Fracture toughness characterization of irradiated, low upper shelf welds

A data base of J − R curve trends is being established for irradiated, light-water reactor pressure vessel steels of low upper shelf toughness. R-curve trends have been developed for several welds made with Linde 80 flux and containing a high copper impurity which enhances the sensitivity to irradiation embrittlement. Tests were conducted using compact toughness specimens of 12.5 mm to 100 mm thickness. These data provide the materials properties necessary for a tearing instability analysis of postulated accident conditions in certain commercial power reactors. Irradiation decreased the level of the R curve significantly in most cases. Furthermore, the average value of tearing modulus (T_avg) was a more discriminating indicator of toughness degradation than the crack initiation toughness (J_Ic). Temperature-dependent correlations between the R-curve parameters (J_Ic, T_avg) and Charpy-V (C_v) upper shelf energy have been suggested here and in other programs. This finding could enhance the significance of C_v reactor surveillance data with respect to structural integrity. However, J_Ic and T_avg have demonstrated an inverse relationship with temperature which is not reflected in C_v upper shelf energy and this must be taken into account in the development of correlations.     

Use of J–R curves in assessing the fracture behavior of low upper shelf toughness materials

The objective of this investigation was to evaluate the use of small specimen J–R curves in assessing the fracture resistance behavior of reactor vessels containing low upper shelf (LUS) toughness weldments. As required by the U.S. Code of Federal Regulations (10 CFR, Part 50), reactor vessel beltline materials must maintain an upper shelf Charpy V-Notch (CVN) energy of at least 50 ft-lbs (68 J) throughout vessel life. If CVN values from surveillance specimens fall below this value, the utility must demonstrate to the U.S. Nuclear Regulatory Commission (NRC) that the lower values will provide “margins of safety against fracture equivalent to those required by Appendix G of the ASME Boiler and Pressure Vessel Code”. This paper will present recommendations regarding the material fracture resistance aspects of this problem and outline an analysis procedure for demonstrating adequate fracture safety based on CVN values.It is recommended that the deformation formulation of the J-integral be used in the analysis described above. For cases where J-integral fracture toughness testing will be required, the ASTM E1152-87 procedure should be followed, however, data should be taken to 50% to 60% of the specimen remaining ligament. Extension of the crack growth validity limits for J–R curve testing, as described in E1152-87, can be justified on the basis of a “J-controlled crack growth zone” analysis which shows an engineering basis for J-control to 25% to 40% of the specimen remaining ligament. If J-R curve extrapolations are required for the analysis, a simple power law fit to data in the extended validity region should be used. The example analysis performed for low upper shelf weld material, showed required CVN values for a reactor vessel with a 7.8 inch (198 mm) thick wall ranging from 32 ft-lbs (43 J) to 48 ft-lbs (65 J), depending on the magnitude of the thermal stress component.     

Small specimen predictions of fracture toughness for nuclear pressure vessel steels

The use of fracture mechanics in the fracture-safe design and continued safe operation of nuclear reactor pressure vessels has provided an incentive for the development of small specimens for obtaining pertinent fracture toughness data. Small specimens are required for economic reasons when a large number of heats are involved and for space limitation reasons such as in surveillance programs. Several approaches to obtaining fracture toughness from small specimens by either direct measurements or indirect correlations and calculations are reviewed, and their merits and limitations are discussed. Emphasis is placed on techniques which have been developed to determine static and dynamic fracture toughness from surveillance-type specimens. Recently developed techniques for obtaining J-initiation values from a single test specimen and methods for estimating lower and upper shelf fracture toughness from tensile properties are also presented.     

