RP-F101

RECOMMENDED PRACTICEDNV-RP-F101

CORRODED PIPELINES

OCTOBER 2004

Since issued in print (October 2004), this booklet has been amended, latest in October 2006. See the reference to Amendments and Corrections on the next page. DET NORSKE VERITAS

FOREWORDDET NORSKE VERITAS (DNV) is an autonomous and independent foundation with the objectives of safeguarding life, prop-erty and the environment, at sea and onshore. DNV undertakes classification, certification, and other verification and consultancyservices relating to quality of ships, offshore units and installations, and onshore industries worldwide, and carries out researchin relation to these functions.DNV Offshore Codes consist of a three level hierarchy of documents: Offshore Service Specifications. Provide principles and procedures of DNV classification, certification, verification and con-

sultancy services. Offshore Standards. Provide technical provisions and acceptance criteria for general use by the offshore industry as well as

the technical basis for DNV offshore services. Recommended Practices. Provide proven technology and sound engineering practice as well as guidance for the higher level

Offshore Service Specifications and Offshore Standards.DNV Offshore Codes are offered within the following areas:

A) Qualification, Quality and Safety MethodologyB) Materials TechnologyC) StructuresD) SystemsE) Special FacilitiesF) Pipelines and RisersG) Asset OperationH) Marine OperationsJ) Wind Turbines

Amendments and Corrections This document is valid until superseded by a new revision. Minor amendments and corrections will be published in a separatedocument normally updated twice per year (April and October). For a complete listing of the changes, see the Amendments and Corrections document located at: http://www.dnv.com/technologyservices/, Offshore Rules & Standards, Viewing Area.The electronic web-versions of the DNV Offshore Codes will be regularly updated to include these amendments and corrections.Comments may be sent by e-mail to [email protected] subscription orders or information about subscription terms, please use [email protected] information about DNV services, research and publications can be found at http://www.dnv.com, or can be obtained from DNV, Veritas-veien 1, NO-1322 Hvik, Norway; Tel +47 67 57 99 00, Fax +47 67 57 99 11.

Det Norske Veritas. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, including pho-tocopying and recording, without the prior written consent of Det Norske Veritas.

Computer Typesetting (FM+SGML) by Det Norske Veritas.Printed in Norway

If any person suffers loss or damage which is proved to have been caused by any negligent act or omission of Det Norske Veritas, then Det Norske Veritas shall pay compensation to such personfor his proved direct loss or damage. However, the compensation shall not exceed an amount equal to ten times the fee charged for the service in question, provided that the maximum compen-sation shall never exceed USD 2 million.In this provision "Det Norske Veritas" shall mean the Foundation Det Norske Veritas as well as all its subsidiaries, directors, officers, employees, agents and any other acting on behalf of DetNorske Veritas.

AcknowledgementsThis Recommended Practice is based upon a project guideline developed in a co-operation between BG Technology andDNV. The results from their respective Joint Industry Projects (JIP) have been merged and form the technical basis for thisRecommended Practice.

We would like to take this opportunity to thank the sponsoring companies / organisations for their financial and technicalcontributions (listed in alphabetical order):

BG plc BP Amoco Health and Safety Executive, UK Minerals Management Service (MMS) Norwegian Petroleum Directorate (NPD) PETROBRAS Phillips Petroleum Company Norway and Co-Ventures Saudi Arabian Oil Company Shell UK Exploration and Production, Shell Global Solutions, Shell International Oil Products B.V. Statoil Total Oil Marine plc

DNV is grateful for valuable co-operations and discussions with the individual personnel of these companies.

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 3CONTENTS

1. GENERAL .............................................................. 51.1 Introduction .............................................................51.2 Update year 2004 .....................................................51.3 BG plc and DNV research projects........................51.4 Application ...............................................................51.5 Structure of RP........................................................51.6 Applicable defects....................................................51.7 Applied loads............................................................61.8 Exclusions.................................................................71.9 Other failure modes.................................................71.10 Tiered approach and further assessment ..............71.11 Responsibility...........................................................71.12 Validation .................................................................71.13 Definitions ................................................................71.14 Symbols and abbreviations.....................................81.15 Units ..........................................................................9

2. METHODOLOGY................................................. 92.1 Capacity equation....................................................92.2 Sizing accuracy and uncertainties..........................92.3 Part A, calibrated safety factors ..........................102.4 Part B, allowable stress approach........................102.5 Onshore pipelines ..................................................102.6 Characteristic material properties.......................102.7 Pressure reference height and static head...........102.8 Probabilistic assessments ......................................11

3. CALIBRATED SAFETY FACTOR (PART A) 113.1 Introduction ...........................................................113.2 Reliability levels .....................................................113.3 Partial safety factors and fractile values .............113.4 Circumferential corrosion ....................................133.5 Usage factors for longitudinal stress....................133.6 System effect ..........................................................133.7 Supplementary material requirements ...............13

4. ASSESSMENT OF A SINGLE DEFECT (PART A)............................................................... 14

4.1 Requirements .........................................................144.2 Longitudinal corrosion defect, internal pressure

loading only ............................................................14

4.3 Longitudinal corrosion defect, internal pressure and superimposed longitudinal compressive stresses.................................................................... 14

4.4 Circumferential corrosion defects, internal pressure and superimposed longitudinal compressive stresses .............................................. 15

5. ASSESSMENT OF INTERACTING DEFECTS (PART A)........................................... 16

5.1 Requirements......................................................... 165.2 Allowable corroded pipe pressure estimate........ 16

6. ASSESSMENT OF COMPLEX SHAPED DEFECTS (PART A)........................................... 21

6.1 Requirements......................................................... 216.2 Allowable corroded pipe pressure estimate........ 21

7. ALLOWABLE STRESS APPROACH (PART B) .............................................................. 24

7.1 Introduction........................................................... 247.2 Total usage factor.................................................. 24

8. ASSESSMENT OF A SINGLE DEFECT (PART B) .............................................................. 24

8.1 Requirements......................................................... 248.2 Safe working pressure estimate - Internal

pressure only.......................................................... 248.3 Safe working pressure estimate - Internal

pressure and combined compressive loading ..... 24

9. ASSESSMENT OF INTERACTING DEFECTS (PART B) ........................................... 26

9.1 Requirements......................................................... 269.2 Safe working pressure estimate ........................... 26

10. ASSESSMENT OF A COMPLEX SHAPED DEFECT (PART B) ............................................. 28

10.1 Requirements......................................................... 2810.2 Safe working pressure estimate ........................... 28

11. REFERENCES..................................................... 30

APP. A EXAMPLES FOR PART A............................... 31

APP. B EXAMPLES FOR PART B............................... 35

APP. C DETAILED CALCULATION OF MEASUREMENT ACCURACIES .................. 41DET NORSKE VERITAS

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 4 see note on front coverDET NORSKE VERITAS

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 51. General1.1 IntroductionThis document provides recommended practice for assessingpipelines containing corrosion. Recommendations are givenfor assessing corrosion defects subjected to:

1) Internal pressure loading only.2) Internal pressure loading combined with longitudinal

compressive stresses.

This Recommended Practice (RP) document describes twoalternative approaches to the assessment of corrosion, and thedocument is divided into two parts. The main differencebetween the two approaches is in their safety philosophy:The first approach, given in Part A, includes calibrated safetyfactors taking into account the natural spread in material prop-erties and wall thickness and internal pressure variations.Uncertainties associated with the sizing of the defect and thespecification of the material properties are specifically consid-ered in determination of the allowable operating pressure. Thispart of the RP is also a supplement to DNV-OS-F101. Proba-bilistic calibrated equations (with partial safety factors) for thedetermination of the allowable operating pressure of a cor-roded pipeline are given. The second approach, given in Part B, is based on the ASD(Allowable Stress Design) format. The failure pressure (capac-ity) of the corrosion defect is calculated, and this failure pres-sure is multiplied by a single usage factor based on the originaldesign factor. Consideration of the uncertainties associatedwith the sizing of the corrosion defect is left to the judgementof the user.

1.2 Update year 2004The RP was first issued in 1999 and updated in 2004 (this ver-sion). The update is based on experience and feedback fromfour years of use. The update covers:

Sec.3 (previously Sec.2) concerning Part A safety factorshas been rewritten and a simplified approach for consider-ing the inspection accuracy is given

a new section describing the methodology and the simpli-fied capacity equation is included

recommended limitations for Charpy values are included a recommendation for probabilistic calculations is

included recommendations for temperature de-rating for SMYS

and SMTS are included.

The update includes the following few technical corrections ofwhich the user should be aware:

the calculation of fully correlated depth measurement forinteraction defects Part A (Step 9 in Sec.5.2 and Step 12 inSec.6) is modified, and is less strict (the 1999 version isconservative)

UTS in Part B is changed to SMTS and fu.

1.3 BG plc and DNV research projectsThis document is a result of co-operation between BG Tech-nology (part of BG plc) and DNV. The results from theirrespective joint industry projects have been merged, and formthe technical basis for this recommended practice (/3/, /4/ and

The BG technology project generated a database of more than70 burst tests on pipes containing machined corrosion defects(including single defects, interacting defects and complexshaped defects), and a database of linepipe material properties.In addition, a comprehensive database of 3D non-linear finiteelement analyses of pipes containing defects was produced.Criteria were developed for predicting the remaining strengthof corroded pipes containing single defects, interacting defectsand complex shaped defects.The DNV project generated a database of 12 burst tests onpipes containing machined corrosion defects, including theinfluence of superimposed axial and bending loads on the fail-ure pressure. A comprehensive database of 3D non-linearfinite element analyses of pipes containing defects was alsoproduced. Probabilistic methods were utilised for code calibra-tion and the determination of partial safety factors.

