See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/281768369

Assessment of corrosion damage acceptance criteria in API579-ASME/1 code

Article  in  International Journal of Mechanics and Materials in Design · September 2014

DOI: 10.1007/s10999-014-9278-6

CITATION READS

1 828

4 authors, including:

Mohammad S. Attia M. Ammar Darwish

GE Global Research University of Atlanta
21 PUBLICATIONS   153 CITATIONS    182 PUBLICATIONS   1,737 CITATIONS   

SEE PROFILE SEE PROFILE

S.Suriya Prakash Sundram

2 PUBLICATIONS   2 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Development, Design, and Analysis of Wind Turbine Jointed Blades View project

All content following this page was uploaded by Mohammad S. Attia on 22 September 2015.

The user has requested enhancement of the downloaded file.

Int J Mech Mater Des
DOI 10.1007/s10999-014-9278-6

Assessment of corrosion damage acceptance criteria

in API579-ASME/1 code
M. S. Attia • M. M. Megahed • M.
Ammar Darwish • S. Sundram

Received: 17 August 2014 / Accepted: 15 September 2014

Ó Springer Science+Business Media Dordrecht 2014

Abstract This study presents a comparative evalu- continued operation at the original design pressure for
ation of the various acceptance criteria in API579- shorter period against a pressure de-rating procedure
ASME/1 fitness-for-service (FFS) code for equipment for extended operation.
suffering from metal loss, using a case of severe
corrosion in a pipe subjected to internal pressure and Keywords Fitness-for-service  Corrosion 
supplemental loads. All three assessment levels were Remaining life assessment  Limit analysis 
conducted to evaluate the remaining life of the pipe Finite element analysis
under various acceptance criteria assuming constant
corrosion rate. In particular, Level-3 assessment was
List of symbols
performed using three-dimensional parametric limit
D Inside shell diameter corrected for LOSS or
finite element analysis to evaluate the remaining life.
FCA
The model accounts for the observed metal loss, with
Do Outside diameter
the aim of evaluating the plastic collapse pressure of
Pd Design pressure
the corroded pipe. The remaining life was estimated
Pc Current operational pressure
using both global and local failure criteria and the
PL Limit pressure of the structure
conservatism of various methods and criteria is
tc Net remaining thickness excluding FCA
assessed. Results showed a potential trade-off of
away from metal loss region
tmin Minimum required wall thickness of the
component
M. S. Attia (&)  M. M. Megahed trd Uniform measured thickness away from the
Faculty of Engineering, Cairo University, Giza 12613, local metal loss location
Egypt tsl Thickness required to withstand loads other
e-mail: m.attia@utoronto.ca
than internal pressure
M. M. Megahed tlim Limiting thickness, for piping components:
e-mail: mmegahed@eng.cu.edu.eg
tlim = max (0.2tnom, 2.5mm)
M. A. Darwish  S. Sundram tnom Nominal pipe thickness corrected for mill
Zakum Development Company, under-tolerance
P.O. Box 46808, Abu-Dhabi, UAE Ls Recommended thickness measurements
e-mail: mdarwish@zadco.ae
inspection grid size per API-579/ASME-1
S. Sundram FCA Future corrosion allowance
e-mail: ssundram@zadco.ae

