Precision Castparts (United Kingdom)

Mechanical experiments and transmission electron microscope (TEM) observations indicate that single-phase γ-TiAl does exhibit primary, secondary, and inverse creep, but not steady-state creep. Constant stress creep tests of γ-Ti-51Al-2Mn conducted at 550 °C, 597 °C, and 703 °C have been interrupted at different stages in the creep process. The TEM observations of these specimens were used to document the microstructural evolution that occurs during creep. Superdislocation motion was activated and subsequently exhausted during primary creep. Ordinary dislocations were observed to be pinned during primary creep, but with time, these dislocations began to bow past their pinning points. The extended region of inverse creep has been related to the bowing and multiplication of these ordinary dislocations. Quantitative measurements of dislocation density were performed, and while the density of superdislocations remained constant, the density of ordinary dislocations increased by an order of magnitude during the life of a creep test. The acceleration in the creep rate has been related to this increase in the density of ordinary dislocations, but the change in dislocation density was not high enough to account for the increase in the creep rate. This suggests that both the mobility and density of ordinary dislocations increase as creep progresses.

Abstract Nomination sampling is a sampling process in which every observation is the maximum of a random sample from some population. Assuming that all samples are taken from a single underlying distribution F, data may be viewed as consisting of pairs (Xi , Ki ), where Ki is the size of the ith sample and, given Ki = ki, Xi is distributed according to F ki. Willemain (1980), who discussed nomination sampling in the context of health care delivery, proposed an estimator for the median of F under the assumption that Ki = N, a fixed integer. In this article, the assumption of a fixed sample size N from each population is relaxed; with K taken as random, the problem of nonparametric estimation of the distribution function F is considered. The nonparametric maximum likelihood estimator of F is derived, its consistency is demonstrated, and its asymptotic behavior as a stochastic process is identified. Conditions are given under which these asymptotic results hold for nonrandom K. A by-product of this developme...