1.4 ApplicationThe methods provided in this document are intended to be usedon corrosion defects in carbon steel pipelines (not applicablefor other components) that have been designed to the DNVOffshore Standard DNV-OS-F101 Submarine Pipeline Sys-tems, /8/, /9/ or other recognised pipeline design code as e.g.ASME B31.4 /1/, ASME B31.8 /2/, BS8010 /5/, IGE/TD/1/10,ISO/DIS 13623 /11/, CSA Z662-94 /7/, provided that thesafety philosophy in the design code is not violated.When assessing corrosion, the effect of continued corrosiongrowth should be considered. If a corroded region is to be leftin service then measures should be taken to arrest further cor-rosion growth, or an appropriate inspection programme shouldbe adopted. The implications of continuing defect growth areoutside the scope of this document.This RP does not cover every situation that requires a fitness-for-purpose assessment and further methods may be required.

1.5 Structure of RPThe RP describes two alternative approaches. The firstapproach is given in Part A, which consists of Sec.3 throughSec.6. The second approach is given in Part B, which consistsof Sec.7 through Sec.10.A flow chart describing the assessment procedure (for bothPart A and Part B) is shown in Fig.1-1.Worked examples are given in Appendix A for the methodsdescribed in Part A and Appendix B for the methods describedin Part B.

1.6 Applicable defectsThe following types of corrosion defect can be assessed usingthis document:

Internal corrosion in the base material. External corrosion in the base material. Corrosion in seam welds. Corrosion in girth welds. Colonies of interacting corrosion defects. Metal loss due to grind repairs (provided that the grinding

leaves a defect with a smooth profile, and that the removalof the original defect has been verified using appropriateNDT methods).

When applying the methods to corrosion defects in seam weldsand girth welds, it should be demonstrated that there are no sig-nificant weld defects present that may interact with the corro-sion defect, that the weld is not undermatched, and that theDET NORSKE VERITAS

/16/). weld has an adequate toughness.

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 6 see note on front coverFigure 1-1Flowchart of the assessment procedure

1.7 Applied loadsInternal pressure, and axial and/or bending loads may influ-ence the failure of a corroded pipeline. The following combi-nations of loading/stresses and defects are covered by this RP:Internal pressure loading for:

Single defect. Interacting defects. Complex shaped derfects.

Internal pressure loading and combined with longitudinal com-pressive stresses for:

The compressive longitudinal stress can be due to axial loads,bending loads, temperature loads etc.The recommended practice given in this document is confinedto the effects of internal pressure and compressive longitudinalloading on longitudinal failure because the validation of theseeffects was addressed in the DNV and BG Technologyprojects. The behaviour of corrosion defects under combined internalpressure and bending loads, and/or tensile longitudinal loads,was outside the scope of the DNV and BG Technology projectsand, therefore, this loading combination has not been includedas part of the RP. Methods for assessing defects under com-bined internal pressure and bending loads, and/or tensile longi-tudinal loads, are recommended in other documents (e.g. /6/

Analyse all Corrosion Damage Sites as Isolated Single Defects

using Section 4 (or 8).

Check for Possible Interactions Between Sites using Section 5

(or 9).

Analyse Corrosion Sites as Colony of Interacting Defects


Are Allowable Corroded Pipe Pressures (Safe Working Pressures) Acceptable?

Are Allowable Corroded PipePressures (Safe WorkingPressures) Acceptable?

Are Defect Profiles Available?

Analyse Corrosion Sites as aComplex Shaped Defects using

Section 6 (or 10).

Best Estimate of AllowableCorroded Pipe Pressures (Safe

Working Pressure).

INTERACTION

NO INTERACTION

NO NO

YESYES YES

NO

Identify Type of Loading

Analyse all Corrosion Damage Sites as Isolated Single Defects


PRESSURE ONLY COMBINELOADING

START

Best Estimate of Allowable Corroded Pipe Pressure (Safe

Working Pressure). DET NORSKE VERITAS

Single defects. and /12/).

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 71.8 ExclusionsThe following are outside the scope of this document:

1) Materials other than carbon linepipe steel.2) Linepipe grades in excess of X80 1).3) Cyclic loading.4) Sharp defects (i.e. cracks) 2).5) Combined corrosion and cracking.6) Combined corrosion and mechanical damage.7) Metal loss defects attributable to mechanical damage (e.g.

gouges) 3).8) Fabrication defects in welds.9) Defect depths greater than 85% of the original wall thick-

ness (i.e. remaining ligament is less than 15% of the orig-inal wall thickness).

The assessment procedure is only applicable to linepipe steelsthat are expected to fail through plastic collapse. Modern pipe-line steel materials normally have sufficient toughness toexpect plastic collapse failure. Studies have recommendedCharpy V-notch value as lower bound for the material tough-ness for plastic collapse /18/ and /19/.The procedure is not recommended for applications wherefracture is likely to occur. These may include:

10) Materials with Charpy values less than 27 J (20 ftlbf) fullsize test (equivalient 2/3 scale is 18 J, 13 ftlbf). For theweld a minimum full size Charpy value of 30 J is recom-mended.

11) Any material that has been shown to have a transition tem-perature above the operating temperature.

12) Material of thickness greater than 12.7 mm (1/2"), unlessthe transition temperature is below the operating tempera-ture.

13) Defects in bond lines of flash welded (FW) pipe.14) Lap welded or furnace butt welded pipe.15) Semi-killed steels.

1.9 Other failure modesOther failure modes, such as buckling, wrinkling, fatigue andfracture, may need to be considered. These failure modes arenot addressed in this document, and other methods may beapplicable, ref. /6/, /12/ and /14/.

1.10 Tiered approach and further assessmentThe intent of this RP is to provide tiered procedures for theassessment of corroded pipe. The first tier level is the simpli-fied approach for single defect assessment, where total lengthand maximum depth of the defect and the material specifica-tion are used.If the defect is not found to be acceptable a more refinedassessment including the profile of the defect can be per-formed, provided that information of the profile is available. Furthermore, if the corrosion defects are still not found to beacceptable using the procedures given in this RP, the user hasthe option of considering an alternative course of action tomore accurately assess the remaining strength of the corrodedpipeline. This could include, but is not limited to, detailed

scale testing, and is outside the scope of this document. If analternative course is selected, the user should document thereliability of the results, and this can often be a very challeng-ing task.

1.11 ResponsibilityIt is the responsibility of the user to exercise independent pro-fessional judgement in application of this recommended prac-tice. This is particularly important with respect to thedetermination of defect size and associated sizing uncertain-ties.

1.12 ValidationThe methods given in this RP for assessing corrosion underonly internal pressure loading have been validated against 138full scale vessel tests, including both machined defects and realcorrosion defects. The range of test parameters is summarisedbelow:

(Shortest defect was l = 2.1 t)For nomenclature, see Sec.1.14.The method for assessing corrosion defects under internalpressure and compressive longitudinal loading has been vali-dated against seven full scale tests on 324 mm (12 inch) nom-inal diameter, 10.3 mm nominal wall thickness, Grade X52linepipe. The method for assessing fully circumferential corrosion underinternal pressure and compressive longitudinal loading hasbeen validated against three full scale tests on 324 mm nominaldiameter, 10.3 mm nominal wall thickness, Grade X52 line-pipe. The validation of this method is not as comprehensive asthe validation of the method for assessing a single longitudinalcorrosion defect subject to internal pressure loading only. Thepartial safety factors have not been derived from an explicitprobabilistic calibration.The validation of the methods described in this document forthe assessment of corrosion defects subject to internal pressureloading plus compressive longitudinal stress (see Sec.4.3 and4.4, is not as comprehensive as the validation of the methodsfor the assessment of corrosion defects subject to internal pres-sure loading alone. The acceptance equation has not been validated for defectsdimensions where the breadth (circumferential extent) of thedefect exceeding the length of the defect. The partial safetyfactors for combined loading have not been derived from anexplicit probabilistic calibration.

1.13 DefinitionsA Single Defect is one that does not interact with a neighbour-ing defect. The failure pressure of a single defect is independ-ent of other defects in the pipeline.An Interacting Defect is one that interacts with neighbouringdefects in an axial or circumferential direction. The failurepressure of an interacting defect is lower than it would be if theinteracting defect was a single defect, because of the interac-

1) The validation of the assessment methods comprised full scale tests on grades up to X65, and material tests on grades up to X80 (inclusive).

2) Cracking, including environmentally induced cracking such as SCC (stress corrosion cracking), is not considered here. Guidance on the assessment of crack-like corrosion defects is given in References 8, 9, 10.

3) Metal loss defects due to mechanical damage may contain a work hard-ened layer at their base and may also contain cracking.

Pipeline:Pipe Diameter, mm 219.1 (8") to 914.4 (36")Wall Thickness, mm 3.40 to 25.40D/t ratio 8.6 to 149.4Grade (API/5L) X42 to X65Defects:d/t 0 to 0.97l/(Dt)0.5 0.44 to 35c/t (circumferential) 0.01 to 22DET NORSKE VERITAS

finite element analysis, probabilistic assessments and/or full tion with neighbouring defects.

