ALBERT

Description: The paper presents the descriptions of recently developed numerical algorithms that prove to be useful for the solution of the free vibration problem of spinning structures. First, a generalized procedure for the computation of nodal centrifugal forces in a finite element owing to any specified spin rate is derived in detail. This is followed by a description of an improved eigenproblem solution procedure that proves to be economical for the free vibration analysis of spinning structures. Numerical results are also presented which indicate the efficacy of the currently developed procedures.

Description: The analysis of empirical modal models of structural components using techniques based on the modal assurance criterion (MAC) defined by Ewins (1984) is discussed and demonstrated for the case of the Space Shuttle Orbiter aft bulkhead. Modal models of the bulkhead are constructed before and after environmental acoustic tests equivalent to 30 or 100 missions and compared (by computing their MACs) to detect significant structural changes; subdivision of the modal vector followed by computation of the MAC for each part is used to localize the changes. Numerical data demonstrating the usefulness of the MAC approach are presented in tables and diagrams.

Description: An assessment is made of the potential of different global-local analysis strategies for predicting the nonlinear and postbuckling responses of structures. Two postbuckling problems of composite panels are used as benchmarks and the application of different global-local methodologies to these benchmarks is outlined. The key elements of each of the global-local strategies are discussed and future research areas needed to realize the full potential of global-local methodologies are identified.

Description: Formal approaches are summarized to evaluate design concepts and perform sensitivity analyses on design parameters of composite structural components for vehicles. The formal approaches include structural analyses coupled with composite micromechanics to assess the structural response of beams made from various intraply hybrids, finite element analysis in conjunction with composite mechanics to assess the structural response of panels made from strip hybrids, and sensitivity analysis through optimization to assess the effects of various design parameters on the optimum design of a panel made from angleplied composite laminates. Results obtained from these approaches are presented in graphical and tabular form to illustrate parametric studies and acceptable ranges of various design parameters.

Description: Vibration absorbers are introduced into an asymmetric configuration of thin cylinders and tori enclosing an acoustic medium. The absorbers consist of thin axial strips bonded to the cylinder with a thin viscoelastic layer. The constrained layer dissipates the energy of relative motions between strip and cylinder. The absorber is most effective on response modes with two or more circumferential waves. The use of transfer matrices is extended to the coupled cylinder-absorber system.

Description: An arbitrarily laminated, anisotropic cylindrical shell of finite length, under uniform internal pressure, is analyzed using Love-Timoshenko's kinematic relations and under the framework of classical lamination theory. The previously obtained solutions for asymmetrically laminated orthotropic (cross-ply) as well as unbalanced-symmetric and balanced-unsymmetric (angle-ply) cylindrical shells under the same loading conditions have been shown to be special cases of the present closed-form solution. Numerical results have been presented for a two-layer cylindrical shell and compared with those obtained using finite element solutions based on the layerwise constant shear-angle theory. These are expected to serve as benchmark solutions for future comparisons and to facilitate the use of unsymmetric lamination in design.

Description: The stress-intensity factors are determined for a cracked orthotropic sheet adhesively bonded to an orthotropic stringer where the adhesive layer is modeled with a nonlinear stress-strain curve. By the use of Green's functions and the complex variable theory of orthotropic elasticity, a set of integral equations is obtained. The integral equations are replaced by an equivalent set of algebraic equations, which are solved to obtain the shear stress distribution in the adhesive layer, with which the crack-tip stress-intensity factors are found. When the adhesive was modeled with a nonlinear stress-strain curve, the peak shear stresses in the adhesive were considerably reduced in comparison to the solution for the linear elastic adhesive. This resulted in increases in the stress-intensity factors for the nonlinear adhesive solution compared to the linear adhesive solution. The nonlinear adhesive has no significant effect on the stress-intensity factor unless the near crack tip is beneath the stringer. It is assumed that the adhesive bond remains intact and it is predicted that onset of adhesive failure occurs at decreasing levels of applied stress as the crack propagates beneath the stringer.

Description: Unified constitutive material models were developed for structural analyses of aircraft gas turbine engine hot section components with particular application to an isotropic material used for combustor liners. Differential forms of models independently developed were considered in this study. These models combine the interactions of time-dependent (creep) and time-independent (plasticity) inelastic behavior of a material. Predicted stress-strain responses from these models were evaluated against cyclic isothermal and nonisothermal test results for uniaxial specimens of a nickel-base superalloy. The unified models were implemented in a nonlinear structural analysis code. Two unique NASA Lewis test facilities were used in the evaluation of the models for complex geometry specimens and evaluation of advanced temperature and high-temperature strain measurement instrumentation. Predicted nonlinear structural responses from one of the models for a flat plate and a segment of a conventional combustor liner are presented.

Description: This paper outlines a technique for preliminary sizing of the adherends for cracked-lap-shear specimen to assure delamination instead of adherend failure. The cracked-lap-shear specimen is a popular specimen used for determination of mixed mode interlaminar fracture toughness of composites. Thin adherends, on the order of three plies thick, have been used successfully in the past for determination of fracture toughness of rather brittle matrix systems. However, the tougher matrices now being considered for composite applications require thicker adherends in order to get composite delamination rather than adherend failure. The optimum thickness is obtained on the basis that the critical value of the strain energy release rate is reached at the delamination front before the ultimate strength is reached in the adherend. By proper design the specimen would delaminate and not use excessive material or load. A simple analysis for the optimum adherend thickness and experimental verification are included.

Description: Future space structures will have a low mass density and high flexibility, with ground test dynamic behavior differing significantly from that in zero-G orbit. Attention is presently given to the vibration behavior of a beam deformed by its own weight; the results obtained by the differential equations for both the static and dynamic responses of a large, simply supported beam, which are derived and solved analytically, allow ground test experiment measurements to be used for orbital dynamic characteristics verification efforts.

Description: The delamination resistance of graphite-reinforced PEEK composites was quantified by conducting static and cyclic edge delamination tests on (35n/-35n/0n/90n)s AS4/PEEK laminates, where n = 1, 2. The experimentally determined mechanical delamination onset strains were used to calculate the critical strain-energy release rate for delamination onset as a function of fatigue cycle. The delamination onset strains decreased dramatically with fatigue cycles and then began to level off to an endurance limit at 1 million cycles. Although the static interlaminar fracture toughness of the AS4/PEEK composite is much greater than the toughness of graphite epoxy composites, the delamination fatigue threshold, calculated from the cyclic strain endurance limit at 1 million cycles, was only slightly greater than the threshold for graphite epoxy composites. The contribution of residual thermal stresses to delamination in the AS4/PEEK is substantial due to the large temperature range between the manufacture and the room temperatures.

Description: A model is described for estimating the impact damage of fiber reinforced composite plates. The displacements and stresses are calculated by a three dimensional transient, finite element method of solution of the governing equations applicable to a linearly elastic body. The region in which damage occurs is estimated using the Tsai-Wu failure criterion. A computer code was developed which can be used to calculate the impact force, displacements and velocities of the plate and the impact body, stresses and strains in the plate, and the damage area. Sample numerical results are presented illustrating the type of information provided by the code. Comparisons between measured and calculated damage areas are also given.

