Coke drums undergo cyclic operations typically in temperatures ranging from the ambient temperature to about 500°C. Generally, coke drums are fabricated from clad plates consisting of chromium-molybdenum low alloy steel as the base metal and ferritic stainless steel typically Type 410S stainless steel (TP.410S) as cladding material. TP.410S cladding material is used for protection against sulfide corrosion. Coke drums frequently display major cracking on the inside surface of their circumferential weld along the fusion line between restoration weld and the TP.410S cladding. Moreover, an array of cracks commonly referred to as “elephant skin” are typically observed proximity to the major cracks. These cracks have been reported by multiple papers, particularly in the API Technical Report 934-G, up to the present and primarily attributed to the thermal fatigue which occurs from the repeated thermal loads experienced during each operating cycle. When the existence of severe cracks is recognized, users are forced to schedule extensive repairs. The authors conducted fatigue tests employing coupons made from two kinds of cladding material, TP.410S and Alloy 625, which simulate the circumferential weld seam of coke drums. The paper summarizes the essential properties for the cladding material in order to delay cracking on the inside surface and proposes new common cladding materials which could be employed for coke drums based on test results.

In order to develop new design fatigue curves for austenitic carbon steels & low alloy steels and stainless steels and a new design fatigue evaluation method that are rational and have clear design basis, Design Fatigue Curve (DFC) Phase 1 subcommittee and Phase 2 subcommittee were established in the Atomic Energy Research Committee in the Japan Welding Engineering Society. The study on design fatigue curves was actively performed in the subcommittees. In the subcommittees, domestic and foreign fatigue data of small test specimens in air were collected and a comprehensive fatigue database was constructed. Using this fatigue database, the accurate best-fit curves of austenitic carbon steels & low alloy steels and stainless steels were developed by applying tensile strength to a parameter of the curve. Regarding design factors on design fatigue curves, data scatter, mean stress correction, surface finishing effect, size effect and variable loading effect were investigated and each design factor was decided to be individually considered on the design fatigue curves. A Japanese utility project performed large scale fatigue tests using austenitic stainless steel piping and carbon and low-alloy steel flat plates as well as fatigue tests using small specimens to obtain not only basic data but also fatigue data of mean stress effect and surface finishing effect. Those test results were provided to the subcommittee and utilized the above studies. In this paper, the large scale fatigue tests using austenitic stainless steel piping and the best-fit curve of austenitic stainless steel are discussed.

In Japan, the Design Fatigue Curve (DFC) Phase 1 and Phase 2 subcommittees were organized under the Atomic Energy Research Committee in the Japan Welding Engineering Society and have proposed new design fatigue curves for carbon, low-alloy, and austenitic stainless steels. To confirm the validity of the proposed design fatigue curves, a Japanese utility collaborative project was launched. In this project, fatigue tests were conducted on large-scale and small-sized specimens, and the test data were provided to the DFC Phase 2 subcommittee. This paper discusses the best-fit curves proposed by the DFC Phase 1 subcommittee, focusing on the results of large-scale fatigue tests for carbon steel and low-alloy steel plates. The fatigue test results for large-scale specimens were compared with the best-fit curve proposed by the DFC Phase 1 subcommittee. This comparison revealed that the fatigue lives given by the proposed curves correspond to those of approximately 1.5–4.0-mm-deep crack initiation in large-scale specimens. In this program, fatigue tests with a mean strain were also carried out on large-scale specimens. These tests found that the fatigue lives were almost equivalent to those of approximately 4.4–7.0-mm-deep crack initiation in large-scale specimens. In determining a design fatigue curve, strain-controlled tests are usually performed on small-sized specimens, and the fatigue life is then defined by the 25% load drop. It is reported that the cracks reach nearly 3–4-mm depth under those 25% drop cycles. The test results confirm that the fatigue lives of large-scale specimens agree with those given by the best-fit curve for carbon and low-alloy steels, and no remarkable size effects exist for the crack depths compared in this study.

