Conventional roll-forming practices are used by many major steel companies for the manufacture of thin-walled ERW steel pipe. Modem trends in the rolling industry have been towards producing thinner-walled pipe with higher strength steels, but since many of the existing conventional rolling mills were designed to accomodate heavier-walled pipe, roll forming the lighter sections can result in quality-control problems. As the plate, or skelp, becomes thinner its lateral stiffness decreases, increasing the occurrence of edge buckling that can reduce the overall weldquality. Although some progress has been made towards developing an analysis technique for this very difficult problem, the accuracy of the solution is dependent on an accurate description of the deformed-skelp geometry.
This paper describes an experimental investigation to check the accuracy of a skelp profilegeometry prediction that can be used to model the strain history during forming. Skelp geometries predicted from roll-stand geometries were compared to profiles measured on an actual 16 inch rolling mill. Through this comparison a method of correcting the theoretically-predicted geometry was developed and found to produce good agreement with the mill-measured geometry. The effect that the correction has on the predicted axial-strain is significant, indicating a need for an accurate deformed-skelp description. Conversely, it is suggested that the skelp geometry predicted from the theoretical roll-stand geometries is not sufficiently accurate to adequately define the strain state in the deformed skelp.