123
 M. S. Attia et al.

MAWP Maximum allowable working pressure evaluate the burst pressure for thin piping with metal
RSF Remaining strength factor loss defects. Furthermore, they outline the role of
assumed flow stress on the assessment results. Sim-
plified assessment rules commonly employ limit load
1 Introduction analysis to evaluate the burst pressure assuming elastic
perfectly-plastic material models BS 7910 2005. In
The concept of Fitness-For-Service (FFS) pertains to contrast, plastic collapse load analysis using nonlinear
the development of quantitative tools to evaluate the stress–strain material data and advanced numerical
ability of existing equipment that suffered one or more techniques such as FEA yields more realistic and less
forms of defects and/or damage to remain in service. conservative results (Choi et al. 2003; Chiodo and
The earliest FFS standard appeared in the mid 1970s in Ruggieri 2009; Netto et al. 2005; Stephens and Leis
the UK to evaluate the effect of welding defects in 1997; Roy et al. 1997; Kamaya et al. 2008; Khyabani
nuclear pressure vessels (R6–Rev.4, 2001). It has then and Sadrnejad 2009). This can be explained due to the
spread to refinery industry equipment; in particular it non-zero post-yield stiffness, which allows an entirely
has attracted an increased attention over the past two plastic region to sustain some of the increase in post-
decades (Anderson and Osage 2000) in terms of defect yield load. Similar to limit load evaluation technique,
types, coverage and analysis fidelity. Currently, there the plastic collapse load is evaluated as the highest
exist a number of FFS standards in various interna- convergent load increment in a well-configured finite
tional jurisdictions (API 579-1/ASME FFS-1 2007; element analysis with sufficiently small load
FITNET 2008; BS 7910; SINTAP 1999; ASME increments.
2012). The decision of an FFS assessment is to either
keep equipment in-service under original design
conditions, re-rating for a specified remaining life, or
to be completely retired. In situations where the 2 Problem statement
equipment is kept in-service, FFS tools provide means
to evaluate its remaining life under original or re-rated The objective of this study is to conduct a comparative
operational conditions. In particular, API-579/ASME- evaluation of various acceptance criteria in API-579/
1 FFS standard has been first released as an API ASME-1 FFS standard (American Society of Mechan-
recommended practice (RP) document in 2000, and ical Engineers 2007) for equipment suffering metal
was later released as a standard under joint collabo- loss. This is demonstrated through a case study of a
ration between API and ASME in 2007 (American 4-inch, Sch. 80 seamless pipe spool suffering from
Society of Mechanical Engineers 2007). It provides a metal loss. Figure 1 illustrates a schematic of the
comprehensive coverage of various forms damage problem under consideration. The spool is subject to
such as crack defects, dents, gouges, general and local mechanical loading due to internal pressure P, fluid
metal loss, among others. The standard provides three and valve attachments weight W, simultaneously. The
assessment levels for each defect type, with reduced spool is attached to a vertical 1200 pipe run with a
conservatism and increased analysis fidelity and reinforcement plate. Evidently, the 1200 pipe is much
inspection requirements as the analysis level is stiffer than the spool under consideration, which
increased from Level-1 to 3. makes it possible to assume fixed end conditions at
A survey of literature showed numerous studies on their junction. Table 1 summarizes the main design
the FFS and acceptance criteria of piping suffering data of the pipe leg. Based on periodic wall thickness
metal loss. A comprehensive overview of existing measurements, an annual corrosion rate of 1.2 mm has
assessment methods and comparison of various been identified. This severe rate prompted an FFS
acceptance criteria is presented in (Turbak and Sims assessment of the pipe spool to evaluate its condition
1994; Stephens et al. 1997; Cosham et al. 2007; and remaining life.
Brighenti 2001). These studies highlight the use of Figure 2 summarizes the inspection results along a
plastic instability (Turbak and Sims 1994; Stephens 150 mm distance between stations S-1 and S-2 using
et al. 1997; Cosham et al. 2007) and buckling pulse-echo ultrasonic (UT) transducer. The inspection
(Brighenti 2001) as governing failure criteria to grid size was 10 mm in the length direction and

