A Study of Blended Woolen Structures

Abstract

At the time of publication of Parts I and II [1, 2], the mathematical approach to establishment of criteria for judging blend distribution had been governed by two factors. First, the experimental procedures involved microscopic examination and counting of fiber sections at yarn cross sections. Second, the materials under exami nation were stock-blended woolen yarns where the manifest objective of the fiber mixing processes was to achieve random intermingling. It was desired to extend the appropriate probability theory considerations to a more general level. This has now been accomplished, and suitable statistical criteria have been derived for use with experimental methods of measuring blend composition which involve gravimetric analysis of finite lengths of yarn, roving, or sliver. To this extent the utility of the earlier work has been considerably expanded. Beyond this, however, the question has been examined as to whether or not the same basic criteria of blend uniformity apply to worsted-type yarns where blending is accom plished by doubling of slivers. It is shown that such methods can be expected to yield no better than a random blended yarn and therefore are subject to the same limiting ideal.