This paper reports the results of a collaborative investigation of an ex-service grade 91 bend carried out by the UK generating companies Centrica, SSE, Engie and RWE. As part of the handover exercise for Centrica’s Langage power station in 2009 a number of routine checks were carried out on the main steam and hot reheat grade 91 steam pipework. In some cases low hardness readings were found with subsequent metallurgical replication showing the presence of an aberrant non martensitic microstructure. This led to a more extensive inspection programme on the steam lines and the discovery of other areas of suspect material. A review of the operating capability of the plant, including detailed pipework stress analysis and a pipework peaking assessment, along with the assumption that lower strength grade 91 material was present, led to the steam lines being down rated and returning to service under these revised conditions. At the first C inspection in December 2012, after the HRSG and associated pipework had operated for 18720 hours, a bend with a soft weld, along with a section of the straight pipe on either side, was removed from service. An investigation was undertaken to establish how long this component would have survived, had it been left in service, and to consider the implications for the future operation of the plant.

This paper reports the results of a collaborative small scale creep testing exercise carried out by the UK generating companies Centrica, SSE, Engie and RWE as part of an investigation of an ex-service grade 91 bend.

The impression creep test method using a rectangular indenter has been well established and the applicability of the technique has been supported by the test data for a number of metallic materials at different temperatures and stresses. The technique has proved to be particularly useful in providing material data for on-site creep strength assessments of power plant components operating in the creep regime. Due to these reasons, “standard” assessment procedures using the impression testing method are needed in order for the technique to be more widely used. This paper will first address some key issues related to the use of the impression creep test method, involving the data conversion method, typical test types and validity of the test technique etc. Then some recommendations on a number of practical aspects, such as the basic requirements of test rigs, “standard” specimen geometry, indenter dimensions, sampling procedures for scoop samples, specimen preparation, temperature and loading control, and displacement measurement, are briefly addressed. Finally, applications of the test data to assist with the risk management and life assessment programme of power plant components, particularly those with service-exposed materials, using data obtained from scoop samples, are described. Proposals for future exploitation and for improvement of the technique are addressed.

The small punch (SP) test technique can generate tensile, toughness, and creep strength data from very small specimens, enabling nondestructive sampling from large components. This has accelerated industry interest in SP, particularly in the energy sector. A round-robin for the SP creep test was organized within the EPERC network on a 1CrMoV rotor steel, primarily to contribute to harmonizing the test procedure for this emerging method. This presentation analyzes the SP creep test results performed at several temperatures and one load level to assess their inter-comparability and interpret the results concerning conventional uniaxial creep tests. The SP loading geometry appeared to be a key parameter in evaluating the SP results, particularly for correlation with uniaxial creep behavior. A standardization initiative for the SP test method, including creep as well as low-temperature testing for determining tensile and toughness properties of metallic materials, has recently started via a CEN Workshop.

Small punch creep testing (SPCT) is a small-scale, accelerated creep test that allows for the determination of creep data using a limited amount of material. The question, however, remains how the data generated by this technique correlate to more established techniques such as uniaxial testing and ultimately to predictions regarding the remaining service life of a plant component. This empirical study investigated the microstructure-to-property relationship of welded 9-12%Cr steels as measured using SPCT. Virgin P91 (X10CrMoVNb9-1) steel was joined to service exposed X20 (X20CrMoV12-1) steel using two different filler materials (X20 and P91) via fusion welding. Site-specific samples were extracted from the parent plates, heat affected zones and weld metals using electro-discharge machining. Small punch creep testing were performed using a 276 N load at a temperature of 625°C. The untested sample microstructures were quantitatively characterized using a range of electron microscopy techniques to determine the precipitate (M 23 C 6 , MX) spacing, subgrain sizes and dislocation densities for each region of the weldments. Multiple linear regression analysis found that the subgrain size (λsg) played the largest contribution to the SPCT rupture life. The heat affected zones had the lowest SPCT rupture times (49-68 hours), which corresponded to the largest subgrain sizes (1.1-1.3 μm). The P91 parent plate material had the longest SPCT rupture time (349 hours), which corresponded to the lowest subgrain size (0.8 μm). The P91 weld metal sample showed lower initial deflection rates during the SPC testing, however the presence of non-metallic SiO 2 inclusions in this zone contributed to accelerated brittle failure.

