Sliver Structure Analysis by Air Suction

Abstract

Slivers were separated into small fiber assemblies by suction air. The size of separated fiber assemblies flyng in a transport tube was sensitively measured by a photoelectric method and signal was stored in a digital memory. Time series data were subjected to fast Fourier transformation and then to the inverse Fourier transformation.The variation in separation behaviour was analyzed in terms of power spectrum and autocorrelation.The distribution function of the number of fiber assemblies separated within a short period of time exhibits Poisson's distribution. The distribution of separation time interval shows an exponential curve. Power spectrum indicates a tendency of pinknoise and the autocorrelation function decreases to zero after short period of time. From these results, it is concluded that the separation behaviour within a short period of time is random independently of time. This suggests that fibers and fiberassemblies are arranged at random in a short length of sliver.