The Extent of Property Variability on Commercial Paper Machines

For many decades the paper industry has desired to monitor the performance of important paper performance properties continuously in real time. This aim, until now, has been impossible to achieve as modern paper machines have lacked the methods to reliably determine the changes in common paper properties during paper manufacture. This paper describes the technology path taken by researchers aiming to produce a practical on-line paper strength sensor, the issues faced and a demonstration that these problems have now been overcome. On-line, continuous measurement of paper strength allows high data rates facilitating the ability to discern quality trends within the reel and the ability to use standard control methods to remove variations in specified paper properties. These new capabilities allow endemic over-specification to be removed and deliver improved control strategies and shorter control action times. This paper sets out to validate the theoretical approach developed for on-line, real time paper property measurement using data taken from working property sensor installations on commercial paper machines and, supported by historical data gleaned from the available public literature. The work also demonstrates the high property variability within a machine paper roll for commonly specified 'strength' properties. A discussion of the difficulty in setting a robust methodology for the sensor validation and designing an unambiguous demonstration of an on-line strength sensor is presented. The discussion highlights the uncertainty in traditional paper manufacturing performance brought about by the available low sampling rates (top of reel testing) and the poor statistical confidence inherent in the traditional paper test. Finally, analysis of strength property variation with respect to the time scale of the variation shows a power order relationship that will be useful in characterising paper machine performance.