The Creep Strength Enhanced Ferritic steel grade 91 is widely used for both retrofit applications and primary construction on high temperature power plant. Although to date most structural integrity issues with this material have been associated with welds, as the operating hours of these plants accumulate, there will be a growing need for remanent creep life assessment of the base material. Arguably this is already the case for aberrant grade 91 material entering service in an incorrectly heat treated condition. In these circumstances the strength may fall below the normally accepted lower bound of the creep strength range and some indication of actual strength may be required. One strategy to address potential base material failure is to use small scale sampling of individual components, followed by small scale creep testing, to investigate the current creep strength present. The data can be compared with the equivalent data produced for well characterised material known to be at the lower bound of the creep strength range. This paper describes a methodology for using the impression creep data obtained to provide both creep strength ranking and an estimate of absolute creep strength for individual grade 91 components. This will enable appropriate judgements to be made by plant operators on repair/run decisions. For those components remaining in service, it allows for the weakest items to be given priority for early re-inspection at future outages. The ultimate goal is to identify base material creep damage development at as early a stage as possible and well in advance of failure in service.

This paper investigates the effect of high temperature tensile strain on subsequent creep strength in grade 91 steel. Failed hot tensile specimens have been sectioned at various positions along the specimen axis, and therefore at different levels of hot tensile strain, to obtain material for creep strength evaluation. Because of the limited amount of material available for creep testing obtained in this way, creep testing has been carried out using the specialised small-scale impression creep testing technique. The grade 91 material has been tested in both the normal martensitic condition and in an aberrant mis-heat treated condition in which the microstructure is 100% Ferrite. The latter condition is of interest because of its widespread occurrence on operating power plant with grade 91 pipework systems.

Remaining-life assessment of high temperature components using the small punch (SP) creep testing technique necessitates the evaluation of SP load (F)/uniaxial stress (σ) conversion factor, F/σ, obtained by comparing the SP and uniaxial creep test results. In the present study, the SP creep tests were carried out at 850°C on various Ni-base alloys having different reduction of area in the range of 0.05-0.67 to investigate the influence of creep ductility on the value of F/σ. The F/σ value was determined for each alloy by correlating SP creep rupture data with corresponding uniaxial creep ones. The experimental results revealed that the F/σ value was not well correlated with Vickers hardness, but it increased almost linearly with increasing reduction of area up to around 0.4. This result indicated that the SP creep rupture data could be converted to the uniaxial data if the creep ductility on a given material was available.

Creep properties of 9Cr heat resistant steels can be improved by the addition of boron and nitrogen to produce martensitic boron-nitrogen strengthened steels (MarBN). The joining of this material is a crucial consideration in the material design since welds can introduce relatively weak points in the structural material. In the present study, creep tests of a number of MarBN weld filler metals have been carried out to determine the effect of chemistry on the creep life of weld metal. The creep life of the weld metals was analysed, and the evolution of creep damage was investigated. Significant differences in the rupture life during creep have been observed as a function of boron, nitrogen and molybdenum concentrations in the weld consumable composition. Although the creep lives differed, the particle size and number in the failed creep tested specimens were similar, which indicates that there is a possible critical point for MarBN weld filler metal creep failure.

Due to their excellent high temperature oxidation resistance, utilities worldwide are adopting advanced austenitic stainless steels (A-ASS) for critical plant components, such as heat exchangers, as they aim to achieve higher operating conditions. However, challenges may be encountered in developing life assessment and life management strategies for such components. This is because conventional methods used for life assessment, such as measuring steam side oxide scale thickness in ferritic and conventional austenitic material to predict tube metal temperature, may not be successfully applied to A-ASS. In such instances, tracking the formation and evolution of microstructural features during service, may offer a possible method to predict the temperature of these steels. For such metallurgy based lifing strategy to be successful, it is essential to develop a good understanding of microstructure evolution in these steels. In this work one heat of Super 304H, that has been creep tested at 600°C, 650°C and 700°C, with applied stress ranging from 110 to 340 MPa, is characterized using a combination of advanced characterization tools and image analysis methods. The amount of sigma phase formed at the gauge and grip sections of the samples is quantified and the methodology used to quantify this phase is presented. From the results, a time-temperature-transformation diagram for sigma formation is developed.

A trial weld joint of COST F and COST FB2 steels was produced using the GTAW HOT-WIRE method in conditions used in industry for production of welding steam turbine rotors. Conventional long-term creep tests (CCT) to the rupture of this weldment and the base materials were carried out at temperatures ranging from 550 °C to 650 °C in the stress range from 70 to 220 MPa (the longest time to rupture was above 52,000 hours). Creep rupture strength was evaluated using Larson-Miller parameter model. Assessment of microstructure was correlated with the creep strength. Precipitation of Laves phase and structure recovery during creep exposures were the main reasons for the failure which occurred in the heat affected zone of steel COST F. The recently developed simulative accelerated creep testing (ACT) on thermal-mechanical simulator allows the microstructural transformation of creep-resisting materials in a relatively short time to a state resembling that of multiyear application under creep conditions. ACT of samples machined from various positions in the weldment was performed at 600 °C under 100 MPa. Changes in the hardness and the microstructures of the samples, which underwent both types of creep tests, were compared. Small sample creep test (SPCT), another alternative method how to obtain information about the creep properties of materials when only a limited amount of test material is at disposal, were performed. It was shown that the same stress-temperature dependence and relationships are valid in the SPCT as in the CCT. Using a simple load-based conversion factor between the SPCT test and the CCT test with the same time to rupture, the results of both test types can be unified.

