Assessment Procedure for Multiple Crack-Like Flaws in Fracture Assessment Diagram (FAD)

Assessment of multiple discrete crack-like flaws is one of the most common problems relating to pressure vessels and piping components. Under the current Fitness for Service (FFS) rules, such as ASME, BS and so on, multiple crack-like flaws are usually recharacterized as an enveloping crack (defined as a single larger crack), following their assessment rules. The procedure, however, varies significantly in these FFS codes. In this paper, the interaction between non-aligned multiple unequal cracks is clarified by applying the body force method. Based on the interaction which indicates the magnification and shielding effects and the reference stress solutions, a newly developed assessment procedure for multiple discrete crack-like flaws in the Fracture Assessment Diagram (FAD) is proposed.

A Probabilistic Evaluation Model for Welding Residual Stress Distribution at Piping Joint in Probabilistic Fracture Mechanics Analysis

Stress corrosion cracking (SCC) has been observed at some piping joints made by Austenitic stainless steel in BWR plants. In JAEA, we have been developing probabilistic fracture mechanics (PFM) analysis methods for aged piping based on latest aging knowledge and an analytical code, PASCAL-SP. PASCAL-SP evaluates the failure probability of piping at aged welded joints under SCC by a Monte Carlo method. We proposes a simplified probabilistic model which can be applied to the failure probability analysis based on PFM for welded joint of piping considering the uncertainty of welding residual stress. And the probabilistic evaluation model is introduced to PASCAL-SP. A parametric PFM analysis concerning uncertainties of residual stress distribution using PASCAL-SP was performed. The PFM analysis showed that the uncertainties of residual stress distribution largely influenced break probability. The break probability increased with increasing the uncertainties of residual stress.

Experimental Determination of J-R Curves Using SENB Specimens and P-CMOD Data

Fracture toughness and J-R curves of ductile materials are often measured under the guidance of ASTM standard E1820 using the single specimen technique and the elastic unloading compliance method. For the standard single-edge notched bend [SENB] specimens, the load, load-line displacement (LLD), and crack-mouth opening displacement (CMOD) are required being measured simultaneously. The load-CMOD data are used to determine the crack extension, and the load-LLD data together with the crack extension are used to determine the J-integral values in a J-R curve test. Experiments have indicated that the CMOD measurement is very accurate, but the LLD measurement is difficult and less accurate in a fracture test on the SENB specimen. If the load-CMOD records is used to determine the crack extension and the J-integral values, experimental accuracies for the J-R curve testing would be increased, and the test costs can be reduced. To this end, this paper develops a simple relationship between LLD and CMOD that is used to convert the measured CMOD record to the corresponding LLD data, and then to calculate the J values for a growing crack in a J-R curve test on the SENB specimen using one single specimen technique. The proposed method is then verified by the experimental data of J-R curves for HY80 steel using the SENB specimens and the load-CMOD data only. The results show that the proposed method is more accurate and more cost-effective for the J-R curve testing.

Small Punch Creep Properties of Heat Affected Zones of Reduced Activation Ferritic Steel

The small punch (SP) creep test was carried out at the temperatures of 823∼923 K by using a further miniaturized specimen, namely, TEM disk-type specimen (φ 3.0×t0.25 mm). The tests were applied to the fine grain heat affected zone (FGHAZ), tempered HAZ (THAZ) and base metal (BM), respectively, which were removed from the joint of the reduced activation ferritic steel welded by an electron beam welding, in order to investigate the creep properties of such local regimes. The results obtained from the SP creep test were correlated with those of uniaxial creep tests using the base metal (BM) and welded joint (WJ). Experimental results revealed that there were no large differences between the SP creep rupture strengths of the FGHAZ and THAZ and that of the BM at the relatively high load levels. This result was in good agreement with the fact that the uniaxial creep strength of the WJ was almost coincident with that of the BM at the relatively high stress levels. In addition, the ratio of load (P) to stress (σ), which gave same rupture time, was calculated by using the creep rupture data of the BMs. As a result, the ratio was determined to be 0.43, resulting in the following equation; P = 0.43 σ.

An Analysis of Residual Stress Improvement for a Dissimilar Metal Nozzle-to-Pipe Weld by Using the Heat Sink Method

This study is concerned with the mechanics analysis of residual stress improvement by the heat sink method applied to a dissimilar metal weld (DMW) for the use in nuclear power plants. The DMW joint considered here is composed of ferritic low-alloy steel nozzle, austenitic stainless steel safe-end, and nickel-base alloy A52 weld metal. To prepare the DMW joint with a narrow-gap, the gas tungsten arc welding (GTAW) process is utilized, and the heat sink method is employed to control thermal gradients developed in the critical region of work pieces during welding. Weld residual stresses are computed by the non-linear thermal elasto-plastic analysis using the axisymmetric finite element (FE) model, for which temperature-dependent thermal and mechanical properties of the materials are considered. A full-scale mock-up test is conducted to validate analytical solution for the DMW joint, and residual stresses are measured by using the hole-drilling method. Results of the FE modeling and mock-up test for the DMW joint are compared and effects of the heat sink method are discussed. It is found that a significant amount of residual compressive stresses can be developed on the inner surface of the DMW joint by using the heat sink method, which can effectively reduce the susceptibility of the welded materials to stress corrosion or fatigue cracking.

