ALBERT

Description: The following constitutes a summary of this paper: on-orbit identification methodology starts with nonparametric techniques for a priori system identification; development of the nonparametric identification and model determination experiment software has been completed; the validation experiments to be performed on the JPL Control and Identification Technology Validation Laboratory have been designed.

Description: An integrated approach to dynamic testing and mathematical model analysis is described. The overall approach addresses four key tasks, namely, pretest planning and analysis, test data acquisition, data reduction and analysis, and test/analysis correlation and mathematical model updates. Several key software programs are employed to accomplish this task. They are a leading finite element code, a sophisticated data analysis processor and a graphical pre- and post-processor along with an advanced interface utility. Several practical structures are used to illustrate tools and concepts employed in the integrated test analysis process.

Description: The paper covers two distinct parts: theory and application. The goal of this work was the reduction of model size with an increase in eigenvalue/vector accuracy. This method is ideal for the condensation of large truss- or beam-type structures. The theoretical approach involves the conversion of a continuum transfer matrix beam element into an 'Exact' dynamic stiffness element. This formulation is implemented in a finite element environment. This results in the need to solve a transcendental eigenvalue problem. Once the eigenvalue is determined the eigenvectors can be reconstructed with any desired spatial precision. No discretization limitations are imposed on the reconstruction. The results of such a combined finite element and transfer matrix formulation is a much smaller FEM eigenvalue problem. This formulation has the ability to extract higher eigenvalues as easily and as accurately as lower eigenvalues. Moreover, one can extract many more eigenvalues/vectors from the model than the number of degrees of freedom in the FEM formulation. Typically, the number of eigenvalues accurately extractable via the 'Exact' element method are at least 8 times the number of degrees of freedom. In contrast, the FEM usually extracts one accurate (within 5 percent) eigenvalue for each 3-4 degrees of freedom. The 'Exact' element results in a 20-30 improvement in the number of accurately extractable eigenvalues and eigenvectors.

Description: A solution to the correlation between structural dynamic test results and finite element analyses of the same components is presented in this paper. Basically, the method can be categorized as a Levenberg-Marquardt type Gauss-Newton method which requires only the differences between FE modal analyses and test results and their first derivatives with respect to preassigned design variables. With proper variable normalization and equation scaling, the method has been made numerically better-conditioned and the inclusion of the Levenberg-Marquardt technique overcomes any remaining difficulty encountered in inverting singular or near-singular matrices. An important feature is that each iteration requires only one function evaluation along with the associated design sensitivity analysis and so the procedure is computationally efficient.

Description: A model identification methodology for structural dynamics has been applied to simulated vibrational data as a first step in evaluating its accuracy. The evaluation has taken into account a wide variety of factors which affect the accuracy of the procedure. The effects of each of these factors were observed in both the response time histories and the estimates of the parameters of the model by comparing them with the exact values of the system. Each factor was varied independently but combinations of these have also been considered in an effort to simulate real situations. The results of the tests have shown that for the chain model, the procedure yields robust estimates of the stiffness parameters under the conditions studied whenever uniqueness is ensured. When inaccuracies occur in the results, they are intimately related to non-uniqueness conditions inherent in the inverse problem and not to shortcomings in the methodology.

Description: An elastic-plastic finite-element analysis with a critical crack-tip-opening displacement criterion was used to simulate fracture of various size compact and bend specimens made of HY-130 steel. From the calculated load-crack-extension and load-displacement curves, J-resistance (J-R) curves were determined by several methods. The simulated 3-R curves were insensitive to specimen size up to maximum load but were sensitive to specimen configuration for crack extensions greater than 10 percent of the initial uncracked ligament length.

Description: Identification of large space structures' distributed mass, stiffness, and energy dissipation characteristics poses formidable analytical, numerical, and implementation difficulties. Development of reliable on-orbit structural identification methods is important for implementing active vibration suppression concepts which are under widespread study in the large space structures community. Near the heart of the identification problem lies the necessity of making a large number of spatially distributed measurements of the structure's vibratory response and the associated force/moment inputs with sufficient spatial and frequency resolution. In the present paper, we discuss a method whereby tens of active or passive (retro-reflecting) targets on the structure are tracked simultaneously by the focal planes of two or more video cameras mounted on an adjacent platform. Triangulation (optical ray intersection) of the conjugate image centroids yield inertial trajectories of each target on the structure. Given the triangulated motion of the targets, we apply and extend methodology developed by Creamer, Junkins, and Juang to identify the frequencies, mode shapes, and updated estimates for the mass/stiffness/damping parameterization of the structure. The methodology is semi-automated, for example, the post experiment analysis of the video imagery to determine the inertial trajectories of the targets typically requires less than thirty minutes of real time. Using methodology discussed herein, the frequency response of a large number of points on the structure (where reflective targets are mounted) on the structure can be determined from optical measurements alone. For comparison purposes, we also utilize measurements from accelerometers and a calibrated impulse hammer. While our experimental work remains in a research stage of development, we have successfully tracked and stereo triangulated 20 targets (on a vibrating cantilevered grid structure) at a sample frequency of 200 HZ, and have established conclusively the feasibility and desirability of this approach. We discuss, in summary, recent advances in analog and digital video processing methodology, actuation methods, and bring them to bear on the structural identification problem. We include a brief discussion of our experimental hardware and some recent experimental results which support the practical feasibility of this structural vibration sensing approach.

Description: Limitations of the frequency domain methods in analyzing structura1 vibrations has created an awareness of the comparative merits of the time domain methods. Although time domain methods would be ideal for modeling large precisions space systems, the popular methods based on fitting theoretical response to actual data by least squares are too sensitive to noise and require too much data to be suitable for orbiting space crafts. This paper briefly reviews the theory and illustrative applications of a time domain methodology called Data Dependent Systems (DDS) that eliminates these limitations. Simulation results are presented to demonstrate a better than 4-place accuracy in the identifications of all system parameters, both modal (frequencies, damping ratios, and mode shapes) and physical (mass, stiffness, and damping matrices).

Description: The objective of the 12-meter truss modal test is to experimentally determine the frequencies, damping, and mode shapes for the first 6 modes in both principle axes and to use this information to update the FEM. These objectives will lead us to our goal of actively controlling the flexible modes of the truss. A secondary objective is to evaluate our capabilities to ground test this class of structures.

Description: Adaptive p-version based hierarchical finite element formulations in conjunction with a posteriori error estimation concepts are described with emphasis on applicability for thermal modeling/analysis of structural configurations. The basic concepts and formulations of hierarchical p-versiion finite element for thermal analysis are first described. A posteriori error estimation features are utililzed to steer the process of adaptive refinement. Several configurations comprised of one-dimensional structures are evaluated to validate the applicability of the proposed formulations and to demonstrate the potential of the p-version adaptive formulations for thermal modeling/analysis. The methodology offers potential and promises to be an attractive alternative to conventional finite element thermal modeling/analysis approaches.

