DNV-RP-F101 DET NORSKE VERITAS

RECOMMENDED PRACTICE
DET NORSKE VERITAS
DNV-RP-F101
CORRODED PIPELINES
OCTOBER 2004

© 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 person
for 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 Det
Norske Veritas.
FOREWORD
DET 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 consultancy
services relating to quality of ships, offshore units and installations, and onshore industries worldwide, and carries out research
in 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 Methodology
B)Materials Technology
C)Structures
D)Systems
E)Special Facilities
F)Pipelines and Risers
G)Asset Operation
H)Marine Operations
J)Wind Turbines
Amendments and Corrections
This document is valid until superseded by a new revision. Minor amendments and corrections will be published in a separate
document normally updated twice per year (April and October).
For a complete listing of the changes, see the “Amendments and Corrections” document located at:
, “Offshore Rules & Standards”, “Viewing Area”.
The electronic web-versions of the DNV Offshore Codes will be regularly updated to include these amendments and corrections.
Acknowledgements
This Recommended Practice is based upon a project guideline developed in a co-operation between BG Technology and
DNV. The results from their respective Joint Industry Projects (JIP) have been merged and form the technical basis for this
Recommended Practice.
We would like to take this opportunity to thank the sponsoring companies / organisations for their financial and technical
contributions (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.DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 3
CONTENTS
1.GENERAL.............................................................. 5
1.1Introduction.............................................................5
1.2Update year 2004.....................................................5
1.3BG plc and DNV research projects........................5
1.4Application...............................................................5
1.5Structure of RP........................................................5
1.6Applicable defects....................................................5
1.7Applied loads............................................................6
1.8Exclusions.................................................................7
1.9Other failure modes.................................................7
1.10Tiered approach and further assessment ..............7
1.11Responsibility...........................................................7
1.12Validation.................................................................7
1.13Definitions................................................................7
1.14Symbols and abbreviations.....................................8
1.15Units..........................................................................9
2.METHODOLOGY................................................. 9
2.1Capacity equation....................................................9
2.2Sizing accuracy and uncertainties..........................9
2.3Part A, calibrated safety factors..........................10
2.4Part B, allowable stress approach........................10
2.5Onshore pipelines..................................................10
2.6Characteristic material properties.......................10
2.7Pressure reference height and static head...........10
2.8Probabilistic assessments......................................11
3.CALIBRATED SAFETY FACTOR (PART A) 11
3.1Introduction...........................................................11
3.2Reliability levels.....................................................11
3.3Partial safety factors and fractile values.............11
3.4Circumferential corrosion....................................13
3.5Usage factors for longitudinal stress....................13
3.6System effect..........................................................13
3.7Supplementary material requirements...............13
4.ASSESSMENT OF A SINGLE DEFECT
(PART A)............................................................... 14
4.1Requirements.........................................................14
4.2Longitudinal corrosion defect, internal pressure
loading only............................................................14
4.3Longitudinal corrosion defect, internal pressure
and superimposed longitudinal compressive
stresses....................................................................14
4.4Circumferential corrosion defects, internal
pressure and superimposed longitudinal
compressive stresses..............................................15
5.ASSESSMENT OF INTERACTING
DEFECTS (PART A)........................................... 16
5.1Requirements.........................................................16
5.2Allowable corroded pipe pressure estimate........16
6.ASSESSMENT OF COMPLEX SHAPED
DEFECTS (PART A)........................................... 21
6.1Requirements.........................................................21
6.2Allowable corroded pipe pressure estimate........21
7.ALLOWABLE STRESS APPROACH
(PART B).............................................................. 24
7.1Introduction...........................................................24
7.2Total usage factor..................................................24
8.ASSESSMENT OF A SINGLE DEFECT
(PART B).............................................................. 24
8.1Requirements.........................................................24
8.2Safe working pressure estimate - Internal
pressure only..........................................................24
8.3Safe working pressure estimate - Internal
pressure and combined compressive loading.....24
9.ASSESSMENT OF INTERACTING
DEFECTS (PART B)........................................... 26
9.1Requirements.........................................................26
9.2Safe working pressure estimate...........................26
10.ASSESSMENT OF A COMPLEX SHAPED
DEFECT (PART B)............................................. 28
10.1Requirements.........................................................28
10.2Safe working pressure estimate...........................28
11.REFERENCES..................................................... 30
APP. AEXAMPLES FOR PART A............................... 31
APP. BEXAMPLES FOR PART B............................... 35
APP. CDETAILED CALCULATION OF
MEASUREMENT ACCURACIES .................. 41DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 4 see note on front coverDET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 5
1.  General
1.1  Introduction
This document provides recommended practice for assessing
pipelines containing corrosion. Recommendations are given
for 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 two
alternative approaches to the assessment of corrosion, and the
document is divided into two parts. The main difference
between the two approaches is in their safety philosophy:
The first approach, given in Part A, includes calibrated safety
factors 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 the
specification of the material properties are specifically consid-
ered in determination of the allowable operating pressure. This
part of the RP is also a supplement to DNV-OS-F101. Proba-
bilistic calibrated equations (with partial safety factors) for the
determination 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 original
design factor. Consideration  of the uncertainties associated
with the sizing of the corrosion defect is left to the judgement
of the user.
1.2  Update year 2004
The RP was first issued in 1999 and updated in 2004 (this ver-
sion). The update is based on experience and feedback from
four years of use.
The update covers:
—Sec.3 (previously Sec.2) concerning Part A safety factors
has 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 of
which the user should be aware:
—the calculation of fully correlated depth measurement for
interaction defects Part A (Step 9 in Sec.5.2 and Step 12 in
Sec.6) is modified, and is less strict (the 1999 version is
conservative)
—UTS in Part B is changed to SMTS and “fu”.
1.3  BG plc and DNV research projects
This document is a result of co-operation between BG Tech-
nology (part of BG plc) and DNV. The results from their
respective joint industry projects have been merged, and form
the technical basis for this recommended practice (/3/, /4/ and
/16/).
The BG technology project generated a database of more than
70 burst tests on pipes containing machined corrosion defects
(including single defects, interacting defects and complex
shaped defects), and a database of linepipe material properties.
In addition, a comprehensive database of 3D non-linear finite
element analyses of pipes containing defects was produced.
Criteria were developed for predicting the remaining strength
of corroded pipes containing single defects, interacting defects
and complex shaped defects.
The DNV project generated a database of 12 burst tests on
pipes containing machined corrosion defects, including the
influence of superimposed axial and bending loads on the fail-
ure pressure. A comprehensive database of 3D non-linear
finite element analyses of pipes containing defects was also
produced. Probabilistic methods were utilised for code calibra-
tion and the determination of partial safety factors.
1.4  Application
The methods provided in this document are intended to be used
on corrosion defects in carbon steel pipelines (not applicable
for other components) that have been designed to the DNV
Offshore 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/, CSAZ662-94 /7/, provided that the
safety philosophy in the design code is not violated.
When assessing corrosion, the  effect of continued corrosion
growth should be considered. If a corroded region is to be left
in service then measures should be taken to arrest further cor-
rosion growth, or an appropriate inspection programme should
be adopted. The implications of continuing defect growth are
outside 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 RP
The RP describes two alternative approaches. The first
approach is given in Part A, which consists of Sec.3 through
Sec.6. The second approach is given in Part B, which consists
of Sec.7 through Sec.10.
A flow chart describing the assessment procedure (for both
Part A and Part B) is shown in Fig.1-1.
Worked examples are given in Appendix A for the methods
described in Part A and Appendix B for the methods described
in Part B.
1.6  Applicable defects
The following types of corrosion defect can be assessed using
this 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 removal
of the original defect has been verified using appropriate
NDT methods).
When applying the methods to corrosion defects in seam welds
and 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 the
weld has an adequate toughness.DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 6 see note on front cover
Figure 1-1
Flowchart of the assessment procedure
1.7  Applied loads
Internal 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:
—Single defects.
The compressive longitudinal stress can be due to axial loads,
bending loads, temperature loads etc.
The recommended practice given in this document is confined
to the effects of internal pressure and compressive longitudinal
loading on longitudinal failure because the validation of these
effects was addressed in  the DNV and BG Technology
projects.
The behaviour of corrosion defects under combined internal
pressure and bending loads, and/or tensile longitudinal loads,
was outside the scope of the DNV and BG Technology projects
and, therefore, this loading combination has not been included
as 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/
and /12/).
  
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
using Section 5 (or 9).
Are Allowable Corroded Pipe
Pressures (Safe Working
Pressures) Acceptable?
Are Allowable Corroded Pipe
Pressures (Safe Working
Pressures) Acceptable?
Are Defect Profiles Available?
Analyse Corrosion Sites as a
Complex Shaped Defects using
Section 6 (or 10).
Best Estimate of Allowable
Corroded Pipe Pressures (Safe
Working Pressure).
INTERACTION
NO INTERACTION
NO  NO
YES
YES  YES
NO
Identify Type of Loading
Analyse all Corrosion Damage
Sites as Isolated Single Defects
using Section 4 (or 8).
PRESSURE ONLY
COMBINE
LOADING
START
Best Estimate of Allowable
Corroded Pipe Pressure (Safe
Working Pressure). DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 7
1.8  Exclusions
The 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 steels
that are expected to fail through plastic collapse. Modern pipe-
line steel materials normally have sufficient toughness to
expect plastic collapse failure. Studies have recommended
Charpy V-notch value as lower bound for the material tough-
ness for plastic collapse /18/ and /19/.
The procedure is not recommended for applications where
fracture is likely to occur. These may include:
10)Materials with Charpy values less than 27 J (20 ftlbf) full
size test (equivalient 2/3 scale is 18 J, 13 ftlbf). For the
weld 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"), unless
the 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 modes
Other failure modes, such as buckling, wrinkling, fatigue and
fracture, may need to be considered. These failure modes are
not addressed in this document, and other methods may be
applicable, ref. /6/, /12/ and /14/.
1.10  Tiered approach and further assessment
The intent of this RP is to provide tiered procedures for the
assessment of corroded pipe. The first tier level is the simpli-
fied approach for single defect assessment, where total length
and maximum depth of the defect and the material specifica-
tion are used.
If the defect is not found to be acceptable a more refined
assessment 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 be
acceptable using the procedures given in this RP, the user has
the option of considering an alternative course of action to
more accurately assess the remaining strength of the corroded
pipeline. This could include, but is not limited to, detailed
finite element analysis, probabilistic assessments and/or full
scale testing, and is outside the scope of this document. If an
alternative course is selected, the user should document the
reliability of the results, and this can often be a very challeng-
ing task.
1.11  Responsibility
It 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 the
determination of defect size and associated sizing uncertain-
ties.
1.12  Validation
The methods given in this RP for assessing corrosion under
only internal pressure loading have been validated against 138
full scale vessel tests, including both machined defects and real
corrosion defects. The range of test parameters is summarised
below:
(Shortest defect was l = 2.1 t)
For nomenclature, see Sec.1.14.
The method for assessing corrosion defects under internal
pressure and compressive longitudinal loading has been vali-
dated against seven full scale tests on 324 mm (12inch) nom-
inal diameter, 10.3 mm nominal wall thickness, Grade X52
linepipe.
The method for assessing fully circumferential corrosion under
internal pressure and compressive longitudinal loading has
been validated against three full scale tests on 324 mm nominal
diameter, 10.3 mm nominal wall thickness, Grade X52 line-
pipe. The validation of this method is not as comprehensive as
the validation of the method for assessing a single longitudinal
corrosion defect subject to internal pressure loading only. The
partial safety factors have not been derived from an explicit
probabilistic calibration.
The validation of the methods described in this document for
the assessment of corrosion defects subject to internal pressure
loading plus compressive longitudinal stress (see Sec.4.3 and
4.4, is not as comprehensive as the validation of the methods
for the assessment of corrosion defects subject to internal pres-
sure loading alone.
The acceptance equation has not  been validated for defects
dimensions where the breadth (circumferential extent) of the
defect exceeding the length of  the defect. The partial safety
factors for combined loading have not been derived from an
explicit probabilistic calibration.
1.13  Definitions
A 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 neighbouring
defects in an axial or circumferential direction. The failure
pressure of an interacting defect is lower than it would be if the
interacting defect was a single defect, because of the interac-
tion with neighbouring defects.