Current Wall Thinning Measured on Piping System of Main or Auxiliary Boiler Plant in Ships

A disaster happened in a nuclear power plant in Japan in August 2004, which was caused by failure of condensation water pipe in the secondary line. Shipping industries were concerned for possibility of occurrence of such a disaster in ships due to its construction similarity to marine boiler plant in steam, feed water and condensation piping for main or auxiliary boilers. Nippon Kaiji Kyokai has therefore investigated and gathered data of piping lines corrosion in ships collaborated with major Japanese ship owners right after the disaster. The results show that similar corrosion failure as in the nuclear power plant has occurred in shipboard steam/feed water/condensation water pipes for main and auxiliary boiler plants without causing severe consequences. The wall thickness measurements on actual pipe lines of steam, feed water and condensation water at bend parts, at T-junction, behind orifices, behind valves and at diffusers/reducers with a ultrasonic thickness gauge show a very definite evidence of a reduction in wall thickness of carbone steel pipes. It was confirmed that the amount of actual reduction in wall thickness could be well predicted by Kastner Equation [2–3].

UK Research Programme on Residual Stresses: A Review of Progress

A long-term UK research programme on residual stresses was launched in 2004. It involves Rolls-Royce plc and Serco Assurance, supported by UK industry and academia. The programme is aimed at progressing the understanding of weld residual stresses and the implementation of finite element simulation and residual stress measurement for assessing the integrity of engineering structures. Following on from this, the intention is then to develop improved guidance on residual stress modelling techniques and then to provide methods and analysis tools for design in order to control and minimise residual stress. The focus of the work to date has been to develop modelling guidelines which can be used by a finite element analyst to predict the residual stresses in austenitic welded components. These guidelines are now drafted and will be incorporated into the next issue of the British Energy R6 defect assessment procedure following peer review. The guidelines have been developed based on the experience that has been attained using various modelling techniques. To support this development, a series of welded mock-ups have been manufactured. The residual stresses in these welds have been measured using various techniques (diffraction and strain relaxation). These measurements are being used to validate the predicted stresses. It is only by corroborating each other that the resulting residual stresses can be confidently used for assessment. Mock-ups are also being used to develop material models for ferritic steel which undergo phase transformations, and to investigate how various weld parameters affect the magnitude and distribution of residual stress. Similarly, mock-ups have been manufactured to investigate the effect of start-stops on residual stresses. The programme is also supported by experimental testing to develop physical and mechanical properties which are required for analysis, i.e. up to melting temperature. Both conventional and miniaturised testing has been used to measure properties in ferritic and austenitic steels. A task has also been undertaken to develop a methodology for providing upper bound residual stress profiles which can be used as an initial estimate of stress for use in structural assessment.

A Fatigue Crack Driving Force Parameter

Most of the previous parameters that utilized as a crack driving force were established in modifying the parameter Kop in Elber’s effective SIF range (ΔKeff = Kmax–Kop). This paper focuses on the physical meaning of compliance changes caused by plastic deformation at the crack tip, the test was carried out for structural steel under constant amplitude loading, and differences of several parameter ΔKeff in literature are analyzed quantificationally. The effect of actual stress amplitude at the crack tip on fatigue crack growth is investigated, and improved two-parameter driving force model ΔKdrive(=Kmax)n(ΔK^)1−n) has been proposed. Experimental data for several different types of materials taken from literature were used in the analyses. Presented results indicate that the parameter ΔKdrive is equally effective or better than ΔK(=Kmax-Kmin), ΔKeff(=Kmax-Kop) and ΔK*(=(Kmax)α(ΔK+)1−α) in correlating and predicting the R-ratio effects on fatigue crack growth rate.

Hydrogen Effects on the Burst Properties of Type 304L Stainless Steel Flawed Vessels

The effects of hydrogen and burst media on the burst properties of Type 304L stainless steel vessels were investigated. The purpose of the study was to compare the burst properties of hydrogen-charged stainless steel vessels burst with different media: water, helium gas, and deuterium gas. A second purpose was to provide data to improve an existing finite-element model for predicting burst behavior. Burst tests were conducted on hydrogen-charged and uncharged axially-flawed cylindrical vessels. The results indicate that samples burst pneumatically had lower volume ductility than those tested hydraulically. For pneumatic burst tests, samples burst with deuterium gas had slightly lower ductility than helium gas tests. For uncharged samples, burst pressure was not affected by burst media. For samples pre-charged with hydrogen, deuterium burst pressures were about 80% of the hydraulic or helium burst pressures. Hydrogen-charged samples had lower volume ductility and slightly higher burst pressures than uncharged samples. The results of the tests were used to verify and improve a previously developed predictive finite-element model. The existing finite-element model can qualitatively predict the expected changes in burst properties with hydrogen or tritium service, but a better material property database is required for quantitative predictions.

Prediction of Brittle Fracture Under Generalised Elastic Plastic Loading

This article describes an investigation into the ability of a number of different fracture mechanics approaches to predict failure by brittle fracture under general elastic/plastic loading. Data obtained from C(T) specimens of A508 ferritic steel subjected to warm pre-stressing and side punching were chosen as such prior loadings produce considerably non-proportionality in the resulting stress states. In addition, failure data from a number of round notched bar specimens of A508 steel were considered for failure with and without prior loading. Failure prediction, based on calibration to specimens in the as received state, was undertaken using two methods based on the J integral and two based on local approach methodologies.