Using roll hardness to screen for excessive web bagginess

Abstract

Various methods are available for quantifying bagginess. But they can be tedious and impractical in a production setting. Often one just needs to know if a roll is too baggy to run without wrinkling during converting. Screening rolls offline can save time and reduce scrap when the alternative is loading the roll and making machine adjustments on the fly to eliminate wrinkles from baggy edges.

In this paper a method for correlating average roll hardness and cross direction hardness profile to runnability using intraclass correlation and binary logistic regression tools is described. Here runnability means running near target line speed without wrinkling. The outcome is the probability that a particular roll will exhibit enough bagginess to impact runnability. Then the decision to reject the roll without running can be made depending on how much risk is acceptable. This information can then be relayed to the web supplier to help reduce bagginess at its source.