is written in VAX FORTRAN 77 for interactive or batch execution and has been implemented on a DEC VAX-11/780 operating under VMS. The IDSP system generates graphics output for a variety of graphics systems. The program requires the use of Versaplot and Template plotting routines and IMSL Math/Library routines. These software packages are not included in IDSP. The virtual memory requirement for the program is approximately 2.36 MB. The IDSP system was developed in 1982 and was last updated in 1986. Versaplot is a registered trademark of Versatec Inc. Template is a registered trademark of Template Graphics Software Inc. IMSL Math/Library is a registered trademark of IMSL Inc.

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 collection of programs comprises The UNIX Based Data Management System for the Pilot Field Experiment (PiFEx) which is an attempt to mimic the Mobile Satellite (MSAT) scenario. The major purposes of PiFEx are to define the mobile communications channels and test the workability of new concepts used to design various components of the receiver system. The results of the PiFex experiment are large amounts of raw data which must be accessed according to a researcher's needs. This package provides a system to manage the PiFEx data in an interactive way. The system not only provides the file handling necessary to retrieve the desired data, but also several FORTRAN programs to generate some standard results pertaining to propagation data. This package assumes that the data file initially generated from the experiment has been already converted from binary to ASCII format. The Data Management system described here consists of programs divided into two categories: those programs that handle the PiFEx generated files and those that are used for number-crunching of these files. Five FORTRAN programs and one UNIX shell script file are used for file manipulation purposes. These activities include: calibration of the acquired data; and parsing of the large data file into datasets concerned with different aspects of the experiment such as the specific calibrated propagation data, dynamic and static loop error data, statistical data, and temperature and spatial data on the hardware used in the experiment. The five remaining FORTRAN programs are used to generate usable information about the data. Signal level probability, probability density of the signal fitting the Rician density function, frequency of the data's fade duration, and the Fourier transform of the data can all be generated from these data manipulation programs. In addition, a program is provided which generates a downloadable file from the signal levels and signal phases files for use with the plotting routine AKPLOT (NPO-16931). All programs in this package are written in either FORTRAN-77 or UNIX shell-scripts. The package does not include test data. The programs were developed in 1987 for use with a UNIX operating system on a DEC MicroVAX computer.

Description: The Optical Communication Link Analysis Program, OPTI, analyzes optical and near-infrared communication links that use pulse position modulation (PPM) and direct detection. The program prompts for inputs of system component parameters, modulation format and other operational parameters, background noise sources, and the desired link bit error rate. From these inputs, link margin is determined and a link design control table (DCT) is generated. The program also allows the user to save sets of input parameters defining a given link and read them back into the program at a later time. Further, the program has the capability of altering automatically any of the input parameters to achieve a desired link margin. The program provides a table of extended background sources, e.g. planets, the moon, and the sun. To compute background noise, only the distance from the receiver to the noise source(s) must be entered. This determines whether or not the whole object is in the field-of-view. The program assumes that each object is a blackbody (at 5900K) with an overall visible magnitude scaled to match the tabulated data. Also provided is a table of 19 bright stars. If the noise source is one or several of these, then only the name(s) of the star(s) are required. Noise sources that are not among those contained in the program can be entered as "additional noise sources". In this case required information includes whether or not the source is a point or extended source, its radiance (for extended sources) or irradiance, and receiver to source distance in A.U. (for point sources). The OPTI program is written in FORTRAN-77 and was designed to be used on the IBM PC and PC/AT personal computers. (Note: The 8087/80287 math coprocessor option is highly recommended for use with this program.) The program will also compile under UNIX 4.3 BSD FORTRAN-77 with minor changes. OPTI was developed in 1987.