123
Assessment of corrosion damage

Fig. 1 Schematic of the

corroded 4 inch pipe leg

Table 1 Design data of the corroded pipe leg 3 FFS assessment procedure
Pipe dimensions NPS 4–Sch. 80
According to the rules of API-579/ASME-1, the
Fluid crude oil
current problem can be studied within the framework
Design pressure 90 barg.
of either Part-4 (General metal loss) or Part-5 (Local
Current operating pressure 15 barg.
metal loss). The selection of a specific procedure
Design temperature -20–400 °F
depends on the resolution of available inspection data.
Operating temperature 120 °F
The recommended grid size Ls per code requirements
Material API 5L Grade B
is
Yield strength 207 MPa h pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffii
Insulation None Ls ¼ min 0:36 Dtmin ; 2trd ð1Þ
Thermal stresses Negligible
which gives Ls & 7 mm for the current pipe geom-
etry. However, the adopted grid size in the UT
inspection is much coarser as can be seen from
46 mm (equivalent to 45°) in the circumferential
Fig. 2. Therefore, it is not possible to evaluate the
direction, covering the lower half of the pipe circum-
nature of metal loss, whether global or local. Hence,
ference surrounding the metal loss region. The UT
resort was made to perform FFS assessment using the
measurements indicated that the metal loss was
rules of Part-4 since it provides a more conservative
confined to the region between 135°–225°, hence
assessment, even if further more detailed inspection
Fig. 2 shows the thickness readings within this range.
proves that the metal loss to be of localized nature.
The greatest metal loss occurred at 180° (6 o’clock)
Accordingly, FFS assessment will be performed in the
with some limited spread toward 3 o’clock and 9
following sequence:
o’clock. The minimum remaining thickness at 180°
was 3 mm (metal loss of 5.56 mm), as the nominal 1. Conduct Level-1 assessment first neglecting the
thickness is 8.56 mm. At 135°, 225° minimum effect of sustained mechanical loads other than
remaining thicknesses were 3.5 mm and 4.8 mm internal pressure,
respectively. The following sections provide a detailed 2. Conduct piping stress analysis to estimate the
rationale for FFS assessment, followed by analysis supplemental wall thickness required to resist
results and discussion. mechanical loads,

123
 M. S. Attia et al.

Fig. 2 UT thickness
measurements in the metal
loss zone

3. Conduct Level-2 assessment taking into account membrane analysis according to the following equa-
the effect of sustained mechanical loads, tions (API: 579-1/ASME FFS-1, 2007)
4. Perform remaining life assessment (RLA) using PDo
C
Level-2 results based on current operating tmin ¼ ð2Þ
2ðSE þ 0:4PÞ
pressure,
5. Conduct Level-3 assessment using non-linear PDo
L L
finite elements and limit load analysis according tmin tmin ¼ þ tsl ð3Þ
4ðSE þ 0:4PÞ
to the procedures of Annex-B1 of ASME/API 579
(API579-1/ASME FFS-1, 2007). which gives tCmin and tLmin values of 3.76 and 1.88 mm,
respectively. Noteworthy, Level-1 assessment does
not take into account supplemental loads thickness, i.e.
tsl = 0; with the rationale behind this to simplify
3.1 Level-1 assessment analysis requirements. The next stage is to evaluate
longitudinal and circumferential critical thickness
The first step in Level-1 assessment is to evaluate the profiles (CTP) using the inspection data summarized
L
minimum thickness values in longitudinal ‘‘tmin ’’ and in Fig. 2. These profiles trace the minimum measured
C thickness in both directions. These profiles are then
circumferential ‘‘tmin ’’ directions using thin-wall

