High Speed Tensile Testing of Fibers

Abstract

The requirements placed on the stress recording device of a tensile tester (usually a stress-transducer and recorder) become more stringent as the time to complete the test decreases. Untess these requirements are met, the difference between the true stress- strain curve and the indicated stress-strain curve becomes so large as to make the latter meaningless. In this paper, the response of a stress-strain tester to the forces arising in testing a sample having a linear stress-strain curve is examined theoretically. An equation is derived which gives the indicated stress in terms of the true stress, the strain, the rate of strain, the resonance frequency and damping of the stress-transducer, and the upper and lower cut-off frequencies or bandwidth of the amplifier-recorder system. Wave effects, however, are not considered. This equation has been solved for a variety of conditions using a small scale digital computer. Indicated stress-strain curves are presented showing their increased deviation from the true stress-strain curve as the rate of testing is increased, as the resonance frequency is lowered. or as the bandwidth is decreased. These results show that certain minimum requirements involving the ratio of the rate of strain to the resonance frequency, damping. and upper and lower cut-off frequencies must be met for accurate indication of stress. As each requirement fails to be met, certain characteristic faults in the recorded stress-strain curve appear. The principles developed above were put to use in designing and constructing a simple stress-strain tester capable of true stress recording at a strain rate of 104 % min.. using a 5-in. gauge length. This tester is described and data taken on textile fibers at 1, 102, and 105 % min. strain rates presented by way of example. The structural basis of fiber behavior at different experimental time-scales is discussed briefty.