Brian N. Leis

Progress in development of acceptance criteria for local thinned areas in pipe and piping components

Over the last 30 years there has been a considerable amount of research conducted on the effect of corrosion on the burst strength of buried gas and oil transmission pipelines. The results of numerous burst tests on artificial flaws and corroded pipe removed from service were used to validate an empirical analysis that was essentially the limit–load solution for an axial crack in a pipe under pressure loading. This basic concept led to acceptance standards in ANSI B31G, and a more recent modified B31G criterion using the RSTRENG computer program developed at Battelle. This program takes into account variable flaw depths rather than the parabolic flaw shape assumed in the original B31G criterion. Since that time, more fundamental research has been conducted to develop a more accurate and theoretically based failure criterion. The Battelle/Pipeline Research Committee International PCORR computer program is an example of a special purpose shell-element based, finite element, PC criterion for the evaluation of local thinned area (LTA) flaws. This program has evolved with time from linear-elastic to elastic-plastic stress with provisions for axial as well as hoop stresses. The development and new insights into blunt flaw behavior resulting from this program will be one aspect covered in this paper. In the nuclear industry erosion-corrosion, or flow-accelerated corrosion, in single-phase liquid lines has become a major problem. Computer programs, such as the EPRI Checworks program, have been developed to assist the plant operators with deciding where to focus their inspections. However, to date no generally validated acceptance criteria have been developed for the plant piping. Plant piping, whether in nuclear power plants, fossil power plants, or petrochemical plants, have several differences from buried pipelines which need to be considered. The buried pipelines typically have low longitudinal stresses that frequently are compressive, and have no pipe fittings such as tees, elbows, and reducers except at compressor stations. Plant piping needs to consider hoop stresses and axial tension loads from the pressure, as well as, bending stresses from dead-weight loads, thermal expansion stresses, and seismic loads. In an effort to develop flaw acceptance criteria for Section XI of the ASME Boiler and Pressure Vessel Code, the criteria in Code Case N-480 have been revised and implemented into a new code case (the number has not yet been assigned). These criteria essentially use either the ANSI B31G approach for axial flaws, or the ANSI B31.1 or ASME Section III stress analysis rules to show that the residual strength of the thinned region meets the initial design stress limits. This paper presents some of the validation efforts recently undertaken to determine the inherent margins in the design stress equation approach compared with the applied safety factors in the axial and circumferential flaw limit–load solutions in: (i) the gas and oil pipeline industries; (ii) the proposed criteria in Belgium for the nuclear industry and other criteria, and (iii) the preliminary criteria from a recently proposed ASME Code Case on erosion/corrosion acceptance criteria and the ASME Appendix H criteria for flawed ferritic nuclear pipe.

Mixed-mode stress intensity factors for interacting semi-elliptical surface cracks in a plate

Abstract Mixed-mode stress intensity factor solutions are presented for two parallel, but not necessarily coplanar, interacting surface cracks in a plate. The plate is subjected to a remote uniaxial stress acting normal to the crack planes. The crack separation distances, measured normal and parallel to the crack planes, are varied to provide 24 crack geometries. The smallest separation distance is equal to a quarter of the plate thickness. The equal sized interacting cracks have fixed depth-to-thickness ratios of 0.5 and fixed half-width-to-depth ratios of 3. The finite element alternating method is used to develop the solutions. The tendency for the cracks to grow apart or to coalesce is discussed.

