No description available.

In several applications of casting, dendritic microstructure is not desirable as it results in poor mechanical properties. Enhancing fluid flow in the mushy zone by stirring is one of the means to suppress this dendritic growth. Strong fluid flow detaches the dendrites formed at the solid-liquid interface and carries them into the mould to form slurry. When this slurry solidifies, the microstructure is characterized by globular, non-dendritic primary phase particles, separated and enclosed by a near-eutectic lower-melting secondary phase. This property represents a great potential for further processing in semisolid forming (SSF) by various techniques such as pressure die casting and forging. Among all currently available methods, linear electromagnetic (EM) stirring is considered as one of the most suitable routes for large scale production of semisolid feed stock. One of the biggest advantages of EM stirring is that the stirring intensity and direction can be modulated externally and in a non-intrusive manner. With this viewpoint, the primary objective of the present research is to investigate rheocasting using linear electromagnetic stirring. A systematic development of a linear electromagnetic stirrer (LEMS) is the subject of the first part of the thesis. The LEMS consists of a set of six primary coils displaced in space. These coils are excited by a three-phase power supply to produce an axially travelling magnetic field. The metal to be stirred is placed in the annular space of the stirrer. The travelling field induces secondary current in the molten metal. The current and magnetic fields interact to generate a net mechanical force in the metal, commonly termed as the Lorentz force. The molten metal is stirred under the influence of this force. Two prototype stirrers, one for low melting alloys and the other for aluminium alloys are developed. The stirrers are characterized by measuring forces on low melting point alloy and on solid aluminum cylinders placed inside the annular space of the stirrer. As an outcome of these tests, a non-intrusive method of detecting stirring of liquid metal is developed. The development of a rheocasting mould for the LEMS forms the second part of the work presented in the thesis. The mould design and cooling arrangement are such that solidification in the mould is primarily unidirectional. Heat from the solidifying metal is extracted at the bottom of the mould, so that the axisymmetric EM stirring effectively shears the dendrites formed at the solid-liquid interface. The outer surface of the mould is cooled with water or air exiting from 64 jets, each of 4 mm diameter. Such an arrangement provides a high heat transfer coefficient and a wide range of cooling rate in the metal ranging from 0.01 to 10 K/s. Temperature is measured at various depths in the solidifying melt and at other key locations in the mould to assess the various heat transfer mechanisms. The results from the rheocasting experiments using the above mould and LEMS are presented in the third and final part of the thesis. Such studies are required for understanding the solidification process in presence of electromagnetic stirring and for highlighting the important issues connecting solidification, fluid flow, dendrite fragmentation and the resulting microstructure. A series of experiments are performed with A356 (Al-7Si-0.3Mg) alloy. Experiments are conducted with various combinations of operating parameters, and the resulting microstructures and cooling curves at various locations are examined. The key process parameters are stirring current, cooling rate, pouring temperature, and stirring current frequency. The parametric studies also include the case without EM stirring in which liquid aluminium is poured into the rheocast mould without powering the LEMS. It is found that stirring at high currents produces non-dendritic microstructures at all locations of the billet. For lower currents, however, dendritic microstructures are observed in regions outside the zone of active stirring. Stirring also enhances heat loss from the exposed top surface, leading to solid front advancement from the top as well. Without EM stirring, microstructures are found to be dendritic everywhere. The percentage of primary α-Al phase and its number density are found to increase with stirring intensity. With a decrease in cooling rate with air as the coolant, the average grain size of primary α-Al phase increases. Excitation frequency is found to be an important parameter, with lower frequencies generating a more uniform force field distribution, and higher frequencies enhancing induction heating. At higher frequencies, the effect of higher induction heating results in the formation of larger and coarser primary phase grains. This phenomenon has led to the development of a one-step process for rheocasting and heat treatment of billets.

