The Creep Recovery of a 0.17 per cent Carbon Steel

Abstract

Since 1876 occasional interest has been shown by investigators in the phenomenon of creep recovery, or what has been sometimes referred to as delayed elasticity or subpermanent elasticity. This may be defined as the negative strain, increasing with time, which occurs when a specimen strained under load, as in a creep test, is unloaded and maintained at its working temperature. Theories have been advanced to account for the phenomenon, but sufficiently full data have not been obtained for any of the commonly used commercial materials to provide means of checking the theories. The present paper, for the case of a 0·17 per cent carbon steel, investigates the relation between recovery and the stress, temperature, period of test, and strain, in the creep test preceding the recovery, and presents data by means of which the existing theories are compared. Creep and recovery tests were made in creep units capable of measuring creep rates of the order 10–8 in. per in. per hour, the test temperature being maintained constant to ±0·5 deg. C. At each of the temperatures 425, 455, and 485 deg. C., groups of creep tests—continued to periods of 0·2, 2, 20, and 145 hours, respectively—were made, each group consisting of tests at three different stresses. Each test piece on unloading was allowed to recover for 100 hours. One or two similar tests were also made at 350 and 550 deg. C. In addition, at 425, 455, and 485 deg. C., groups of creep tests were made, and were carried to the stage of minimum creep rate; the test piece was then unloaded, and allowed to complete its recovery. These groups of tests also consisted of tests at two or three stresses. The author discusses the relation of creep recovery to stress, temperature, period of creep, and strain in the initial creep test, and finally tabulates his principal conclusions which, he points out, apply only to the steel tested, and are not to be taken as generally true for other materials. A bibliography is also included in an Appendix.