To evaluate the structural integrity of nuclear power plant piping, it is important to predict ductile tearing of circumferential cracked pipe from the view point of Leak-Before-Break concept under seismic conditions. CRIEPI (Central Research Institute of Electric Power Industry) conducted fracture test on Japanese carbon steel (STS410) circumferential through-wall cracked pipes under monotonic or cyclic bending load in room temperature. Cyclic loading test conducted variable experimental conditions considering effect of stress ratio and amplitude. In the previous study, monotonic fracture pipe test was simulated by modified stress-strain ductile damage model determined by C(T) specimen fracture toughness test. And, ductile fracture of pipe under cyclic loading simulated using damage criteria based on fracture strain energy from C(T) specimen test data. In this study, monotonic pipe test result is applied to determination of damage model based on fracture strain energy, using finite element analysis, without C(T) specimen fracture toughness test. Ductile fracture of pipe under variable cyclic loading conditions simulates using determined fracture energy damage model from monotonic pipe test.

When cracked bodies are subjected to cyclic loading, fatigue crack growth evaluation is often required from the viewpoint of the assurance of fitness-for-service. For cyclic loading with constant amplitude, crack growth can be calculated by integrating a fatigue crack growth rate law provided by the Paris law in terms of stress intensity factor range. However, the cause of cyclic loading supposed in actual structures is not simple. For an example of LWR (Light Water Reactor) plants, a number of datasets of cyclic loading with different amplitudes is necessary for specific transient events. And the chronological order of individual transients cannot be determined. In this paper, a universal procedure to deal with multiple transients was developed in case that the chronological order of the transients was indefinite. Cyclic loading sequence such that the loadings were lined up in the order corresponding to larger amplitude gave the most conservative crack growth prediction among the possible sequences from the set of the specific transient events. Nevertheless the effect of the sequence was quite limited and the differences in fatigue crack growth were much less than the accuracy of the analysis outputs.

In nuclear power plants, inner surface of reactor pressure vessel, which is made of low-alloy steel, is cladded by austenitic stainless steel to improve corrosion resistance of vessel. In an evaluation of fatigue crack propagation behavior at J-welding portions of bottom mounted instrumentation in reactor pressure vessel, crack propagation crossing dissimilar materials interface between base metal of low-alloy steel and stainless steel cladding is assumed. Fatigue crack growth rate of the cladding is predicted to be captured by that of the base metal because thickness of the cladding is very thin compared that of the base metal. In this study, fatigue crack propagation analyses of cladded plates under cyclic loading were performed using XFEM, which can model crack independently of finite elements, as a preliminary study for crack propagation analyses considering dissimilar materials interface.

In this study, some benchmark problems on fracture tests for circumferentially through-wall/surface cracked pipes were provided. The participants predicted the ductile crack propagation behavior by their own approaches, including nucleation, growth, and coalescence of voids simulated by Gurson model, ductile crack propagation using stress modified fracture strain (SMFS) model, J-integral based ductile crack propagation using XFEM, CTOA based ductile crack propagation using FEM, stress triaxiality and plastic strain (STPS) based ductile crack propagation using FEM, and ductile crack propagation using peridynamics. Among them, GTN, CTOA and STPS models were not applied to surface crack problems. Discrepancies between the experimental maximum loads and calculated maximum loads were within 10% in most cases and 25% in the maximum case. Element size dependency of analysis parameters were considered in SMFS and GTN models while those were determined from independent material tests. Gurson model can predict slanting crack propagation directions. XFEM which did not need analysis fitting parameters cannot analyze beyond the peaks of load-LPD curves. Crack propagation directions were given and fixed in both CTOA and STPS models. Parameters in Gurson model and peridynamics were optimized to reproduce load-LPD curve in one of the benchmark problems.