123
Assessment of corrosion damage

used to evaluate average measured thicknesses tSam and section, a complete Level-2 assessment will be
tCam based on longitudinal and circumferential CTPs. conducted, including the effects of supplemental
Thickness averaging is performed over characteristic loads.
length L, given as:
8 vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
#ffi
> u " 2 3.2 Level-2 assessment
>
< u 1  R
1:123tDtc
t
 1 Rt RSFa
L¼ 1  Rt =RSFa The transition from Level-1 to Level-2 FFS assess-
>
> pffiffiffiffiffiffiffi
: ment necessitates higher analysis fidelity and usually
50 Dtc Rt [RSFa
provides a less conservative assessment results and
ð4Þ
extended component life. In particular, it is necessary
where Rt is the net thickness ratio between the metal to account for mechanical loads other than internal
loss area to the nominal thickness, and RSFa is the pressure. The associated reduced conservatism is
allowable remaining strength factor defined as the mainly due to accounting for the RSFa, thereby
ratio between limit or plastic collapse loads of obtaining credit for the nonlinear deformation of the
damaged to undamaged component. It aims at reduc- component. In the current case, these include the
ing the analysis conservatism by taking into account weight of the filled corroded pipe, and the weight of
the nonlinear material behaviour rather than the elastic attached valves at the tip of the cantilevered pipe
regime only. The current recommendation for RSFa spool. Figure 3 depicts a schematic of the piping leg
for piping components in API-579/ASME-1 is 0.9 showing the applied loads at their respective center of
(API579-1/ASME FFS-1, 2007). Hence, the values of gravity. The piping leg is rigidly attached to a 12’’ line
tSam and tCam can be evaluated for various FCA levels to at its far end, with a reinforcement plate at the junction
evaluate the acceptability conditions and remaining location.
life of the component. In order to estimate the stresses in the vicinity of the
Table 2 summarizes the results of Level-1 accord- corroded area resulting from the sustained mechanical
ing to various acceptability criteria in (API579-1/ loads, piping stress analysis criteria in ASME B31.3
ASME FFS-1, 2007) for FCA between 0.4–2.8 mm, (2004) are used. The essential design requirement in
which is equivalent to 4–28 months of operation at a this regard is that the sum of longitudinal stresses
corrosion rate of 1.2 mm/year. The acceptability arising from both pressure and mechanical loads
conditions are shown for each case as True ‘‘T’’ or should not exceed the allowable stress S, i.e.:
False ‘‘F’’. According to Level-1 requirements, the
rPa þ rsla  S ð5Þ
corroded pipe should not be operated at the original
design pressure Pd = 90 bar. However, it is permis- where rPa and rsla are the axial stresses due to pressure
sible to continue operation at a de-rated pressure and supplemental mechanical loads, respectively.
Pc = 15 bar for a period of 24 months. It is worth These stresses can be evaluated using either mechan-
mentioning that Level-1 assessment does not account ics of materials approach using Fig. 3, or through
for supplemental loads requirements. In the following finite element analysis. In order to quantify the effect

Table 2 Level-1 assessment results at various FCA

FCA (mm) 0.4 0.8 1.2 1.6 2.0 2.4 2.8
Months of operation 4 8 12 16 20 24 28
Acceptability of average measured thickness
Condition-1: tSam-FCA [ tCmin F F F F F F F
Condition-2: tCam-FCA [ tLmin T T F F F F F
Acceptability of MAWP
Condition-3/1: MAWPr [ Pd F F F F F F F
Condition-3/2: MAWPr [ Pc T T T T T T F
Acceptability of minimum measured thickness
Condition-4: tmm-FCA B tlim T F F F F F F

123
 M. S. Attia et al.

Fig. 3 Schematic of the

Piping Leg and Acting
Sustained Loads

Table 3 Level-2 assessment results at various FCA

FCA (mm) 0.4 0.8 1.2 1.6 2 2.4 2.8
Months of operation 4 8 12 16 20 24 28
Acceptability of average measured thickness
Condition-1: tsam-FCA [ RSFa 9 tcmin F F F F F F F
Condition-2: tcam-FCA [ RSFa 9 tLmin T F F F F F F
Acceptability of MAWP
Condition-3/1: MAWPr [ Pd F F F F F F F
Condition-3/2: MAWPr [ Pc T T T T F F F
Acceptability of minimum measured thickness
Condition-4: tmm-FCA B tlim T F F F F F F

of supplemental loads, the required thickness to resist RSFa effect. Furthermore, both Level-1 and Level-2
them tsl was evaluated equal to 0.76 mm, almost 30 % show that it is not possible to operate the piping
of the minimum required thickness of 2.54 mm. This component at the original design pressure Pd, while it
significant effect of supplemental loads was reflected is possible to operate at a reduced pressure Pc for 24
in Level-2 assessment results as will be shown later. and 16 months, respectively. This result is counter-
Table 3 summarizes Level-2 assessment results for intuitive since Level-2 is supposed to provide less
the corroded piping leg. Compared to Level-1 conservative results through RSFa concept. However,
acceptability criteria, the main difference is the the effect of incorporating supplemental loads on the
introduction of remaining strength factor (RSFa) in minimum required thickness actually caused Level-2
Level-2 assessment to reduce the assessment conser- thickness acceptability criteria to be more conserva-
vatism in acceptance conditions 1 through 3. It is tive than Level-1. The above conclusions are more
observed that Level-2 assessment consistently yielded accurate than those obtained from Level-1. Consid-
more conservative results than Level-1 in condi- eration of supplemental loads has proved to be useful
tions 1, 2, and 3, while it provided similar conserva- and has its subtle influence on the calculations of
tism in condition-4, which does not incorporate the Level-2 assessment.

