| Summary: | A mathematical model has been developed to predict the temperature distribution through a slab during rough rolling. The heat transfer model is based on two-dimensional heat flow, and includes the bulk heat flow due to high speed slab motion, but ignores heat conduction in the rolling direction. At high temperature, an oxide scale forms due to exposure to air and must be considered in the heat transfer analysis. The thick scale which forms during reheating helps to insulate the slab during transportation of the slab to the rolling stands. Prior to rolling, the heavy oxide is removed by the descale sprays, and a much thinner oxide layer is formed during the very short exposure of the slab to air during rough rolling. The temperature results predicted by the model have been validated by comparison with another model. The results show that the work roll chilling has a significant effect on the temperature distribution in the slab in the roll gap and approximately 33%of the total heat lost by the slab is extracted by the work rolls; however, the chilling affect is confined to a very thin surface layer on the slab, approximately 2.5%
of the slab thickness.
To measure the roll chilling effect, pilot mill tests have been conducted at CANMET and UBC. In these tests, the surface and the interior temperatures of specimens during rolling have been recorded using a data acquisition system. The corresponding heat transfer coefficients in the roll bite have been back-calculated by a trial-and-error method using the heat transfer model developed. The heat transfer coefficient has been found to increase along the arc of contact and reaches a maximum and then declines until the exit of the roll bite. It is important to note that the mean heat transfer coefficient in the roll gap is strongly dependent on the mean roll pressure. At low mean roll pressure, such as in the case of rolling plain carbon steels at elevated temperature, the maximum heat transfer coefficient in the roll bite is in 25-50kW/m2-°C range. As the roll pressure increases, the maximum heat transfer coefficient also increases to approximately 700kW/m2-°C. Obviously, the high pressure improves the contact between the roll and the slab surface thereby reducing the resistance to heat flow. The mean roll gap heat transfer coefficient at the interface (HTC) has been shown to be linearly related to the mean roll pressure. These results were employed to calculate the thermal history of the slab during industrial rough rolling; the results are in good agreement with the data in the literature.
In addition to the thermal history, the strain and strain rate distribution also affect the evolution of microstructure of rolled steels. In the present project, heat transfer and deformation during rough rolling of a slab have been analyzed with the aid of a coupled finite element model based on the flow formulation approach. In the model, sliding friction is assumed to prevail along the arc of contact and the effect of roll flattening has been incorporated. The model has been validated by comparing the results from the pilot mill tests. It confirms that the deformation of a slab in the roll gap is inhomogeneous and just beneath the surface very high strain rates of approximately 5-10times the nominal strain rate are reached due to the redundant shearing. The maximum strain rate is attained at the entrance to the roll bite just beneath the rolls. The corresponding strain distribution through the thickness is also non-uniform, being lowest at the center and highest at the surface. The temperature gradient near the surface of the slab is very large due to work roll chilling; this is consistent with results obtained from the finite-difference model. The predicted roll forces are in good agreement with the measured values for the 9-pass schedule currently employed on the roughing mill at Stelco's Lake Erie Works and the pilot mill tests.
|
|---|