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 8 see note on front coverA Complex Shaped Defect is a defect that results from combin-ing colonies of interacting defects, or a single defect for whicha profile is available.

1.14 Symbols and abbreviations

A = Projected area of corrosion in the longitudinal plane through the wall thickness (mm2).

Ac = Projected area of corrosion in the circumferential plane through the wall thickness (mm2).

Ai,pit = Area of the ith idealised pit in a complex shaped defect (mm2).

Apatch = Area of an idealised patch in a complex shaped defect (mm2).

Ar = Circumferential area reduction factor.= 1-Ac/ Dt 1-(d/t)

D = Nominal outside diameter (mm).F = Total usage factor.

= F1F2F1 = Modelling factor.F2 = Operational usage factor.FX = External applied longitudinal force (N).H1 = Factor to account for compressive longitudinal

stresses.H2 = Factor to account for tensile longitudinal

stresses.MY = External applied bending moment (Nmm).N = Number of defects in a colony of interacting

defects.Pcomp = Failure pressure of the corroded pipe for a single

defect subject to internal pressure and compres-sive longitudinal stresses (N/mm2).

Pf = Failure pressure of the corroded pipe (N/mm2).= Failure pressure for jth depth increment in a

progressive depth analysis of a complex shaped defect (N/mm2).

Pnm = Failure pressure of combined adjacent defects n to m, formed from a colony of interacting defects (N/mm2).

Ppatch = Failure pressure of an idealised patch in a com-plex shaped defect (N/mm2).

Ppress = Failure pressure of the corroded pipe for a single defect subject to internal pressure only (N/mm2).

Psw = Safe working pressure of the corroded pipe (N/mm2).

Ptensile = Failure pressure of the corroded pipe for a single defect subject to internal pressure and tensile longitudinal stresses (N/mm2).

Ptotal = Failure pressure of a complex shaped defect when treated as a single defect (N/mm2).

Pi = Failure pressures of an individual defect forming part of a colony of interacting defects (N/mm2).

RP = Recommended PracticeQ = Length correction factor.Qi = Length correction factor of an individual defect

forming part of a colony of interacting defects.Qnm = Length correction factor for a defect combined

from adjacent defects n to m in a colony of inter-acting defects.

Qtotal = Length correction factor for the total longitudinal length of a complex shaped defect (mm).

SMTS = Specified minimum tensile strength (N/mm2).SMYS = Specified minimum yield stress (N/mm2).ULS = Ultimate Limit State

Pfj

Z = Circumferential angular spacing between projec-tion lines (degrees).

E[X] = Expected value of random variable X.StD[X] = Standard deviation of random variable X.CoV[X] = Coefficient of variation of random variable X.

= StD[X]/E[X](X)* = Characteristic value of X.XM = Model uncertainty factorc = Circumferential length of corroded region (mm).d = Depth of corroded region (mm).dave = Average depth of a complex shaped defect (mm).

= A/ltotal dei = The depth of the ith idealised pit in a pipe

with an effectively reduced wall thickness due to a complex corrosion profile (mm).

de,nm = Average depth of a defect combined from adja-cent pits n to m in a colony of interacting defects in the patch region of a complex corrosion pro-file (mm).

di = Depth of an individual defect forming part of a colony of interacting defects (mm). Average depth of ith idealised pit in a progressive depth analysis of a complex shaped defect (mm).

dj = The jth depth increment in a progressive depth analysis of a complex shaped defect (mm).

dnm = Average depth of a defect combined from adja-cent defects n to m in a colony of interacting defects (mm).

dpatch = Average depth of an idealised patch in a com-plex shaped defect (mm).

(d/t)meas = Measured (relative) defect depth(d/t)meas,acc = Maximum acceptable measured (relative) defect

depthfu = Tensile strength to be used in designfy = Yield strength to be used in designi = Isolated defect number in a colony of N interact-

ing defects.j = Increment number in a progressive depth analy-

sis of a complex shaped defect.l = Longitudinal length of corroded region (mm).li = Longitudinal length of an individual defect form-

ing part of a colony of interacting defects (mm). Longitudinal length of ith idealised pit in a progressive depth analysis of a complex shaped defect (mm).

lj = Longitudinal length increment in a progressive depth analysis of a complex shaped defect (mm).

lnm = Total longitudinal length of a defect combined from adjacent defects n to m in a colony of inter-acting defects, including the spacing between them (mm).

ltotal = Total longitudinal length of a complex shaped defect (mm).

pmao = Maximum allowable operating pressure (N/mm2).

pcap,patch = Capacity pressure of an idealised patch in a complex shaped defect (N/mm2).

pcorr = Allowable corroded pipe pressure of a single lon-gitudinal corrosion defect under internal pressure loading (N/mm2).

= Allowable corroded pressure for jth depth increment in a progressive depth analysis of a complex shaped defect (N/mm2).

pcorr,circ = Allowable corroded pipe pressure of a single cir-cumferential corrosion defect (N/mm2).

pcorr,comp = Allowable corroded pipe pressure of a single lon-gitudinal corrosion defect under internal pressure and superimposed longitudinal compressive

2

pcorrjDET NORSKE VERITAS

UTS = Ultimate Tensile Strength (N/mm2) stresses (N/mm ).

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 91.15 UnitsThe units adopted throughout this document are N and mm,unless otherwise specified.

2. Methodology2.1 Capacity equationThe expression of the burst capacity for a single longitudinallyoriented, rectangular shaped, corrosion defect was developedbased on a large number of FE analyses, and a series of full

each important parameter was investigated, while the accuracyof the analyses was verified by a large number of full-scaleburst tests. The equations used in the development of this RPand in the calibration are fairly complex. For practical use asimplified capacity equation is give below. For more detailssee /16/ and /17/.The simplified capacity equation of a single rectangularshaped defect is given as:

where

This capacity equation represents the mean (best) estimate ofthe capacity of a pipe with a rectangular shaped corrosion(metal loss) defect. This implies that on average the equationshould represent the capacity of the pipe but that some of thedefects will fail at a slightly lower pressure, and some at aslightly higher pressure, than predicted.Since the equation is simplified, some effects, and combina-tion of effects, are not represented in detail. This includes e.g.yield to tensile ratio, D/t ratio, and length and depth effect. Forexample it is known that the equation over-predicts the failurepressure (capacity) for medium long defect with high yield totensile ratio (high grade steel), and under-predict the failurepressure for low yield to tensile ratio (low grade steel).The accuracy of the capacity equation had to be known forestablishing the appropriate safety factors, and the above men-tioned effects were accounted for.The factor 1.05 in the capacity equation is determined fromcomparison with laboratory test results with rectangularshaped metal loss defects, see /17/.If the equation is used for irregular or parabolic defect shapes,and the maximum depth and lengths are used, the equation willin general underestimate the failure pressure, as the defect isnot as large as the rectangular shaped defect assumed in thecapacity equation. This will result in a conservative estimate ofthe failure pressure capacity for defects shapes other than rec-tangular.

Figure 2-1Illustration of irregular and rectangular defects

2.2 Sizing accuracy and uncertaintiesFor known defect size, pipe dimensions and material proper-ties, the capacity equation predicts the burst capacity with agood accuracy. However, these input parameters usuallyinclude a certain degree of uncertainty, and this should beaccounted for in calculating the acceptable operating pressureof the corroded pipeline.A high level of safety (reliability) is required for pipelines.This is obtained by using safety factors in combination withthe capacity equation.For example, in an assessment of a defect only the materialgrade (giving SMTS and SMYS) will usually be available. The

pi = Allowable corroded pipe pressures of individual defects forming a colony of interacting defects (N/mm2).

pnm = Allowable corroded pressure of combined adja-cent defects n to m, formed from a colony of interacting defects (N/mm2).

ppatch = Allowable corroded pipe pressure of an idealised patch in a complex shaped defect (N/mm2).

ptotal = Allowable corroded pipe pressure of a complex shaped defect when treated as a single defect (N/mm2).

r = Remaining ligament thickness (mm).s = Longitudinal spacing between adjacent defects

(mm).si = Longitudinal spacing between adjacent defects

forming part of a colony of interacting defects (mm).

t = Uncorroded, measured, pipe wall thickness, or tnom (mm).

te = Equivalent pipe wall thickness used in a progres-sive depth analysis of a complex shaped defect (mm).

d = Factor for defining a fractile value for the corro-sion depth.

= Circumferential angular spacing between adja-cent defects (degrees).

d = Partial safety factor for corrosion depth.m = Partial safety factor for longitudinal corrosion

model prediction.mc = Partial safety factor for circumferential corrosion

model prediction. = Partial safety factor for longitudinal stress for

circumferential corrosion. = Ratio of circumferential length of corroded

region to the nominal outside circumference of the pipe, (c/D).

A = Longitudinal stress due to external applied axial force, based on the nominal wall thickness (N/mm2).

B = Longitudinal stress due to external applied bend-ing moment, based on the nominal wall thickness (N/mm2).

L = Combined nominal longitudinal stress due to external applied loads (N/mm2).

u = Ultimate tensile strength (N/mm2).1 = Lower bound limit on external applied loads

(N/mm2).2 = Upper bound limit on external applied loads

(N/mm2). = Usage factor for longitudinal stress.

