A new algorithm for simulating mold filling of castings has been developed, based on the Direct Finite Difference Method (DFDM) and orthogonal-nonorthogonal-mixed meshing. In order to increase the accuracy of simulated results and to use the advantages of orthogonal meshing, i. e. shorter CPU time and smaller memory size, nonorthogonal elements were introduced at the boundary regions of curved or sloped mold and cavity, while orthogonal meshing were used in the inner region. The accuracy of the method was examined by comparing the simulated results with the experimental ones. It was found that as compared with the method using orthogonal elements, the proposed method resulted in better agreement with the experimental one, while the CPU time was twice longer, and the memory size was almost the same. The proposed method may be more practical than the conventional one.

抄録全体を表示

  Three point bending tests of U-notched, V-notched and unnotched rectangular bar of ferritic spheroidal graphite cast iron were carried out. The bending strength of both notched and unnotched specimens is generally larger than the tensile strength and roughly equal to the fracture strength of the iron. The notch bending strength ratio of the U-notched specimen is 0.9 to 1.0, while that of the V-notched specimen 0.8 to 0.9. The crack initiation energy U_i, that is the bending energy required prior to the maximum load, is proportional to W₀^1.5, where W₀ stands for the width of the minimum cross section. And the ratios of U_i for U-notched and V-notched specimens to that for unnotched are 0.13 and 0.05, respectively. The crack propagation energy U_p, that is the bending energy required posterior to the maximum load, is expressed as U_p = 0.105 W₀² for both types of notches. From the results of interrupted bending tests, the critical stress intensity factor seems to remain almost unchanged during crack propagation. The load-deflection relationship in crack propagation derived from the strength and energy also agrees well with the experimental data.

抄録全体を表示

  The relationship between the dissolved oxygen content (sol-O), eutectic temperature (TE), and the graphite shape was inspected by adding a spheroidaizer, mischmetal and tellurium to molten cast iron and holding the melt. The results were as follows : 1) In case of sol-O < 0.0001 %, higher TE led to a spheroidal the graphite shape and lower TE led the graphite shape to the vermicular. 2) In the case of sol-O > 0.0001 %, the graphite shape became type-A when TE was high, and type-D when TE was low. 3) It was also found that it became cementite when TE was close to an unstable eutectic temperature regardless of sol-O. These results show that the graphite shape can be estimated by the measurement of sol-O and TE.

抄録全体を表示

  The authors prepared various oolitics sand by coating Na and Ca bentonite oven silica sand within 4∼16 mass % and burning it at 1273 K. According to the specific gravity and water absorption quantity of ootitics, it was calculated that oolitics had a 5∼15 vol % void which absorbs water. According to the change of the grain fineness number, that the average oolitics thickness was calculated to be 4∼7μm. The authors prepared green sand by adding bentonite to oolitics sand. Green sand properties such as moisture, test piece density, dry compression strength, maximum expansion and scabbing time changed in proportion to the oolitics quantity. Green compression strength and surface stability were increased by adhering oolitics, however the ratio was low when more oolitics adhered. The authors have arrived concluded that oditics affected the properties of green sand due to (a) Water absorption to void part, (b) Decrease of bulk specific gravity, (c) Increase of friction resistance, (d) Granulation, (e) Expansion absorption.

抄録全体を表示