• PDF / 1,517,279 Bytes
• 12 Pages / 583.28 x 792 pts Page_size
• 5 Downloads / 237 Views

DOWNLOAD

REPORT

---

I.

INTRODUCTION

IN earlier publications 1'2 the

thermal and mechanical behavior of the billet mold have been discussed in considerable detail. Mathematical analyses have been applied in these studies to predict the temperature distribution in the mold wall during casting, and the distortion that arises from differential thermal expansion, as a function of mold design and operating variables. It has been shown that a significant quantity of heat is conducted longitudinally, as well as transversely, in the mold wall near the meniscus with the result that the maximum temperature in the wall is not at the meniscus, but roughly 30 mm below it. The mold temperature distribution was also seen to be strongly influenced by the velocity of the cooling water, wall thickness, and the carbon content of the steel being cast. The stress analysis revealed that during casting, the thermal-expansion forces cause the mold to assume a characteristic shape with an outward bulge, away from the steel, in the meniscus region. The point of maximum bulging was found to occur below the meniscus in the area of the peak wall temperature. Above this level the mold has a negative taper of 1 to 2 pct/m, while below it the predicted taper is positive, with a value of roughly 0.4 pct/m. In the region of negative taper the corners were found to move farther away from the steel than the mid face. It was also shown that in the meniscus region, the mold material may yield and thus distort plastically. The extent of plastic distortion was found to depend on the mold wall thickness, type of mold constraint, meniscus level, and cooling water velocity. In general, the dominant factor affecting distortion is thermal expansion in the transverse plane rather than bending caused by the temperature gradients through the thickness of the mold wall. These results provide an interesting glimpse into the thermal and mechanical working of the mold, but are most valuable if they can be applied to the solution of quality and operating problems that are known to be mold related. These problems, the most important of which are longitudinal I.V. SAMARASEKERA, Lecturer, and J. K. BRIMACOMBE, Stelco Professor of Process Metallurgy, are both with the Department of Metallurgical Engineering, University of British Columbia, Vancouver, BC, V6T 1W5, Canada. Manuscript submitted December 29, 1980. METALLURGICAL TRANSACTIONS B

9

comer cracks, rhomboidity, and breakouts, have been discussed at length in the literature but have not been linked mechanistically to mold behavior. The objectives of the present study, therefore, have been to develop such links and thereby to provide solutions to mold-related production problems. During the course of this work it was found that nucleate boiling, which had not been considered in the earlier study of heat flow,~ was an important factor affecting mold behavior. Thus, this phenomenon has been examined in considerable detail in the present paper. The incorporation of boiling in the heat flow analysis has then led to a reformulation of th