Normalizing the influence of flaw length on failure pressure of straight pipe with wall-thinning
Introduction
Structural integrity of wall-thinned pipes in nuclear components has continuously been an issue of interest to engineers. However, in the boiler and pressure vessel code section XI of the American Society of Mechanical Engineers (ASME) (ASME, 2004), only a Code Case for evaluation of wall-thinned pipe is provided (ASME, 2003). The code mainly prescribes assessment rules for flawed components containing crack-like flaws. The British Standard (BS) also provides an assessment method for general corrosion in pipes and pipelines (BS, 2005), though it is still an Annex. In order to enhance the rules for non-crack-like flaws, the failure strength of flawed components has to be clarified. Especially, strength against internal pressure is important to prevent catastrophic accidents due to burst of flawed pipe (NRC, 2006).
A number of experimental studies have been conducted to assess the integrity of pipes containing wall-thinning under internal pressure. Most of these studies focused on pipe line steel using full-scale specimens containing artificial wall-thinning in addition to corroded pipe taken from actual lines (Vieth and Kiefner, 1993). Empirical and semi-empirical failure criteria were developed based on the results for fitness-for-service (Turbak and Sims, 1994), and some modifications were made in order to reduce conservativeness included in the criteria (Vieth and Kiefner, 1989). Meanwhile, theoretical and numerical approaches were made based on elastic shell theory (Folias, 1965, Kanninen et al., 1992) and finite element analysis (FEA), respectively. Because of limitations in the number of experiments and experimental conditions, numerical analyses are inevitable to complement experimental results. Especially, the elastic-plastic FEA based on shell modeling (Sims et al., 1992, Stephens and Leis, 1997) and three-dimensional solid modeling (Roy et al., 1997, Chouchaoui et al., 1992, Netto et al., 2005, Oh et al., 2006) seemed give good agreement with experimental results. In the FEA, several analysis procedures (criteria) were used as follows.
Limit load (LL) analysis is widely used to assess the strength of components. In LL analysis, an elastic-perfect-plastic stress–strain curve is usually used together with the twice-elastic slope criterion. This approach is easy to perform due to the simple material properties and analysis procedure, and has been used in studies of wall-thinning problems under internal pressure (Kim et al., 2002a, Kim et al., 2004, Choi et al., 2005). Large deformation analysis using work-hardening stress–strain curves is also a practical procedure. By using an actual stress–strain relation of the material, deformation of a pipe due to applied internal pressure can be simulated accurately. The problem with this approach is definition of the failure pressure of the pipe, which corresponds to the burst pressure in experiments. Since, in general, it is difficult to obtain the stress–strain relation above the ultimate strength of material, deformation above the ultimate strength is difficult to simulate. Shim et al. (2004) defined failure pressure as the pressure when stress reached ultimate strength at the deepest portion of wall-thinning. Roy et al. (1997) assumed a stress–strain curve above the ultimate strength. Miyazaki et al. (2002) also extrapolated the stress–strain curve and defined failure by considering the multi-axial stress condition, although they treated failure due to bending load. Thus, there are various approaches to the evaluation of failure strength of wall-thinned pipe.
The objective of the current study is to quantify the influence of length of wall-thinning on burst pressure. At the same time, the validity of various FEA procedures is discussed with comparisons with experimental results. The experiments were carried out using pipes of carbon steel for high-temperature use containing artificial wall-thinning of different lengths. In these experiments, in order to compare the burst pressure with that obtained by FEA, the precise thickness of pipe at wall-thinned portion was identified. Three-dimensional elastic-plastic FEA was then performed for wall-thinned pipes. Several analysis procedures were applied and their results were compared with those of experiments. Based on the FEA results, the influences of flaw shape on failure pressure were examined, and discussions were made for how to prescribe the influence of flaw length for assessments of failure pressure of straight pipe with wall-thinning.