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, mm219.1 (8")to914.4 (36")
Wall Thickness, mm3.40to25.40
D/t ratio8.6to149.4
Grade (API/5L)X42toX65
Defects:
d/t 0to0.97
l/(Dt)0.5 0.44to35
c/t (circumferential)0.01to22DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 8 see note on front cover
A Complex Shaped Defect is a defect that results from combin-
ing colonies of interacting defects, or a single defect for which
a 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 ‘i’th 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/π Dt
≈ 1-(d/t)θ
D=Nominal outside diameter (mm).
F=Total usage factor.
=F1F2
F1 =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 ‘j’th 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 Practice
Q=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
UTS=Ultimate Tensile Strength (N/mm2)
Pf
j
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 factor
c=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 ‘i’th 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 ‘i’th idealised ‘pit’ in a progressive
depth analysis of a complex shaped defect (mm).
dj
=The ‘j’th 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
depth
fu =Tensile strength to be used in design
fy =Yield strength to be used in design
i=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 ‘i’th 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 ‘j’th 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
stresses (N/mm2).
pcorr
jDET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 9
1.15  Units
The units adopted throughout this document are N and mm,
unless otherwise specified.
2.  Methodology
2.1  Capacity equation
The expression of the burst capacity for a single longitudinally
oriented, rectangular shaped, corrosion defect was developed
based on a large number of FE analyses, and a series of full
scale burst tests.  By using finite element analyses the effect of
each important parameter was investigated, while the accuracy
of the analyses was verified by a large number of full-scale
burst tests. The equations used in the development of this RP
and in the calibration are fairly complex. For practical use a
simplified capacity equation is give below. For more details
see /16/ and /17/.
The simplified capacity equation of a single rectangular
shaped defect is given as:
where 
This capacity equation represents the mean (best) estimate of
the capacity of a pipe with a  rectangular shaped corrosion
(metal loss) defect. This implies that on average the equation
should represent the capacity of the pipe but that some of the
defects will fail at a slightly lower pressure, and some at a
slightly 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.  For
example it is known that the equation over-predicts the failure
pressure (capacity) for medium long defect with high yield to
tensile ratio (high grade steel), and under-predict the failure
pressure for low yield to tensile ratio (low grade steel).
The accuracy of the capacity equation had to be known for
establishing the appropriate safety factors, and the above men-
tioned effects were accounted for.
The factor 1.05 in the capacity equation is determined from
comparison with laboratory test results with rectangular
shaped 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 will
in general underestimate the failure pressure, as the defect is
not as large as the rectangular  shaped defect assumed in the
capacity equation. This will result in a conservative estimate of
the failure pressure capacity for defects shapes other than rec-
tangular.
Figure 2-1
Illustration of irregular and rectangular defects
2.2  Sizing accuracy and uncertainties
For known defect size, pipe dimensions and material proper-
ties, the capacity equation predicts the burst capacity with a
good accuracy. However, these  input parameters usually
include a certain degree of uncertainty, and this should be
accounted for in calculating the acceptable operating pressure
of the corroded pipeline.
A high level of safety (reliability) is required for pipelines.
This is obtained by using safety factors in combination with
the capacity equation.
For example, in an assessment of a defect only the material
grade (giving SMTS and SMYS) will usually be available. The
actual material properties at the location of the defect will not
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).
εd =Factor for defining a fractile value for the corro-
sion depth.
φ =Circumferential angular spacing between adja-
cent defects (degrees).
γd =Partial safety factor for corrosion depth.
γm =Partial safety factor for longitudinal corrosion
model prediction.
γmc =Partial safety factor for circumferential corrosion
model prediction.
η =Partial safety factor for longitudinal stress for
circumferential corrosion.
θ =Ratio of circumferential length of corroded
region to the nominal outside circumference of
the pipe, (c/πD).
σA =Longitudinal stress due to external applied axial
force, based on the nominal wall thickness
(N/mm2).
σB =Longitudinal stress due to external applied bend-
ing moment, based on the nominal wall thickness
(N/mm2).
σL =Combined nominal longitudinal stress due to
external applied loads (N/mm2).
σu =Ultimate tensile strength (N/mm2).
σ1 =Lower bound limit on external applied loads
(N/mm2).
σ2 =Upper bound limit on external applied loads
(N/mm2).
ξ =Usage factor for longitudinal stress.
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
⋅
=
Q
t d
t d
t D
t
P u
cap
) / (
1
) / ( 1 2
05 . 1
σ
2
31 . 0 1 ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
L
Q
 
Rectangular shaped metal loss defectDET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 10 see note on front cover
be known. Furthermore, the defect sizing will be determined
with some level of uncertainty. The defect can be shallower, or
deeper, than the measured value, as illustrated in Fig.2-2. This
depth uncertainty has to be considered in the assessment of the
allowable pressure.
Figure 2-2
Measured defect depth and sizing accuracy
2.3  Part A, calibrated safety factors
The effect of the inspection accuracy, combined with the other
uncertainties described above, is accounted for in the calibra-
tion of the safety factor.  Although a single safety factor to
account for these uncertainties would give simpler calcula-
tions, several partial safety factors were introduced to give
results with a consistent reliability level for the validity range
of input parameters. If a single safety factor should cover the
full range of input parameters, this would give results with a
varying reliability level depending on the input parameters. If
the safety factor should be selected such that the minimum
required reliability level is satisfied in all cases, the code would
be undesirably conservative for some combinations of the
input parameters.
Results of FE analyses and laboratory tests, together with sta-
tistical data of material properties, pressure variations and
selected levels of uncertainties in the defect sizing, form the
required basis for a reliability code calibration where appropri-
ated safety factors were defined.
The maximum allowable operating pressure for a pipeline with
a corrosion defect is given by the acceptance equation with the
safety factors:
where
The safety factors are described in Sec.3.
2.4  Part B, allowable stress approach
The approach given in Part B is based on the ASD (Allowable
Stress Design) format. The failure pressure (capacity) of the
pipeline with the corrosion defect is calculated, and multiplied
by a safety factor to obtain a safe working pressure.  Often the
original design factor is used as the safety factor.
However, when assessing corrosion defects, due consideration
should be given to the measurement uncertainty of the defect
dimensions and the pipeline geometry. In contrast to Part A,
these uncertainties are not included in the Part B approach, and
are left to the user to consider and account for in the assess-
ment.
2.5  Onshore pipelines
Design codes for onshore pipelines allow in general a lower
utilisation of the material compared to offshore codes, i.e. the
safety factors are higher.  These factors probably implicitly
cover other loads and degradation mechanisms than consid-
ered in this RP, and if using Part A this could be in conflict
with the safety philosophy in the original design code.  Part B
could be more appropriate for onshore pipelines, where the
user have to account for these additional failures aspects.
However, when using Part B it is recommended that the user
also check according to Part A.  If this yields stricter results,
considerations should be made.
2.6  Characteristic material properties
The specified minimum tensile strength (SMTS) is used in the
acceptance 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 the
assessment of the metal loss defects. The material grades refer
to mechanical properties at  room temperature, and possible
temperature effects on the material properties should also be
considered.
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. In
lack of any material information the values in Fig.2-3 can be
used for both yield stress and tensile strength for temperatures
above 50°C.
Figure 2-3
Proposed de-rating values
2.7  Pressure reference height and static head
The 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 loads
should be used. The pressure reference height and the eleva-
tion of the defect must be known.
For offshore pipelines the benefit of external water pressure
can be utilised, and the increased pressure due to the internal
static head has to be included.
For onshore pipelines only the internal static head is to be
included.
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
Q
t d
t d
t D
SMTS t
p
d
d
m corr
* ) / (
1
* ) / ( 1 2
γ
γ
γ
] t / d [ StD ) t / d ( )* t / d ( d meas ⋅ + = ε
Table 2-1  Characteristic material properties
fy = SMYS – fy,temp
fu = SMTS – fu,temp
De-rating yield stress and tensile strength
0
20
40
60
80
100
050100150200
Temperature deg C
Stess de-rating (MPa)
CMnDET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 11
The calculated pressures, e.g. pcorr in this RP refer to the local
differential pressure load, and when determining pmao
(MAOP) the internal and external static head should be
included.
2.8  Probabilistic assessments
The safety factors in this RP are derived from probabilistic cal-
ibrations, and based on a set of input parameter distributions
that 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 the
assessment of metal loss defects.
Probabilistic assessment is outside the scope of this RP, and
the rest of Sec.2.8 is given for information only.
Probabilistic assessments of pipes with metal loss defects can
be based on the following limit state function:
g = Pcap - PINT
where
Pcap =the burst pressure capacity, but where the 1.05 factor is
replaced by XM
PINT=the annual maximum differential pressure.
The parameters in the limit state should be modelled with their
actual distributions, and considerations should be given to the
inspection sizing accuracy. A set of input parameter distribu-
tions considered to be representative for pipelines were used in
the 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 be
used as a fixed value, rather than a variable with associated dis-
tribution. However, the distributions (uncertainties) in the
model and the sizing accuracy have to be included in a proba-
bilistic assessment, where it is often seen that only one of these
are accounted for. The model uncertainty XM for the DNV-RP-
F101 capacity equation is given in the Table 2-2, while the
uncertainty in the sizing accuracy is given in Sec.3 of this RP.
In addition to the inspection accuracy, the corrosion rate will
also add to the uncertainty of the future defect size.
3.  Calibrated safety factor (Part A)
3.1  Introduction
The approach given in Part A  includes calibrated safety fac-
tors. Uncertainties associated with the sizing of the defect
depth and the material properties are specifically considered.
Probabilistic calibrated equations for the determination of the
allowable operating pressure of a corroded pipeline are given.
These equations are based on the LRFD (Load and Resistance
Factor Design) methodology.
Partial safety factors are given for two general inspection
methods (based on relative measurements e.g. magnetic flux
leakage, and based on absolute measurements e.g. ultrasonic),
four different levels of inspection accuracy, and three different
reliability levels.
3.2  Reliability levels
Pipeline design is normally to be based on Safety/Location
Class, Fluid Category and potential failure consequence for
each failure mode, and to be classified into safety classes.
Subsea oil and gas pipelines, where no frequent human activity
is anticipated, will normally be classified as Safety Class Nor-
mal.  Safety Class High is used for risers and the parts of the
pipeline close to platforms, or in areas with frequent human
activity.  Safety Class Low can be considered for e.g. water
injection pipelines. For more details see ref /8/ and other rele-
vant onshore and offshore pipeline codes.
3.3  Partial safety factors and fractile values
The partial safety factors are given as functions of the sizing
accuracy of the measured defect depth for inspections based on
relative depth measurements and for inspections based on
absolute depth.  For inspections based on relative depth meas-
urements the accuracy is normally quoted as a fraction of the
wall 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 with
the inspection tool provider.
The acceptance equation is based on two partial safety factors
and 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 γm  is given in Table 3-2 for inspection results
based on relative depth measurements, (e.g. Magnetic Flux
Leakage (MFL) intelligent pig measurements), and for abso-
lute depth measurements (e.g. Ultrasonic Wall Thickness or
Wall Loss Measurements). MFL  is a relative measurement
where the defect depth measurement and the accuracy are
given as a fraction of the wall thickness. The UT is an absolute
measurement where the local wall thickness, the defect depth
measurement and the accuracy are given directly.
Table 2-2  Parameters in the modelling of the burst limit state
VariableDistributionMeanUncertainty
PINT Gumbel1.05 MAOPCoV= 3.0%
DDeterministicActual -
tNormalNominal CoV =3.0%
σu Normal1.09 SMTSCoV = 3.0% and  6.0%
Lmeas NormalMeasured valueSpecified
d/tNormalMeasured valueSpecified
XM Normal1.05StD = 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   < 10-5
Normal   < 10-4
Low   < 10-3
γm =Partial safety factor for model prediction.
γd =Partial safety factor for corrosion depth.
εd =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 γm
Inspection method
Safety Class
LowNormalHigh
Relative (e.g. MFL) γm = 0.79 γm = 0.74 γm = 0.70
Absolute (e.g. UT) γm = 0.82 γm = 0.77 γm = 0.72DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 12 see note on front cover
The factors for absolute measurement are higher since it is
assumed that the pipe wall thickness around the corroded area
is measured with at least the same accuracy as the corrosion
depth. The measured values of the wall thickness (t) should be
used 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. The
standard deviation is further used to determine the  γd safety
factor and the εd 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 normal
cumulative distribution at probability x.