( )( )

=Q

tdtd

tDtP ucap )/(1

)/(1205.1

2

31.01

+=

DtLQ

Rectangular shaped metal loss defectDET NORSKE VERITAS

scale burst tests. By using finite element analyses the effect of actual material properties at the location of the defect will not

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 10 see note on front coverbe known. Furthermore, the defect sizing will be determinedwith some level of uncertainty. The defect can be shallower, ordeeper, than the measured value, as illustrated in Fig.2-2. Thisdepth uncertainty has to be considered in the assessment of theallowable pressure.

Figure 2-2Measured defect depth and sizing accuracy

2.3 Part A, calibrated safety factorsThe effect of the inspection accuracy, combined with the otheruncertainties described above, is accounted for in the calibra-tion of the safety factor. Although a single safety factor toaccount for these uncertainties would give simpler calcula-tions, several partial safety factors were introduced to giveresults with a consistent reliability level for the validity rangeof input parameters. If a single safety factor should cover thefull range of input parameters, this would give results with avarying reliability level depending on the input parameters. Ifthe safety factor should be selected such that the minimumrequired reliability level is satisfied in all cases, the code wouldbe undesirably conservative for some combinations of theinput parameters.Results of FE analyses and laboratory tests, together with sta-tistical data of material properties, pressure variations andselected levels of uncertainties in the defect sizing, form therequired basis for a reliability code calibration where appropri-ated safety factors were defined.The maximum allowable operating pressure for a pipeline witha corrosion defect is given by the acceptance equation with thesafety factors:

where

The safety factors are described in Sec.3.

2.4 Part B, allowable stress approachThe approach given in Part B is based on the ASD (AllowableStress Design) format. The failure pressure (capacity) of thepipeline with the corrosion defect is calculated, and multipliedby a safety factor to obtain a safe working pressure. Often theoriginal design factor is used as the safety factor.However, when assessing corrosion defects, due considerationshould be given to the measurement uncertainty of the defectdimensions and the pipeline geometry. In contrast to Part A,these uncertainties are not included in the Part B approach, andare left to the user to consider and account for in the assess-ment.

2.5 Onshore pipelinesDesign codes for onshore pipelines allow in general a lowerutilisation of the material compared to offshore codes, i.e. thesafety factors are higher. These factors probably implicitlycover other loads and degradation mechanisms than consid-

with the safety philosophy in the original design code. Part Bcould be more appropriate for onshore pipelines, where theuser have to account for these additional failures aspects.However, when using Part B it is recommended that the useralso check according to Part A. If this yields stricter results,considerations should be made.

2.6 Characteristic material propertiesThe specified minimum tensile strength (SMTS) is used in theacceptance equation. This is given in the linepipe steel mate-rial specification (e.g. API 5L , /15/) for each material grade.The characteristic material properties are to be used in theassessment of the metal loss defects. The material grades referto mechanical properties at room temperature, and possibletemperature effects on the material properties should also beconsidered.

where fy,temp and fu,temp = de-rating value of the yield stress and ten-sile strength due to temperature.The de-rating is highly material dependent and should prefer-ably be based on detailed knowledge of the actual material. Inlack of any material information the values in Fig.2-3 can beused for both yield stress and tensile strength for temperaturesabove 50C.

Figure 2-3Proposed de-rating values

2.7 Pressure reference height and static headThe assessment of corrosion defects should consider the pres-sure load at the location of the defect, both internal and exter-nal. If this effect is not included, conservative pressure loadsshould be used. The pressure reference height and the eleva-tion of the defect must be known.For offshore pipelines the benefit of external water pressurecan be utilised, and the increased pressure due to the internalstatic head has to be included.For onshore pipelines only the internal static head is to be

( )( )

=

Qtdtd

tDSMTStp

d

dmcorr *)/(1

*)/(12

]t/d[StD)t/d()*t/d( dmeas +=

Table 2-1 Characteristic material propertiesfy = SMYS fy,tempfu = SMTS fu,temp

De-rating yield stress and tensile strength

0

20

40

60

80

100

0 50 100 150 200

Temperature deg C

Stes

s de

-rat

ing

(MPa

)

CMnDET NORSKE VERITAS

ered in this RP, and if using Part A this could be in conflict included.

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 11The calculated pressures, e.g. pcorr in this RP refer to the localdifferential pressure load, and when determining pmao(MAOP) the internal and external static head should beincluded.

2.8 Probabilistic assessmentsThe safety factors in this RP are derived from probabilistic cal-ibrations, and based on a set of input parameter distributionsthat are considered to be representative.When more accurate knowledge of the distributions is known,or if further growth of the metal loss defects is to be included,probabilistic calculations can provide a strong tool for theassessment of metal loss defects.Probabilistic assessment is outside the scope of this RP, andthe rest of Sec.2.8 is given for information only.Probabilistic assessments of pipes with metal loss defects canbe based on the following limit state function:

g = Pcap - PINTwhere

Pcap = the burst pressure capacity, but where the 1.05 factor isreplaced by XM

PINT= the annual maximum differential pressure.

The parameters in the limit state should be modelled with theiractual distributions, and considerations should be given to theinspection sizing accuracy. A set of input parameter distribu-tions considered to be representative for pipelines were used inthe calibration of the safety factors included in DNV-RP-F101,and presented in Table 2-2. For details see ref. /16 and 17/.

The significance of each parameter varies, and some may beused as a fixed value, rather than a variable with associated dis-tribution. However, the distributions (uncertainties) in themodel and the sizing accuracy have to be included in a proba-bilistic assessment, where it is often seen that only one of theseare accounted for. The model uncertainty XM for the DNV-RP-F101 capacity equation is given in the Table 2-2, while theuncertainty in the sizing accuracy is given in Sec.3 of this RP.In addition to the inspection accuracy, the corrosion rate willalso add to the uncertainty of the future defect size.

3. Calibrated safety factor (Part A)3.1 IntroductionThe approach given in Part A includes calibrated safety fac-tors. Uncertainties associated with the sizing of the defectdepth and the material properties are specifically considered.Probabilistic calibrated equations for the determination of theallowable operating pressure of a corroded pipeline are given.

Factor Design) methodology.Partial safety factors are given for two general inspectionmethods (based on relative measurements e.g. magnetic fluxleakage, and based on absolute measurements e.g. ultrasonic),four different levels of inspection accuracy, and three differentreliability levels.

3.2 Reliability levelsPipeline design is normally to be based on Safety/LocationClass, Fluid Category and potential failure consequence foreach failure mode, and to be classified into safety classes.

Subsea oil and gas pipelines, where no frequent human activityis anticipated, will normally be classified as Safety Class Nor-mal. Safety Class High is used for risers and the parts of thepipeline close to platforms, or in areas with frequent humanactivity. Safety Class Low can be considered for e.g. waterinjection pipelines. For more details see ref /8/ and other rele-vant onshore and offshore pipeline codes.

3.3 Partial safety factors and fractile valuesThe partial safety factors are given as functions of the sizingaccuracy of the measured defect depth for inspections based onrelative depth measurements and for inspections based onabsolute depth. For inspections based on relative depth meas-urements the accuracy is normally quoted as a fraction of thewall thickness. For inspections based on absolute depth meas-urements the accuracy is normally quoted directly. An appro-priate sizing accuracy should be selected in consultation withthe inspection tool provider.The acceptance equation is based on two partial safety factorsand corresponding fractile levels for the characteristic values.

The safety factors are determined based on:

safety class (or equivalent), usually from design inspection method, relative or absolute inspection accuracy and confidence level.

Safety factor m is given in Table 3-2 for inspection resultsbased on relative depth measurements, (e.g. Magnetic FluxLeakage (MFL) intelligent pig measurements), and for abso-lute depth measurements (e.g. Ultrasonic Wall Thickness orWall Loss Measurements). MFL is a relative measurementwhere the defect depth measurement and the accuracy aregiven as a fraction of the wall thickness. The UT is an absolutemeasurement where the local wall thickness, the defect depthmeasurement and the accuracy are given directly.

Table 2-2 Parameters in the modelling of the burst limit stateVariable Distribution Mean UncertaintyPINT Gumbel 1.05 MAOP CoV= 3.0%D Deterministic Actual -t Normal Nominal CoV =3.0% u Normal 1.09 SMTS CoV = 3.0% and 6.0%Lmeas Normal Measured value Specifiedd/t Normal Measured value SpecifiedXM Normal 1.05 StD = 10%CoV is normalised standard deviation (CoV = StD/mean)

Table 3-1 Safety Class and target annual failure probability for Ultimate Limit State (ULS)

Safety Class Indicating a target annual failure probability of:High < 10-5Normal < 10-4Low < 10-3

m = Partial safety factor for model prediction.d = Partial safety factor for corrosion depth.d = Factor for defining a fractile value for the cor-

rosion depth.StD[d/t] = Standard deviation of the measured (d/t) ratio

(based on the specification of the tool).

