ALBERT

Description: System studies have shown the benefits of reducing blade tip clearances in modern turbine engines. Minimizing blade tip clearances throughout the engine will contribute materially to meeting NASA s Ultra-Efficient Engine Technology (UEET) turbine engine project goals. NASA GRC is examining two candidate approaches including rub-avoidance and regeneration which are explained in subsequent slides.

Description: As the aviation industry moves toward higher efficiency electrical power generation, all electric aircraft, or zero emissions and more quiet aircraft, fuel cells are sought as the technology that can deliver on these high expectations. The hybrid solid oxide fuel cell system combines the fuel cell with a micro-turbine to obtain up to 70% cycle efficiency, and then distributes the electrical power to the loads via a power distribution system. The challenge is to understand the dynamics of this complex multidiscipline system and the design distributed controls that take the system through its operating conditions in a stable and safe manner while maintaining the system performance. This particular system is a power generation and a distribution system, and the fuel cell and micro-turbine model fidelity should be compatible with the dynamics of the power distribution system in order to allow proper stability and distributed controls design. The novelty in this paper is that, first, the case is made why a high fidelity fuel cell mode is needed for systems control and stability designs. Second, a novel modeling approach is proposed for the fuel cell that will allow the fuel cell and the power system to be integrated and designed for stability, distributed controls, and other interface specifications. This investigation shows that for the fuel cell, the voltage characteristic should be modeled but in addition, conservation equation dynamics, ion diffusion, charge transfer kinetics, and the electron flow inherent impedance should also be included.

Description: The usage and integrated vehicle health management of the NASA C-17. Propulsion health management flight objectives for the aircraft include mapping of the High Pressure Compressor in order to calibrate a Pratt and Whitney engine model and the fusion of data collected from existing sensors and signals to develop models, analysis methods and information fusion algorithms. An additional health manage flight objective is to demonstrate that the Commercial Modular Aero-Propulsion Systems Simulation engine model can successfully execute in real time onboard the C-17 T-1 aircraft using engine and aircraft flight data as inputs. Future work will address aircraft durability and aging, airframe health management, and propulsion health management research in the areas of gas path and engine vibration.

Description: A vane has an airfoil shell and a spar within the shell. The vane has an outboard shroud at an outboard end of the shell and an inboard platform at an inboard end of the shell. The spar has a first chamber essentially along the suction side and a second chamber along the pressure side opposite the first chamber.

Description: Key aspects of the design of sealing systems for On Rotor Combustion/Wave Rotor (ORC/WR) systems were addressed. ORC/WR systems generally fit within a broad class of pressure gain Constant Volume Combustors (CVCs) or Pulse Detonation Combustors (PDCs) which are currently being considered for use in many classes of turbine engines for dramatic efficiency improvement. Technology readiness level of this ORC/WR approaches are presently at 2.0. The results of detailed modeling of an ORC/WR system as applied to a regional jet engine application were shown to capture a high degree of pressure gain capabilities. The results of engine cycle analysis indicated the level of specific fuel consumption (SFC) benefits to be 17 percent. The potential losses in pressure gain due to leakage were found to be closely coupled to the wave processes at the rotor endpoints of the ORC/WR system. Extensive investigation into the sealing approaches is reported. Sensitivity studies show that SFC gains of 10 percent remain available even when pressure gain levels are highly penalized. This indicates ORC/WR systems to have a high degree of tolerance to rotor leakage effects but also emphasizes their importance. An engine demonstration of an ORC/WR system is seen as key to progressing the TRL of this technology. An industrial engine was judged to be a highly advantageous platform for demonstration of a first generation ORC/WR system. Prior to such a demonstration, the existing NASA pressure exchanger wave rotor rig was identified as an opportunity to apply both expanded analytical modeling capabilities developed within this program and to identify and fix identified leakage issues existing within this rig. Extensive leakage analysis of the rig was performed and a detailed design of additional sealing strategies for this rig was generated.

Description: Computational fluid dynamics (CFD) analysis has been performed to study the plume effects on sonic boom signature for isolated nozzle configurations. The objectives of these analyses were to provide comparison to past work using modern CFD analysis tools, to investigate the differences of high aspect ratio nozzles to circular (axisymmetric) nozzles, and to report the effects of underexpanded nozzle operation on boom signature. CFD analysis was used to address the plume effects on sonic boom signature from a baseline exhaust nozzle. Near-field pressure signatures were collected for nozzle pressure ratios (NPRs) between 6 and 10. A computer code was used to extrapolate these signatures to a ground-observed sonic boom N-wave. Trends show that there is a reduction in sonic boom N-wave signature as NPR is increased from 6 to 10. The performance curve for this supersonic nozzle is flat, so there is not a significant loss in thrust coefficient as the NPR is increased. As a result, this benefit could be realized without significant loss of performance. Analyses were also collected for a high aspect ratio nozzle based on the baseline design for comparison. Pressure signatures were collected for nozzle pressure ratios from 8 to 12. Signatures were nearly twice as strong for the two-dimensional case, and trends also show a reduction in sonic boom signature as NPR is increased from 8 to 12. As low boom designs are developed and improved, there will be a need for understanding the interaction between the aircraft boat tail shocks and the exhaust nozzle plume. These CFD analyses will provide a baseline study for future analysis efforts.

Description: Helicopter Health Usage Monitoring Systems (HUMS) have potential for providing data to support increasing the service life of a dynamic mechanical component in the transmission of a helicopter. Data collected can demonstrate the HUMS condition indicator responds to a specific component fault with appropriate alert limits and minimal false alarms. Defining thresholds for specific faults requires a tradeoff between the sensitivity of the condition indicator (CI) limit to indicate damage and the number of false alarms. A method using Receiver Operating Characteristic (ROC) curves to assess CI performance was demonstrated using CI data collected from accelerometers installed on several UH60 Black Hawk and AH64 Apache helicopters and an AH64 helicopter component test stand. Results of the analysis indicate ROC curves can be used to reliably assess the performance of commercial HUMS condition indicators to detect damaged gears and bearings in a helicopter transmission.

Description: Flaps (or half wedges) attached to the sides of a pylon are shown to result in a small but clear noise benefit. Noise radiated towards the ground is reduced apparently through a deflection and thickening of the fan stream underneath. Based on results from the current as well as concurrent investigations at the University of California at Irvine, it is recommended that further tests in a larger facility simulating realistic engine conditions be considered.

Description: The objective is to provide turbine-cooling technologies to meet Propulsion 21 goals related to engine fuel burn, emissions, safety, and reliability. Specifically, the GE Aviation (GEA) Advanced Turbine Cooling and Thermal Management program seeks to develop advanced cooling and flow distribution methods for HP turbines, while achieving a substantial reduction in total cooling flow and assuring acceptable turbine component safety and reliability. Enhanced cooling techniques, such as fluidic devices, controlled-vortex cooling, and directed impingement jets, offer the opportunity to incorporate both active and passive schemes. Coolant heat transfer enhancement also can be achieved from advanced designs that incorporate multi-disciplinary optimization of external film and internal cooling passage geometry.

Description: The objective of the Advanced Turbine Cooling and Thermal Management program is to develop intelligent control and distribution methods for turbine cooling, while achieving a reduction in total cooling flow and assuring acceptable turbine component safety and reliability. The program also will develop embedded sensor technologies and cooling system models for real-time engine diagnostics and health management. Both active and passive control strategies will be investigated that include the capability of intelligent modulation of flow quantities, pressures, and temperatures both within the supply system and at the turbine component level. Thermal management system concepts were studied, with a goal of reducing HPT blade cooling air supply temperature. An assessment will be made of the use of this air by the active clearance control system as well. Turbine component cooling designs incorporating advanced, high-effectiveness cooling features, will be evaluated. Turbine cooling flow control concepts will be studied at the cooling system level and the component level. Specific cooling features or sub-elements of an advanced HPT blade cooling design will be downselected for core fabrication and casting demonstrations.

Description: For the Intelligent Engine System (Propulsion 21) study, each technology was evaluated to determine the impact to fuel burn, acoustics, and NOx emissions. The optimum combination of technologies and their overall benefits to the system were also evaluated, resulting in noise improvement potential of 1.89 EPNdB cumulative margin,-1.34 percent fuel burn, and 50 percent NOx reduction from the 2015 UEET-QAT baseline. All the technology evaluations, except T18-20D, were based on newengines, where the engine was resized to obtain the maximum system benefit while maintaining the same cycle parameters as the 2015 UEET-QAT baseline. The impact of turbine clearance control on deteriorated engines, T18-20D, was also evaluated. Recommendations for future system study work include, but were not limited to, validation of a university-developed engine deterioration model and customer value analysis as figures of merit beside fuel burn, emissions, and acoustics.

Description: An adaptive controls method for instability suppression in gas turbine engine combustors has been developed and successfully tested with a realistic aircraft engine combustor rig. This testing was part of a program that demonstrated, for the first time, successful active combustor instability control in an aircraft gas turbine engine-like environment. The controls method is called Adaptive Sliding Phasor Averaged Control. Testing of the control method has been conducted in an experimental rig with different configurations designed to simulate combustors with instabilities of about 530 and 315 Hz. Results demonstrate the effectiveness of this method in suppressing combustor instabilities. In addition, a dramatic improvement in suppression of the instability was achieved by focusing control on the second harmonic of the instability. This is believed to be due to a phenomena discovered and reported earlier, the so called Intra-Harmonic Coupling. These results may have implications for future research in combustor instability control.

Description: Low-noise fan exit guide vanes are disclosed. According to the present invention a fan exit guide vane has an outer shell substantially shaped as an airfoil and defining an interior cavity. A porous portion of the outer shell allows communication between the fluctuations in the air passing over the guide vane and the interior cavity. At least one acoustically resonant chamber is located within the interior cavity. The resonant chamber is in communication with the porous portion of the outer perimeter. The resonant chamber is configured to reduce the noise generated at a predetermined frequency. In various preferred embodiments, there is a plurality of acoustically resonant chambers located within the interior cavity. The resonant chambers can be separated by one or more partitions within the interior cavity. In these embodiments, the resonant chambers can be configured to reduce the noise generated over a range of predetermined frequencies.

Description: Pulsed combustion is receiving renewed interest as a potential route to higher performance in air breathing propulsion systems. Pulsejets offer a simple experimental device with which to study unsteady combustion phenomena and validate simulations. Previous computational fluid dynamic (CFD) simulation work focused primarily on the pulsejet combustion and exhaust processes. This paper describes a new inlet sub-model which simulates the fluidic and mechanical operation of a valved pulsejet head. The governing equations for this sub-model are described. Sub-model validation is provided through comparisons of simulated and experimentally measured reed valve motion, and time averaged inlet mass flow rate. The updated pulsejet simulation, with the inlet sub-model implemented, is validated through comparison with experimentally measured combustion chamber pressure, inlet mass flow rate, operational frequency, and thrust. Additionally, the simulated pulsejet exhaust flowfield, which is dominated by a starting vortex ring, is compared with particle imaging velocimetry (PIV) measurements on the bases of velocity, vorticity, and vortex location. The results show good agreement between simulated and experimental data. The inlet sub-model is shown to be critical for the successful modeling of pulsejet operation. This sub-model correctly predicts both the inlet mass flow rate and its phase relationship with the combustion chamber pressure. As a result, the predicted pulsejet thrust agrees very well with experimental data.

Description: Mission Support Features: a) Shirtsleeve environment, . 18 scientists; b) worldwide deployment experience; c) Extensive modifications to support in-situ and remote sensing instruments 1) zenith and nadir viewports; 2) modified power systems; 3) 19 inch rack mounting; 4) on-board data acquisition network.

Description: The Model Based Fault Tolerant Control (MBFTC) task was conducted under the NASA Aviation Safety and Security Program. The goal of MBFTC is to develop and demonstrate real-time strategies to diagnose and accommodate anomalous aircraft engine events such as sensor faults, actuator faults, or turbine gas-path component damage that can lead to in-flight shutdowns, aborted take offs, asymmetric thrust/loss of thrust control, or engine surge/stall events. A suite of model-based fault detection algorithms were developed and evaluated. Based on the performance and maturity of the developed algorithms two approaches were selected for further analysis: (i) multiple-hypothesis testing, and (ii) neural networks; both used residuals from an Extended Kalman Filter to detect the occurrence of the selected faults. A simple fusion algorithm was implemented to combine the results from each algorithm to obtain an overall estimate of the identified fault type and magnitude. The identification of the fault type and magnitude enabled the use of an online fault accommodation strategy to correct for the adverse impact of these faults on engine operability thereby enabling continued engine operation in the presence of these faults. The performance of the fault detection and accommodation algorithm was extensively tested in a simulation environment.

Description: An assessment was made of the capability of jet noise prediction codes over a broad range of jet flows, with the objective of quantifying current capabilities and identifying areas requiring future research investment. Three separate codes in NASA s possession, representative of two classes of jet noise prediction codes, were evaluated, one empirical and two statistical. The empirical code is the Stone Jet Noise Module (ST2JET) contained within the ANOPP aircraft noise prediction code. It is well documented, and represents the state of the art in semi-empirical acoustic prediction codes where virtual sources are attributed to various aspects of noise generation in each jet. These sources, in combination, predict the spectral directivity of a jet plume. A total of 258 jet noise cases were examined on the ST2JET code, each run requiring only fractions of a second to complete. Two statistical jet noise prediction codes were also evaluated, JeNo v1, and Jet3D. Fewer cases were run for the statistical prediction methods because they require substantially more resources, typically a Reynolds-Averaged Navier-Stokes solution of the jet, volume integration of the source statistical models over the entire plume, and a numerical solution of the governing propagation equation within the jet. In the evaluation process, substantial justification of experimental datasets used in the evaluations was made. In the end, none of the current codes can predict jet noise within experimental uncertainty. The empirical code came within 2dB on a 1/3 octave spectral basis for a wide range of flows. The statistical code Jet3D was within experimental uncertainty at broadside angles for hot supersonic jets, but errors in peak frequency and amplitude put it out of experimental uncertainty at cooler, lower speed conditions. Jet3D did not predict changes in directivity in the downstream angles. The statistical code JeNo,v1 was within experimental uncertainty predicting noise from cold subsonic jets at all angles, but did not predict changes with heating of the jet and did not account for directivity changes at supersonic conditions. Shortcomings addressed here give direction for future work relevant to the statistical-based prediction methods. A full report will be released as a chapter in a NASA publication assessing the state of the art in aircraft noise prediction.

