Analysis of longitudinal stress imparted to fibers in twisting an optical communication cable unit

In the exploratory fiber optic cables used in the Atlanta Fiberguide System Experiment, 12 optical fiber ribbons each containing 12 fibers are stacked one on top of the other to form a rectangular array of 144 optical fibers. Just prior to sheathing, the array is twisted to a given period (lay) in order to improve its bending properties. Moreover, good cable bending properties dictate short lay lengths. However, short lay lengths result in high longitudinal (axial) stresses and strains in the optical fibers. To obtain high fiber yield in cable manufacturing, such strains must be well below 0.3 percent for current 35 ksi fibers. A model which assumes that each fiber follows a helical space curve is used to calculate an upper bound on the axial stress imparted by the twisting operation. The intent was to use the results to choose a lay length short enough to give acceptable bending properties yet long enough to avoid endangering fiber survival in cable manufacture. Model predictions based on a cable design similar to the one in the Atlanta Fiberguide System Experiment lead to the conclusion that the twist period should be not less than 4 inches.