In several applications of casting, dendritic microstructure is not desirable as it results in poor mechanical properties. Enhancing the fluid flow in the mushy zone by stirring is one of the means to suppress this dendritic growth. The strong fluid flow detaches the dendrites from the solid-liquid interface and carries them into the mold to form slurry. The detached dendrites coarsen in the slurry and form into rosette or globular particles based on processing conditions. This slurry offers less resistance to flow even at a high solid fraction and easily flow into the die-cavity. The above principle is the basis of a new manufacturing technology called “semi-sold forming” (SSF), in which metal alloys are cast in the semi-solid state. This technique has several advantages over other existing commercial casting processes, such as reduction of macrosegregation, reduction of porosity and low forming efforts. A major challenge existing in semisolid manufacturing is the production of metallic slurry in a consistent manner. The main difficulty arises because of the presence of a wide range of process parameters affecting the quality of the final product. An established method of producing slurry is by stirring the alloy using an electromagnetic stirrer. From an elaborate review of literature, it is apparent that solidification in presence of electromagnetic stirring involves a wide range of shear and cooling rates variation. However, the CFD models found in the literature are generally not based on accurate rheological properties, which are known to be functions of the relevant process parameters. Hence, there is a clear need for a comprehensive numerical model for such a solidification process, involving accurate rheological data for the semisolid slurry subjected to a range of processing conditions. The objective of the present work is to develop a numerical model for studying the transport phenomena during solidification with linear electromagnetic stirring. The study is presented in the context of a billet making process in a cylindrical mould using linear electromagnetic stirring. The mould consists of two parts: the upper part of the mould is surrounded by a linear electromagnetic stirrer forming the zone of active stirring, and the lower part of the mould is used to cool the liquid metal. The material chosen for the study is Al-7.32%Si (A356) alloy, commonly used for die casting applications. A complete numerical model will therefore have two major components: one dealing with rheological behavior of the semisolid slurry, and the other involving macroscopic modeling of the process using computational fluid dynamics (CFD) techniques. For the latter part of the model, determination of rheological behavior of the slurry is a pre-requisite. The rheological characteristics of the stirred slurry, as a function of shear rate and cooling rate, is determined experimentally using a concentric cylinder viscometer. Two different series of experiments are performed. In the first series, the liquid metal is cooled at a constant cooling rate and sheared with different shear rates to get the effect of shear rate on viscosity. In the second series of experiments, the liquid metal is cooled at different cooling rates and sheared at a constant shear rate to obtain the effect of cooling rate on viscosity. During all these experiments, the shear rate is calculated from the measured angular velocity of spindle using inductive position sensor; viscosity of the slurry is calculated based on the torque applied to the slurry and angular velocity of the spindle; and the solid fraction is calculated from measured temperature of the slurry based on Schiel equation. From these data, a constitutive relation for variable viscosity is established, which is subsequently used in a numerical model for simulating the transport phenomena associated with the solidification process. The numerical model uses a set of single-phase governing equations of mass, momentum, energy and species conservation. The set of governing equations is solved using a pressure based finite volume technique, along with an enthalpy based phase change algorithm. The numerical simulation of this process also involves modeling of Lorentz force field. The numerical study involves prediction of temperature, velocity, species and solid fraction distribution. First, studies are performed for a base case with a moderate stirring intensity of 250A primary current and 50 Hz frequency. It is found that the electromagnetic forces have maximum values near the mould periphery, which results in an ascending movement of the slurry near the mould periphery. Because of continuity, this slurry comes down along the axis of the mould. Stirring produces a strong fluid flow which results good mixing in the melt. Correspondingly, a homogenized temperature distribution is found in the domain. Because of strong stirring, the solid fraction in the slurry is found to be distributed almost uniformly. It is also found that fragmentation of dendrites increases solid fraction in the slurry with processing time. During processing, the continuous rejection of solute makes the liquid progressively solute enriched. It is predicted from the present study that the remaining liquid surrounding the primary solid phase finally solidifies with a near-eutectic composition, which is desirable from the point of view of semisolid casting. Correspondingly, a set of experiments are performed to validate the numerically predicted results. The numerical predictions of temperature variations are in good agreement with experiments, and the predicted flow field evolution correlate well with the microstructures obtained through experiments at various locations, as observed in the numerical results. Subsequently the study is extended to predict the effect of process parameters such as stirring intensity and cooling rate on the distributions of solid fraction and solute in the domain. It is found, from the simulation, that the solidification process is significantly affected by stirring intensity. At increasing primary excitation current, the magnitude of Lorentz force increases and results in increase of slurry velocity. Correspondingly, the fragmentation of dendrites from the solid/liquid is more during solidification at higher stirring intensity, which increases the fraction of solid in the slurry to a high value. It is also found that the solute and fraction of solid in the liquid mixes well under stirring action. Thus, a near uniform distribution of solute and solid fraction is found in the domain. It is found that stirring at high currents produces high solid fraction in the liquid. Also, at very low cooling rate, the solid fraction in the liquid increases. The present study focuses on the model development and experimental validation for solidification with linear electromagnetic stirring for producing a rheocast billet. Further studies highlighting the effects of various process parameters on the thermal history and microstructure formation are also presented.