Description: A concept for mitigating the adverse effects of jet vorticity and liftoff at high blowing ratios for turbine film cooling flows has been developed and studied at NASA Glenn Research Center. This "anti-vortex" film cooling concept proposes the addition of two branched holes from each primary hole in order to produce a vorticity counter to the detrimental kidney vortices from the main jet. These vortices typically entrain hot freestream gas and are associated with jet separation from the turbine blade surface. The anti-vortex design is unique in that it requires only easily machinable round holes, unlike shaped film cooling holes and other advanced concepts. The anti-vortex film cooling hole concept has been modeled computationally for a single row of 30deg angled holes on a flat surface using the 3D Navier-Stokes solver Glenn-HT. A modification of the anti-vortex concept whereby the branched holes exit adjacent to the main hole has been studied computationally for blowing ratios of 1.0 and 2.0 and at density ratios of 1.0 and 2.0. This modified concept was selected because it has shown the most promise in recent experimental studies. The computational results show that the modified design improves the film cooling effectiveness relative to the round hole baseline and previous anti-vortex cases, in confirmation of the experimental studies.

Description: Recent technology reviews have identified the need for objective assessments of engine health management (EHM) technology. The need is two-fold: technology developers require relevant data and problems to design and validate new algorithms and techniques while engine system integrators and operators need practical tools to direct development and then evaluate the effectiveness of proposed solutions. This paper presents a publicly available gas path diagnostic benchmark problem that has been developed by the Propulsion and Power Systems Panel of The Technical Cooperation Program (TTCP) to help address these needs. The problem is coded in MATLAB (The MathWorks, Inc.) and coupled with a non-linear turbofan engine simulation to produce "snap-shot" measurements, with relevant noise levels, as if collected from a fleet of engines over their lifetime of use. Each engine within the fleet will experience unique operating and deterioration profiles, and may encounter randomly occurring relevant gas path faults including sensor, actuator and component faults. The challenge to the EHM community is to develop gas path diagnostic algorithms to reliably perform fault detection and isolation. An example solution to the benchmark problem is provided along with associated evaluation metrics. A plan is presented to disseminate this benchmark problem to the engine health management technical community and invite technology solutions.

Description: An accurate indication of available power is required for helicopter mission planning purposes. Available power is currently estimated on U.S. Army Blackhawk helicopters by performing a Maximum Power Check (MPC), a manual procedure performed by maintenance pilots on a periodic basis. The MPC establishes Engine Torque Factor (ETF), an indication of available power. It is desirable to replace the current manual MPC procedure with an automated approach that will enable continuous real-time assessment of available power utilizing normal mission data. This report presents an automated power assessment approach which processes data currently collected within helicopter Health and Usage Monitoring System (HUMS) units. The overall approach consists of: 1) a steady-state data filter which identifies and extracts steady-state operating points within HUMS data sets; 2) engine performance curve trend monitoring and updating; and 3) automated ETF calculation. The algorithm is coded in MATLAB (The MathWorks, Inc.) and currently runs on a PC. Results from the application of this technique to HUMS mission data collected from UH-60L aircraft equipped with T700-GE-701C engines are presented and compared to manually calculated ETF values. Potential future enhancements are discussed.

Description: In this paper, a baseline system which utilizes dual-channel sensor measurements for aircraft engine on-line diagnostics is developed. This system is composed of a linear on-board engine model (LOBEM) and fault detection and isolation (FDI) logic. The LOBEM provides the analytical third channel against which the dual-channel measurements are compared. When the discrepancy among the triplex channels exceeds a tolerance level, the FDI logic determines the cause of the discrepancy. Through this approach, the baseline system achieves the following objectives: (1) anomaly detection, (2) component fault detection, and (3) sensor fault detection and isolation. The performance of the baseline system is evaluated in a simulation environment using faults in sensors and components.

Description: Recent developments in gas foil bearing technology have led to numerous advanced high-speed rotating system concepts, many of which have become either commercial products or experimental test articles. Examples include Oil-Free microturbines, motors, generators and turbochargers. The driving forces for integrating gas foil bearings into these high-speed systems are the benefits promised by removing the oil lubrication system. Elimination of the oil system leads to reduced emissions, increased reliability, and decreased maintenance costs. Another benefit is reduced power plant weight. For rotorcraft applications, this would be a major advantage, as every pound removed from the propulsion system results in a payload benefit. Implementing foil gas bearings throughout a rotorcraft gas turbine engine is an important long-term goal that requires overcoming numerous technological hurdles. Adequate thrust bearing load capacity and potentially large gearbox applied radial loads are among them. However, by replacing the turbine end, or hot section, rolling element bearing with a gas foil bearing many of the above benefits can be realized. To this end, engine manufacturers are beginning to explore the possibilities of hot section gas foil bearings in propulsion engines. This paper presents a logical follow-on activity by analyzing a conceptual rotorcraft engine to determine the feasibility of a foil bearing supported core. Using a combination of rotordynamic analyses and a load capacity model, it is shown to be reasonable to consider a gas foil bearing core section.

Description: A methodology for the design and construction of simple foil thrust bearings intended for parametric performance testing and low marginal costs is presented. Features drawn from a review of the open literature are discussed as they relate to bearing performance. The design of fixtures and tooling required to fabricate foil thrust bearings is presented, using conventional machining processes where possible. A prototype bearing with dimensions drawn from the literature is constructed, with all fabrication steps described. A load-deflection curve for the bearing is presented to illustrate structural stiffness characteristics. Start-top cycles are performed on the bearing at a temperature of 425 C to demonstrate early-life wear patterns. A test of bearing load capacity demonstrates useful performance when compared with data obtained from the open literature.

Description: CFD calculations using high-performance parallel computing were conducted to simulate the pre-stall flow of a transonic compressor stage, NASA compressor Stage 35. The simulations were run with a full-annulus grid that models the 3D, viscous, unsteady blade row interaction without the need for an artificial inlet distortion to induce stall. The simulation demonstrates the development of the rotating stall from the growth of instabilities. Pressure-rise performance and pressure traces are compared with published experimental data before the study of flow evolution prior to the rotating stall. Spatial FFT analysis of the flow indicates a rotating long-length disturbance of one rotor circumference, which is followed by a spike-type breakdown. The analysis also links the long-length wave disturbance with the initiation of the spike inception. The spike instabilities occur when the trajectory of the tip clearance flow becomes perpendicular to the axial direction. When approaching stall, the passage shock changes from a single oblique shock to a dual-shock, which distorts the perpendicular trajectory of the tip clearance vortex but shows no evidence of flow separation that may contribute to stall.

Description: A model has been developed to simulate a fixed-exit porous bleed system for supersonic inlets. The fixed-exit model allows the amount of bleed flow to vary according to local flow conditions and fixed-exit characteristics of the bleed system. This variation is important for the control of shock-wave/boundary-layer interactions within the inlet. The model computes the bleed plenum static pressure rather than requiring its specification. The model was implemented in the Wind-US computational fluid dynamics code. The model was then verified and validated against experimental data for bleed on a flat plate with and without an impinging oblique shock and for bleed in a Mach 3.0 axisymmetric, mixed-compression inlet. The model was able to accurately correlate the plenum pressures with bleed rates and simulate the effect of the bleed on the downstream boundary layer. Further, the model provided a realistic simulation of the initiation of inlet unstart. The results provide the most in-depth examination to date of bleed models for use in the simulation of supersonic inlets. The results also highlight the limitations of the models and aspects that require further research.

Description: Meeting future goals for aircraft and air traffic system performance will require new airframes with more highly integrated propulsion. Previous studies have evaluated hybrid wing body (HWB) configurations with various numbers of engines and with increasing degrees of propulsion-airframe integration. A recently published configuration with 12 small engines partially embedded in a HWB aircraft, reviewed herein, serves as the airframe baseline for the new concept aircraft that is the subject of this paper. To achieve high cruise efficiency, a high lift-to-drag ratio HWB was adopted as the baseline airframe along with boundary layer ingestion inlets and distributed thrust nozzles to fill in the wakes generated by the vehicle. The distributed powered-lift propulsion concept for the baseline vehicle used a simple, high-lift-capable internally blown flap or jet flap system with a number of small high bypass ratio turbofan engines in the airframe. In that concept, the engine flow path from the inlet to the nozzle is direct and does not involve complicated internal ducts through the airframe to redistribute the engine flow. In addition, partially embedded engines, distributed along the upper surface of the HWB airframe, provide noise reduction through airframe shielding and promote jet flow mixing with the ambient airflow. To improve performance and to reduce noise and environmental impact even further, a drastic change in the propulsion system is proposed in this paper. The new concept adopts the previous baseline cruise-efficient short take-off and landing (CESTOL) airframe but employs a number of superconducting motors to drive the distributed fans rather than using many small conventional engines. The power to drive these electric fans is generated by two remotely located gas-turbine-driven superconducting generators. This arrangement allows many small partially embedded fans while retaining the superior efficiency of large core engines, which are physically separated but connected through electric power lines to the fans. This paper presents a brief description of the earlier CESTOL vehicle concept and the newly proposed electrically driven fan concept vehicle, using the previous CESTOL vehicle as a baseline.

Description: The purpose of this cooperative agreement was to develop a foundation of intelligent propulsion technologies for NASA and industry that will have an impact on safety, noise, emissions, and cost. These intelligent engine technologies included sensors, electronics, communications, control logic, actuators, smart materials and structures, and system studies. Furthermore, this cooperative agreement helped prepare future graduates to develop the revolutionary intelligent propulsion technologies that will be needed to ensure pre-eminence of the U.S. aerospace industry. This Propulsion 21 - Phase 11 program consisted of four primary research areas and associated work elements at Ohio universities: 1.0 Turbine Engine Prognostics, 2.0 Active Controls for Emissions and Noise Reduction, 3.0 Active Structural Controls and Performance, and 4.0 System Studies and Integration. Phase l, which was conducted during the period August 1, 2003, through September 30, 2004, has been reported separately.

Description: The Advanced Thermally Actuated Clearance Control System underwent several studies. Improved flow path isolation quantified what can be gained by making the HPT case nearly adiabatic. The best method of heat transfer was established, and finally two different borrowed air cooling circuits were evaluated to be used for the HPT Active Clearance Control System.

Description: This presentation reviews recent progress made under NASA s Subsonic Rotary Wing (SRW) propulsion research activities. Advances in engines, drive systems and optimized propulsion systems are discussed. Progress in wide operability compressors, modeling of variable geometry turbine performance, foil gas bearings and multi-speed transmissions are presented.

Description: Far-field noise sound power level (PWL) spectra and overall sound pressure level (OASPL) directivities were compared for three significantly different model fan stages which were tested in the NASA Glenn 9 15 Low Speed Wind Tunnel. The test fans included the Advanced Ducted Propulsor (ADP) Fan1, the baseline Source Diagnostic Test (SDT) fan, and the Quiet High Speed Fan2 (QHSF2). These fans had design rotor tangential tip speeds from 840 to 1474 ft/s and stage pressure ratios from 1.29 to 1.82. Additional parameters included rotor-stator spacing, stator sweep, and downstream support struts. Acoustic comparison points were selected on the basis of stage thrust. Acoustic results for the low tip speed/low pressure ratio fan (ADP Fan1) were thrust-adjusted to show how a geometrically-scaled version of this fan might compare at the higher design thrust levels of the other two fans. Lowest noise levels were typically observed for ADP Fan1 (which had a radial stator) and for the intermediate tip speed fan (Source Diagnostics Test, SDT, R4 rotor) with a swept stator. Projected noise levels for the ADP fan to the SDT swept stator configuration at design point conditions showed the fans to have similar noise levels. However, it is possible that the ADP fan could be 2 to 3 dB quieter with incorporation of a swept stator. Benefits of a scaled ADP fan include avoidance of multiple pure tones associated with transonic and higher blade tip speeds. Penalties of a larger size ADP fan would include increased nacelle size and drag.

Description: Control of jet noise continues to be an important research topic. Exhaust-nozzle chevrons have been shown to reduce jet noise, but parametric effects are not well understood. Additionally, thrust loss due to chevrons at cruise suggests significant benefit from active chevrons. The focus of this study is development of an active chevron concept for the primary purpose of parametric studies for jet noise reduction in the laboratory and secondarily for technology development to leverage for full scale systems. The active chevron concept employed in this work consists of a laminated composite structure with embedded shape memory alloy (SMA) actuators, termed a SMA hybrid composite (SMAHC). SMA actuators are embedded on one side of the neutral axis of the structure such that thermal excitation, via joule heating, generates a moment and deflects the structure. The performance of two active chevron concepts is demonstrated in the presence of representative flow conditions. One of the concepts is shown to possess significant advantages for the proposed application and is selected for further development. Fabrication and design changes are described and shown to produce a chevron prototype that meets the performance objectives.

Description: Researchers at NASA Glenn Research Center have been investigating high temperature shape memory alloys as potential damping materials for turbomachinery rotor blades. Analysis shows that a thin layer of SMA with a loss factor of 0.04 or more would be effective at reducing the resonant response of a titanium alloy beam. Two NiTiHf shape memory alloy compositions were tested to determine their loss factors at frequencies from 0.1 to 100 Hz, at temperatures from room temperature to 300 C, and at alternating strain levels of 34-35x10(exp -6). Elevated damping was demonstrated between the M(sub s) and M(sub f) phase transformation temperatures and between the A(sub s) and A(sub f) temperatures. The highest damping occurred at the lowest frequencies, with a loss factor of 0.2-0.26 at 0.1 Hz. However, the peak damping decreased with increasing frequency, and showed significant temperature hysteresis in heating and cooling. Keywords: High-temperature, shape memory alloy, damping, aircraft engine blades, NiTiHf

Description: Ceramic thermal and environmental barrier coatings (TEBCs) are used in gas turbine engines to protect engine hot-section components in the harsh combustion environments, and extend component lifetimes. Advanced TEBCs that have significantly lower thermal conductivity, better thermal stability and higher toughness than current coatings will be beneficial for future low emission and high performance propulsion engine systems. In this paper, ceramic coating design and testing considerations will be described for turbine engine high temperature and high-heat-flux applications. Thermal barrier coatings for metallic turbine airfoils and thermal/environmental barrier coatings for SiC/SiC ceramic matrix composite (CMC) components for future supersonic aircraft propulsion engines will be emphasized. Further coating capability and durability improvements for the engine hot-section component applications can be expected by utilizing advanced modeling and design tools.

