Abstract
Stepped creep bars providing four different stresses on each specimen have been tested at 525° C for up to 850 h. Density changes in each section following creep are used to determine the effect of the test variables on V, the total void volume/unit weight. It is shown that the density changes are described by the relationship V = 1·52 × 10−38 ɛ t σ7·5, where ɛ is the strain, t the time in h, and σ the initial stress in lbf/in2. This equation is valid to the point of rupture and calculated rupture times are in good agreement with actual values. However, since a knowledge of the value of V at rupture is required for such a prediction, it is doubtful whether the approach will be suitable for purposes of extrapolation. It is demonstrated that some of the assumptions made in the past regarding cavitation damage are untenable. In particular, the amount of damage at failure or at any stage of the creep curve is dependent on the test conditions. There is no reason to suspect that the development of cavitation is related to any stage of the creep curve or that necking is always associated with a particular stage of cavitation.
Get full access to this article
View all access options for this article.
