Menghuai Wu

Influence of convection and grain movement on globular equiaxed solidification

Abstract A two-phase volume averaging model was used to study convection and grain movement, and their influence on the globular equiaxed solidification. Both liquid and solid phases were treated as separate interpenetrating continua. The mass, momentum, species and enthalpy conservation equations for each phase and a grain transport equation were coupled. An ingot casting (Al–4 wt.% Cu) with near globular solidification morphology was simulated. Case studies with different modeling assumptions such as with and without grain movement, and with slip and non-slip boundary conditions for solid phase were presented and compared. Understanding of grain evolution and macrosegregation formation in globular equiaxed solidification was improved.

Modelling the solidification of hypermonotectic alloys

A two-phase model is developed to simulate the decomposition and spatial phase separation (microstructure evolution) during solidification of hypermonotectic alloys. The minority liquid phase, decomposing in morphology of droplets from the parent melt, is treated as the second phase, L2, while the parent melt, including the solidified monotectic matrix, is the first phase, L1. The conservation equations of mass, momentum, solute and enthalpy for both phases, and an additional transport equation for the droplets are solved. Nucleation of the droplets, diffusion-controlled growth (coarsening) and dissolution of the droplets, interphase interactions such as Marangoni (thermocapillary) force, Stokes force, solute partitioning and heat release of decomposition are modelled by the corresponding source and exchange terms in the conservation equations. The monotectic reaction is modelled by adding the latent heat on the L1 phase and applying a suitable large viscosity to the solidified monotectic matrix. The simulation results of a two-dimensional square casting with hypermonotectic composition (Al–10 wt.% Bi) under normal terrestrial and weightless conditions are presented and discussed.

Computer aided prediction and control of shrinkage porosity in titanium dental castings

OBJECTIVES The main objectives were to investigate the possibility and reliability of quantitative prediction and control of the concentrated shrinkage porosity (macroporosity) in titanium dental castings by means of a numerical simulation technique; and finally to optimize the filling and feeding system design for dental castings. METHODS A commercial software, MAGMASOFT (Giessereitechnologie GmbH, Germany), was employed to simulate the mold filling and solidification process, and predict the shrinkage tendency in a sample dental casting, two simplified tooth crowns with a connector bar between them. The numerically predicted shrinkages were compared with the experimental results. The experiments were carried out on a centrifugal casting machine. The same geometric and processing parameters of the casting as in the simulations were strictly controlled. RESULTS The computer predicted shrinkage porosity coincided with the performed experiments, demonstrating the reliability of the numerical model and the thermal physical data chosen for the calculations. Based on the above numerical model, several filling and feeding systems for the same casting were numerically simulated and compared. Finally an optimized design for this sample casting was proposed, and porosity-free castings were obtained. SIGNIFICANCE It was expected that the numerical simulation technique could be further developed for dental laboratories to aid the real dental casting design.

Numerical study of porosity in titanium dental castings

A commercial software package, MAGMASOFT (MAGMA Giessereitechnologie GmbH, Aachen, Germany), was used to study shrinkage and gas porosity in titanium dental castings. A geometrical model for two simplified tooth crowns connected by a connector bar was created. Both mold filling and solidification of this casting model were numerically simulated. Shrinkage porosity was quantitatively predicted by means of a built-in feeding criterion. The risk of gas pore formation was investigated using the numerical filling and solidification results. The results of the numerical simulations were compared with experiments, which were carried out on a centrifugal casting machine with an investment block mold. The block mold was made of SiO2 based slurry with a 1 mm thick Zr2 face coat to reduce metal–mold reactions. Both melting and casting were carried out under protective argon (40 kPa). The finished castings were sectioned and the shrinkage porosity determined. The experimentally determined shrinkage porosity coincided with the predicted numerical simulation results. No apparent gas porosity was found in these model castings. Several running and gating systems for the above model casting were numerically simulated. An optimized running and gating system design was then experimentally cast, which resulted in porosity-free castings.

Numerical simulation of the casting process of titanium tooth crowns and bridges

The objectives of this paper were to simulate the casting process of titanium tooth crowns and bridges; to predict and control porosity defect. A casting simulation software, MAGMASOFT, was used. The geometry of the crowns with fine details of the occlusal surface were digitized by means of laser measuring technique, then converted and read in the simulation software. Both mold filling and solidification were simulated, the shrinkage porosity was predicted by a “feeding criterion”, and the gas pore sensitivity was studied based on the mold filling and solidification simulations. Two types of dental prostheses (a single-crown casting and a three-unit-bridge) with various sprue designs were numerically “poured”, and only one optimal design for each prosthesis was recommended for real casting trial. With the numerically optimized design, real titanium dental prostheses (five replicas for each) were made on a centrifugal casting machine. All the castings endured radiographic examination, and no porosity was detected in the cast prostheses. It indicates that the numerical simulation is an efficient tool for dental casting design and porosity control.

Numerical simulation of the casting process of titanium removable partial denture frameworks

The objective of this work was to study the filling incompleteness and porosity defects in titanium removal partial denture frameworks by means of numerical simulation. Two frameworks, one for lower jaw and one for upper jaw, were chosen according to dentists’ recommendation to be simulated. Geometry of the frameworks were laser-digitized and converted into a simulation software (MAGMASOFT). Both mold filling and solidification of the castings with different sprue designs (e.g. tree, ball, and runner-bar) were numerically calculated. The shrinkage porosity was quantitatively predicted by a feeding criterion, the potential filling defect and gas pore sensitivity were estimated based on the filling and solidification results. A satisfactory sprue design with process parameters was finally recommended for real casting trials (four replica for each frameworks). All the frameworks were successfully cast. Through X-ray radiographic inspections it was found that all the castings were acceptably sound except for only one case in which gas bubbles were detected in the grasp region of the frame. It is concluded that numerical simulation aids to achieve understanding of the casting process and defect formation in titanium frameworks, hence to minimize the risk of producing defect casting by improving the sprue design and process parameters.

Simulation of casting, homogenization, and hot rolling: consecutive process and microstructure modelling for aluminium sheet production

An overview of simulation of casting, homogenization, and hot rolling of an aluminium alloy is addressed in this paper. The microstructure models used to describe casting, solidification, precipitation (growth and coarsening) during homogenization, deformation texture evolution, and the work hardening behaviour are presented as well as their respective theoretical backgrounds. Emphasis is placed on interfacing the microstructure models with each other between the processing steps. This makes it possible to take into account microstructural changes that occur early during processing during later production steps. Along with this overview, reference will be made to previously presented simulation and experimental results—for validation—where appropriate.