In Japan, a new standard of an assessment procedure for crack-like flaws in pressure equipment at elevated temperature is now under development in the High Pressure Institute of Japan (HPI). In this standard development, it is needed to adopt reference stress solutions for crack-like flaws in pressure equipment being subjected to membrane stress and/or bending stress. Such reference stress solutions have been proposed in various references such as ASME FFS-1/API579-1, BS7910, R5, FBR draft guideline, HPIS Z101-2, etc. A comparative study of those reference stress solutions was conducted in order to select appropriate one. As a result, reference stress solutions in HPIS Z101-2 were adopted. The sufficiency of adopted reference stress solutions was introduced in this paper. Also, the reference stress solutions for axially and circumferentially through-wall rectangular flawed cylinders, which were not provided in the HPIS Z101-2 standard but were utilized to derive those solutions adopted in the standard, were introduced in this paper. These solutions should be adopted in a new HPI standard for crack-like flaws in pressure equipment at elevated temperature.

In the 2012 version of the JSME Fitness-for-Service Rules for Nuclear Power Plants, the procedure to calculate screening parameter, SC , which is used for selecting the analysis method (limit load controlled by plastic collapse, elastic-plastic fracture mechanics, or linear elastic fracture mechanics), has been revised to reflect a semi-elliptical surface crack. Both limit load solution and stress intensity factor solution are needed to calculate SC , and the solutions for a semi-elliptical surface crack are different from those for a fan-shape surface crack. In this study, the effect of the difference in crack shape on SC is investigated. Through the results on the sensitivity analysis, the adequacy of the evaluation procedure of SC is ascertained.

The Master Curve approach is a powerful tool to evaluate material-specific fracture toughness of ferritic steels, such as RPV steels, using a limited number of specimens. However, preparing a sufficient number of standard fracture toughness test specimens is difficult for irradiated RPV steels of existing surveillance programs. Utilization of miniature specimens that can be machined from broken halves of standard Charpy specimens is a possible solution to address this issue. CRIEPI has been working on the test technique utilizing a miniature C(T) (Mini-C(T)) specimens, whose dimensions are 4 × 10 × 9.6 mm (0.16 inch thickness specimen). The basic applicability of the Mini-C(T) Master Curve approach has been confirmed [1] for the base metals of typical Japanese RPV steels. International round robin tests confirmed the reproducibility of fracture toughness data obtained by Mini-C(T) specimens [2–4]. Ensuring the applicability of the Mini-C(T) Master Curve approach to weld metals and heat affected zone materials is of great importance to meet the future demand from the RPV surveillance programs for over 40 or 60 years’ reactor operation. For a weld metal deposit, we verified that valid reference temperature, T o , can be estimated using the Mini-C(T) specimens and the statistics of the fracture toughness data [5] show good conformity to the assumption of the Master Curve method [6]. In the present paper, fracture toughness of a weld joint, which consists of two different heats of RPV plate material was examined. Five sets of Mini-C(T) specimens taken from two base metals, their heat affected zones (HAZ) and weld metal deposit, were subjected to the fracture toughness test. 0.5T-C(T) specimens taken from similar locations were also subjected to the fracture toughness tests to investigate specimen size effect. All the Mini-C(T) data sets taken from base metal, HAZ and weld metal were eligible for the determination of valid T o with each 12 to 16 Mini-C(T) specimens. The relevance of the specimen size correction in the Master Curve method was confirmed for two base metals and a weld metal. The fracture toughness data for HAZ materials gave a reasonable agreement with the specific Weibull distribution assumed in the Master Curve method. Nevertheless, T o values of four data sets of HAZ materials, including two Mini-C(T) datasets and two 0.5T-C(T) datasets, showed larger variation than that of the base metals or the weld metal. The crack initiation sites of HAZ specimens were all within so-called fine grain HAZ. However the HAZ width near the crack initiation site was dependent on the individual specimens. Higher fracture toughness tended to be gained from the specimens with narrower HAZ width. The resulting T o values for HAZ material were close to or lower than that for base metals. The results suggest that the HAZ material gives equivalent or higher fracture toughness than in base metals.

In petrochemical and nuclear power plants, inner surface of components, such as pressure vessels and piping, which are made of carbon steels or low-alloy steels, is often cladded by austenitic stainless steels to improve the corrosion resistance of those components. In the evaluation of a crack postulated near the inner surface of cladded components, the following two different kinds of cracks are often assumed: One is a surface crack penetrated through cladding and the other is a subsurface crack under cladding. In order to evaluate the structural integrity of those components in a rational manner, it is important to investigate the crack propagation behavior. In particular, it should be evaluated whether a subsurface crack penetrates through cladding or is arrested at the interface between cladding and base metal. In this study, ductile crack propagation analysis for a subsurface elliptical crack in cladded plates was performed using elastic-plastic XFEM, which can model the crack independent of finite elements.