The confidence level indicates the portion of the measure-
ments that will fall within the given sizing accuracy. A selected
set 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 a
confidence level of 80%. A Normal distribution is assumed.
Figure 3-1
Example 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 the
detailed expressions of, the standard deviations, see
AppendixC, of absolute measurements used in the 1999 ver-
sion of this RP. The detailed expressions may also be used. The
simplification conservatively assumes d = t in the calculation
of StD[d/t].
A selected set of calculated standard deviations for absolute
sizing accuracy is given in Table 3-4 through Table 3-6 for a
wall thickness of 6.35 mm, 12.7 mm and 19.05 mm.
Safety factor γd and fractile value εd:
The   γd safety factor and the  εd fractile values are given in
Table 3-7 for various levels of inspection accuracy (defined in
terms of the standard deviation) and Safety Class:
Polynomial equations can be used to determine the appropriate
partial safety factors and fractile values for intermediate values
of StD[d/t] and are given in Table 3-8.  The polynomial equa-
tions are curve fits based on the calibrated factors given in
Table 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 assumed
that the standard deviation in the length measurement is less
than 20 times the standard deviation in the depth measurement.
Table 3-3  Standard deviation and confidence level
Relative sizing
accuracy
Confidence level
80% (0.80)90% (0.90)
Exact ±  (0. 0 of t)StD[d/t] = 0.00StD[d/t] = 0.00
±  0.05 of tStD[d/t] = 0.04StD[d/t] = 0.03
±  0.10 of tStD[d/t] = 0.08StD[d/t] = 0.06
±  0.20 of tStD[d/t] = 0.16StD[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 mm
Absolute sizing
accuracy
Confidence level
80% (0.80)90% (0.90)
Exact ±  (0 mm)StD[d/t] = 0.000StD[d/t] = 0.000
±  0.25 mmStD[d/t] = 0.043StD[d/t] = 0.034
±  0.5 mmStD[d/t] = 0.087StD[d/t] = 0.068
±  1.0 mmStD[d/t] = 0.174StD[d/t] = 0.135
Table 3-5  Standard deviation and confidence level,
t = 12.7 mm
Absolute sizing
accuracy
Confidence level
80% (0.80)90% (0.90)
Exact ±  (0 mm)StD[d/t] = 0.000StD[d/t] = 0.000
±  0.25 mmStD[d/t] = 0.022StD[d/t] = 0.017
±  0.5 mmStD[d/t] = 0.043StD[d/t] = 0.034
±  1.0 mmStD[d/t] = 0.087StD[d/t] = 0.068
Table 3-6  Standard deviation and confidence level,
t =19.05 mm
Absolute sizing
accuracy
Confidence level
80% (0.80)90% (0.90)
Exact ±  (0 mm)StD[d/t] = 0.000StD[d/t] = 0.000
±  0.25 mmStD[d/t] = 0.014StD[d/t] = 0.011
±  0.5 mmStD[d/t] = 0.029StD[d/t] = 0.023
±  1.0 mmStD[d/t] = 0.058StD[d/t] = 0.045
Table 3-7  Partial safety factor and fractile value
Inspection sizing
accuracy, StD[d/t] εd
Safety Class
LowNormalHigh
(exact) 0.000.0 γd = 1.00 γd = 1.00 γd = 1.00
 0.040.0 γd = 1.16 γd = 1.16 γd = 1.16
 0.081.0 γd = 1.20 γd = 1.28 γd = 1.32
 0.162.0 γd = 1.20 γd = 1.38 γd = 1.58DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 13
The variation of the partial safety factors εd and γd with
StD[d/t] are shown in Fig.3-2 and Fig.3-3:
Figure 3-2
Partial safety factor γd with StD[d/t].
Figure 3-3
Safety factor εd with StD[d/t].
3.4  Circumferential corrosion
Partial safety factors factor γmc and η are given in Table 3-9 for
a 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 inspection
accuracy.
3.5  Usage factors for longitudinal stress
The usage factors for longitudinal stress are given in
Table 3-10
3.6  System effect
The 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 defect
will also govern the reliability level of the pipeline for failure
due to corrosion.  In the case of several corrosion defects each
with approximately the same allowable corroded pipe pres-
sure, or a pipeline with a large number of corrosion defects, the
system effect must be accounted for when determining the reli-
ability level of the pipeline.  Adding the failure probability of
each defect will conservatively assess the system effect.
3.7  Supplementary material requirements
The safety factors in Table 3-11 and Table 3-12 can be used if
the 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 be
used if it is explicitly documented that the supplementary
material requirements are fulfilled.
Table 3-8  Polynomial Equations for Partial Safety Factor and
Fractile Value, see Table 3-7
Substitute “a” with “StD[d/t]”
Safety
Class
γd and εd Range
Low
Normal
High
(all)
a d 0 . 4 0 . 1 + = γ 04 . 0 < a
2
5 . 37 5 . 5 1 a a d − + = γ 08 . 0 04 . 0 < ≤ a
2 . 1 = d γ 16 . 0 08 . 0 ≤ ≤a
2
9 . 13 6 . 4 1 a a d − + = γ 16 . 0 ≤ a
2
1 . 4 3 . 4 1 a a d − + = γ 16 . 0 ≤ a
0 = d ε
2
2 . 104 5 . 37 33 . 1 a a d − + − = ε
04 . 0 ≤ a
16 . 0 04 . 0 ≤ <a
0.9
1
1.1
1.2
1.3
1.4
1.5
1.6
1.7
00.020.040.060.080.10.120.140.160.18
StD [d/t]
d
Safety Class LOW
Safety Class NORMAL
Safety Class HIGH
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2.2
00.020.040.060.080.10.120.140.160.18
StD [d/t]
d
Table 3-9  Partial safety factors γmc
and η
Safety ClassFactor γmc Factor η
Low γmc = 0.81 η = 0.96
Normal γmc = 0.76 η = 0.87
High γmc = 0.71 η = 0.77
Table 3-10  Usage factors ξ
Safety ClassUsage Factor ξ
Low ξ = 0.90
Normal ξ = 0.85
High ξ = 0.80
Table 3-11  Partial safety factor γm for pipelines with
supplemtary material requirements
Inspection method
Safety Class
LowNormalHigh
Relative (e.g. MFL) γm = 0.82 γm = 0.77 γm = 0.73
Absolute (e.g. UT) γm = 0.85 γm = 0.80 γm = 0.75
Table 3-12  Partial safety factors γmc
and η for pipelines with
supplementary material requirements
Safety ClassFactor γmc Factor η
Low γmc = 0.85 η = 1.00
Normal γmc = 0.80 η = 0.90
High γmc = 0.75 η = 0.80DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 14 see note on front cover
4.  Assessment of a Single Defect (Part A)
4.1  Requirements
Isolated metal loss defects are  to be individually assessed as
single defects, see Fig.4-1. 
Adjacent defects can interact to produce a failure pressure that
is lower than the individual failure pressures of the isolated
defects treated as single defects.  For the case where interaction
occurs, the single defect equation is no longer valid and the
procedure given in Sec.5 must be applied.  Fig.4-2 shows the
key dimensions for defect interaction.
A defect can be treated as an  isolated defect, and interaction
with other defects need not be considered, if either of the fol-
lowing conditions is satisfied:
1)The circumferential angular spacing between adjacent
defects, φ (degrees):
2)Or, the axial spacing between adjacent defects, s:
4.2  Longitudinal corrosion defect, internal pressure
loading only
4.2.1  Acceptance equation
The allowable corroded pipe pressure of a single metal loss
defect subject to internal pressure loading is given by the fol-
lowing acceptance equation. The acceptance equation has not
been validated for defects dimensions where the breadth (cir-
cumferential extent) of the defect exceeding the length of the
defect.
where:
If  γd (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 thickness
is not accepted.
4.2.2  Alternative applications
The form of the acceptance equation is made to determine the
acceptable operating pressure for a measured corrosion defect
in a pipeline. The equation can be re-arranged to determine the
acceptable measured defect size for a specified operational
pressure. 
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 “γd (d/t)* ≥ 1 then pcorr = 0” considers the
confidence in the sizing of the defect depth, and can also be
expressed 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 acceptable
defect depth:
a)Measured defect depth shall not exceed 85% of the wall
thickness, i.e. minimum remaining wall thickness ≥ 15%
of the (nominal) wall thickness.
b)The measured defect depth  plus the uncertainty in the
defect sizing can not exceed the wall thickness, with the
reliability level applicable for the defect, identified by the
safety or location class.
The maximum acceptable measured defect depths are depend-
ent on the inspection method, sizing capabilities and safety or
location 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 a
10mm wall thickness pipeline the minimum requirement may
be only 1.5 mm.  Special attention should be given to these
defects, both in term of reliability of the inspection methods
and potential further growth.
4.3  Longitudinal corrosion defect, internal pressure
and superimposed longitudinal compressive stresses
This 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 longitudinal
corrosion defect subject to internal pressure and longitudinal
compressive stresses can be estimated using the following pro-
cedure:
D
t
360 > φ
Dt s 2 >
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
Q
t d
t d
t D
f t
p
d
d u
m corr
* ) / (
1
* ) / ( 1 2
γ
γ
γ
2
31 . 0 1 ⎟
⎠
⎞
⎜
⎝
⎛
+ =
Dt
l
Q
] / [ StD ) / ( )* / ( t d t d t d d meas ε + =
()
] / [ StD
/
1
/ 1 1
) / (
0
0
,
t d
Q
p p
p p
t d d
oper
oper
d
acc meas ⋅ −
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
= ε
γ
Table 4-1  Maximum acceptable measured depth, selected
examples
Safety
Class
Inspection
method Accuracy
Conf.
level
Max acceptable
measured depth
NormalMFL+/- 5%80%0.86 t
1)
NormalMFL+/- 10%80%0.70 t
HighMFL+/- 10%80%0.68 t
NormalMFL+/- 20%80%0.41 t
1)
 Limited to maximum 0.85 t, see a) above.
STEP 1Determine 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:
() t D
f t
p u
m −
=
2
0 γ
] / [ StD / 1 ) / ( ,
t d t d d d acc meas
ε γ − ≤
() t t D
FX
A
−
=
π
σ
() t t D
4M
2
Y
B
−
=
π
σDET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 15
4.4  Circumferential corrosion defects, internal pres-
sure and superimposed longitudinal compressive
stresses
The acceptance equation given below is not valid for full cir-
cumference corrosion defects with a longitudinal length
exceeding 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
Single defect dimensions
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 1Determine 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:
B A L σ σ σ + =
()
()
1 ,
* ) / (
1
* ) / ( 1 2
H
Q
t d
t d
t D
f t
p
d
d u
m comp corr
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
γ
γ
γ
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
+
=
Q
t d
t d
A
A f
H
d
d
r
m
r u
L
* ) / (
1
* ) / ( 1
2
1
1
1
1
γ
γ
ξ
γ
ξ
σ
⎟
⎠
⎞
⎜
⎝
⎛
− = θ
t
d
Ar
1
The combined nominal longitudinal stress is: 
STEP 2If 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 η fy, in tension or in
compression.  The longitudinal pipe wall stress
shall include the effect of all loads, including the
pressure.
where: σL-nom is the longitudinal stress in the
nominal pipe wall.
() t t D
FX
A
−
=
π
σ
() t t D
4M
2
Y
B
−
=
π
σ
B A L σ σ σ + =
() ()
⎟ ⎟
⎟
⎟
⎟
⎠
⎞
⎜ ⎜
⎜
⎜
⎜
⎝
⎛
−
⎟
⎟
⎟
⎟
⎠
⎞
⎜
⎜
⎜
⎜
⎝
⎛
−
+
−
=
t D
f t
A
A f
t D
f t
p u
mc
r
mc
r u
L
u
mc circ corr
2
,
1
2
1
1
1
2
min ,
γ
ξ
γ
ξ
σ
γ
⎟
⎠
⎞
⎜
⎝
⎛
− = θ
t
d
Ar
1
() ) / ( 1 t d f y nom L − ≤ − η σ
t
t
d
A
l
c
Ac
d
t
t
d
A
l
c
Ac
dDET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 16 see note on front cover
Figure 4-2
Interacting defect dimensions
5.  Assessment of Interacting Defects
(Part A)
5.1  Requirements
The interaction rules are strictly valid for defects subject to
only 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 these
interaction rules may be non-conservative for other loading
conditions. The minimum information required comprises:
—The angular position of each defect around circumference
of 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 estimate
The partial safety factors for interacting defects have not been
derived from an explicit probabilistic calibration.  The partial
safety factors for a single defect subject to internal pressure
loading 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 all
combinations of adjacent defects, are considered in order to
determine 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 the
corroded 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
φ
STEP 1For 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 2The 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 3Construct a series of axial projection lines with a
circumferential angular spacing of:
 (degrees)
STEP 4Consider 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 5Where 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  =  1…N
where:
If γd
(d/t)* ≥ 1 then pi
 = 0.
