The effect of sand material type on the heat absorbing properties of molds in the expendable pattern casting (EPC) process of aluminum alloy castings was investigated experimentally. The bulk densities and thermal conductivities of dry packed beds composed of several types of sand were measured. Because the true density of sands used in this study was higher than that of natural silica sand, the bulk density of the sand packed beds was also higher. The thermal conductivities of the packed beds of zircon sand and steel shot used in this study were almost the same as that of natural silica sand. The thermal conductivities of the dry packed beds composed of artificial sand were slightly smaller than that of the natural silica sand. Using test sands used in this study, an aluminum alloy plate was cast by the EPC process. The solidification time was obtained from the cooling curve. Compared to the solidification time for natural silica sand, the times for artificial sands were longer, while those for zircon sand and steel shot were shorter. Considering the interfacial resistance at the molten metal - sand mold interface, the solidification time and heat diffusivity showed relatively good correlation.

  The effects of the sand material and vibration condition of flask on the appearance of aluminum alloy castings in the expendable pattern casting (EPC) process were investigated experimentally. The sand fluidity under vibration was measured in the vibrating flask for fused artificial sand and natural silica sand. It was confirmed that the sand fluidity was best under rotary motion vibration in vertical plane (rotary vibration), followed by horizontal vibration, then vertical vibration. When fused artificial sand with a nearly spherical shape was used, the sand fluidity under vibration was better than that of natural silica sand. The simple shape aluminum alloy castings with ribs was cast by the EPC process using natural silica sand and artificial sand. The casting appearance and dimensional difference with the expendable polystyrene (EPS) pattern were examined. For both natural silica sand and artificial sand, the casting appearance and dimensional difference with EPS pattern were best under rotary vibration, followed by horizontal vibration, then vertical vibration. When the vibration frequency was sufficiently high at 45Hz, the appearance and dimensional accuracy of castings were good, regardless of the sand material. When the vibration frequency was low at 35Hz, leaching of the molten metal occurred due to cracks in the coat wall, causing the positive dimensional differences in castings with EPS pattern. However, when artificial sand was filled, even at the slightly low vibration frequency of 40Hz, the casting appearance was relatively good, and the dimensional differences of the castings ware insignificant. The change in the sand pressure between the ribs of EPS pattern was measured using a pressure sensor. Castings with good appearance were obtained even with slightly increased sand pressure due to the inflow of sand.

  In order to decrease the amount of air in a polystyrene foam pattern, we replaced the gas in the pattern with argon, and then measured the pattern decomposition rate and mold filling rate in the evaporative pattern casting of molten cast iron. The gas substitution in the pattern was performed by keeping the pattern in an argon atmosphere after the pattern was placed in a vacuum vessel under reduced pressure. The amount of air in the pattern after gas substitution was found to decrease with increasing holding times at reduced pressure and in an argon atmosphere. The pattern decomposition rate immediately after pouring tended to increase with increasing holding times at reduced pressure and in the argon atmosphere, however the decomposition rate slowed down and the thermal decomposition gas layer thickness increased with increasing time after pouring. In the initial stage of the mold filling, the mold filling rate tended to be faster with increasing argon atmosphere holding time during gas substitution, and slower with longer flow length of the molten metal. The decomposition rate of the pattern was slower when the pattern was coated after gas substitution than when the pattern was coated before gas substitution, suggesting that gas substitution efficiency is higher without coatings.

  Mold strength is an important control point for mold management. Shell mold strength varies under the influence of the resin coating that accumulates in sand grain contacts. Therefore, quantitative evaluation of the resin coating is important for mold management. In this study, in order to perform quantitative evaluation, we cut and ground shell molds to produce smooth coating surfaces. A scanning electron microscope was used to obtain backscattered electron images of the smooth surfaces. The cross-section of the resin coating and contact length of the sand coating containing sand grains (observed in the backscattered electron images) were measured. Due to the difference in the perimeter of each sand grain, the contact lengths of the resin coating with the sand grains varied. The results indicated a high correlation between “resin contact length/mean perimeter × 100” and mold strength. However, this correlation was negative for samples with varying blow pressure, and positive for constant blow pressure. This indicates that by varying the blow pressure and filling molds, different sized sites can be formed to make resin coating accumulate in the sand grain contacts. These sites were the first to affect the mold strength. Additionally, it was found that the resin quantity, resin type, silica sand type, silica sand grain size, artificial and special sand, and shell type at these sites are factors influencing the formation of resin coating and regulating mold strength.