Table 3-2 Partial safety factor mInspection method

Safety ClassLow Normal High

Relative (e.g. MFL) m = 0.79 m = 0.74 m = 0.70DET NORSKE VERITAS

These equations are based on the LRFD (Load and Resistance Absolute (e.g. UT) m = 0.82 m = 0.77 m = 0.72

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 12 see note on front coverThe factors for absolute measurement are higher since it isassumed that the pipe wall thickness around the corroded areais measured with at least the same accuracy as the corrosiondepth. The measured values of the wall thickness (t) should beused in the calculation of the allowable pressure.From the inspection accuracy and confidence level the stand-ard deviation in the sizing accuracy can be determined. Thestandard deviation is further used to determine the d safetyfactor and the d fractile value.The approach to calculate the standard deviation StD[d/t],where a Normal distribution is assumed, is:StD[d/t] for relative (e.g. MFL):The approach to calculate the standard deviation StD[d/t],where a Normal distribution is assumed, is:

StD[d/t] = acc_rel/NORMSINV(0.5 + conf/2)acc_rel = the relative depth accuracy, e.g. 0.2 (0.2 t)conf = the confidence level, e.g. 0.8 (80%)NORMSINV = a Microsoft Excel function. NORMSINV(x)

returns the inverse of the standard normalcumulative distribution at probability x.

The confidence level indicates the portion of the measure-ments that will fall within the given sizing accuracy. A selectedset of calculated standard deviations for relative sizing accu-racy is given in Table 3-3.

Fig.3-1 illustrates a sizing accuracy of 5% of t, quoted with aconfidence level of 80%. A Normal distribution is assumed.

Figure 3-1Example of a sizing accuracy of 5% of t, quoted with a confi-dence level of 80%

StD[d/t] for absolute (e.g. UT):

StD[d/t] = acc_abs/(t NORMSINV(0.5 + conf/2))acc_abs = the absolute depth accuracy, e.g. 0.5 (0.5

mm) conf = the confidence level, e.g. 0.8 (80%)

NORMSINV = a Microsoft Excel function.Note that the expression is dependent on the wall thickness.This function is a slightly conservative approximation of thedetailed expressions of the standard deviations, seeAppendix C, of absolute measurements used in the 1999 ver-sion of this RP. The detailed expressions may also be used. Thesimplification conservatively assumes d = t in the calculationof StD[d/t].A selected set of calculated standard deviations for absolutesizing accuracy is given in Table 3-4 through Table 3-6 for awall thickness of 6.35 mm, 12.7 mm and 19.05 mm.

Safety factor d and fractile value d:The d safety factor and the d fractile values are given inTable 3-7 for various levels of inspection accuracy (defined interms of the standard deviation) and Safety Class:

Polynomial equations can be used to determine the appropriatepartial safety factors and fractile values for intermediate valuesof StD[d/t] and are given in Table 3-8. The polynomial equa-tions are curve fits based on the calibrated factors given inTable 3-7. The curves are also shown in Fig.3-2 and Fig.3-3.In the determination of the partial safety factors it is assumedthat the standard deviation in the length measurement is lessthan 20 times the standard deviation in the depth measurement.

Table 3-3 Standard deviation and confidence levelRelative sizing accuracy

Confidence level80% (0.80) 90% (0.90)

Exact (0. 0 of t) StD[d/t] = 0.00 StD[d/t] = 0.00 0.05 of t StD[d/t] = 0.04 StD[d/t] = 0.03 0.10 of t StD[d/t] = 0.08 StD[d/t] = 0.06 0.20 of t StD[d/t] = 0.16 StD[d/t] = 0.12

[d/t][d/t] + 0.05[d/t]

10%10%

actual - 0.05 actual

actual

80%

2

Table 3-4 Standard deviation and confidence level, t = 6.35 mmAbsolute sizing accuracy


Exact (0 mm) StD[d/t] = 0.000 StD[d/t] = 0.000 0.25 mm StD[d/t] = 0.043 StD[d/t] = 0.034 0.5 mm StD[d/t] = 0.087 StD[d/t] = 0.068 1.0 mm StD[d/t] = 0.174 StD[d/t] = 0.135Table 3-5 Standard deviation and confidence level, t = 12.7 mmAbsolute sizing accuracy


Exact (0 mm) StD[d/t] = 0.000 StD[d/t] = 0.000 0.25 mm StD[d/t] = 0.022 StD[d/t] = 0.017 0.5 mm StD[d/t] = 0.043 StD[d/t] = 0.034 1.0 mm StD[d/t] = 0.087 StD[d/t] = 0.068Table 3-6 Standard deviation and confidence level, t =19.05 mmAbsolute sizing accuracy


Exact (0 mm) StD[d/t] = 0.000 StD[d/t] = 0.000 0.25 mm StD[d/t] = 0.014 StD[d/t] = 0.011 0.5 mm StD[d/t] = 0.029 StD[d/t] = 0.023 1.0 mm StD[d/t] = 0.058 StD[d/t] = 0.045

Table 3-7 Partial safety factor and fractile valueInspection sizing accuracy, StD[d/t] d Safety ClassLow Normal High

(exact) 0.00 0.0 d = 1.00 d = 1.00 d = 1.00 0.04 0.0 d = 1.16 d = 1.16 d = 1.16 0.08 1.0 d = 1.20 d = 1.28 d = 1.32 0.16 2.0 d = 1.20 d = 1.38 d = 1.58DET NORSKE VERITAS

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 13The variation of the partial safety factors d and d with StD[d/t] are shown in Fig.3-2 and Fig.3-3:

Figure 3-2Partial safety factor d with StD[d/t].

Figure 3-3Safety factor d with StD[d/t].

3.4 Circumferential corrosionPartial safety factors factor mc and are given in Table 3-9 fora single circumferential corrosion defect under internal pres-sure and longitudinal compressive stresses.

The calibration of the partial safety factors for a single circum-ferential corrosion defect under internal pressure and longitu-dinal compressive stresses did not consider the inspectionaccuracy.

3.5 Usage factors for longitudinal stressThe usage factors for longitudinal stress are given in Table 3-10

3.6 System effectThe target reliability levels are for a single metal loss defect.If the defect in question is clearly the most severe defect gov-erning the allowable corroded pipe pressure, then this defectwill also govern the reliability level of the pipeline for failuredue to corrosion. In the case of several corrosion defects eachwith approximately the same allowable corroded pipe pres-sure, or a pipeline with a large number of corrosion defects, thesystem effect must be accounted for when determining the reli-ability level of the pipeline. Adding the failure probability ofeach defect will conservatively assess the system effect.

3.7 Supplementary material requirementsThe safety factors in Table 3-11 and Table 3-12 can be used ifthe material requirements are documented with increased con-fidence for the yield and ultimate strength as given in e.g.DNV-OS-F101 additional material requirement U, or equiva-lent.The safety factors in Table 3-11 and Table 3-12 may only beused if it is explicitly documented that the supplementarymaterial requirements are fulfilled.

Table 3-8 Polynomial Equations for Partial Safety Factor and Fractile Value, see Table 3-7Substitute a with StD[d/t]Safety Class d and d Range

Low

Normal

High

(all)

ad 0.40.1 += 04.0

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 14 see note on front cover4. Assessment of a Single Defect (Part A)4.1 RequirementsIsolated metal loss defects are to be individually assessed assingle defects, see Fig.4-1. Adjacent defects can interact to produce a failure pressure thatis lower than the individual failure pressures of the isolateddefects treated as single defects. For the case where interactionoccurs, the single defect equation is no longer valid and theprocedure given in Sec.5 must be applied. Fig.4-2 shows thekey dimensions for defect interaction.A defect can be treated as an isolated defect, and interactionwith other defects need not be considered, if either of the fol-lowing conditions is satisfied:

1) The circumferential angular spacing between adjacentdefects, (degrees):

2) Or, the axial spacing between adjacent defects, s:

4.2 Longitudinal corrosion defect, internal pressure loading only

4.2.1 Acceptance equationThe allowable corroded pipe pressure of a single metal lossdefect subject to internal pressure loading is given by the fol-lowing acceptance equation. The acceptance equation has notbeen validated for defects dimensions where the breadth (cir-cumferential extent) of the defect exceeding the length of thedefect.

where:

If d (d/t)* 1 then pcorr = 0. pcorr is not allowed to exceed pmao. The static head and pres-sure reference height should be accounted for.Measured defects depths exceeding 85% of the wall thicknessis not accepted.

4.2.2 Alternative applicationsThe form of the acceptance equation is made to determine theacceptable operating pressure for a measured corrosion defectin a pipeline. The equation can be re-arranged to determine theacceptable measured defect size for a specified operationalpressure. By setting the specified operating pressure poper equal to pcorr,the equation can be re-arranged to calculate maximum accept-able measured defect depths:

where

(The limitations in the equation are not explicitly given)

4.2.3 Maximum acceptable defect depth The requirement d (d/t)* 1 then pcorr = 0 considers theconfidence in the sizing of the defect depth, and can also beexpressed as:

(The expression can also be determined from the above equa-tion where short defect is assumed and hence Q = 1.)The RP includes two requirements for maximum acceptabledefect depth:

a) Measured defect depth shall not exceed 85% of the wallthickness, i.e. minimum remaining wall thickness 15%of the (nominal) wall thickness.

b) The measured defect depth plus the uncertainty in thedefect sizing can not exceed the wall thickness, with thereliability level applicable for the defect, identified by thesafety or location class.

The maximum acceptable measured defect depths are depend-ent on the inspection method, sizing capabilities and safety orlocation class. Selected examples are given in Table 4-1.

If the wall thickness is close to the required minimum remain-ing wall thickness, special care should be given, e.g. for a10mm wall thickness pipeline the minimum requirement maybe only 1.5 mm. Special attention should be given to thesedefects, both in term of reliability of the inspection methodsand potential further growth.