Section snippets
Experimental procedure
Six experiments were carried out using pipes made of carbon steel for high-temperature use (STPT370 in JIS) with artificially introduced wall-thinning (hereafter, flaw). The chemical compositions and mechanical properties of the material are shown in Table 1, Table 2, respectively. The geometry of tested pipe is depicted in Fig. 1.
The nominal outer diameter and thickness of the pipes were Do = 107.1 mm and t = 4 mm, respectively. In order to obtain uniform dimensions of diameter and thickness, the
Experimental results and discussions
The burst pressures obtained by the tests were Pf = 17.49 MPa (S1A) and 17.46 MPa (S1B) in the case of So = 72.5 mm, 18.17 MPa (S2A) and 19.50 MPa (S2B) in the case of So = 50 mm and 23.90 MPa (S3A) and 24.16 MPa (S3B) in the case of So = 25 mm, respectively. The burst pressure decreased as flaw length increased. It was also influenced by flaw depth. In the cases of So = 50 and 25 mm, burst pressure was lower in thinner pipe at z = 0 position, although it was almost the same in the case of So = 72.5 mm, in which the
Procedure of finite element analysis
The three-dimensional elastic-plastic FEA was performed using the general purpose program ABAQUS, Version 6.5 (ABAQUS Inc., 2005). The geometry of a pipe with wall-thinning (flaw) is schematically shown in Fig. 5. The half-length of the flaw is expressed by So with an edge length of Se. Pipe length was set to L = 5Do, which is enough to ignore the boundary effect at the end of the specimen. The pipe was modeled by 8-node solid elements as shown in Fig. 6. At the bottom of the flaw, the model was
Prediction of experimental results
The FEA results simulating those of the experiments are shown in Table 4 together with the experimental results. The failure pressures obtained by LL analyses give conservative results, except those obtained using σb as the yield strength. The evaluated failure pressures are less than 30% of those obtained by the experiments in some cases. With use of σb, results are less conservative, although, in general, σb is not used as the yield strength in LL analyses. The safety margin changes with the
Conclusions
Burst tests using wall-thinned pipe of carbon steel for high-temperature use were conducted in order to examine the influence of the length of wall-thinning on failure pressure. Then, three-dimensional elastic-plastic FEA was performed to predict the burst pressure in the tests and investigate the influence of flaw shape on failure pressure. The following conclusions were obtained:
- (a)
The burst pressure of wall-thinned pipe in tests decreased as flaw length increased, and was accurately predicted
References (33)
- et al.
Ductile fracture theories for pressurized pipes and containers
Int. J. Pressure Vessel Piping
(1976) - et al.
Plastic limit pressures for cracked pipes using finite element limit analyses
Int. J. Pressure Vessels Piping
(2002) - et al.
The effect of corrosion defects on the burst pressure of pipelines
J. Construct. Steel Res.
(2005) ABAQUS/Standard User's Manual Ver. 6.5
(2005)- ANSI/ASME, 1984. Manual for Determining the Remaining Strength of Corroded Pipelines, ANSI/ASME...
Fitness-For-Service API579
(2000)- ASME, 2003. Requirements for Analytical Evaluation of Pipe Wall Thinning (Cases of ASME Boiler and Pressure Vessel Code...
- ASME, 2004. Boiler and Pressure Vessel Code Section XI, ASME, New York,...
- BS, 2005. Guide to Methods for Assessing the Acceptability of Flaws in Metallic Structures BS7910: 2005,...
- et al.
Limit load analyses for wall-thinned elbow with different bend angles subjected to internal pressure
Burst pressure predictions of line pipe containing single corrosion pits using the finite element method
OMAE
An axial crack in a pressurized cylindrical shell
Int. J. Fract. Mech.
Criteria for crack extension in cylindrical pressure vessel
Int. J. Fract. Mech.