50 . Dt 25 . Dt
D
t
Z 360 =
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
i
i d
i d u
m i
Q
t d
t d
t D
f t
p
* ) / (
1
* ) / ( 1 2
γ
γ
γ
Q
l
Dt
i
i
=+
⎛
⎝
⎜
⎞
⎠
⎟ 1031
2
.
] / [ StD ) / ( )* / ( t d t d t d d meas i i
ε + =DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 17
Guidance note:
Steps 7 to 9 estimate the allowable corroded pipe pressure of all
combinations of adjacent defects.  The allowable corroded pipe
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.
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
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 the
applied measurement uncertainty need to be assessed and docu-
mented it the reduced uncertainty is to be used. In cases where
the conditions are not known it is recommended to assume fully
correlated depth measurements.
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
Figure 5-1
Corrosion 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  =  1…N
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 =
1…N
where:
() llls nmmii
in
im
=++
=
=−
∑
1
d
dl
l
nm
ii
in
im
nm
= =
=
∑
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
nm
nm d
nm d u
m nm
Q
t d
t d
t D
f t
p
* ) / (
1
* ) / ( 1 2
γ
γ
γ
2
31 . 0 1 ⎟
⎠
⎞
⎜
⎝
⎛
+ =
Dt
l
Q nm
nm
] / [ StD ) / ( )* / ( t d t d t d nm d meas nm nm ε + =
If γd
(d/t)* ≥ 1 then pcorr = 0.
Note that εd and γd 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
i i
m i
n i
nm
l
t d l
t d
/ StD
/ StD
∑
=
=
=
) , ,... , min( 2 1 nm N corr
p p p p p =
  l
d
t
< 0.1t  DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 18 see note on front cover
Figure 5-2
Projection of circumferentially interacting defects
Figure 5-3
Projection 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 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 19
Figure 5-4
Projection of overlapping internal and external defects onto a single projection line and the formation of a composite defect
Figure 5-5
Combining interacting defects
d
1
li
Projection Line
Section Through Projection Line
d
2
2 1 d d d i
+  =
 
sn  ln  lm  ln+1
dm
sm-1
(  )  l  l  l  s
nm  m  i  i
i  n
i  m
=  +  +
=
=  −
∑
1
d
d  l
l
nm
i i
i n
i m
nm
=
=
=
∑
lnm
dn+1
dn DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 20 see note on front cover
Figure 5-6
Example 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 2Defect 3Defect 4DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 21
6.  Assessment of Complex Shaped Defects
(Part A)
6.1  Requirements
This 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.  The
length 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 circumference
for that axial length (i.e. a river bottom profile of the
defect).
2)The length of the profile must include all material between
the start and end of the complex shaped defect.
6.2  Allowable corroded pipe pressure estimate
The partial safety factors for a complex shaped defect have not
been derived from an explicit probabilistic calibration.  The
partial safety factors for a single defect subject to internal pres-
sure loading have been used. 
The allowable corroded pipe pressure of a complex shaped
defect can be estimated using the following procedure:
Guidance note:
The principle underlying the complex shaped defect method is to
determine whether the defect  behaves as a  single irregular
‘patch’, 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 defect
is divided into a number of increments based on depth.
At each depth increment the corrosion defect is modelled by an
idealised ‘patch’ containing a number of idealised ‘pits’.  The
‘patch’ is the material loss shallower than the given increment
depth.  The ‘pits’ are defined by the areas which are deeper than
the increment depth, see Fig.6-1 and Fig.6-2. The allowable cor-
roded pipe pressure of the ‘pits’ within the ‘patch’ is estimated
by considering an equivalent pipe of reduced wall thickness.  The
capacity (failure pressure) of the equivalent pipe is equal to the
capacity of the ‘patch’.
The idealised ‘pits’ in the equivalent pipe are assessed using the
interacting defect method (see Sec.5).
The estimated allowable corroded pipe pressure at a given depth
increment, 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 total
length and average depth.
The procedure is repeated for all depth increments in order to
determine the minimum predicted allowable corroded pipe pres-
sure.  This is the allowable corroded pipe pressure of the complex
shaped defect.
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
Guidance note:
Note that εd and γd 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 the
applied measurement uncertainty need to be assessed and docu-
mented if the reduced uncertainty is to be used. In cases where
the conditions are not known, it is recommended to assume fully
correlated depth measurements.
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
STEP 1Calculate the average depth (dave) of the complex
shaped defect as follows:
STEP 2Calculate the allowable corroded pipe pressure of
the total profile (ptotal
), using dave and ltotal
 in the
single defect equation:
where:
d
A
l
ave
total
=
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
total
ave d
ave d u
m total
Q
t d
t d
t D
f t
p
* ) / (
1
* ) / ( 1 2
γ
γ
γ
If γd
(dave/t)* ≥ 1 then ptotal
 = 0.
Fully correlated depth measurements:
STEP 3Divide 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 4Calculate the average depth of an idealised
‘patch’ as follows (see Fig.6-2):
STEP 5Calculate 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 γd
(dpatch/t)* ≥ 1 then ppatch = 0.
2
31 . 0 1 ⎟ ⎟
⎠
⎞
⎜ ⎜
⎝
⎛
+ =
Dt
l
Q total
total
] / [ StD ) / ( )* / ( t d t d t d ave d meas ave ave ε + =
] / [ StD ] / [ StD t d t dave =
total
patch
patch
l
A
d =
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
total
patch d
patch d u
m patch
Q
t d
t d
t D
f t
p
* ) / (
1
* ) / ( 1 2
γ
γ
γ
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
total
patch
patch u
patch cap
Q
t d
t d
t D
f t
p
) / (
1
) / ( 1 2
09 . 1 ,
2
31 . 0 1 ⎟ ⎟
⎠
⎞
⎜ ⎜
⎝
⎛
+ =
Dt
l
Q total
total
] / [ StD ) / ( )* / ( t d t d t d patch d meas patch patch ε + =DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 22 see note on front cover
Guidance note:
Note that εd and γd 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 the
applied measurement uncertainty need to be assessed and docu-
mented if the reduced uncertainty is to be used. In cases where
the conditions are not known, it is recommended to assume fully
correlated depth measurements.
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
Guidance note:
Steps 10 to 12 estimate the allowable corroded pipe pressures of
all combinations of adjacent defects.  The allowable corroded
pipe 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.
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
Guidance note:
The formula for StD[de,nm/t] assumes fully correlated depth
measurements.  In the case that the measurements are not fully
correlated the uncertainty is reduced. The estimate and the effect
of the applied measurement uncertainty need to be assessed and
documented if the reduced uncertainty is to be used. In cases
where the conditions are not known it is recommended to assume
fully correlated depth measurements.
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
Fully correlated depth measurements:
STEP 6For 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 =  1…N
STEP 7Estimate 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 8The average depth of each ‘pit’ is corrected for
the effective thickness (te) using:
STEP 9Calculate 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 = 1…N
where:
If γd
(dei
/te)* ≥ 1 then pi
 = 0.
] / [ StD ] / [ StD t d t d patch =
i
pit i
i
l
A
d
,
=
()
 
) 09 . 1 ( 2 ,
,
patch cap u
patch cap
e
p f
D p
t
+ ⋅
⋅
=
() e i ei
t t d d − − =
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
i
e ei d
e ei d
e
u e
m i
Q
t d
t d
t D
f t
p
* ) / (
1
* ) / ( 1 2
γ
γ
γ
2
31 . 0 1
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
e
i
i
Dt
l
Q
] / [ StD ) / ( )* / ( t d t d t d d meas ei e ei
ε + =
STEP 10Calculate 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  =  1…N
STEP 11Calculate 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 12Calculate 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 =
1…N
where:
If γd
(de,nm/te)* ≥ 1 then pnm = 0.
Note that εd and γd 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:
() ∑
− =
=
+ + =
1 m i
n i
i i m nm s l l l
nm
m i
n i
i ei
nm e
l
l d
d
∑
=
= = ,
()
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
nm
e nm e d
e nm e d
e
u e
m nm
Q
t d
t d
t D
f t
p
* ) / (
1
* ) / ( 1 2
,
,
γ
γ
γ
2
31 . 0 1
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
e
nm
nm
Dt
l
Q
] / [ StD ) / ( )* / ( , , ,
t d t d t d nm e d meas nm e e nm e ε + =
[]
[]
nm
ei i
m i
n i
nm e
l
t d l
t d
/ StD
/ StD ,
∑
=
= =
) , , , ,... , min( 2 1 total patch nm N corr p p p p p p p j
=DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 23
Figure 6-1
Subdivision of complex shape into idealised 'patch' and 'pits'
Figure 6-2
Definition of Apatch and Apit for subdivision of complex shape into idealised 'patch' and 'pits'
STEP 14Repeat the Steps 4 to 13 for the next interval of
depth increment (dj
) until the maximum depth of
corrosion profile has been reached.
STEP 15Calculate 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 16The 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
CurrentDepthIncrement,  d
d
j
j
Apatch
Apit
t
CurrentDepthIncrement,  d
d
j
j
Apatch
Apit
tDET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 24 see note on front cover
7.  Allowable Stress Approach (Part B)
7.1  Introduction
The approach given in Part B is based on the ASD (Allowable
Stress Design) format.  The failure pressure (capacity) of the
pipeline with the corrosion defect is calculated, and this failure
pressure is multiplied by a single safety factor based on the
original design factor. 
When assessing corrosion defects, due consideration should be
given to the measurement uncertainty of the defect dimensions
and the pipeline geometry.
7.2  Total usage factor
The usage factor to be applied in determining the safe working
pressure has two components:
The Total Usage Factor (F) to be applied to determine the safe
working pressure should be calculated from:
F = F1F2
8.  Assessment of a Single Defect (Part B)
8.1  Requirements
Isolated metal loss defects are  to be individually assessed as
single defects, see Fig.4-1. 
Adjacent defects can interact to produce a failure pressure that
is lower than the individual failure pressures of the isolated
defects treated as single defects.  For the case where interaction
occurs, the single defect equation is no longer valid and the
procedure given in Sec.9 must be applied.  Fig.4-2 shows the
key dimensions for defect interaction.
A defect can be treated as an  isolated defect, and interaction
with other defects need not be considered, if either of the fol-
lowing conditions is satisfied:
1)The circumferential angular spacing between adjacent
defects, φ:
2)The axial spacing between adjacent defects, s:
8.2  Safe working pressure estimate - Internal pres-
sure only
The safe working pressure of a single defect subject to internal
pressure 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 a
10mm wall thickness pipeline the minimum requirement is
only 1.5 mm.  Special attention should be given to these
defects, both in term of reliability of the inspection methods
and result and potential further growth.
8.3  Safe working pressure estimate - Internal pres-
sure and combined compressive loading
The validation of the method for assessing corrosion defects
subject to internal pressure  and longitudinal compressive
stresses is not as comprehensive as the validation of the
method for assessing corrosion defects under internal pressure
loading 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 subject
to internal pressure and longitudinal compressive stresses can
be 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 1Calculate the failure pressure of the corroded
pipe (Pf
):
(degrees) 360
D
t
> φ
Dt s 0 . 2 >
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
=
tQ
d
t
d
t D
f t
P u
f
1
1
2
where:
STEP 2Calculate the safe working pressure of the cor-
roded pipe (Psw):
Measured defects depths exceeding 85% of the
wall thickness is not accepted.
STEP 1Determine 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 . 0 1 ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
l
Q
f sw P F P =
() t t D
FX
A
−
=
π
σ
() t t D
4M
2
Y
B
−
=
π
σ
B A L σ σ σ + =
1 σ σ > LDET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 25
If the above condition is satisfied then Step 4 can be neglected.