Description: The "Data Management System for Mobile Satellite Propogation" package is a collection of FORTRAN programs and UNIX shell scripts designed to handle the huge amounts of data resulting from Mobile Satellite propogation experiments. These experiments are designed to assist in defining channels for mobile satellite systems. By understanding multipath fading characteristics of the channel, doppler effects, and blockage due to manmade objects as well as natural surroundings, characterization of the channel can be realized. Propogation experiments, then, are performed using a prototype of the system simulating the ultimate product environment. After the data from these experiments is generated, the researcher must access this data with a minimum of effort and to derive some standard results. The programs included in this package manipulate the data files generated by the NASA/JPL Mobile Satellite propogation experiment on an interactive basis. In the experiment, a transmitter operating at 869 MHz was carried to an altitude of 32Km by a stratospheric balloon. A vehicle within the line-of-sight of the transmitter was then driven around, splitting the incoming signal into I and Q channels, and sampling the resulting signal strength at 1000 samples per second. The data was collected at various antenna elavation angles and different times of day generating the ancillary data for the experiment. This package contains a program to convert the binary format of the data generated into standard ASCII format suitable for use with a wide variety of machine architectures. Also included is a UNIX shell-script designed to parse this ASCII file into those records of data that match the researcher's desired values for the ancillary data parameters. In addition, four FORTRAN programs are included to obtain standard quantities from the data. Quantities such as probability of signal level greater than or equal to a specified signal level, probability density of the signal levels, frequency of fade duration, and Fourier Transforms of the sampled data can be generated from the propogation experiment data. All programs in this package are written in either FORTRAN 77 or UNIX shell-scripts. The package does not include test data. The programs were developed in 1987 for use with a UNIX operating system on a DEC MicroVAX computer.

Description: The Distributed Architecture for Phased Array Antennas program, DISTAR, is a simulation tool used to study the implementation of distributed phased array hardware. DISTAR allows the placement of possibly faulty transmit/receive modules (T/R) at locations throughout the array. Variations in amplifiers and phase shifters may degrade antenna performance, depending on environmental conditions and array architecture. DISTAR enables the antenna designer to examine array characteristics and how they effect both type and extent of antenna failures. General specifications for amplifier and phase shifter tolerances may also be determined for various architectures. The array geometry is assumed to be rectangular with optional staggering in the X or Y direction. The array is divided into various subgroups both geometrically and electronically. Each T/R drives a subgroup of elemental radiators. Both hard and soft failures of the T/R amplifiers are modeled. Hard failures are catastrophic as no power is transmitted to those elements controlled by that particular amplifier. Soft, or non-catastrophic, failures are simulated by a modified Gaussian distribution. Flat, warped, and parabolic arrays can be modeled. The user must supply array geometry, amplitude and phase for each row and column, and location of all T/Rs. Specific or randomly chosen amplifiers may be deemed faulty. DISTAR outputs the normalized antenna gain patterns in the form of tables, and 2-D or 3-D graphs. DISTAR is written in FORTRAN 77 for interactive execution and has been implemented on a DEC VAX 11/780 computer operating under VMS 4.4 with a central memory requirement of approximately 35K of 8 bit bytes (without graphics). DISTAR requires the commercial DISSPLA plot library for full graphics implementation. This program was developed in 1986.

Description: RHCP, the Right-Handed, Circularly Polarized Microstrip Antenna program, aids in the design of a rectangular antenna element, given the desired frequency of operation and substrate characteristics. RHCP begins the design calculations based on a square element with linear polarization. The effective dielectric constant and changes in electrical length due to fringing at the radiating element edges are taken into account. A coaxial feed is inset with 50 ohms input impedance. By placing the feed such that two orthonormal modes are produced in the antenna cavity, right- or left-handed circular polarization is obtained. Input to RHCP consists of desired frequency, dielectric constant, and substrate thickness. Output consists of the final rectangular geometry, the proposed feed inset placement, and actual input impedance. RHCP has been used successfully for frequencies between 2 and 15 GHz for thin substrates. This program was used to fabricate antenna elements for the S-band quad antennas on board the Space Shuttle, and is a part of the design project for the S-band phased array antenna radiating aperture. RHCP is written in FORTRAN 77 for interactive execution and has been implemented on a DEC VAX series computer operating under VMS. This program was developed in 1985.