Results and conclusions from the first pressurized-thermal-shock experiment

The first pressurized-thermal-shock test of a 148 mm thick steel pressure vessel with a 1 m long flaw was performed to investigate fracture behavior of a vessel under conditions relevant to a flawed nuclear reactor pressure vessel during an overcooling accident. The objectives were to observe crack arrest and stability on the ductile upper shelf and the effects of warm prestressing on crack initiation. Three coordinated pressure and thermal transients were imposed on the vessel, which was preheated to 290°C. Two episodes of crack propagation and arrest occurred. The thermal transients were induced by coolant at −29 to 15°C. Pressure transients were as high as 94.4 MPa. The experimental objectives were attained. The inhibiting effects of warm prestressing were definitely demonstrated. Crack propagation was nearly pure cleavage, and arrest at 30 K above the onset of the Charpy upper-shelf was experienced in a positive K₁ gradient and with K₁ = 300 MPa√m. Fracture-mechanics analysis of brittle fracture based on small-specimen toughness measurements was reasonably accurate. Flaw evaluation by procedures of the ASME Boiler and Pressure Vessel Code conservatively predicted vessel failure, which did not occur. No ductile tearing occurred after each crack arrest, although some stable tearing had been predicted on the basis of tearing resistance data.     

Pressure vessel steel fracture toughness in the regime from room temperature to 400°C

The fracture toughness of steels that are susceptible to dynamic strain aging shows a minimum at temperatures higher than the upper shelf starting temperature. This phenomenon is caused simultaneously by strain aging and plastic deformation. The first aim of the present work is to analyze the effect of dynamic strain aging on the fracture toughness values of three pressure vessel steels in the temperature range between room temperature and 400°C. Fracture mechanics tests were carried out on A533 GB, A516 G70 and 304L steels to obtain the following parameters: J_IC, CTOD_m and the J-R curves. These values were compared against those available in the present references, and good agreement was found. Charpy V notch tests were also carried out on A516 G70 steel at the same test temperatures as for the fracture mechanics tests to analyze the effect of the strain rate. The critical wide stretch zones of the 304L steel specimens were also measured to verify another author's hypothesis about a toughness drop at the upper shelf temperature.     

Ontario hydro's leak before break approach: Application to the darlington (candu) nuclear generating station a

Ontario Hydro has developed a leak before break (LBB) approach for application to the large diameter heat transport piping for Darlington NGS A as an alternative to the provision of pipewhip restraints. This approach has been applied to pipe sizes which are equal to or greater than 530 mm (21 in. NPS). The proposed LBB approach incorporates assessments at several levels to provide assurance against catastrophic rupture. A comprehensive and systematic review of pipe failure mechanisms is considered the first important step in establishing role and applicability of the LBB concept. The elements integral to the approach are those related to demonstration of crack stability utilizing fracture mechanics methods and those related to leak rate predictions and leakage detection capability. For evaluation of crack stability the J-integral/tearing modulus (J/T) method has been selected. Results from an extensive material test program from actual heat transport piping, forgings, associated welds and heat affected zones as inpur to EPFM analyses provide the J-resistance and J—T curves. The details of EPFM analyses for a straight pipe with a circumferential crack and a piping elbow with a central longitudinal throughwall crack are presented here. Additionally, results of crack opening detail, the effects of crack face pressure, the predictions of LEAK RATE code and an assessment of the leakage detection capability are presented.     

Elastic-plastic fracture toughness behavior of heavy section steels for nuclear pressure vessels

Fracture toughness tests were performed in the transition region for ASTM A508 Class 3 steel using about 160 specimens. The K_J-values which are converted from J_c of the smaller specimens indicated a wide scatter ranging from below the K_Ic-value to much higher toughness. The fast brittle fracture behavior in the transition regime can be divided into two regions: (1) the region where fracture occurs on a blunting line (Region I) and (2) the region where fracture occurs on an R-curve (Region II). The scatter of the K_J-values in each region is caused by the amount of crack extension contained in the specimens. The methods to obtain the fracture toughness equivalent to the K_Ic from the K_J values were also presented.In the upper shelf region, the ductile fracture behavior of A508 Class 3 base metal and weldments was investigated. The 25% side grooved specimen was recommended for measuring the resistance against ductile crack growth. The weld heat affected zone (HAZ) has comparatively higher tearing modulus, whereas the weld metal shows the lowest one.     

