Salinas provides a massively parallel implementation of structural dynamics finite element analysis, required for high fidelity, validated models used in modal, vibration, static and shock analysis of weapons systems. This document provides a users guide to the input for Salinas. Details of input specifications for the different solution types, output options, element types and parameters are included. The appendices contain detailed examples, and instructions for running the software on parallel platforms.
In an attempt to automatically produce high-quality all-hex meshes, we investigated a mesh improvement strategy: given an initial poor-quality all-hex mesh, we iteratively changed the element connectivity, adding and deleting elements and nodes, and optimized the node positions. We found a set of hex reconnection primitives. We improved the optimization algorithms so they can untangle a negative-Jacobian mesh, even considering Jacobians on the boundary, and subsequently optimize the condition number of elements in an untangled mesh. However, even after applying both the primitives and optimization we were unable to produce high-quality meshes in certain regions. Our experiences suggest that many boundary configurations of quadrilaterals admit no hexahedral mesh with positive Jacobians, although we have no proof of this.
Predicting the behavior of ion channel proteins is important for understanding biological effects of drugs and toxins. These problems involve steady state transport of ions through very small (1-2 atoms wide) pores. FY99 LDRD funding was used to begin investigations of ion channel proteins using a molecular theory approach. Much of our efforts involved establishing the soundness of the approach by direct comparison with grand canonical molecular dynamics simulations of simple model systems. In addition, several dimensional ion channel models have been implemented to demonstrate the viability of the approach, The seed funding provided by this LDRD grant resulted in 50K of DOWOBER funds for FY99, an invitation to submit a full length 0(500K) proposal for consideration to DOWOBER, and start a larger LDRD effort in computational biophysics beginning in FY00.
Direct metal deposition technologies produce complex, near net shape components from Computer Aided Design (CAD) solid models. Most of these techniques fabricate a component by melting powder in a laser weld pool, rastering the weld bead to form a layer, and additively constructing subsequent layers. This report will describe anew direct metal deposition process, known as WireFeed, whereby a small diameter wire is used instead of powder as the feed material to fabricate components. Currently, parts are being fabricated from stainless steel alloys. Microscopy studies show the WireFeed parts to be filly dense with fine microstructural features. Mechanical tests show stainless steel parts to have high strength values with retained ductility. A model was developed to simulate the microstructural evolution and coarsening during the WireFeed process. Simulations demonstrate the importance of knowing the temperature distribution during fabrication of a WireFeed part. The temperature distribution influences microstructural evolution and, therefore, must be controlled to tailor the microstructure for optimal performance.
Current supercomputers use large parallel arrays of tightly coupled processors to achieve levels of performance far surpassing conventional vector supercomputers. Shock-wave physics codes have been developed for these new supercomputers at Sandia National Laboratories and elsewhere. These parallel codes run fast enough on many simulations to consider using them to study the effects of varying design parameters on the performance of models of conventional munitions and other complex systems. Such studies maybe directed by optimization software to improve the performance of the modeled system. Using a shaped-charge jet design as an archetypal test case and the CTH parallel shock-wave physics code controlled by the Dakota optimization software, we explored the use of automatic optimization tools to optimize the design for conventional munitions. We used a scheme in which a lower resolution computational mesh was used to identify candidate optimal solutions and then these were verified using a higher resolution mesh. We identified three optimal solutions for the model and a region of the design domain where the jet tip speed is nearly optimal, indicating the possibility of a robust design. Based on this study we identified some of the difficulties in using high-fidelity models with optimization software to develop improved designs. These include developing robust algorithms for the objective function and constraints and mitigating the effects of numerical noise in them. We conclude that optimization software running high-fidelity models of physical systems using parallel shock wave physics codes to find improved designs can be a valuable tool for designers. While current state of algorithm and software development does not permit routine, ''black box'' optimization of designs, the effort involved in using the existing tools may well be worth the improvement achieved in designs.
A set of linear and nonlinear stability analysis tools have been developed to analyze steady state incompressible flows in 3D geometries. The algorithms have been implemented to be scalable to hundreds of parallel processors. The linear stability of steady state flows are determined by calculating the rightmost eigenvalues of the associated generalize eigenvalue problem. Nonlinear stability is studied by bifurcation analysis techniques. The boundaries between desirable and undesirable operating conditions are determined for buoyant flow in the rotating disk CVD reactor.