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: MOM3D (LAR-15074) is a FORTRAN method-of-moments electromagnetic analysis algorithm for open or closed 3-D perfectly conducting or resistive surfaces. Radar cross section with plane wave illumination is the prime analysis emphasis; however, provision is also included for local port excitation for computing antenna gain patterns and input impedances. The Electric Field Integral Equation form of Maxwell's equations is solved using local triangle couple basis and testing functions with a resultant system impedance matrix. The analysis emphasis is not only for routine RCS pattern predictions, but also for phenomenological diagnostics: bistatic imaging, currents, and near scattered/total electric fields. The images, currents, and near fields are output in form suitable for animation. MOM3D computes the full backscatter and bistatic radar cross section polarization scattering matrix (amplitude and phase), body currents and near scattered and total fields for plane wave illumination. MOM3D also incorporates a new bistatic k space imaging algorithm for computing down range and down/cross range diagnostic images using only one matrix inversion. MOM3D has been made memory and cpu time efficient by using symmetric matrices, symmetric geometry, and partitioned fixed and variable geometries suitable for design iteration studies. MOM3D may be run interactively or in batch mode on 486 IBM PCs and compatibles, UNIX workstations or larger computers. A 486 PC with 16 megabytes of memory has the potential to solve a 30 square wavelength (containing 3000 unknowns) symmetric configuration. Geometries are described using a triangular mesh input in the form of a list of spatial vertex points and a triangle join connection list. The EM-ANIMATE (LAR-15075) program is a specialized visualization program that displays and animates the near-field and surface-current solutions obtained from an electromagnetics program, in particular, that from MOM3D. The EM-ANIMATE program is windows based and contains a user-friendly, graphical interface for setting viewing options, case selection, file manipulation, etc. EM-ANIMATE displays the field and surface-current magnitude as smooth shaded color fields (color contours) ranging from a minimum contour value to a maximum contour value for the fields and surface currents. The program can display either the total electric field or the scattered electric field in either time-harmonic animation mode or in the root mean square (RMS) average mode. The default setting is initially set to the minimum and maximum values within the field and surface current data and can be optionally set by the user. The field and surface-current value are animated by calculating and viewing the solution at user selectable radian time increments between 0 and 2pi. The surface currents can also be displayed in either time-harmonic animation mode or in RMS average mode. In RMS mode, the color contours do not vary with time, but show the constant time averaged field and surface-current magnitude solution. The electric field and surface-current directions can be displayed as scaled vector arrows which have a length proportional to the magnitude at each field grid point or surface node point. These vector properties can be viewed separately or concurrently with the field or surface-current magnitudes. Animation speed is improved by turning off the display of the vector arrows. In RMS modes, the direction vectors are still displayed as varying with time since the time averaged direction vectors would be zero length vectors. Other surface properties can optionally be viewed. These include the surface grid, the resistance value assigned to each element of the grid, and the power dissipation of each element which has an assigned resistance value. The EM-ANIMATE program will accept up to 10 different surface current cases each consisting of up to 20,000 node points and 10,000 triangle definitions and will animate one of these cases. The capability is used to compare surface-current distribution due to various initial excitation directions or electric field orientations. The program can accept up to 50 planes of field data consisting of a grid of 100 by 100 field points. These planes of data are user selectable and can be viewed individually or concurrently. With these preset limits, the program requires 55 megabytes of core memory to run. These limits can be changed in the header files to accommodate the available core memory of an individual workstation. An estimate of memory required can be made as follows: approximate memory in bytes equals (number of nodes times number of surfaces times 14 variables times bytes per word, typically 4 bytes per floating point) plus (number of field planes times number of nodes per plane times 21 variables times bytes per word). This gives the approximate memory size required to store the field and surface-current data. The total memory size is approximately 400,000 bytes plus the data memory size. The animation calculations are performed in real time at any user set time step. For Silicon Graphics Workstations that have multiple processors, this program has been optimized to perform these calculations on multiple processors to increase animation rates. The optimized program uses the SGI PFA (Power FORTRAN Accelerator) library. On single processor machines, the parallelization directives are seen as comments to the program and will have no effect on compilation or execution. MOM3D and EM-ANIMATE are written in FORTRAN 77 for interactive or batch execution on SGI series computers running IRIX 3.0 or later. The RAM requirements for these programs vary with the size of the problem being solved. A minimum of 30Mb of RAM is required for execution of EM-ANIMATE; however, the code may be modified to accommodate the available memory of an individual workstation. For EM-ANIMATE, twenty-four bit, double-buffered color capability is suggested, but not required. Sample executables and sample input and output files are provided. Electronic documentation is provided for both EM-ANIMATE and MOM3D in PostScript format. Documentation for EM-ANIMATE is also provided in the form of IRIX man pages. The standard distribution medium for COS-10048 is a .25 inch streaming magnetic IRIX tape cartridge in UNIX tar format. MOM3D and EM-ANIMATE are also available separately as LAR-15074 and LAR-15075, respectively. MOM3D was developed in 1992. EM-ANIMATE was developed in 1993.