Investigation of the ductile fracture properties of type 304 stainless steel plate, welds, and 4-inch pipe

J-integral fracture toughness tests were performed on welded 304 stainless steel 2-inch plate and 4-inch diameter pipe. The 2-inch plate was welded using a hot-wire automatic gas tungsten arc process. This weldment was machined into 1T and 2T compact specimens for single specimen unloading compliance J-integral tests. The specimens were cut to measure the fracure toughness of the base metal, weld metal and the heat affected zone (HAZ). The tests were performed at 550°F, 300°F and room temperature. The results of the J-integral tests indicate that the J_Ic of the base plate ranged from 4400 to 6100 in lbs/in² at 550°F. The J_Ic values for the tests performed at 300°F and room temperature were beyond the measurement capacity of the specimens and appear to indicate that J_Ic was greater than 8000 in lb/in². The J-integral tests performed on the weld metal specimens indicate that the J_Ic values ranged from 930 to 2150 in lbs/in² at 550°F. The J_Ic values of the weld metal specimens tested at 300°F and room temperature were 2300 and 3000 in lbs/in² respectively. One HAZ specimen was tested at 550°F and found to have a J_Ic value of 2980 in lbs/in² which indicates that the HAZ is an average of the base metal and weld metal thoughness. These test results indicate that there is a significant reduction in the initiation fracture toughness as a result of welding.The second phase of this task dealt with the fracture toughness testing of 4-inch diameter 304 stainless steel pipes containing a gas tungsten arc weld. The pipes were tested at 550°F in four point bending. Three tests were performed, two with a through wall flaw growing circumferentially and the third pipe had a part through radial flaw in combination with the circumferential flaw. These tests were performed using unloading compliance and d.c. potential drop crack length estimate methods. The results of these test indicate that the presence of a complex crack (radial and circumferential) reduces in the initiation toughness and the tearing modulus of the pipe material compared to a pipe with only a circumferentially growing crack.     

Ductile fracture parameters of RPV-materials

Ductile fracture material parameters have been determined for a reactor pressure vessel material to characterize its fracture resistance in the upper shelf toughness regime. Three different methods (the multiple specimen unloading (MSU), direct current potential drop (DCPD) and single specimen partial unloading compliance (SSPUC) methods) have been applied to test different CT-specimen geometries at temperatures between 25 and 300°C.It is shown that there are principle differences between J-R-curves measured by different experimental procedures, because of different methods for the measurement of crack lengths and crack growth. For instability analyses, using a complete J-R-curve, these differences seem to be negligible. For the determination of critical material parameters at or close to initiation of stable crack growth these differences may cause systematic errors tending to higher values for DCPD as compared to MSU-results and to lower values for SSPUC respectively.Procedures can be defined to evaluate comparable critical material parameters from the different experimental procedures, if J_i is known in a good approximation allowing to consider only the real crack extension without blunting, or if in addition the real (or realistically modelled) blunting and the effective blunting of the specific method are known. The differences in material parameters will depend quantitatively on the type of material and its toughness (slope of J-R-curve). They may be in the range of the experimental scatter observed in testing and seem to be negligible, but their systematic character should be kept in mind, e.g. when ranking different materials according to their critical parameters determined by different methods.     

Evaluation of throughwall crack pipes under displacement controlled loading

Tearing modulus solutions are developed for flawed throughwall pipes subjected to displacement controlled loading. Two cases of loading were considered: (1) a displacement controlled bending loading, and (2) a displacement controlled axial tension loading. A revised version of the EPRI J-integral estimation scheme is used in the development of the solutions. These solutions can be used for the entire range of elastic-plastic loading, from linear elastic, contained yielding, to large scale yielding of the crack section. Experimental data from pipes in bending were used to assess the accuracy of the compliant loading solutions. The evaluations were performed using elastic plastic J-integral (J) and tearing modulus (T) analysis methods. These solutions are shown to have good accuracy when used to predict the experimental results. The methodology and procedure can also be applied to part-throughwall cracks. These solutions have application to the leak before break fracture mechanics analyses.     

