By assuming a linear cumulative damage hypothesis for strain cycling, theories are developed in this paper for fatigue life under varying strain amplitude, residual ductility after a given history of strain cyling, and life under mean strain condition. These predictions are based on uniaxial completely reversed cycling behavior. The expressions obtained were compared with the experimental results on 2024-T351 aluminum alloy as well as other data available in the literature. The agreements between theories and experiments were found to be satisfactory.

The anelastic behavior of polystyrene has been studied by means of creep tests under long-time load application and by means of damping capacity tests under rapidly varying repeated loading. Tensile creep data taken at various stress amplitudes reveal that the log of the creep rate (at 1000 hours) varies linearly with the log of the stress amplitude. A similar type of variation is obtained when damping capacity or energy absorbed per cycle is plotted against stress amplitude. From these two sets of data, the creep rate is found to be proportional to the square of the damping capacity. It would thus appear possible, for polystyrene at least, to predict 1000-hour creep rates from short-time measurements of absorbed energy under dynamic loading conditions. The data obtained from the creep and damping tests, together with additional data from short-time tension and compression tests, seem to be consistent with an internal structure in which the linear polymer chains and groups of chains are in ordered or partially extended positions, but in which, in the absence of stress, no preference is shown for any particular direction. Under the action of stress—particularly if the stress is maintained for a long period of time—a tendency exists for the ordered regions to orient in the direction of the applied stress. The so-called ``crazing'' condition which has been observed to occur in the creep specimens is probably a manifestation of this orientation. X-ray evidence appears to support this point of view.

No abstract

No abstract

No abstract