Search

Galerkin finite-element approximations and Newton's method for solving free boundary problems are combined with computer-implemented techniques from nonlinear perturbation analysis to study solidification problems with natural convection in the melt. The Newton method gives rapid convergence to steady state velocity, temperature and pressure fields and melt-solid interface shapes, and forms the basis for algebraic methods for detecting multiple steady flows and assessing their stability. The power of this combination is demonstrated for a two-phase Rayleigh-Benard problem composed of melt and solid in a veritical cylinder with the thermal boundary conditions arranged so that a static melt with a flat melt-solid interface is always a solution. Multiple cellular flows bifurcating from the static state are detected and followed as Rayleigh number is varied. Changing the boundary conditions to approach those appropriate for the vertical Bridgman solidification system causes imperfections that eliminate the static state. The flow structure in the Bridgman system is related to those for the Rayleigh-Benard system by a continuous evolution of the boundary conditions.

The structure of axisymmetric buoyancy-driven convection in a vertical cylinder heated from below is probed by finite element solution of the Boussinesq equations coupled with computed-implemented perturbation techniques for detecting and tracking multiple flows and for determining flow stability. Results are reported for fluids with Prandtl number of one and for cylinders with aspect ratio (Lambda) (defined as the height to radius of the cylinder) between 0.5 and 2.25. Extensive calculations of the neutral stability curve for the static solution and of the nonlinear motions along the bifurcating flow families show a continuous evolution of the primary cellular motion from a single toroidal cell to two and three cells nested radially in the cylinder, instead of the sharp transitions found for a cylinder with shear-free sidewalls. The smooth transitions in flow structure with Rayleigh number and lambda are explained by nonlinear connectivity between the first two bifurcating flow families formed either by a secondary bifurcation point for Lambda or = Lambda * approximately 0.80 or by a limit point for Lambda Lambda *. The transition between these two modes may be described by the theory of multiple limit point bifurcation.

Two iterative schemes based on the mixed finite element method are developed for analyzing steady natural convection in a melt adjacent to its solid phase. The simplest method decouples the calculation of the field variables and the shape of the melt/solid interface into two interlocked iterations that are performed successively. The second method uses Newton's iteration to solve simultaneously for both types of unknowns and has a quadratic convergence rate. Results for a model problem of melt and solid in a cylindrical ampoule show the Newton algorithm to be a factor of three more efficient.

The roles of natural convection in the melt and the shape of the melt/solid interface on radial dopant segregation are analyzed for a prototype of vertical Bridgman crystal growth system by finite element methods that solve simultaneously for the velocity field in the melt, the shape of the solidification isotherm, and the temperature distribution in both phases. Results are presented for crystal and melt with thermophysical properties similar to those of gallium-doped germanium in Bridgman configurations with melt below (thermally destabilizing) and above (stabilizing) the crystal. Steady axisymmetric flow are classified according to Rayleigh number as either being nearly the growth velocity, having a weak cellular structure or having large amplitude cellular convention. The flows in the two Bridgman configurations are driven by different temperature gradients and are in opposite directions. Finite element calculations for the transport of a dilute dopant by these flow fields reveal radial segregation levels as large as sixty percent of the mean concentration. Segregation is found most severe at an intermediate value of Rayleigh number above which the dopant distribution along the interface levels as the intensity of the flow increases.

The results of two lines of research aimed at calculating the structure of the flows driven by buoyancy in small-scale crystal growth systems and at understanding the coupling between these flows, the shape of the solidification interface, and dopant segregation in the crystal are reviewed. First, finite-element methods are combined with computer-aided methods for detecting multiple steady solutions to analyze the structure of the buoyancy-driven axisymmetric flows in a vertical cylinder heated from below. This system exhibits onset of convection, multiple steady flows, and loss of the primary stable flow beyond a critical value of the Rayleigh number. Second, results are presented for calculations of convection, melt/solid interface shape, and dopant segregation within a vertical ampoule with thermal boundary conditions that represent a prototype of the vertical Bridgman growth system.

The state of the art is assessed in substructure coupling for dynamic analysis. A general formulation, which permits all previously described methods to be characterized by a few constituent matrices, is developed. Limited results comparing the accuracy of various methods are presented.

Fixed interface and free interface methods of substructure coupling for dynamic analysis are discussed. Three methods for reducing the number of coordinates required by fixed interface methods are introduced. Matrix ordinary differential equations are employed to improve accuracy in free interface substructure coupling methods.

Substructure coupling or component-mode synthesis may be employed in the solution of dynamics problems for complex structures. Although numerous substructure-coupling methods have been devised, little attention has been devoted to methods employing attachment modes. In the present paper the various mode sets (normal modes, constraint modes, attachment modes) are defined. A generalized substructure-coupling procedure is described. Those substructure-coupling methods which employ attachment modes are described in detail. One of these methods is shown to lead to results (e.g., system natural frequencies) comparable to or better than those obtained by the Hurty (1965) method.

Benfield, et al. (1972) showed that among fixed-interface, free-interface, and hybrid substructure coupling methods, the fixed-interface methods as the most accurate and the free-interface methods are the least accurate. In the present note, a substructure coupling method is proposed which employs free-interface substructure modes supplemented by 'reduced flexibility.' Substructure coupling based on the improved substructure model is discussed, and a numerical comparison with Hou's free-interface method is given. To simplify representation of the method proposed, the substructure equations are developed first for constrained substructures, and then the equations representing substructures with rigid-body modes are given. Finally, the equations for coupling of substructures are derived. Example calculations are included.

The derivations of mathematical models in Vol. 1 are presented along with descriptions of all peripheral studies related to the primary study objectives.

The optimization technique is described which was used in the study for applying modern optimal control technology to the design of shuttle booster engine reaction control systems and aerodynamic control systems. Complete formulations are presented for both the ascent and reentry portions of the study. These formulations include derivations of the 6D perturbation equations of motion and the process followed in the control and blending law selections. A total hybrid software concept applied to the study is described in detail. Conclusions and recommendations based on the results of the study are included.

Prototype vehicle features exceptional obstacle-negotiating and slope-climbing capabilities plus high propulsive efficiency. Concept should interest designers of polar or ocean-bottom research vehicles. Also, its large footprint and low ground pressure will minimize ecological damage on terrain with low bearing strength, as in off-the-road application.