123
Assessment of corrosion damage

Fig. 4 Finite element mesh

of the corroded pipe leg

3.3 Level-3 assessment The corroded metal area was modeled using a
perfectly square mesh with an element size of 5 mm
3.3.1 Finite element modeling covering the inspection area completely. In order to
incorporate the metal loss into the FE model, the
Level-3 FFS assessment requires detailed stress ultrasonic thickness measurement results summarized
analysis calculations to assess the acceptability of in Fig. 1 were mapped on the FE mesh and were linearly
the equipment and evaluate remaining life. Limit interpolated at respective element centroids. This ensures
load analysis of the corroded piping component was smooth thickness variation throughout the corroded area
performed using three-dimensional nonlinear finite and is more consistent with the general metal loss
element analysis (COSMOSM, 2004). The adopted assessment criteria adopted herein Carpenter et al.
meshis illustrated in Fig. 4. Thin-wall, full-integra- (2014). Figure 5 depicts the three-dimensional thickness
tion four-node shell elements with six degrees of profile in the corroded area as obtained from MATLAB
freedom per node were used to model the piping (2008).
under consideration. The loads acting on the piping
3.3.2 Evaluation of limit pressure
are identical to those adopted in Level-2 assessment
as detailed in Fig. 4, Hence, the corresponding
The piping under consideration is acted upon by
boundary conditions on the FE model consist of
sustained loads W and internal pressure P. Limit
internal pressure P on all pipes and axial traction ra
pressure evaluation involves first applying the sus-
along the circumference at the pipe ends to simulate
tained loads W to the system, followed by incremental
the axial membrane stresses due to P. The pipe and
increase of P until the material exhibits plastic
fluid weights were included in the effective density
instability. This is typically evidenced by uncontrolled
of the pipe wall material. Remote concentrated
increase in strain levels with virtually no increase in
forces W1 and W2 simulating the two valves
pressure. In order to verify the FE model, limit
attachments were applied to simulate the weights
analysis was performed in the uncorroded conditions
of these valves at their respective centers of gravity.
ignoring supplemental loads effect. The value of limit
Finally, the pipe was assumed fully constrained at
pressure PL in this case was evaluated using FEA as
the 1200 line junction. In accordance with code
37 MPa. The theoretical value of yield and limit
recommendations for limit load analysis in Annex
pressure for a thin-walled cylinder; without consider-
B1 (API579-1/ASME FFS-1, 2007), the material
ation of any sustained loads is given by:
behavior was modeled as an elastic-perfectly-plastic
material with small strain plasticity formulation. The 2 t
PL ¼ pffiffiffi Y ð6Þ
material yield strength Y = 207 MPa and the elastic 3R
modulus E = 208 GPa. The tangent modulus Et was which gives a theoretical PL = 38.5 MPa. This indi-
set equal to 0.01 % of the elastic modulus to avoid cates a good match between numerical and theoretical
numerical stability issue. estimates.

123
 M. S. Attia et al.

Fig. 5 Three-dimensional corroded thickness profile

Fig. 6 Evaluation of limit

pressure ‘‘PL’’ for both
corroded and as-new pipes

Figure 6 depicts the variation of effective stress onset of plastic instability amounted to 220 MPa,
with internal pressure in a corroded pipe with which indicates significant yielding was attained due
FCA = 0.4 mm. The limit pressure PL was evaluated to the limited extent of the plastic instability zone.
to correspond to the last stable solution time step Comparing the solver behaviour in this case with the
before divergence. Compared to the uncorroded uncorroded case, it was observed that the peak
condition, the corroded pipe with FCA = 0.4 mm effective stress in the uncorroded case barely reached
has PL & 18 MPa, thus indicating a RSFa of 49 %. 209 MPa before the solution diverged due to excessive
Figure 7a, b illustrates the effective stress distribution plastic strain. The reason for is that the instability in
in the corroded pipe at the onset of yield and plastic the uncorroded condition is global, while it was
instability, respectively. Evidently, yielding is local- localized in the corroded condition. Hence, the
ized to the location of metal loss area, while the rest of surrounding material was able to provide sufficient
the pipe remained largely elastic. It is interesting to stiffness to the component to prevent global instability
observe that the peak effective stress distribution at the (American Society of Mechanical Engineers 2007).