When a crack is detected in a piping line during in-service inspections, failure estimation method provided in ASME Boiler and Pressure Vessel Code Section XI or JSME Rules on Fitness-for-Service for Nuclear Power Plants can be applied to evaluate the structural integrity of the cracked pipe. The failure estimation method in the current codes accounts for the bending moment and axial force due to pressure. The torsion moment is not considered. Recently, analytical investigations have been carried out by several authors on the limit load of cracked pipes considering multi-axial loads including torsion and two failure estimation methods for multi-axial loads including torsion moment with different ranges of values have been proposed. In this study, to investigate the failure behavior of cracked pipes subjected to multi-axial loads including the torsion moment and to provide experimental support for the failure estimation methods, failure experiments were performed on 20 mm diameter stainless steel pipes with a circumferential surface crack or a through-wall crack under combined axial force and bending and torsion moments. Based on the experimental results, the proposed failure estimation methods were confirmed to be applicable to cracked pipes subjected to multi-axial loads.

Piping lines in nuclear power plants may experience multi-axial loads including tensile force, bending and torsion moments during operation. There is a lack of guidance for failure evaluation under the multi-axial loads including torsion moment. ASME B&PV Code Section XI Working Group is currently developing fully plastic failure evaluation procedures for pressurized piping items containing local wall thinning subjected to multi-axial loads. A failure estimation method for locally wall thinned pipes subjected to multi-axial loads including torsion moment has been proposed through numerical analyses. In this study, in order to investigate the failure behavior of the pipes with local wall thinning subjected to multi-axial loads including the torsion, failure experiments were performed on 20 mm diameter carbon steel pipes with a local wall thinning. Based on the experimental results, the proposed failure estimation method is confirmed to be applicable to pipes with local wall thinning.

The Master Curve approach using the miniature C(T) specimens with 4 mm-thickness is a promising method for the direct determination of the reference temperature of reactor pressure vessel steels, because they can be taken from the broken halves of the Charpy specimens used for surveillance program to monitor neutron irradiation embrittlement. The dimensional tolerances of standard C(T) specimens are relatively provided in the present standard, ASTM E1921, consequently, absolute dimensional tolerances are severer for smaller specimens. In this paper, the effect of the tolerances of key dimensions on the elastic-plastic equivalent stress intensity factor derived from the J-integral, K J , was calculated using three-dimensional finite element analysis. Even if the dimensional tolerances for the miniature C(T) specimens based on the present standard were mitigated in some degree (as examples, the tolerance of specimen thickness of ±0.08 mm was mitigated to ±0.1 mm, the tolerance of specimen width of ±0.04 mm was mitigated to ±0.1 mm), the variations of K J and reference temperature were negligibly small. The use of the mitigated dimensional tolerances with adequate accuracy of fracture toughness evaluation was ascertained.

The Master Curve approach for the fracture toughness evaluation is expected to be a powerful tool to ensure the reliability of long term used reactor pressure vessel (RPV) steels. In order to get sufficient number of data for the Master Curve approach coexistent with the present surveillance program for RPVs, the utilization of miniature specimens that can be taken from the broken halves of the surveillance Charpy specimens is important. CRIEPI has developed the test technique for the miniature C(T) specimens, whose dimensions are 4 × 10 × 9.6 mm, and has verified the basic applicability of the Master Curve approach by means of the miniature C(T) for the determination of the fracture toughness of typical Japanese RPV steel base metals [1]. A series of round robin tests on RPV steel base metals [2–4] demonstrated that the miniature C(T) specimen can be used for the determination of the reference temperature ( T o ) with no specific difficulties in test techniques. The present paper addresses the applicability of the fracture toughness evaluation by the miniature C(T) specimens on a RPV weld metal with multi-layer weld bead structure. The distribution of the fracture toughness and the trend in fracture toughness change with temperature were confirmed to show a good agreement with the assumption of the Master Curve method [5]. Fracture surface of the specimens were in cleavage fracture mode regardless of the difference in fracture toughness level. The relevance of the specimen size correction in the Master Curve method was confirmed. The difference of T o values were only in a few degrees Celsius between the data obtained with 0.5 inch-thickness C(T) specimens and the miniature C(T) specimens. The effect of local loss of constraint nearby the specimen side surface was examined by comparing with the datasets from the specimens with and without side grooves. The difference of T o was only 3 degree centigrade and no remarkable effect of side grooving could be seen. From overall examination results, it was concluded that the miniature C(T) specimen can be used for the Master Curve evaluation of tested PRV weld metal.