Description: The dynamic behavior of a spinning linear-elastic paraboloid subject to nonaxisymmetric deformation is investigated analytically, applying the Rayleigh-Ritz procedure described by Utku et al. (1983). Energy-density, strain-displacement, and velocity-displacement expressions are generated; expressions for the generalized strain and position vector are derived; and the discretized dynamic equations are obtained. Numerical results obtained with a computer-program implementation of the method are presented in extensive tables and graphs. The effects of spin rate and bending rigidity and results for the special case of a spinning disk are included.

Description: This report examines tuning a finite element model using vector optimization techniques. Structural models using finite element theory often need to be adjusted so they can accurately simulate the real structure. The goal is to tune the model such that it will reproduce data derived from structural tests. First, the performance indices are extremized using multiple objective optimization theory, producing a set of possible solutions. Next, the solutions are rank ordered according to a decision maker's preferences to select the best answer. The tuning process was applied to a T-38 horizontal stabilizer. Numerous weighted solutions contained a best static deformation model, a best frequency model and three intermediate combinations of these two models. This automated procedure proved to be a versatile method capable of producing solutions for many types of tuning problems.

Description: A method which has evolved in the laboratories for the past 20 yr is re-examined with the intent of improving its accuracy and simplicity of application to engineering problems. Several modifications are introduced both to the analytical formulation of the Damage Curve Approach, and to the procedure for modifying this approach to achieve a Double Linear Damage Rule formulation which immensely simplifies the calculation. Improvements are also introduced in the treatment of mean stress for determining fatigue life of the individual events that enter into a complex loading history. While the procedure is completely consistent with the results of numerous two level tests that have been conducted on many materials, it is still necessary to verify applicability to complex loading histories. Caution is expressed that certain phenomenon can also influence the applicability - for example, unusual deformation and fracture modes inherent in complex loading especially if stresses are multiaxial. Residual stresses at crack tips, and metallurgical factors are also important in creating departures from the cumulative damage theories; examples of departures are provided.

Description: The static stiffnesses for torsional and flexural deformation of a tapered beam under axial loading are determined analytically, extending the Bernoulli-Euler/Bessel-function approach of Banerjee and Williams (1985). The derivation of the expressions is explained in detail, and numerical results for sample problems are presented in tables and shown to be in excellent agreement with those obtained using stepped beams of 400 or 500 elements.

Description: A study was undertaken to characterize the debond growth mechanism of adhesively bonded composite joints under mode I, mixed mode I-II, and mode II static loadings. The bonded system consisted of graphite/epoxy (T300/5208) composite adherends bonded with a toughened epoxy (EC 3445) adhesive. The mode I, mode II and mixed-mode I-II fracture energies of the tested adhesive were found to be equal to each other. Furthermore, the criterion for mixed mode fracture in composite bonded joints was determined.

Description: Elastic displacements and stress intensity measurements for a mode II specimen have been obtained over a range of a/W values between 0.500 and 0.900 using the MARC general purpose finite element program. Stress intensity factors were experimentally determined using load point displacement values. Good general agreement between numerical and experimental results for crack mouth, crack surface, and load point displacements, and for stress intensity factors, demonstrates the accuracy of the present method.

Description: It has been observed that the residual stresses and dislocations during the silicon crystal growth for photovoltaic applications are caused by thermal stresses. The temperatures along the boundaries of the silicon crystal ribbon are prescribed to meet the requirements of the crystal growth. It is shown that by allowing the temperatures to satisfy a second-order partial differential equation in the ribbon, all thermal stresses, and others induced by them, may be eliminated for the stress-free growth of the silicon crystal.

Description: An improved plate theory that accounts for the transverse shear deformation is presented, and mixed and displacement finite element models of the theory are developed. The theory is based on an assumed displacement field in which the inplane displacements are expanded in terms of the thickness coordinate up to the cubic term and the transverse deflection is assumed to be independent of the thickness coordinate. The governing equations of motion for the theory are derived from the Hamilton's principle. The theory eliminates the need for shear correction factors because the transverse shear stresses are represented parabolically. A mixed finite element model that uses independent approximations of the displacements and moments, and a displacement model that uses only displacements as degrees of freedom are developed. A comparison of the numerical results for bending with the exact solutions of the new theory and the three-dimensional elasticity theory shows that the present theory (and hence the finite element models) is more accurate than other plate-theories of the same order.

Description: Computational finite-element and boundary-element methods are reviewed, and their application to the mechanics of solids is discussed. Stability conditions for general FEMs are considered in addition to the use of least-order, stable, invariant, or hybrid/mixed isoparametric elements as alternatives to the displacement-based isoparametric elements. The use of symbolic manipulation, adaptive mesh refinement, transient dynamic response, and boundary-element methods for linear elaslticity and finite-strain problems of inelastic materials are also discussed.

Description: Methods for calculating sensitivity derivatives for discrete structural systems are surveyed, primarily covering literature published during the past two decades. Methods are described for calculating derivatives of static displacements and stresses, eigenvalues and eigenvectors, transient structural response, and derivatives of optimum structural designs with respect to problem parameters. The survey is focused on publications developed in nonstructural fields such as electronics, controls, and physical chemistry which are directly applicable to structural problems. Most notable among the nonstructural-based methods are the adjoint variable technique from control theory, and the Green's function and FAST methods from physical chemistry.

Description: A list of books, reports, periodicals, and conference proceedings, as well as individual papers, centered on specific aspects of fracture phenomenon has been compiled by the ASTM Committee E-24 on Fracture Testing. A list of basic references includes the articles on the development of fracture toughness, evaluation of stress intensity factors, fatigue crack growth, fracture testing, fracture of brittle materials, and fractography. Special attention is given to the references on application of fracture mechanics to new designs and on reevaluation of failed designs, many of them concerned with naval and aircraft structures.

Description: A standard stiffness matrix procedure which permits any combination of rigid, elastic, pinned or sliding connections of the degrees of freedom at the ends of a member to the nodes of its parent structure is described, in order to show how easily it can be extended to allow an existing algorithm to be used to ensure that no eigenvalues of the parent structure can be missed even when 'exact' member theory is used. The eigenvalues are the natural frequencies of undamped free vibration analyses or the critical load factors of buckling problems. The method preserves the exactness of the member theory and an efficient method for computer application is indicated. The theory also permits any combination of rigid, elastic, pinned or sliding connections between the freedoms of a substructure and those of its parent structure.

Description: A simple rectangular finite element was developed for two-dimensional analysis of laminated composite materials. The rectangular laminated composite element eliminates the need to add elements to a model simply to account for the material properties of various laminae. This is particularly advantageous for thick laminates with many lamina. Explicit integration in terms of generalized displacements minimizes the algebraic effort required to derive the element stiffness and the thermal load vector. A substitute shape function technique is used to improve the performance of the element in modeling bending type deformation. Results for several example problems are discussed.

Description: A new higher order theory has been proposed for the analysis of composite cylindrical shells. The formulation allows for arbitrary variation of inplane displacements. Governing equations are presented in the form of a hierarchy of sets of partial differential equations. Each set describes the shell behavior to a certain degree of approximation. The natural frequencies of simply-supported isotropic and laminated shells and stresses in a ring loaded composite shell have been determined to various orders of approximation and compared with three dimensional solutions. These numerical studies indicate the improvements achievable in estimating the natural frequencies and the interlaminar shear stresses in laminated composite cylinders.