4.3 Longitudinal corrosion defect, internal pressure and superimposed longitudinal compressive stressesThis method is only valid for single defects.The development of the method is outlined in ref. /17/.The allowable corroded pipe pressure of a single longitudinalcorrosion defect subject to internal pressure and longitudinalcompressive stresses can be estimated using the following pro-cedure:

Dt360>

Dts 2>

( )( )

=

Qtdtd

tDftp

d

dumcorr *)/(1

*)/(12

2

31.01

+=DtlQ

]/[StD)/()*/( tdtdtd dmeas +=

( )]/[StD

/1

/11)/(0

0, td

Qpppp

td doper

oper

daccmeas

=

Table 4-1 Maximum acceptable measured depth, selected examples

Safety Class

Inspection method Accuracy

Conf. level

Max acceptable measured depth

Normal MFL +/- 5% 80% 0.86 t 1)Normal MFL +/- 10% 80% 0.70 tHigh MFL +/- 10% 80% 0.68 tNormal MFL +/- 20% 80% 0.41 t1) Limited to maximum 0.85 t, see a) above.

STEP 1 Determine the longitudinal stress, at the location of the corrosion defect, from external loads, as for instance axial, bending and temperature loads on the pipe. Calculate the nominal longitudinal elas-tic stresses in the pipe at the location of the corro-sion defect, based on the nominal pipe wall thickness:

( )tDft

p um =2

0

]/[StD/1)/( , tdtd ddaccmeas

( ) ttDFX

A = 4M

2Y

B =DET NORSKE VERITAS

( ) ttD

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 154.4 Circumferential corrosion defects, internal pres-sure and superimposed longitudinal compressive stressesThe acceptance equation given below is not valid for full cir-cumference corrosion defects with a longitudinal lengthexceeding 1.5t. The allowable corroded pipe pressure of a single circumferen-tial corrosion defect can be estimated using the following pro-cedure:

Figure 4-1

The combined nominal longitudinal stress is:

STEP 2 If the combined longitudinal stress is compres-sive, then calculate the allowable corroded pipe pressure, including the correction for the influ-ence of compressive longitudinal stress:

where:

pcorr,comp is not allowed to exceed pcorr.

STEP 1 Determine the longitudinal stress, at the location of the corrosion defect, from external loads, as for instance axial, bending and temperature loads on the pipe. Calculate the nominal longitudinal elas-tic stresses in the pipe, based on the nominal pipe wall thickness:

BAL +=

( )( )

1, *)/(1

*)/(12 H

Qtdtd

tDftp

d

dumcompcorr

=

( )

+=

Qtdtd

A

AfH

d

d

r

m

ru

L

*)/(1

*)/(12

1

11

1

= tdAr 1

The combined nominal longitudinal stress is:

STEP 2 If the combined longitudinal stress is compres-sive, then calculate the allowable corroded pipe pressure, including the correction for the influ-ence of compressive longitudinal stress:

where:

pcorr,circ is not allowed to exceed pmao.

The longitudinal pipe wall stress in the remaining ligament is not to exceed fy, in tension or in compression. The longitudinal pipe wall stress shall include the effect of all loads, including the pressure.

where: L-nom is the longitudinal stress in the nominal pipe wall.

( ) ttDFX

A =

( ) ttD4M

2Y

B =

BAL +=

( ) ( )

+= tD

ft

A

AftDftp umc

r

mc

ru

L

umccirccorr

2,12

1

112min,

= tdAr 1

( ))/(1 tdf ynomL

t

t

d

Al

c

Ac

d

t

t

d

Al

c

Ac

dDET NORSKE VERITAS

Single defect dimensions

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 16 see note on front coverFigure 4-2Interacting defect dimensions

5. Assessment of Interacting Defects (Part A)5.1 RequirementsThe interaction rules are strictly valid for defects subject toonly internal pressure loading. The rules may be used to deter-mine if adjacent defects interact under other loading condi-tions, at the judgement of the user. However, using theseinteraction rules may be non-conservative for other loadingconditions. The minimum information required comprises:

The angular position of each defect around circumferenceof the pipe.

The axial spacing between adjacent defects. Whether the defects are internal or external. The length of each individual defect. The depth of each individual defect. The width of each individual defect.

5.2 Allowable corroded pipe pressure estimateThe partial safety factors for interacting defects have not beenderived from an explicit probabilistic calibration. The partialsafety factors for a single defect subject to internal pressureloading have been used. The allowable corroded pipe pressure of a colony of interact-ing defects can be estimated using the following procedure:

Guidance note:Within the colony of interacting defects, all single defects, and allcombinations of adjacent defects, are considered in order todetermine the minimum predicted failure pressure.Combined defects are assessed with the single defect equation,using the total length (including spacing) and the effective depth(based on the total length and a rectangular approximation to thecorroded area of each defect within the combined defect).

---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---

Axis

s l1 l2

Defect 1

Defect 2

d1 d2

STEP 1 For regions where there is background metal loss (less than 10% of the wall thickness) the local pipe wall thickness and defect depths can be used (see Fig.5-1).

STEP 2 The corroded section of the pipeline should be divided into sections of a minimum length of

, with a minimum overlap of . Steps 3 to 12 should be repeated for each sectioned length to assess all possible interactions.

STEP 3 Construct a series of axial projection lines with a circumferential angular spacing of:

(degrees)

STEP 4 Consider each projection line in turn. If defects lie within Z, they should be projected onto the cur-rent projection line (see Fig.5-2).

STEP 5 Where defects overlap, they should be combined to form a composite defect. This is formed by taking the combined length, and the depth of the deepest defect (see Fig.5-3). If the composite defect con-sists of an overlapping internal and external defect then the depth of the composite defect is the sum of the maximum depth of the internal and external defects (see Fig.5-4).

STEP 6 Calculate the allowable corroded pipe pressure (p1, p2 pN) of each defect, to the Nth defect, treating each defect, or composite defect, as a single defect:

i = 1N

where:

If (d/t)* 1 then p = 0.

50. Dt 2 5. Dt

DtZ 360=

( )( )

=

i

id

idumi

Qtdtd

tDftp

*)/(1

*)/(12

QlDtii= +

1 031

2

.

]/[StD)/()*/( tdtdtd dmeasii +=DET NORSKE VERITAS

d i

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 17Guidance note:Steps 7 to 9 estimate the allowable corroded pipe pressure of allcombinations of adjacent defects. The allowable corroded pipepressure of the combined defect nm (i.e. defined by single defectn to single defect m, where n = 1 N and m = n N) is denotedpnm.


Guidance note:The formula for StD[dnm/t] assumes fully correlated depth meas-urements. In the case that the measurements are not fully corre-lated the uncertainty is reduced. The estimate and the effect of theapplied measurement uncertainty need to be assessed and docu-mented it the reduced uncertainty is to be used. In cases wherethe conditions are not known it is recommended to assume fullycorrelated depth measurements.


Figure 5-1Corrosion depth adjustment for defects with background corrosion

STEP 7 Calculate the combined length of all combina-tions of adjacent defects (see Fig.5-5 and Fig.5-6). For defects n to m the total length is given by:

n,m = 1N

STEP 8 Calculate the effective depth of the combined defect formed from all of the interacting defects from n to m, as follows (see Fig.5-5):

STEP 9 Calculate the allowable corroded pipe pressure of the combined defect from n to m (pnm) (see Fig.5-6, using lnm and dnm in the single defect equation:

n,m = 1N

where:

( )l l l snm m i ii n

i m

= + +=

= 1

dd l

lnmi i

i n

i m

nm= =

=

( )( )

=

nm

nmd

nmdumnm

Qtdtd

tDftp

*)/(1

*)/(12

2

31.01

+=Dt

lQ nmnm

]/[StD)/()*/( tdtdtd nmdmeasnmnm +=

If d (d/t)* 1 then pcorr = 0.Note that d and d are functions of StD[dnm/t].

Fully correlated depth measurements:

STEP 10 The allowable corroded pipe pressure for the current projection line is taken as the minimum of the failure pressures of all of the individual defects (p1 to pN), and of all the combinations of individual defects (pnm), on the current pro-jection line.

pcorr is not allowed to exceed pmao.

STEP 11 The allowable corroded pipe pressure for the section of corroded pipe is taken as the mini-mum of the allowable corroded pipe pressures calculated for each of the projection lines around the circumference.

STEP 12 Repeat Steps 3 to 11 for the next section of the corroded pipeline.

[ ] [ ]nm

ii

mi

ninm l

tdltd

/StD/StD

===

),,...,min( 21 nmNcorr ppppp =

ld

t

< 0.1t DET NORSKE VERITAS

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 18 see note on front coverFigure 5-2Projection of circumferentially interacting defects

Figure 5-3Projection of overlapping sites onto a single projection line and the formation of a composite defect

Axial Projection Lines

Box Enclosing Defect

Project onto Line

Z

Z

di

li si

Projection Line

Section Through Projection Line DET NORSKE VERITAS

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 19Figure 5-4Projection of overlapping internal and external defects onto a single projection line and the formation of a composite defect

Figure 5-5Combining interacting defects

d 1

li

Projection Line

Section Through Projection Line

d2

21 ddd i + =

sn ln lm ln+1

dm

sm-1

( ) l l l s nm m i ii n

i m = + +

=

= 1 d d l l nmi i

i n

i m

nm= =

=

lnm

dn+1dn DET NORSKE VERITAS

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 20 see note on front coverFigure 5-6Example of the grouping of adjacent defects for interaction to find the grouping that gives the lowest estimated failure pressure

1-2-3

1-2-3-4

1

1-2

2-3-4

3

2

2-3

3-4

4

GROUP

Defect 1 Defect 2 Defect 3 Defect 4DET NORSKE VERITAS

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 216. Assessment of Complex Shaped Defects (Part A)6.1 RequirementsThis method must only be applied to defects subjected to inter-nal pressure loading only.The minimum information required comprises:

1) A length and depth profile for the complex shape. Thelength must be the axial length along the axis of the pipe.The defect depth, at a given axial length along the defect,should be the maximum depth around the circumferencefor that axial length (i.e. a river bottom profile of thedefect).