Irradiation experiments in the testing nuclear power plant VAK

Within the German Research Programme “Integrity of Components” the first two capsules were irradiated in the Testing Nuclear Power Reactor VAK. The materials are of the 22 NiMoCr 3 7 and 20 MnMoNi 5 5 types and represent the lower bound of the base material regarding upper shelf energy and chemical composition (Cu, S, P), as well as a state of material which does not meet both chemical and toughness requirements (low upper shelf test melt). Tensile, Charpy, drop-weight, and fracture mechanics specimens were irradiated up to a range of 1.5 to 2 × 10¹⁹ cm⁻² (E > 1 MeV). Despite the materials being at or beyond the specification limits, the results show irradiation sensitivity which can be predicted from the US Reg. Guide Trend Curves (1.99) and KWU Trend Curves in a conservative manner. The procedure to determine the adjusted reference temperature RT_NDT (adj.) on the basis of ΔT_41J (following ASTM E 185) could also be confirmed as conservative by comparing the different criteria derived from Charpy and drop weight tests in the unirradiated and irradiated condition.The results of fracture mechanics testing in the linear elastic range show a remarkable temperature margin to the K_Ic-curve of ASME XI.Prestrained compact tension specimens CT 40 mm made of 22 NiMoCr 3 7 material with an upper shelf energy of approx. 100 J were wedge loaded in a range up to 30 MPa m and exposed to the water environment during radiation. Macroscopic examination gave no indications of stress corrosion cracking.From tests of these specimens in the linear elastic range, a fracture toughness K_Ic^*, which was not affected by the prestrain and environment history, was found depending only on the overload applied during the prestraining procedure.     

The application of constraint based fracture mechanics to engineering structures

Elastic-plastic crack tip fields can be characterised by two parameters, J and T/Q which describes crack tip constraint. This forms the foundation of a constraint based fracture mechanics in which toughness is expressed as a function of constraint in the form of a J–(T/Q) locus. The enhanced toughness associated with shallow cracks and defects can be used in a systematic manner by constraint matching, and implemented through a simple modification to a failure assessment diagram. This methodology enables the conservatism associated with the use of deeply cracked fracture toughness measurements to be relaxed. In this work these methods have been applied to a tubular welded joint.     

The application of elastic-plastic fracture mechanics parameters in fracture safe design

The report summarizes some of the methods which are currently used for assessing the fracture toughness of materials under elastic and elastic-plastic conditions. The main parameters which are considered are (1) plane strain fracture toughness (K_Ic), (2) equivalent energy (K_Icd), (3) contour integral (J) and (4) crack opening displacement (COD). Gross strain crack tolerance and stress concentration methods are also discussed.It is concluded that of these parameters, the contour integral and the crack opening displacement have most potential for future development. These two parameters are shown to be equivalent, however, at the present stage of development the COD concept has several advantages over the J concept. Firstly, the COD concept is able to take into account, secondary stresses, such as welding residual stresses. Because these stresses are in equilibrium, they do not appear in energy measurements to evaluate J. Secondly, the COD value is a physical measure of the crack tip conditions which includes the effect of stress state and thickness. It is, therefore, possible to measure and calculate COD levels for cracks in real structures. It is not possible to evaluate J for real structures since J methods are appropriate only to in-plane problems. This also means that partial wall (thumbnail) flaws are better characterized by the COD concept.The COD concept has been developed to a stage where it is possible to estimate the significance of flaws in welded structures provided the toughness of the material and the acting stresses or strains are known. This development is described and the method used to analyze tests on model pressure vessels with 6″ thick walls. A comparison is made with other methods, and it is concluded that although the COD analysis gives conservative estimates of the flaw size to cause failure, further work is necessary to be able to predict vessel burst conditions when failure is preceded by extensive plasticity and stable ductile tearing. A simple nomogram to determine COD levels to ensure leak before break conditions is also developed.     

