Validation of constraint‐based methodology in structural integrity of ferritic steels for nuclear reactor pressure vessels

Abstract

VOCALIST (validation of constraint-based methodology in structural integrity) was a shared cost action project co-financed by DG Research of the European Commission under the Fifth Framework of the European Atomic Energy Community (EURATOM). The motivation for VOCALIST was based on the understanding that the pattern of crack-tip stresses and strains causing plastic flow and fracture in components is different to that in test specimens. This gives rise to the so-called constraint effect. Crack-tip constraint in components is generally lower than in test specimens. Effective toughness is correspondingly higher. The fracture toughness measured on test specimens is thus likely to underestimate that exhibited by cracks in components. The purpose of VOCALIST was to develop validated models of the constraint effect and associated best practice advice, with the objective of aiding improvements in defect assessment methodology for predicting safety margins and making component lifetime management decisions. The main focus in VOCALIST was an assessment of constraint effects on the cleavage fracture toughness of ferritic steels used in the fabrication of nuclear reactor pressure vessels, because of relevance to the development of improved safety assessments for plant under postulated accident conditions. This paper provides a detailed summary of the main results and conclusions from VOCALIST and points out their contribution to advances in constraint-based methodology for structural integrity assessment. In particular, the output from VOCALIST has improved confidence in the use of KJ-Tstress and KJ-Q approaches to assessments of cleavage fracture where the effects of in-plane constraint are dominant. Cleavage fracture models based on the Weibull stress, ?W, have been shown to be reliable, although current best practice advice suggests that ?W should be computed in terms of hydrostatic stress (as distinct from maximum principal stress) for problems involving out-of-plane loading. Correspondingly, the results suggest that the hydrostatic parameter, QH, is the appropriate one with which to characterize crack-tip constraint in analysing such problems. The materials characterization test results generated as part of VOCALIST have provided added confidence in the use of sub-size specimens to determine the Master Curve reference temperature, T0, for as-received and degraded ferritic RPV materials. The usefulness of correlating the Master Curve reference temperature, T0, with the constraint parameter, Q, has been demonstrated; however, the trend curves derived require further development and validation before they can be used in fracture analyses. The output from VOCALIST has contributed in providing the validation of methodology necessary to underpin the diffusion of constraint-based fracture mechanics arguments in RPV safety cases, with potential applications including WWER as well as Western-style LWR reactor types

Constraint issues in cleavage fracture and matters arising from practical application

“…The ferritic material has been used as a test material in various national and international projects 1–5 . The conducted experiments within these projects form the basis for intense research work on transferability issues of specimen to component behaviour.…”

“…Three point bend specimens (in this case pre‐cracked Charpy size) in two constraint conditions and at one temperature provide the basis for the calibration (tuning) of the micromechanical models as described in the section, ‘Local approach model based on the Weibull stress’. These results have been gained from the European project VOCALIST 2 . The following results are used: shallow flaw a/W = 0.1 and deep flaw a/W = 0.5 specimens.…”

“…The cruciform specimens were tested under biaxial (1:1) loading. In the project VOCALIST 2 seven cruciform specimens were tested 8 , six of the cruciform specimens were tested as foreseen under biaxial (1:1) loading, with equal loads applied to each beam arm. However, for one specimen the load application system malfunctioned, such that the load ratio was variable.…”

“…The experimental data used for the calibration should have a visible fracture toughness shift at the same test temperature to achieve different crack‐tip constraint conditions. In this analysis the experiments on pre‐cracked Charpy specimens with different aspect ratios from the project VOCALIST 2 are used. One set (set A) is machined and tested with a ratio a/W = 0.1–0.14 being representative for low constraint levels, the second set (set B) has a standard a/W = 0.5 ratio to produce the high constraint data.…”

“…From the calculations of the pre‐cracked deep flaw Charpy bend specimen and the C(T)‐specimens a shift of Δ T 0 =−8.5 °C is obtained. Experimental evidence of the numerical outcome is given by results obtained in 1 and 2 of the German ferritic RPV steel in the initial condition, which has also been used for the calibration of the numerical model. Here a shift in reference temperature Δ T 0 =−10 °C has been observed.…”