Description: A pressure-gain combustor comprised of a mechanically valved, liquid fueled pulsejet, an ejector, and an enclosing shroud, was coupled to a small automotive turbocharger to form a self-aspirating, thrust producing gas turbine engine. The system was constructed in order to investigate issues associated with the interaction of pulsed combustion devices and turbomachinery. Installed instrumentation allowed for sensing of distributed low frequency pressure and temperature, high frequency pressure in the shroud, fuel flow rate, rotational speed, thrust, and laboratory noise. The engine ran successfully and reliably, achieving a sustained thrust of 5 to 6 lbf, and maintaining a rotor speed of approximately 90,000 rpm, with a combustor pressure gain of approximately 4 percent. Numerical simulations of the system without pressure-gain combustion indicated that the turbocharger would not operate. Thus, the new combustor represented a substantial improvement in system performance. Acoustic measurements in the shroud and laboratory indicated turbine stage sound pressure level attenuation of 20 dB. This is consistent with published results from detonative combustion experiments. As expected, the mechanical reed valves suffered considerable damage under the higher pressure and thermal loading characteristics of this system. This result underscores the need for development of more robust valve systems for this application. The efficiency of the turbomachinery components did not appear to be significantly affected by unsteadiness associated with pulsed combustion, though the steady component efficiencies were already low, and thus not expected to be particularly sensitive.

Description: Reliable engine-weight estimation at the conceptual design stage is critical to the development of new aircraft engines. It helps to identify the best engine concept amongst several candidates. At NASA Glenn (GRC), the Weight Analysis of Turbine Engines (WATE) computer code, originally developed by Boeing Aircraft, has been used to estimate the engine weight of various conceptual engine designs. The code, written in FORTRAN, was originally developed for NASA in 1979. Since then, substantial improvements have been made to the code to improve the weight calculations for most of the engine components. Most recently, to improve the maintainability and extensibility of WATE, the FORTRAN code has been converted into an object-oriented version. The conversion was done within the NASA s NPSS (Numerical Propulsion System Simulation) framework. This enables WATE to interact seamlessly with the thermodynamic cycle model which provides component flow data such as airflows, temperatures, and pressures, etc. that are required for sizing the components and weight calculations. The tighter integration between the NPSS and WATE would greatly enhance system-level analysis and optimization capabilities. It also would facilitate the enhancement of the WATE code for next-generation aircraft and space propulsion systems. In this paper, the architecture of the object-oriented WATE code (or WATE++) is described. Both the FORTRAN and object-oriented versions of the code are employed to compute the dimensions and weight of a 300- passenger aircraft engine (GE90 class). Both versions of the code produce essentially identical results as should be the case. Keywords: NASA, aircraft engine, weight, object-oriented

Description: The NASA Glenn Research Center (GRC) is developing a high-power-density switched-reluctance cryogenic motor for all-electric and pollution-free flight. However, cryogenic operation at higher rotational speeds markedly shortens the life of mechanical rolling element bearings. Thus, to demonstrate the practical feasibility of using this motor for future flights, a non-contact rotor-bearing system is a crucial technology to circumvent poor bearing life that ordinarily accompanies cryogenic operation. In this paper, a bearingless motor control technology for a 12-8 (12 poles in the stator and 8 poles in the rotor) switched-reluctance motor operating in liquid nitrogen (boiling point, 77 K (-196 C or -321 F)) was presented. We pushed previous disciplinary limits of electromagnetic controller technique by extending the state-of-the-art bearingless motor operating at liquid nitrogen for high-specific-power applications. The motor was levitated even in its nonlinear region of magnetic saturation, which is believed to be a world first for the motor type. Also we used only motoring coils to generate motoring torque and levitation force, which is an important feature for developing a high specific power motor.

Description: The General Aviation Propulsion (GAP) Program Turbine Engine Element focused on the development of an advanced small turbofan engine. Goals were good fuel consumption and thrust-to-weight ratio, and very low production cost. The resulting FJX-2 turbofan engine showed the potential to meet all of these goals. The development of the engine was carried through to proof of concept testing of a complete engine system. The proof of concept engine was ground tested at sea level and in altitude test chambers. A turboprop derivative was also sea-level tested.

Description: A new hypersonic inlet for a turbine-based combined-cycle (TBCC) engine has been designed. This split-flow inlet is designed to provide flow to an over-under propulsion system with turbofan and dual-mode scramjet engines for flight from takeoff to Mach 7. It utilizes a variable-geometry ramp, high-speed cowl lip rotation, and a rotating low-speed cowl that serves as a splitter to divide the flow between the low-speed turbofan and the high-speed scramjet and to isolate the turbofan at high Mach numbers. The low-speed inlet was designed for Mach 4, the maximum mode transition Mach number. Integration of the Mach 4 inlet into the Mach 7 inlet imposed significant constraints on the low-speed inlet design, including a large amount of internal compression. The inlet design was used to develop mechanical designs for two inlet mode transition test models: small-scale (IMX) and large-scale (LIMX) research models. The large-scale model is designed to facilitate multi-phase testing including inlet mode transition and inlet performance assessment, controls development, and integrated systems testing with turbofan and scramjet engines.

Description: The primary objective of this effort was to demonstrate active control of combustion instabilities in a direct-injection gas turbine combustor that accurately simulates engine operating conditions and reproduces an engine-type instability. This report documents the second phase of a two-phase effort. The first phase involved the analysis of an instability observed in a developmental aeroengine and the design of a single-nozzle test rig to replicate that phenomenon. This was successfully completed in 2001 and is documented in the Phase I report. This second phase was directed toward demonstration of active control strategies to mitigate this instability and thereby demonstrate the viability of active control for aircraft engine combustors. This involved development of high-speed actuator technology, testing and analysis of how the actuation system was integrated with the combustion system, control algorithm development, and demonstration testing in the single-nozzle test rig. A 30 percent reduction in the amplitude of the high-frequency (570 Hz) instability was achieved using actuation systems and control algorithms developed within this effort. Even larger reductions were shown with a low-frequency (270 Hz) instability. This represents a unique achievement in the development and practical demonstration of active combustion control systems for gas turbine applications.

Description: The acoustic liner system designed for use in the High Speed Civil Transport (HSCT) was tested in a thermal-acoustic environment. Five ceramic matrix composite (CMC) acoustic tile configurations, five bulk acoustic absorbers, and one thermal protection system design were tested. The CMC acoustic tiles were subjected to two 2 3/4 hr ambient temperature acoustic exposures to measure their dynamic response. One exposure was conducted on the tiles alone and the second exposure included the tiles and the T-foam bulk absorber. The measured tile RMS strains were small. With or without the T-foam absorber, the dynamic strains were below strain levels that would cause damage during fatigue loading. After the ambient exposure, a 75-hr durability test of the entire acoustic liner system was conducted using a thermal-acoustic cycle that approximated the anticipated service cycle. Acoustic loads up to 139 dB/Hz and temperatures up to 1670 F (910 C) were employed during this 60 cycle test. During the durability test, the CMC tiles were exposed to temperatures up to 1780 F and a transient through thickness gradient up to 490 F. The TPS peak temperatures on the hot side of the panels ranged from 750 to 1000 F during the 60 cycles. The through thickness delta T ranged from 450 to 650 F, varying with TPS location and cycle number. No damage, such as cracks or chipping, was observed in the CMC tiles after completion of the testing. However, on tile warped during the durability test and was replaced after 43 or 60 cycles. No externally observed damage was found in this tile. No failure of the CMC fasteners occurred, but damage was observed. Cracks and missing material occurred, only in the fastener head region. No indication of damage was observed in the T-foam acoustic absorbers. The SiC foam acoustic absorber experienced damage after about 43 cycles. Cracking in the TPS occurred around the attachment holes and under a vent. In spite of the development of damage, the TPS maintained its insulative capability throughout the durability test. The durability test results demonstrate damage-tolerant CMC tile, CMC fastener, TPS, and T-foam absorber designs for the combined thermal and acoustic engine nozzle environment.

Description: This study was motivated by a goal to understand the mixing and emissions in the Rich-burn/Quick-mix/Lean-burn (RQL) combustor scheme that has been proposed to minimize the formation of oxides of nitrogen (NOx) in gas turbine combustors. The study reported herein was a reacting jet-in-crossflow experiment at atmospheric pressure. The jets were injected from the perimeter of a cylindrical duct through round-hole orifices into a fuel-rich mainstream flow. The number of orifices investigated in this study gave over- to optimum to underpenetrating jets at a jet-to-mainstream momentum-flux ratio of J = 57. The size of individual orifices was decreased as the number of orifices increased to maintain a constant total area; the jet-to-mainstream mass-flow ratio was constant at MR = 2.5. The experiments focused on the effects of the number of orifices and inlet air preheat and were conducted in a facility that provided the capability for independent variation of jet and main inlet air preheat temperature. The number of orifices was found to have a significant effect on mixing and the distributions of species, but very little effect on overall NOx emissions, suggesting that an aerodynamically optimum mixer might not minimize NOx emissions. Air preheat was found to have very little effect on mixing and the distributions of major species, but preheating both main and jet air did increase NOx emissions significantly. Although the air jets injected in the quick-mix section of an RQL combustor may comprise over 70 percent of the total air flow, the overall NOx emission levels were found to be more sensitive to main stream air preheat than to jet stream air preheat.

Description: The NASA John H. Glenn Research Center has a wealth of experience in Halbach array technology through the Fundamental Aeronautics Program. The goals of the program include improving aircraft efficiency, reliability, and safety. The concept of a Halbach magnetically levitated electric aircraft motor will help reduce harmful emissions, reduce the Nation s dependence on fossil fuels, increase efficiency and reliability, reduce maintenance and decrease operating noise levels. Experimental hardware systems were developed in the GRC Engineering Development Division to validate the basic principles described herein and the theoretical work that was performed. A number of Halbach Magnetic rotors have been developed and tested under this program. A separate test hardware setup was developed to characterize each of the rotors. A second hardware setup was developed to test the levitation characteristics of the rotors. Each system focused around a unique Halbach array rotor. Each rotor required original design and fabrication techniques. A 4 in. diameter rotor was developed to test the radial levitation effects for use as a magnetic bearing. To show scalability from the 4 in. rotor, a 1 in. rotor was developed to also test radial levitation effects. The next rotor to be developed was 20 in. in diameter again to show scalability from the 4 in. rotor. An axial rotor was developed to determine the force that could be generated to position the rotor axially while it is rotating. With both radial and axial magnetic bearings, the rotor would be completely suspended magnetically. The purpose of this report is to document the development of a series of Halbach magnetic rotors to be used in testing. The design, fabrication and assembly of the rotors will be discussed as well as the hardware developed to test the rotors.

Description: A key technological concept for producing reliable engine diagnostics and prognostics exploits the benefits of fusing sensor data, information, and/or processing algorithms. This report describes the development of a hybrid engine model for a propulsion gas turbine engine, which is the result of fusing two diverse modeling methodologies: a physics-based model approach and an empirical model approach. The report describes the process and methods involved in deriving and implementing a hybrid model configuration for a commercial turbofan engine. Among the intended uses for such a model is to enable real-time, on-board tracking of engine module performance changes and engine parameter synthesis for fault detection and accommodation.

Description: The purpose of this engine feasibility study was to determine the benefits that can be achieved by incorporating positive displacement axial vane compression and expansion stages into high bypass turbofan engines. These positive-displacement stages would replace some or all of the conventional compressor and turbine stages in the turbine engine, but not the fan. The study considered combustion occurring internal to an axial vane component (i.e., Diesel engine replacing the standard turbine engine combustor, burner, and turbine); and external continuous flow combustion with an axial vane compressor and an axial vane turbine replacing conventional compressor and turbine systems.

Description: This paper at first describes the fluid network approach recently implemented into the National Combustion Code (NCC) for the simulation of transport of aerosols (volatile particles and soot) in the particulate sampling systems. This network-based approach complements the other two approaches already in the NCC, namely, the lower-order temporal approach and the CFD-based approach. The accuracy and the computational costs of these three approaches are then investigated in terms of their application to the prediction of particle losses through sample transmission and distribution lines. Their predictive capabilities are assessed by comparing the computed results with the experimental data. The present work will help establish standard methodologies for measuring the size and concentration of particles in high-temperature, high-velocity jet engine exhaust. Furthermore, the present work also represents the first step of a long term effort of validating physics-based tools for the prediction of aircraft particulate emissions.

Description: Model predictive control is a strategy well-suited to handle the highly complex, nonlinear, uncertain, and constrained dynamics involved in aircraft engine control problems. However, it has thus far been infeasible to implement model predictive control in engine control applications, because of the combination of model complexity and the time allotted for the control update calculation. In this paper, a multiplexed implementation is proposed that dramatically reduces the computational burden of the quadratic programming optimization that must be solved online as part of the model-predictive-control algorithm. Actuator updates are calculated sequentially and cyclically in a multiplexed implementation, as opposed to the simultaneous optimization taking place in conventional model predictive control. Theoretical aspects are discussed based on a nominal model, and actual computational savings are demonstrated using a realistic commercial engine model.

Description: In several recent studies and on-going developments for advanced rotorcraft, the need for variable or multi-speed capable rotors has been raised. A speed change of up to 50 percent has been proposed for future rotorcraft to improve overall vehicle performance. Accomplishing rotor speed changes during operation requires both a rotor that can perform effectively over the operation speed/load range, and a propulsion system that can enable these speed changes. A study has been completed to investigate possible drive system arrangements that can accommodate up to the 50 percent speed change. Several concepts will be presented and evaluated. The most promising configurations will be identified and developed for future testing in a sub-scaled test facility to validate operational capability.

Description: Current collaborative research with Pratt & Whitney on Ultra High Bypass Engine Cycle noise, performance and emissions improvements as part of the Subsonic Fixed Wing Project Ultra High Bypass Engine Partnership Element is discussed. The Subsonic Fixed Wing Project goals are reviewed, as well as their relative technology level compared to previous NASA noise program goals. Progress toward achieving the Subsonic Fixed Wing Project goals over the 2008 fiscal year by the UHB Partnership in this area of research are reviewed. The current research activity in Ultra High Bypass Engine Cycle technology, specifically the Pratt & Whitney Geared Turbofan, at NASA and Pratt & Whitney are discussed including the contributions each entity bring toward the research project, and technical plans and objectives. Pratt & Whitney Geared Turbofan current and future technology and business plans are also discussed, including the role the NASA SFW UHB partnership plays toward achieving those goals.

Description: Current collaborative research with General Electric Aviation on Open Rotor propulsion as part of the Subsonic Fixed Wing Project Ultra High Bypass Engine Partnership Element is discussed. The Subsonic Fixed Wing Project goals are reviewed, as well as their relative technology level compared to previous NASA noise program goals. The current Open Rotor propulsion research activity at NASA and GE are discussed including the contributions each entity bring toward the research project, and technical plans and objectives. GE Open Rotor propulsion technology and business plans currently and toward the future are also discussed, including the role the NASA SFW UHB partnership plays toward achieving those goals.