Effect of dynamic strain aging on the leak-before-break analysis in SA106 Gr.C piping steel

The characteristics of dynamic strain aging (DSA) on material properties used in leak-before-break (LBB) analysis were discussed. Using these material data, the effect of DSA on the LBB analysis was estimated through the evaluation of leakage-size crack and flaw stability in SA106 Gr.C piping steel. Also, the results were represented as a form of ‘LBB allowable load window'. In the DSA temperature region, the leakage-size crack length was smaller than that at other temperatures and it increased with increasing tensile strain rate. In the results of flaw stability analysis, the lowest instability load appeared at the temperature corresponding to the minimum J–R curve which was caused by DSA. The instability load depended on the loading rate of J–R data, and decreased with increasing tensile strain rate at the plant operating temperature. These are due to the strain hardening characteristic and strain rate sensitivity of DSA. In the ‘LBB allowable load window', the LBB allowable region at the temperature and loading conditions where DSA occurs was decreased by about 30% compared with that in other conditions.     

Evaluation of dynamic crack initiation and growth toughness by computer aided charpy impact testing system

The validity of the newly developed CAI system ( omputer ided nstrumented Charpy Impact Testing ) for the evaluation of the dynamic elastic-fracture toughness parameters, i.e. the dynamic crack initiation toughness (K_Id, J_Id) and growth toughness (J_R curve, tearing modulus: T_mat) was investigated in various materials such as steel, aluminum alloy, titanium alloy, cast iron and so on.The dynamic elastic-plastic fracture toughness parameters (J_{Id and} T_mat) values in various materials could be easily obtained from the single load-deflection curve of the precracked specimen using the CAI system. The introduction of the side grooves was found to be effective to obtain the valid J_Id value using the standard size Charpy specimen. In addition, the strain rate dependency of the specimen compliance (C₃), and the effect of the specimen size on J_d value and impact velocity on the geometry of the load-deflection curve were studied.     

Analysis of stable crack growth of a semi-elliptical surface crack by numerical simulation

In the present paper, the canoeing effect during stable crack growth of a semi-elliptical surface crack in a side-grooved panel under tension was investigated by means of a three-dimensional elastic-plastic finite element analysis. The numerical crack growth simulation was performed by using crack mouth opening displacement resistance curves obtained from experiment on panels by the multi-specimen procedure. The influence of the crack tip constraint and the stress triaxiality on the ductile crack resistance property was studied. It is shown that the tearing modulus T_JR increases proportionally with decreasing stress triaxiality implying constraint dependent J_R-curves.     

Reactor pressure vessel structural integrity research

Development continues on the technology used to assess the safety of irradiation embrittled nuclear reactor pressure vessels (RPVs) containing flaws. Fracture mechanics tests on RPV steel, coupled with detailed elastic-plastic finite element analyses of the crack-tip stress fields, have shown that (1) constraint relaxation at the crack-tip of shallow surface flaws results in increased data scatter but no increase in the lower-bound fracture toughness, (2) the nil-ductility temperature (NDT) performs better than the reference temperature for nil-ductility transition (RT_NDT) as a normalizing parameter for shallow flaw fracture toughness data, (3) biaxial loading can reduce the shallow flaw fracture toughness, (4) stress based dual-parameter fracture toughness correlations cannot predict the effect of biaxial loading on shallow flaw fracture toughness because in-plane stresses at the crack-tip are not influenced by biaxial loading, and (5) an implicit strain based dual-parameter fracture toughness correlation can predict the effect of biaxial loading on shallow flaw fracture toughness. Experimental irradiation investigations have shown that (1) the irradiation induced shift in Charpy V-notch vs. temperature behavior may not be adequate to conservatively assess fracture toughness shifts due to embrittlement, and (2) the wide global variations of initial chemistry and fracture properties of a nominally uniform material within a pressure vessel may confound accurate integrity assessments that require baseline properties.     

