An improved method for the determination of the maximum thermal stress induced during a quench test

Abstract

In the evaluation of the thermal shock resistance of a material, a standard procedure is to heat a number of bend specimens in a furnace to a series of selected temperatures, quench the specimens into a cooling medium, and then determine their bending strengths. The thermal-shock resistance is related to that temperature difference, DTc, between that of the furnace and the cooling medium at which the bending strength is abruptly lowered as the result of the formation of thermal cracks. The larger the value of DTc, the better the thermal shock resistance is considered to be (1–5). However, because the amount of data obtained in this type of testing is limited, it is difficult to investigate the details of the circumstances leading to the development of thermal stresses and the formation of thermal-shock induced cracks by this method. Moreover, uncertainty with respect to the thermal boundary conditions between the specimen surface and the cooling medium inhibits a theoretical analysis because of the dependence of heat transfer on time, specimen size and shape, the characteristics of the cooling medium, and the specimen’s surface condition (6–7). A modified test procedure is needed in order to obtain the data needed for understanding of the circumstances resulting in thermal crack formation. In order to use a material safely under thermal shock loading, the effects of repeated thermal shocks as well as of a single thermal shock on crack growth behavior and fracture toughness are of interest, and a critical parameter is the value of thermal stress developed under thermal shock conditions. Ishihara et al. (7–10) have utilized an improved thermal shock test which involved the determination of the transient thermal stresses immediately following a quench. These thermal stresses were calculated based upon measured temperature distributions which were obtained as a function of time following a quench. In the present paper a more convenient method for obtaining the maximum thermal stress in a thermal shock test without the need for measuring the temperature distributions will be described. In order to obtain experimental data for use in the analysis, thermal shock data using the improved thermal shock test procedure were obtained. Three types of hard, brittle materials were used in this study, a cemented carbide, a cermet and a ceramic (SN-220) which were produced by the Kyocera Corp. In addition tests were also carried out on a medium carbon steel, JIS45C, for purposes of comparison. Pergamon Scripta Materialia, Vol. 41, No. 5, pp. 553–559, 1999 Elsevier Science Ltd Copyright © 1999 Acta Metallurgica Inc. Printed in the USA. All rights reserved. 1359-6462/99/$–see front matter PII S1359-6462(99)00185-2