Fatigue growth of initially physically short cracks in notched aluminum and steel plates

Much of the fatigue life of engineering structures is spent initiating cracks at a notch and propagating one (or more) of these cracks into and through the notch stress field. Experimental results and empirical analysis suggest that early in this process the physically small crack sometimes propagates in a manner inconsistent with analysis based on linear elastic fracture mechanics (LEFM). Available empirical adjustments for this apparent aberration postulate that the data can be made consistent with LEFM by the addition of a constant, with length dimensions, to the current crack length, resulting in a pseudo-crack. The present paper introduces and analyzes an extensive data set pertinent to this so-called short-crack problem. Included are results for notched plates made from two aluminum alloys and a steel. These plates, which contain either circular or one of two different elliptical notches, have been tested under load or displacement control and encompass both confined flow and gross-section yield. Results for crack lengths as small as 20 μm (0.0008 in.) are reported. Data presented and analyzed do indeed show trends which differ from those of LEFM analysis of so-called long cracks. More significantly, they show that the so-called short-crack behavior does not occur only for physically short cracks. Results presented indicate that cracks as large as 2.5 mm (0.1 in.) in the aluminum alloys and 1.25 cm (0.5 in.) in the steel also exhibit aberrations in their behavior as compared with longer-crack trends. Also, the results presented suggest that the data cannot be made consistent with the LEFM trend through the addition of a constant pseudo-crack length to the current crack size as has been suggested in one of the currently popular empirical models. It is postulated that the aberrations observed are a consequence of the failure of the LEFM-based analyses to recognize that crack growth in the plastic field of the notch is dominantly displacement-controlled local to the crack tip. Other relevant aspects of the problem are also discussed.

Fatigue crack propagation through inelastic gradient fields

Abstract This paper examines the growth rate behaviour of cracks growing in the inelastic gradient field of a notch for a range of crack sizes, stress levels and notch severities. Experimental data used as a vehicle to explore the utility of K as a correlating parameter show that linear elastic fracture mechanics (LEFM) errs significantly. Previously suggested pseudo-plastic forms of K , as well as a parameter based on the J integral, are also shown not to consolidate these data during crack growth through the inelastic field of the notch. Calculations of growth intervals are made using conventional analysis procedures and contrasted with that experimentally observed to illustrate possible errors that may be made in practical applications of LEFM to ductile materials. The significance of these errors is discussed with regard to current code specified procedures based on LEFM, whose use to calculate in-service inspection intervals in ductile piping systems may lead to non-conservative estimates.

Investigating a Mechanism for Transgranular Stress Corrosion Cracking on Buried Pipelines in Near-Neutral pH Environments

Abstract Buried carbon steel fuel transmission pipelines, protected by external coatings and cathodic protection, are known to experience transgranular stress corrosion cracking (TGSCC). Failure an...

Relationship Between Apparent (Total) Charpy Vee-Notch Toughness and the Corresponding Dynamic Crack-Propagation Resistance

The consequences of a dynamic fracture in a gas-transmission pipeline require that pipelines be designed to avoid such incidents at a high level of certainty. For this reason, the related phenomonology has been studied since the early 1970s when the possibility of a dynamic ductile fracture was recognized. Full-scale experiments were done to characterize the fracture and gas dynamics associated with this process and empirical models were developed as a means to represent these experiments in a design or analysis setting. Such experiments focused on pure methane gas, and in the early days used steels with toughnesses less than 100 J, consistent with the steel making capabilities of the 1970s. Subsequently, interest shifted to larger diameter, higher pressure, higher BTU “rich” gases requiring higher toughness steels. The full-scale tests conducted to validate the arrest toughness levels determined that these empirical models were non-conservative.This paper presents a relationship between the dynamic crack propagation resistance and the apparent crack propagation resistance as measured by Charpy vee-notch (CVN) test specimens. This relationship is used in conjunction with the existing Battelle empirical criterion for dynamic-fracture arrest to determine the apparent toughness required to arrest a propagating ductile fracture in gas-transmission pipelines. The validity of this relationship is illustrated by successful predictions of arrest toughness in pipelines under a range of conditions including rich gases and high-toughness steels, including those showing a rising upper-shelf behavior.Copyright

A new model for characterizing primary creep damage

Primary creep damage may occur at a crack tip in steel at room temperature and below. The effect of this time-dependent damage is generally neglected. Recently developed experimental data clearly show that, for some materials, neglecting time-dependent deformation and damage may be quite nonconservative and dangerous in certain practical applications.In this paper, a new theory is developed which incorporates the effects of time-dependent damage into the crack growth and failure process. The predictive ability of the model is verified first by finite element analyses and then by comparison to experimental data. It is clearly shown that, for the material considered here, time-dependent damage effects must be considered or the crack growth process is not adequately accounted for.