Description: Basic algorithms for unstructured mesh generation and fluid flow calculation are discussed. In particular the following are addressed: preliminaries of graphs and meshes; duality and data structures; basic graph operations important in CFD (Computational Fluid Dynamics); triangulation methods, including Varonoi diagrams and Delaunay triangulation; maximum principle analysis; finite volume schemes for scalar conservation law equations; finite volume schemes for the Euler and Navier-Stokes equations; and convergence acceleration for steady state calculations.

Description: It is shown that to satisfy the general accepted compressible law of the wall derived from the Van Driest transformation, turbulence modeling coefficients must actually be functions of density gradients. The transformed velocity profiles obtained by using standard turbulence model constants have too small a value of the effective von Karman constant kappa in the log-law region (inner layer). Thus, if the model is otherwise accurate, the wake component is overpredicted and the predicted skin friction is lower than the expected value.

Description: The multigrid method has been applied to an existing three-dimensional compressible Euler solver to accelerate the convergence of the implicit symmetric relaxation scheme. This lower-upper symmetric Gauss-Seidel implicit scheme is shown to be an effective multigrid driver in three dimensions. A grid refinement study is performed including the effects of large cell aspect ratio meshes. Performance figures of the present multigrid code on Cray computers including the new C90 are presented. A reduction of three orders of magnitude in the residual for a three-dimensional transonic inviscid flow using 920 k grid points is obtained in less than 4 min on a Cray C90.

Description: The present paper explores the use of large-eddy simulations as a tool for predicting noise from first principles. A high-order numerical scheme is used to perform large-eddy simulations of a supersonic jet flow with emphasis on capturing the time-dependent flow structure representating the sound source. The wavelike nature of this structure under random inflow disturbances is demonstrated. This wavelike structure is then enhanced by taking the inflow disturbances to be purely harmonic. Application of Lighthill's theory to calculate the far-field noise, with the sound source obtained from the calculated time-dependent near field, is demonstrated. Alternative approaches to coupling the near-field sound source to the far-field sound are discussed.

Description: The steady state solution of the system of equations consisting of the full Navier-Stokes equations and two turbulence equations has been obtained using a multigrid strategy of unstructured meshes. The flow equations and turbulence equations are solved in a loosely coupled manner. The flow equations are advanced in time using a multistage Runge-Kutta time-stepping scheme with a stability-bound local time step, while turbulence equations are advanced in a point-implicit scheme with a time step which guarantees stability and positivity. Low-Reynolds-number modifications to the original two-equation model are incorporated in a manner which results in well-behaved equations for arbitrarily small wall distances. A variety of aerodynamic flows are solved, initializing all quantities with uniform freestream values. Rapid and uniform convergence rates for the flow and turbulence equations are observed.

Description: Numerical results obtained with direct simulation Monte Carlo and Navier-Stokes methods are presented for a Mach-20 nitrogen flow about a 70-deg blunted cone. The flow conditions simuulated are those that can be obtained in existing low-density hypersonic wind tunnels. Three sets of flow conditions are considered with freestream Knudsen numbers ranging from 0.03 to 0.001. The focus is on the wake structure: how the wake structure changes as a function of rarefaction, what the afterbody levels of heating are, and to what limits the continuum models are realistic as rarefaction in the wake is progressively increased. Calculations are made with and without an afterbody sting. Results for the after body sting are emphasizes in anticipation of an experimental study for the current flow conditions and model configuration. The Navier-Stokes calculations were made with and without slip boundary conditions. Comparisons of the results obtained with the two simulation methodologies are made for both flowfield structure and surface quantities.

Description: The transformation validity question utilizing resulting data from direct numerical simulations (DNS) of supersonic, isothermal cold wall channel flow was investigated. The DNS results stood for a wide scope of parameter and were suitable for the purpose of examining the generality of Van Driest transformation. The Van Driest law of the wall can be obtained from the inner-layer similarity arguments. It was demonstrated that the Van Driest transformation cannot be incorporated to collapse the sublayer and log-layer velocity profiles simultaneously. Velocity and temperature predictions according to the preceding composite mixing-length model were presented. Despite satisfactory congruity with the DNS data, the model must be perceived as an engineering guide and not as a rigorous analysis.

Description: A Monte Carlo solution technique has been formulated to predict the radiative heat transfer in three-dimensional, inhomogeneous participating media which exhibit spectrally dependent emission and absorption and anisotropic scattering. Details of the technique and selected numerical sensitivities are discussed. The technique was applied to a problem involving a medium composed of a gas mixture of carbon dioxide and nitrogen and suspended carbon particles. A homogeneous medium was modeled to examine the effect of total pressure and carbon-particle concentration on radiative heat transfer. Variation in total pressure, over the range studied, had minimal effect on the amount of heat radiated to the enclosure walls and on the radiative-flux distribution within the medium. Increases in the carbon particle concentration produced significantly higher heat fluxes at the boundaries and altered the radiative flux distribution. The technique was then applied to an inhomogeneous medium to examine effects of specific temperature and carbon particle concentration distributions on radiative heat transfer. For the inhomogeneous conditions examined, the largest radiative flux divergence occurs near the center of the medium and the regions near some enclosure walls act as energy sinks.

Description: Numerical results obtained with direct simulation Monte Carlo and Navier-Stokes methods are presented for a Mach-20 nitrogen flow about a 70-deg blunted cone. The flow conditions simulated are those that can be obtained in existing low-density hypersonic wind tunnels. Three sets of flow conditions are considered with freestream Knudsen numbers ranging from 0.03 to 0.001. The focus is on the wake structure: how the wake structure changes as a function of rare faction, what the afterbody levels of heating are, and to what limits the continuum models are realistic as rarefunction in the wake is progressively increased. Calculations are made with and without an afterbody sting. Results for the afterbody sting are emphasized in anticipation of an experimental study for the current flow conditions and model configuration. The Navier-Stokes calculations were made with and without slip boundary conditions. Comparisons of the results obtained with the two simulation methodologies are made for both flowfield structure and surface quantities.

Description: The Simplified Shuttle Payload Thermal Analyzer program (SSPTA) was developed to aid in the evaluation of thermal design concepts of instruments to be flown in the Space Shuttle cargo bay. SSPTA consists of a collection of programs that are currently used in the thermal analysis of spacecraft and have been modified for quick, preliminary analysis of payloads. SSPTA includes a reduced math model of the Shuttle cargo bay to simplify use of the program for payload analysis. One of the prime objectives in developing SSPTA was to create a program which was easy to use. With SSPTA, the user required input is simple and the user is free from many of the concerns of computer usage such as disk space handling, tape usage, and complicated program control. Although SSPTA was designed primarily to analyze Shuttle payloads, it can easily be used to perform thermal analysis in other situations. SSPTA is comprised of a system of data files called 'bins', a master program, and a set of thermal subprograms. The bin system is a collection of disk files which contain data required by or computed by the thermal subprograms. SSPTA currently has the capability of handling 50 bins. The master program serves primarily as a manager for the bin system and its interaction with the thermal subprograms. Input to the master program consists of simple user commands which direct the data manipulation procedures, prepare the data for these procedures, and call the appropriate thermal subprograms. The subprograms of SSPTA are all based on programs which have been used extensively in the analysis of orbiting spacecraft and space hardware. Subprogram CONSHAD uses the user supplied geometric radiation model to compute black body view factors, shadow factors, and a description of the surface model. The subprogram WORKSHEET uses the surface model description, optical property data, and node assignment data to prepare input for SCRIPTF. Subprogram SCRIPTF computes the inverses of the infrared (IR) and ultraviolet (UV) radiation transfer equations; it also computes the radiation coupling between nodes in the thermal model. Subprogram ORBITAL uses the shadow tables to compute incident flux intensities on each surface in the geometric model. Subprogram ABSORB uses these flux intensities combined with the IR and UV inverses to compute the IR and UV fluxes absorbed by each surface. The radiation couplings from SCRIPTF and the absorbed fluxes from ABSORB are used by subprogram TTA to compute the temperature and power balance for each node in the thermal model. Output consists of tabulated data from each of the subprograms executed during a particular analysis. Due to the modular form of SSPTA, analyses may be run in whole or in part, and new subprograms may be added by the user. SSPTA is written in FORTRAN for use on a DEC VAX-11/780. SSPTA was originally developed in 1977 for use on IBM 370 series computers. This version is an update which was ported to the VAX in 1980.

Description: Remote Interactive Particle-tracing (RIP) is a distributed-graphics program which computes particle traces for computational fluid dynamics (CFD) solution data sets. A particle trace is a line which shows the path a massless particle in a fluid will take; it is a visual image of where the fluid is going. The program is able to compute and display particle traces at a speed of about one trace per second because it runs on two machines concurrently. The data used by the program is contained in two files. The solution file contains data on density, momentum and energy quantities of a flow field at discrete points in three-dimensional space, while the grid file contains the physical coordinates of each of the discrete points. RIP requires two computers. A local graphics workstation interfaces with the user for program control and graphics manipulation, and a remote machine interfaces with the solution data set and performs time-intensive computations. The program utilizes two machines in a distributed mode for two reasons. First, the data to be used by the program is usually generated on the supercomputer. RIP avoids having to convert and transfer the data, eliminating any memory limitations of the local machine. Second, as computing the particle traces can be computationally expensive, RIP utilizes the power of the supercomputer for this task. Although the remote site code was developed on a CRAY, it is possible to port this to any supercomputer class machine with a UNIX-like operating system. Integration of a velocity field from a starting physical location produces the particle trace. The remote machine computes the particle traces using the particle-tracing subroutines from PLOT3D/AMES, a CFD post-processing graphics program available from COSMIC (ARC-12779). These routines use a second-order predictor-corrector method to integrate the velocity field. Then the remote program sends graphics tokens to the local machine via a remote-graphics library. The local machine interprets the graphics tokens and draws the particle traces. The program is menu driven. RIP is implemented on the silicon graphics IRIS 3000 (local workstation) with an IRIX operating system and on the CRAY2 (remote station) with a UNICOS 1.0 or 2.0 operating system. The IRIS 4D can be used in place of the IRIS 3000. The program is written in C (67%) and FORTRAN 77 (43%) and has an IRIS memory requirement of 4 MB. The remote and local stations must use the same user ID. PLOT3D/AMES unformatted data sets are required for the remote machine. The program was developed in 1988.

Description: This software package includes two programs, the KPD12 and the KPD12P. Both programs utilizes the vortex-blob method to simulate flow around solid bodies, in an unbounded domain using the KPD12, with periodicity in one direction using the KPD12P. The main advantage of the vortex-blob method is the ability to handle situations involving arbitrary shapes including multiple bodies. The user just supplies points on the solid boundaries; there is no grid. The KPD12 program has worked successfully on bluff bodies, stalled wings, and multiple-element airfoils. The KPD12P program has been used successfully on high-solidity separated cascades and on cases of rotating stall in cascades of thin airfoils. However, they do not capture subtle viscous effects such as incipient separation and friction drag. The KPD12 and the KPD12P programs apply the vortex-blob method to time-dependent, high-Reynolds-number flows around solid bodies. Both programs solve the two-dimensional incompressible Navier-Stokes equations, neglecting the viscous effects away from the walls. By creating new vortices along the wall at every time step, they treat the no-penetration and no-slip boundary conditions while using an influence matrix. The code automatically controls the number of vortices. Furthermore, the code has the option of treating the boundary layers by simple integral methods to determine the separation points. The KPD12 outputs forces, moments, and pressure distributions on the bodies. The KPD12P also outputs the turning angle and loss of total pressure. The source code is in Cray FORTRAN and contains a few calls to Cray vector functions which are vectorized with the Cray compiler. However, substitutes for these vector functions are provided. The code is set up to plot the bodies, vortex positions, and streamlines using the DISSPLA graphics software. The software requires a mainframe computer with at least 589k of memory available running under COS 1.16. KPD12 was developed in 1988.

Description: The Steady State Thermal Analysis Program (STEADY) provides the thermal designer with a quick and convenient method for calculating heat loads and temperatures. STEADY can be used on small nodal networks for conceptual or preliminary thermal design and analysis. STEADY will accept up to 20 nodes of fixed or variable temperature, with constant or temperature-dependent thermal conductivities, and any set of consistent units. In a steady state thermal network, the heat balance on each variable temperature node must sum to zero. The general heat transfer equations are solved with a Newton-Raphson technique and refined by a fourth order quartic solution. Input data includes the number of nodes, number of boundary nodes, the fixed temperatures at all boundary nodes, initial temperature guesses for variable nodes, impressed heat loads, conduction and radiation coefficients, and control parameters such as convergence criteria, maximum iterations, and damping factors. The output is stored in a print file and tabulates final temperatures and heat flows for all nodes. STEADY is menu driven and allows the user to save files for future modification. STEADY is written in FORTRAN 77 (Ryan McFarland's RMFORTRAN) for interactive execution and has been implemented on the IBM PC computer series under DOS with a central memory requirement of approximately 92K of 8 bit bytes using a math coprocessor, and 103K bytes without the coprocessor. This program was developed in 1987.