123
Assessment of corrosion damage

Fig. 7 Effective stress

distribution in the corroded
pipe leg: a at the onset of
yielding, b at the onset of
plastic instability

Table 4 Level-3 global and local instability assessment

results
FCA Global Local Decision
(mm) instability instability

Design pressure 0.4 T T Pass

Pd = 90 barg. 1.0 T F Fail
1.6 T F Fail
Reduced operating 0.4 T T Pass
pressure 1.0 T T Pass
Pc = 15 barg.
1.6 T T Pass

3.3.3 Level-3 acceptability criteria

Level-3 assessments for both global and local failure

under applied loads are summarized in Eq. 7. In this Fig. 8 Level-2 and Level-3 Remaining Life Estimates
equation ‘‘D’’ denotes the dead weight of the piping
and attachments, while ‘‘P’’ denotes the internal
The global instability criterion in Eq. 7(a) is derived
pressure. If either load combination causes global
based on the load combination (Pressure and dead
instability, the component is considered unsafe for
loads) required to cause incipient yielding in the
operation under the applied pressure.
structure. These design loads were originally evalu-
Global instability : 1:5ðP þ DÞ  RSFa ated based on stress allowable S = Y/1.5. Hence,
ð7Þ
Local instability : 1:7ðP þ DÞ  RSFa multiplying them by 1.5 would cause primary stresses

