Computer simulation of solidification of casting has been newly developed and successfully applied industrially. Forward finite difference method was used for the analysis of two-dimensional and three-dimensional radial symmetric heat transfer in sand mold casting. As for the analysis of complicated shapes, a two-dimensional converting method based on Chvorinov’s rule was newly proposed. Data input in this numerical simulation is interactively performed by the computer graphics system consisting of a tablet/digitizer and a graphic display, and the result is graphically output in three forms : temperature distribution, solidification time and time-temperature plot. The simulation result was compared with that which was experimentally obtained in iron castings and a satisfactory coincidence was obtained for demonstrating the capability of predicting solidification process and shrinkage defect.

View full abstract

  Mechanical properties of the 17-4 PH cast steel, designated as SCS 24 in JIS, sometimes fail to meet the specification values, even with a chemical composition and heat treatment within the ranges dictated in the specification. The results of an experiment indicated that ranges of Ni and Cr contents in the steel narrower than the specification were necessary, if sufficient mechanical properties were to be assured. Ranges of temperatures of heat treatment must also be made narrower than the specification. Extremely low ductility of the steel occasionally found in practice was due to an abnormally high concentration of minor impurities of Sn and Pb. The relation was partly confirmed by an experiment where elongation decreased when Sn content exceeded 0.025%, but the Pb content below 0.028% did not apparently affect the ductility of the steel. Satisfactory properties are being consistently obtained in the shop, after strictly controlling the chemistry and heat treatment following the results of the present study.

View full abstract

  In permanent mold castings, the accuracy of the size of casting is greatly affected by the deformation of the mold. The deformation behavior is influenced by the degree of plastic deformation in the mold surface. The relation between the surface condition of the mold and the deformation behavior in laboratory test using a simple-shaped mold was examined. A test piece having a flat plate form was included in the mold. The one end of the test piece was fixed with a bolt, and the other end was made to shift with the deformation of the test piece. The magnitude of the deformation was measured by a dial guage. The surface condition of the test piece was varied by enveloping and grooving. The magnitude of deformation was markedly influenced by the surface condition. When the surface of steel plate was displaced with a gray iron by enveloping, the magnitude of deformation was minimal when the thickness of the surface layer of gray iron was made to be about 1/3 of the whole. When grooving on some steel plates, markedly diminished the deformation.

View full abstract

  A plain carbon steel was poured at 1,560°C into Y-block CO₂ process mold in which 50φ×50mm sand test pieces were set to make its end surface contact with molten steel. The 50φ sand mold surfaces with zircon facing and without facing were observed by scanning electron microscope in order to obtain new information. In the green sand mold, the sticky features due to combination with bentonite were recognized in 1 to 6 hrs after molding, and the dry features in 24 hrs. When α-starch was added, the sticky features were recognized even 24 hrs. In the CO₂ process mold more firm features were recognized. In the case of zircon facing, the mold surface became flat, hence the difference of the surface feature between green sand mold and CO₂ process mold descreased. The feature of steel casting surface was changed remarkably by zircon facing. In the case of zircon facing, the mold-metal interface became flat, and the mold-metal interface layer easily peeled off.

View full abstract

  In order to investigate the behavior of chromium during eutectic growth, eutectic cast irons containing about 15%Cr and 30%Cr were quenched into water spray while freezing unidirectionally and the distribution of chromium in the melts around the eutectic cells were quantitatively analyzed with electron probe microanalyzer. As the carbide spacing of the quenched eutectic varied from 1 to 4μ and the chromium concentration of the carbides was much higher than that of the matrix, the intensity of characteristic radiation of chromium fluctuated remarkably when the eutectic was analyzed with finely focused beam. The size of the region from which the X-ray is irradiated should be three times as large as the carbide spacing in order to make the fluctuation of the intensity of chromium small enough to get accurate chromium concentration of the eutectic.
  The chromium concentration of the melt was nearly uniform in the plane perpendicular to the freezing direction. When proceeding from the freezing front to the bottom of the cell boundary, the concentration gradually decreased from 14.8% to 8.0% in the 15%Cr iron and from 28.3% to 19.6% in the 30%Cr iron. This distribution of chromium in the freezing direction was in good accordance with that expected from the Fe-Cr-C diagram.

View full abstract

  Rotary bending fatigue tests were carried out on a series of spheroidal graphite cast irons with various matrix hardness ranging from about 170 to 640 in Vickers hardness. Fatigue crack initiation and propagation were observed at the root of the notched test pieces. Many cracks were observed to initiate at the edges of graphite nodules by stress concentrating effect of graphite nodules. These cracks propagated to the final failure by connecting the graphite nodules when the applied stress was higher than fatigue limit. On the other hand, non-propagating cracks were found when the stress was lower than fatigue limit. By analyzing the fatigue test data using the theory of fracture mechanics, a few equations were derived, which relate the fatigue limit and the notch effect of graphite nodules to the length of non-propagating crack. These equations indicate that the notch effect of graphite nodules on the fatigue limit may be ascribed to the size and is not related to the stress concentration.

View full abstract