Description: The Thermal Radiation Analyzer System, TRASYS, is a computer software system with generalized capability to solve the radiation related aspects of thermal analysis problems. TRASYS computes the total thermal radiation environment for a spacecraft in orbit. The software calculates internode radiation interchange data as well as incident and absorbed heat rate data originating from environmental radiant heat sources. TRASYS provides data of both types in a format directly usable by such thermal analyzer programs as SINDA/FLUINT (available from COSMIC, program number MSC-21528). One primary feature of TRASYS is that it allows users to write their own driver programs to organize and direct the preprocessor and processor library routines in solving specific thermal radiation problems. The preprocessor first reads and converts the user's geometry input data into the form used by the processor library routines. Then, the preprocessor accepts the user's driving logic, written in the TRASYS modified FORTRAN language. In many cases, the user has a choice of routines to solve a given problem. Users may also provide their own routines where desirable. In particular, the user may write output routines to provide for an interface between TRASYS and any thermal analyzer program using the R-C network concept. Input to the TRASYS program consists of Options and Edit data, Model data, and Logic Flow and Operations data. Options and Edit data provide for basic program control and user edit capability. The Model data describe the problem in terms of geometry and other properties. This information includes surface geometry data, documentation data, nodal data, block coordinate system data, form factor data, and flux data. Logic Flow and Operations data house the user's driver logic, including the sequence of subroutine calls and the subroutine library. Output from TRASYS consists of two basic types of data: internode radiation interchange data, and incident and absorbed heat rate data. The flexible structure of TRASYS allows considerable freedom in the definition and choice of solution method for a thermal radiation problem. The program's flexible structure has also allowed TRASYS to retain the same basic input structure as the authors update it in order to keep up with changing requirements. Among its other important features are the following: 1) up to 3200 node problem size capability with shadowing by intervening opaque or semi-transparent surfaces; 2) choice of diffuse, specular, or diffuse/specular radiant interchange solutions; 3) a restart capability that minimizes recomputing; 4) macroinstructions that automatically provide the executive logic for orbit generation that optimizes the use of previously completed computations; 5) a time variable geometry package that provides automatic pointing of the various parts of an articulated spacecraft and an automatic look-back feature that eliminates redundant form factor calculations; 6) capability to specify submodel names to identify sets of surfaces or components as an entity; and 7) subroutines to perform functions which save and recall the internodal and/or space form factors in subsequent steps for nodes with fixed geometry during a variable geometry run. There are two machine versions of TRASYS v27: a DEC VAX version and a Cray UNICOS version. Both versions require installation of the NASADIG library (MSC-21801 for DEC VAX or COS-10049 for CRAY), which is available from COSMIC either separately or bundled with TRASYS. The NASADIG (NASA Device Independent Graphics Library) plot package provides a pictorial representation of input geometry, orbital/orientation parameters, and heating rate output as a function of time. NASADIG supports Tektronix terminals. The CRAY version of TRASYS v27 is written in FORTRAN 77 for batch or interactive execution and has been implemented on CRAY X-MP and CRAY Y-MP series computers running UNICOS. The standard distribution medium for MSC-21959 (CRAY version without NASADIG) is a 1600 BPI 9-track magnetic tape in UNIX tar format. The standard distribution medium for COS-10040 (CRAY version with NASADIG) is a set of two 6250 BPI 9-track magnetic tapes in UNIX tar format. Alternate distribution media and formats are available upon request. The DEC VAX version of TRASYS v27 is written in FORTRAN 77 for batch execution (only the plotting driver program is interactive) and has been implemented on a DEC VAX 8650 computer under VMS. Since the source codes for MSC-21030 and COS-10026 are in VAX/VMS text library files and DEC Command Language files, COSMIC will only provide these programs in the following formats: MSC-21030, TRASYS (DEC VAX version without NASADIG) is available on a 1600 BPI 9-track magnetic tape in VAX BACKUP format (standard distribution medium) or in VAX BACKUP format on a TK50 tape cartridge; COS-10026, TRASYS (DEC VAX version with NASADIG), is available in VAX BACKUP format on a set of three 6250 BPI 9-track magnetic tapes (standard distribution medium) or a set of three TK50 tape cartridges in VAX BACKUP format. TRASYS was last updated in 1993.

Description: TDIGG is a fast and versatile program for generating two-dimensional computational grids for use with finite-difference flow-solvers. Both algebraic and elliptic grid generation systems are included. The method for grid generation by algebraic transformation is based on an interpolation algorithm and the elliptic grid generation is established by solving the partial differential equation (PDE). Non-uniform grid distributions are carried out using a hyperbolic tangent stretching function. For algebraic grid systems, interpolations in one direction (univariate) and two directions (bivariate) are considered. These interpolations are associated with linear or cubic Lagrangian/Hermite/Bezier polynomial functions. The algebraic grids can subsequently be smoothed using an elliptic solver. For elliptic grid systems, the PDE can be in the form of Laplace (zero forcing function) or Poisson. The forcing functions in the Poisson equation come from the boundary or the entire domain of the initial algebraic grids. A graphics interface procedure using the Silicon Graphics (GL) Library is included to allow users to visualize the grid variations at each iteration. This will allow users to interactively modify the grid to match their applications. TDIGG is written in FORTRAN 77 for Silicon Graphics IRIS series computers running IRIX. This package requires either MIT's X Window System, Version 11 Revision 4 or SGI (Motif) Window System. A sample executable is provided on the distribution medium. It requires 148K of RAM for execution. The standard distribution medium is a .25 inch streaming magnetic IRIX tape cartridge in UNIX tar format. This program was developed in 1992.

Description: The Systems Improved Numerical Fluids Analysis Code, SINFAC, consists of additional routines added to the April 1983 revision of SINDA, a general thermal analyzer program. The purpose of the additional routines is to allow for the modeling of active heat transfer loops. The modeler can simulate the steady-state and pseudo-transient operations of 16 different heat transfer loop components including radiators, evaporators, condensers, mechanical pumps, reservoirs and many types of valves and fittings. In addition, the program contains a property analysis routine that can be used to compute the thermodynamic properties of 20 different refrigerants. SINFAC can simulate the response to transient boundary conditions. SINFAC was first developed as a method for computing the steady-state performance of two phase systems. It was then modified using CNFRWD, SINDA's explicit time-integration scheme, to accommodate transient thermal models. However, SINFAC cannot simulate pressure drops due to time-dependent fluid acceleration, transient boil-out, or transient fill-up, except in the accumulator. SINFAC also requires the user to be familiar with SINDA. The solution procedure used by SINFAC is similar to that which an engineer would use to solve a system manually. The solution to a system requires the determination of all of the outlet conditions of each component such as the flow rate, pressure, and enthalpy. To obtain these values, the user first estimates the inlet conditions to the first component of the system, then computes the outlet conditions from the data supplied by the manufacturer of the first component. The user then estimates the temperature at the outlet of the third component and computes the corresponding flow resistance of the second component. With the flow resistance of the second component, the user computes the conditions down stream, namely the inlet conditions of the third. The computations follow for the rest of the system, back to the first component. On the first pass, the user finds that the calculated outlet conditions of the last component do not match the estimated inlet conditions of the first. The user then modifies the estimated inlet conditions of the first component in an attempt to match the calculated values. The user estimated values are called State Variables. The differences between the user estimated values and calculated values are called the Error Variables. The procedure systematically changes the State Variables until all of the Error Variables are less than the user-specified iteration limits. The solution procedure is referred to as SCX. It consists of two phases, the Systems phase and the Controller phase. The X is to imply experimental. SCX computes each next set of State Variables in two phases. In the first phase, SCX fixes the controller positions and modifies the other State Variables by the Newton-Raphson method. This first phase is the Systems phase. Once the Newton-Raphson method has solved the problem for the fixed controller positions, SCX next calculates new controller positions based on Newton's method while treating each sensor-controller pair independently but allowing all to change in one iteration. This phase is the Controller phase. SINFAC is available by license for a period of ten (10) years to approved licensees. The licenced program product includes the source code for the additional routines to SINDA, the SINDA object code, command procedures, sample data and supporting documentation. Additional documentation may be purchased at the price below. SINFAC was created for use on a DEC VAX under VMS. Source code is written in FORTRAN 77, requires 180k of memory, and should be fully transportable. The program was developed in 1988.

Description: The Thermal Radiation Analyzer System, TRASYS, is a computer software system with generalized capability to solve the radiation related aspects of thermal analysis problems. TRASYS computes the total thermal radiation environment for a spacecraft in orbit. The software calculates internode radiation interchange data as well as incident and absorbed heat rate data originating from environmental radiant heat sources. TRASYS provides data of both types in a format directly usable by such thermal analyzer programs as SINDA/FLUINT (available from COSMIC, program number MSC-21528). One primary feature of TRASYS is that it allows users to write their own driver programs to organize and direct the preprocessor and processor library routines in solving specific thermal radiation problems. The preprocessor first reads and converts the user's geometry input data into the form used by the processor library routines. Then, the preprocessor accepts the user's driving logic, written in the TRASYS modified FORTRAN language. In many cases, the user has a choice of routines to solve a given problem. Users may also provide their own routines where desirable. In particular, the user may write output routines to provide for an interface between TRASYS and any thermal analyzer program using the R-C network concept. Input to the TRASYS program consists of Options and Edit data, Model data, and Logic Flow and Operations data. Options and Edit data provide for basic program control and user edit capability. The Model data describe the problem in terms of geometry and other properties. This information includes surface geometry data, documentation data, nodal data, block coordinate system data, form factor data, and flux data. Logic Flow and Operations data house the user's driver logic, including the sequence of subroutine calls and the subroutine library. Output from TRASYS consists of two basic types of data: internode radiation interchange data, and incident and absorbed heat rate data. The flexible structure of TRASYS allows considerable freedom in the definition and choice of solution method for a thermal radiation problem. The program's flexible structure has also allowed TRASYS to retain the same basic input structure as the authors update it in order to keep up with changing requirements. Among its other important features are the following: 1) up to 3200 node problem size capability with shadowing by intervening opaque or semi-transparent surfaces; 2) choice of diffuse, specular, or diffuse/specular radiant interchange solutions; 3) a restart capability that minimizes recomputing; 4) macroinstructions that automatically provide the executive logic for orbit generation that optimizes the use of previously completed computations; 5) a time variable geometry package that provides automatic pointing of the various parts of an articulated spacecraft and an automatic look-back feature that eliminates redundant form factor calculations; 6) capability to specify submodel names to identify sets of surfaces or components as an entity; and 7) subroutines to perform functions which save and recall the internodal and/or space form factors in subsequent steps for nodes with fixed geometry during a variable geometry run. There are two machine versions of TRASYS v27: a DEC VAX version and a Cray UNICOS version. Both versions require installation of the NASADIG library (MSC-21801 for DEC VAX or COS-10049 for CRAY), which is available from COSMIC either separately or bundled with TRASYS. The NASADIG (NASA Device Independent Graphics Library) plot package provides a pictorial representation of input geometry, orbital/orientation parameters, and heating rate output as a function of time. NASADIG supports Tektronix terminals. The CRAY version of TRASYS v27 is written in FORTRAN 77 for batch or interactive execution and has been implemented on CRAY X-MP and CRAY Y-MP series computers running UNICOS. The standard distribution medium for MSC-21959 (CRAY version without NASADIG) is a 1600 BPI 9-track magnetic tape in UNIX tar format. The standard distribution medium for COS-10040 (CRAY version with NASADIG) is a set of two 6250 BPI 9-track magnetic tapes in UNIX tar format. Alternate distribution media and formats are available upon request. The DEC VAX version of TRASYS v27 is written in FORTRAN 77 for batch execution (only the plotting driver program is interactive) and has been implemented on a DEC VAX 8650 computer under VMS. Since the source codes for MSC-21030 and COS-10026 are in VAX/VMS text library files and DEC Command Language files, COSMIC will only provide these programs in the following formats: MSC-21030, TRASYS (DEC VAX version without NASADIG) is available on a 1600 BPI 9-track magnetic tape in VAX BACKUP format (standard distribution medium) or in VAX BACKUP format on a TK50 tape cartridge; COS-10026, TRASYS (DEC VAX version with NASADIG), is available in VAX BACKUP format on a set of three 6250 BPI 9-track magnetic tapes (standard distribution medium) or a set of three TK50 tape cartridges in VAX BACKUP format. TRASYS was last updated in 1993.

Description: INS3D computes steady-state solutions to the incompressible Navier-Stokes equations. The INS3D approach utilizes pseudo-compressibility combined with an approximate factorization scheme. This computational fluid dynamics (CFD) code has been verified on problems such as flow through a channel, flow over a backwardfacing step and flow over a circular cylinder. Three dimensional cases include flow over an ogive cylinder, flow through a rectangular duct, wind tunnel inlet flow, cylinder-wall juncture flow and flow through multiple posts mounted between two plates. INS3D uses a pseudo-compressibility approach in which a time derivative of pressure is added to the continuity equation, which together with the momentum equations form a set of four equations with pressure and velocity as the dependent variables. The equations' coordinates are transformed for general three dimensional applications. The equations are advanced in time by the implicit, non-iterative, approximately-factored, finite-difference scheme of Beam and Warming. The numerical stability of the scheme depends on the use of higher-order smoothing terms to damp out higher-frequency oscillations caused by second-order central differencing. The artificial compressibility introduces pressure (sound) waves of finite speed (whereas the speed of sound would be infinite in an incompressible fluid). As the solution converges, these pressure waves die out, causing the derivation of pressure with respect to time to approach zero. Thus, continuity is satisfied for the incompressible fluid in the steady state. Computational efficiency is achieved using a diagonal algorithm. A block tri-diagonal option is also available. When a steady-state solution is reached, the modified continuity equation will satisfy the divergence-free velocity field condition. INS3D is capable of handling several different types of boundaries encountered in numerical simulations, including solid-surface, inflow and outflow, and far-field boundaries. Three machine versions of INS3D are available. INS3D for the CRAY is written in CRAY FORTRAN for execution on a CRAY X-MP under COS, INS3D for the IBM is written in FORTRAN 77 for execution on an IBM 3090 under the VM or MVS operating system, and INS3D for DEC RISC-based systems is written in RISC FORTRAN for execution on a DEC workstation running RISC ULTRIX 3.1 or later. The CRAY version has a central memory requirement of 730279 words. The central memory requirement for the IBM is 150Mb. The memory requirement for the DEC RISC ULTRIX version is 3Mb of main memory. INS3D was developed in 1987. The port to the IBM was done in 1990. The port to the DECstation 3100 was done in 1991. CRAY is a registered trademark of Cray Research Inc. IBM is a registered trademark of International Business Machines. DEC, DECstation, and ULTRIX are trademarks of the Digital Equipment Corporation.

Description: SAGE, Self Adaptive Grid codE, is a flexible tool for adapting and restructuring both 2D and 3D grids. Solution-adaptive grid methods are useful tools for efficient and accurate flow predictions. In supersonic and hypersonic flows, strong gradient regions such as shocks, contact discontinuities, shear layers, etc., require careful distribution of grid points to minimize grid error and produce accurate flow-field predictions. SAGE helps the user obtain more accurate solutions by intelligently redistributing (i.e. adapting) the original grid points based on an initial or interim flow-field solution. The user then computes a new solution using the adapted grid as input to the flow solver. The adaptive-grid methodology poses the problem in an algebraic, unidirectional manner for multi-dimensional adaptations. The procedure is analogous to applying tension and torsion spring forces proportional to the local flow gradient at every grid point and finding the equilibrium position of the resulting system of grid points. The multi-dimensional problem of grid adaption is split into a series of one-dimensional problems along the computational coordinate lines. The reduced one dimensional problem then requires a tridiagonal solver to find the location of grid points along a coordinate line. Multi-directional adaption is achieved by the sequential application of the method in each coordinate direction. The tension forces direct the redistribution of points to the strong gradient region. To maintain smoothness and a measure of orthogonality of grid lines, torsional forces are introduced that relate information between the family of lines adjacent to one another. The smoothness and orthogonality constraints are direction-dependent, since they relate only the coordinate lines that are being adapted to the neighboring lines that have already been adapted. Therefore the solutions are non-unique and depend on the order and direction of adaption. Non-uniqueness of the adapted grid is acceptable since it makes possible an overall and local error reduction through grid redistribution. SAGE includes the ability to modify the adaption techniques in boundary regions, which substantially improves the flexibility of the adaptive scheme. The vectorial approach used in the analysis also provides flexibility. The user has complete choice of adaption direction and order of sequential adaptions without concern for the computational data structure. Multiple passes are available with no restraint on stepping directions; for each adaptive pass the user can choose a completely new set of adaptive parameters. This facility, combined with the capability of edge boundary control, enables the code to individually adapt multi-dimensional multiple grids. Zonal grids can be adapted while maintaining continuity along the common boundaries. For patched grids, the multiple-pass capability enables complete adaption. SAGE is written in FORTRAN 77 and is intended to be machine independent; however, it requires a FORTRAN compiler which supports NAMELIST input. It has been successfully implemented on Sun series computers, SGI IRIS's, DEC MicroVAX computers, HP series computers, the Cray YMP, and IBM PC compatibles. Source code is provided, but no sample input and output files are provided. The code reads three datafiles: one that contains the initial grid coordinates (x,y,z), one that contains corresponding flow-field variables, and one that contains the user control parameters. It is assumed that the first two datasets are formatted as defined in the plotting software package PLOT3D. Several machine versions of PLOT3D are available from COSMIC. The amount of main memory is dependent on the size of the matrix. The standard distribution medium for SAGE is a 5.25 inch 360K MS-DOS format diskette. It is also available on a .25 inch streaming magnetic tape cartridge in UNIX tar format or on a 9-track 1600 BPI ASCII CARD IMAGE format magnetic tape. SAGE was developed in 1989, first released as a 2D version in 1991 and updated to 3D in 1993.