Description: Two coupled, nonlinear differential equations are proposed for the modeling of the elastic and rate (time) dependent inelastic behavior of structural metals in the absence of recovery and aging. The structure of the model is close to the unified theories but contains essential differences. It is shown that the model reproduces almost elastic regions upon initial loading and in the unloading regions of the hysteresis loop. Under loading, unloading and reloading in strain control the model simulated the experimentally observed sharp transition from nearly elastic to inelastic behavior. When a formulation akin to existing unified theories is adopted the almost elastic regions reduce the points and the transition upon reloading is very gradual. For different formulations the behavior under sudden in(de)creases of the strain rate by two orders of magnitude is simulated by numerical experiments and differences are noted. The model represents cyclically neutral behavior and contains three constants and two positive, decreasing functions. The determination of constants and functions from monotonic loading with strain rate changes and relaxation periods is described.

Description: An experimental study of cracked-lap-shear specimens was conducted to determine the influence of adherend stacking sequence on debond initiation and damage growth in a composite-to-composite bonded joint. Specimens consisted of quasi-isotropic graphite/epoxy adherends bonded together with either FM-300 or EC 3445 adhesives. The stacking sequence of the adherends was varied such that 0 deg, 45 deg, or 90 deg plies were present at the adherend-adhesive interfaces. Fatigue damage initiated in the adhesive layer in those specimens with 0 deg and 45 deg interface plies. Damaage initiated in the form of ply cracking in the strap adherend for the specimens with 90 deg interface plies. The fatigue-damage growth was in the form of delamination within the composite adherends for specimens with the 90 deg and 45 deg plies next to the adhesive, while debonding in the adhesive resulted for the specimens with 0 deg plies next to the adhesive. Those joints with the 0 deg and 45 deg plies next to either adhesive has essentially the same fatigue-damage-initiation stress levels. These stress levels were 13 and 71 percent higher, respectively, than those for specimens with 90 deg plies next to the EC 3445 and FM-300 adhesives.

Description: Steady state harmonic stress waves in an isotropic elastic plate excited on one face by a circular transducer are analyzed theoretically. The transmitting transducer transforms an electrical voltage into a uniform normal stress at the top of the plate. To solve the boundary value problem,the radiation into a half-space is considered. The receiving transducer produces an electrical voltage proportional to the average spatially integrated normal stress over its face due to an incident wave. A numerical procedure is given to evaluate the frequency response at a receiving point due to a multiply reflected wave in the near field. Its stability and convergence are discussed. Parameterization plots which determine the particular wave whose frequency response has maximum magnitude compared with other multiple reflected waves are given for a range of values of dimensionless parameters. The effects of changes in the values of the parameters are discussed.

Description: Accelerometer data were collected during on-orbit structural dynamic testing of the Solar Array Flight Experiment aboard the Space Shuttle, and were analyzed at Lockheed Missile and Space Co. to determine the amount of damping present in the structure. The results of this analysis indicated that the damping present in the fundamental in-plane mode of the structure substantially exceeded that of the fundamental out-of-plane mode. In an effort to determine the source of the higher in-plane damping, a test was performed involving a small device known as a constant-force spring motor or constant-torque mechanism. Results from this test indicate that this discrete device is at least partially responsible for the increased in-plane modal damping of the Solar Array Flight Experiment structure.

Description: The objective of this program is to produce a series of new computer codes that permit more accurate and efficient three-dimensional inelastic structural analysis of combustor liners, turbine blades, and turbine vanes. Each code embodies a progression of mathematical models for increasingly comprehensive representation of the geometrical features, loading conditions, and forms of nonlinear material response that distinguish these three groups of hot section components.

Description: The accomplishments achieved under the isotropic creep-fatigue crack initiation life prediction program are summarized. A sizeable creep-fatigue crack initiation data base was generated on the nickel-base superalloy, B-1900. Companion constitutive modeling programs have also generated extensive data bases on the same heat of material. The crack initiation results have formed the basis of a new approach to creep-fatigue life prediction. The term Cyclic Damage Accumulation (CDA) was coined for the method, which was evaluated under isothermal, uniaxial conditions. Stringent laboratory verification experiments were used to test the accuracy of the method. Considering the quite limited material property data needed to evaluate the constants in the approach, the prediction accuracy is acceptable. At the expense of the larger data base required, Lewis developed total strain- strainrange partitioning method (TS-SRP) is capable of a higher degree of accuracy.

Description: Hot-section components of aircraft gas turbine engines are subjected to severe thermal structural loading conditions, especially during the startup and takeoff portions of the engine cycle. The most severe and damaging stresses and strains are those induced by the steep thermal gradients induced during the startup transient. These transient stresses and strains are also the most difficult to predict, in part because the temperature gradients and distributions are not well known or readily predictable and, in part, because the cyclic elastic-viscoplastic behavior of the materials at these extremes of temperature and strain are not well known or readily predictable. A broad spectrum of structures related technology programs is underway to address these deficiencies at the basic as well as the applied level. The three key program elements in the HOST structural analysis program are computations, constitutive modeling, and experiments for each research activity. Also shown are tables summarizing each of the activities.

Description: The objectives of the high-temperature structures program are threefold: to assist in the development of analytical tools needed to improve design analyses and procedures for the efficient and accurate prediction of the nonlinear structural response of hot-section components; to aid in the calibration, validation, and evaluation of the analytical tools by comparing predictions with experimental data; and to evaluate existing as well as advanced temperature and strain measurement instrumentation. As the analytical tools, test methods, tests, instrumentations, as well as data acquisition, management, and analysis methods are developed and evaluated, a proven, integrated analysis and experiment method will result in a more accurate prediction of the cyclic life of hot section components.

Description: In most high-temperature engineering applications, components are subjected to complex combinations of thermal and mechanical loading during service. A number of viscoplastic constitutive models were proposed which potentially can provide mathematical descriptions of material response under such conditions. Implementation of these models into large finite element codes such as MARC has already resulted in much improved inelastic analysis capability for hot-section aircraft engine components. However, a number of questions remain regarding the validity of methods adopted in characterizing these constitutive models for particular high-temperature materials. One area of concern is that the majority of experimental data available for this purpose are determined under isothermal conditions. This is in contrast to service conditions which, as noted above, almost always involve some form of thermal cycling. The obvious question arises as to whether a constitutive model characterized using an isothermal data base can adequately predict material response under thermomechanical conditions. An experimental program was initiated within the HOST program to address this particular concern. The results of the most recent isothermal and thermomechanical experiments are described.

Description: The natural frequencies and associated mode shapes for three thick open cantilevered cylindrical shells were determined both numerically and experimentally. The shells ranged in size from moderately to very thick with length to thickness ratios of 16, 8 and 5.6, the independent dimension being the shell thickness. The shell geometry is characterized by a circumferential angle of the 142 degrees and a ratio of length to inner radii arc length near 1.0. The finite element analysis was performed using NASTRAN's (COSMIC) triangular plate bending element CTRIA2, which includes membrane effects. The experimental results were obtained through holographic interferometry which enables one to determine the resonant frequencies as well as mode shapes from photographs of time-averaged holograms.