Based on the precedent design fatigue curves and recent fatigue data obtained from materials with different mechanical properties, new design fatigue curves with high general versatility in air have been developed by The Japan Welding Engineering Society (JWES). Structural materials with different tensile strength are utilized in fatigue tests to verify the validity of these design fatigue curves and discuss the mean stress effect. The materials employed in this study are austenitic stainless steel (SS) SUS316LTP, carbon steel (CS) STPT370, low-alloy steels (LASs) SQV2A and SCM435H, all of which are used in the structural components of nuclear power plants of Japan. The best-fit curves (BFCs) are formulated by using the parameter of tensile strength to describe the relationship between strain (stress) amplitude and fatigue life [1]. The results of fully reversed axial fatigue tests conducted with small-scale test specimens of those materials in air at ambient temperature show good agreement with the developed BFCs. The results of fatigue tests also indicate that the mean stress effect is remarkable in materials with higher tensile strength. The applicability of Modified Goodman and Smith-Watson-Topper (SWT) approaches to the design fatigue curves is compared and discussed when considering mean stress effect. The correction of mean stress effect with SWT approach shows a good agreement with the developed BFCs.

Composite Reinforced Accumulator (CRA) is widely used in hydrogen stations. A high-cost pressure cycle test is mandatory to ensure the safety of accumulator in present regulations. To reduce the high cost, the aim is to develop a methodology of numerical fatigue life prediction of CRA with results of pressure cycle tests. An axisymmetric finite element model for the Type III accumulator is created and actual loading process including autofrettage pressure is simulated. Stress amplitude caused by pressure cycle is evaluated based on the instructions in KD-3 of ASME BPVC VIII 3-2015. By comparing stress amplitude distributions with the leak positions after the pressure cycle test, and plotting the results in the design fatigue curve, it could be shown that fatigue life prediction of Type III accumulator can be done by precise finite element analysis of the liner including dome part, where the principal axes of stress change in pressure cycle.

Operation flexibility and the high efficiency of thermal cycle are the two hot research topics for the steam turbine unit. For the operation flexibility, it requires the turbine unit be able to start up quickly and frequently, and good thermal fatigue properties of the components are a must. On the other hand, the trend to higher temperature with ultra supercritical (USC) steam is the key drive of ongoing development for the validation of a method to improve the thermal cycle efficiency, but the application of the higher steam temperature will intensify the component’s thermal fatigue. In this paper, a high pressure inner casing for the new designed solar steam turbine was studied, and the Finite Element Method (FEM) combined with linear elastic material was applied to simulate the transient stress and temperature fields during the daily warm/hot startup process. On the basis of the stress spectrum in the critical zone, the standard DIN EN-12952-3 and the fatigue curve were used to evaluate the low cycle fatigue life consumption in the transient process.

Delayed coker drums are unique in hydrocarbon processing facilities in that estimating their true design and service life has been problematic. Generally, pressure containing equipment in these facilities is designed using the notion of design life based on required pressure thickness and corrosion allowance considerations. Hence, pressure containing equipment is routinely monitored by facility inspectors for wall thickness. Although many analysts have ascribed coke drum failure to “thermal stress cycling”, the difficulty posed by the operation of coke drums results in an inability to measure or calculate the magnitude of the thermo-mechanical “stresses” and the actual number of significant exposures, that is, cycles causing fatigue damage. As well, the use of Code construction practices has been generally misapplied, for this specific equipment, as the practices are intended to define a safe design life rather than a service life. Indirect measures of service life based on shell bulge severity have fallen from favor by being ineffective. A trend to use a strain index method is somewhat more appealing but is based on static load and monotonic material property considerations rather than those properties indicative of thermal cyclic operation. Recent work has shown that thermo-mechanical strain cycling can be characterized quantitatively and used to determine a cyclic service life for both undamaged and damaged coke drums. This paper discusses some of the engineering specifics to generate a high probability estimate of coke drum fatigue service life for a new drum, a damaged-stable drum, drums with weld overlay and for drums exhibiting incremental damage.

