No abstract

No abstract

We performed environmental fatigue testing in simulated primary water reactor (PWR) primary water and reference fatigue testing in air in the framework of an international, collaborative project (INCEFA-PLUS), where the effects of mean strain and stress, hold time, strain amplitude and surface finish on fatigue life of austenitic stainless steels in light water reactor environments are being studied. Our fatigue lives obtained on machined specimens in air at 300 °C lie close to the NUREG/CR6909 mean air fatigue curve and are in line with INCEFA-PLUS air fatigue lives. Our environmental fatigue lives obtained in simulated PWR primary water at 300 °C lie relatively close to the NUREG/CR6909 mean fatigue curve; derived from the NUREG/CR6909 mean air fatigue curve and the applicable environmental correction factor (Fen). The PWR results show that (1) a polished surface finish has a slightly higher and a ground surface finish a slightly lower fatigue life than the NUREG/CR6909 prediction; (2) the ratio of polished to ground specimen life is ~1.37 at 300 °C and ~1.47 at 230 °C; (3) holds—at zero strain after a positive strain-rate—have a slightly detrimental effect on fatigue life. These results are in line with the INCEFA-PLUS PWR fatigue lives. A novel gauge-strain extensometer was deployed in order to perform a true gauge-strain-controlled fatigue test in simulated PWR primary water.

The penetrations in the early Pressurized Water Reactors Vessels are characterized by Alloy 600 tubes, welded by Alloy 182/82. The Alloy 600 tubes have been shown to be susceptible to PWSCC (Primary Water Stress Corrosion Cracking) which may lead to crack forming. The cracking mechanism is driven mainly by the welding residual stresses and, in a second place, by the operational stresses in the weld region. It is therefore of big interest to quantify the weld residual stresses correctly. In order to determine the welding residual stresses, the weld procedure is simulated numerically by finite elements analysis. In the article, central as well as eccentric sidehill nozzles on the vessel head are analyzed. For the former a 2-dimensional axisymmetrical finite element model is used, whereas for the latter a 3-dimensional model is set up. A nonlinear transient thermo-mechanical analysis is performed, which is preceded by a transient thermal analysis simulating the heating during the multipass welding. Weld beads are deposited “all-at-once”. Different positions on the vessel head are compared and the influence of the sidehill effect is illustrated. The methodology is applied to the reactor vessels of the Belgian nuclear power plants by Tractebel Engineering (Belgium). The results are compared with literature. The global approach in both cases is very similar but is applied to different configurations, specific for each plant.

The penetrations in the early Pressurized Water Reactors Vessels are characterized by Alloy 600 tubes, welded by Alloy 182/82. The Alloy 600 tubes have been shown to be susceptible to PWSCC (Primary Water Stress Corrosion Cracking) which may lead to crack forming. The cracking mechanism is driven mainly by the welding residual stress and, in a second place, by the operational stress in the weld region. It is therefore of big interest to quantify the weld residual stress field correctly. In this paper the weld residual stress field is calculated by finite elements, using a common approach well known in nuclear domain. It includes a transient thermal analysis simulating the heating during the multipass welding, followed by a transient thermo-mechanical analysis for the determination of the stresses involved with it. The welding consists of a sequence of weld beads, each of which is deposited in its entirety, at once, instead of gradually. Central as well as eccentric sidehill nozzles on the vessel head are analyzed in the paper. For the former a 2-dimensional axisymmetrical finite element model is used, whereas for the latter a 3-dimensional model is set up. Different positions on the vessel head are compared and the influence of the sidehill effect is illustrated. In the framework of a common project for Angra 1, Tractebel Engineering (Belgium) and Eletronuclear (Nuclear Utility, Brazil) had the opportunity to compare their analysis method, which they applied to the Belgian and the Brazilian nuclear reactors, respectively. The global approach in both cases is very similar but is applied to different configurations, specific for each NPP. In the article the results of both cases are compared.