Description: NASTRAN and COPES/CONMIN were used in the early-stage design optimization of a propeller and shaft. The work was undertaken, in part, to assess the performance of these programs for such a task. While the optimization was successful, some drawbacks to the approach surfaced and are discussed.

Description: Using two commercial programs an application was developed to aid in generating a run-ready NASTRAN data deck on the PC. Macros are used to access relevant reference material and card files while editing the deck. The application can be easily customized to suit individual or group needs.

Description: An interface between Cosmic/NASTRAN and GIFTS has recently been released, combining the powerful pre- and post-processing capabilities of GIFTS with Cosmic/NASTRAN's analysis capabilities. The interface operates on a wide range of computers, even linking Cosmic/NASTRAN and GIFTS when the two are on different computers. GIFTS offers a wide range of elements for use in model construction, each translated by the interface into the nearest Cosmic/NASTRAN equivalent; and the options of automatic or interactive modelling and loading in GIFTS make pre-processing easy and effective. The interface itself includes the programs GFTCOS, which creates the Cosmic/NASTRAN input deck (and, if desired, control deck) from the GIFTS Unified Data Base, COSGFT, which translates the displacements from the Cosmic/NASTRAN analysis back into GIFTS; and HOSTR, which handles stress computations for a few higher-order elements available in the interface, but not supported by the GIFTS processor STRESS. Finally, the versatile display options in GIFTS post-processing allow the user to examine the analysis results through an especially wide range of capabilities, including such possibilities as creating composite loading cases, plotting in color and animating the analysis.

Description: Investigations were conducted on the 64-meter antenna hydrostatic bearing oil film thickness under a variety of loads and elastic moduli. These parametric studies used a NASTRAN pedestal structural model to determine the deflections under the hydrostatic bearing pad. The deflections formed the input for a computer program to determine the hydrostratic bearing oil film thickness. For the future 64-meter to 70-meter antenna extension and for the 2.2-meter (86-in.) haunch concrete replacement cases, the program predicted safe oil film thickness (greater than 0.13 mm (0.005 in.) at the corners of the pad). The effects of varying moduli of elasticity for different sections of the pedestal and the film height under stressed runner conditions were also studied.

Description: Some techniques for comparing structural vibration data determined from test and analysis are discussed. Orthogonality is a general category of one group, correlation is a second, synthesis is a third and matrix improvement is a fourth. Advantages and short-comings of the methods are explored with suggestions as to how they can complement one another. The purpose for comparing vibration data from test and analysis for a given structure is to find out whether each is representing the dynamic properties of the structure in the same way. Specifically, whether: mode shapes are alike; the frequencies of the modes are alike; modes appear in the same frequency sequence; and if they are not alike, how to judge which to believe.

Description: Differential stiffness as developed in NASTRAN is a linear change in stiffness caused by applied loads. Examples of differential stiffness are the stiffening effects of gravity forces in a pendulum, centrifugal forces in rotor blades and pressure loading of shell structures. In cases wherein this stiffness caused by a load is destabilizing, the differential stiffness concept lends itself to nonlinear structural analysis. Rigid Formats 4 (static analysis with differential stiffness) and 13 (normal modes with differential stiffness) are specifically designed to account for such stiffness changes. How pressure loading may be treated in these rigid formats is clarified. This clarification results from modal correlation of Ground Vibration Test (GVT) results from the empty and pressurized Filament Wound Case (FWC) quarter-scale Space Shuttle solid rocket booster (QSSRB). A sketch of the QSSRB cantilevered to the floor by its external tank attachments is shown.

Description: The IAC program provides the framework required for the development of an extensive multidisciplinary analysis capability. Several NASTRAN related capabilities were developed which can all be expanded in a routine manner to meet in-house unique needs. Plans are to complete the work discussed herein and to provide it to the engineering community through COSMIC. Release is to be after the current IAC Level 2 contract work on the IAC executive system is completed and meshed with the interfacing modules and analysis capabilities under development at the GSFC.

Description: Parametic investigations are performed using several two dimensional finite element formulations to determine their suitability for use in predicting extremum stresses in marine propellers. Comparisons are made of two NASTRAN elements (CTRIM6 and CTRAIA2) wherein elasticity properties have been modified to yield plane strain results. The accuracy of the elements is investigated by comparing finite element stress predictions with experimentally determined stresses in two classical cases: (1) tension in a flat plate with a circular hole; and (2) a filleted flat bar subjected to in-plane bending. The CTRIA2 element is found to provide good results. The displacement field from a three dimensional finite element model of a representative marine propeller is used as the boundary condition for the two dimensional plane strain investigations of stresses in the propeller blade and fillet. Stress predictions from the three dimensional analysis are compared with those from the two dimensional models. The validity of the plane strain modifications to the NASTRAN element is checked by comparing the modified CTRIA2 element stress predictions with those of the ABAQUS plane strain element, CPE4.

Description: Reviewed is the enhancement to NASTRAN program performed by NUK (Nippon Univac Kaisha, Ltd.) added to Level 15.5. Features discussed include intermediate checkpoint-restart in triangular decomposition, I/O improvement, multibanked memory and new plate element. The first three improvements provide the capability to solve significantly large size problems, while the new elements release the analyst from the cumbersome work to constrain the singularities caused by the lack of stiffness of inplane rotation of old plate elements.

Description: This work traces the response of a granular material via the Ten Coefficient Truesdell rate-type constituitive model into the simplest meaningful loading: the triaxial test configuration. A functional relation has been posed for computing the rather peculiar relation between average applied stress and average porosity. Using that relation an attack has been mounted on the dilemma that exists between dynamic and constitutive use of the pressure variable; that is relating dynamic pressure, thermodynamic pressure, stress deviator and higher stress invariants. The resolution was as a linear superposition with a one-way feedback, in that while the dynamic component could not effect the constituitive component, the converse was not true since density appears in the momentum transport relation.

Description: It was found that the sampling rates from flight recorders on a remotely piloted transport aircraft that crashed into the ground were too low, although they were higher than those required now by regulations. For example, the sampling rate for roll angle was one per second. The sampling rate for normal acceleration was also fairly low. Existing regulations require only 4 samples per second; researchers had 16 samples per second. Some data was lost during the initial impact. The frequency response data was adequate.

Description: The operation of a radio beacon position locator during and after the remotely controlled transport aircraft is discussed. The radio beacon transmission was actuated and was picked up by the Navy P-3A chase aircraft for a short time, after which reception was lost. The pilot reported that he received a signal on both 121.5 MHz and 243 MHz for a period of approximately 5 seconds. Five minutes after the crash a portable direction finding unit located on the roof of the NASA Dryden Flight Research Facility, 4 miles distant from the crash, was unable to pick up the beacon transmission. The fire crews started fighting the fires approximately 90 seconds after the time of impact. Navy personnel access to the crash site was allowed on the morning of December 2, 1984. Radio beacon locator was found resting top side up, 15 feet forward and 13 feet perpendicular from the tray location the starboard side of the aircraft. An immediate inspection indicated the airfoil suffered moderate fire damage with paint peeling but not intumescing. The visual marker strobe lamp housings were intact but extensively burned such that it was impossible to see if the lamps had survived. The airfoil suffered minor structural damage, with assorted dents, etc. The extended plunger on the ARU-21 release unit indicated that the pyrotechnic deployment system operated. The radio beacon base (tray) suffered some heat and fire damage, and was charred and blackened by smoke. The frangible switch in the nose survived and the switch in the belly was recovered and found to have actuated. It is assumed that this switch fired the ARU-21 squib. There were no other release switches installed in the normally open system in the aircraft.

