In heavy-walled pearlitic spheroidal graphite iron castings, ferrite precipitates along the prior austenite grain boundaries. Using specimens with a modulus of 1.25 to 5 cm, we investigated the cooling curve, microstructure, and element segregation and clarified the causes. The results revealed that the most influential cause was not alloy elements such as Mn and Cu, but the long solidification cooling time in the mold. When the modulus exceeds 3 cm, network ferrite precipitated, along the prior austenite grain boundaries in a network manner. Since this network morphology and distribution may affect the mechanical properties, it is necessary to avoid the precipitation of network ferrite to ensure the required mechanical properties of castings.

  To evaluate and improve the toughness of heavy-walled pearlitic spheroidal graphite iron castings, Charpy impact tests were conducted on specimens with casting modulus of 1.25 to 5 cm and different solidification cooling rates to simulate actual products. As a result, the absorbed energy in the transition region and ductile region tended to decrease as the solidification cooling rate decreased in the as-cast material. The impact transition temperature was higher than room temperature. However, after normalizing, the absorbed energy in the transition region improved and the impact transition temperature shifted to a lower temperature, suggesting that improving the matrix structure is effective for improving the impact properties of heavy-walled pearlitic spheroidal graphite iron castings.

  In this study, we investigated the effects of homogenizing heat treatment on the coefficient of thermal expansion and martensitic transformation behavior of low thermal expansive ductile cast iron (Fe-2.4C-1.6Si-0.1Mn-32Ni-5Co) with eutectic composition. By increasing the holding time at a homogenizing heat treatment temperature of 1100℃, the segregation degree (δ) (which is defined as the ratio of the difference in maximum and minimum concentrations (ΔC) in heat-treated samples to the ΔC in as-cast samples) of several alloying elements decreased to 0.2-0.4 at a holding time of 4 hours, and became 0 at a holding time of 100 hours, achieving complete homogenization. With the decrease in δ, the coefficient of thermal expansion also decreased from 4 × 10^-6/℃ to 3 × 10^-6/℃. However, when the heat treatment time exceeded 4 hours, despite the decrease in δ, the coefficient of thermal expansion remained constant at 3 × 10^-6/℃. We conducted investigations on the possible causes, and found that in as-cast samples, martensitic transformation could be observed in the micro final solidification region. However, after homogenizing heat treatment at 1100℃ for more than 4 hours, martensitic transformation hardly occurred even in liquid nitrogen (-196℃). On the other hand, the sample homogenized at 1100℃ for 1 hour showed an increased martensite area fraction compared to the as-cast sample. This may be because the redistribution of alloy elements during the homogenizing heat treatment process increases the region where the Ms point is higher than the cooling temperature.

  In sand molds made by additive manufacturing, stair-steps are formed on the slope. The effects of outer-shape slope angle θ, pouring temperature and shot blasting on the edge slope angle φ and surface roughness Rz_s of stair-steps were analyzed for iron castings produced by additively manufactured sand molds using statistical analysis methods.

  For φ, a significant main effect of shot blasting treatment was found as a result of a 3-ways ANOVA test. The change in the value of φ due to casting is explained by the drop out of sand caused by molten metal and the transfer of its sand mold shape. The larger the φ, the greater the effect of crushing by shot blasting, but the effect was smaller than that of casting under the conditions of this experiment.

  The means of Rz_s were approximately consistent with the larger one between the Rz_s geometrically-calculated using the mean value of φ and the Rz_s of the horizontal plane. In addition, for the condition of θ ≦ 15°, almost all the measured Rz_s fell within the range defined by the 95% prediction interval of φ. For Rz_s, there was a significant main effect of outer-shape slope angle and a significant trend toward shot blast treatment having a main effect as shown by the 3-ways ANOVA test.