Description: Different elementary clamping models are discussed for a three layer crossply laminate to study the sensitivity of clamping to the definition of cross-sectional rotation. All of these models leave a considerable residual warping at the edges. Using a complimentary energy principle and principle of superposition, an analysis is conducted to reduce this residual warping. This led to the identification of exact interior solution corresponding to the ideal clamping. This study also suggests a presence of stress singularities at the corners and between different layers near the fixed edge.

Description: This paper is concerned with the development of an efficient eigenproblem solution algorithm and an associated computer program for the economical solution of the free vibration problem of complex practical spinning structural systems. Thus, a detailed description of a newly developed block Lanczos procedure is presented in this paper that employs only real numbers in all relevant computations and also fully exploits sparsity of associated matrices. The procedure is capable of computing multiple roots and proves to be most efficient compared to other existing similar techniques.

Description: A three-dimensional virtual crack-closure technique is presented which calculates the strain energy release rates and the stress intensity factors using only nodal forces and displacements from a standard finite element analysis. The technique is an extension of the Rybicki-Kanninen (1977) method, and it assumes that any continuous function can be approximated by a finite number of straight line segments. Results obtained by the method for surface cracked plates with and without notches agree favorably with previous results.

Description: The Rayleigh-Ritz method was used to find the postbuckling static displacement pattern of a composite plane (CFRP) under uniaxial in-plane compression of uniform edge-shortening. The resonance frequencies and mode shapes at various postbuckled states are then evaluated by eigenvalue analysis of the dynamical matrix equation consisting of up-dated tangential stiffness matrix at corresponding static configuration. The theoretical results are compared with experimental results obtained in 16-layered CFRP laminate of aspect ratio 1.5. The resonance frequencies and mode shapes obtained are used to interpret the multimodal and nonlinear strain responses to high level of acoustic excitation. The dominance of second-mode contribution and softening-spring behavior are found in the strain response of postbuckled plates.

Description: This paper represents an attempt to apply extensions of a hybrid transfinite element computational approach for accurately predicting thermoelastic stress waves. The applicability of the present formulations for capturing the thermal stress waves induced by boundary heating for the well known Danilovskaya problems is demonstrated. A unique feature of the proposed formulations for applicability to the Danilovskaya problem of thermal stress waves in elastic solids lies in the hybrid nature of the unified formulations and the development of special purpose transfinite elements in conjunction with the classical Galerkin techniques and transformation concepts. Numerical test cases validate the applicability and superior capability to capture the thermal stress waves induced due to boundary heating.

Description: The antiplane strain problem of an edge cracked elastic plate subjected to a surface displacement load is solved, and the surface line loads are obtained using conformal mapping techniques. Results are presented for the yield, stress, strain and displacement distributions, and stress intensity factors in the plate. A superposition technique and Green's functions are used to determine the strain field on the upper plate surface due to arbitrarily applied tensile stress on the lower plate surface.

Description: The sublaminate or ply-level analysis of composite structures is presently undertaken by a computational procedure yielding the stresses in regions affected by delaminations, transverse cracks, and discontinuities that are related to material properties, geometries, and loads. Attention is given to layers or groups of layers that are immediately affected by flaws; these are analyzed as if they were homogeneous bodies in equilibrium, in isolation from the rest of the laminate. Computed stresses agree with those from a three-dimensional FEM analysis.

Description: A computational method/procedure is described which can be used to simulate individual and mixed mode interlaminar fracture progression in fiber composite laminates. Different combinations of Modes 1, 2, and 3 fracture are simulated by varying the crack location through the specimen thickness and by selecting appropriate unsymmetric laminate configurations. The contribution of each fracture mode to strain energy release rate is determined by the local crack closure methods while the mixed mode is determined by global variables. The strain energy release rates are plotted versus extending crack length, where slow crack growth, stable crack growth, and rapid crack growth regions are easily identified. Graphical results are presented to illustrate the effectiveness and versatility of the computational simulation for: (1) evaluating mixed-mode interlaminar fracture, (2) for identifying respective dominant parameters, and (3) for selecting possible simple test methods.

Description: Mode II and Mode I calibrations of the NASA Lewis Research Center Mode II fatigue specimen were performed experimentally over crack length to specimen width ratios (a/W) of 0.5 to 0.9. Mode II displacements were measured both at the specimen notch mouth and at the intersection of the notch with the centerline of the loading pin holes. Mode I displacements were measured across the span of the specimen at the loading pins' centerline. Analytical stress intensity factor coefficients for both Mode II and Mode I are also presented.

Description: Fatigue crack growth rate changes caused by single overloads and an overload followed by an underload have been studied in the high-strength aluminum alloy 7091. Crack-tip plasticity parameters were measured at each step in the loading sequence using the stereoimaging technique. Effective stress-intensity factor was measured with high resolution immediately at the crack tip both before an overload and during the subsequent growth rate retardation period. Crack-tip opening displacement and strain were altered in the same way as growth rate following the overload/underload events, and similitude with respect to crack-tip plasticity was preserved during the growth retardation period. Effective stress-intensity factor was found to correlate well with crack growth rate. Strains within the plastic zone were converted to stress at each step in load change. These stresses were then summed to determine the zone of residual stress caused by the overload. The region of compressive residual stress ahead of the crack tip was found to increase with overload severity but was limited to yield stress in magnitude. An underload shrank the spatial extent of the residual stresses caused by the overload.

Description: The fracture process in compact and bend specimens was simulated using a two-dimensional finite-element analysis of the J-resistance (J-R) curve and a CTOD criterion. The J-R curves were calculated from the numerical results for each specimen type using several different methods. In general, the J-R curves obtained for the bend specimens were found to be higher than those for the compact specimens, especially beyond maximum load. However, below the maximum load, the modified deformation theory of plasticity and the contour-integral J(r) method resulted in very similar J-R curves for both specimen types.

Description: The Boundary Force Method (BFM) was formulated for the two-dimensional stress analysis of complex crack configurations. In this method, only the boundaries of the region of interest are modeled. The boundaries are divided into a finite number of straight-line segments, and at the center of each segment, concentrated forces and a moment are applied. This set of unknown forces and moments is calculated to satisfy the prescribed boundary conditions of the problem. The elasticity solution for the stress distribution due to concentrated forces and a moment applied at an arbitrary point in a cracked infinite plate are used as the fundamental solution. Thus, the crack need not be modeled as part of the boundary. The formulation of the BFM is described and the accuracy of the method is established by analyzing several crack configurations for which accepted stress-intensity factor solutions are known. The crack configurations investigated include mode I and mixed mode (mode I and II) problems. The results obtained are, in general, within + or - 0.5 percent of accurate numerical solutions. The versatility of the method is demonstrated through the analysis of complex crack configurations for which limited or no solutions are known.