Super 304H is a new generation of advanced austenitic stainless steels that is increasingly being used in superheater/ reheater (SH/RH) sections of thermal ultra-supercritical steam power plants due to its high creep strength combined with good oxidation resistance and microstructure stability. However, recent studies have shown significant microstructural changes and associated degradation in creep performance during long-term service exposure in this alloy. Microstructure evolution during service and its effect on the long-term creep performance has not been comprehensively assessed. In this work, variations in the microstructure of long-term service exposed Super 304H RH tubes (~99,600 hours at 596°C steam temperature) are documented. The results for the ex-service material are compared to well-documented laboratory studies to provide perspective on improved life management practices for this mainstay advanced stainless steel.

Grade 91 steel has been widely utilized in power plants over the last 20 years. Its specification worldwide has dramatically increased since the acceptance of Code Case 1943 for this material in 1983. Recent evaluation of a combination of ex-service Grade 91 steel components and virgin material has provided a unique opportunity to independently assess the performance of a combination of base metal and weldments. This approach has been grounded in the fundamental objective of linking metallurgical risk factors in Grade 91 steel to the cross-weld creep performance. Establishing critical risk factors in 9Cr steels is regarded as a key consideration in the integration of a meaningful life management strategy for these complex steels. The potential metallurgical risk factors in Grade 91 steel have been fundamentally divided into factors which affect strength, ductility or both. In this study, two heats of ex-service Grade 91 steel which exhibit dramatic differences in strength and ductility have been evaluated in the ex-service condition and re-heat treated to establish a relevant set of strength:ductility variables. This set of variables includes [strength:ductility]: low:low, medium:low, low:high and medium:high. The influence of these strength:ductility variables were investigated for feature type cross-weld creep tests to better evaluate the influence of the initial base material condition on cross-weld creep performance.

Creep-fatigue crack formation (endurance) and crack growth rate data are necessary inputs for assessing the structural integrity and for estimating the design life of high temperature components in power generation and aircraft engine industries. Ensuring consistency in the reported test data, as well as an understanding of the inherent scatter and its source in the data, are both necessary for assuring quality and limitations of the analyses that rely on the data. In 2008, the American Society for Testing and Materials (ASTM) under the umbrella of its subcommittees E08.05 on Cyclic Deformation and Crack Formation and E08.06 on Crack Growth, and the sponsorship of Electric Power Research Institute (EPRI) through its international experts’ working group on creep-fatigue embarked on the task of developing separate standard test methods for creep-fatigue crack formation and creep-fatigue crack growth. The first standard entitled, “E-2714-09: Standard Test Method for Creep-fatigue Testing” was developed in 2009 and was followed up with a round-robin consisting of 13 laboratories around the world for testing the newly developed standard. This paper discusses the results of this round-robin concluded in 2012 using the widely used P91 steel that led to the formulation of the Precision and Bias statement contained in the version of the ASTM standard E2714 that was successfully balloted in the year 2013.

Boron and nitride additions are emerging as a promising design concept for stabilizing the microstructure of creep-resistant martensitic high-chromium steels. This approach, known as MarBN steel (martensitic steel strengthened by boron and nitrogen), combines the benefits of solid solution strengthening from boron with precipitation strengthening from nitrides. However, initial welding trials revealed challenges in achieving a uniform fine-grained region in the heat-affected zone (HAZ), which is crucial for mitigating Type IV cracking and ensuring creep strength. Despite these initial hurdles, preliminary creep test results for welded joints have been encouraging. This study presents an improved MarBN steel formulation and its investigation through uniaxial creep tests. Base material and welded joints were subjected to creep tests at 650°C for up to 25,000 hours under varying stress levels. The analysis focused not only on the creep strength of both the base material and welded joints but also on the evolution of damage. Advanced techniques like synchrotron micro-tomography and electron backscatter diffraction were employed to understand the underlying creep damage mechanisms. By combining long-term creep testing data with 3D damage investigation using synchrotron micro-tomography, this work offers a novel perspective on the fundamental failure mechanisms occurring at elevated temperatures within the HAZ of welded joints in these advanced steels.