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: A summary is presented of vortex control applications and current techniques for the control of longitudinal vortices produced by bodies, leading edges, tips and intersections. Vortex control has up till now been performed by many approaches in an empirical fashion, assisted by the essentially inviscid nature of much of longitudinal vortex behavior. Attention is given to Reynolds number sensitivities, vortex breakdown and interactions, vortex control on highly swept wings, and vortex control in juncture flows.

Description: The flow-field within an axial flow turbomachine, such as a turbine or compressor, is extremely complex because of three-dimensional features such as hub-corner stall, tip-leakage flows, and airfoil wakes. These flow features interact with each other and with rotor and stator airfoils inducing time-varying forces on the airfoils. These complicated rotor-stator interactions must be understood in order to design turbomachines that are light and compact as well as reliable and efficient. Two codes, STAGE-2 and STAGE-3, have been developed to compute these unsteady rotor-stator interaction flows in multistage turbomachines. An implicit, thin-layer Euler/Navier-Stokes zonal algorithm is used to compute the unsteady flow-field within both turbine and compressor configurations. Results include surface pressures and wake profiles for two-dimensional turbine and compressor configurations and surface pressures for a three-dimensional single-stage turbine configuration. The results compare well with experimental data and other unsteady computations.

Description: Jet noise and jet-induced structural loads have become key issues in the design of commercial and military aircraft. Computational Fluid Dynamics (CFD) can be of use in predicting the underlying jet shear-layer instabilities and, in conjunction with classical acoustic theory, jet noise. The computational issues involved in the resolution of high Reynolds number unsteady jet flows are addressed in this paper. Once these jet flows can be accurately resolved, it should be possible to use acoustic theory to extract, for example, the far-field jet noise. An assessment of future work and computational resources required for directly computing far-field jet noise is also presented.

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: The phased array has become a key component in the design of advanced antenna systems. This computer program was developed to examine the radiation characteristics of a generalized phased array antenna. Using a very efficient numerical technique, this program calculates the array's radiated power and its directivity. The results can be used to determine the radiation pattern of a generalized phased array at near- or far-field observation points. This program is a key research tool at the NASA Lewis Research Center for analyzing advanced space communication antenna systems. Results from this program compare favorably with experimental Lewis results for arrays of 2x2 and 3x3 elements. Given the array geometry and element characteristics, generalized phased array attributes can be broken into two areas: 1) the power radiated and its directivity at any given point, and 2) the co- and cross-polarization field components. This program allows arbitrarily located source elements with an analytically described cosine pattern. The formulation is based on a Romberg integration scheme and takes into account arbitrary element polarization, E and H plane element patterns, and mutual coupling. The input consists of the array geometry; phase, amplitude, linear and circular polarization of each source element; and the cosine functions of the element pattern. The output is a series of observation angles with their associated field magnitude and polarizations. Total radiated power and peak directivity are also calculated. This program is written in FORTRAN IV for batch execution and has been implemented on an IBM 370 computer operating under TSS with a central memory requirement of approximately 22K of 8 bit bytes. The IBM Scientific Subroutine Package (SSP) is required to run the program. This program was developed in 1986.