In order to develop new design fatigue curves for carbon steels & low-alloy steels and austenitic stainless steels and a new design fatigue evaluation method that are rational and have clear design basis, Design Fatigue Curve (DFC) Phase 1 subcommittee and Phase 2 subcommittee were established in the Atomic Energy Research Committee in the Japan Welding Engineering Society (JWES). The study on design fatigue curves was actively performed in the subcommittees. In the subcommittees, domestic and foreign fatigue data of small test specimens in air were collected and a comprehensive fatigue database (≈6000 data) was constructed and the accurate best-fit curves of carbon steels & low-alloy steels and austenitic stainless steels were developed. Design factors were investigated. Also, a Japanese utility collaborative project performed large scale fatigue tests using austenitic stainless steel piping and low-alloy steel flat plates as well as fatigue tests using small specimens to obtain not only basic data but also fatigue data of mean stress effect, surface finish effect and size effect. Those test results were provided to the subcommittee and utilized the above studies. Based on the above studies, a new fatigue evaluation method has been developed.

This paper proposes simplified methods to evaluate fatigue damage in a component subjected to cyclic thermal loading, in order to visualize the distribution of usage factor using a graphical user interface (GUI) incorporated in a widely-used commercial CAE. The objective is to perform the evaluation and visualization using a standard desktop PC. In the previous paper, three simplified methods based on elastic finite-element analysis (FEA) were proposed in place of the method in the procedures employed in ASME Section III Subsection NH. In this paper, the methods have been improved for elastic-plastic FEA. A previously performed thermal fatigue test with a type 304 stainless steel cylinder was simulated. Heat transfer, elastic, and inelastic analyses were conducted. Simultaneously with the analyses performed, the equivalent total strain ranges and fatigue usage factor distributions were calculated using user subroutines developed in this study including three newly proposed simplified and ASME NH-based methods. These distributions can be visualized on a GUI incorporated in a commercial FEA code. The calculation results were consistent with the distribution of cracks observed. In addition, by using these, the analysts can visualize these distributions using their familiar CAE system.

This work analyzes the applicability of the ASME Boiler and Pressure Vessel Code procedures to calculate fatigue crack initiation under multiaxial stresses and/or strains, in particular when caused by non-proportional loads that lead the principal directions at the critical point to vary with time, e.g. due to out-of-phase bending and torsion loads induced by independent sources. Classic uniaxial fatigue damage models are usually inappropriate for analyzing multiaxial loads, since they can generate highly inaccurate predictions. Moreover, it is shown that the ASME procedures can lead to non-conservative results for non-proportional load histories.

Change in the fatigue life due to application of the mean strain was investigated for Type 316 stainless steel in simulated pressurized water rector (PWR) primary water environment. The tests were conducted by controlling the strain range to 1.2% for different strain rates of 0.4, 0.004, or 0.001%/s. The applied mean strain was 15% in nominal strain. In addition, cold worked specimens were also subjected to the tests without applying the mean strain. The tests using the cold worked specimens were regarded as the tests with the mean strain without increase in surface roughness due to application of plastic deformation. By inducing the cold working at low temperature, the effect of martensitic phase on the fatigue life was also examined. It was shown that the fatigue life of the stainless steel was reduced in the PWR water environment and the degree of the fatigue life reduction was consistent with the prediction model prescribed in the code issued by the Japan Society of Mechanical Engineers (JSME) and NUREG/CR-6909. Increases in peak stress and stress range due to cold working did not cause any apparent influence on the fatigue life. It was also shown that the 10.5 wt% martensitic phase induced by the low temperature cold working and the increase in the surface roughness caused by application of 15% mean strain did not bring about further fatigue life reduction. It was concluded that the effects of the mean strain, cold working, and martensitic phase were minor on the fatigue life in the PWR water environment. The current JSME and NUREG/CR-6909 models were applicable for predicting the reduction in fatigue due to the PWR water environment even if the mean strain or cold working was applied.

