Masao Itatani

Fatigue crack growth behavior of weld heat-affected zone of type 304 stainless steel in high temperature water

Abstract The fatigue crack growth behavior of the weld heat-affected zone (HAZ) of type 304 stainless steel in high temperature water which simulates the boiling-water reactor environment was investigated to clarify the effects of welding residual stress, cyclic frequency f and thermal aging on crack growth rate. A lower crack growth rate of the HAZ than of the base metal was observed in both the high temperature water and the ambient air caused by the compressive residual stress. The crack closure point was measured in the high temperature water. The effect of the welding residual stress on the crack growth rate of the HAZ can be evaluated separately from the environmental effect through the crack closure behavior. The high temperature water increased the crack growth rate at a cyclic frequency of 0.0167 Hz but did not affect it much at 3 and 5 Hz. The crack growth behavior of the thermally aged HAZ at 400 °C for 1800 h was almost the same as that of the unaged material tested at 0.0167 and 5 Hz in the high temperature water.

Evaluation of Fracture Characteristics of Ni-Base Weld Metal for BWR Components

Fracture behavior of Ni-base weld metals used for boiling water reactor (BWR) was investigated. The elastic-plastic fracture toughness (JIc ) tests were conducted for Alloys 182 and 82 in room temperature (R.T.) and 300°C air using 2TCT specimen. It was found that the ductile crack growth resistance of Alloy 82 is higher than that of Alloy 182, and also the ductile crack growth resistances at 300°C are higher than those at R.T. for both Alloys 182 and 82. The fracture loads of CT specimen were predicted using existing limit load equations. The estimated limit load Pc using flow stress which is defined as the average of 0.2% proof stress and tensile strength coincided well with the experimental maximum load Pmax . It was confirmed that the conservative limit load estimation is possible when using 2.7Sm as the effective yield stress in the limit load equation.Copyright

Revision of Flaw Evaluation Methods of Pipes Having a Circumferential Flaw in JSME Fitness-for-Service Codes

This paper shows the technical basis of revision to the JSME Fitness-for-Service Code (the FFS code) for flaw evaluation methods of pipes which have a very shallow circumferential flaw. When a flaw in a pipe is very small, the allowable stress between the FFS code and the JSME Design and Construction Code (the design code) is mismatched. Fracture strength of a pipe with small flaw depends on the tensile strength accompanied by large deformation. Therefore, fracture mechanics is not applicable in such a case. This mismatch has been resolved for an axial crack assessment by improving definition of flow stress for shallow crack. In this study, the authors investigated this mismatch in the allowable stress in the flaw assessment for a pipe with a circumferential crack. Some past fracture test results of pipes showed that flawed pipes did not fracture at the flaw section when the circumferential flaw size was small and they failed by oval deformation or plastic buckling. Allowable stress for such behavior has been incorporated in some existing design codes as a restriction for plastic collapse. Through the reevaluation of the existing piping fracture test results, the applicability of fracture evaluation methods defined in the FFS code was examined for the case that flaw size was very small. As a result, the fracture evaluation method based on flow stress was found not to be applicable when flaw size was very small, and the failure criterion in this case depended on the collapse strength accompanying with ovalization. Revisions of the FFS code reflecting these examination results were proposed in this paper.Copyright

Re-Evaluation of Fatigue Crack Growth Curve for Austenitic Stainless Steels in BWR Environment

The Rules on Fitness-for-Service for Nuclear Power Plants of the Japan Society of Mechanical Engineers (JSME Code) has the reference fatigue crack growth curve for austenitic stainless steels in BWR environment. This reference curve was determined as the upper bound of crack growth data excluding the outlier data. However, the other reference curves for fatigue crack growth rate such as austenitic stainless steels and ferritic steels in air environment and ferritic steels in water environment in the ASME Boiler and Pressure Vessel Code, Section XI and the JSME Code, austenitic stainless steels in PWR environment in the JSME Code and Ni-base alloys in PWR environment in the JSME Code Case are determined based on the 95% upper confidential limit by statistic data treatment. In the present study, the fatigue crack growth data of austenitic stainless steels in BWR environment were re-evaluated statistically. It was found that the current reference curve almost coincides with 95% upper confidential limit of fatigue crack growth data in the Paris region. Consequently, the current reference fatigue crack growth curve for austenitic stainless steels in BWR environment in the JSME Code can be regarded to stand on the same technical bases with other reference fatigue crack growth curves. Furthermore, the authors proposed to extend applicable upper bound of load rising time tr from 1000 s to 32000 s.Copyright

Development of Load Multipliers (Z-Factors) in Elastic-Plastic Fracture Mechanics Evaluation in Rules on Fitness-for-Service for Nuclear Power Plants

Not only nonmandatory Appendix C, “Evaluation of Flaws in Piping,” in ASME Boiler & Pressure Vessel Code Section XI but also Appendix E-9, “Elastic-Plastic Fracture Mechanics Evaluation,” in the JSME Rules on Fitness-for-Service for Nuclear Power Plants use the load multiplier Z-factor that is applied to elastic-plastic fracture mechanics evaluation for a circumferential flaw of austenitic stainless steel piping and ferritic steel piping. The Z-factor is defined as the ratio of the limit load to the load at fracture load. Basically, the Z-factor equations were conservatively formulated by using the Z-factors for circumferential through-wall flaws. However, the Codes require flaw evaluation for circumferential surface flaws. Accordingly, Z-factors for circumferential surface flaws should be developed to have the consistency. Therefore Z-factor equations of austenitic stainless steel piping and ferritic steel piping have been developed for circumferential surface flaws.© 2012 ASME

