A mathematical model is developed in this study to simulate the filling pattern in lost foam casting and validated by comparing the simulated results to the experimental measurements. A special treatment is devised to handle the unique problem of back-pressure generated due to the evaporation of polystyrene during filling in lost foam casting. Experiments are also conducted with thermocouples embedded in the pattern of lost foam casting. With the measured temperature data, filling pattern can be derived. The mathematical model is then tested on several lost foam castings, where experimental measurements are also conducted. As the simulated filling patterns are compared with the experimental measurements, good agreement is observed.

The interfacial heat transfer coefficient (HTC) between the steel mold and AZ91D magnesium (Mg) alloy that cast under various initial solid fractions were investigated in this study.The interfacial HTC was determined by using an Inverse Method with measured temperature data and known thermo-physical properties. A Computer Aided Cooling Curve Analysis technique was used to determine the solid fraction versus temperature relationship. To comply with the requirements of the Inverse Method, a one-dimensional heat transfer system from the casting to the mold was designed for the permanent mold casting of AZ91D Mg alloy in molten and semisolid states.Experiments were conducted with different initial solid fractions of 0, 30, and 50%. The results indicate that the HTC profile of molten AZ91D during solidification can be divided into five stages, while casting with semisolid AZ91D only into three. During each stage, the casting/ mold interfacial conditions vary, which in turn causes the HTC values to vary. These data are critical for any solidification model of permanent mold casting and semisolid casting to obtain a reliable prediction of the thermal profile inside the solidifying casting and its freezing time.

An interactive computer simulation system has been developed in this study to aid the determination of the pressure±time relationship during the ®lling of a low pressure casting to eliminate ®lling-related defects while maintaining its productivity. The pressure required to ®ll a casting in a low pressure casting process can be separated into two stages. The ®rst stage is to exert pressure to force the molten metal to rise in the riser tube up to the gate of the casting die, which varies from casting to casting due to the drop of the level of the molten metal in the furnace, whilst the second stage is to add an additional pressure to push the molten metal into the die cavity in a way that will not cause much turbulence and have the proper ®lling pattern to avoid the entrapment of gas while maintaining productivity.One of the major efforts in this study is to modify the ®lling simulation system with the capability to directly predict the occurrence of gas porosity developed earlier to interactively determine the proper gate velocity for each and every part of the casting. The pressure required to ®ll the die cavity can then be obtained from the simulations.The operation principles and the interactive analysis system developed are then tested on an automotive wheel made by the low pressure casting process to demonstrate how the system can aid in determining the proper pressure±time relations, the p±t curve, required to produce a sound casting without sacri®cing productivity.