An examination of the influence of residual stresses on the Fatigue and fracture of railroad rail

Fatigue cracks in rails can be the source of failures and subsequent derailments. The present paper examines the effect of residual stresses on the fatigue and fracture resistance of rail steel materials and the probable influence of residual stresses on the fatigue performance of rails in service. The magnitudes of residual stresses in rails were measured as a part of this study. Tensile and compressive residual (mean) stresses are shown to have a significant effect on the fatigue resistance of rail steels through tests using cylindrical smooth and compact tension specimens (free of inherent residuals) over a range of mean stress conditions. Fatigue and fatigue crack growth rate data developed are analyzed using specific damage parameters to consolidate the effect of mean stress. These data are then used in combination with measured residual stresses on rails to estimate the influence of service/fabrication induced, residual stresses in actual rail.

Constraint Corrected J-R - Curve and its Application to Fracture Assessment for X80 Pipelines

Single specimen J-R curve testing for X80 pipeline steel was conducted using single edge notched bend (SENB) and single edge notched tension (SENT) specimens with various crack lengths. Test data indicate that the J-R curves for this steel are strongly constraint dependent. To facilitate transferability of experimental J-R curves to those for actual cracked components, this paper develops a constraint corrected J-R curve for X80 steel. A modified J-Q theory that can consider the global bending stress influence is proposed so as to correctly quantify constraint effect on the crack-tip fields and the J-R curves. Results show that the modified J-Q solution can well match numerical crack-tip fields for bending specimens, with Q being a load-independent constraint parameter under large scale yielding. Based on the experimental data and numerical analysis, a constraint corrected J-R curve is formulated as a function of the parameter Q and crack extension Δa for X80 steel. A general procedure to predict J-R curves for actual cracked components is then outlined. Comparison indicates that the predicted J-R curves developed in this paper agree well with the experimental data for both SENB and SENT specimens. To demonstrate its application in failure assessment, the constraint corrected J-R curve for X80 steel is used to determine failure loads for a surface cracked pipeline. Reasonable agreement to available analytic solution is achieved.

Initiation of Stress-Corrosion Cracking on Gas Transmission Piping

This paper explores the initiation of stress-corrosion cracking (SCC) on the exterior of gas transmission piping. Initiation is taken here to include processes leading to the formation of a crack-like feature as well as early growth. Initiation is characterized in terms of the microstructural factors and the mechanics conditions that influence initiation and subsequent growth. Field and laboratory data are introduced to identify features typical of, but somewhat unique, to SCC initiation on gas-transmission pipelines. It is shown that the cracking behavior can be grouped as a function of the crack spacing perpendicular to the maximum principal stress. Patches of cracks are defined as sparse or dense in terms of this spacing, with a circumferential crack spacing on the order of 0.2x (wall thickness) separating these two cracking patterns. These results show that cracks with dense spacings tend to dormancy whereas the sparse spacings continue to grow. Fracture mechanics analysis is used to rationalize the field cracking patterns. It is shown that compliance changes due to the presence of adjacent crack tips account for the field cracking behavior. The results show that initiation patterns, which lead to closely spaced crack arrays, will tend to a dormant state as the cracks grow, because this growth leads to shielding of the tips from the service stresses. In contrast, growth can continue in arrays where the conditions at initiation give rise to sparse crack spacings. In addition, the fracture mechanics results are used to evaluate the suitability of the tapered-tension test-a commonly used procedure to characterize SCC initiation. Laboratory and modeling results are shown to be consistent with the field behavior.