Fracture Assessment of Pipes Having Multiple Flaws Based on Ramberg–Osgood-Type Stress–Strain Relationships

This paper describes a fracture assessment method for a pipe having multiple circumferential flaws. According to Fitness-for-Service (FFS) codes for nuclear facilities published by the Japanese Society of Mechanical Engineers (JSME), the fracture strength of a high-ductility pipe having a circumferential flaw is evaluated using the limit load assessment method assuming the elastic–perfectly-plastic stress–strain relationship. In this assessment, flow stress is used as a proportional stress. However, previous experimental results [1, 2, 3] show that a crack penetrates before the entire flawed pipe section reaches the flow stress. Therefore, stress concentration at a flaw was evaluated on the basis of the Dugdale model [4], and the fracture strength of the crack-ligament was evaluated. This model can predict test results with high accuracy when the ligament fracture strength is assumed to be tensile strength. Based on this examination, a fracture assessment method for pipes having multiple flaws was developed considering the stress concentration in the crack-ligament by using the realistic stress–strain relationship (Ramberg–Osgood-type stress–strain curve). The fracture strength of a multiple-flawed pipe estimated by the developed method was compared with previous experimental results. When the stress concentration in the crack-ligament was taken into consideration, the fracture strength estimated using the Ramberg–Osgood-type stress–strain curve was in good agreement with experimental results, confirming the validity of the proposed method.© 2011 ASME

Fracture Assessment of Austenitic Stainless Steel Piping With Multiple Flaws in Heat-Affected Zone

Stress corrosion cracking (SCC) has been observed as circumferential multiple flaws in the weld heat-affected zone of primary loop recirculation system piping and core shrouds made of low carbon stainless steel. In the Japan Society of Mechanical Engineers code, Rules on Fitness-for-Service for Nuclear Power Plants, there is no fracture assessment of piping with multiple flaws which are not subject to flaw combination rule criteria. Through fracture testing of piping with two circumferential flaws in the weld heat-affected zone, the limit load estimation method was used for fracture assessment of stainless steel piping.Copyright

Fracture Assessment of Austenitic Stainless Steel Piping With Twin Flaws in TIG Weld

Fracture behavior of austenitic stainless steel piping for boiling water reactor (BWR) internals with circumferential through wall twin flaws at the weld was investigated. A 150A Sch.40 piping of type 316L stainless steel which has an outer diameter of 165.2 mm and a thickness of 7.1 mm was butt welded by tungsten inert gas (TIG) weld and single or twin through wall slits were introduced by an electro discharge machining (EDM) on the weld bead. Four point bending tests were conducted and failure stress was evaluated by currently proposed limit load equation for a piping with multiple flaws. The fracture loads obtained by the test were higher than the limit load based on the recently proposed equation for pipe with multiple flaws using 2.7Sm . It was concluded that the limit load criterion is able to be applied to the fracture assessment of austenitic stainless steel piping with twin flaws in the TIG weld. Through the pipe fracture test, it was found that the crack tends to grow in base metal rather than weld metal.© 2009 ASME

Proposal of Common Rules for Fatigue Crack Growth Law of the JSME Rules on Fitness-for-Service

The JSME Codes and Standards Committee have provided many kinds of crack growth rates of SCC and fatigue cracking as Appendix in the code and the code cases. However each code has been separately investigated and there is no common rule among SCC growth rates or fatigue crack growth rates. The working group (WG) members discussed to construct the common rules for fatigue crack growth. Sensitivity study for loading rise time in PWR environment was investigated with slower rise time. Using the proposed common rule, the subgroup (SG) on Evaluation and Standard will transfer the fatigue crack growth rates in the code cases to the code.Copyright

Evaluation of Crack Growth of Ni-Base Alloys Under Long Term Cyclic Loading in BWR Environment

Crack growth test data of Ni-base alloys under cyclic loading in simulated boiling water reactor (BWR) environment including the effects of load rising time (tr) were evaluated in the view points of both fatigue and stress corrosion cracking (SCC). When the test data were plotted in the relationship between da/dt and Kmax, da/dt monotonically decreased with increasing tr and the stress ratio (R). For alloy 182 weld metal under short tr and/or low R, the crack growth rate assuming SCC is much lower than those of the test data. For alloy 182 under tr = 30 and 1000 s at R = 0.8, the crack growth rate assuming SCC almost coincided with test data. For heat affected zone (HAZ) of alloy 600 base metal (600HAZ), the crack growth rate assuming SCC had much different slope of da/dN-ΔK relationship compared with the test data in the tested range of tr up to 3000 s. From these observations, the contribution of SCC is relatively small and the main mechanism of crack growth is thought to be fatigue for the tested range (tr=1 to 1000 s for weld metal, tr=1 to 3000 s for base metal and R = 0.1 to 0.8). It was assured that the fatigue crack growth formula proposed by the authors accounts the effect of SCC adequately at long tr. Additionally, the applicability of the fatigue crack growth rate formula for austenitic stainless steels to the long term cyclic load was investigated and it was found that the formula can be applied to tr=30000 s.Copyright