Strain controlled LCF testing extended to 10 million cycles revealed an abrupt endurance limit enforced by secondary hardening. In elevated temperatures the ε-N curve is rotated and endurance limit is lowered, but not vanished. When very low strain rates are applied at 325°C in simulated PWR environment, fatigue life is reduced, but far less than predicted according to NUREG/CR-6909. It is possible, but not probable that the difference is due to different stainless grades studied. We assume that the test method plays a more important role. We have repeatedly demonstrated in different tests campaigns that interruptions of straining with holds aiming to simulate steady state normal operation between fatigue relevant cycles can notably extend the fatigue endurance. Further proof is again presented in this paper. The suspected explanation is prevention of strain localization within the material microstructure and also in geometric strain concentrations. This actually suggests, that hold effects should be even more pronounced in real components. Cyclic behavior of austenitic steels is very complex. Transferability of laboratory data to NPP operational conditions depends on test environment, temperature, strain rate and holds in many ways not considered in current fatigue assessment procedures. In addition to penalty factors, also bonus factors are needed to improve transferability. Furthermore, it seems that the load carrying capacity of fatigued stainless steel is not compromised before the crack growth phase. Tensile tests performed after fatigue tests interrupted shortly before end-of-life condition in 325°C (N ≈ 0.85 × N 25 ) showed strength and ductility almost identical to virgin material. This paper provides new experimental results and discusses previous observations aiming to sum up a state of the art in fatigue performance of German NPP primary loop materials.

INCEFA-PLUS stands for INcreasing safety in nuclear power plants by Covering gaps in Environmental Fatigue Assessment. It is a five year project supported by the European Commission HORIZON2020 program that commenced in mid-2015 and in which sixteen organizations from across Europe participate. Specifically, the effects of mean strain/stress, hold time, strain amplitude and surface finish on fatigue life of austenitic stainless steels in light water reactor environments are being studied, these being issues of common interest to all participants. The project will develop proposals for improvements to methods for environmental fatigue assessment of nuclear plant components. Therefore, extensive testing capacity is being solicited in various laboratories across Europe in order to add to the existing amount of published data on environmentally assisted fatigue. Since there currently is no standard on environmental fatigue testing, it was imperative to come up with and agree upon a testing procedure within the consortium to minimize lab-to-lab variations in test results. This was done prior to the first phase of testing, but an update of the procedure was required after review of initial results, when additional potential lab-to-lab differences were identified. The current status of the so-called test protocol, and the key areas of difference found between different testing facilities, will be discussed. Due to the large test matrix within INCEFA-PLUS, distributed amongst various test laboratories, it has been necessary to develop a method to assign a data quality level to each test result, and a minimum data quality requirement for results that will be included in the project’s datasets used for analysis. Furthermore, the project has triggered international interest in facilitating mutual data access, and this requires data is gathered in a common database with data quality ratings applied. Ways to address the evaluation of data quality will be discussed. In a way, both activities, on a test protocol and on data review, jointly contribute to data quality by, respectively, ensuring a pre-test, common test procedure and a post-test, harmonized data evaluation. The large number of participants in the INCEFA-PLUS project presents a unique opportunity to gain consensus on light water reactor environment fatigue testing procedures and data quality assessment from experts working in a range of different organizations. The test protocol and data quality ratings developed within the INCEFA-PLUS project could be adopted by other organizations, or possibly used as the basis for future testing standards documents to harmonize approaches across the nuclear industry.