Description: Results of the crash test of a remotely piloted transport aircraft instrumented to measure a NASA energy-absorbing transport seat are given. Human tolerance limits to acceleration and a dynamic response index model are discussed. It was found that the acceleration levels at the rear of the airplane were quite low and were below the stroking threshold of the NASA EA-seat. Therefore, dummies in the standard and EA-seat responded approximately the same. All longitudinal accelerations were quite low for the primary impact with very low forces measured in the lap belts. The vertical (spineward) acceleration levels measured in the dummies were also relatively low and very survivable from an impact tolerance standpoint. The pilot with an 18 G peak acceleration received by far the highest vertical acceleration and could have possibly received slight spinal injury.

Description: The FAA restraint system experiments consisted of 24 standard and modified seats, 2 standard galleys and 2 standard overhead compartments. Under the controlled impact demonstration (CID) program, the experimental objective was to demonstrate the effectiveness of individual restraint system designs when exposed to a survivable air-to-ground impact condition. What researchers were looking for was the performance exhibited by standard and modified designs, performance differences resulting from their installed cabin location, and interrelating performance demonstrated by test article and attaching floor and/or fuselage structure. The other restraint system experiment consisted of 2 standard overhead stowage compartments and 2 galley modules. Again, researchers were concerned with the retention of stowed equipment and carry-on articles. The overhead compartments were loaded with test weights up to their maximum capacity, and each of the galleys was filled with test articles: aft with normal galley equipment, forward with hazardous material test packages. A breakdown of instrumentation and distribution is given beginning with 11 instrumented type anthropomorphic dummies and 185 sensors which provided for acceleration and load measurements at the various experiment and associated structure locations. The onboard cameras provided additional coverage of these experiments, including the areas of cabin which were not instrumented. Test results showing the window-side leg forces versus pulse duration are given.

Description: The testing conducted on LaRC-developed hardware for the controlled impact demonstration transport aircraft is discussed. To properly develop flight qualified crash systems, two environments were considered: the aircraft flight environment with the focus on vibration and temperature effects, and the crash environment with the long pulse shock effects. Also with the large quantity of fuel in the wing tanks the possibility of fire was considered to be a threat to data retrieval and thus fire tests were included in the development test process. The aircraft test successfully demonstrated the performance of the LaRC developed heat shields. Good telemetered data (S-band) was received during the impact and slide-out phase, and even after the aircraft came to rest. The two onboard DAS tape recorders were protected from the intense fire and high quality tape data was recovered. The complete photographic system performed as planned throughout the 40.0 sec of film supply. The four photo power distribution pallets remained in good condition and all ten onboard 16 mm high speed (400 frames/sec) cameras produced good film data.

Description: The high-environment flight instrumentation system was designed to acquire Langley's structural response data during the full scale transport-controlled impact demonstration test. There was only one opportunity for data acquisition. Thus, a high reliability and crashworthy design approach was implemented. The approach featured multi-level redundancy and a vigorous quality assurance testing program. Complying with an accelerated schedule, the instrumentation system was developed, tested and shipped within 18 months to Dryden Flight Research Facility. The flight instrumentation system consists of two autonomous data systems, DAS #1 and #2, and an excellent checkout subsystem. Each data system is partitioned into four pallets. The system was designed to operate on manned and unmanned flights. There are 176 data channels per data system. These channels are sequentially sampled and encoded into 1 megabit/sec pulse code modulation (PCM) data signal. To increase the probability of success, a special PCM distribution subsystem was developed. This subsystem distributes the PCM signal to two transmitters, one delay memory, and eight recorder tracks. The data on four of these trackes was digitally delayed approximately 300 msec to maximize data acquisition during impact. Therefore each data system's data is redundantly recorded onboard and on the ground. There are two time code generators. Parallel time from each is encoded into both data systems. Serial time from each is redundantly recorded on both onboard recorders. Instrumentation power is independent of aircraft power and self-contained.

Description: The calibration of the KRASH and DYCAST models for transport aircraft is discussed. The FAA uses computer analysis techniques to predict the response of controlled impact demonstration (CID) during impact. The moment bridges can provide a direct correlation between the predictive loads or moments that the models will predict and what was experienced during the actual impact. Another goal is to examine structural failure mechanisms and correlate with analytical predictions. The bending bridges did achieve their goals and objectives. The data traces do provide some insight with respect to airframe loads and structural response. They demonstrate quite clearly what's happening to the airframe. A direct quantification of metal airframe loads was measured by the moment bridges. The measured moments can be correlated with the KRASH and DYCAST computer models. The bending bridge data support airframe failure mechanisms analysis and provide residual airframe strength estimation. It did not appear as if any of the bending bridges on the airframe exceeded limit loads. (The observed airframe fracture was due to the fuselage encounter with the tomahawk which tore out the keel beam.) The airframe bridges can be used to estimate the impact conditions and those estimates are correlating with some of the other data measurements. Structural response, frequency and structural damping are readily measured by the moment bridges.

Description: From a total of 351 instrumentation channels, 341 channels (97%) were in operation during the initial impact of the airplane. Both NASA seats, the energy absorbing seat and the standard seat, maintained their integrity during the impact. The floor accelerations at the seat locations were lower than the accelerations required for the energy absorbers to stroke; consequently, the energy absorbing seat did not stroke. The two seats remained firm in place during the crash and no seat attachment failures were observed. Due to the low accelerations experienced during the crash, both seats performed as standard seats. In the airplane structure, the accelerations were higher at both the point of impact in the left wing and at the forward end of the fuselage. The accelerations on the floor were higher toward the front than toward the rear and the floor accelerations on the left side were higher than on the right side at the front of the fuselage, but toward the rear they evened out.

Description: The planned versus the actual results of the controlled impact demonstration of a transport aircraft are discussed. Remote control systems, site selection, manned flight tests, and wreckage distribution are discussed.

Description: On December 1, 1984, NASA and the Federal Aviation Administration (FAA) conducted the first remotely piloted air-to-ground crash test of a transport category aircraft. The Full-Scale Transport Controlled Impact Demonstration (CID) was the culmination of 4 years of effort by the two agencies. NASA and the FAA had many objectives during the joint planning and execution of the Controlled Impact Demonstration. The structural loads experiment was very successful. Ninety-seven percent of the channels were active at impact. The data is still being assessed. Only a portion of the data is presented here; approximately 80 channels of data are available. Analysis of the remaining data is in progress. Interior photography was also very successful. One hundred percent of the cameras functioned. The film contains unique information on the development of fire and smoke in the interior of the aircraft. From a human tolerance point of view, the CID was simulation of a survivable crash.