Description: INS3D computes steady-state solutions to the incompressible Navier-Stokes equations. The INS3D approach utilizes pseudo-compressibility combined with an approximate factorization scheme. This computational fluid dynamics (CFD) code has been verified on problems such as flow through a channel, flow over a backwardfacing step and flow over a circular cylinder. Three dimensional cases include flow over an ogive cylinder, flow through a rectangular duct, wind tunnel inlet flow, cylinder-wall juncture flow and flow through multiple posts mounted between two plates. INS3D uses a pseudo-compressibility approach in which a time derivative of pressure is added to the continuity equation, which together with the momentum equations form a set of four equations with pressure and velocity as the dependent variables. The equations' coordinates are transformed for general three dimensional applications. The equations are advanced in time by the implicit, non-iterative, approximately-factored, finite-difference scheme of Beam and Warming. The numerical stability of the scheme depends on the use of higher-order smoothing terms to damp out higher-frequency oscillations caused by second-order central differencing. The artificial compressibility introduces pressure (sound) waves of finite speed (whereas the speed of sound would be infinite in an incompressible fluid). As the solution converges, these pressure waves die out, causing the derivation of pressure with respect to time to approach zero. Thus, continuity is satisfied for the incompressible fluid in the steady state. Computational efficiency is achieved using a diagonal algorithm. A block tri-diagonal option is also available. When a steady-state solution is reached, the modified continuity equation will satisfy the divergence-free velocity field condition. INS3D is capable of handling several different types of boundaries encountered in numerical simulations, including solid-surface, inflow and outflow, and far-field boundaries. Three machine versions of INS3D are available. INS3D for the CRAY is written in CRAY FORTRAN for execution on a CRAY X-MP under COS, INS3D for the IBM is written in FORTRAN 77 for execution on an IBM 3090 under the VM or MVS operating system, and INS3D for DEC RISC-based systems is written in RISC FORTRAN for execution on a DEC workstation running RISC ULTRIX 3.1 or later. The CRAY version has a central memory requirement of 730279 words. The central memory requirement for the IBM is 150Mb. The memory requirement for the DEC RISC ULTRIX version is 3Mb of main memory. INS3D was developed in 1987. The port to the IBM was done in 1990. The port to the DECstation 3100 was done in 1991. CRAY is a registered trademark of Cray Research Inc. IBM is a registered trademark of International Business Machines. DEC, DECstation, and ULTRIX are trademarks of the Digital Equipment Corporation.

Description: The ability to treat arbitrary boundary shapes is one of the most desirable characteristics of a method for generating grids, including those about airfoils. In a grid used for computing aerodynamic flow over an airfoil, or any other body shape, the surface of the body is usually treated as an inner boundary and often cannot be easily represented as an analytic function. The GRAPE computer program was developed to incorporate a method for generating two-dimensional finite-difference grids about airfoils and other shapes by the use of the Poisson differential equation. GRAPE can be used with any boundary shape, even one specified by tabulated points and including a limited number of sharp corners. The GRAPE program has been developed to be numerically stable and computationally fast. GRAPE can provide the aerodynamic analyst with an efficient and consistent means of grid generation. The GRAPE procedure generates a grid between an inner and an outer boundary by utilizing an iterative procedure to solve the Poisson differential equation subject to geometrical restraints. In this method, the inhomogeneous terms of the equation are automatically chosen such that two important effects are imposed on the grid. The first effect is control of the spacing between mesh points along mesh lines intersecting the boundaries. The second effect is control of the angles with which mesh lines intersect the boundaries. Along with the iterative solution to Poisson's equation, a technique of coarse-fine sequencing is employed to accelerate numerical convergence. GRAPE program control cards and input data are entered via the NAMELIST feature. Each variable has a default value such that user supplied data is kept to a minimum. Basic input data consists of the boundary specification, mesh point spacings on the boundaries, and mesh line angles at the boundaries. Output consists of a dataset containing the grid data and, if requested, a plot of the generated mesh. The GRAPE 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 135K (octal) of 60 bit words. For plotted output the commercially available DISSPLA graphics software package is required. The GRAPE program was developed in 1980.

Description: The Thermal Radiation Analyzer System, TRASYS, is a computer software system with generalized capability to solve the radiation related aspects of thermal analysis problems. TRASYS computes the total thermal radiation environment for a spacecraft in orbit. The software calculates internode radiation interchange data as well as incident and absorbed heat rate data originating from environmental radiant heat sources. TRASYS provides data of both types in a format directly usable by such thermal analyzer programs as SINDA/FLUINT (available from COSMIC, program number MSC-21528). One primary feature of TRASYS is that it allows users to write their own driver programs to organize and direct the preprocessor and processor library routines in solving specific thermal radiation problems. The preprocessor first reads and converts the user's geometry input data into the form used by the processor library routines. Then, the preprocessor accepts the user's driving logic, written in the TRASYS modified FORTRAN language. In many cases, the user has a choice of routines to solve a given problem. Users may also provide their own routines where desirable. In particular, the user may write output routines to provide for an interface between TRASYS and any thermal analyzer program using the R-C network concept. Input to the TRASYS program consists of Options and Edit data, Model data, and Logic Flow and Operations data. Options and Edit data provide for basic program control and user edit capability. The Model data describe the problem in terms of geometry and other properties. This information includes surface geometry data, documentation data, nodal data, block coordinate system data, form factor data, and flux data. Logic Flow and Operations data house the user's driver logic, including the sequence of subroutine calls and the subroutine library. Output from TRASYS consists of two basic types of data: internode radiation interchange data, and incident and absorbed heat rate data. The flexible structure of TRASYS allows considerable freedom in the definition and choice of solution method for a thermal radiation problem. The program's flexible structure has also allowed TRASYS to retain the same basic input structure as the authors update it in order to keep up with changing requirements. Among its other important features are the following: 1) up to 3200 node problem size capability with shadowing by intervening opaque or semi-transparent surfaces; 2) choice of diffuse, specular, or diffuse/specular radiant interchange solutions; 3) a restart capability that minimizes recomputing; 4) macroinstructions that automatically provide the executive logic for orbit generation that optimizes the use of previously completed computations; 5) a time variable geometry package that provides automatic pointing of the various parts of an articulated spacecraft and an automatic look-back feature that eliminates redundant form factor calculations; 6) capability to specify submodel names to identify sets of surfaces or components as an entity; and 7) subroutines to perform functions which save and recall the internodal and/or space form factors in subsequent steps for nodes with fixed geometry during a variable geometry run. There are two machine versions of TRASYS v27: a DEC VAX version and a Cray UNICOS version. Both versions require installation of the NASADIG library (MSC-21801 for DEC VAX or COS-10049 for CRAY), which is available from COSMIC either separately or bundled with TRASYS. The NASADIG (NASA Device Independent Graphics Library) plot package provides a pictorial representation of input geometry, orbital/orientation parameters, and heating rate output as a function of time. NASADIG supports Tektronix terminals. The CRAY version of TRASYS v27 is written in FORTRAN 77 for batch or interactive execution and has been implemented on CRAY X-MP and CRAY Y-MP series computers running UNICOS. The standard distribution medium for MSC-21959 (CRAY version without NASADIG) is a 1600 BPI 9-track magnetic tape in UNIX tar format. The standard distribution medium for COS-10040 (CRAY version with NASADIG) is a set of two 6250 BPI 9-track magnetic tapes in UNIX tar format. Alternate distribution media and formats are available upon request. The DEC VAX version of TRASYS v27 is written in FORTRAN 77 for batch execution (only the plotting driver program is interactive) and has been implemented on a DEC VAX 8650 computer under VMS. Since the source codes for MSC-21030 and COS-10026 are in VAX/VMS text library files and DEC Command Language files, COSMIC will only provide these programs in the following formats: MSC-21030, TRASYS (DEC VAX version without NASADIG) is available on a 1600 BPI 9-track magnetic tape in VAX BACKUP format (standard distribution medium) or in VAX BACKUP format on a TK50 tape cartridge; COS-10026, TRASYS (DEC VAX version with NASADIG), is available in VAX BACKUP format on a set of three 6250 BPI 9-track magnetic tapes (standard distribution medium) or a set of three TK50 tape cartridges in VAX BACKUP format. TRASYS was last updated in 1993.

Description: INS3D computes steady-state solutions to the incompressible Navier-Stokes equations. The INS3D approach utilizes pseudo-compressibility combined with an approximate factorization scheme. This computational fluid dynamics (CFD) code has been verified on problems such as flow through a channel, flow over a backwardfacing step and flow over a circular cylinder. Three dimensional cases include flow over an ogive cylinder, flow through a rectangular duct, wind tunnel inlet flow, cylinder-wall juncture flow and flow through multiple posts mounted between two plates. INS3D uses a pseudo-compressibility approach in which a time derivative of pressure is added to the continuity equation, which together with the momentum equations form a set of four equations with pressure and velocity as the dependent variables. The equations' coordinates are transformed for general three dimensional applications. The equations are advanced in time by the implicit, non-iterative, approximately-factored, finite-difference scheme of Beam and Warming. The numerical stability of the scheme depends on the use of higher-order smoothing terms to damp out higher-frequency oscillations caused by second-order central differencing. The artificial compressibility introduces pressure (sound) waves of finite speed (whereas the speed of sound would be infinite in an incompressible fluid). As the solution converges, these pressure waves die out, causing the derivation of pressure with respect to time to approach zero. Thus, continuity is satisfied for the incompressible fluid in the steady state. Computational efficiency is achieved using a diagonal algorithm. A block tri-diagonal option is also available. When a steady-state solution is reached, the modified continuity equation will satisfy the divergence-free velocity field condition. INS3D is capable of handling several different types of boundaries encountered in numerical simulations, including solid-surface, inflow and outflow, and far-field boundaries. Three machine versions of INS3D are available. INS3D for the CRAY is written in CRAY FORTRAN for execution on a CRAY X-MP under COS, INS3D for the IBM is written in FORTRAN 77 for execution on an IBM 3090 under the VM or MVS operating system, and INS3D for DEC RISC-based systems is written in RISC FORTRAN for execution on a DEC workstation running RISC ULTRIX 3.1 or later. The CRAY version has a central memory requirement of 730279 words. The central memory requirement for the IBM is 150Mb. The memory requirement for the DEC RISC ULTRIX version is 3Mb of main memory. INS3D was developed in 1987. The port to the IBM was done in 1990. The port to the DECstation 3100 was done in 1991. CRAY is a registered trademark of Cray Research Inc. IBM is a registered trademark of International Business Machines. DEC, DECstation, and ULTRIX are trademarks of the Digital Equipment Corporation.

Description: The ability to treat arbitrary boundary shapes is one of the most desirable characteristics of a method for generating grids. 3DGRAPE is designed to make computational grids in or about almost any shape. These grids are generated by the solution of Poisson's differential equations in three dimensions. The program automatically finds its own values for inhomogeneous terms which give near-orthogonality and controlled grid cell height at boundaries. Grids generated by 3DGRAPE have been applied to both viscous and inviscid aerodynamic problems, and to problems in other fluid-dynamic areas. 3DGRAPE uses zones to solve the problem of warping one cube into the physical domain in real-world computational fluid dynamics problems. In a zonal approach, a physical domain is divided into regions, each of which maps into its own computational cube. It is believed that even the most complicated physical region can be divided into zones, and since it is possible to warp a cube into each zone, a grid generator which is oriented to zones and allows communication across zonal boundaries (where appropriate) solves the problem of topological complexity. 3DGRAPE expects to read in already-distributed x,y,z coordinates on the bodies of interest, coordinates which will remain fixed during the entire grid-generation process. The 3DGRAPE code makes no attempt to fit given body shapes and redistribute points thereon. Body-fitting is a formidable problem in itself. The user must either be working with some simple analytical body shape, upon which a simple analytical distribution can be easily effected, or must have available some sophisticated stand-alone body-fitting software. 3DGRAPE does not require the user to supply the block-to-block boundaries nor the shapes of the distribution of points. 3DGRAPE will typically supply those block-to-block boundaries simply as surfaces in the elliptic grid. Thus at block-to-block boundaries the following conditions are obtained: (1) grids lines will match up as they approach the block-to-block boundary from either side, (2) grid lines will cross the boundary with no slope discontinuity, (3) the spacing of points along the line piercing the boundary will be continuous, (4) the shape of the boundary will be consistent with the surrounding grid, and (5) the distribution of points on the boundary will be reasonable in view of the surrounding grid. 3DGRAPE offers a powerful building-block approach to complex 3-D grid generation, but is a low-level tool. Users may build each face of each block as they wish, from a wide variety of resources. 3DGRAPE uses point-successive-over-relaxation (point-SOR) to solve the Poisson equations. This method is slow, although it does vectorize nicely. Any number of sophisticated graphics programs may be used on the stored output file of 3DGRAPE though it lacks interactive graphics. Versatility was a prominent consideration in developing the code. The block structure allows a great latitude in the problems it can treat. As the acronym implies, this program should be able to handle just about any physical region into which a computational cube or cubes can be warped. 3DGRAPE was written in FORTRAN 77 and should be machine independent. It was originally developed on a Cray under COS and tested on a MicroVAX 3200 under VMS 5.1.