The challenges of weight reduction and good strength in automotive industry have drawn considerable interest in thixocasting technologies. Joining of such components with conventional fusion welding creates voids, hot cracking, distortion in shape, and more importantly evolution of dendritic microstructure that ultimately would lead to inferior mechanical properties of the weld region. Thus, the purpose of making thixocast component is lost. The friction welding which is a solid state joining process can avoid defects associated with melting and solidification in a typical fusion weld and can be a promising alternative. This process produces a weld under compressive force at the contact of workpieces rotating or moving relative to one another to produce heat and plastically displacing material from the faying surfaces. Research on semisolid processing has its origin in the early 1970s. However, from the literature survey on semisolid processing it is clear that, till date, not much work has been done in field of joining of semisolid processed components. In the area of solid state welding, in particular, it is not at all explored. In view of this, the present work is focused on exploration of joining of Thixocast A356 Aluminium alloy component by friction welding and comparison of its performance with friction weld of conventionally cast sample of the same alloy. The study is carried out experimentally as well as numerically. Moreover, the material behaviour of thixocast component at elevated temperature in solid state is also described with the help of processing maps and constitutive modelling. The hot workability of thixocast and conventionally cast A356 alloy is evaluated with the help of processing maps developed on the basis of the dynamic materials model approach using the flow stress data obtained from the isothermal compression test in wide range of temperature (300-500℃) and strain rates (0.001s-1-10s-1). The domains of the processing map are interpreted in terms of the associated microstructural mechanism. On comparing the flow stress at elevated temperature of thixocast and conventionally cast A356 alloy samples, it is observed that the flow stress of the latter showed higher value at different strain level, temperature and strain rates. This indicates that the flow property of the thixocast alloy sample is better than that of the conventionally cast one (i.e. response to plastic flow is better for the former); while at room temperature thixocast sample has higher strength. Moreover to investigate the general nature of the influence of strain, strain rate and temperature on the compressive deformation characteristics of thixocast A356 and conventionally cast A356 aluminium alloy, a comprehensive model describing the relationship of the flow stress, strain rate and temperature of the alloys at elevated temperatures is proposed by hyperbolic-sine Arrhenius-type equation and Johnson-Cook model. The validity of descriptive results based on the proposed constitutive equation is also investigated and a comparison between two constitutive models is also made. In order to numerically model the friction welding process of a thixocast A356 aluminium alloy and conventionally cast alloy of same material using a finite element method (FEM), temperature dependent physical properties, mechanical properties as well as viscoplastic constitutive equations were used in the model. A two- dimensional axisymmetric finite element model has been developed. The modelling is based on a coupled thermomechanical approach. First, a nonlinear, transient two-dimensional heat transfer model is developed to determine the temperature fields. Later, the temperature fields are used as input for a nonlinear, two-dimensional structural model in order to predict the distortions and von Mises stress. The finite element models are parametrically built using APDL (ANSYS Parametric Design Language) provided by ANSYS. The validation of the model is carried out by comparing with the experiment. Once validated, the thermomechanical model was used to perform parametric studies in order to investigate effects of various process parameters on temperature and stress distribution in the workpieces. This helps in deciding the range of parameters for friction welding experiments in order to get good weld. Both thixocast and conventionally cast samples exhibited similar temperature distribution during the friction welding process, because of identical thermophysical properties. The magnitude of von Mises stress distribution during friction welding of thixocast A356 sample is found to be lower than that of the conventionally cast sample. It is because of their different constitutive behaviour at elevated temperature. Moreover, the developed FEM model can be successfully used to predict the residual stress at various locations for different set of parameters and geometry for friction welding of thixocast and conventionally cast A356 alloy. This helps in reducing time consuming and expensive experiments on residual stress measurement. The chosen experiments based on Taguchi L27 orthogonal array were conducted on the friction welding machine which works on the principles of continuous drive-mechanism. The experimental specimens were machined from thixocast A356 aluminium alloy connecting rods as well as conventionally cast A356 aluminium alloy ingot in the form of cylindrical bars of dimensions 85mm length and 20mm diameter. The parameters used for welding were friction pressure, rpm, forge pressure, burn-off, and upset pressure. The effects of welding parameters on performance characteristics (i.e. tensile strength and weld efficiency) were evaluated. Taguchi method was applied to investigate the influence of each parameter on strength of joints and evaluate the combination of parameters that leads to the highest weld strength. Accordingly optimum process parameters was identified which helps in achieving the tensile strength of more than parent material. The optimized process parameters for friction welding of thixocast A356 aluminium alloy are rpm = 500, friction pressure = 60, upset time = 5, upset pressure = 100 and burn off = 5. The empirical relationships were also developed to predict the tensile strength. The developed relationship can be effectively used to predict the tensile strength of welded joint with a correlation coefficient of 0.86, which indicates the strong positive relationship between predicted and experimental data. Friction welding of thixocast A356 aluminium alloy helps to achieve very fine eutectic silicon particles of the order of 0.4 at the interface due to severe plastic deformation taking place during welding. Obtaining such fine eutectic silicon particles is difficult to be achieved within few seconds of processing by any other method. The hardness variation of friction welded thixocast alloy shows higher value as compared to that of a conventionally cast sample in the heat affected zone, which indicates better weld strength of the former. This was also confirmed by the tensile strength studied and fatigue test. This indicates that weldability of cast alloys will get improved if the microstructure is modified to globular type.