Description: The fracture resistance curve method of Newman (1985), based on the crack-tip-opening displacement, V(R), for a 'stationary' crack, was applied to various crack configurations in 2024-T351 and 7075-T651 aluminum alloys tested at room temperature. Using a stationary crack solution, the crack-tip-displacement was calculated at the current crack length for the crack configurations which included compact, middle-crack, single-edge-crack, and three-hole-crack tension specimens. The results showed that the V(R) resistance curves are insensitive to crack length, specimen width, and specimen type up to maximum load. After the maximum load is reached, the V(R) remains nearly constant; this constant depends only on specimen type, specimen width, and crack length. The V(R) resistance curve method can be used with the strip-yield analyses to accurately predict stable crack growth and instability of cracked metallic materials.

Description: In this paper, the viscoelastic boundary element method is used to estimate the opening displacement and the envelope stress on the surface of an isolated crack-induced-craze system. To predict the propagation history of both the crack and the craze in a polymer sheet, the material properties of the glassy polymers are represented by a generalized linear viscoelastic model. Results are compared with the theoretical micromechanics predictions. Good agreements are obtained.

Description: An important problem that has emerged from combined analytical/experimental investigations is the task of identifying and quantifying the differences between results predicted by F.E. analysis and results obtained from experiment. The objective of this study is to extend and evaluate the procedure developed by Sidhu for correlation of linear F.E. and modal test data to include structures with viscous damping. The desirability of developing this procedure is that the differences are identified in terms of physical mass, damping, and stiffness parameters instead of in terms of frequencies and modes shapes. Since the differences are computed in terms of physical parameters, locations of modeling problems can be directly identified in the F.E. model. From simulated data it was determined that the accuracy of the computed differences increases as the number of experimentally measured modes included in the calculations is increased. When the number of experimental modes is at least equal to the number of translational degrees of freedom in the F.E. model both the location and magnitude of the differences can be computed very accurately. When the number of modes is less than this amount the location of the differences may be determined even though their magnitudes will be underestimated.

Description: The space shuttle Challenger accident investigation focused on the failure of a tang-clevis joint on the right solid rocket motor. The existence of relative motion between the inner arm of the clevis and the O-ring sealing surface on the tang has been identified as a potential contributor to this failure. This motion can cause the O-rings to become unseated and therefore lose their sealing capability. Finite element structural analyses have been performed to predict both deflections and stresses in the joint under the primary, pressure loading condition. These analyses have demonstrated the difficulty of accurately predicting the structural behavior of the tang-clevis joint. Stresses in the vicinity of the connecting pins, obtained from elastic analyses, considerably exceed the material yield allowables indicating that inelastic analyses are probably necessary. Two modifications have been proposed to control the relative motion between the inner clevis arm and the tang at the O-ring sealing surface. One modification, referred to as the capture feature, uses additional material on the inside of the tang to restrict motion of the inner clevis arm. The other modification uses external stiffening rings above and below the joint to control the local bending in the shell near the joint. Both of these modifications are shown to be effective in controlling the relative motion in the joint.

Description: The problem of an orthotropic cantilevered plate subjected to a uniformly distributed end load is solved by the Rayleigh-Ritz energy method. The result is applied to laminated composite, double cantilevered specimens to estimate the effect of crack tip constraint on the transverse curvature, deflection and energy release rate. The solution is also utilized to determined finite width correction factors for fracture energy characterization tests in which neither plane stress nor plane strain conditions apply.

Description: A mathematical model based on the Euler-Bernoulli beam theory is proposed for predicting the effective Young's moduli of piece-wise isotropic composite laminates with wavy patterns in the main load-carrying layers. Strains in corrugated layers, in-phase layers, and out-of-phase layers are predicted for various geometries and material configurations by assuming matrix layers as elastic foundations. Experimental results obtained from corrugated aluminum specimens and aluminum/epoxy specimens with in-phase and out-of-phase wavy patterns coincide very well with the predictions. The work represents a preliminary effort toward further generalization of the model for two-dimensional anisotropic laminates containing wavy patterns in the main load-carrying layers.

Description: Results of the ratio of stress intensity factor to crack-mouth displacement as a function of crack length are presented for the wedge-loaded compact specimen. Comparisons are made between experimental compliance results, numerical results from collocation methods, and deep-crack limit-solution results. Applications are for crack-arrest and stress-corrosion-cracking tests for metals and other materials under predominantly linear elastic conditions.

Description: Practical and reliable estimators of the discretization errors in engineering problems are developed. Error indicators for identifying the regions or elements of the solution domain which are likely to have the largest discretization errors are presented, and a simple computational procedure for improving the accuracy of the finite element solutions for shell problems is given. The similarities between the proposed procedure and a preconditioned conjugate gradient (PCG) technique are identified and exploited to generate pointwise error indicators from the PCG technique. Numerical examples in the linear static analysis of shells are presented.

Description: Four computational simulation methods with different levels of sophistication were used to simulate thermal behavior and structural changes of composite sandwich panels with a honeycomb core subjected to a variety of environmental effects. The models on thich these methods are based include three-dimensional finite-element modeling, three-dimensional finite-element modeling assuming a homogeneous core, laminate theory, and simple equations for predicting the equivalent properties of the honeycomb core. A procedure was developed and embedded in a composite mechanics computer code, which made it possile to conduct parametric studies to determine 'optimum' composite sandwich configurations for specific applications. The procedure was applied for the evaluation of composite sandwich behavior at the global, local, laminate, ply, and micromechanics levels when the composite sandwich is subjected to hygral, thermal, and mechanical loading environments.

Description: The demand on safety performance of launching structure and equipment system from impulsive excitations necessitates a study which predicts the maximum response of the system as well as the maximum stresses in the system. A method to extract higher modes and frequencies for a class of multiple degree-of-freedom (MDOF) Structure system is proposed. And, along with the shock spectra derived from a linear oscillator model, a procedure to obtain upper bound solutions for maximum displacement and maximum stresses in the MDOF system is presented.

Description: A computerized data base of crack growth properties of materials was developed for use in fracture control analysis of rocket engine components and other NASA space hardware. The software system has files of basic crack growth rate data, other fracture mechanics material properties such as fracture toughness and environmental crack growth threshold values, and plotting and fitting routines for deriving material properties for use in fracture control analysis. An extensive amount of data was collected and entered, and work is continuing on compiling additional data. The data base and software codes are useful both for fracture control analysis and for evaluation or development of improved crack growth theories.

Description: A multi-input/output random vibration control algorithm was developed based on system identification concepts derived from random vibration spectral analysis theory. The unique features of the algorithm are: (1) the number of input excitors and the number of output control responses need not be identical; (2) the system inverse response matrix is obtained directly from the input/output spectral matrix; and (3) the system inverse response matrix is updated every control loop cycle to accommodate system amplitude nonlinearities. A laboratory demonstration case of two imputs with three outputs is presented to demonstrate the system capabilities.