2) The length of the profile must include all material betweenthe start and end of the complex shaped defect.

6.2 Allowable corroded pipe pressure estimateThe partial safety factors for a complex shaped defect have notbeen derived from an explicit probabilistic calibration. Thepartial safety factors for a single defect subject to internal pres-sure loading have been used. The allowable corroded pipe pressure of a complex shapeddefect can be estimated using the following procedure:

Guidance note:The principle underlying the complex shaped defect method is todetermine whether the defect behaves as a single irregularpatch, or whether local pits within the patch dominate the fail-ure. Potential interaction between the pits has also to be assessed.A progressive depth analyses is performed. The corrosion defectis divided into a number of increments based on depth.At each depth increment the corrosion defect is modelled by anidealised patch containing a number of idealised pits. Thepatch is the material loss shallower than the given incrementdepth. The pits are defined by the areas which are deeper thanthe increment depth, see Fig.6-1 and Fig.6-2. The allowable cor-roded pipe pressure of the pits within the patch is estimatedby considering an equivalent pipe of reduced wall thickness. Thecapacity (failure pressure) of the equivalent pipe is equal to thecapacity of the patch.The idealised pits in the equivalent pipe are assessed using theinteracting defect method (see Sec.5).The estimated allowable corroded pipe pressure at a given depthincrement, is the minimum of the allowable corroded pipe pres-sure of the patch, the idealised pits, and the allowable cor-roded pipe pressure of the total corroded area based on its totallength and average depth.The procedure is repeated for all depth increments in order todetermine the minimum predicted allowable corroded pipe pres-sure. This is the allowable corroded pipe pressure of the complexshaped defect.


Guidance note:Note that d and d are functions of StD[dave/t].The formula for StD[dave/t] assumes fully correlated depth meas-urements. In the case that the measurements are not fully corre-lated the uncertainty is reduced. The estimate and the effect of theapplied measurement uncertainty need to be assessed and docu-mented if the reduced uncertainty is to be used. In cases wherethe conditions are not known, it is recommended to assume fullycorrelated depth measurements.


STEP 1 Calculate the average depth (dave) of the complex shaped defect as follows:

STEP 2 Calculate the allowable corroded pipe pressure of the total profile (ptotal), using dave and ltotal in the single defect equation:

dA

lave total=

( )( )

=

total

aved

avedumtotal

Qtdtd

tDftp

*)/(1

*)/(12

If d (dave/t)* 1 then ptotal = 0.Fully correlated depth measurements:

STEP 3 Divide the maximum defect depth into incre-ments, and perform the below calculations for all depth increments (dj) (see Fig.6-1). Each subdi-vision of the profile separates the profile into an idealised patch portion, shallower than the depth subdivision (i.e. the maximum depth of the patch is dj), and into pits which are deeper than the subdivision (see Fig.6-2). The recom-mended number of increments is between 10 and 50.

STEP 4 Calculate the average depth of an idealised patch as follows (see Fig.6-2):

STEP 5 Calculate the allowable corroded pipe pressure of the idealised patch (ppatch) and the predicted failure pressure (capacity) of the idealised patch (pcap,patch), using ltotal and dpatch in the single defect equation:

Calculate also for use in Step 7:

where:

If (d /t)* 1 then p = 0.

2

31.01

+=Dt

lQ totaltotal

]/[StD)/()*/( tdtdtd avedmeasaveave +=

]/[StD]/[StD tdtdave =

total

patchpatch l

Ad =

( )( )

=

total

patchd

patchdumpatch

Qtdtd

tDftp

*)/(1

*)/(12

( )( )

=

total

patch

patchupatchcap

Qtdtd

tDftp

)/(1

)/(1209.1,

2

31.01

+=Dt

lQ totaltotal

]/[StD)/()*/( tdtdtd patchdmeaspatchpatch +=DET NORSKE VERITAS

where: d patch patch

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 22 see note on front coverGuidance note:Note that d and d are functions of StD[dpatch/t].The formula for StD[dave/t] assumes fully correlated depth meas-urements. In the case that the measurements are not fully corre-lated the uncertainty is reduced. The estimate and the effect of theapplied measurement uncertainty need to be assessed and docu-mented if the reduced uncertainty is to be used. In cases wherethe conditions are not known, it is recommended to assume fullycorrelated depth measurements.


Guidance note:Steps 10 to 12 estimate the allowable corroded pipe pressures ofall combinations of adjacent defects. The allowable corrodedpipe pressure of the combined defect nm (i.e. defined by singledefect n to single defect m, where n = 1 N and m = n N) isdenoted pnm.

Guidance note:The formula for StD[de,nm/t] assumes fully correlated depthmeasurements. In the case that the measurements are not fullycorrelated the uncertainty is reduced. The estimate and the effectof the applied measurement uncertainty need to be assessed anddocumented if the reduced uncertainty is to be used. In caseswhere the conditions are not known it is recommended to assumefully correlated depth measurements.


Fully correlated depth measurements:

STEP 6 For each of the idealised pits, calculate the area loss in the nominal thickness cylinder, as shown in Fig.6-2, for the current depth interval, and esti-mate the average depth of each of the idealised pits from:

i = 1N

STEP 7 Estimate the effective thickness of an equiva-lent pipe with the same failure pressure as the patch, (pcap,patch), as calculated in Step 5 (see Fig.6-1).

STEP 8 The average depth of each pit is corrected for the effective thickness (te) using:

STEP 9 Calculate the corroded pipe pressure of all indi-vidual idealised pits (p1, p2, pN) as isolated defects, using the corrected average depth (dei), and the longitudinal length of the each idealised pit (li) in the single defect equation:

i = 1N

where:

If d (dei/te)* 1 then pi = 0.

]/[StD]/[StD tdtd patch =

i

pitii l

Ad ,=

( ) )09.1(2 ,, patchcapupatchcape pfDp

t +=

( )eiei ttdd =

( )( )

=

i

eeid

eeid

e

uemi

Qtdtd

tDftp

*)/(1

*)/(12

2

31.01

+=

e

ii Dt

lQ

]/[StD)/()*/( tdtdtd dmeaseieei +=

STEP 10 Calculate the combined length of all combina-tions of adjacent defects (see Fig.5-5 and Fig.5-6). For defects n to m the total length is given by:

N,m = 1N

STEP 11 Calculate the effective depth of the combined defect formed from all of individual idealised pits from n to m, as follows (see Fig.5-5):

STEP 12 Calculate the allowable corroded pipe pressure of the combined defect from n to m (pnm) (see Fig.5-6), using lnm , te and de,nm in the single defect equation:

n,m = 1N

where:

If d (de,nm/te)* 1 then pnm = 0.Note that d and d are functions of StD[de,nm/t].Fully correlated depth measurements:

STEP 13 The allowable corroded pipe pressure for the cur-rent depth increment is taken as the minimum of all the allowable corroded pipe pressures from above:

( )==

++=1mi

niiimnm slll

nm

mi

niiei

nme l

ld

d===,

( )( )

=

nm

enmed

enmed

e

uemnm

Qtdtd

tDftp

*)/(1

*)/(12

,

,

2

31.01

+=

e

nmnm Dt

lQ

]/[StD)/()*/( ,,, tdtdtd nmedmeasnmeenme +=

[ ][ ]

nm

eii

mi

ninme l

tdl

td

/StD

/StD ,===

),,,,...,min( 21 totalpatchnmNcorr ppppppp =DET NORSKE VERITAS

---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e--- j

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 23Figure 6-1Subdivision of complex shape into idealised 'patch' and 'pits'

Figure 6-2Definition of Apatch and Apit for subdivision of complex shape into idealised 'patch' and 'pits'

STEP 14 Repeat the Steps 4 to 13 for the next interval of depth increment (dj) until the maximum depth of corrosion profile has been reached.

STEP 15 Calculate the allowable pipe pressure according to the single defect equation in Sec.4.2 using the maximum defect depth and the total length of the defect.

STEP 16 The allowable corroded pipe pressure of the complex shaped defect (pcorr) should be taken as the minimum of that from all of the depth inter-vals, but not less than the allowable pressure for a single defect calculated in Step 15.pcorr is not allowed to exceed pmao.

Current Depth Increment dj

dj

li si

de,i

dpatch

te

ltotal

di

Current Depth Increment, d

d j

j

Apatch

Apit

t

Current Depth Increment, d

d j

j

Apatch

Apit

tDET NORSKE VERITAS

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 24 see note on front cover7. Allowable Stress Approach (Part B)7.1 IntroductionThe approach given in Part B is based on the ASD (AllowableStress Design) format. The failure pressure (capacity) of thepipeline with the corrosion defect is calculated, and this failurepressure is multiplied by a single safety factor based on theoriginal design factor. When assessing corrosion defects, due consideration should begiven to the measurement uncertainty of the defect dimensionsand the pipeline geometry.

