Метод екстраполяції діаграм тривалої міцності жароміцних матеріалів

Abstract

Наведено метод прогнозування тривалої міцності матеріалів, виходячи з концепції базових діаграм. Запропоновано прості співвідношення, що зв’язують розрахункову довговічність і відповідне їй діюче напруження, визначене по базовій діаграмі. Ефективність методу проілюстровано при розрахунках довговічності сталей SUS 316-HP (18Cr-12Ni-Mo) і JIS SCMV 4NT (2,25Cr-1Mo) для досить широких діапазонів діючих напружень і температур. Показано перевагу даного підходу відповідно найпоширеніших параметричних методів Ларсона-Міллера, Орра-Шербі-Дорна, Менсона-Саккопа, Труніна й ін.

The method to predict the stress-rupture strength of materials using 3000–7000 h test data and the master-curve concept is presented. Simple formulas relating the theoretical lifetime and the corresponding stress determined from the master curve are proposed. To determine accurately the difference between the true stress and the stress found from the master curve, an approximating function of β based on available test data is used. The quantity β is the cotangent of the angle between the ordinate axis and a segment of the stress-rupture curve plotted on log–log scale. The parameters of the approximating function can be assumed to be either invariant to the test temperature or determined from the difference between the experimental and calculated values for each isotherm. The assumption of invariance is equivalent to the hypothesis that the family of stress-rupture curves can be reduced to a unified curve. The latter approach results in a better agreement between calculated and experimental data. In some cases, however, the assumption that the parameters of the approximating function are constant allows obtaining reliable results with accuracy sufficient for practical purposes. The method is capable of extrapolating the lifetime to values that are more than in 30 times greater than the available experimental data. The efficiency of the method is illustrated by calculating the lifetime of steels SUS316-HP (18Cr–12Ni–Mo) and JIS SCMV 4NT (2.25Cr–1Mo) for quite wide ranges of stress and temperature. The linear dependence of the mismatch function on β is used as the approximating function. Using a function rather than a set of constant parameters, as in the master curve method, always results in a better agreement between calculated and experimental data when extrapolating stress-rupture curves. It is shown that the method has advantages over the master-curve approach and popular parameters such as Larson-Miller, Orr-Sherby-Dorn, Manson-Succop, Conrad’s, Korchynsky-Clauss, and Trunin’s. The parameters of the linear models are calculated using multifactor linear regression.