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: The EM-ANIMATE program is a specialized visualization program that displays and animates the near-field and surface-current solutions obtained from an electromagnetics program, in particular, that from MOM3D (LAR-15074). The EM-ANIMATE program is windows based and contains a user-friendly, graphical interface for setting viewing options, case selection, file manipulation, etc. EM-ANIMATE displays the field and surface-current magnitude as smooth shaded color fields (color contours) ranging from a minimum contour value to a maximum contour value for the fields and surface currents. The program can display either the total electric field or the scattered electric field in either time-harmonic animation mode or in the root mean square (RMS) average mode. The default setting is initially set to the minimum and maximum values within the field and surface current data and can be optionally set by the user. The field and surface-current value are animated by calculating and viewing the solution at user selectable radian time increments between 0 and 2pi. The surface currents can also be displayed in either time-harmonic animation mode or in RMS average mode. In RMS mode, the color contours do not vary with time, but show the constant time averaged field and surface-current magnitude solution. The electric field and surface-current directions can be displayed as scaled vector arrows which have a length proportional to the magnitude at each field grid point or surface node point. These vector properties can be viewed separately or concurrently with the field or surface-current magnitudes. Animation speed is improved by turning off the display of the vector arrows. In RMS modes, the direction vectors are still displayed as varying with time since the time averaged direction vectors would be zero length vectors. Other surface properties can optionally be viewed. These include the surface grid, the resistance value assigned to each element of the grid, and the power dissipation of each element which has an assigned resistance value. The EM-ANIMATE program will accept up to 10 different surface current cases each consisting of up to 20,000 node points and 10,000 triangle definitions and will animate one of these cases. The capability is used to compare surface-current distribution due to various initial excitation directions or electric field orientations. The program can accept up to 50 planes of field data consisting of a grid of 100 by 100 field points. These planes of data are user selectable and can be viewed individually or concurrently. With these preset limits, the program requires 55 megabytes of core memory to run. These limits can be changed in the header files to accommodate the available core memory of an individual workstation. An estimate of memory required can be made as follows: approximate memory in bytes equals (number of nodes times number of surfaces times 14 variables times bytes per word, typically 4 bytes per floating point) plus (number of field planes times number of nodes per plane times 21 variables times bytes per word). This gives the approximate memory size required to store the field and surface-current data. The total memory size is approximately 400,000 bytes plus the data memory size. The animation calculations are performed in real time at any user set time step. For Silicon Graphics Workstations that have multiple processors, this program has been optimized to perform these calculations on multiple processors to increase animation rates. The optimized program uses the SGI PFA (Power FORTRAN Accelerator) library. On single processor machines, the parallelization directives are seen as comments to the program and will have no effect on compilation or execution. EM-ANIMATE is written in FORTRAN 77 for implementation on SGI IRIS workstations running IRIX 3.0 or later. A minimum of 55Mb of RAM is required for execution of this program; however, the code may be modified to accommodate the available memory of an individual workstation. For program execution, twenty-four bit, double-buffered color capability is suggested, but not required. Sample input and output files and a sample executable are provided on the distribution medium. Electronic documentation is provided in PostScript format and in the form of IRIX man pages. The standard distribution medium for EM-ANIMATE is a .25 inch streaming magnetic IRIX tape cartridge in UNIX tar format. EM-ANIMATE is also available as part of a bundled package, COS-10048 that includes MOM3D, an IRIS program that produces electromagnetic near field and surface current solutions. This program was developed 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: Unsteady flow behavior and load characteristics of a 2D VR-7 airfoil with and without a leading-edge slat were studied in the water tunnel of the Aeroflightdynamics Directorate, NASA Ames Research Center. Both airfoils were oscillated sinusoidally between 5 and 25 deg at Re = 200,000 to obtain the unsteady lift, drag, and pitching moment data. A fluorescent dye was released from an orifice located at the leading edge of the airfoil for the purpose of visualizing the boundary layer and wake flow. The flowfield and load predictions of an incompressible Navier-Stokes code based on a velocity-vorticity formulation were compared with the test data. The test and predictions both confirm that the slatted VR-7 airfoil delays both static and dynamic stall as compared to the VR-7 airfoil alone.

Description: The problem of the hypersonic double ellipse in rarefied flow is treated by a particle method using the collision model first described by McDonald (1988). In the approach used here, the computational overhead is reduced by using simple cubic cells. The problem of the definition of complex geometries is addressed by developing an algorithm to define the relation of a body surface to the network of cells.