Description: Several models were built at NASA Langley and used to demonstrate the following nonlinear behavior: internal resonance in a free response, principal parametric resonance and subcritical instability in a cantilever beam-lumped mass structure, combination resonance in a parametrically excited flexible beam, autoparametric interaction in a two-degree-of-freedom system, instability of the linear solution, saturation of the excited mode, subharmonic bifurcation, and chaotic responses. A video tape documenting these phenomena was made. An attempt to identify a simple structure consisting of two light-weight beams and two lumped masses using the Eigensystem Realization Algorithm showed the inherent difficulty of using a linear based theory to identify a particular nonlinearity. Preliminary results show the technique requires novel interpretation, and hence may not be useful for structural modes that are coupled by a guadratic nonlinearity. A literature survey was also completed on recent work in parametrically excited nonlinear system. In summary, nonlinear systems may possess unique behaviors that require nonlinear identification techniques based on an understanding of how nonlinearity affects the dynamic response of structures. In this was, the unique behaviors of nonlinear systems may be properly identified. Moreover, more accutate quantifiable estimates can be made once the qualitative model has been determined.

Description: A deployable structure which meets stringent thermal and strength requirements in a space environment was developed. A mast with a very low coefficient of thermal expansion (CTE) was required to limit the movement from thermal distortion over the temperature range of -200 C to 80 C to .064 cm (.025 in). In addition, a high bending strength over the temperature range and weight less than 18.1 kg (40 lbs) was needed. To meet all of the requirements, a composite, near-zero CTE structure was developed. The measured average CTE over the temperature range for the mast was .70 x .000001/C (.38 x .000001/F). The design also has the advantage of being adjustable to attain other specific CTE if desired.

Description: The Space Transportation System (STS) is a very complex and expensive flight system which is intended to carry payloads into low Earth orbit and return. A catastrophic failure of the STS (such as experienced in the 51-L incident) results in the loss of both human life as well as very expensive hardware. One impact of this incident was to reaffirm the need to do everything possible to insure the integrity and reliability of the STS is sufficient to produce a safe flight. One means of achieving this goal is to expand the number of inspection technologies available for use on the STS. The purpose was to begin to evaluate the possible use of assessing the structural integrity of STS components for which Marshall Space Flight Center (MSFC) has responsibility. This entailed reviewing the available literature and determining a low-level experimental program which could be performed by MSFC and would help establish the feasibility of using this technology for structural fault detection.

Description: One of the most difficult problems in engine structural component durability analysis is the determination of the temperatures and fluxes in the structural components directly in contact with the hot gas flow path. Currently there exists no rational analytical or numerical technique which can effectively deal with this problem. Since the temperature distribution in the structural components are strongly influenced by both the fluid flow and the deformation as well as the cooling system in the structure, the only effective way to deal with this problem is to develop an integrated solid mechanics, fluid mechanics and heat transfer analysis for this problem. Herein, the Boundary Element Method (BEM) is chosen as the basic analysis tool principally because the definition of quantities like fluxes, temperatures, displacements, and velocities are very precise on a boundary based discretization scheme. One fundamental difficulty is that a BEM analysis requires a considerable amount of analytical work which is not present in other numerical methods. During the past year, all of this analytical work was completed and a two dimensional, general purpose code was written. A portion of the work is summarized.

Description: Reusable space propulsion hot gas-path components are required to operate under severe thermal and mechanical loading conditions. These operating conditions produce elevated temperature and thermal transients which results in significant thermally induced inelastic strains, particularly, in the turbopump turbine blades. An inelastic analysis for this component may therefore be necessary. Anisotropic alloys such as MAR M-247 or PWA-1480 are being considered to meet the safety and durability requirements of this component. An anisotropic inelastic structural analysis for an SSME fuel turbopump turbine blade was performed. The thermal loads used resulted from a transient heat transfer analysis of a turbine blade. A comparison of preliminary results from the elastic and inelastic analyses is presented.

Description: The service life calculations of two computer codes, NASCRAC and NASA/FLAGRO, are compared. The analysis technique is based on linear elastic fracture mechanics (LEFM), in which stresses remain below the yield strength of an elastic/plastic material. To perform service life calculations, a relationship expressing incremental crack growth, DA/DN, as a function of loading, geometry, and material properties is necessary. Load and geometry are expressed in terms of the cyclic stress intensity factor, delta K. The crack growth rate as a function of delta K is then determined by material tests, plotting DA/DN versus delta K for the given material, loading condition, and environment. Crack growth rate equations such as the Paris, Walker, and modified Forman equations are used to obtain a best fit curve to the laboratory DA/DN versus delta K data.

Description: The requirements for non-contact temperature measurement capabilities for electronic materials processing in space are assessed. Non-contact methods are probably incapable of sufficient accuracy for the actual absolute measurement of temperatures in most such applications but would be useful for imaging in some applications.

Description: An overview is given of the use of MSC/NASTRAN in the analysis of rotating flexible blades. The geometrically nonlinear analysis using NASTRAN Solution Sequence 64 is discussed along with the determination of frequencies and mode shapes using Solution Sequence 63. Items unique to rotating blade analysis, such as setting angle, centrifugal softening effects, and hub flexibility, are emphasized.

Description: A study was performrd to determine the dynamic characteristics of the Space Shuttle Main Engine high pressure fuel turbopump (HPFTP) blades made of single crystal (SC) material. The first and second stage drive turbine blades of HPFTP were examined. The nonrotating natural frequencies were determined experimentally and analytically. The experimental results of the SC second stage blade were used to verify the analytical procedures. The study examined the SC first stage blade natural frequencies with respect to crystal orientation at typical operating conditions. The SC blade dynamic response was predicted to be less than the directionally solidified base. Crystal axis orientation optimization indicated that the third mode interference will exist in any SC orientation.

Description: Lane's (1957) analytical formulation of the unsteady pressure distribution on an oscillating two-dimensional flat plate cascade in supersonic axial flow has been developed into a computer code. This unsteady aerodynamic code has shown good agreement with other published data. This code has also been incorporated into an existing aeroelastic code to analyze the NASA Lewis supersonic through-flow fan design.

Description: Three semi-empirical aerodynamic stall models are compared with respect to their lift and moment hysteresis loop prediction, limit cycle behavior, easy implementation, and feasibility in developing the parameters required for stall flutter prediction of advanced turbines. For the comparison of aeroelastic response prediction including stall, a typical section model and a plate structural model are considered. The response analysis includes both plunging and pitching motions of the blades. In model A, a correction of the angle of attack is applied when the angle of attack exceeds the static stall angle. In model B, a synthesis procedure is used for angles of attack above static stall angles, and the time history effects are accounted for through the Wagner function.

