ALBERT

Description: The effect of heat treatment on material properties of Sic/Ti-15-3 was measured by vibration tests. Heat treatment changes the microstructure, which stiffens the matrix and reduces its damping capacity. Test results illustrate how the changes in matrix material affect the stiffness and damping properties of the composite. Damping was found to be more sensitive than stiffness to microstructural changes in the matrix. Effects of heat treatment temperature and exposure time are presented.

Description: A two-dimensional model for metalorganic chemical vapor deposition of GaAs in a horizontal reactor is presented. The model is characterized by the following parameters: reactor geometry and operating pressure, thermal boundary conditions, ratio of reactants, chemical reactions, total inlet gas flow rate, as well as molecular weights, thermal conductivities, heat capacities, viscosities, and binary diffusion coefficients of the gas-phase species. Film thickness profiles predicted by the model are compared with those of GaAs thin films grown in the modeled reactor. Results obtained show a good agreement between the predictions and data over the entire length of the deposition region for the low pressure and high flow rate run. Attention is also given to the reactor design and growth conditions.

Description: We have defined an object-oriented software architecture for Intelligent Tutoring Systems (ITS's) to facilitate the rapid development, testing, and fielding of ITS's. This software architecture partitions the functionality of the ITS into a collection of software components with well-defined interfaces and execution concept. The architecture was designed to isolate advanced technology components, partition domain dependencies, take advantage of the increased availability of commercial software packages, and reduce the risks involved in acquiring ITS's. A key component of the architecture, the Executive, is a publish and subscribe message handling component that coordinates all communication between ITS components.

Description: The Hubble Space Telescope was designed to be an orbiting astronomical observatory which could be operated in the same manner as ground based observatories. The design drivers for the pointing control system's hardware and software were the requirements of an absolute pointing accuracy of 4.8E-8 radians and pointing stability (jitter) of 3.4E-8 radians. Of comparable importance was the objective of providing a flexible command methodology and structure to enable seven day operational planning employing stored program command and real time command capability. The pointing control system hardware, software, safemode control schemes, ground system monitoring capability, and in-orbit results are reviewed.

Description: The Hubble Space Telescope (HST) was designed to maintian a pointing stability (jitter) of 0.007 arc seconds rms throughout every observing period, which can last from a few seconds to several orbits. On-orbit measurements indicate that the hardware excitation induced by the reaction wheels. gyros, high gain antennae, science instrument mechanisms and tape recorders are well within specifications. Unexpectedly, the solar arrays because the dominant source of jitter. Every passage through an orbital terminator produces vibrations which emanate from the solar arrays due to thermal effects, which affect the relative positional stability. Broadband frequencies centered about 0.11 and 0.65 Hz were detected in the frequency content of the vehicle jitter. On-board modifications to the control law have attenuated the disturbance torques and reduced the vehicle jitter close to specification. Replacement of the solar arrays in December, 1993, should eliminate the torque distubances. Astrometric science observations are extremely susceptible to corruption from vehicle jitter. The removal of vehicle jitter from astrometric Transfer function scans of binary stars is explained in detail. A binary star separation of 16 milli-seconds of arc has been achieved, a separation resolution of 10 to 12 milli-seconds of arc appears feasible, with a binary star magnitude of 9 m(sub V). The achievement of this resolution is in part due to vehicle jitter removal. Comparison of vehicle jitter measurements from the position path of the vehicle control law, or from the guiding Fine Guidance Sensors (FGS), are shown to be equivalent to approximately 0.001 arc second.

Description: The low cycle fatigue behavior of polycrystalline NiAl was determined at 300 and 1000 K - temperatures below and above the brittle- to-ductile transition temperature (BDTT). Fully reversed, plastic strain-controlled fatigue tests were conducted on two differently fabricated alloy samples: hot isostatically pressed (HIP'ed) prealloyed powder and hot extruded castings. HIP'ed powder (HP) samples were tested only at 1000 K, whereas the more ductile cast-and-extruded (C+E) NiAl samples were tested at both 1000 and 300 K. Plastic strain ranges of 0.06 to 0.2 percent were used. The C+E NiAl cyclically hardened until fracture, reaching stress levels approximately 60 percent greater than the ultimate tensile strength of the alloy. Compared on a strain basis, NiAl had a much longer fatigue life than other B2 ordered compounds in which fracture initiated at processing-related defects. These defects controlled fatigue life at 300 K, with fracture occurring rapidly once a critical stress level was reached. At 1000 K, above the BDTT, both the C+E and HP samples cyclically softened during most of the fatigue tests in air and were insensitive to processing defects. The processing method did not have a major effect on fatigue life; the lives of the HP samples were about a factor of three shorter than the C+E NiAl, but this was attributed to the lower stress response of the C+E material. The C+E NiAl underwent dynamic grain growth, whereas the HP material maintained a constant grain size during testing. In both materials, fatigue life was controlled by intergranular cavitation and creep processes, which led to fatigue crack growth that was primarily intergranular in nature. Final fracture by overload was transgranular in nature. Also, HP samples tested in vacuum had a life three times longer than their counterparts tested in air and, in contrast to those tested in air, hardened continuously over half of the sample life, thereby indicating an environmentally assisted fatigue damage mechanism. The C+E samples were tested only in air. At 1000 K, NiAl exhibited a superior fatigue life when compared to most superalloys on a plastic strain basis, but was inferior to most superalloys on a stress basis.