STEP 3Calculate the failure pressure of the single corro-
sion defect under internal pressure only, using the
following equation:
where:
STEP 4Calculate the failure pressure for a longitudinal
break, including the correction for the influence
of compressive longitudinal stress (Fig.8-2):
where:
STEP 5Determine the failure pressure of a single corro-
sion defect subjected to internal pressure loading
combined with compressive longitudinal
stresses:
STEP 6Calculate the safe working pressure of the cor-
roded pipe (Psw):
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
− =
tQ
d
t
d
fu
1
1
5 . 0 1 σ
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
=
tQ
d
t
d
t D
f t
P u
press
1
1
2
2
31 . 0 1 ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
l
Q
()
1
1
1
2
H
tQ
d
t
d
t D
f t
P u
comp
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
=
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
+
=
tQ
d
t
d
A
A f
H
r
r u
L
1
1
2
1
1
1
1
1
σ
⎟
⎠
⎞
⎜
⎝
⎛
− = θ
t
d
Ar
1
) , min comp press f
P (P     P =
f sw P F P =DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 26 see note on front cover
Figure 8-1
Range of superimposed longitudinal and/or bending loads that will not influence the failure pressure
Figure 8-2
Influence of applied loads on the failure mode of a corrosion defect
9.  Assessment of Interacting Defects
(Part B)
9.1  Requirements
The interaction rules are strictly valid for defects subject to
only 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 these
interaction rules may be non-conservative for other loading
conditions.  The methods given in Sec.8 for assessing corro-
sion defects under combined loads are only valid for single
defects.
The minimum information required comprises:
—The angular position of each defect around circumference
of 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 estimate
The safe working pressure can be estimated from the following
procedure:
σHOOP
σLONGITUDINAL 
Arfu
σ1 σ2
External axial or bending stresses
do not influence failure pressure, if
greater than σ1 and less than σ2
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
=
tQ
d
t
d
fu
1
1
FAILURE SURFACE
σHOOP
σLONGITUDINAL 
Arfu
σ1 σ2
External axial or bending stresses
do not influence failure pressure, if
greater than σ1 and less than σ2
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
=
tQ
d
t
d
fu
1
1
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
=
tQ
d
t
d
fu
1
1
FAILURE SURFACE
σHOOP
σLONGITUDINAL
CIRCUMFERENTIAL BREAK
LONGITUDINAL BREAK
REDUCED PRESSURE
LONGITUDINAL BREAK
BUCKLING / WRINKLING
FAILURE SURFACEDET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 27
Guidance note:
Within the colony of interacting defects, all single defects, and all
combinations of adjacent defects, are considered in order to
determine 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 to
the corroded area of each defect within the combined defect).
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
Guidance note:
Steps 7 to 9 estimate the failure pressures of all combinations of
adjacent 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  Requirements
This method must only be applied to defects subjected to inter-
nal pressure loading only.
STEP 1For 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 2The 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 3Construct a series of axial projection lines with a
circumferential angular spacing of:
 
(degrees)
STEP 4Consider 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 5Where 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 6Calculate the failure pressures (P1, P2 … PN) of
each defect, to the Nth defect, treating each defect,
or composite defect, as a single defect:
  
i  =  1…N
where:
STEP 7Calculate 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  =  1…N
Dt 0 . 5 Dt 5 . 2
D
t
Z 360 =
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
=
i
i
i
u
i
tQ
d
t
d
t D
f t
P
1
1
2
2
31 . 0 1 ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
l
Q i
i
() ∑
− =
=
+ + =
1 m i
n i
i i m nm s l l l
STEP 8Calculate 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 9Calculate 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 10The 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 11Calculate the safe working pressure (Psw) of the
interacting defects on the current projection line:
STEP 12The 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 13Repeat Steps 3 to 12 for the next section of the
corroded pipeline.
nm
m i
n i
i i
nm
l
l d
d
∑
=
=
=
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
=
nm
nm
nm
u
nm
tQ
d
t
d
t D
f t
P
1
1
2
2
31 . 0 1 ⎟
⎠
⎞
⎜
⎝
⎛
+ =
Dt
l
Q nm
nm
) , ,... , ( 2 1 nm N f
P P P P MIN P =
f sw P F P =DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 28 see note on front cover
The minimum information required comprises:
1)A length and depth profile for the complex shape.  The
length must be the axial length along the axis of the pipe.
The depth, at a given axial length along the defect, should
be the maximum depth around the circumference for that
axial length (i.e. a river bottom profile of the defect).
2)The length of the profile must include all material between
the start and end of the complex shaped defect.
10.2  Safe working pressure estimate
The safe working pressure of a complex shaped defect can be
estimated from the following procedure:
Guidance note:
The principle underlying the complex shaped defect method is to
determine whether the defect behaves as a single irregular
‘patch’, or whether local ‘pits’ within the patch dominate the fail-
ure.  Potential interaction between pits is also to be assessed.
A progressive depth analyses is performed.  The corrosion defect
is divided into a number of increments based on depth.
At each depth increment the corrosion defect is modelled by an
idealised ‘patch’ containing a number of idealised ‘pits’.  The<, BR>‘patch’ is the material loss shallower than the given increment
depth.  The ‘pits’ are defined by the areas which are deeper than
the increment depth, see Fig.6-1 and Fig.6-2.  The failure pres-
sure of the ‘pits’ within the ‘patch’ is estimated by considering an
equivalent pipe of reduced wall thickness.  The failure pressure
of the equivalent pipe is equal to the failure pressure of the
‘patch’.
The idealised ‘pits’ in the equivalent pipe are assessed using the
interacting defect method (see Sec.9).
The estimated failure pressure at a given depth increment, is the
minimum of the failure pressure of the ‘patch’, the idealised
‘pits’, and the failure pressure of the total corroded area based on
its total length and average depth.
The procedure is repeated for all depth increments in order to
determine the minimum predicted failure pressure.  This is the
failure pressure of the complex shaped defect.
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
STEP 1Calculate the average depth (dave) of the complex
shaped defect as follows:
STEP 2Calculate the failure pressure of the total profile
(Ptotal
), using dave and ltotal
 in the single  defect
equation:
where:
total
ave
l
A
d =
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
=
total
ave
ave
u
total
tQ
d
t
d
t D
f t
P
1
1
2
2
31 . 0 1 ⎟ ⎟
⎠
⎞
⎜ ⎜
⎝
⎛
+ =
Dt
l
Q total
total
STEP 3Divide 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 4Calculate the average depth of an idealised
‘patch’ as follows (see Fig.6-2):
STEP 5Calculate the failure pressure of the idealised
‘patch’ (Ppatch), using ltotal
 and dpatch in the sin-
gle defect equation:
where:
STEP 6For 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  =  1…N
STEP 7Estimate the effective thickness of an ‘equiva-
lent’ pipe with the same failure pressure as the
‘patch’, (Ppatch), as calculated in Step 5 (see
Fig.6-1).
STEP 8The average depth of each ‘pit’ is corrected for
the effective thickness (te) using:
STEP 9Calculate the failure pressure of all individual
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:
total
patch
patch
l
A
d =
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
=
total
patch
patch
u
patch
tQ
d
t
d
t D
f t
P
1
1
2
2
31 . 0 1 ⎟ ⎟
⎠
⎞
⎜ ⎜
⎝
⎛
+ =
Dt
l
Q total
total
i
pit i
i
l
A
d
,
=
() patch u
patch
e
P f
D P
t
+ ⋅
⋅
=
) 09 . 1 ( 2
() e i ei
t t d d − − =
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
=
i e
ei
e
ei
e
u e
i
Q t
d
t
d
t D
f t
P
1
1
2DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 29
Guidance note:
Steps 10 to 12 estimate the failure pressures of all combinations
of adjacent 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
---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---
 
11.  References
/1/
where:
STEP 10Calculate 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  =  1…N
STEP 11Calculate 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 failure 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:
where:
STEP 13 The failure pressure for the current depth incre-
ment is taken as the minimum of all the failure
pressures from above:
STEP 14Repeat the Steps 4 to 13 for the next interval of
depth increment (dj
) until the maximum depth of
corrosion profile has been reached.
STEP 15Calculate the failure pressure according to the
single defect equation in Sec.8.2, Step 1, using
the maximum defect depth and the total length of
the defect.
2
31 . 0 1
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
e
i
i
Dt
l
Q
() ∑
− =
=
+ + =
1 m i
n i
i i m nm s l l l
nm
m i
n i
i ei
nm e
l
l d
d
∑
=
= = ,
()
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
=
nm e
nm e
e
nm e
e
u e
nm
Q t
d
t
d
t D
f t
P
,
,
1
1
2
2
31 . 0 1
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
e
nm
nm
Dt
l
Q
) , , ,... , min( 2 1 total patch nm N f
P P P P P P P j
=
STEP 16The failure pressure of the complex shaped defect
(Pf
) should be taken as the minimum of that from
all of the depth intervals, but not less than the fail-
ure pressure for a single defect calculated in Step
15.
STEP 17Calculate the safe working pressure (Psw) of the
complex shaped defect:
f sw P F P =DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 30 see note on front cover
ASME Code For Pressure Piping, B31 Liquid Petroleum
Transportation Piping Systems, ASME B31.4 - 1989 Edition.
/2/
ASME Code For Pressure Piping, B31 Gas Transmission And
Distribution Piping Systems, ASME B31.8 - 1992 Edition,
Incorporating ASME B31.8b - 1994 Addenda.
/3/
Bjørnøy, O.H., Fu, B., Sigurdsson, G., Cramer, E.H.,
Ritchie,D., (1999), Introduction and background to DNV RP-
F101 "Corroded Pipelines", ISOPE 1999, Brest, France.
/4/
Bjørnøy, O.H., Sigurdsson,  G., Cramer, E.H., Fu, B.,
Ritchie,D., (1999), Introduction to DNV RP-F101 "Corroded
Pipelines", OMAE 1999, Newfoundland, Canada.
/5/
British Standard Code of Practice for Pipelines, BS8010: 1989,
British Standards Institute, 1989.
/6/
BSI STANDARDS; Guide  on Methods for Assessing the
Acceptability of Flaws in Fusion Welded Structures, BS7910,
British Standards Institute, London, UK, 1999.
/7/
Canadian Standards Association (CSA) 1994, Oil and Gas
Pipeline Systems, Z662-94, Rexdale, Ontario.
/8/
DNV Offshore Standard,  DNV-OS-F101, Submarine Pipeline
Systems, Det Norske Veritas, 2000.
/9/
DNV “Rules for Submarine Pipeline Systems” (DNV'96), Det
Norske Veritas, 1996, Norway.
/10/
IGE/TD/1 Edition 3: 1993; Steel Pipelines for High Pressure
Gas Transmission, Recommendations on Transmission and
Distribution Practice, Institute of Gas Engineers, Communica-
tion 1530, 1993.
/11/
ISO/DIS 13623 "Petroleum and Natural Gas Industries - Pipe-
line Transportation Systems", 1997.
/12/
MILNE,I., AINSWORTH,R.A., DOWLING,A.R. and STEW-
ART,A.T.; Assessment of the Integrity of Structures Contain-
ing Defects - Revision 3, Central Electricity Generation Board
Report R/H/R6, Revision 3, 1987.
/13/
MILLER,A.G.; Review of Test Results for Ductile Failure
Pressure of Cracked Spherical and Cylindrical Pressure Ves-
sels, Central Electricity Generating Board (CEGB), TPRD/B/
0489/N84, July 1984.
/14/
ROSENFELD, M. J., and KIEFNER, J. F.; Proposed Fitness-
for-Purpose Appendix to ASME B31 Code for Pressure Pip-
ing, Section B31.8, Gas Transmission and Distribution Sys-
tems, Final Report to ASME, January 13, 1995.
/15/
Specification for Line Pipe, Exploration and Production
Department, American Petroleum Institute, API Specification
5L, Forty First Edition, April 1, 1995.
/16/
Sigurdsson G., Cramer E.H., Bjørnøy O.H., Fu B., RitchieD.,
(1999), Background to DNV RP-F101 "Corroded Pipelines",
OMAE 1999, Newfoundland, Canada.
/17/
Bjørnøy, O.H., Sigurdsson, G., Marley, M.J.,(2001), Back-
ground and Development of DNV-RP-F101 "Corroded Pipe-
lines", ISOPE 2001, Stavanger, Norway.
/18/
Cosham, A., Hopkins, P., 2002, “The Pipeline Defect Assess-
ment Manual” IPC02-27067, Proceedings of IPC 2002, Inter-
national Pipeline Conference, ASME, Calgary, Canada.