Description: A structural and aeroelastic analysis of the SR-7L advanced turboprop is presented. Analyses were conducted for several cases at different blade pitch angles, blade support conditions, rotational speeds, free-stream Mach numbers, and number of blades. A finite element model of the final blade design was used to determine the blade's vibration behavior and its sensitivity to support stiffness. A computer code which was based on three-dimensional, subsonic, unsteady lifting surface aerodynamic theory, was used for the aeroelastic analysis to examine the blade's stability at a cruise condition of Mach 0.8 at 1700 rpm. The results showed that the calculated frequencies and mode shapes obtained agreed well with the published experimental data and that the blade is stable for that operating point.

Description: A modal forced response method for propfans in yawed flow is presented. This capability exists in the Aeroelastic Stability and Response of Propfan (ASTROP3) code developed at the Lewis Research Center. The code uses three-dimensional steady and unsteady cascade aerodynamics by Williams and Hwang (1986) and a NASTRAN finite element model to represent the blade structure. In addition, many utility programs exist in ASTROP3 that help in both the preprocessing of the NASTRAN model and the postprocessing of modal response results. The postprocessing work that computes the blade vibratory displacements and stresses in yawed flow are highlighted here.

Description: This study explores combining Component Mode Synthesis methods for coupling structural components with Parameter Identification procedures for improving the analytical modeling of the connections. Improvements in the connection stiffness and damping properties are computed in terms of physical parameters so that the physical characteristics of the connections can be better understood, in addition to providing improved input for the system model.

Description: The NASA Lewis aeroelastic research program is focused on unstalled and stalled flutter, forced response, and whirl flutter of turborotors and propfans. The objectives are to understand the physical phenomena of cascade flutter and response including blade mistuning.

Description: In structural dynamics the equations are usually expressed as finite elements. Neighbor elements need not be connected. The process of condensing a fine model into a coarse model and interpolating the low-frequency solution to the fine model is studied.

Description: Four current research projects are summarized: (1) active control of rotor system dynamics; (2) attenuation of rotor vibration using controlled pressure hydrostatic bearings; (3) a new seal test facility for measuring isotropic and anisotropic linear rotordynamic characteristics; and (4) the use of rotordynamic instability thresholds to accurately measure bearing rotordynamic characteristics.

Description: A procedure is developed to determine approximate periodic solutions of autonomous and non-autonomous systems. The trignometric collocation method (TCM) is formalized to allow for the analysis of relatively small order systems directly in physical coordinates. The TCM is extended to large order systems by utilizing modal analysis in a component mode synthesis strategy. The procedure was coded and verified by several check cases. Numerical results for two small order mechanical systems and one large order rotor dynamic system are presented. The method allows for the possibility of approximating periodic responses for large order forced and self-excited nonlinear systems.

Description: An overall picture of the impact damper is obtained by using time-history solutions of the system motion for the oscillator in free decay. The impactor behavior depends very strongly on oscillator amplitude, and free decay can sample the full range of behavior from an infinite number of impacts per cycle at high amplitude to no impacts at low amplitude. The overall picture cannot be obtained by analysis of steady-state forced response. Yet, the predictions are relevant to forced response behavior when the damping is relatively light. Three major regimes of impact behavior are shown to exist: low, middle and high amplitude ranges.

Description: A bithermal fatigue test technique was proposed as a simplified alternative to the thermomechanical fatigue test. Both the thermomechanical cycle and the bithermal technique can be used to study nonisothermal fatigue behavior. The difference between the two cycles is that in a conventional thermomechanical fatigue cycle the temperature is continuously varied concurrently with the applied mechanical strains, but in the bithermal fatigue cycle the specimen is held at zero load during the temperature excursions and all the loads are applied at the two extreme temperatures of the cycle. Experimentally, the bithermal fatigue test technique offers advantages such as ease in synchronizing the temperature and mechanical strain waveforms, in minimizing temperature gradients in the specimen gauge length, and in reducing and interpreting thermal fatigue such as the influence of alternate high and low temperatures on the cyclic stress-strain response characteristics, the effects of thermal state, and the possibility of introducing high- and low-temperature deformation mechanisms within the same cycle. The bithermal technique was used to study nonisothermal fatigue behavior of alloys such as single-crystal PWA 1480, single-crystal Rene N4, cast B1900+Hf, and wrought Haynes 188.

Description: The problem of calculating expected component life under fatigue loading conditions is complicated by the fact that component loading histories contain, in many cases, cyclic loads of widely varying amplitudes. In such a case a cumulative damage model is required, in addition to a fatigue damage criterion, or life relationship, in order to compute the expected fatigue life. The traditional cumulative damage model used in design is the linear damage rule. This model, while being simple to use, can yield grossly unconservative results under certain loading conditions. Research at the NASA Lewis Research Center has led to the development of a nonlinear cumulative damage model, named the double damage curve approach (DDCA), that has greatly improved predictive capability. This model, which considers the life (or loading) level dependence of damage evolution, was applied successfully to two polycrystalline materials, 316 stainless steel and Haynes 188. The cumulative fatigue behavior of the PWA 1480 single-crystal material is currently being measured to determine the applicability of the DDCA for this material.

Description: A fatigue loading stage inside a scanning electron microscopy (SEM) was developed. The stage allows dynamic and static high-magnification and high-resolution viewing of the fatigue crack initiation and crack propagation processes. The loading stage is controlled by a closed-loop servohydraulic system. Maximum load is 1000 lb (4450 N) with test frequencies ranging up to 30 Hz. The stage accommodates specimens up to 2 inches (50 mm) in length and tolerates substantial specimen translation to view the propagating crack. At room temperature, acceptable working resolution is obtainable for magnifications ranging up to 10,000X. The system is equipped with a high-temperature setup designed for temperatures up to 2000 F (1100 C). The signal can be videotaped for further analysis of the pertinent fatigue damage mechanisms. The design allows for quick and easy interchange and conversion of the SEM from a loading stage configuration to its normal operational configuration and vice versa. Tests are performed entirely in the in-situ mode. In contrast to other designs, the NASA design has greatly extended the life of the loading stage by not exposing the bellows to cyclic loading. The loading stage was used to investigate the fatigue crack growth mechanisms in the (100)-oriented PWA 1480 single-crystal, nickel-based supperalloy. The high-magnification observations revealed the details of the crack growth processes.

Description: Ceramics materials have the potential for use in high-temperature, fuel-efficient engines. However, because these materials are brittle, their fracture characteristics must be well documented prior to their application. Thus Lewis is working to understand the fracture and strength properties of brittle ceramic and ceramic matrix materials. An understanding of fracture properties aids both designers who are attempting to design high-temperature structures and materials scientists who seek to design more temperature-resistant materials. Both analytical and experimental approaches to fracture analysis are being taken. Methods for testing fracture toughness, crack growth resistance, and strength are being developed. The failure mechanisms at both room and elevated temperatures are also being investigated. Such investigations aid materials scientists in developing better high-temperature materials. Of concern is the anisotropy of ceramic materials and the experimental verification of ceramic design codes that will allow brittle material behavior to be accurately predicted at high temperature.

