Random winding is widely used for the preparation of yarn packages used in a variety of textile processes. Ribboning caused by overwinding of yarn turns at certain places
within the package is a basic problem encountered in this process. Standard random winders incorporate some means of ribbon breaking, which strictly speaking have limited effectiveness.
The research reported in thesis was undertaken to realise a random winder capable of detecting the occurrence of ribboning, and taking the ribbon breaking action at these
precise times, and thereby achieve 'active' ribbon breaking. Such a winder was realised by the addition of suitable transducers to a standard random winder, which could be controlled by a PC. Winding trials carried out using this equipment established the effectiveness of the concept and also the greater freedom from ribboning the method achieved in comparison with other available methods. Unwinding trials carried out by suitably modifying the winder for measuring yarn unwinding tension were also used to establish the effectiveness of active ribbon breaking.
Preliminary experiments carried out on the above apparatus showed the possibility of constructing a Angle of Double Traverse (ADT) diagram on the VDU of the PC during
winding operations, which could serve as a useful aid to follow the progress of the winding operation. It was very useful for showing the occurrence of major and minor
ribboning in the wound package, and also for visually indicating the effectiveness of the ribbon breaking procedures.
The diagram was of further use in studying the nature of package driving by the grooved drum. The case of driving a deformable body such as a yarn package on a comparatively non-deformable grooved drum is important for understanding how the package rotates without any obvious slipping. By varying the winding conditions, it was in fact shown that some measurable slip occurs, and these relationships were
experimentally established.
Computer simulation is useful to determine the stresses within a wound package, particularly as these cannot be measured using conventional techniques. The results of
such a simulation carried out on a random wound package and the comparison of the results with those obtained for a precision wound package are also presented.