/19/
Cosham, A., Hopkins, P., 2003, “The Assessment of corrosion
in pipelines – Guidance in the pipeline defect assessment man-
ual, (PDAM)”, International Colloquium- Reliability of High
Pressure Pipelines, Prague, Czech Republic.DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 31
APPENDIX A
EXAMPLES FOR PART A
A.1  Single defect assessment
Example 1
This example is for the assessment of an isolated corrosion
defect under internal pressure loading (see Sec.4.2), using rel-
ative depth measurements.
The dimensions and material properties are summarised as fol-
lows:
The defect dimensions have been taken from the results of an
internal inspection using a magnetic flux intelligent pig.  The
inspection accuracy quoted by the inspection tool provider is
that the defect depth will be reported with a ±10% tolerance.
This sizing accuracy is quoted with a confidence level of 80%.
The maximum allowable operating pressure is 150 bar.
The Safety Class is assumed to be Normal.
StD[d/t] = 0.08 (from Table 3-3)
(alt. calc: StD[d/t] = acc_rel / NORMSINV(0.5+conf/2) =
0.1 / NORMSINV(0.5 + 0.8/2) = 0.0780)
Taking the partial safety factors from Table 3-2 and Table 3-7
γm =0.74
γ
d =1.28
εd =1.0
Using the procedure for assessing single defects given in
Sec.4.2.
fu = SMTS
The allowable corroded pipe pressure is 15.94 N/mm2
(159.4bar).  Therefore, the corrosion defect is acceptable, at
the current time, for the maximum allowable operating pres-
sure of 150 bar.
Example 2
This example is for the assessment of an isolated corrosion
defect under internal pressure loading (see Sec.4.2), using
absolute depth measurements.
The dimensions and material properties are summarised as fol-
lows:
The defect dimensions have been taken from the results of an
internal inspection using an ultrasonic intelligent pig.  The
inspection accuracy quoted by the inspection tool provider is
that the defect depth will be reported with a ±1.0mm tolerance.
This sizing accuracy is quoted with a confidence level of 80%.
The maximum allowable operating pressure is 150 bar.
The Safety Class is assumed to be Normal.
Calculation of standard deviation (this example is also
included in Table 3-6):
StD[d/t] =  acc_abs / (t · NORMSINV(0.5 + conf/2))
=  1.0 / (19.1*NORMSINV(0.5 + 0.8/2)) = 0.058
(The more detailed calculation of the standard deviation would
be 0.0422, see app. C or the 1999 version of the RP)
Taking the partial safety factors from Table 3-8
Using the procedure for assessing single defects given in
Sec.4.2 .
The allowable corroded pipe pressure is 17.40N/mm2
(174.0bar).  Therefore, the corrosion defect is acceptable, at
the current time, for the maximum allowable operating pres-
sure of 150 bar.
Example 3
This example is for the assessment of an isolated longitudinal
corrosion defect under internal pressure loading and superim-
posed longitudinal compressive stresses (see Sec.4.3 ).
The dimensions and material properties are summarised as fol-
lows:
The pipe is subject to a compressive longitudinal stress of
magnitude 200 N/mm2.
Outside diameter=812.8 mm
Wall thickness=19.10 mm
SMTS=530.9 N/mm2  (X65)
Defect length (max)=200 mm
Defect depth (max)=25% of wall thickness
Outside diameter=812.8 mm
Wall thickness=19.10 mm
3412 . 1 31 . 0 1
2
= ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
l
Q
0.33 0.08 0 . 1 25 . 0 )* / ( = × + = t d
()
() 2
/ 94 . 15
* ) / ( 28 . 1
1
* ) / ( 28 . 1 1 2
74 . 0 mm N
Q
t d
t d
t D
tSMTS
pcorr =
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
SMTS=530.9 N/mm2  (X65)
Defect length (max)=200 mm
Defect depth (max)=4.8 mm (~ 25%)
Outside diameter=219.0 mm
Original wall thickness=14.5 mm
SMTS=455.1 N/mm2  (X52)
Defect length (max)=200.0 mm
Defect width (max)=100.0 mm
Defect depth (max)=62% of wall thickness
2
2
[][] 22 . 1 / StD 9 . 13 / StD 6 . 4 1 2
= − + = t d t d d γ
[][] 49 . 0 / StD 2 . 104 / StD 5 . 37 33 . 1 2
= − + − = t d t d d ε
3412 . 1 31 . 0 1
2
= ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
l
Q
0.2546 0.058 49 . 0 25 . 0 )* / ( = × + = t d
()
() 2
/ 40 . 17
* ) / ( 17 . 1
1
* ) / ( 17 . 1 1 2
77 . 0 m N
Q
t d
t d
t D
tSMTS
pcorr =
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 32 see note on front cover
The defect dimensions have been taken from the results of an
internal inspection using a magnetic flux intelligent pig.  The
inspection accuracy quoted by the inspection tool provider is
that the defect depth will be reported with a ±10% tolerance.
This sizing accuracy is quoted with a confidence level of 80%.
The maximum allowable operating pressure is 150 bar.
The Safety Class is assumed to be Normal.
From Table 3-3 (assuming that the sizing accuracy follows a
Normal distribution).
StD[d/t] = 0.08
Taking the partial safety factors from Tables 3-2, 3-7 and 3-10:
γm = 0.74
γd = 1.28
εd = 1.0
ξ = 0.85
Using the procedure for assessing single defects given in
Sec.4.2.
Where fu = SMTS
Using the procedure given in Sec.4.3.
Step 1
Calculate the nominal longitudinal elastic stresses in the pipe,
based on the nominal pipe wall thickness:
σL = –200   N/mm2
Step 2
Calculate the allowable corroded pipe pressure, including the
correction for the influence of compressive stresses:
The allowable corroded pipe pressure is 3.93 N/mm2
(39.3bar).  This is less than the maximum allowable operating
pressure of 150 bar.  Therefore the pipeline must be downrated
to 39 bar, until the corrosion defect is repaired.
A.2  Interacting defects
Example 4
This example is for a pair of rectangular patches 200 mm and
150 mm in length, respectively, and separated axially by 100
mm.  The longer defect is 20% of the wall thickness deep and
the shorter defect is 30% of the wall thickness deep.
The basic properties required by the assessment are:
The defect dimensions have been taken from the results of an
internal inspection using a magnetic flux intelligent pig.  The
inspection accuracy quoted by the inspection tool provider is
that the defect depth will be reported with a ±10% tolerance.
This sizing accuracy is quoted with a confidence level of 80%.
The maximum allowable operating pressure is 150 bar.
The Safety Class is assumed to be High.
From Table 3-3, (assuming that the sizing accuracy follows a
Normal distribution).
StD[d/t] = 0.08
Taking the partial safety factors from Tables 3-2 and 3-7
γ
m = 0.70
γd = 1.32
εd = 1.0
Using the procedure for assessing interacting defects given in
Sec.5:
The defects should be grouped into axial projections as
described in Steps 1 to 5 of Sec.5.2.
Step 6 is to estimate the failure pressure of both defects, when
treated as isolated defects.  The allowable corroded pipe pres-
sures are 16.47 N/mm2 and 16.19N/mm2 respectively.
Applying the rules for defect interactions in Steps 7 to 9 (Sec.9)
gives:
Assuming that the defect depth measurements are fully corre-
lated:
Taking the partial safety factors from Tables 3-2 and 3-7
γm = 0.70
γd = 1.25
2147 . 2 31 . 0 1
2
= ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
l
Q
0.70 0.08 0 . 1 62 . 0 )* / ( = × + = t d
()
() 2
/ 34 . 8
* ) / ( 28 . 1
1
* ) / ( 28 . 1 1 2
74 . 0 m N
Q
t d
t d
t D
tSMTS
pcorr =
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
1453 . 0 = =
D
c
π
θ
() 9098 . 0 ) / ( 1 = − = θ meas r t d A
2147 . 2 31 . 0 1
2
= ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
l
Q
0.70 0.08 0 . 1 62 . 0 )* / ( = × + = t d
()
4711 . 0
* ) / ( 28 . 1
1
* ) / ( 28 . 1 1
85 . 0 2
74 . 0
1
1
85 . 0
1
1 =
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
×
−
+
=
Q
t d
t d
A
A SMTS
H
r
r
L σ
Outside Diameter =812.8 mm
Original Wall Thickness=20.1 mm
SMTS=530.9 N/mm2  (X65)
Combined length (Step 7)=450 mm
Effective depth (Step 8)=0.19t
()
() 2
1 ,
/ 93 . 3
* ) / ( 28 . 1
1
* ) / ( 28 . 1 1 2
74 . 0 mm N H
Q
t d
t d
t D
tSMTS
p comp corr =
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
−
−
=
[]
[]
0622 . 0
/ StD
/ StD = =
∑
=
=
nm
i i
m i
n i
nm l
t d l
t dDET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 33
εd = 0.60
Step 10 is to select the minimum allowable corroded pipe pres-
sure of the individual and combined defects.  In this case, the
allowable corroded pipe pressure of the combined defect is less
than that of either of the single defects, which indicates that the
defects interact.
The allowable corroded pipe pressure is 15.40 N/mm2 (154.0
bar).
A.3  Complex shaped defect
Example 5
The following worked example  is for an actual corrosion
defect for which the profile has been measured using a depth
micrometer, (measured d and t)
The pipeline geometry and properties are summarised as fol-
lows:
The inspection accuracy quoted by the inspection tool provider
is that the defect depth will be reported with a ±0.1mm toler-
ance.  This sizing accuracy is quoted with a confidence level of
90%.
The maximum allowable operating pressure is 70 bar.
The Safety Class is assumed to be Normal.
The defect profile is shown in Fig.A-1 and the defect depths
are tabulated in Table A-1.  It is assumed that the depth meas-
urements are fully correlated.
As single defect:
Using the procedure for assessing single defects given in
Sec.4, with a total length of 289 mm and maximum depth of
2.8 mm.
Calculation of standard deviation:
StD[d/t] =  acc_abs / (t · NORMSINV(0.5 + conf/2))
=  1.0 / 8.2*NORMSINV(0.5 + 0.9/2) = 0.0105
(The more detailed calculation of the standard deviation would
be 0.0078, see Appendix C or the 1999 version of the RP).
Taking the partial safety factor from Table 3-2.
γm = 0.77
Taking the partial safety factors from Table 3-8.
Allowable Corroded Pipe Pressure=  8.12 N/mm2
When the complex shaped defect is assessed as a single defect,
using the total length and maximum depth, then the allowable
corroded pipe pressure is 8.12 N/mm2.
As single defect with average depth:
Using the procedure for assessing complex shaped defects
given in Sec.6:
Single Defect Solution (Steps 1 to 2)
Step 1 is to calculate the average depth of the defect from the
projected total area loss of the defect. The total projected metal
loss area is calculated to be 421.94 mm2, resulting in an aver-
age depth of 1.46mm for the length of 289 mm.
Step 2 is to estimate the allowable corroded pipe pressure of
the defect from the average depth and the total length.
Assuming that the defect depth measurements are fully corre-
lated:
StD[dave/t] = StD[d/t] = 0.0105
Safety factors as above.
Progressive Depth Analysis (Steps 3 to 15)
The profile was sectioned at 50 levels and the allowable cor-
roded pipe pressure was estimated for each increment.  Fig.A-
2 shows the variation of the allowable corroded pipe pressure
estimate with depth.  The minimum allowable corroded pipe
pressure estimate was 9.19N/mm2 (91.9 bar).  The section
depth was 1.06 mm, which corresponds to the natural division
between patch and pit, which can be seen in Fig.A-1.  The
effect of the relatively distinct change in profile at this depth
produces a sharp change in the estimated allowable corroded
pipe pressure curve, as shown in Fig.A-2.
Assuming that the defect depth measurements are fully corre-
lated:
StD[dpatch/t] = StD[d/t] = 0.0105
Safety factors as above
Steps 6 to 12 are to estimate the allowable corroded pipe pres-
sure of the idealised pits.
Step 9 is to estimate the allowable corroded pipe pressure of all
individual idealised pits.
Step 12 is to estimate the allowable corroded pipe pressure of
the combined defect from n to m.
Step 13 is to estimate the allowable corroded pipe pressure for
the current horizontal step depth from the minimum of the
patch and pit estimates.  In this case the minimum allowable
corroded pipe pressure is from the pit:
Minimum allowable corroded pipe pressure (Step 13) =
9.17N/mm2.