Description: A simple two node, linear, finite strip plate bending element based on Mindlin-Reissner plate theory for the analysis of very thin to thick bridges, plates, and axisymmetric shells is presented. The new transverse shear strains are assumed for constant distribution in the two node linear strip. The important aspect is the choice of the points that relate the nodal displacements and rotations through the locking transverse shear strains. The element stiffness matrix is explicitly formulated for efficient computation and ease in computer implementation. Numerical results showing the efficiency and predictive capability of the element for analyzing plates with different supports, loading conditions, and a wide range of thicknesses are given. The results show no sign of the shear locking phenomenon.

Description: The structural performance of a space station thermal energy storage (TES) canister subject to orbital solar flux variation and engine cold start up operating conditions was assessed. The impact of working fluid temperature and salt-void distribution on the canister structure are assessed. Both analytical and experimental studies were conducted to determine the temperature distribution of the canister. Subsequent finite element structural analyses of the canister were performed using both analytically and experimentally obtained temperatures. The Arrhenius creep law was incorporated into the procedure, using secondary creep data for the canister material, Haynes 188 alloy. The predicted cyclic creep strain accumulations at the hot spot were used to assess the structural performance of the canister. In addition, the structural performance of the canister based on the analytically determined temperature was compared with that based on the experimentally measured temperature data.

Description: Three dimensional finite element analyses using MSC/NASTRAN and MARC are performed to predict the thermal and structural response of various cooling schemes under high heat loads. Steady state heat transfer analyses and elastic stress analyses are performed using MSC/NASTRAN. Elastic/plastic analyses are done using MARC. To help verify these analyses experimentally, a hydrogen-oxygen rocket engine was modified to use the exhaust stream as a high enthalpy, high heat flux source to evaluate various actively cooled, simulated cowl lip (leading edges) segments as well as flat structural segments. Cross flow and parallel flow cooling configurations were tested and analyzed using cooling fluids of water and gaseous hydrogen. In addition, various material types, including high conductivity copper, nickel, and a copper and graphite metal matrix composite were tested and compared.

Description: The effects of actual variations, also called uncertainties, in geometry and material properties on the structural response of a space shuttle main engine turbopump blade are evaluated. A normal distribution was assumed to represent the uncertainties statistically. Uncertainties were assumed to be totally random, partially correlated, and fully correlated. The magnitude of these uncertainties were represented in terms of mean and variance. Blade responses, recorded in terms of displacements, natural frequencies, and maximum stress, was evaluated and plotted in the form of probabilistic distributions under combined uncertainties. These distributions provide an estimate of the range of magnitudes of the response and probability of occurrence of a given response. Most importantly, these distributions provide the information needed to estimate quantitatively the risk in a structural design.

Description: General purpose finite element computer codes that can model inelastic material behavior have been available for more than a decade. However, these codes have not been accurate enough for use in analyzing hot section engine components. To correct this problem, General Electric developed a series of nine new stand-alone computer codes for NASA. Because of the large temperature excursions associated with hot section engine components, these codes have been designed to accommodate broad variations in material behavior, including plasticity and creep. The capabilities of these computer codes are summarized.

Description: The Mechanics of Materials Model (MOMM) is one of a series of new stand-alone three dimensional nonlinear structural analysis codes. Incorporation of a general purpose finite element computer code into the hot section design process was severely limited by the high costs involved. MOMM is a stiffness method finite element code that uses an internally generated network of beams to characterize hot section component behavior. The method was proposed as a fast, easy to use, computationally efficient tool for approximate analyses. MOMM incorporates a wide variety of analysis capabilities, material models, and load type specifiers instrumental for the analysis of hot section components.

Description: A need for development of realistic constitutive models for structural components operating at high temperatures, accompanied by appropriate solution technologies for stress/life analyses of these components is studied. Viscoplastic models provide a better description of inelastic behavior of materials, but their mathematical structure is very complex. The highly nonlinear and stiff nature of the constitutive equations makes analytical solutions difficult. Therefore, suitable solution, finite element or other numerical, technologies must be developed to make these models adaptable for better and rational designs of components. NASA-Lewis has developed several solution technologies and successfully applied them to the solution of a number of uniaxial and multiaxial problems. Some of these solution technologies are described along with the models and representative results. The solution technologies developed and presented encompass a wide range of models, such as, isotropic, anisotropic, metal matrix composites, and single crystal models.

Description: Complex states of stress and strain are introduced into components during service in engineering applications. It follows that analysis of such components requires material descriptions, or constitutive theories, which reflect the tensorial nature of stress and strain. For applications involving stress levels above yield, the situation is more complex in that material response is both nonlinear and history dependent. This has led to the development of viscoplastic constitutive theories which introduce time by expressing the flow and evolutionary equation in the form of time derivatives. Models were developed here which can be used to analyze high temperature components manufactured from advanced composite materials. In parallel with these studies, effort was directed at developing multiaxial testing techniques to verify the various theories. Recent progress in the development of constitutive theories from both the theoretical and experimental viewpoints are outlined. One important aspect is that material descriptions for advanced composite materials which can be implemented in general purpose finite element codes and used for practical design are verified.

Description: Many high temperature aircraft and rocket engine components experience large mechanical loads as well as severe thermal gradients and transients. These nonisothermal conditions are often large enough to cause inelastic deformations, which are the ultimate cause for failure in those parts. A way to alleviate this problem is through improved engine designs based on better predictions of thermomechanical material behavior. To address this concern, an experimental effort was recently initiated within the Hot Section Technology (HOST) program at Lewis. As part of this effort, two new test systems were added to the Fatigue and Structures Lab., which allowed thermomechanical tests to be conducted under closely controlled conditions. These systems are now being used for thermomechanical testing for the Space Station Receiver program, and will be used to support development of metal matrix composites.

Description: Structural alloys embody internal mechanisms that allow recovery of state with varying stress and elevated temperature, i.e., they can return to a softer state following periods of hardening. Such material behavior is known to strongly influence structural response under some important thermomechanical loadings, for example, that involving thermal ratchetting. The influence of dynamic and thermal recovery on the creep buckling of a column under variable loading is investigated. The column is taken as the idealized (Shanley) sandwich column. The constitutive model, unlike the commonly employed Norton creep model, incorporates a representation of both dynamic and thermal (state) recovery. The material parameters of the constitutive model are chosen to characterize Narloy Z, a representative copper alloy used in thrust nozzle liners of reusable rocket engines. Variable loading histories include rapid cyclic unloading/reloading sequences and intermittent reductions of load for extended periods of time; these are superimposed on a constant load. The calculated results show that state recovery significantly affects creep buckling under variable loading.