Predicting the properties of nonequilibrium systems from molecular simulations is a growing area of interest. One important class of problems involves steady state diffusion. To study these cases, a grand canonical molecular dynamics approach has been developed by Heffelfinger and van Swol [J. Chem. Phys., 101, 5274 (1994)]. With this method, the flux of particles, the chemical potential gradients, and density gradients can all be measured in the simulation. In this paper, we present a complementary approach that couples a nonlocal density functional theory (DFT) with a transport equation describing steady-state flux of the particles. We compare transport-DFT predictions to GCMD results for a variety of ideal (color diffusion), and nonideal (uphill diffusion and convective transport) systems. In all cases excellent agreement between transport-DFT and GCMD calculations is obtained with diffusion coefficients that are invariant with respect to density and external fields.
The emerging field of haptics represents a fundamental change in human-computer interaction (HCI), and presents solutions to problems that are difficult or impossible to solve with a two-dimensional, mouse-based interface. To take advantage of the potential of haptics, however, innovative interaction techniques and programming environments are needed. This paper describes FGB (FLIGHT GHUI Builder), a programming tool that can be used to create an application specific graphical and haptic user interface (GHUI). FGB is itself a graphical and haptic user interface with which a programmer can intuitively create and manipulate components of a GHUI in real time in a graphical environment through the use of a haptic device. The programmer can create a GHUI without writing any programming code. After a user interface is created, FGB writes the appropriate programming code to a file, using the FLIGHT API, to recreate what the programmer created in the FGB interface. FGB saves programming time and increases productivity, because a programmer can see the end result as it is created, and FGB does much of the programming itself. Interestingly, as FGB was created, it was used to help build itself. The further FGB was in its development, the more easily and quickly it could be used to create additional functionality and improve its own design. As a finished product, FGB can be used to recreate itself in much less time than it originally required, and with much less programming. This paper describes FGB's GHUI components, the techniques used in the interface, how the output code is created, where programming additions and modifications should be placed, and how it can be compared to and integrated with existing API's such as MFC and Visual C++, OpenGL, and GHOST.
Traditionally law enforcement agencies have relied on basic measurement and imaging tools, such as tape measures and cameras, in recording a crime scene. A disadvantage of these methods is that they are slow and cumbersome. The development of a portable system that can rapidly record a crime scene with current camera imaging, 3D geometric surface maps, and contribute quantitative measurements such as accurate relative positioning of crime scene objects, would be an asset to law enforcement agents in collecting and recording significant forensic data. The purpose of this project is to develop a feasible prototype of a fast, accurate, 3D measurement and imaging system that would support law enforcement agents to quickly document and accurately record a crime scene.
The objective of the SIERRA framework is to provide a common software infrastructure for massively parallel computational mechanics applications. The SIERRA framework consolidates the mechanics-independent computational services required by a diverse set of mechanics applications into a shared framework. Consolidation of these computational services eliminates their redundant development and maintenance efforts and streamlines the coupling of independently developed computational mechanics capabilities into integrated multi-mechanics applications.
In this letter we discusses the first application of 3-dimensional nonlocal density functional calculations to the interactions of solvated rigid polymers. The three cases considered are cylindrical polymers, bead-chain polymers, and periodic polymers. We calculate potentials of mean force, and show that polymer surface structure plays a critical role in determining the solvation energy landscape which in turn controls routes to assembly of the macromolecules.
Current hexahedral mesh generation techniques rely on a set of meshing tools, which when combined with geometry decomposition leads to an adequate mesh generation process. Of these tools, sweeping tends to be the workhorse algorithm, accounting for at least 50% of most meshing applications. Constraints which must be met for a volume to be sweepable are derived, and it is proven that these constraints are necessary but not sufficient conditions for sweepability. This paper also describes a new algorithm for detecting extruded or sweepable geometries. This algorithm, based on these constraints, uses topological and local geometric information, and is more robust than feature recognition-based algorithms. A method for computing sweep dependencies in volume assemblies is also given. The auto sweep detect and sweep grouping algorithms have been used to reduce interactive user time required to generate all-hexahedral meshes by filtering out non-sweepable volumes needing further decomposition and by allowing concurrent meshing of independent sweep groups. Parts of the auto sweep detect algorithm have also been used to identify independent sweep paths, for use in volume-based interval assignment.
Sweeping algorithms have become very mature and can create a semi-structured mesh on a large set of solids. However, these algorithms require that all linking surfaces be mappable or submappable. This restriction excludes solids with imprints or protrusions on the linking surfaces. The grafting algorithm allows these solids to be swept. It then locally modifies the position and connectivity of the nodes on the linking surfaces to align with the graft surfaces. Once a high-quality surface mesh is formed on the graft surface, it is swept along the branch creating a 2 3/4-D mesh.