A theoretical analysis for the residual strength of corroded gas and oil transmission pipelines
Corrosion Forms and Control for Infrastructure, ASTM Special Technical Publication
Cited by (16)
Efficient elastic stress analysis method for piping system with wall-thinning and reinforcement
2022, Nuclear Engineering and TechnologyCitation Excerpt :When it is found during service, the continued operation can be checked using some certified codes and standards, for instance, using ASME Code Case N-597-2 [5] for nuclear piping, or using more advanced analysis results. Various works have been done to develop acceptance criteria for wall-thinned pipes; for instance, the effect of wall-thinning on pipe failure pressures [6–10] and bending strength [11–13]. In case when acceptance criteria cannot be met, the defective pipe should be either replaced or repaired.
The effect of load-controlled bending load on the failure pressure of wall-thinned pipe elbows
2013, Nuclear Engineering and DesignCitation Excerpt :Integrity evaluation procedures were also proposed. Recently, several studies have focused on the assessment of the failure pressure of nuclear piping components with local wall thinning under internal pressure (Kamaya et al., 2008; Kim et al., 2009b; Meshii, 2010; Tsuji and Meshii, 2013). Kamaya et al. (2008) and Tsuji and Meshii (2013) conducted burst tests on 50A–200A carbon steel pipe specimens with simulated wall-thinning defects and investigated the influences of the wall-thinning geometry on the failure pressure and failure mode of straight pipes.
Proposal of limit moment equation applicable to planar/non-planar flaw in wall thinned pipes under bending
2011, Nuclear Engineering and DesignCitation Excerpt :Currently, much research is being conducted to establish and improve the safety assessment of pipes with local wall thinning (Janelle, 2005; Wilkowski et al., 2000). For pipes with local wall thinning that are both under pressure (Ahn et al., 2002; API/ASME, 2007; Kamaya et al., 2008a,b; Kim and Park, 2003a,b; Hasegawa et al., 2011) as well as a bending moment (Han et al., 1999; Kim and Park, 2003a,b; Zheng et al., 2004), the fracture behavior (Ahn et al., 2002; Miyazaki et al., 1999; Takahashi et al., 2007; Tsuji and Meshii, 2011) and a safety assessment have been established and modified by the finite element method and through experiment. For the case of evaluating the limit moment of wall-thinned straight pipes for a circumferential crack inside a cylinder (Fig. 1, top), a limit moment equation such as Kanninen's equation (Kanninen et al., 1982) is widely used (ANSI/ASME B31.G., 1991).
Safety assessment of pipes with multiple local wall thinning defects under pressure and bending moment
2011, Nuclear Engineering and DesignCitation Excerpt :Currently, dozens of research work has been well done to establish and improve the safety assessment of pipes with local wall thinning (Janelle, 2005; Wilkowski et al., 2000). For pipes with single local wall thinning, both under pressure (Ahn et al., 2002; API/ASME, 2007; Kamaya et al., 2008a,b; Kim and Park, 2003a,b) and bending moment (Han et al., 1999; Kim and Park, 2003a,b; Miyazaki et al., 1999, 2002; Shen and Zheng, 2004; Shim et al., 2004; Zheng et al., 2004), as well as the fracture behavior, the safety assessment has been established and modified by finite element method and experiment. For pipes with multiple local wall thinning defects, however, the safety assessment has not been discussed thoroughly.
A procedure for determining the true stress-strain curve over a large range of strains using digital image correlation and finite element analysis
2011, Mechanics of MaterialsCitation Excerpt :To assess ductile crack growth, the stress–strain curve of the material is used for evaluating the driving force of ductile crack propagation (J-integral value) by finite element analysis (FEA). The stress–strain curve is also necessary for evaluating the change in the load-carrying capacity of pipes with wall-thinning, which is a major degradation issue in nuclear power plants (NRC, 2006; Kamaya, 2008). Although the stress–strain curve is usually obtained by a tensile test using cylindrical specimens, it is difficult to obtain the curve for a large range of strains exceeding the strain for the tensile strength (hereafter, post-necking strain).