An internal pressure creep test has been carried out on a Gr. 91 steel longitudinal welded pipe at 650°C to examine the type IV failure behavior of actual pipes, using a large-scale experiment facility “BIPress”, which can load internal pressure and bending force on large diameter pipes at high temperatures. The creep test was also interrupted three times to measure hardness and voids density in the HAZ region of the outer surface of the test pipe. Results of the measurement of the hardness and voids density at the interruption did not indicate creep damage accumulation. The welded pipe suddenly ruptured with large deformation, which caused crushing damage to the surrounding facility. Type IV cracking occurred in the longitudinal welded portion of the test pipe, and the length of the crack reached 5000mm. SEM observation was carried out at the cross section of the welded portion of the test pipe and voids density was measured along the thickness direction in the HAZ region. To clarify the stress/strain distribution in the welded portion, creep analysis was conducted on the test pipe, where the materials are assumed to consist of base metal, weld metal and HAZ. After stress redistribution due to creep deformation, stress and strain concentrations were observed inside the HAZ region. Then, the authors' creep life prediction model was applied to the creep test result to examine its validity to actual size pipes. It was demonstrated that the life prediction model can evaluate damage of the Gr. 91 steel longitudinal welded pipe with sound accuracy.

Alloy 718 is an important class of Nb-bearing Ni-based superalloys for high-temperature applications, such as compressor disks/blades and turbine disks in gas turbine systems. The service temperature of this alloy is, however, limited below 650 °C probably due to the degradation of its strengthening phase γ"-Ni3Nb. Aiming at understanding and improving creep properties of 718-type alloys, we investigated creep behaviors of alloy 718 and alloy Ta-718 where different types of γ" phases, Ni3Nb and Ni3Ta, were precipitated, respectively. Creep tests were conducted at 700 °C under stress conditions of 400 and 500 MPa for the two alloys in aged conditions. It was found that while the minimum creep rates were comparable in the two alloys, the creep rate acceleration was lower in alloy Ta-718 than in alloy 718 under the creep conditions studied. Microstructural observations on the specimens before and after the creep tests suggested that the γ" precipitates were distinguishably finer in alloy Ta-718 than in alloy 718 throughout the creep tests. The formation of planar defects and shearing of γ" precipitates occurred frequently in the alloy 718 specimen. The observed creep deformations were discussed in terms of the critical resolved shear stress due to shearing of γ" particles by strongly paired dislocations.

In order to study the effect of precipitation strengthening by MX precipitates on the restriction of microstructure degradation in 9 mass% Cr ferritic heat-resistant steels, V, Nb additioned model steels were evaluated by microstructure analysis through TEM and EBSD with reference to the creep test and creep interrupting test. VN precipitation increased the creep strength if the content was higher than 0.02%. Simultaneous addition of Nb and V in the specimen resulted in the complex NbC-VN precipitates even in the as-heat-treated specimens. The coherent and fine-needle-type VN was also detected in the steel. These precipitates are expected to increase the creep strength according to the creep strain curves. V variation up to 0.02% did not affect the crystallographic character of the grain boundary in the as-heat-treated specimens. Nb variation affected the crystallographic character of the grain boundary significantly because of the grain refinement effect of NbC. VN precipitation during the creep test restricted the crystallographic misorientation-angle-profile degradation. Integrating all intragranular precipitates, VN, restricts the crystallographic degradation significantly. The long-term creep test results and the precise precipitation analysis will be disclosed by the presentation.

A constitutive equation, with parameters derived from the interpolation of primary and steady state stages of constant load creep curves, has been utilized to estimate the stress relaxation behavior of the martensitic steel X20Cr13, alloy used in many high temperature applications, including heavy duty gas turbines. Creep and stress relaxation tests have been performed at 350°C, close to the negligible creep temperature of the studied alloy for stresses of interest for engineering applications. The creep tests were carried out at stresses below and above the yield stress, whereas, for the relaxation stress tests, the imposed strain was in the range 0.2% to 1.2% with the purpose to have, at the beginning of the tests, the same initial stresses of the performed creep tests. After a stress relaxation period, lasting between 10 to 1000 hours, each specimen was generally reloaded at the initial stress and a new relaxation test, on the same specimen, was carried out. This “reloading procedure”, simulating the re-tightening of bolts, has been repeated several times. The proposed equation has shown to well predict the experimental creep and stress relaxation behavior of the steel under investigation.

This study conducted creep tests, microstructural, and hardness analyses on SA213T23-TP347H dissimilar weld joints of long-term serviced coal-fired boiler final superheater tube. The welded joint (SA213 T23-TP347H) of the superheater tube, after approximately 105,000 hours of service, was sampled for creep life assessment and maintenance planning. Creep tests were conducted at 600°C under three stress conditions: 100, 140, and 160MPa. Most cracks were observed in the heat-affected zone of T23, and compared to unused tubes, the creep life consumption rate was approximately 90%. All dissimilar weld joints used welding rods similar in chemical composition to T23, and significant hardness reduction occurred in the flame-affected zone.