Description: Program LAURA (Langley Aerothermodynamic Upwind Relaxation Algorithm) is an upwind-biased, point-implicit relaxation algorithm for obtaining the numerical solution to the governing equations for 3D viscous hypersonic flows in chemical and thermal nonequilibrium. The algorithm is derived using a finite-volume formulation in which the inviscid components of flux across cell walls are described with a modified Roe's averaging and with second-order corrections based on Yee's Symmetric Total Variation Diminishing scheme. The code has been applied to Problem 8.2 of this workshop for the case of thermochemical nonequilibrium flow through a nozzle. Chemical reaction rates are defined with the model of Park (1987). Thermal nonequilibrium is modeled using a two-temperature approximation in which the vibrational energies of all molecules are assumed to be in equilibrium at a single temperature which is generally different from the translational-rotational temperature. Two grids were used to define the flow for the original problem, with a stagnation temperature of 6500 K. A third case with a stagnation temperature of 10,000 K is also presented. The solution domain includes the converging nozzle, subsonic flow domain in which the gas is substantially in thermochemical equilibrium and the diverging nozzle, hypersonic flow domain in which the gas is substantially in thermochemical nonequilibrium.

Description: Solutions have been computed and results are presented for Problem 1, the case of Mach 9 transitional flow past a 7 deg half-angle cone at zero incidence. The solutions were computed using a code developed for the integration of the parabolized Navier-Stokes equations. The algorithm employed in the code is based on a Roe-type flux-difference-splitting scheme applied following a finite-volume approach. The basic algorithm has been modified to make it implicit and second-order accurate in the crossflow directions. Results are presented in terms of surface pressure and heat transfer as well as boundary layer profiles of pitot pressure, Mach number, and tangential velocity. The case was recalculated several times in an effort to determine sensitivities to such parameters as grid density, wall temperature, turbulence model parameters, as well as freestream expansion. Comparisons with the experimental data are presented and discussed.

Description: A development status evaluation is presented for the theoretical understanding and design conceptualization of boundary layer control (BLC) systems applicable to supersonic transports, such as the currently envisioned NASA High Speed Civil Transport. By reducing fuel burned, supersonic BLC techniques could expand ranges to Pacific-crossing scales, while lowering sonic boom effects and upper-atmosphere pollution and even reducing skin friction temperature. The critical consideration for supersonic BLC is the presence of wave effects.

Description: Attention is drawn to the influence of preexisting finite-amplitude instabilities on the growth of other disturbances; current design tools for LFC take no notice of this kind of interaction. When a rational accounting is accomplished for the evolution of incoming disturbances in finite-amplitude solutions of the equations of motion, future transition-prediction methods will need to take these wave interactions into account. Attention is given here to interactions in the presence of crossflow vortices and interactions involving Goertler vortices.

Description: A development history and a development-trends evaluation are presented for laminar flow controlled airfoil technologies and design concepts, including the search for 'natural' laminar flow and actively controlled flow via suction through small pores on the airfoil surface. While most NASA activities in this field have been concerned with subsonic aircraft, it has been projected that the control of boundary layer turbulence may be even more critical to the aerodynamic efficiency of supersonic aircraft. Developmental programs for these techniques have been conducted with several modified conventional aircraft.

Description: An account is given of the development history of natural laminar-flow (NLF) airfoil profiles under guidance of an experimentally well-verified theoretical method for the design of airfoils suited to virtually all subcritical applications. This method, the Eppler Airfoil Design and Analysis Program, contains a conformal-mapping method for airfoils having prescribed velocity-distribution characteristics, as well as a panel method for the analysis of potential flow about given airfoils and a boundary-layer method. Several of the NLF airfoils thus obtained are discussed.

Description: Airloads measured on a two-bladed helicopter rotor in flight, from the Tip Aerodynamic and Acoustic Test, are compared with calculations from a comprehensive helicopter analysis (CAMRAD/JA), and the pressures compared with calculations from a full-potential rotor code (FPR). The flight test results cover an advance ratio range from 0.19 to 0.38. The lowest speed case is characterized by the presence of significant blade-vortex interactions. Good correlation of peak-to-peak vortex-induced loads and the corresponding pressures is obtained. The results of the correlation for this two-bladed rotor are substantially similar to the results for three- and four-bladed rotors, concerning the tip vortex core size for best correlation, calculation of the peak-to-peak loads on the retreating side, and calculation of vortex-induced loads on inboard radial stations.