In Step 15 the defect is calculated as a single defect with the
total length and the maximum depth.  The allowable pressure
is calculated as 8.12 N/mm2 (81.2 bar).
Allowable corroded pipe pressure (Step 9)=15.40 N/mm2
Outside diameter=611.0 mm
Wall thickness=8.20 mm
SMTS=517.1 N/mm2  (X60)
Table A-1  Tabulated profile for actual corrosion defect
Length
(mm)
Depth
(mm)
00
28.91
57.81.1
86.71.1
115.61.1
144.51.3
173.41.8
202.32.8
231.22.8
260.11.6
2890
2
2
Allowable Corroded Pipe Pressure=9.52 N/mm2
Patch allowable corroded pipe pressure
(Step 5)=9.99 N/mm2
Patch capacity pressure (Step 5)=14.20 N/mm2
Effective reduced thickness (Step 7)=7.60 mm
[][] 046 . 1 / StD 9 . 13 / StD 6 . 4 1 2
= − + = t d t d d γ
0 . 0 = d εDET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 34 see note on front cover
Step 16 is to estimate the allowable corroded pipe pressure of
the complete defect as the minimum of all the minimum esti-
mates for each horizontal step, i.e. the minimum of all Step 13
results (see Fig.A-2), but not less than the pressure from Step
15.
Analysis of the defect as a complex profile, using the progres-
sive depth method, gives an allowable corroded pipe pressure
estimate of 9.17N/mm2.
The allowable corroded pipe pressure is 9.17 N/mm2
(91.7bar), if it is assumed that the depth measurements are
fully correlated.  Therefore, the corrosion defect is acceptable,
at the current time, for the maximum allowable operating pres-
sure of 70 bar.
Figure A-1 
Profile for actual corrosion defect - example assessment
Figure A-2 
Variations of the estimated failure pressure for actual corrosion defect - Example assessment
050100150200250300
-8
-6
-4
-2
0
Length (mm)
Complex Shaped Corrosion Defect
Pressure Estimates at Depth Increments
9.0
9.1
9.2
9.3
9.4
9.5
9.6
0.00.51.01.52.02.53.0
Depth increment (mm)
Predicted Allowable Corroded Pressure (MPa)
Allowable Corroded Pipe Pressure = 9.19 MPa
Increment depth = 1.06 mmDET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 35
APPENDIX B
EXAMPLES FOR PART B
B.1  Single defect assessment
B.1.1 Example 6
This example is for the assessment of an isolated corrosion
defect under internal pressure loading only (see Sec.8.2).
The dimensions and material properties are summarised as fol-
lows:
Using the procedure for assessing single defects given in
Sec.8.2.
Step 1 - Calculate the failure pressure using:
fu = SMTS
Step 2 - Calculate a safe working pressure based on the factors
of safety, and assuming a design factor of 0.72, gives:
The safe working pressure:
(This compares with a burst pressure of 20.50 N/mm2 from a
full scale test, with measured ultimate tensile strength of
608MPa.  Using the ultimate tensile strength and the capacity
equation including the 1.05 factor this will result in a capacity
prediction of 19.1 N/mm2, a deviation of about 7%).
B.1.2 Example 7
This example is for the assessment of an isolated corrosion
defect under internal pressure and compressive longitudinal
loading (see Sec.8.3).
The dimensions and material properties are summarised as fol-
lows:
The pipe is subject to a compressive longitudinal stress of
magnitude 200 N/mm2.
Using the procedure for assessing single defects given in
Sec.8.3
Step 1 - Calculate the nominal longitudinal elastic stresses in
the pipe, based on the nominal pipe wall thickness:
Step 2 - Assess whether it is necessary to consider the external
loads:
Because σL < σ1, Step 4 cannot be neglected.
Step 3 - Calculate the failure pressure under the influence of
internal pressure loading only:
Step 4 - Calculate the failure pressure for a longitudinal break,
including the correction for  the influence of compressive
stresses:
Outside diameter=812.8 mm
Original wall thickness=19.10 mm
SMTS =530.9 N/mm2 (X65)
Defect length (max)=203.2 mm
Defect depth (max)=13.4 mm
Outside diameter=219.0 mm
Original wall thickness=14.5 mm
SMTS (= fu)=455.1 N/mm2  (X52)
Defect length (max)=200.0 mm
Defect width (max)=100.0 mm
Defect depth (max)=62% of wall thickness
350 . 1 31 . 0 1
2
= ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
l
Q
()
2
/ 87 . 15
1
1
2
mm N
Q t
d
t
d
t D
f t
P u
f
=
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
=
2
/ 28 . 10 ) 72 . 0 )( 9 . 0 ( mm N P P f sw = =
2
/ 200 mm N L − = σ
1453 . 0 = =
D
c
π
θ
9098 . 0 1 = ⎟
⎠
⎞
⎜
⎝
⎛
− = θ
t
d
Ar
2147 . 2 31 . 0 1
2
= ⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
Dt
l
Q
2
1 / 92 . 119
1
1
5 . 0 mm N
tQ
d
t
d
SMTS − =
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
− = σ
2147 . 2 = Q
()
2
/ 01 . 34
1
1
2
mm N
tQ
d
t
d
t D
tSMTS
Ppress =
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
=
7277 . 0
1
1
2
1
1
1
1
1 =
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
+
=
tQ
d
t
d
A
A SMTS
H
r
r
L σ
()
2
1 / 75 . 24
1
1
2
mm N H
tQ
d
t
d
t D
tSMTS
Pcomp =
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−
−
=DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 36 see note on front cover
Step 5 - Calculate the failure pressure:
Step 6 - Calculate a safe working pressure based on the factors
of safety, and assuming a design factor of 0.72, gives:
The safe working pressure is 16.04 N/mm2.
B.2  Interacting defects
B.2.1 Example 8
This example is for a pair of rectangular patches 203.2 mm in
length and separated axially by 81.3 mm.  One defect is
14.2mm deep and the other is 13.7 mm deep.
The basic properties required by the assessment are:
Using the procedure for assessing interacting defects given in
Sec.9:
The defects should be grouped into axial projections as
described in Steps 1 to 5 of Sec.9.2.
Step 6 is to estimate the failure pressure of both defects, when
treated as isolated defects.  These pressures are 19.73 N/mm2
and 20.59N/mm2 respectively.
Applying the rules for defect interactions in Steps 7 to 9 (Sec.9)
for the combined defect gives :
Step 10 is to select the minimum of the individual and com-
bined defects as the failure pressure.  In this case, the failure
pressure of the combined defect is less than the single defect
solutions, indicating interaction.  The failure pressure Pf of the
defect is therefore 17.71 N/mm2.
Step 11 is to calculate the safe working pressure from the esti-
mated failure pressure, by applying the appropriate safety fac-
tors.  For a design factor of 0.72, the safe working pressure is
11.48 N/mm2.
B.2.2 Example 9
This example is for a pair of rectangular patches 203.2 mm in
length and separated axially by 203.2 mm.  The defects are
14.1 mm and 14.2 mm deep respectively.
The basic properties required by the assessment are:
Using the procedure for assessing interacting defects given in
Sec.9:
Steps 1 to 5 would be used to group the defects along a gener-
ator and estimate the projected profiles.
Step 6 is to estimate the failure pressure of both defects, when
treated as isolated defects.  The failure pressures are 19.90N/
mm2 and 19.73N/mm2 respectively.
Applying the rules for defect interactions in Steps 7 to 9 (Sec.9)
gives:
Step 10 is to select the minimum of the individual and com-
bined defects as the failure pressure.  In this case, the failure
pressure of the combined defect is slightly greater than that of
either of the single defects, which suggests that there will be no
interaction and that the pipe will fail at 19.73N/mm2.
Step 11 is to calculate the safe working pressure by applying
the appropriate safety factors.  For a design factor of 0.72, the
safe working pressure is 12.79 N/mm2.
B.3  Complex shaped defect
B.3.1 Example 10
This example is an analysis of the failure pressure of a complex
shaped defect (see Sec.10).  It is a large rectangular patch con-
taining two adjacent deeper circular defects with semi-ellipti-
cal profiles.
The dimensions and material properties are summarised as fol-
lows, and a schematic of the defect is given in Fig. B-1:
The defect profile is shown in Fig.B-1 and the exact depths are
tabulated in Table B-1.
Outside diameter =812.8 mm
Original wall thickness=20.1 mm
SMTS=624.2 N/mm2
Combined length (Step 7)=487.7 mm
Combined area=5669 mm2
Effective depth (Step 8)=11.62 mm
Failure pressure (Step 9)=17.71 N/mm2
Outside diameter =812.8 mm
Original wall thickness=20.1 mm
SMTS (=fu)=624.2 N/mm2
2
N/mm 24.75 ) , min = = comp press f
P (P     P
2
/ 04 . 16 ) 72 . 0 )( 9 . 0 ( mm N P P f sw = =
Combined length (Step 7)=609.6 mm
Combined area=5751 mm2
Effective depth (Step 8)=9.43 mm
Failure pressure (Step 9)=20.13 N/mm2
Outside diameter=762.0 mm
Original wall thickness=22.1 mm
SMTS (=fu)=525.3 N/mm2
Table B-1  Tabulated profile complex shaped defect
Length
(mm)
Depth
(mm)
00
03.9
0.87.39
1.68.7
2.49.61
3.210.3
410.83
4.811.23
5.611.53
6.411.74
7.211.86
811.9
16311.9
169.212.42
175.413.41
181.514.28
187.715.04
193.915.67
20016.19
206.216.59
212.316.87
218.417.04
224.517.1
230.617.04
236.716.87
242.816.59
24916.19DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 37
Using the procedure for assessing complex shaped defects
given in Sec.10:
Single Defect Solution (Steps 1 to 2)
Step 1 is to calculate the average depth of the defect from the
projected total area loss of the defect.
Step 2 is to estimate the failure pressure of the defect from the
average depth and the total length.
Progressive Depth Analysis (Steps 3 to 16)
The failure pressure was estimated for 50 increments in a pro-
gressive depth analysis.  The variation in the failure pressure
estimate, with respect to each step, is shown in Fig.B-2.
Step 3 is to subdivide the defect  into horizontal sections or
depth increments and estimate the failure pressure for each
section from Steps 4 to 12.
Two examples of the analysis at various depths of horizontal
section are given below:
Steps 6 to 12 are to estimate the failure pressure of the idealised
pits.
Step 7 is to estimate the effective thickness of the pipe for the
remaining pits.
Pit interactions based on the reduced thickness pipe.
Step 13 is to estimate the failure pressure for the current hori-
zontal step depth from the minimum of the patch and pit esti-
mates.  In this case the minimum pressure is from the pit:
Number of Pits  = 2
Pit Interactions Based on the Reduced Thickness Pipe
Step 13 is to estimate the failure pressure for the current hori-
zontal step depth from the minimum of the patch and pit esti-
mates.  In this case, the minimum pressure is from the pit
interaction between pits 1 and 2:
In Step 15 the defect is calculated as a single defect with the
255.115.67
261.315.04
267.514.28
273.613.41
279.812.42
28611.3
292.212.42
298.413.41
304.514.28
310.715.04
316.915.67
32316.19
329.216.59
335.316.87
341.417.04
347.517.1
353.617.04
359.716.87
365.816.59
37216.19
378.115.67
384.315.04
390.514.28
396.613.41
402.812.42
40911.9
56411.9
564.811.86
565.611.74
566.411.53
567.211.23
56810.83
568.810.3
569.69.61
570.48.7
571.27.39
5723.9
5720
Total length=572.0 mm
Maximum depth=17.1 mm
Total projected area loss=7584.6 mm2
Average depth =13.26 mm
Failure pressure =16.23 N/mm2
Table B-1  Tabulated profile complex shaped defect
 (Continued)
Depth of increment no. 12=4.1 mm
Patch average area (Step 4)=2347 mm2
Patch length =572.0 mm
Patch average depth (Step 4)=4.1 mm
Patch failure pressure (Step 5)=27.47 N/mm2
Number of Pits =1
Effective reduced thickness=19.42 mm
Pit
Average
Depth
(mm)
(Step 6)
Average Depth In
Reduced Wall
(mm)
(Step 8)
Length
 (mm)
Failure
Pressure
(N/mm2)
(Step 9)
113.2610.58571.915.54
Minimum pressure=15.54 N/mm2
Depth of increment no. 38
(This is the section that gives the mini-
mum pressure).
