No abstract

In PVP2011-57942 we reported improved endurance in fatigue tests with intermediate annealing to roughly simulate steady state operation between fatigue transients in NPP components. Quantification of this effect is in focus of our continued research on fatigue performance of niobium stabilized stainless steel (1.4550, X6CrNiNb1810mod). Similar effect is expected in nuclear power plants during normal operation — e.g. in a PWR surge line or in pressurizer spray lines. Holds affect cyclic stress strain response. Stress amplitude, tensile mean stress and apparent elastic modulus are increased immediately after a hold, while decreased by cycles in between. Axial shortening is measured during hot holds at zero stress. This all suggest cyclic accumulation of lattice defects and recovery during holds. Recovery may occur through thermally activated dislocation migration together with diffusion, grouping and annihilation of lattice defects. More than one thermally activated processes control the rates of contraction during hold periods at elevated temperatures. Hold hardening delays crack formation by preventing plastic strain localization, in components also on macroscopic level. A mechanism informed model is sought for transferring laboratory data to real plant components in terms of improving accuracy of numerical fatigue usage assessment. Anticipated mechanisms behind gradual changes in material responses are discussed in relation to quantitative effects of holds.

Our experimental research on fatigue performance of stainless steels and transferability of laboratory data to plant operational conditions focuses in niobium stabilized stainless steel (1.4550, X6CrNiNb1810mod) taken from a pipe manufactured as primary piping for a German NPP Good fatigue performance both in air and in PWR water was reported in previous PVP papers. The NUREG/CR-6909 report proposes Fen factors based on stroke controlled experiments in hot water for non-stabilized steels. Since PVP2013-97500 we have new data in 200°C PWR water to compare with predictions by NUREG/CR-6909. Our strain controlled tests in 325°C and 200°C PWR water give longer lives resp. smaller Fen factors. For the slowest tested strain rate 4·10−6 in 325°C water the prediction according to NUREG/CR-6909 goes just below the current ASME design curve, but our results remains well above. Including also the relevant design temperature effect, our result Fen = 4 is well below the predicted Fen = 14,5. The gap is smaller for higher strain rates and low Fen values. Simplified simulations of fatigue transients combined with normal operation indicated that relevant loading patterns as hold-time effects may result to notably longer lives than in standard laboratory tests. A concern was raised on transferability of data to thermal transients separated with months of normal operation. Cyclic strain (transients) followed by hot holds (normal operation) lead to time and temperature dependent hardening with reduction in cyclic plastic strain and fatigue usage, i.e. extension of life. This paper reports new data, challenges met and our progress towards developing realistic design factors for effects both reducing and extending fatigue endurance in nuclear power plant operational conditions.

No abstract

Fatigue of Niobium stabilized austenitic stainless steel (X6CrNiNb1810 mod) was studied using specimens extracted from a solution annealed and quenched primary piping material sample. This paper reports and discusses results of non-standard experiments to complement previously published test data. The NPP primary piping components spend long times in operation temperature between fatigue cycles originating from thermal transients. This was roughly simulated by fatigue tests periodically interrupted for intermediate annealing in elevated temperature. Fatigue endurance was notably increased when low strain amplitudes were used. The life extension is explained by the cyclic stress strain response. Hardening followed by slow cyclic softening was consistently observed after annealing. It is generally assumed that cumulative accumulation of fatigue damage occurs at a wide range of loading amplitudes. We performed two level and spectrum straining tests combining amplitudes above and below the (Nf > 107) endurance limit. The endurance limit seems to be effective also in variable amplitude loading. In terms of modified Miner rule, even “negative damage” was obtained in two level tests below and above the constant amplitude endurance limit. This behavior is linked to prominent secondary hardening of the steel.