Description: A new CFD potential code, FPX (eXtended Full-Potential), has been developed for application to both helicopters and tilt-rotors. The code solves the unsteady, three-dimensional full potential equation and is an extension of the rotor code, FPR. Both entropy and viscosity corrections are included to enhance the physical modeling capabilities. A number of efficiency related modifications have yielded a factor of two speed-up in the code. An axial flow capability has been added to treat tilt-rotor in forward flight (cruise mode). In order to employ streamwise periodicity and accurately solve for the propagation of acoustic signals in the tip region, an H-H topology has been added to the basic O-H grid system. Computations are performed for the XV-15 Standard and ATB blades at high-speed conditions. Comparisons are made for the blade aerodynamics and the induced fuselage cabin pressure for a range of Mach numbers. Grid generation, wake treatment, and far-field wall treatment are identified as problem areas with recommendations for future research.

Description: A preliminary test/theory correlation evaluation is conducted for wake measurement test results obtained by LDV for a B360 helicopter rotor, at conditions critical to the understanding of wake-rollup and blade-vortex interaction phenomena. The LDV data were complemented by acoustic, blade pressure, rotor performance, and blade/control load measurements.

Description: Analytic expressions have been derived for estimating the nonablating laminar and turbulent boundary-layer convective heating rates on inclined flat surfaces for the Martian atmosphere in thermochemical equilibrium. The equations are valid in the speed and altitude regime where aerobraking would occur at Mars. Comparisons with limited experimental measurements and calculations for CO2 (the Martian atmosphere is 95.6 percent CO2) yielded reasonably good agreement, especially for the ratios of heating rates in CO2 to those in air at the same conditions. In the aerobraking speed regime, the laminar flat surface boundary layer heating rates are 15-25 percent greater at Mars than in air. The differences between the turbulent heating rates are even more pronounced. The turbulent heating rates can be over 50 percent greater at Mars than in air at the same flight conditions.

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: An incompressible three-dimensional laminar boundary-layer flow over a swept wing is used as a model to study both the wall-curvature and streamline-curvature effects on the stationary crossflow instability. The basic state is obtained by solving the full Navier-Stokes (N-S) equations numerically. The linear disturbance equations are cast on a fixed, body-intrinsic, curvilinear coordinate system. Those nonparallel terms which contribute mainly to the streamline-curvature effect are retained in the formulation of the disturbance equations and approximated by their local finite difference values. The resulting eigenvalue problem is solved by a Chebyshev collocation method. The present results indicate that the convex wall curvature has a stabilizing effect, whereas the streamline curvature has a destabilizing effect. A validation of these effects with an N-S solution for the linear disturbance flow is provided.

Description: The paper presents a general approach to constructing mean velocity profiles for compressible turbulent boundary layers with isothermal or adiabatic walls. The theory is based on a density-weighted transformation that allows the extension of the incompressible similarity laws of the wall to the compressible regions. The velocity profile family is compared to a range of experimental data, and excellent agreement is obtained. A self-consistent skin friction law, which satisfies the proposed velocity profile family, is derived and compared with the well-known Van Driest II theory for boundary layers in zero pressure gradient. The results are found to be at least as good as those obtained by using the Van Driest II transformation.

Description: A number of chemical-kinetic problems related to phenomena occurring behind a shock wave surrounding an object flying in the earth atmosphere are discussed, including the nonequilibrium thermochemical relaxation phenomena occurring behind a shock wave surrounding the flying object, problems related to aerobraking maneuver, the radiation phenomena for shock velocities of up to 12 km/sec, and the determination of rate coefficients for ionization reactions and associated electron-impact ionization reactions. Results of experiments are presented in form of graphs and tables, giving data on the reaction rate coefficients for air, the ionization distances, thermodynamic properties behind a shock wave, radiative heat flux calculations, Damkoehler numbers for the ablation-product layer, together with conclusions.