The use of miniature C(T) specimens makes it possible for the direct determination of the reference temperature of reactor pressure vessel steels, because they can be taken from the broken halves of the Charpy specimens used for surveillance program to monitor neutron irradiation embrittlement. Fracture toughness tests using C(T) specimens usually need the measurement of load-line displacement, however, it is difficult to mount a clip gauge inside the miniature specimen due to the limitation of the specimen size. A pair of knife edges is machined at the front face of the miniature C(T) specimen to mount a clip gauge with razor blade tips, and the front-face displacement is translated to the load-line displacement. When front-face displacement measurements are made, the load-line displacement can be inferred by multiplying the measured values by the constant 0.73. This conversion factor has been simply derived from the assumption of the linear deformation around a supposed point of rotation. In this study, the conversion factor was directly evaluated by using a three-dimensional elastic-plastic finite element analysis for the miniature C(T) specimens, and the adequacy of the conversion factor was investigated.

The rules on fitness for service for nuclear power plants of JSME are applied for flaw evaluation after detecting defects in the operating nuclear plants in Japan. The rules mainly focus on simple geometry such as straight pipes or vessels and do not provide the evaluation procedure for complex structures. The authors made a draft of flaw acceptance rule for J-groove weld of a bottom mounted instrumentation nozzle at application of the cap repair. The rule contains flaw modeling, fatigue and SCC crack growth calculation and flaw instability assessment. After detecting a defect on a J-groove weld, a flaw will be modeled in the whole of J-groove weld region because of fast SCC crack propagation in the weld region. Due to complex configuration of the evaluation location, FE analysis is needed for obtaining stress intensity factors ( K s) to calculate the crack growth and flaw instability. The proposed rule has a guidance for K calculation by FE analysis with the aim of decreasing dependence of individuals for calculation. The authors performed benchmark analyses to confirm the guidance applicability. The calculation results by three participants agreed within several percent.

The Master Curve approach for the fracture toughness evaluation is expected to be a powerful tool to ensure the reliability of long-term used RPV steels. In order to get sufficient number of data for the Master Curve approach coexistent with the present surveillance program for RPVs, the utilization of miniature specimens, which can be taken from broken halves of surveillance Charpy specimens, is important. CRIEPI developed the test technique for the miniature C(T) specimens (Mini-CT), whose dimensions are 4 × 10 × 10 mm, and verified the basic applicability of Master Curve approach by means of Mini-CT for the determination of fracture toughness of typical Japanese RPV steels. A round robin program is organized in order to assure the robustness of the testing procedure to the difference in testing machines or operators. The first and second round robin tests (PVP2012-78661 [1], PVP2013-97936 [2]) suggested that the reference temperature T 0 evaluation technique by Mini-CT specimen potentially is fairly robust in regard to difference in testing machines and operators, and gives similar loading rate dependency to the larger C(T) specimens. As the final year of the round robin program, “blind tests” were carried out. Here, detailed material information such as the type of materials, estimated T 0 , existing fracture toughness data for the material, were not given with the specimen, and 6 organizations independently selected the test temperature based on Charpy full curve of the tested material. The selection of test temperature has the variation of −120 °C to −150 °C among the organizations. 8 to 20 specimens in a set were subjected to the Master Curve evaluation and all the 6 organizations successfully obtained valid T 0 . The scatter range in T 0 was at most 16 °C, which was within the acceptable scatter range specified in ASTM E1921-10e1. The selection of test temperature seems to give limited effect as like as that in larger specimens.