Description: This algorithm has been developed for calculating both the quantity of compressor bleed flow required to cool a turbine and the resulting decrease in efficiency due to cooling air injected into the gas stream. Because of the trend toward higher turbine inlet temperatures, it is important to accurately predict the required cooling flow. This program is intended for use with axial flow, air-breathing jet propulsion engines with a variety of airfoil cooling configurations. The algorithm results have compared extremely well with figures given by major engine manufacturers for given bulk metal temperatures and cooling configurations. The program calculates the required cooling flow and corresponding decrease in stage efficiency for each row of airfoils throughout the turbine. These values are combined with the thermodynamic efficiency of the uncooled turbine to predict the total bleed airflow required and the altered turbine efficiency. There are ten airfoil cooling configurations and the algorithm allows a different option for each row of cooled airfoils. Materials technology is incorporated and requires the date of the first year of service for the turbine stator vane and rotor blade. The user must specify pressure, temperatures, and gas flows into the turbine. This program is written in FORTRAN IV for batch execution and has been implemented on an IBM 3080 series computer with a central memory requirement of approximately 61K of 8 bit bytes. This program was developed in 1980.

Description: A fast algorithm has been developed for accurately generating boundary-conforming, three-dimensional consecutively refined computational grids applicable to arbitrary wing-body and axial turbomachinery geometries. This algorithm has been incorporated into the GRID3O computer program. The method employed in GRID3O is based on using an analytic function to generate two-dimensional grids on a number of coaxial axisymmetric surfaces positioned between the centerbody and the outer radial boundary. These grids are of the O-type and are characterized by quasi-orthogonality, geometric periodicity, and an adequate resolution throughout the flow field. Because the built-in nonorthogonal coordinate stretching and shearing cause the grid lines leaving the blade or wing trailing-edge to end at downstream infinity, use of the generated grid simplifies the numerical treatment of three-dimensional trailing vortex sheets. The GRID3O 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 450K of 8 bit bytes. The GRID3O program was developed in 1981.

Description: As turbine-engine core operating conditions become more severe, designers must develop more effective means of cooling blades and vanes. In order to design reliable, cooled turbine blades, advanced transient thermal calculation techniques are required. The TACT1 computer program was developed to perform transient and steady-state heat-transfer and coolant-flow analyses for cooled blades, given the outside hot-gas boundary condition, the coolant inlet conditions, the geometry of the blade shell, and the cooling configuration. TACT1 can analyze turbine blades, or vanes, equipped with a central coolant-plenum insert from which coolant-air impinges on the inner surface of the blade shell. Coolant-side heat-transfer coefficients are calculated with the heat transfer mode at each station being user specified as either impingement with crossflow, forced convection channel flow, or forced convection over pin fins. A limited capability to handle film cooling is also available in the program. The TACT1 program solves for the blade temperature distribution using a transient energy equation for each node. The nodal energy balances are linearized, one-dimensional, heat-conduction equations which are applied at the wall-outer-surface node, at the junction of the cladding and the metal node, and at the wall-inner-surface node. At the mid-metal node a linear, three-dimensional, heat-conduction equation is used. Similarly, the coolant pressure distribution is determined by solving the set of transfer momentum equations for the one-dimensional flow between adjacent fluid nodes. In the coolant channel, energy and momentum equations for one-dimensional compressible flow, including friction and heat transfer, are used for the elemental channel length between two coolant nodes. The TACT1 program first obtains a steady-state solution using iterative calculations to obtain convergence of stable temperatures, pressures, coolant-flow split, and overall coolant mass balance. Transient calculations are based on the steady-state solutions obtained. Input to the TACT1 program includes a geometrical description of the blade and insert, the nodal spacing to be used, and the boundary conditions describing the outside hot-gas and the coolant-inlet conditions. The program output includes the value of nodal temperatures and pressures at each iteration. The final solution output includes the temperature at each coolant node, and the coolant flow rates and Reynolds numbers. 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 480K of 8 bit bytes. The TACT1 program was developed in 1978.

Description: This is a finite-difference program for calculating the viscous compressible boundary layer flow over either planar or axisymmetric surfaces. The flow may be initially laminar and progress through a transitional zone to a fully turbulent flow, or it may remain laminar, depending on the imposed boundary conditions, laws of viscosity, and numerical solution of the momentum and energy equations. The flow may also be forced into a turbulent flow at a chosen spot by the data input. The input may contain factors of arbitrary Reynolds number, free-stream Mach number, free stream turbulence, wall heating or cooling, longitudinal wall curvature, wall suction or blowing, and wall roughness. The solution may start from an initial Falkner-Skan similarity profile, an approximate equilibrium turbulent profile, or an initial arbitrary input profile. This program has been implemented on the IBM 7094/7044 Direct Couple System. This program is written in FORTRAN IV and was developed in 1974.

Description: A computer program has been developed for the design of sharp-edged throat supersonic nozzles where losses are accounted for by correcting the ideal nozzle geometry for boundary layer displacement thickness. The ideal nozzle is designed by the method of characteristics to produce uniform parallel flow at the nozzle exit in the smallest possible distance. Boundary-layer parameters (displacement and momentum thicknesses) are calculated for the ideal nozzle, and the final nozzle geometry is obtained by adding the displacement thickness to the ideal nozzle coordinates. The boundary layer parameters are also used to calculate the aftermixing conditions downstream of the nozzle assuming the flow mixes to a uniform state. The computer program input consists essentially of the nozzle-exit Mach number, specific-heat ratio, nozzle angle, throat half-height, nozzle subsonic section coordinates and corresponding pressure ratios, total temperature and pressure, gas constant, and initial momentum or displacement thickness. The program gas properties are set up for air; for other gases, changes are required to the program. The computer program output consists of the corrected nozzle coordinates, the principal boundary-layer parameters, and the aftermixing conditions. This program has been implemented on the IBM 7094/7044 Direct Couple System.

Description: A computer program has been developed which analyzes by means of mathematical models the temperature profiles in the contents of a filled propellant tank. In designing space vehicles using cryogenic liquid propellants, it is necessary to know how heat transferred from the tank walls and heat absorbed internally affect the temperature distribution with the tank contents. The mathematical flow model is based on results from small-scale experiments. The results showed that when a subcooled fluid is subject to both nonuniform internal heating and wall heating, two distinct temperature regions are developed. In the lower region, the fluid is thoroughly mixed and maintains a uniform temperature profile. In the upper region, a stratified layer develops, and a temperature gradient is formed from the accumulation of warm fluid from the boundary layer along the tank walls; it also indicated that the temperature profiles in the stratified layer exhibited similarity. This concept was developed primarily for internal heating caused by nuclear radiation. However, the theory and computer program are applicable for any form of internal or bulk heating. This program is written in FORTRAN IV for batch execution and has been implemented on the IBM 7094. This program was developed in 1970.

Description: This program solves the two-dimensional, compressible laminar or turbulent boundary-layer equations in an arbitrary pressure gradient. Cohen and Reshotko's method is used for the laminar boundary layer, Sasman and Cresci's method for the turbulent boundary layer, and the Schlichting-Ulrich-Granville method to predict transition. Transition may also be forced at any point by the user. Separation, if it occurs, is predicted for both laminar and turbulent flow. The user may begin values for displacement thickness and momentum thickness in either laminar or turbulent flow. This program was implemented on the IBM 7094.

Description: New research on hypersonic vehicles, such as the National Aero-Space Plane (NASP), has raised concerns about the effects of shock-wave interference on various structural components of the craft. State-of-the-art aerothermal analysis software is inadequate to predict local flow and heat flux in areas of extremely high heat transfer, such as the surface impingement of an Edney-type supersonic jet. EASI revives and updates older computational methods for calculating inviscid flow field and maximum heating from shock wave interference. The program expands these methods to solve problems involving the six shock-wave interference patterns on a two-dimensional cylindrical leading edge with an equilibrium chemically reacting gas mixture (representing, for example, the scramjet cowl of the NASP). The inclusion of gas chemistry allows for a more accurate prediction of the maximum pressure and heating loads by accounting for the effects of high temperature on the air mixture. Caloric imperfections and specie dissociation of high-temperature air cause shock-wave angles, flow deflection angles, and thermodynamic properties to differ from those calculated by a calorically perfect gas model. EASI contains pressure- and temperature-dependent thermodynamic and transport properties to determine heating rates, and uses either a calorically perfect air model or an 11-specie, 7-reaction reacting air model at equilibrium with temperatures up to 15,000 K for the inviscid flowfield calculations. EASI solves the flow field and the associated maximum surface pressure and heat flux for the six common types of shock wave interference. Depending on the type of interference, the program solves for shock-wave/boundary-layer interaction, expansion-fan/boundary-layer interaction, attaching shear layer or supersonic jet impingement. Heat flux predictions require a knowledge (from experimental data or relevant calculations) of a pertinent length scale of the interaction. Output files contain flow-field information for the various shock-wave interference patterns and their associated maximum surface pressure and heat flux predictions. EASI is written in FORTRAN 77 for a DEC VAX 8500 series computer using the VAX/VMS operating system, and requires 75K of memory. The program is available on a 9-track 1600 BPI magnetic tape in DEC VAX BACKUP format. EASI was developed in 1989. DEC, VAX, and VMS are registered trademarks of the Digital Equipment Corporation.

Description: Current investigations of the hydrogen-fueled supersonic combustion ramjet engine have delineated several technological problem areas. One area, the analysis of the injection, turbulent mixing, and combusiton of hydrogen, requires the accurate calculation of the supersonic combustion flow fields. This calculation has proven difficult because of an interesting phenomena which makes possible the transition from supersonic to subsonic flow in the combustion field, due to the temperature transitions which occur in the flow field. This computer program was developed to use viscous characteristics theory to analyze supersonic combustion flow fields with imbedded subsonic regions. Intended to be used as a practical design tool for two-dimensional and axisymmetric supersonic combustor development, this program has proven useful in the analysis of such problems as determining the flow field of a single underexpanded hydrogen jet, the internal flow of a gas sampling probe, the effects of fuel-injector strut shape, and the effects of changes in combustor configuration. Both combustion and diffusive effects can significantly alter the wave pattern in a supersonic field and generate significant pressure gradients in both the axial and radial directions. The induced pressure, in turn, substantially influences the ignition delay and reaction times as well as the velocity distribution. To accurately analyze the flow fields, the effects of finite rate chemistry, mixing, and wave propagation must be properly linked to one another. The viscous characteristics theory has been used in the past to describe flows that are purely supersonic; however, the interacting pressure effects in the combustor often allow for the development of shock waves and imbedded subsonic regions. Numerical investigation of these transonic situations has required the development of a new viscous characteristics procedure which is valid within the subsonic region and can be coupled with the standard viscous characteristics procedure in the supersonic region. The basic governing equations used are the 'viscous-inviscid' equations, similar to those employed in higher-order boundary layer analyses, with finite rate chemistry terms included. In addition, the Rankine-Hugoniot and Prandtl-Meyer relations are used to compute shock and expansion conditions. The program can handle up to 20 simultaneous shock waves. Chemistry terms are computed for a 7-species 8-mechanism hydrogen-air reaction scheme. The user input consists of a physical description of the combustor and flow determination parameters. Output includes detail flow parameter values at selected points within the flow field. This computer program is written in FORTRAN IV for batch execution and has been implemented on a CDC CYBER 175 with a central memory requirement of approximately 114K (octal) of 60 bit words. The program was developed in 1978.

Description: SINDA, the Systems Improved Numerical Differencing Analyzer, is a software system for solving lumped parameter representations of physical problems governed by diffusion-type equations. SINDA was originally designed for analyzing thermal systems represented in electrical analog, lumped parameter form, although its use may be extended to include other classes of physical systems which can be modeled in this form. As a thermal analyzer, SINDA can handle such interrelated phenomena as sublimation, diffuse radiation within enclosures, transport delay effects, and sensitivity analysis. FLUINT, the FLUid INTegrator, is an advanced one-dimensional fluid analysis program that solves arbitrary fluid flow networks. The working fluids can be single phase vapor, single phase liquid, or two phase. The SINDA'85/FLUINT system permits the mutual influences of thermal and fluid problems to be analyzed. The SINDA system consists of a programming language, a preprocessor, and a subroutine library. The SINDA language is designed for working with lumped parameter representations and finite difference solution techniques. The preprocessor accepts programs written in the SINDA language and converts them into standard FORTRAN. The SINDA library consists of a large number of FORTRAN subroutines that perform a variety of commonly needed actions. The use of these subroutines can greatly reduce the programming effort required to solve many problems. A complete run of a SINDA'85/FLUINT model is a four step process. First, the user's desired model is run through the preprocessor which writes out data files for the processor to read and translates the user's program code. Second, the translated code is compiled. The third step requires linking the user's code with the processor library. Finally, the processor is executed. SINDA'85/FLUINT program features include 20,000 nodes, 100,000 conductors, 100 thermal submodels, and 10 fluid submodels. SINDA'85/FLUINT can also model two phase flow, capillary devices, user defined fluids, gravity and acceleration body forces on a fluid, and variable volumes. SINDA'85/FLUINT offers the following numerical solution techniques. The Finite difference formulation of the explicit method is the Forward-difference explicit approximation. The formulation of the implicit method is the Crank-Nicolson approximation. The program allows simulation of non-uniform heating and facilitates modeling thin-walled heat exchangers. The ability to model non-equilibrium behavior within two-phase volumes is included. Recent improvements to the program were made in modeling real evaporator-pumps and other capillary-assist evaporators. SINDA'85/FLUINT is available by license for a period of ten (10) years to approved licensees. The licensed program product includes the source code and one copy of the supporting documentation. Additional copies of the documentation may be purchased separately at any time. SINDA'85/FLUINT is written in FORTRAN 77. Version 2.3 has been implemented on Cray series computers running UNICOS, CONVEX computers running CONVEX OS, and DEC RISC computers running ULTRIX. Binaries are included with the Cray version only. The Cray version of SINDA'85/FLUINT also contains SINGE, an additional graphics program developed at Johnson Space Flight Center. Both source and executable code are provided for SINGE. Users wishing to create their own SINGE executable will also need the NASA Device Independent Graphics Library (NASADIG, previously known as SMDDIG; UNIX version, MSC-22001). The Cray and CONVEX versions of SINDA'85/FLUINT are available on 9-track 1600 BPI UNIX tar format magnetic tapes. The CONVEX version is also available on a .25 inch streaming magnetic tape cartridge in UNIX tar format. The DEC RISC ULTRIX version is available on a TK50 magnetic tape cartridge in UNIX tar format. SINDA was developed in 1971, and first had fluid capability added in 1975. SINDA'85/FLUINT version 2.3 was released in 1990.