Description: The applicability of a classical constitutive model for stress-strain analysis of a nickel base superalloy, Rene' 80, in the gas turbine thermomechanical fatigue (TMF) environment is examined. A variety of tests were conducted to generate basic material data and to investigate the material response under cyclic thermomechanical loading. Isothermal stress-strain data were acquired at a variety of strain rates over the TMF temperature range. Creep curves were examined at 2 temperature ranges, 871 to 982 C and 760 to 871 C. The results provide optimism on the ability of the classical constitutive model for high temperature applications.

Description: A viscoplastic material model for the high temperature turbine airfoil material B1900 + Hf was developed and was demonstrated in a three dimensional finite element analysis of a typical turbine airfoil. The demonstration problem is a simulated flight cycle and includes the appropriate transient thermal and mechanical loads typically experienced by these components. The Walker viscoplastic material model was shown to be efficient, stable and easily used. The demonstration is summarized and the performance of the material model is evaluated.

Description: A unified numerical method for the integration of stiff time dependent constitutive equations is presented. The solution process is directly applied to a constitutive model proposed by Bodner. The theory confronts time dependent inelastic behavior coupled with both isotropic hardening and directional hardening behaviors. Predicted stress-strain responses from this model are compared to experimental data from cyclic tests on uniaxial specimens. An algorithm is developed for the efficient integration of the Bodner flow equation. A comparison is made with the Euler integration method. An analysis of computational time is presented for the three algorithms.

Description: An automated procedure is presented for evaluating the material parameters in Walker's exponential viscoplastic constitutive model for metals at elevated temperature. Both physical and numerical approximations are utilized to compute the constants for Inconel 718 at 1100 F. When intermediate results are carefully scrutinized and engineering judgement applied, parameters may be computed which yield stress output histories that are in agreement with experimental results. A qualitative assessment of the theta-plot method for predicting the limiting value of stress is also presented. The procedure may also be used as a basis to develop evaluation schemes for other viscoplastic constitutive theories of this type.

Description: A global/local time incrementing scheme for viscoplastic analysis of structures is presented. The scheme is very efficient and useful for conducting large scale nonlinear finite element analysis involving viscoplastic materials.

Description: An experimentally based unified creep-plasticity constitutive model was implemented for 1070 steel. Accurate rate and temperature effects were obtained for isothermal and thermo-mechanical loading by incorporating deformation mechanisms into the constitutive equations in a simple way.

Description: Four current viscoplastic models are compared experimentally for Inconel 718 at 593 C. This material system responds with apparent negative strain rate sensitivity, undergoes cyclic work softening, and is susceptible to low cycle fatigue. A series of tests were performed to create a data base from which to evaluate material constants. A method to evaluate the constants is developed which draws on common assumptions for this type of material, recent advances by other researchers, and iterative techniques. A complex history test, not used in calculating the constants, is then used to compare the predictive capabilities of the models. The combination of exponentially based inelastic strain rate equations and dynamic recovery is shown to model this material system with the greatest success. The method of constant calculation developed was successfully applied to the complex material response encountered. Backstress measuring tests were found to be invaluable and to warrant further development.

Description: From cyclic, strain controlled, nonproportional tests on type 304 stainless steel and Hastelloy-X, the following statements may be made: (1) A dynamic recovery term is essential to properly model the backstress evolution. (2) From analysis of Hastelloy-X data obtained at 649 C, the inelastic strain rate appears to be a satisfactory directional index for direct hardening, but the backstress appears to be an inappropriate directional index of dynamic recovery. (3) Sinusoidal, 90 deg out-of-phase axial torsional tests can be very useful in aiding determination of backstress evolution functions, including both directional indices and scalar hardening functions, by virtue of the associated approximately constant magnitudes of overstress, inelastic strain rate, and effective stress. Such tests have previously been associated with the study of nonproportional hardening effects but have more far ranging applications.

Description: The utility of advanced constitutive models and structural analysis methods are evaluated for predicting the cyclic life of an air-cooled turbine blade for a gas turbine aircraft engine. Structural analysis methods of various levels of sophistication were exercised to obtain the cyclic stress-strain response at the critical airfoil location. Calculated strain ranges and mean stresses from the stress-strain cycles were used to predict crack initiation lives by using the total strain version of the strain range partitioning life prediction method. The major results are given and discussed.

Description: A constitutive theory for use in structural and durability analyses of high temperature isotropic alloys is presented. Constitutive equations based upon a potential function are determined from conditions of stability and physical considerations. The theory is self-consistent; terms are not added in an ad hoc manner. It extends a proven viscoplastic model by introducing the Kachanov-Rabotnov concept of net stress. Material degradation and inelastic deformation are unified; they evolve simultaneously and interactively. Both isotropic hardening and material degradation evolve with dissipated work which is the sum of inelastic work and internal work. Internal work is a continuum measure of the stored free energy resulting from inelastic deformation.

Description: A simplified uniaxial strain controlled creep damage law is deduced with the use of experimental observation from a more complex strain dependent law. This creep damage law correlates the creep damage, which is interpreted as the density variation in the material, directly with the accumulated creep strain. Based on the deduced uniaxial strain controlled creep damage law, a continuum mechanical creep rupture analysis is carried out for a beam resting on a high temperature elastic (Winkler) foundation. The analysis includes the determination of the nondimensional time for initial rupture, the propagation of the rupture front with the associated thinning of the beam, and the influence of creep damage on the deflection of the beam. Creep damage starts accumulating in the beam as soon as the load is applied, and a creep rupture front develops at and propagates from the point at which the creep damage first reaches its critical value. By introducing a series of fundamental assumptions within the framework of technical Euler-Bernoulli type beam theory, a governing set of integro-differential equations is derived in terms of the nondimensional bending moment and the deflection. These governing equations are subjected to a set of interface conditions at the propagating rupture front. A numerical technique is developed to solve the governing equations together with the interface equations, and the computed results are presented and discussed in detail.

Description: Development and solution of coupled thermomechanical equations at elevated temperature and/or high strain rates are discussed. Three main considerations are presented: development of the coupled thermomechanical equations by means of the rational theory of thermodynamics, development of a thermoviscoplastic constitutive equation which is congruous with the developed coupled equations, and the applicability of the developed equations to the treatment by the finite element method.

Description: The design of kinematic supports using elastic elements is reviewed. The two standard methods (cone, Vee and flat and three Vees) are presented and a design example involving a machine tool metrology bench is given. Design goals included thousandfold strain attenuation in the bench relative to the base when the base strains due to temperature variations and shifting loads. Space applications are also considered.

Description: A method is presented for coupling a broad class of component dynamic models in the manner of direct stiffness assembly. The method is implemented in a general matrix manipulation program.

