Influence of the casting-nozzle shape on the jet degassing in vacuum casting of metal

Abstract

545 In recent years, the significant expansion of large� ingot production has been associated with problems of metal quality in superlarge components for the power industry and heavy machinery. One challenge is effec� tive degassing of vacuumcast metal so as to reduce the nonmetallic inclusions. The technical specifications for critical components strictly regulate the quantity, shape, and size of such inclusions. Vacuum treatment of metal in ladle-furnace systems and other equip� ment significantly reduces the residual gas content in the steel; the hydrogen content is 1 cm 3 /100 g or less. However, the effectiveness and intensity of degassing largely depend on the nucleation and formation of gas bubbles in the casting nozzles. On the basis of indus� trial data and calculation results, it has been estab� lished that the metal-gas surface in the jet is greatly increased when the steel passes through the internal cavity of the casting nozzle in vacuum casting (1). Most researchers relate the effectiveness of bubble nucleation and development to the state of the noz� zle's internal surface, its roughness, and the refractory composition. However, practically nothing is known about the influence of the nozzle's internal configura� tion on gas liberation from the jet. Round casting noz� zles are used to obtain a compact stable jet when cast� ing steel in air and do not intensify the degassing of the vacuumcast metal. In the present work, we determine the optimum internalcavity configuration of the casting nozzle, corresponding to maximum gas removal from the jet in vacuum transfer of the solution that simulates liquid steel. Comparability of the vacuum processes in steel casting and in the physical model constructed in the laboratory is ensured on the basis of similarity analysis. The laboratory apparatus is shown in Fig. 1. Dimensional analysis allows us to identify the fol� lowing relevant dimensionless variables: Re, the Rey� nolds number; We, the Weber number; and Dif, the diffusion number. The table presents their values in the sample and in the model.