Description: The transient response of a thermal protection material to heat applied to the surface can be calculated using the CHAP III computer program. CHAP III can be used to analyze pyrolysis gas chemical kinetics in detail and examine pyrolysis reactions-indepth. The analysis includes the deposition of solid products produced by chemical reactions in the gas phase. CHAP III uses a modelling technique which can approximate a wide range of ablation problems. The energy equation used in CHAP III incorporates pyrolysis (both solid and gas reactions), convection, conduction, storage, work, kinetic energy, and viscous dissipation. The chemically reacting components of the solid are allowed to vary as a function of position and time. CHAP III employs a finite difference method to approximate the energy equations. Input values include specific heat, thermal conductivity, thermocouple locations, enthalpy, heating rates, and a description of the chemical reactions expected. The output tabulates the temperature at locations throughout the ablator, gas flow within the solid, density of the solid, weight of pyrolysis gases, and rate of carbon deposition. A sample case is included, which analyzes an ablator material containing several pyrolysis reactions subjected to an environment typical of entry at lunar return velocity. CHAP III is written in FORTRAN IV for batch execution and has been implemented on a CDC CYBER 170 series computer operating under NOS with a central memory requirement of approximately 102K (octal) of 60 bit words. This program was developed in 1985.

Description: This program performs a one-dimensional numerical analysis of the transient thermal response of multi-layer insulative systems. The analysis can determine the temperature distribution through a system consisting of from one to four layers, one of which can be an air gap. Concentrated heat sinks at any interface can be included. The computer program based on the analysis will determine the thickness of a specified layer that will satisfy a temperature limit criterion at any point in the insulative system. The program will also automatically calculate the thickness at several points on a system and determine the total system mass. This program was developed as a tool for designing thermal protection systems for high-speed aerospace vehicles but could be adapted to many areas of industry involved in thermal insulation systems. In this package, the equations describing the transient thermal response of a system are developed. The governing differential equation for each layer and boundary condition are put in finite-difference form using a Taylor's series expansion. These equations yield an essentially tridiagonal matrix of unknown temperatures. A procedure based on Gauss' elimination method is used to solve the matrix. This program is written in FORTRAN IV for the CDC RUN compiler and has been implemented on a CDC 6000 series machine operating under SCOPE 3.0. This program requires a minimum of 44K (octal) of 60 bit words of memory.

Description: A computer program has been developed to analyze the transient response of an ablating axisymmetric body, including the effect of shape change. The governing differential equation, the boundary conditions for the analysis on which the computer program is based, and the method of solution of the resulting finite-difference equations are discussed in the documentation. Some of the features of the analysis and the associated program are (1) the ablation material is considered to be orthotropic with temperature-dependent thermal properties; (2) the thermal response of the entire body is considered simultaneously; (3) the heat transfer and pressure distribution over the body are adjusted to the new geometry as ablation occurs; (4) the governing equations and several boundary-condition options are formulated in terms of generalized orthogonal coordinates for fixed points in a moving coordinate system; (5) the finite-difference equations are solved implicitly; and (6) other instantaneous body shapes can be displayed with a user-supplied plotting routine. The physical problem to be modeled with the analysis is described by FORTRAN input variables. For example, the external body geometry is described in the W, Z coordinates; material density is given; and the stagnation cold-wall heating rate is given in a time-dependent array. Other input variables are required which control the solution, specify boundary conditions, and determine output from the program. The equations have been programmed so that either the International System of Units or the U. S. Customary Units may be used. This program is written in FORTRAN IV for batch execution and has been implemented on a CDC 6000 Series computer. This program was developed in 1972.

Description: The GTRAN program was developed to solve transient, as well as steady state, problems for gas piping systems. GTRAN capabilities allow for the analysis of a variety of system configurations and components. These include: multiple pipe junctions; valves that change position with time; fixed restrictions (orifices, manual valves, filters, etc.); relief valves; constant pressure sources; and heat transfer for insulated piping and piping subjected to free or forced convection. In addition, boundary conditions can be incorporated to simulate specific components. The governing equations of GTRAN are the one-dimensional transient gas dynamic equations. The three equations for pressure, velocity, and density are reduced to numerical equations using an implicit Crank-Nicholson finite difference technique. Input to GTRAN includes a description of the piping network, the initial conditions, and any events (e.g. valve closings) occuring during the period of analysis. Output includes pressure, velocity, and density versus time. GTRAN is written in FORTRAN 77 for batch execution and has been implemented on a DEC VAX series computer. GTRAN was developed in 1983.

Description: Results are presented for the numerical simulation of unsteady viscous incompressible flow past thick airfoils. Specifically, flow past a NACA 4424 at an angle of attack of 2.5 deg and Reynolds numbers in the range of 1700-4000 has been simulated using the spectral element method. At these conditions the flow is separatedd and an unsteady wake is formed. Application of the method of empirical eigenfunction reveals the structure of the most energetic components of the flow. These are found to occur in pairs that, through phase exchange, are responsible for the vortex shedding. A set of ordinary differential equations is obtained for the amplitudes of these eigenfunctions by a Galerkin projection of the Navier-Stokes equations. The solutions of the model system are compared with the full simulation. The work is of relevance to the transition process and observed routes to chaos in airfoil wakes.

Description: The objective of this study is to develop a reduced mechanism for ethylene oxidation. The authors are interested in a model with a minimum number of species and reactions that still models the chemistry with reasonable accuracy for the expected combustor conditions. The model will be validated by comparing the results to those calculated with a detailed kinetic model that has been validated against the experimental data.

Description: Artificial viscosity is added either implicity or explicitly in practically every numerical scheme for suppressing spurious oscillations in the solution of fluid-dynamics equations. In the present central-difference scheme, artificial viscosity is added explicitly for suppressing high-frequency oscillations and achieving good convergence properties. The amount of artificial viscosity added is controlled through the use of preselected coefficients. In the standard scheme, scalar coefficients based on the spectral radii of the Jacobian of the convective fluxes are used. However, this can add too much viscosity to the slower waves. Hence, the use of matrix-valued coefficients, which give appropriate viscosity for each wave component, is suggested. With the matrix-valued coefficients, the central-difference scheme produces more accurate solutions on a given grid, particularly in the vicinity of shocks and boundary layers, while still maintaining good convergence properties.

Description: A potential flow based three-dimensional panel method was modified to treat time-dependent conditions in which several submerged bodies can move within the fluid along different trajectories. This modification was accomplished by formulating the momentary solution in an inertial frame of reference, attached to the undisturbed stationary fluid. Consequently, the numerical interpretation of the multiple-body, solid-surface boundary condition and the viscous wake rollup was considerably simplified. The usteady capability of this code was calibrated and validated by comparing computed results with closed-form analytical results available for an airfoil, which was impulsively set into a constant speed forward motion. To demonstrate the multicomponent capability, computations were made for two wings following closely intersecting paths (i.e., simulations aimed at avoiding mid-air collisions) and for a flowfield with relative rotation (i.e., the case of a helicopter rotor rotating relative to the fuselage). Computed results for the cases were compared to experimental data, when such data was available.

Description: An approximate method for calculating heating rates on three-dimensional vehicles at angle of attack is presented. The method is based on the axisymmetric analog for three-dimensional boundary layers and uses a generalized body-fitted coordinate system. Edge conditions for the boundary-layer solution are obtained from an inviscid flowfield solution, and because of the coordinate system used, the method is applicable to any blunt body geometry for which an inviscid flowfield solution can be obtained. The method is validated by comparing with experimental heating data and with thin-layer Navier-Stokes calculations on the shuttle orbiter at both wind-tunnel and flight conditions and with thin-layer Navier-Stokes calculations on the HL-20 at wind-tunnel conditions.

Description: This paper presents the results of a study on the variation of the critical Marangoni number (Ma(sub c)) for the onset of Benard convection in a finite liquid layer bounded horizontally as well as from below. A direct-numerical-simulation procedure is devised to determine the Ma(sub c) for aspect ratios (Ar) ranging from 0.8 to 10. The results predict a strong increase of Ma(sub c) as Ar decreases to below 2. A dip of Ma(sub c) occurs between Ar = 1.45 and 1.3, which is accompanied by a pattern transition from a two-cell convection to a unicellular flow. For Ar above 4, the calculated Ma(sub c) shows little change and asymptotically approach a value of 116.15, with Biot number (Bi) equal to 1.

Description: An experimental investigation carried out to determine aerodynamic and acoustic characteristics of a low area ratio rectangular jet ejector is reported. A supersonic primary jet issuing from a rectangular convergent-divergent nozzle of aspect ratio 4, into a rectangular duct of area ratio 3, was used. Improved performance was found when the ejector screech tone is most intense and appears to match the most unstable Strouhal number of the free rectangular jet. When the primary jet was operating at over and ideally expanded conditions, significant noise reduction was obtained with the ejector as compared to a corresponding free jet. Application of particle image velocimetry to high speed ejector flows was demonstrated through the measurement of instantaneous two dimensional velocity fields.

Description: Particle imaging velocimetry, which is rapidly becoming an essential tool for the synoptic measurement of two dimensional velocity fields, is addressed. This rapid development is carried out by a growing number of fluid mechanics experimentalists who recognize the unique capabilities of the technique to measure velocity fields in both space and time, and is supported by the concomitant development and availability of microcomputer hardware and software. Various applications of the technique to map different flow regimes are described and illustrated. Hardware implementations of the technique which utilize both conventional photography for image acquisition as well as digital 'online' methods for integrated image acquisition and processing are discussed. Recommendations to further enhance the technique and make it possible to map the three velocity components in three dimensional flow regions are given.

Description: One of the key technical elements in NASA's high speed research program is reducing the noise level to meet the federal noise regulation. The dominant noise source is associated with the supersonic jet discharged from the engine exhaust system. Whereas the turbulence mixing is largely responsible for the generation of the jet noise, a broadband shock-associated noise is also generated when the nozzle operates at conditions other than its design. For both mixing and shock noise components, because the source of the noise is embedded in the jet plume, one can expect that jet noise can be predicted from the jet flowfield computation. Mani et al. developed a unified aerodynamic/acoustic prediction scheme by applying an extension of Reichardt's aerodynamic model to compute turbulent shear stresses which are utilized in estimating the strength of the noise source. Although this method produces a fast and practical estimate of the jet noise, a modification by Khavaran et al. has led to an improvement in aerodynamic solution. The most notable feature in this work is that Reichardt's model is replaced with the computational fluid dynamics (CFD) solution of Reynolds-averaged Navier-Stokes equations. The major advantage of this work is that the essential, noise-related flow quantities such as turbulence intensity and shock strength can be better predicted. The predictions were limited to a shock-free design condition and the effect of shock structure on the jet mixing noise was not addressed. The present work is aimed at investigating this issue. Under imperfectly expanded conditions the existence of the shock cell structure and its interaction with the convecting turbulence structure may not only generate a broadband shock-associated noise but also change the turbulence structure, and thus the strength of the mixing noise source. Failure in capturing shock structures properly could lead to incorrect aeroacoustic predictions.

Description: A dimensionless group, called a pressure loss modulus (N(sub PL)), is introduced that, in conjunction with an appropriately defined Reynolds number, is of considerable engineering utility in correlating steady-state Delta p vs flow calibration data and subsequently as a predictor, using the same or a different fluid, in uniformly distributed pressure loss devices. It is particularly useful under operation in the transition regime. Applications of this simple bivariate correlation to three diverse devices of particular interest for small liquid rocket engine fluid systems are discussed: large L/D capillary tube restrictors, packed granular catalyst beds, and stacked vortex-loss disk restrictors.

Description: A numerical study of axisymmetric overexpanded nozzle is presented. The flow structure of the startup and throttle-down processes are examined. During the impulsive startup process, observed flow features include the Mach disk, separation shock, Mach stem, vortex core, contact surface, slip stream, initial shock front, and shocklet. Also the movement of the Mach disk is not monotonical in the downstream direction. For a range of pressure ratios, hysteresis phenomenon occurs; different solutions were obtained depending on different processes. Three types of flow structures were observed. The location of separation point and the lower end turning point of hysteresis are closely predicted. A high peak of pressure is associated with the nozzle flow reattachment. The reversed vortical structure and affects engine performance.

Description: Incompresible, thin sheet flows have been of research interest for many years. Those studies were mainly concerned with the stability of the flow in a surrounding gas. Squire was the first to carry out a linear, invicid stability analysis of sheet flow in air and compare the results with experiment. Dombrowski and Fraser did an experimental study of the disintegration of sheet flows using several viscous liquids. They also detected the formulation of holes in their sheet flows. Hagerty and Shea carried out an inviscid stability analysis and calculated growth rates with experimental values. They compared their calculated growth rates with experimental values. Taylor studied extensively the stability of thin liquid sheets both theoretically and experimentally. He showed that thin sheets in a vacuum are stable. Brown experimentally investigated thin liquid sheet flows as a method of application of thin films. Clark and Dumbrowski carried out second-order stability analysis for invicid sheet flows. Lin introduced viscosity into the linear stability analysis of thin sheet flows in a vacuum. Mansour and Chigier conducted an experimental study of the breakup of a sheet flow surrounded by high-speed air. Lin et al. did a linear stability analysis that included viscosity and a surrounding gas. Rangel and Sirignano carried out both a linear and nonlinear invisid stability analysis that applies for any density ratio between the sheet liquid and the surrounding gas. Now there is renewed interest in sheet flows because of their possible application as low mass radiating surfaces. The objective of this study is to investigate the fluid dynamics of sheet flows that are of interest for a space radiator system. Analytical expressions that govern the sheet geometry are compared with experimental results. Since a space radiator will operate in a vacuum, the analysis does not include any drag force on the sheet flow.

Description: The transition of an incompressible three-dimensional boundary layer with strong cross-flow is considered theoretically and computationally in the context of vortex/wave interactions. Specifically the work centers on two lower-branch Tollmien-Schlichting waves which mutually interact nonlinearly to induce a longitudinal vortex flow. The vortex motion in turn gives rise to significant wave modulation via wall-shear forcing. The characteristic Reynolds number is large and, as a consequence, the waves' and the vortex motion are governed primarily by triple deck theory. The nonlinear interaction is captured by a viscous partial-differential system for the vortex coupled with a pair of amplitude equations for each wave pressure. Following analysis and computation over a wide range of parameters, three distinct responses are found to emerge in the nonlinear behavior of the flow solution downstream: an algebraic finite-distance singularity, far-downstream saturation or far-downstream wave decay leaving pure vortex flow. These depend on the input conditions, the wave angles and the size of the cross flow.