Description: The improved cost effectiveness using better models, more accurate and faster algorithms and large scale computing offers more representative dynamic analyses. The band Lanczos eigen-solution method was implemented in Rockwell's version of 1984 COSMIC-released NASTRAN finite element structural analysis computer program to effectively solve for structural vibration modes including those of large complex systems exceeding 10,000 degrees of freedom. The Lanczos vectors were re-orthogonalized locally using the Lanczos Method and globally using the modified Gram-Schmidt method for sweeping rigid-body modes and previously generated modes and Lanczos vectors. The truncated band matrix was solved for vibration frequencies and mode shapes using Givens rotations. Numerical examples are included to demonstrate the cost effectiveness and accuracy of the method as implemented in ROCKWELL NASTRAN. The CRAY version is based on RPK's COSMIC/NASTRAN. The band Lanczos method was more reliable and accurate and converged faster than the single vector Lanczos Method. The band Lanczos method was comparable to the subspace iteration method which was a block version of the inverse power method. However, the subspace matrix tended to be fully populated in the case of subspace iteration and not as sparse as a band matrix.

Description: The salient aspects of the theoretical modeling of a conventional triaxial test (CTC) of a cohesionless granular medium with stress and strain rate loading are described. Included are a controllable gravitational body force and provision for low confining pressure and/or very low intergranular stress. The modeling includes rational, analytic, and numerical phases, all in various stages of development. The numerical evolutions of theoretical models will be used in final design stages and in the analysis of the experimental data. In this the experimental design stage, it is of special interest to include in the candidate considerations every anomaly found in preliminary terrestrial experimentation. Most of the anomalies will be eliminated by design or enhanced for measurement as the project progresses. The main aspect of design being not the physical apparatus but the type and trajectories of loading elected. The major considerations that have been treated are: appearance and growth of local surface aberrations, stress-power coefficients, strain types, optical strain, radial bead migration, and measures of rotation for the proper stress flux.

Description: Under normal one-g conditions immiscible alloys segregate extensively during solidification due to sedementation of the more dense of the immiscible liquid phases. However, under low-g conditions it should be possible to form a dispersion of the two immiscible liquids and maintain this dispersed structure during solidification. Immiscible (hypermonotectic) gold-rhodium alloys were processed in the Marshall Space Flight Center 105 meter drop tube in order to investigate the influence of low gravity, containerless solidification on their microstructure. Hypermonotectic alloys composed of 65 atomic % rhodium exhibited a tendency for the gold rich liquid to wet the outer surface of the containerless processed samples. This tendency led to extensive segregation in several cases. However, well dispersed microstructures consisting of 2 to 3 micron diameter rhodium-rich spheres in a gold-rich matrix were produced in 23.4 atomic % rhodium alloys. This is one of the best dispersions obtained in research on immiscible alloy-systems to data.

Description: The loads analysis for the design of the Galileo structure was done by combining the transient analysis method with methods aimed at reducing cost and schedule impact while increasing confidence in the estimation of the loads. The preliminary sizing was done by the mass acceleration curve. The loads iterations were done by the generalized shock spectra method. Only a few critical subsystems required a coupled transient analysis to show positive margins. Dynamic coupling between the launch vehicle and the spacecraft was done by analytically removing and adding a spacecraft to a given launch vehicle at the modal level, without re-solving the eigenvalue problem. The modal response resulting from the dynamic coupling between the launch vehicle and the spacecraft was used only for the 28 Galileo modes below 40 Hz. In the absence of definition of the Shuttle/Centaur model and the forcing functions above 40 Hz, the mass acceleration curve was used to generate modal bounds between 40 and 80 Hz.

Description: The stress wave factor (SWF) signal is utilized for the nondestructive evaluation of plates containing perpendicular edge cracks. The effects of the existence lateral location and depth of the crack on the magnitude spectra of individual reflections in the SWF signal are studied. If the reflections in the SWF signal are not overlapped the short time Fourier analysis is applied. If the reflections are overlapped the short time homomorphic analysis (cepstrum analysis) is applied. Several reflections which have average resonant frequencies approximately at 0.9, 1.3, and 1.7 MHz are analyzed. It is observed that the magnitude ratios evaluated at average resonant frequencies decrease more with increasing d/h if the crack is located between the transducers, where h is plate thickness and d is crack depth. Moreover, for the plates, crack geometries, reflections, and frequencies considered, the average decibel drop depends mainly on the dimensionless parameter d/h and it is approximately -1 dB per 0.07 d/h. Changes in the average resonant frequencies of the magnitude spectra are also observed due to changes in the location of the crack.

Description: A failure analysis was performed to establish the cause of a 2014-T6 aluminum compressor rotor blade failure in a large NASA-Ames wind tunnel. Metallurgical failure analysis by light photography, fractographic SEM examinations, and fatigue experiments showed that a 0.13-mm-deep scratch in the shank of the blade had acted as an initiation site for a fatigue crack, which subsequently grew by Stage II fatigue before occurrence of the final fracture by unstable crack growth. Studies of modal deformation plots and blade FEM analysis indicated that Mode-3 was the most likely mode responsible for the resonance condition which produced the vibration loading that effected crack growth. Tunnel measurements data showed that the crack growth was controlled by both the tunnel resonance and the dampening from the crack itself.

Description: Recent developments in computational structural mechanics are reviewed with reference to computational needs for future structures technology, advances in computational models for material behavior, discrete element technology, assessment and control of numerical simulations of structural response, hybrid analysis, and techniques for large-scale optimization. Research areas in computational structural mechanics which have high potential for meeting future technological needs are identified. These include prediction and analysis of the failure of structural components made of new materials, development of computational strategies and solution methodologies for large-scale structural calculations, and assessment of reliability and adaptive improvement of response predictions.

Description: The stiffness-matrix formulation for the rectangular finite element described by Melosh (1963) and Weaver and Johnston (1984) is generalized to orthotropic materials with material axes not necessarily coincident with the x and y axes; i.e., the condition d(13) = d(23) = 0 is removed. Also included are explicit expressions for the element load vector associated with nonuniform temperature increase in the element. Applications to the analysis of thermal stresses in thin Si-crystal ribbons subjected to temperature changes with highly nonuniform lengthwise and transverse gradients (Utku et al., 1986) and to the simulation of the thermoviscoelastic behavior of growing Si ribbons (Utku and Ray, 1986) are indicated.

Description: An analytical model is presented for the surface displacement associated with two symmetric cracks radiating from a circular hole in a plate undergoing remote uniform stress. The model equations account for the stress intensity factor (Kic), the boundary conditions enveloping the hole, and the effects of symmetric forces concentrated around the hole. The displacement(s) of the surface are calculated at a given distance from the hole along the crack(s).

Description: A combined experimental and analytical study was conducted to determine the behavior of a graphite/epoxy laminate subjected to combined bearing and bypass loading. Single-fastener quasi-isotropic specimens were loaded at various bearing-bypass ratios until damage was produced at the fastener hole. Damage-onset strengths and damage modes were then analyzed using local hole-boundary stresses calculated by a finite-element analysis. The tension data showed the expected linear interaction for combined bearing and bypass loading with damage developing in the net-section tension mode. However, the compression bearing-bypass strength showed an unexpected interaction involving the bearing mode.