Description: The framework of a mechanics of materials model is proposed for thermomechanical fatigue (TMF) life prediction of unidirectional, continuous-fiber metal matrix composites (MMC's). Axially loaded MMC test samples are analyzed as structural components whose fatigue lives are governed by local stress-strain conditions resulting from combined interactions of the matrix, interfacial layer, and fiber constituents. The metallic matrix is identified as the vehicle for tracking fatigue crack initiation and propagation. The proposed framework has three major elements. First, TMF flow and failure characteristics of in situ matrix material are approximated from tests of unreinforced matrix material, and matrix TMF life prediction equations are numerically calibrated. The macrocrack initiation fatigue life of the matrix material is divided into microcrack initiation and microcrack propagation phases. Second, the influencing factors created by the presence of fibers and interfaces are analyzed, characterized, and documented in equation form. Some of the influences act on the microcrack initiation portion of the matrix fatigue life, others on the microcrack propagation life, while some affect both. Influencing factors include coefficient of thermal expansion mismatch strains, residual (mean) stresses, multiaxial stress states, off-axis fibers, internal stress concentrations, multiple initiation sites, nonuniform fiber spacing, fiber debonding, interfacial layers and cracking, fractured fibers, fiber deflections of crack fronts, fiber bridging of matrix cracks, and internal oxidation along internal interfaces. Equations exist for some, but not all, of the currently identified influencing factors. The third element is the inclusion of overriding influences such as maximum tensile strain limits of brittle fibers that could cause local fractures and ensuing catastrophic failure of surrounding matrix material. Some experimental data exist for assessing the plausibility of the proposed framework.

Description: The energy density and power density of light weight aerospace batteries utilizing the nickel oxide electrode are often limited by the microstructures of both the collector and the resulting active deposit in/on the collector. Heretofore, these two microstructures were intimately linked to one another by the materials used to prepare the collector grid as well as the methods and conditions used to deposit the active material. Significant weight and performance advantages were demonstrated by Britton and Reid at NASA-LeRC using FIBREX nickel mats of ca. 28-32 microns diameter. Work in our laboratory investigated the potential performance advantages offered by nickel fiber composite electrodes containing a mixture of fibers as small as 2 microns diameter (Available from Memtec America Corporation). These electrode collectors possess in excess of an order of magnitude more surface area per gram of collector than FIBREX nickel. The increase in surface area of the collector roughly translates into an order of magnitude thinner layer of active material. Performance data and advantages of these thin layer structures are presented. Attributes and limitations of their electrode microstructure to independently control void volume, pore structure of the Ni(OH)2 deposition, and resulting electrical properties are discussed.

Description: The nonlinear stress strain behavior of 90 degree/0 degree sub 2s, SiC/Ti-15-3 composite laminate was numerically investigated with a finite element, unit cell approach. Tensile stress-strain curves from room temperature experiments depicted three distinct regions of deformation, and these regions were predicted by finite element analysis. The first region of behavior, which was linear elastic, occurred at low applied stresses. As applied stresses increased, fiber/matrix debonding in the 90 degree plies caused a break in the stress-strain curve and initiated a second linear region. In this second region, matrix plasticity in the 90 degree plies developed. The third region, which was typified by nonlinear, stress-strain behavior occr red at high stresses. In this region, the onset of matrix plasticity in the 0 degree plies stiffened the laminate in the direction transverse to the applied load. Metallographic sections confirmed the existence of matrix plasticity in specific areas of the structure. Finite element analysis also predicted these locations of matrix slip.

Description: An experimental method is described which allows for the observation of slip bands due to matrix plasticity in the SiC/Ti-15-3 composite system. A post-test heat treatment and subsequent chemical etch is employed to reveal slip bands in the titanium matrix. Composite specimens of various laminates were examined after tensile testing at room temperature. This method definitively shows that matrix plasticity has occurred in all the laminates investigated and at load/strain levels which were insufficient to cause fiber breakage.