Following the flaw indications found in summer 2012 in two Belgian Reactors Pressure Vessels (RPV), WENRA recommended [1] the nuclear safety authorities in Europe to verify the material quality and integrity of the RPV in a 2-step approach: 1) a comprehensive review of the manufacturing and inspection records of the forgings of the RPV, 2) an additional UT examination of the base material of the vessels if needed. In this context, and to consolidate scientific basis on this issue, IRSN, the French technical safety organization, conducted, with CEA support, a test program aiming at studying the consequences of hydrogen flakes in large forgings of primary equipment (RPV, steam generator, pressurizer). Framatome provided the material to be investigated, namely two blocks of a steam generator vessel shell in 18MND5 steel: a block without flake — the reference block — and a block including a high density of hydrogen flakes. This shell — so called VB395 — was rejected because of an incident which occurred during the degassing heat treatment. Fracture toughness has been evaluated from 85 tests in the ductile range and the ductile-to-brittle transition range of the material. The test results on usual 0.5T-CT specimens were compared to those on specimens containing a hydrogen flake replacing the fatigue precrack. The latter were interpreted using 3D elastic-plastic X-FEM simulations allowing the modelling of the irregular flake geometry. Furthermore, large scale bending specimens with multiple flakes have been tested at −100°C. These tests were interpreted thanks to 3D X-FEM simulations allowing the analysis of the hydrogen flake interaction in terms of K J .

In order to conduct effective and rational maintenance activity of components in nuclear power plants, it is proposed to manage fatigue degradation based on crack size corresponding to an extent of cumulative fatigue damage. The purpose of this study focuses on the influence of strain rate in simulated reactor coolant environment for fatigue crack initiation and growth. 3-dimensional replica observations were conducted for environmental fatigue test specimens in different strain rates. Crack initiation and growth were observed in the experiments. It is clarified that low strain rate influences crack propagation and coalescence and increases crack growth rate that finally decrease fatigue life.

The update of the ASME III design fatigue curve for stainless steel in conjunction with the F en model described in the NUREG/CR-6909 report has been criticized since publication. Data used to develop curves and models raises more questions than it answers. Material testing in a simulated light water reactor environment is difficult due to the temperature and pressure involved. The experimental challenge makes it tempting to take shortcuts where they should least be taken. Facing and overcoming the challenges, direct strain-controlled fatigue testing has been performed at VTT using a unique tailored-for-purpose EAF facility. The applicable ASTM standards E 606 and E1012 are followed to provide results that are directly compatible with ASME Code Section III. Several earlier PVP papers (PVP2016-63291, PVP2017-65374) report lower than calculated experimental F en factors for stabilized stainless steels. In this paper new results, in line with the previous years’ conclusions, are presented for nonstabilized AISI 304L tested with dual strain rate waveforms. To model environmental effects more accurately, an approach accounting for the damaging effect of plastic strain is proposed. A draft F en model, similar in structure to the NUREG model but with additional parameters, is shown to significantly improve the accuracy of F en prediction.

Unexpected findings on time and temperature dependent behavior have been recorded during our research on fatigue performance of niobium stabilized stainless steel. Cyclic straining at 325°C and low strain rates resulted in higher stress responses than in higher rate tests. This effect is particular strong in PWR water environment. Subsurface bulk effect in environment is in contrast to the assumption on similar responses in air and environment, which is the foundation of the ‘companion specimen’ method where the strain in environment is measured from a parallel specimen similarly tested in air. Our data shows that environmental effects caused by PWR water cannot be isolated as a separate issue. Environment, temperature and strain rate are factors, which interactively affect the cyclic response and fatigue performance of stainless steel in relevant temperatures and loading conditions. The current ASME Code Section III design curve is based on different translation of the laboratory data than that made by Langer et al. The resulting effect is not as radical as caused by replacement of the original test data to a new highly scattered data base. But also the procedural changes have detectable effects and open a door for a provocative question: “is the design curve still compatible with the code?”

