Search

A mechanical modelling is proposed in order to describe viscoplastic behaviour without hardening of a nickel-base super alloy loaded at high temperature (900 °C) with strain rates varying within a wide range (from 10−1 to 10−4 s−1). A mathematical law is associated to the viscoplastic model; the parameters of the law are identified from monotonic biaxial tests of membranes loaded by pressure of inert gas (disk pressure testing under helium). The viscoplastic law provides calculated stresses with accuracy better than 1% at the highest strain rates and 4% at the lowest strain rates; the identified yield stress is a logarithmical function of strain rate as for other metallic alloys studied in the bibliography. The parameters identified from biaxial tensile tests of disks have been successfully used to calculate the stresses during stabilized tension–compression loops of cylindrical specimens. The proposed experimental method and behaviour model are interesting because the disk biaxial testing is much more easily performed at high temperature than the tension–compression testing of cylindrical specimens. [ABSTRACT FROM AUTHOR]

Numerical calculations are becoming more and more efficient in estimating the lifetime of structures under thermomechanical loading. However, these life estimations cannot be reliable if the necessary parameters have not been correctly identified and measured and if all the causes of damage have not been considered. Disk testing under gas pressure is similar to oil bulging testing. However, disk testing can easily be used for the mechanical characterization of materials subject to more varied solicitations: monotone loading (biaxial rupture tests at strain rates from 10−6 to 100 s−1), constant loading under high stresses (sustained load) at elevated temperature (creep tests), cyclic loading (mechanical slow fatigue tests); the temperature may be chosen between 20 and 900 °C and the environment may be studied by comparing the results obtained with either an inert gas or reactive gas. Moreover, disk testing reveals light damage since crossing cracks through the thin membrane create leakages detected by a mass spectrometer. In this paper, we present an original method of calculation developed to determine the true mechanical properties of the pressurized disk; the method of calculation is validated because its numerical results are identical to the measured tensile properties. In addition, the range of uniform deformation is correctly determined; this property is needed to establish sheet formability which is not clearly determined by oil bulging. Of course, the mechanical behaviour can be determined within the whole ranges of temperature and strain rates; such wide ranges cannot be tested by other techniques such as tensile testing or oil bulging. As disk edges are not stressed during testing, the results are very reproducible at any temperature and at any strain rate while the machining or cutting defects initiate very scattered ruptures of tensile specimens tested at high temperature or at high strain rate. The disk and its loading simulate real applications with thin walls embedded by thick parts such as thermal exchangers or spatial engines. The analytical method of calculation may be used for identifying the needed parameters of thermomechanical modelling; it will be optimized by finite elements methods and it would allow a rational quantification of hydrogen embrittlement. [ABSTRACT FROM AUTHOR]