7.2 Total usage factorThe usage factor to be applied in determining the safe workingpressure has two components:

The Total Usage Factor (F) to be applied to determine the safeworking pressure should be calculated from: F = F1F2

8. Assessment of a Single Defect (Part B)8.1 RequirementsIsolated metal loss defects are to be individually assessed assingle defects, see Fig.4-1. Adjacent defects can interact to produce a failure pressure thatis lower than the individual failure pressures of the isolateddefects treated as single defects. For the case where interactionoccurs, the single defect equation is no longer valid and theprocedure given in Sec.9 must be applied. Fig.4-2 shows thekey dimensions for defect interaction.A defect can be treated as an isolated defect, and interactionwith other defects need not be considered, if either of the fol-lowing conditions is satisfied:

1) The circumferential angular spacing between adjacentdefects, :

2) The axial spacing between adjacent defects, s:

8.2 Safe working pressure estimate - Internal pres-sure onlyThe safe working pressure of a single defect subject to internalpressure loading only is given by the following equation:

Due consideration should be given to the measurement uncer-tainty of the defect dimensions and the pipeline geometry,which is not accounted for in the equations.If the wall thickness is close to the required minimum remain-ing wall thickness, special care should be given. E.g. for a10mm wall thickness pipeline the minimum requirement isonly 1.5 mm. Special attention should be given to thesedefects, both in term of reliability of the inspection methodsand result and potential further growth.

8.3 Safe working pressure estimate - Internal pres-sure and combined compressive loadingThe validation of the method for assessing corrosion defectssubject to internal pressure and longitudinal compressivestresses is not as comprehensive as the validation of themethod for assessing corrosion defects under internal pressureloading only.Method for assessing a single defect subject to tensile longitu-dinal and/or bending stresses is given in e.g. refs /6/ and /12/The safe working pressure of a single corrosion defect subjectto internal pressure and longitudinal compressive stresses canbe estimated using the following procedure:

F1 = 0.9 ( Modelling Factor )

F2 = Operational Usage Factor which is introduced to ensure a safe margin between the operating pres-sure and the failure pressure of the corrosion defect (and is normally taken as equal to the Design Fac-tor).

STEP 1 Calculate the failure pressure of the corroded pipe (Pf):

(degrees)360Dt>

Dts 0.2>

( )

=

tQdtd

tDft

P uf1

12

where:

STEP 2 Calculate the safe working pressure of the cor-roded pipe (Psw):

Measured defects depths exceeding 85% of the wall thickness is not accepted.

STEP 1 Determine the longitudinal stress, at the location of the corrosion defect, from external loads, as for instance axial, bending and temperature loads on the pipe. Calculate the nominal longitudinal elas-tic stresses in the pipe at the location of the corro-sion defect, based on the nominal pipe wall thickness:

The combined nominal longitudinal stresses is:

STEP 2 Determine whether or not it is necessary to con-sider the effect of the external compressive longi-tudinal loads on the failure pressure of the single defect (see Fig.8-1).

It is not necessary to include the external loads if the loads are within the following limit:

where:

2

31.01

+=

DtlQ

fsw PFP =

( ) ttDFX

A =

( ) ttD4M

2Y

B =

BAL +=

1 >LDET NORSKE VERITAS

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 25If the above condition is satisfied then Step 4 can be neglected.

STEP 3 Calculate the failure pressure of the single corro-sion defect under internal pressure only, using the following equation:

where:

STEP 4 Calculate the failure pressure for a longitudinal break, including the correction for the influence of compressive longitudinal stress (Fig.8-2):

where:

STEP 5 Determine the failure pressure of a single corro-sion defect subjected to internal pressure loading combined with compressive longitudinal stresses:

STEP 6 Calculate the safe working pressure of the cor-roded pipe (Psw):

=

tQdtd

fu1

15.01

( )

=

tQdtd

tDft

P upress1

12

2

31.01

+=

DtlQ

( ) 11

12

H

tQdtd

tDft

P ucomp

=

+=

tQdtd

A

AfH

r

ru

L

1

1

211

11

1

= tdAr 1

),min comppressf P(P P =

fsw PFP =DET NORSKE VERITAS

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 26 see note on front coverFigure 8-1Range of superimposed longitudinal and/or bending loads that will not influence the failure pressure

Figure 8-2Influence of applied loads on the failure mode of a corrosion defect

9. Assessment of Interacting Defects (Part B)9.1 RequirementsThe interaction rules are strictly valid for defects subject toonly internal pressure loading. The rules may be used to deter-mine if adjacent defects interact under other loading condi-tions, at the judgement of the user. However, using theseinteraction rules may be non-conservative for other loadingconditions. The methods given in Sec.8 for assessing corro-sion defects under combined loads are only valid for single

The minimum information required comprises:

The angular position of each defect around circumferenceof the pipe.

The axial spacing between adjacent defects. Whether the defects are internal or external. The length of each individual defect. The depth of each individual defect. The width of each individual defect.

9.2 Safe working pressure estimateThe safe working pressure can be estimated from the following

HOOP

LONGITUDINAL Arfu 1 2External axial or bending stresses do not influence failure pressure, if greater than 1 and less than 2

=

tQ

d

t

d

fu1

1FAILURE SURFACE

HOOP

LONGITUDINAL Arfu 1 2External axial or bending stresses do not influence failure pressure, if greater than 1 and less than 2

=

tQ

d

t

d

fu1

1

=

tQ

d

t

d

fu1

1FAILURE SURFACE

HOOP

LONGITUDINAL

CIRCUMFERENTIAL BREAK

LONGITUDINAL BREAK

REDUCED PRESSURELONGITUDINAL BREAK

BUCKLING / WRINKLING

FAILURE SURFACEDET NORSKE VERITAS

defects. procedure:

Amended October 2006 Recommended Practice DNV-RP-F101, October 2004see note on front cover Page 27Guidance note:Within the colony of interacting defects, all single defects, and allcombinations of adjacent defects, are considered in order todetermine the minimum safe working pressure. Combined defects are assessed with the single defect equation,using the total length (including spacing) and the effective depth(calculated the total length and a rectangular approximation tothe corroded area of each defect within the combined defect).


Guidance note:Steps 7 to 9 estimate the failure pressures of all combinations ofadjacent defects. The failure pressure of the combined defect nm(i.e. defined by single defect n to single defect m, where n = 1 N and m = n N) is denoted Pnm.

10. Assessment of a Complex Shaped Defect (Part B)10.1 RequirementsThis method must only be applied to defects subjected to inter-

STEP 1 For regions where there is background metal loss (less than 10% of the wall thickness) the local pipe wall thickness and defect depths can be used (see Fig.5-1).

STEP 2 The corroded section of the pipeline should be divided into sections of a minimum length of

with a minimum overlap of . Steps 3 to 12 should be repeated for each sectioned length to assess all possible interactions.

STEP 3 Construct a series of axial projection lines with a circumferential angular spacing of:

(degrees)

STEP 4 Consider each projection line in turn. If defects lie within Z, they should be projected onto the cur-rent projection line (see Fig.5-2).

STEP 5 Where defects overlap, they should be combined to form a composite defect. This is formed by taking the combined length, and the depth of the deepest defect, see Fig.5-3). If the composite defect con-sists of an overlapping internal and external defect then the depth of the composite defect is the sum of the maximum depth of the internal and external defects (see Fig.5-4).

STEP 6 Calculate the failure pressures (P1, P2 PN) of each defect, to the Nth defect, treating each defect, or composite defect, as a single defect:

i = 1N

where:

STEP 7 Calculate the combined length of all combinations of adjacent defects (see Fig.5-5 and Fig.5-6). For defects n to m the total length is given by:

n,m = 1N

Dt0.5 Dt5.2

DtZ 360=

( )

=

i

i

i

ui

tQdt

d

tDft

P1

12

2

31.01

+=

DtlQ ii

( )= ++= 1mi iimnm slll

STEP 8 Calculate the effective depth of the combined defect formed from all of the interacting defects from n to m, as follows (see Fig.5-5):

STEP 9 Calculate the failure pressure of the combined defect from n to m (Pnm) (see Fig.5-6), using lnm and dnm in the single defect equation:

where:

STEP 10 The failure pressure for the current projection line, is taken as the minimum of the failure pres-sures of all of the individual defects (P1 to PN), and of all the combinations of individual defects (Pnm), on the current projection line.

STEP 11 Calculate the safe working pressure (Psw) of the interacting defects on the current projection line:

STEP 12 The safe working pressure for the section of cor-roded pipe is taken as the minimum of the safe working pressures calculated for each of the pro-jection lines around the circumference.

STEP 13 Repeat Steps 3 to 12 for the next section of the corroded pipeline.

nm

mi

niii

nm l

ldd

===

( )

=

nm

nm

nm

unm

tQd

td

tDft

P1

12

2

31.01

+=Dt

lQ nmnm

),,...,( 21 nmNf PPPPMINP =

fsw PFP =DET NORSKE VERITAS

nal pressure loading only.=ni

Recommended Practice DNV-RP-F101, October 2004 Amended October 2006Page 28 see note on front coverThe minimum information required comprises:

1) A length and depth profile for the complex shape. Thelength must be the axial length al

RP-F101

Documents

foundation det norske

omission of det norske

provision det norske

dnv offshore codes

dnv offshore services

offshore standards

offshore units

offshore industry