For several decades the nuclear industry has used structural weld overlays (SWOL) to repair and mitigate cracking within pressurized water reactor (PWR) components such as nozzles, pipes and elbows. There are two known primary mechanisms that have led to cracking within PWR components. One source of cracking has been primary water stress corrosion cracking (PWSCC). Numerous SWOL repairs and mitigations were installed in the early 2000s to address PWSCC in components such as pressurizer nozzles. However, nearly all of the likely candidate components for SWOL repairs have now been addressed in the industry. The other cause for cracking has been by fatigue, which usually results from thermal cycling events such as leakage caused by a faulty valve close to the component. The PWR components of most concern for fatigue cracking are mainly stainless steel. Thus, ASME Section XI Code Case N-504-4 would be a likely basis for SWOL repairs of these components, although this Code Case was originally drafted to address stress corrosion cracking (SCC) in boiling water reactors (BWR). N-504-4 includes the requirements for the SWOL design and subsequent analyses to establish the design life for the overlay based on predicted crack growth after the repair. This paper presents analysis work performed using Code Case N-504-4 to establish the design life of a SWOL repair applied to a boron injection tank (BIT) line nozzle attached to the cold leg of an operating PWR. The overlay was applied to the nozzle to address flaws found within the stainless steel base metal during inservice examination. Analyses were performed to calculate the residual stresses resulting from the original fabrication and the subsequent SWOL repair. In addition, post-SWOL operating stresses were calculated to demonstrate that the overlay does not invalidate the ASME Section III design basis for the nozzle and attached pipe. The operating and residual stresses were also used for input to a fatigue crack growth (FCG) analysis in order to establish the design life of the overlay. Lastly, the weld shrinkage from the application of overlay was evaluated for potential impact on the attached piping, restraints and valves within the BIT line. The combined analyses of the installed SWOL provide a basis for continued operation for the remaining life of the plant.

Commonly used piping vibration screening limits are typically justified by experience and lack a well-documented technical basis. This paper presents technical background for future Level 1 Fitness-for-Service (FFS) vibration screening criteria. The criteria assess the risk of fatigue in process piping due to bending mode type vibrations. Finite element analysis (FEA) of 20,000 randomly generated candidate-piping models and high-cycle welded joint fatigue curves for both constant amplitude and variable amplitude loading form the stress limits and basis for the proposed criteria. Most importantly, the proposed criteria aligns with historically used allowable vibration limits rooted in substantial experience. The allowable stress basis implemented in this paper considers periodic and random vibrations making it applicable to situations of mechanically induced, two-phase flow induced, turbulent-induced vibration of single-phase process fluid, or wind-induced, which may be manifested as either periodic or random. To reduce conservatism, limits are set for butt-welded and non-butt welded mainline piping to prevent use of a single blanket limit that may lead to unnecessary piping support alterations/additions, or costly piping configuration changes and unit downtime. Furthermore, the proposed Level 1-type criteria are consistent with previously proposed FFS Level 2 and 3 piping vibration fatigue evaluations [1] intended for inclusion in the ASME FFS-1/API 579 (API 579) Fitness for Service Standard [2].

This paper describes an experimental validation of the enhanced reference stress method to calculate fatigue J-integral ranges, which are effective in predicting the fatigue crack propagation rate under low–cycle fatigue loadings. Although J-integral type fracture mechanics parameters can be calculated via elastic–plastic finite element analysis (FEA) of the crack geometry, performing such an analysis is costly and requires a high–end computer. A simplified method for estimating the elastic–plastic J-integral is therefore desired. Herein, several representative simplified methods for estimating the elastic–plastic J-integral were applied to crack propagation prediction and compared with each other. The experiments referred to was a previously performed cyclic bending tests using wide–plate specimens containing a semielliptical surface crack. Limit load correction factors to improve the accuracy of the reference stress method were estimated by performing an elastic–plastic FEA. The predicted crack propagation behaviors were compared against the test results.