Pipe fracture safety analysis using limit load and J-integral techniques

J-integral fracture toughness tests were performed on full scale pipe specimens to assess the fracture safety performance of two reactor piping alloys. The two alloys investigated were ASTM A106 Grade B carbon steel and circumferentially welded Type 304 stainless steel.The full scale pipe fracture tests were performed on 1.2 m long, circumferentially cracked pipes loaded in four-point bending on a variably compliant test bed. Results of the experiments were analyzed using the limit load approach currently being considered for inclusion in Section XI of the ASME Code. The results were also evaluated using two tearing instability approaches. One approach assumed elastic-perfectly plastic material behavior and the other accounted for material hardening by requiring actual load and displacement data.The limit load analysis provided a good prediction of the maximum load carrying capacity of the pipe specimens in most cases. The results were especially good for the ASTM A106 steel pipes when the materials property data was used to calculate the flow stress. The J-integral tearing instability analysis was shown to accurately describe the ductile tearing instability behavior of the ASTM A106 steel pipe providing material hardening was taken into account.     

First spinning cylinder test analysis using a local approach to fracture

In recent years, several experimental programmes on large-scale specimens have been organized to evaluate the capabilities of the fracture mechanics concepts employed in structural integrity assessment of pressurized water reactor pressure vessels. During the first spinning cylinder test, a geometry effect was revealed experimentally showing the difficulties of transferring toughness data from small-scale to large-scale specimens. An original analysis of this test, by means of a local approach to fracture, is presented in this paper. Both compact tension specimen and spinning cylinder fracture behaviour were computed using a continuum damage mechanics model developed at EDF. We confirmed by numerical analysis that the cylinder's resistance to ductile tearing was considerably larger than in small-scale fracture mechanics specimen tests, about 50%. The final crack growth predicted by the model was close to the experimental value. Discrepancies in J R curves seemed to be due to an effect of stress triaxiality and plastic zone evolution. The geometry effect inducing differences in resistance to ductile tearing of the material involved in the specimens can be investigated and explained using a local approach to fracture methodology.     

Reactor vessel cladding separate effects studies

The existence of a layer of tough weld overlay cladding on the interior of a light-water reactor pressure vessel could mitigate damage caused during certain overcooling transients. The potential benefit of the cladding is that it could keep a short surface flaw, which would otherwise become long, from growing either by impeding crack initiation or by arresting a running crack. Two aspects critical to cladding behavior will be reported: irradiation effects on cladding toughness and the response of mechanically loaded, flawed structures in the presence of cladding.A two-phase irradiation experiment is being conducted. In the first phase, Charpy impact and tensile specimens from a single wire, submerged-arc stainless steel weld overlay were irradiated to 2 × 10²³ neutrons/m² (>1 MeV) at 288°C. Typical, good quality pressure vessel cladding exhibited very little irradiation-induced degradation. However, ductile-to-brittle transition behavior, caused by temperature-dependent failure of the residual δ-ferrite, was observed. In contrast, specimens from a highly diluted, poor quality weldment were markedly embrittled. In the second phase of irradiations, now in progress, a commercially produced three-wire series arc weldment will be evaluated under identical irradiation and testing conditions as the first series. In addition, 0.5T compact specimens of both weldments and higher fluences will be examined.A two-phase program is also being conducted utilizing relatively large bend specimens that have been clad and flawed on the tension surface. The testing rationale is that if a surface flaw is pinned by the cladding and cannot grow longer, it will also not grow beyond a certain depth, thereby arresting the entire flaw in a stress field in which it would otherwise propagate through the specimen. The results of phase one showed that single wire cladding with low-to-moderate toughness appeared to have a limited ability to mitigate crack propagation. For the second phase, three-wire cladding has been deposited on a base plate with a very high ductile-to-brittle transition temperature allowing testing to ascertain the crack inhibiting capability of tough upper shelf cladding.