Description: An approximate method to compute the maximum deformation and permanent set of a beam subjected to shock wave laoding in vacuo and in water was investigated. The method equates the maximum kinetic energy of the beam (and water) to the elastic plastic work done by a static uniform load applied to a beam. Results for the water case indicate that the plastic deformation is controlled by the kinetic energy of the water. The simplified approach can result in significant savings in computer time or it can expediently be used as a check of results from a more rigorous approach. The accuracy of the method is demonstrated by various examples of beams with simple support and clamped support boundary conditions.

Description: Component mode synthesis was used to analyze different types of structures with MSC NASTRAN. The theory and technique of using Multipoint Constraint Equations (MPCs) to connect substructures to each other or to a common foundation is presented. Computation of the dynamic response of the system from shack spectrum inputs was automated using the DMAP programming language of the MSC NASTRAN finite element code.

Description: The problem of predicting the phase angle of two self-synchronized rotors starting from rest is presented. It is shown that with insufficient power the rotors may not reach the final operating speed of the motors and stay locked at one of the lower natural frequencies of the vibrating system, thus producing large amplitude and failure of the equipment.

Description: The results of a wind tunnel test program to determine the performance loads and dynamic characteristics of the Composite Flexbeam Tail Rotor (CFTR) for the AH-64 Advanced Attack Helicopter are reported. The CFTR uses an elastomeric shear attachment of the flexbeam to the hub to provide soft-inplane S-mode and stiff-inplane C-mode configuration. The properties of the elastomer were selected for proper frequency placement and scale damping of the inplane S-mode. Kinematic pitch-lag coupling was introduced to provide the first cyclic inplane C-mode damping at high collective pitch. The CFTR was tested in a wind tunnel over the full slideslip envelop of the AH-64. It is found that the rotor was aeroelastically stable throughout the complete collective pitch range and up to rotor speeds of 1403 rpm. The dynamic characteristics of the rotor were found to be satisfactory at all pitch angles and rotor speeds of the tunnel tests. The design characteristics of the rotor which permit the high performance characteristics are discussed. Several schematic drawings and photographs of the rotor are provided.

Description: Most dynamic components in helicopters are designed with a safe-life constant-amplitude testing approach that has not changed in many years. In contrast, the fatigue methodology in other industries has advanced significantly in the last two decades. Recent research at the NASA Langley Research Center and the U.S. Army Aerostructures Directorate at Langley are reviewed relative to fatigue and fracture design methodology for metallic components. Most of the Langley research was directed towards the damage tolerance design approach, but some work was done that is applicable to the safe-life approach. In the areas of testing, damage tolerance concepts are concentrating on the small-crack effect in crack growth and measurement of crack opening stresses. Tests were conducted to determine the effects of a machining scratch on the fatigue life of a high strength steel. In the area of analysis, work was concentrated on developing a crack closure model that will predict fatigue life under spectrum loading for several different metal alloys including a high strength steel that is often used in the dynamic components of helicopters. Work is also continuing in developing a three-dimensional, finite-element stress analysis for cracked and uncracked isotropic and anisotropic structures. A numerical technique for solving simultaneous equations called the multigrid method is being pursued to enhance the solution schemes in both the finite-element analysis and the boundary element analysis. Finally, a fracture mechanics project involving an elastic-plastic finite element analysis of J-resistance curve is also being pursued.

Description: A theory for the prediction of the creep and the microyield strength of composite laminates with viscoelastic matrices has been developed and tested. Precise creep curves from three laminates with arbitrary orientations provide the necessary input data. The time-dependent lamina properties are derived for subsequent use with standard laminate codes. Both the level and the shape of creep and strain recovery curves can be predicted. Extensions of the theory to include the effects of microcracking are outlined.

Description: An experimental investigation of adhesively bonded composite joint was conducted to characterize the debond growth mechanism under mode II static and fatigue loadings. For this purpose, end-notched flexure specimens of graphite/epoxy (T300/5208) adherends bonded with EC 3445 adhesive were tested. In all specimen tested, the fatigue failure occurred in the form of cyclic debonding. The present study confirmed the result of previous studies that total strain-energy-release rate is the driving parameter for cyclic debonding. Further, the debond growth resistance under cyclic loading with full shear reversal (i.e., stress ratio, R = -1) is drastically reduced in comparison to the case when subjected to cyclic shear loading with no shear reversal (i.e., R = 0.1).

Description: The location of crack closure with respect to crack wake and specimen thickness under different loading conditions was determined. The rate of increase of K sub CL in the crack wake was found to be significantly higher for plasticity induced closure in comparison to roughness induced closure. Roughness induced closure was uniform throughout the thickness of the specimen while plasticity induced closure levels were 50 percent higher in the near surface region than in the midthickness. The influence of state of stress on low-high load interaction effects was also examined. Load interaction effects differed depending upon the state of stress and were explained in terms of delta K sub eff.

Description: A three-dimensional elastic-plastic finite-element analysis of crack growth and closure under cyclic loading has been performed in order to investigate the behavior of a crack in a finite-thickness middle-crack tension specimen. The cases of both constant-amplitude loading and a single-spike overload are considered. The calculated crack-opening stresses for the exterior of the specimen agree with previous plane-stress results, and those obtained for the interior of the specimen agree with previous plane-strain results.

Description: The fatigue crack opening load is determined as the tangent point on the ascending load-displacement data between the curved portion and the upper linear region. A model for the 'unzipping' behavior of the crack indicates that the curved portion of the load-deflection curve is second order. The opening load is determined by a nonlinear, least squares fit of the data to the model, which optimally locates the tangent point of the two curves. The method provided consistent results for determining opening load P(op) for 7475-T731 aluminum using data from a crack tip opening gage.

Description: Loading rate effects on the mode I delamination fracture toughness of AS4/3501-6 graphite/epoxy are presently studied by means of a height-tapered double-cantilever beam specimen whose height contour is designed to furnish a slightly decreasing compliance with increasing crack length, in order to yield a stable and smooth crack propagation at high loading rates. This specimen geometry also allows much higher crack propagation velocities to be obtained with either uniform or width-tapered double cantilever beam specimens.

Description: A set of thermoviscoplastic nonlinear constitutive relationships (TVP-NCR) developed for application to high-temperature metal matrix composites (HT-MMC) is described. The structural response of a turbine blade, made from fiber-reinforced superalloy HT-MMC and subject to representative loading conditions, is evaluated. Results indicate that this set of TVP-NCR is computationally effective.

Description: Research on the mechanics of composite structures at NASA's Langley Research Center is discussed. The advantages and limitations of special purpose and general purpose analysis tools used in research are reviewed. Future directions in computational structural mechanics are described to address analysis short-comings. Research results on the buckling and postbuckling of unstiffened and stiffened composite structures are presented. Recent investigations of the mechanics of failure in compression and shear are reviewed. Preliminary studies of the dynamic response of composite structures due to impacts encountered during crash-landings are presented. Needs for future research are discussed.