Description: Computations were made for those test cases of Problem 3 which were designated as laminar flows, viz., test cases 3.1, 3.2, 3.4, and 3.5. These test cases corresponded to flows over a flat plate and a compression ramp at high Mach number and at high Reynolds number. The computations over the compression ramps indicate a substantial streamwise extent of separation. Based on previous experience with separated laminar flows at high Mach numbers which indicated a substantial effect with spatial grid refinement, a series of computations with different grid sizes were performed. Also, for the flat plate, comparisons of the results for two different algorithms were made.

Description: A development status evaluation is presented for CFD methods applicable to fuselage-integrated scramjet powerplant incorporating hypersonic vehicles; these methods are critically important due to the unavailability of experimental facilities for such elevated Mach number/high-enthalphy conditions. Advancements are required in algorithm robustness and speed, geometric flexibility, and the inclusion of more complete flow physics. The most serious deficiencies lie in turbulence modeling, the lack of complete transition-prediction methods, and combustion modeling.

Description: An overview is given of research activity on the application of computational fluid dynamics (CDF) for hypersonic propulsion systems. After the initial consideration of the highly integrated nature of air-breathing hypersonic engines and airframe, attention is directed toward computations carried out for the components of the engine. A generic inlet configuration is considered in order to demonstrate the highly three dimensional viscous flow behavior occurring within rectangular inlets. Reacting flow computations for simple jet injection as well as for more complex combustion chambers are then discussed in order to show the capability of viscous finite rate chemical reaction computer simulations. Finally, the nozzle flow fields are demonstrated, showing the existence of complex shear layers and shock structure in the exhaust plume. The general issues associated with code validation as well as the specific issue associated with the use of CFD for design are discussed. A prognosis for the success of CFD in the design of future propulsion systems is offered.

Description: Arrays of slots with arbitrary orientation in a conducting plane which are infinitely periodic in one dimension and finitely periodic in another dimension are considered. The perfect conducting plane extends infinitely and is bounded on each side by dielectric slabs of finite thickness and infinite extent. Single columns of slots are represented by equivalent magnetic scattering currents, which are solved for via the moment method. The mutual coupling (admittance) between slot columns in the presence of the stratified media is found by the array scanning method (ASM), which expresses the admittance as the average of the scan admittance of an artificially constructed doubly infinite array of slots over all real scan angles. The technique avoids the use of Sommerfeld integrals, but still gives rise to singularities at scan angles corresponding to the resonant excitation of surface waves. An analytical approximation is made to remove these surface wave singularities, thus making numerical implementation of the method practical. The method is quite general and may be extended to thin slot elements of arbitrary shape and orientation.

Description: This is an effort aimed at validating recent hover prediction methods. The experimental basis for this validation work is an extensive set of loads, wake and performance data, which were obtained from a pressure instrumented model UH-60 rotor tested at the Sikorsky hover test facility and at Duits-Nederlandse Windtunnel (DNW). This model was equipped with replaceable tips - including a tapered and a BERP-type tip - which permitted studies of the effects of rotor geometry. The central prediction method studied is a free-wake, vortex embedded, full-potential CFD method - called HELIX-I. It is found that the HELIX-I code produces very good comparisons with the data including wake, surface pressure and performance. Comparisons with the measured radial load distributions have permitted an improved understanding of the wake resolution modelling requirements of CFD methods. Since HELIX-I is a combined Eulerian/Lagrangian method, limited comparisons are also made with a Lagrangian boundary element code (called EHPIC) and an Eulerian Navier-Stokes code (called TURNS). In most cases all methods produce good comparisons with the data. It is found that the HELIX-I code provides a good compromise between the speed of boundary integral methods and the comprehensive nature of Navier-Stokes methods.

Description: Shadowgraph flow visualization images were acquired from a 0.184-scale tiltrotor and wing in hover. Measurements and details of the vortex core structure were examined as a function of thrust condition and wake age. Experimental data for the isolated rotor wake geometry and rotor wake interactions with a semi-span wing and image plane were acquired. Quantitative measurements and comparisons of wake geometry and distortion were made for three configurations: the isolated rotor, rotor/wing, and rotor/wing/image plane. Comparisons between tiltrotor and helicopter rotor wake geometry measurements were made. Experimental wake geometry data were also compared with two wake models. Suggestions for improvements to existing prescribed-wake and free-wake models are proposed.