Description: A finite-element analysis of the end-notch flexure specimen for Mode II interlaminar fracture toughness measurement was conducted. The effects of friction between the crack faces and large deflection on the evaluation of G(IIc) from this specimen were investigated. Results of this study are presented in this paper.

Description: First two equations of equilibrium are utilized to compute the transverse shear stress variation through thickness of a thick laminated plate after in-plane stresses have been computed using an assumed quadratic displacement triangular element based on transverse inextensibility and layerwise constant shear angle theory (LCST). Centroid of the triangle is the point of exceptional accuracy for transverse shear stresses. Numerical results indicate close agreement with elasticity theory. An interesting comparison between the present theory and that based on assumed stress hybrid finite element approach suggests that the latter does not satisfy the condition of free normal traction at the edge. Comparison with numerical results obtained by using constant shear angle theory suggests that LCST is close to the elasticity solution while the CST is closer to classical (CLT) solution. It is also demonstrated that the reduced integration gives faster convergence when the present theory is applied to a thin plate.

Description: A Mode II test specimen was developed which has potential application in understanding phemonena associated with mixed mode fatigue failures in high performance aircraft engine bearing races. The attributes of the specimen are: it contains one single ended notch, which simplifies data gathering and reduction; the fatigue crack grows in-line with the direction of load application; a single axis test machine is sufficient to perform testing; and the Mode I component is vanishingly small.

Description: Experiments were conducted to study the growth behavior of surface fatigue cracks in the circumferential plane of solid and hollow cylinders. In the solid cylinders, the fatigue cracks were found to have a circular arc crack front with specific upper and lower limits to the arc radius. In the hollow cylinders, the fatigue cracks were found to agree accurately with the shape of a transformed semiellipse. A modification to the usual nondimensionalization expression used for surface flaws in flat plates was found to give correct trends for the hollow cylinder problem.

Description: Ceramic research at the University of Washington, brittle materials design program, and NASA program research at the University of Washington are outlined.

Description: The application of fracture mechanics to the design of ceramic structures will require the precise measurement of crack growth and fracture resistance of these materials over their entire range of anticipated service temperatures and standardized test methods for making such measurements. The development of a standard test for measuring the plane strain fracture toughness is sought. Stress intensity factor coefficients were determined for three varieties of chevron-notch specimens, and fracture toughness measurements were made on silicon nitrides, silicon carbides, and aluminum oxides to assess the performance of each specimen variety. It was determined that silicon nitride and silicon carbides have flat crack growth resistance curves, but aluminum oxide does not. Additionally, batch-to-batch differences were noticed for the aluminum oxide. Experiments are continuing to explain the rising crack growth resistance and batch-to-batch variations for the aluminum oxide.

Description: Modal identification of seemingly simple structures, such as the generic truss is often surprisingly difficult in practice due to high modal density, nonlinearities, and other nonideal factors. Under these circumstances, different data analysis techniques can generate substantially different results. The initial application of a new hybrid-data method for studying the performance characteristics of various identification techniques with such data is summarized. This approach offers new pieces of information for the system identification researcher. First, it allows actual experimental data to be used in the studies, while maintaining the traditional advantage of using simulated data. That is, the identification technique under study is forced to cope with the complexities of real data, yet the performance can be measured unquestionably for the artificial modes because their true parameters are known. Secondly, the accuracy achieved for the true structural modes in the data can be estimated from the accuracy achieved for the artificial modes if the results show similar characteristics. This similarity occurred in the study, for example, for a weak structural mode near 56 Hz. It may even be possible--eventually--to use the error information from the artificial modes to improve the identification accuracy for the structural modes.

Description: All of the structures considered for the Control of Flexible Structures (COFS) flight experiments are deployable truss structures and their response will be dominated by the structural response of the joints. To prepare for these experiments some fundamental research work is being conducted in the Structures and Dynamics Division at LaRC which will provide insight into structurally efficient and predictable deployable truss joints. This work involves generic studies of the static and dynamic response of joints as well as the development of analytical models which can be used to predict the response of a large multijointed truss. In addition to the generic joint studies, the research effort encompasses the design and fabrication of a 20-meter long deployable truss beam for laboratory evaluation of its structural characteristics and correlation with developed prediction methods. The experimental results have indicated the importance of attention to detail in the design and fabrication of joints for deployable truss structures. The dimensional relations and material considerations for efficient pin-clevis joints have been outlined. Results of tests on the near-center latch are discussed.

Description: COFS 3 is the third project within the Control of Flexible Structures (COFS) program. It deals with developing multibody dynamics and control technology for large space structures. It differs from COFS 1 and 2 in two respects. First, it addresses a more complex class of structure, and second it is basically a scale model ground test and analysis program while COFS 1 and 2 feature Shuttle flight experiments. The specific technology thrusts within COFS 3 are model sensitivities, test methods, analysis validation, systems identification, and vibration suppression. The COFS 3 project will develop the methods for using dynamically scaled models and analysis to predict the structural dynamics of large space structures. The project uses the space station as a focus because it is typical of the structures of interest and provides the first opportunity to obtain full-scale on-orbit dynamics data.

Description: The integration process of the MAST flight system is surveyed. Insight is given into the planned orbital experiment process. The data flow necessary to support the flight operation is outlined.

Description: The MAST Flight System as a test bed for large space structure control algorithms is discussed. An overview is given of the control system architecture. The actuators, the sensors, the control computer, and the baseline damping algorithm are discussed.

Description: An overall understanding of the beam assembly and data with which potential experimenters can begin to conduct analyses relevant to their experiments is given. Data is given on the beam structural concept, the tip remote station layout, the intermediate remote station layout with and without actuators, beam element materials, equivalent beam characteristics, beam element properties, remote station mass properties, and MAST Flight System modal characteristics.

Description: A procedure for determining the scattered pressure field resulting from a monochromatic harmonic wave that is incident upon a layer energy absorbing structure is treated. The situation where the structure is modeled with finite elements and the surrounding acoustic medium (water or air) is represented with either acoustic finite elements, or some type of boundary integral formulation, is considered. Finite element modeling problems arise when the construction of the structure, at the fluid structure interface, are nonhomogeneous and in particular, when the inhomogeneities are small relative to the acoustic wave length. An approximate procedure is presented for replacing the detailed microscopic representation of the layered surface configuration with an equivalent simple surface impedance finite element, which is especially designed to work only at limited frequencies. An example problem is presented using NASTRAN. However, the procedure is general enough to adapt to practically any finite element code having a steady state option.

Description: Several very low aspect ratio flat plate wing configurations are analyzed for their aerodynamic instability (flutter) characteristics. All of the wings investigated are delta planforms with clipped tips, made of aluminum alloy plate and cantilevered from the supporting vehicle body. Results of both subsonic and supersonic NASTRAN aeroelastic analyses as well as those from another version of the program implementing the supersonic linearized aerodynamic theory are presented. Results are selectively compared with the experimental data; however, supersonic predictions of the Mach Box method in NASTRAN are found to be erratic and erroneous, requiring the use of a separate program.

