The effect of material and process parameters on the frictional conditions in hot flat rolling of steels

Abstract

An extensive investigation on the frictional conditions in laboratory and industry is undertaken due to increased demands on accuracy in roll separating force predictions, as well as demands on understanding of roll wear.

In an attempt to clarify the role of process and material parameters, including the role of scale, on the frictional conditions in hot flat rolling, a large set of controlled laboratory investigations is undertaken. Two steels, one low-carbon and one Nb-treated micro-alloyed steel are investigated. The experimental findings from the laboratory are compared to findings obtained by an analysis of hot strip mill logbooks.

In calculating the coefficient of friction, the laboratory investigations are analysed in a unique manner, as a simultaneous match in measured and calculated roll separating force, roll torque, and forward slip is achieved. The effect of scale on the coefficient of friction is investigated after a scale growth kinetics study, which results in an equation relating scale thickness to time, temperature, and environment. The laboratory rolling experiments show that the coefficient of friction increases with decreasing rolling temperatures, decreasing roll velocities, decreasing scale thickness, and to a lesser degree, increasing reductions. Most of these cause increased material flow stress, increased adhesive bond strength and increased number of adhesive bonds.

Similarities are found in the analysis of the mill logbook data, in which the measured and calculated roll separating forces are matched. It is shown that the coefficient of friction increases with decreasing temperature, increasing amount of alloying elements, increasing relative velocity, as well as increasing material flow stress. Moreover, it is shown that the coefficient of friction is low in the first stands, increases stand by stand, and results in sticking friction in F7. The introduction of a dimensionless parameter, ^, a function of the properties of the interface, the strip, and the work rolls, successfully describes the general frictional conditions throughout the finishing train. It is shown that the predominant wear mechanism in the first 4 stands is thermal fatigue, whereas abrasive wear is more severe in the last 3 stands of a 7 stand finishing train. A study of the non-steady-state condition of rolling with freshly ground rolls indicates that high-speed steel work rolls in stand F3 initially produce a low coefficient of friction. As a thin chromium oxide layer develops on the surface, the coefficient of friction increases, reaches a plateau, and then drops to a steady-state value after a total contact time of 200 s.

In comparing the findings from laboratory and industry it is shown that laboratory hot rolling experiments may be compared to data from industry by correcting for the most obvious difference in geometry - the roll radius. Furthermore, it is shown that, by the use of the ^ - parameter, tribology of hot rolling can be considered in terms of regions of different modes of lubrication. These are boundary, mixed, and quasi-hydrodynamic lubrication. Previously, this theory has been limited to the tribology of cold rolling. It is shown that although the temperature has an effect, two parameters, the scale thickness and the relative velocity, appear to control the mode of lubrication.