=13.0 mm
Patch average area (Step 4)=7019 mm2
Patch length (Step 4)=572.0 mm
Patch average depth (Step 4)=12.59 mm
Patch failure pressure (Step 5)=17.65 N/mm2
Effective reduced thickness=12.59 mm
Pit
Average Depth in nominal
Thickness Pipe (mm)
Length
(mm)
Separation to next
pit
115.7310319.6 mm
215.73103-
Start
Pit
End 
Pit
Average Depth In
Reduced Wall (mm)
(Step 6-8)
Overall
Length
(mm)
Failure Pressure
 (N/mm2)
(Step 9 or 10-12) 
116.2210315.56
125.6922613.40
226.2210315.56
Minimum pressure is due to
interaction between pits 1 and 2=13.40 N/mm2DET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 38 see note on front cover
total length and the maximum depth.  Using the procedure for
assessing single defects given in Sec.8.2 .
If this complex shaped defect  is assessed as a single defect,
based on the total length and maximum depth, then the pre-
dicted failure pressure is 10.03 N/mm2.
Step 15 is to estimate the failure pressure of the complete
defect, as the minimum of all the minimum estimates for each
horizontal step, i.e. the minimum of all Step 13 results but not
less than the pressure from Step 15, (see Fig.B-4).
Analysis of the defect as a complex profile using the progres-
sive depth method, without the application of a safety factor,
gives a failure pressure estimate of 13.40 N/mm2 from a sec-
tion depth of 13.0 mm.
Step 17 is to estimate a safe working pressure from the esti-
mated failure pressure.  Applying the safety factors for a
design factor of 0.72:
The safe working pressure is 8.68N/mm2 (86.8 bar).
B.3.2 Example 11
This example is an analysis of the failure pressure of a smooth
shaped complex shaped defect 
The pipeline geometry and properties are summarised as fol-
lows:
The defect profile is shown in Fig.B-3 and the exact depths are
tabulated in Table B-2.
Using the procedure for assessing complex shaped defects
given in Sec.10:
Single Defect Solution (Steps 1 to 2)
Step 1 is to calculate the average depth of the defect from the
projected total area loss of the defect.
Step 2 is to estimate the failure pressure of the defect from the
average depth and the total length.
Progressive Depth Analysis (Steps 3 to 16)
The profile was sectioned at 50 levels and the failure pressure
estimated for each increment.  Fig.B4 shows the variation of
the failure pressure estimate with depth.  The minimum failure
pressure estimate was 13.21N/mm2.  The section depth was
1.09 mm; this corresponds to the natural division between
patch and pit, which can be seen in Fig.B-4.  The effect of the
relatively distinct change in profile at this depth produces a
sharp change in the estimated failure pressure curve, as shown
in Fig.B-4.
The calculations at the section that produced the minimum fail-
ure pressures are presented as follows, as a typical example of
the calculation which had to be performed at each section:
Steps 6 to 12 are to estimate the failure pressure of the idealised
pits.
Step 13 is to estimate the failure pressure for the current hori-
zontal step depth from the minimum of the patch and pit esti-
mates.  In this case the minimum pressure is from the pit:
Step 15 is to estimate the failure pressure of the complete
defect as the minimum of all the minimum estimates for each
horizontal step, i.e. the minimum of all  Step 13 results (see
Fig.B-4).
Analysis of the defect as a complex profile using the progres-
sive depth method, without the application of a safety factor,
gives a failure pressure estimate of 13.21N/mm2.
In Step 15 the defect is calculated as a single defect with the
total length and the maximum depth
Using the procedure for assessing single defects given in
Sec.8.
If this complex shaped defect  is assessed as a single defect,
based on the total length and maximum depth, then the pre-
dicted failure pressure is 11.86 N/mm2.
Step 16 is to estimate the allowable corroded pipe pressure of
the complete defect as the minimum of all the minimum esti-
mates for each horizontal step, i.e. the minimum of all Step 13
results, but not less than the pressure from Step 15.
Step 17 is to calculate the safe working pressure from the esti-
mated failure pressure.  Applying the safety factors for a
Total length=572.0 mm
Maximum depth=17.1 mm
Failure pressure =10.03 N/mm2
Outside diameter=611.0 mm
Wall thickness=8.20 mm
SMTS=571.0 N/mm2
Table B-2  Tabulated profile for actual corrosion defect
Length
(mm)
Depth
(mm)
00
28.91
57.81.1
86.71.1
115.61.1
144.51.3
173.41.8
202.32.8
231.22.8
260.11.6
2890
Total length=289.0 mm
Maximum depth=2.8 mm
Total projected area loss=421.94 mm2
Average depth =1.46 mm
2
/ 68 . 8 ) 72 . 0 )( 9 . 0 ( mm N P P f sw = =
Failure pressure =13.55 N/mm2
Step depth=1.06 mm
Patch average area (Step 4)=280.4 mm2
Patch length=289.0 mm
Patch average depth (Step 4)=0.97 mm
Patch failure pressure (Step 5)=15.68 N/mm2
Effective reduced thickness (Step 7)=7.60 mm
Number of Pits =1
Pit
Average
Depth
(mm)
Average Depth
On Reduced Wall
(mm)
Length
 (mm)
Failure Pressure
(N/mm2)
11.7001.10022213.22
Minimum pressure=13.22 N/mm2
Total length=289.0 mm
Maximum depth=2.8 mm
Failure pressure =11.86 N/mm2DET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 39
design factor of 0.72:The safe working pressure is 8.56N/mm2 (85.6 bar).
Figure B-1 
Profile for complex shaped defect - Example assessment
Figure B-2 
Variations of the estimated failure pressure for complex shaped defect - Example assessment
2
/ 56 . 8 ) 72 . 0 )( 9 . 0 ( mm N P P f sw = =
Illustration of complex shape
0
5
10
15
20
25
-1000100200300400500600700
Length axis (mm)
Complex Shaped Corrosion Defect
Pressure Estimates at Depth Increments
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0
0.02.04.06.08.010.012.014.016.018.0
Depth increment (mm)
Predicted Failure  Pressure (MPa)
Failure Pressure = 13.4 MPa
Increment depth = 13.0 mmDET NORSKE VERITAS
Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 40 see note on front cover
Figure B-3 
Profile for actual corrosion defect - Example assessment
Figure B-4 
Variations of the estimated failure pressure for actual corrosion defect - Example assessment
050100150200250300
-8
-6
-4
-2
0
Length (mm)
Depth (mm)
Complex Shaped Corrosion Defect
Pressure Estimates at Depth Increments
12.5
12.7
12.9
13.1
13.3
13.5
13.7
13.9
0.00.51.01.52.02.53.0
Depth increment (mm)
Predicted Failure  Pressure (MPa)
Failure Pressure = 13.22 MPa
Increment depth = 1.06 mmDET NORSKE VERITAS
Amended October 2006Recommended Practice DNV-RP-F101,  October 2004
see note on front cover Page 41
APPENDIX C
DETAILED CALCULATION OF MEASUREMENT ACCURACIES
C.1  Implications of correlated and uncorrelated wall
loss measurements for the assessment of interacting
defects and complex shaped defects
When assessing interacting or complex shaped defects using
the methods in Part A of this document, it is important to estab-
lish whether the defect depth measurements are correlated or
uncorrelated.  The assessment should be made in consultation
with an appropriate authority on the measurement technique
and procedures used.
The difference between fully correlated measurements and
uncorrelated measurements can be explained from the follow-
ing simple example: two adjacent pits of equal depth.  Fully
correlated measurements of the depth of two adjacent pits of
equal depth would give the same value, because the measure-
ment error would be same.  Therefore it would be known that
the pits were of equal depth, but the actual depth would not be
known with certainty.  Uncorrelated measurements of the same
two pits may give different values for each pit.  If the same
uncorrelated measurement technique was applied to many pits
of the same depth, then the average value of the depth meas-
urements would give an estimate of the actual depth of the pits.
The difference between fully  correlated and uncorrelated
measurements of corrosion profiles can be explained in the
same way.  Fully correlated  measurements of the depth at
points along a uniform depth wall loss would all be the same,
because the measurement error would be the same for each
measurement.  The technique would reveal a uniform depth
wall loss, but the depth would not be known with certainty.  An
uncorrelated technique would produce different depth esti-
mates at each point, because the error might be different for
each individual measurement.  For a long defect with a uni-
form depth profile, if there were a large number of uncorre-
lated measurements, then  the average depth would be
accurately measured, but it would not be apparent that the
defect had a uniform depth profile.
Depth measurements are averaged as part of the assessment of
the interactions between pits and the assessment of complex
profiles.  Correlated measurements give a larger spread in
uncertainty during this process than do uncorrelated measure-
ments.  In practice, measurement errors are neither completely
uncorrelated nor fully correlated, and it is important to take
expert advice to decide which assumption is the most appropri-
ate for a particular inspection technique.  If it is not possible to
establish whether measurements are correlated or uncorre-
lated, then the most conservative assumption is to assume that
they are fully correlated.
C.2  Partial safety factors for absolute depth meas-
urement (e.g. ultrasonic wall thickness or wall loss
measurements)
For known correlation between the pipe wall thickness meas-
urement and the ligament thickness (or corrosion depth) meas-
urements, the following procedure can be used to calculate the
StD[d/t] of the relative corrosion depth from the known uncer-
tainties in the absolute measurements.  The derivation assumes
that d, r and t have LogNormal distributions
C.2.1 Remaining ligament thickness (r) and the wall thick-
ness (t) are measured
where
Z1 =ln(r)
Z2 =ln(t).
The mean value and standard deviation for Z1 and Z2 may be
derived from:
The mean values of the ligament thickness, E[r], and the pipe
wall thickness, E[t], may be approximated by the measured
values.
The CoV is the Coefficient of Variation, defined as the stand-
ard deviation divided by the mean.  The correlation coefficient
between Z1 and Z2, , may be calculated from:
It should be noted that the correlation between Z1 and Z2 is due
to the correlation between r and t. If r and t is uncorrelated, then
Z1 and Z2 is uncorrelated.
C.2.2 Corrosion depth (d) and the wall thickness (t) are
measured
where
Z1 =ln(d)
Z2 =ln(t).
The mean value and standard deviation for Z1 and Z2 may be
derived from:

Recommended Practice DNV-RP-F101,  October 2004Amended October 2006
Page 42 see note on front cover
The mean values of the corrosion depth, E[d], and the pipe wall
thickness, E[t], may be approximated by the measured values.
The CoV is the Coefficient of Variation, defined as the stand-
ard deviation divided by the mean.  The correlation coefficient
between Z1 and Z2, , may be calculated from:
C.3  Application of absolute depth measurement
The acceptance equation require stochastic properties for rela-
tive depth measurements. When absolute measurements are
available the relative corrosion depth needs to be calculated.
Procedures for calculating the mean and the StD[d/t] of the rel-
ative corrosion depth from the known uncertainties in the abso-
lute measurements are given below.
C.3.1 If the remaining ligament thickness (r) and the wall
thickness (t) are measured:
The acceptance equation is only applicable the when following
limitations are fulfilled:
The correlation coefficient is a measure of the mutual linear
dependence between a pair of stochastic variables.  In most
cases, the correlation between the pipe wall thickness measure-
ment and the ligament thickness measurement will not be
known and, therefore, it should be assumed to equal zero (i.e.
no correlation).
For no correlation,  the mean value, E[d/t], and the standard
deviation, StD[d/t], of the relative corrosion depth may be
written as:
The mean values of the ligament thickness, E[r], and the pipe
wall thickness, E[t], may be approximated by the measured
values. 
C.3.2 If the corrosion depth (d) and the wall thickness (t)
are measured:
The acceptance equation is only applicable when the following
limitations are fulfilled:
The correlation coefficient is a measure of the mutual linear
dependence between a pair of stochastic variables.  In most
cases, the correlation between the pipe wall thickness measure-
ment and the metal loss depth measurement will not be known
and, therefore, it should be assumed to equal zero (i.e. no cor-
relation). 
For no correlation,  the mean value, E[d/t], and the sta, ndard
deviation, StD[d/t], of the relative corrosion depth may be
written as:
The mean values of the corrosion depth, E[d], and the pipe wall
thickness, E[t], may be approximated by the measured values.3217