Description: The general purpose Finite Element Program COSMIC NASTRAN currently has the ability to model magnetic circuits with constant permeablilities. An approach was developed which, through small modifications to the program, allows modelling of non-linear magnetic devices including soft magnetic materials, permanent magnets and coils. Use of the NASTRAN code resulted in output which can be used for subsequent mechanical analysis using a variation of the same computer model. Test problems were found to produce theoretically verifiable results.

Description: Today, the increased use of Finite Element Analysis coupled with computer graphics interfaces presents the analyst with many alternatives. The drive toward automation has resulted in the creation of many modeling and post-processing systems that are in use today throughout the world. Many of these systems exist as islands of technology. The need for fully integrated systems is being met by many companies today.

Description: Several new capabilities dealing with the dynamic and aeroelastic analyses of turbosystems have been added as standard features to the April 1986 release of NASTRAN. A brief description is given of these capabilities and their implementation in NASTRAN are outlined.

Description: Improvements that were made to the COSMIC NASTRAN elements CIHEX1 and QDMEM1 are described. These elements are isoparametric representations of solid and membrane elastic behavior. Recent papers by the authors have shown the official COSMIC versions of these elements to be inferior to those available in the MacNeal-Schwendler Corporation (MSC) version of NASTRAN in that they are overly stiff for some loadings. Modifications were made to these elements which reduce the order of integration for shear terms and, for the eight-mode solid element, add additional strain functions. The resulting element formulations give behavior similar to that of the MSC elements. The changes made in the element formulations are discussed and results of test problems are compared with results from the official COSMIC elements and with the MSC elements.

Description: Langley Research Center (LaRC) was responsible for the design, manufacture, and integration of all hardware required for the photographic system used to film the interior of the controlled impact demonstration (CID) B-720 aircraft during actual crash conditions. Four independent power supplies were constructed to operate the ten high-speed 16 mm cameras and twenty-four floodlights. An up-link command system, furnished by Ames Dryden Flight Research Facility (ADFRF), was necessary to activate the power supplies and start the cameras. These events were accomplished by initiation of relays located on each of the photo power pallets. The photographic system performed beyond expectations. All four power distribution pallets with their 20 year old Minuteman batteries performed flawlessly. All 24 lamps worked. All ten on-board high speed (400 fps) 16 mm cameras containing good resolution film data were recovered.

Description: According to preliminary examination of the data, out of 179 data channels that were onboard the aircraft in support of the seat experiments, there is data from 168. There was somewhat more severe environment imposed in the structure of the obstacles than by the ground impact. Therefore, both ground impact and obstacle impact are of interest for crashworthiness experiments. Most of the data channels that were studied are fairly consistent with the physical evidence: they show acceleration levels that are reasonable, and in many cases these integrate out to a reasonable velocity change. Finally, from observation thus far, the ground impact did not fail or significantly damage any seat. Nor did any of the energy absorbers in the modified seats extend. The accelerations do not appear high enough and/or energetic enough to cause this to happen. At this time, the onboard films have not been studied; only some videotape have been viewed. Some of the seats were so badly damaged by the fire that any failures which might have occurred were obscured. A close examination of the onboard films using a stop-action projector will allow a more thorough evaluation.

Description: Papers are presented on cracking at nozzle corners in the nuclear pressure vessel industry, applied fracture mechanics for assessing defect significance in a crude oil pipeline, failure analysis of a large wind-tunnel compressor blade, analysis of a compressor-wheel failure, and preventing fracture by inspection and analysis. Consideration is also given to the fatigue crack growth predictions of welded aircraft structures containing flaws in the residual stress field, the fatigue and fracture mechanics analysis of a compression loaded aircraft structure, fracture of an aircraft horizontal stabilizer, fatigue life analysis of fuel tank skins under combined loads, and aircraft structural maintenance recommendations based on fracture mechanics analysis. Additional papers discuss an analysis of two metal-forming die failures, an analysis of a failed saw arbor, and the role of fracture mechanics in assessing the effect on fatigue life of design changes in welded fabrications.

Description: The convergence properties of three modal methods for solving structural transient response (STR) problems are examined by comparing the accuracy of the total number of modes used to represent the responses. The goal is to identify a method which involves as few degrees of freedom as possible. Two mode superposition methods, the mode displacement method (MDM) and the mode acceleration method (MAM), are fitted with equivalent expressions for damped response. An error norm is employed to compare the accuracies of the methods in calculating displacements, moments, and shear forces for a viscously-damped cantilevered beam experiencing several dynamic loading conditions and levels of damping. The MAM converges quicker and uses fewer dof, mainly because it has a pseudo-static response term which accounts for some of the flexibility of higher modes. A third method, which integrates the convolution integral iteratively over time, generates a higher order approximation to the STR with a reduced number of modes. It converges fastest among the three, but requires definition of an appropriate forcing function. Finally, the MAM and MDM models are applied to study the effects of the forcing frequency and the initial conditions.

Description: The present conference gives attention to topics in the application of fracture mechanics, subcritical crack growth phenomena, fracture testing methods, ductile fracture behavior, and fracture mechanisms and their analysis. Specific papers treat the resistance curve approach to composite materials characterization, fracture toughness in ductile iron and cast steel, hold-time effects in elevated temperature fatigue crack propagation, creep crack growth under nonsteady conditions, viscoplastic fatigue in a superalloy at elevated temperatures, fracture testing with arc bend specimens, one-point bend impact test application, and a compact mode II fracture specimen. Also discussed are the computation of stable crack growth using the J-integral, the use of plastic energy dissipation to characterize crack growth, the extension of surface cracks under cyclic loading, the minimum time criterion for crack instability in structural materials, dynamic crack propagation and branching under biaxial loading, and boundary layer effects in cracked bodies.

Description: The present paper describes an approach for the optimum sizing of single and joined wing structures that is based on representing the built-up finite element model of the structure by an equivalent beam model. The low order beam model is computationally more efficient in an environment that requires repetitive analysis of several trial designs. The design procedure is implemented in a computer program that requires geometry and loading data typically available from an aerodynamic synthesis program, to create the finite element model of the lifting surface and an equivalent beam model. A fully stressed design procedure is used to obtain rapid estimates of the optimum structural weight for the beam model for a given geometry, and a qualitative description of the material distribution over the wing structure. The synthesis procedure is demonstrated for representative single wing and joined wing structures.

Description: The solution of several elasticity problems, and particularly crack problems, can be reduced to the solution of one-dimensional singular integral equations with a Cauchy-type kernel or to a system of uncoupled singular integral equations. Here a method for the numerical solution of random singular integral equations of Cauchy type is presented. The solution technique involves a Chebyshev series approximation, the coefficients of which are the solutions of a system of random linear equations. This method is applied to the problem of periodic array of straight cracks inside an infinite isotropic elastic medium and subjected to a nonuniform pressure distribution along the crack edges. The statistical properties of the random solution are evaluated numerically, and the random solution is used to determine the values of the stress-intensity factors at the crack tips. The error, expressed as the difference between the mean of the random solution and the deterministic solution, is established. Values of stress-intensity factors at the crack tip for different random input functions are presented.