We present a new shape measure for tetrahedral elements that is optimal in the sense that it gives the distance of a tetrahedron from the set of inverted elements. This measure is constructed from the condition number of the linear transformation between a unit equilateral tetrahedron and any tetrahedron with positive volume. We use this shape measure to formulate two optimization objective functions that are differentiated by their goal: the first seeks to improve the average quality of the tetrahedral mesh; the second aims to improve the worst-quality element in the mesh. Because the element condition number is not defined for tetrahedral with negative volume, these objective functions can be used only when the initial mesh is valid. Therefore, we formulate a third objective function using the determinant of the element Jacobian that is suitable for mesh untangling. We review the optimization techniques used with each objective function and present experimental results that demonstrate the effectiveness of the mesh improvement and untangling methods. We show that a combined optimization approach that uses both condition number objective functions obtains the best-quality meshes.
Objective functions for unstructured hexahedral and tetrahedral mesh optimization are analyzed using matrices and matrix norms. Mesh untangling objective functions that create valid meshes are used to initialize the optimization process. Several new objective functions to achieve element invertibility and quality are investigated, the most promising being the ''condition number''. The condition number of the Jacobian matrix of an element forms the basis of a barrier-based objective function that measures the distance to the set of singular matrices and has the ideal matrix as a stationary point. The method was implemented in the Cubit code, with promising results.
Considerable progress has been made on automatic hexahedral mesh generation in recent years. Several automatic meshing algorithms have proven to be very reliable on certain classes of geometry. While it is always worth pursuing general algorithms viable on more general geometry, a combination of the well-established algorithms is ready to take on classes of complicated geometry. By partitioning the entire geometry into meshable pieces matched with appropriate meshing algorithm the original geometry becomes meshable and may achieve better mesh quality. Each meshable portion is recognized as a meshing feature. This paper, which is a part of the feature based meshing methodology, presents the work on shape recognition and volume decomposition to automatically decompose a CAD model into meshable volumes. There are four phases in this approach: (1) Feature Determination to extinct decomposition features, (2) Cutting Surfaces Generation to form the ''tailored'' cutting surfaces, (3) Body Decomposition to get the imprinted volumes; and (4) Meshing Algorithm Assignment to match volumes decomposed with appropriate meshing algorithms. The feature determination procedure is based on the CLoop feature recognition algorithm that is extended to be more general. Results are demonstrated over several parts with complicated topology and geometry.
A new method for lessening skew in mapped meshes is presented. This new method involves progressive subdivision of a surface into loops consisting of four sides. Using these loops, constraints can then be set on the curves of the surface, which will propagate interval assignments across the surface, allowing a mesh with a better skew metric to be generated.
NIST and Sandia have developed a procedure for producing and calibrating critical dimension (CD), or linewidth, reference materials. These reference materials will be used to calibrate metrology instruments used in semiconductor manufacturing. The artifacts, with widths down to 100 nm, are produced in monocrystalline silicon with all feature edges aligned to specific crystal planes. A two-part calibration of these linewidths is used: the primary calibration, with accuracy to within a few lattice plane thicknesses, is accomplished by counting the lattice planes across the sample as-imaged through use of high-resolution transmission electron microscopy (HRTEM). The secondary calibration is the high-precision electrical CD technique. NIST and Sandia are developing critical dimension (CD), or linewidth, reference materials for use by the semiconductor industry. To meet the current requirements of this rapidly changing industry, the widths of the reference features must be at or below the widths of the finest features in production and/or development. Further, these features must produce consistent results no matter which metrology tool (e.g., scanning electron microscope, scanned probe microscope, electrical metrology) is used to make the measurement. This leads to a requirement for the samples to have planar surfaces, known sidewall angles, and uniform material composition. None of the production techniques in use in semiconductor manufacturing can produce features with all these characteristics. In addition, requirements specified in the National Technology Roadmap for Semiconductors indicate that the width of the feature must be accurately calibrated to approximately 1-2 nm, a value well beyond the current capabilities of the instruments used for semiconductor metrology.
This paper presents a new technique for automatically detecting interval constraints for swept volumes with holes. The technique finds true volume constraints that are not necessarily imposed by the surfaces of the volume. A graphing algorithm finds independent, parallel paths of edges from source surfaces to target surfaces. The number of intervals on two paths between a given source and target surface must be equal; in general, the collection of paths determine a set of linear constraints. Linear programming techniques solve the interval assignment problem for the surface and volume constraints simultaneously.
