A B S T R A C T In this paper, the warm pre-stress (WPS) effect on the cleavage fracture of an 18MND5 (A533B) RPV steel is investigated. This effect, which describes the effective enhancement of the cleavage fracture toughness at low temperature following a prior loading at high temperature, has received great interest in light of its significance in the integrity assessment of structures, such as nuclear pressure vessels, subjected to thermal transients. Several loading cycles between room temperature (RT) and −150 • C are considered: Load-Unload-Cool-Fracture (LUCF), Load-Cool-Fracture (LCF) and Load-Cool with Increasing K-Fracture (LCIKF). All experiments complied with the conservative principle, which states that no fracture will occur if the applied stress intensity factor (SIF) decreases (or is held constant) while the temperature at the crack-tip decreases, even if the fracture toughness of the virgin material is exceeded. The experimental results indicate that an effective WPS effect is present even at small pre-load (K wps = 40 MPa √ m), and that a minimum critical slope (− K / T) in the LCIKF cycle has to be exceeded to induce cleavage fracture between RT and −150 • C. Numerical modelling was performed using mixed isotropic and kinematic hardening laws identified on notched tensile (NT) specimens, tested in tension to large strains (up to 40%), followed by large compressive strains. Detailed microstructural investigations on compact tensile (CT) and NT fracture test specimens were performed so as to determine the nature of the cleavage initiation sites, as well as the local mechanical conditions at fracture. Based on this local information, a new cleavage model was calibrated and applied to predict the probability of cleavage fracture after WPS: it is shown that the predictions are in good agreement with the experimental results.a 0 = crack length in CT toughness specimen (mm) d = distance (mm) K = SIF increment between RT and −150 • C (MPa √ m) (− K/ T) = transient slope of the LICKF cycle between RT and −150 • C (MPa √ m· • C −1 ) K, K el = stress intensity factor (MPa √ m) K frac = stress intensity factor at fracture (MPa √ m) K Ic , K Jc = mode I fracture toughness for the virgin material (MPa √ m)K wps = stress intensity factor after warm pre-stressing (MPa √ m)

No abstract

No abstract

Abstract. Reactor pressure vessels are internally clad with austenitic stainless steel. This welding operation generates residual stresses which can have an important role in integrity assessments. In order to evaluate these stresses, an experimental and numerical programme has been conducted. The experiments includes cladding operations, macrographic analyses, temperature and residual stresses measurements with different methods. According to these measurements, transversal stresses (perpendicular to the welding direction) and longitudinal stresses (parallel to the welding direction) are highly tensile in stainless steel and they are compressive in the HAZ. Finite element calculations were used to simulate both welding operations and post weld heat treatment. These calculations coupled the thermal, metallurgical and mechanical aspects in a 2D representation. Different models were studied including effect of generalised plane strain, transformation plasticity, creep and tempering. The transversal stresses calculated are similar to the measured ones, but the longitudinal stresses showed to be very sensitive to the model used. As expected because of the two-dimension model, the longitudinal stresses can't be well estimated. More work is needed to improve measurements of stresses in depth (important differences appeared between the different methods). A predictive model would be also very usehl to determine the thermal loading which is at present dependant on measurements. A 3D calculation appears to be necessary to evaluate longitudinal stresses.

No abstract