123
 M. S. Attia et al.

to reach yield stress level. In an elastic-perfectly- more conservative compared to Level-1, which con-
plastic material, this is the limit load of the structure. tradicts the code philosophy of reduced conservatism
The inclusion of RSFa accounts for the limit load with increased analysis fidelity. It has been shown that
knockdown due to present metal loss damage. The this is due to supplemental loads being ignored in
rationale of the local instability criterion is similar to Level-1 assessment. Level-3 assessment using plastic
the global one, with the exception of using a factor of instability analysis shows that the pipe passes global
1.7 instead of 1.5, hence allowing higher limit load instability failure criterion under the design pressure
levels to account for the local nature of instability, i.e. for a short period, which can be justified by the limited
the surrounding material to the instability zone will extent to metal loss in the pipe. However, the pipe
provide additional resistance to the spread of exces- failed the local instability criterion. In conclusion, this
sive plastic deformation. study reveals that careful consideration of the assump-
Table 4 summarizes Level-3 acceptability criteria tions and restrictions of each level are critical to the
for FCA = 0.4, 1.0 and 1.6 mm under both the design final assessment results.
and current, reduced operating pressures. No global
instability was observed for the FCA values under
consideration. This is attributed to the limited extent of
corrosion although the code treatment of the problem is References
global. Furthermore, the pipe did not exhibit local
Anderson, T.L., Osage, D.: API 579: A comprehensive fitness-
instability for FCA = 0.4 mm. However, a FCA value
for-service guide. Int. J. Press. Vessel Pip. 77, 953–963
of 1 mm resulted in the onset of plastic instability and (2000)
clear instability was observed for FCA 1.6 mm. Hence, API: 579-1/ASME FFS-1- Fitness-for Service. American
it is concluded that the piping component will be safe Society of Mechanical Engineers (2007)
Assessment of the Integrity of Structures Containing Defects,
for Level-3 assessment for FCA values between 0.4 and
R6 Rev. 4, British Energy Generation Ltd, 2001
1 mm under both global and local instability criteria. B31.3:2004: Process Piping—ASME Code for Pressure Piping
Finally, it is noteworthy to compare the remaining life (2004)
estimates of Levels 2 and 3. Figure 8 depicts Level-2 B31G: Manual for determining the remaining strength of cor-
roded pipelines, ASME (2012)
estimates using MAWP criterion, and Table 4 presents
Brighenti, R.: Influence of a central straight crack on the
Level-3 assessments results. While Level-2 estimated buckling behavior of thin plates under tension, compres-
that the component cannot withstand the design sion, or shear loading. Int. J. Mech. Mat. Des. 6(1), 73–87
pressure for more than 1 month, Level-3 estimated that (2001)
BS 7910: Guide on methods for assessing the acceptability of
the design pressure can be tolerated for more than
flaws in fusion welded structures, BSI (2005)
4 months. This ‘‘reduced conservatism’’ comes at the Carpenter, H.W., Reid, R.G., Paskaramoorthy, R.: Extension of
expense of increased analysis complexity and compu- the layer removal technique for the measurement of
tations and, indeed, is in accordance with the inherent residual stresses in layered anisotropic cylinders. Int.
J. Mech. Mat. Des. doi:10.1007/s10999-014-9245-2. (2014
ASME/API -579 code philosophy.
in-press)
Chiodo, M.S.G., Ruggieri, C.: Failure assessments of corroded
pipelines with axial defects using stress-based criteria:
4 Conclusions Numerical studies and verification analyses. Int. J. Press.
Vessel. Pip. 86, 164–176 (2009)
Choi, J.B., Goo, B.K., Kim, J.C., Kim, Y.J., Kim, W.S.:
This study presented a detailed FFS assessment of a Development of limit load solutions for corroded gas
4-inch pipe subjected to internal pressure and supple- pipelines. Int. J. Press. Vessel. Pip. 80, 121–128 (2003)
mental mechanical loads and suffers from metal loss, Cosham, A., Hopkins, P., Macdonald, K.A.: Best practice for the
assessment of defects in pipelines – Corrosion. Eng. Fail.
using the general metal loss guidelines in API-579/
Anal. 14, 1245–1265 (2007)
ASME-1 code. The pipe failed both Level-1 and COSMOSM 2.90 Theory Manual, Structural Research and
Level-2 acceptability conditions regarding average Analysis Corporation (2004)
measured thickness, design pressure and minimum FITNET Fitness-For-Service (FFS): Procedure - Revision MK8.
GKSS Research Centre, Geesthacht (2008)
measured thickness. It passed both levels with varying
Kamaya, M., Suzuki, T., Meshii, T.: Failure pressure of straight
remaining life estimates at a reduced operating pipe with wall thinning under internal pressure. Int.
pressure level. Furthermore, Level-2 assessment was J. Press. Vessel. Pip. 85, 628–634 (2008)

123
Assessment of corrosion damage

Khyabani, A., Sadrnejad, S.A.: Finite element evaluation of CT95-0024. http://www.eurofitnet.org/sintap_Procedure_

residual stresses in thick plates. Int. J. Mech. Mat. Des. version_1a.pdf (1999)
5(3), 253–261 (2009) Stephens, D.R., Leis, B.N.: Material and geometry factors
MATLAB: MATLAB Release 2008. The MathWorks, Inc, controlling the failure of corrosion defects in piping.
Natick (2008) Vessel. Pip. Div. 350, 3–11 (1997)
Netto, T.A., Ferraz, U.S., Estefen, S.F.: The effect of corrosion Stephens, D.R., Krishnaswamy, P., Mohan, R., Osage, D.A.,
defects on the burst pressure of pipelines. J. Construc. Steel Wilkowski, G.M.: A review of analysis methods and
Res. 61, 1185–1204 (2005) acceptance criteria for local thinned areas in piping and
Roy, S., Grigory, S., Smith, M., Kanninen, M.F., Anderson, M.: piping components. Vessel Pip. Div. 359, 55–65 (1997)
Numerical simulations of full-scale corroded pipe tests Turbak, T.A., Sims, J.R.: Comparison of local thin area assess-
with combined loading. J. Press. Vessel. Tech. 119, ment methodologies. Vessel Pip. Div. 288, 307–314 (1994)
457–466 (1997)
SINTAP, Structural Integrity Assessment Procedure for Euro-
pean Industry: SINTAP BRITE-EURAM Project BRPR-

123

View publication stats