Frequency-domain shot-record migration can produce higher quality images than Kirchhoff migration but typically at a greater cost. The computational cost of shot-record migration is the product of the number of shots in the survey and the expense of each individual migration. Many attempts to reduce this cost have focused on the speed of the individual migrations, trying to achieve a better trade-off between accuracy and speed. Another approach is to reduce the number of migrations. We investigate the simultaneous migration of shot records using frequency-domain shot-record migration algorithms. The difficulty with this approach is the production of so-called cross terms between unrelated shot and receiver wavefields, which generate unwanted artifacts or noise in the final image. To reduce these artifacts and obtain an image comparable in quality to the single-shot-per-migration result, we have introduced a process called phase encoding which shifts or disperses these cross terms. The process of phase encoding thus allows one to trade signal-to-noise ratio for the speed of migrating the entire survey. Several encoding functions and two application strategies have been tested. The first strategy, combining multiple shots per migration and using each shot only once, provides a reduction in computation directly related to the number of shots combined. The second strategy, performing multiple migrations of all the shots in the survey, provides a means to reduce the cross-term noise through stacking the resulting images. The additional noise in both strategies may be tolerated if it is no stronger than the inherent seismic noise in the migrated image, and if the final image is achieved with less cost.
Both the increased complexity of integrated circuits, resulting in six or more levels of integration, and the increasing use of flip-chip packaging have driven the development of integrated circuit (IC) failure analysis tools that can be applied to the backside of the chip. Among these new approaches are focused ion beam (FIB) tools and processes for performing chip edits/repairs from the die backside. This paper describes the use of backside FIB for a failure analysis application rather than for chip repair. Specifically, we used FIB technology to prepare an IC for inspection of voided metal interconnects (''lines'') and vias. Conventional FIB milling was combined with a super-enhanced gas assisted milling process that uses XeF{sub 2} for rapid removal of large volumes of bulk silicon. This combined approach allowed removal of the TiW underlayer from a large number of Ml lines simultaneously, enabling rapid localization and plan view imaging of voids in lines and vias with backscattered electron (BSE) imaging in a scanning electron microscope (SEM). Sequential cross sections of individual voided vias enabled us to develop a 3-d reconstruction of these voids. This information clarified how the voids were formed, helping us identify the IC process steps that needed to be changed.
Many applications produce three-dimensional points that must be further processed to generate a surface. Surface reconstruction algorithms that start with a set of unorganized points are extremely time-consuming. Sometimes, however, points are generated such that there is additional information available to the reconstruction algorithm. We present Spiraling Edge, a specialized algorithm for surface reconstruction that is three orders of magnitude faster than algorithms for the general case. In addition to sample point locations, our algorithm starts with normal information and knowledge of each point's neighbors. Our algorithm produces a localized approximation to the surface by creating a star-shaped triangulation between a point and a subset of its nearest neighbors. This surface patch is extended by locally triangulating each of the points along the edge of the patch. As each edge point is triangulated, it is removed from the edge and new edge points along the patch's edge are inserted in its place. The updated edge spirals out over the surface until the edge encounters a surface boundary and stops growing in that direction, or until the edge reduces to a small hole that is filled by the final triangle.
Visualization systems are complex dynamic software systems. Debugging such systems is difficult using conventional debuggers because the programmer must try to imagine the three-dimensional geometry based on a list of positions and attributes. In addition, the programmer must be able to mentally animate changes in those positions and attributes to grasp dynamic behaviors within the algorithm. In this paper we shall show that representing geometry, attributes, and relationships graphically permits visual pattern recognition skills to be applied to the debugging problem. The particular application is a particle system used for isosurface extraction from volumetric data. Coloring particles based on individual attributes is especially helpful when these colorings are viewed as animations over successive iterations in the program. Although we describe a particular application, the types of tools that we discuss can be applied to a variety of problems.
We show that Sorensen's [35] implicitly restarted Arnoldi method (including its block extension) is simultaneous iteration with an implicit projection step to accelerate convergence to the invariant subspace of interest. By using the geometric convergence theory for simultaneous iteration due to Watkins and Elsner [43], we prove that an implicitly restarted Arnoldi method can achieve a super-linear rate of convergence to the dominant invariant subspace of a matrix. Moreover, we show how an IRAM computes a nested sequence of approximations for the partial Schur decomposition associated with the dominant invariant subspace of a matrix.
