NO

TECHNICAL SPECIFICATION I-ET- 0000.00-6500-210-PSQ-001

CLIENT: SHEET
PETROBRAS 1 de 10
PROGRAM:
GENERAL CC:

AREA:
GENERAL PROJECT:

TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
ENGENHARIA
PIPELINES AND RISERS GIRTH WELDS
ET-000 00-6500-210-PSQ-001=0.doc
MS-WORD-2003

INDEX OF REVIEW

REV DESCRIPTION AND/OR REVISED SHEET

0 ISSUED FOR CONSTRUCTION

REV. 0 REV. A REV. B REV. C REV. D REV. E REV. F REV. G REV. H

DATA 09/05/08
PROJETO SL/SEQUI
EXECUO PZJ
VERIFICAO TBK
APROVAO JMBR
AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV. G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA
 NO REV.
ESPECIFICAO TCNICA I-ET- 0000.00-6500-210-PSQ-001 0
PROGRAM SHEET:
GENERAL 2 de 10
TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
PIPELINES AND RISERS GIRTH WELDS

1. SCOPE

1.1.1 This Technical Specification sets forth the technical requirements for issuing Engineering Critical
Assessment (ECA) for pipelines and risers in subsea installations to determine maximum acceptable
flaw criteria in girth welds before installation. This specification is only applicable to C-Mn steels.

1.1.2 ECA shall be issued according to recognized industry practice and shall meet the requirements of the
codes and standards referenced herein.

2. DEFINITIONS AND ABBREVIATIONS

The following definition is used in this Technical Specification:

AUT: Automatic Ultrasonic Test.

COMPANY: PETROBRAS.

CONTRACTOR: The company responsible for the engineering design, procurement and construction of
the respective contract.

CMOD: Crack Mouth Opening Displacement

CTOD: Crack Tip Opening Displacement

ECA: Engineering Critical Assessment.

FCGR: Fatigue Crack Growth Rate

HAZ: heat affected zone of the welded joint.

MAY: Is used where alternatives are equally acceptable within the limits of the specification.

MUST: Is used where a provision is a statutory requirement.

NDT: Non Destructive Test.

PIPELINE: As defined in item C-263 of section 1 of DNV OS-F-101.

RISER: As defined in item C-289 of section 1 of DNV OS-F-101.

SCF: Stress Concentration Factor

SENB: Single Edge Notch Bend fracture toughness specimen.

SENT: Single Edge Notch Tensile fracture toughness specimen.

SHALL: Is used where a provision is mandatory.

AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA
 NO REV.
ESPECIFICAO TCNICA I-ET- 0000.00-6500-210-PSQ-001 0
PROGRAM SHEET:
GENERAL 3 de 10
TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
PIPELINES AND RISERS GIRTH WELDS
SHOULD: Is used where a provision is preferred. Alternative solutions are acceptable if having the same
functionality and quality.

SPECIFICATION: Any or all of the sections, paragraphs, appendices, tables, sketches, figures or
attachments of this or other referenced specification.

WORK: Physic execution of the project.

3. REFERENCES

3.1 CODES, STANDARDS AND SPECIFICATIONS

3.1.1 The latest issue of the following codes, standards and regulations shall form part of this specification.
In the event of revision of any such document written clarification shall be sought before proceeding
with the work.

British Standard Institute (BSI)

BS 7910 Guide on Methods for Assessing the Acceptability of Flaws

in Metallic Structures.

BS 7448 Fracture Mechanics Toughness Test.

BS 7608 Code of Practice for Fatigue Design and Assessment of

Steel Structures

Det Norske Veritas (DNV)

DNV OS-F-101 Submarine Pipeline Systems

DNV OS-F-201 Dynamic Risers

DNV RP-F-108 Fracture Control for Pipeline Installation Methods

Introducing Cyclic Plastic Strain

DNV RP-C-203 Fatigue Design of Offshore Steel Structures

DNV RP-F-105 Free Spanning Pipelines

AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA
 NO REV.
ESPECIFICAO TCNICA I-ET- 0000.00-6500-210-PSQ-001 0
PROGRAM SHEET:
GENERAL 4 de 10
TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
PIPELINES AND RISERS GIRTH WELDS

3.2 COMPLIANCE

3.2.1 This specification shall define the methodology to establish any ECA for determining alternative flaw
acceptance criteria of submarine rigid pipelines and risers girth welds.

3.3 CONFLICT OF INFORMATION AND DOCUMENT APPROVAL

3.3.1 In the event of any conflict between this specification or any other specification and associated
requisition forms, or with any of the applicable codes and regulations, written clarification shall be
sought from COMPANY before proceeding with the WORK. CONTRACTOR shall provide
COMPANY with a written report of conflict. COMPANY will resolve the conflict and provide
written document.

3.3.2 Should CONTRACTORs procedures deviate from this specification, a written report shall be sent to
COMPANY.

3.3.3 COMPANYs approval of the deviation (s) must be in written form.

3.3.4 CONTRACTOR shall submit the ECA report to COMPANY respecting the contractual time limit for
approval.

3.3.5 For the purpose of using ECA, none girth weld shall be executed before the formal written approval
by COMPANY.

3.4 METHODOLOGY

3.4.1 PURPOSE OF ECA

3.4.1.1 The purpose of an ECA is to determine alternative flaw acceptance criteria in submarine rigid
pipelines and risers girth welds. The baseline methodology adopted to assess fatigue and fracture
shall be according to the procedures defined in appendix A of DNV OS-F-101.

3.4.2 ECA LEVEL AND WORKMANSHIP CRITERIA.

3.4.2.1 The Level of assessment adopted regarding BS 7910 procedures, if required, shall be according to
appendix A of DNV OS F-101 pipeline design code. Assessments at Level 2B or at Level 3B of BS
7910 are always required. So, the method shall include the use of the specific-material stress-strain
curve.

3.4.2.2 If experimentally determined values of J (for installation) do not meet the requirements specified in
the tables of appendix A of DNV OS-F-101, a full ECA shall be carried out. In this case, the
maximum acceptable criteria resulted shall be limited to the minimum tabulated flaw sizes presented
in the tables regarding the respective maximum total strain. If total longitudinal strain is lower than
0.4% the workmanship criteria from appendix D of DNV OS-F-101 may be used or alternatively an
ECA may establish the flaw acceptance criteria.

AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA
 NO REV.
ESPECIFICAO TCNICA I-ET- 0000.00-6500-210-PSQ-001 0
PROGRAM SHEET:
GENERAL 5 de 10
TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
PIPELINES AND RISERS GIRTH WELDS
3.4.2.3 For pipelines and risers installed by the reeling method and also other methods introducing large
plastic strains, the total plastic straining shall be limited to 3% nominal strain. In case total plastic
straining exceeds this value, written agreement from COMPANY shall be required. The total crack
extension for all events of reeling and unreeling is limited to 1 mm.

3.4.2.4 In any case, the weld strength shall over-match the parent metal strength. For girth welds exposed to
total longitudinal strain equal or larger than 0.4%, it is required that the actual yield stress of the weld
metal is 120 MPa higher than SMYS of the base material according to item B-703 of section 6 of
DNV OS-F-101. For girth welds exposed to total longitudinal strain lower than 0.4%, it is required
that the actual yield stress of the weld metal be minimum 80 MPa above the SMYS of the base
material according to item C-302 of appendix C of DNV OS-F-101. For pipelines exposed to total
longitudinal strain equal or larger than 0.4%, it is recommended to limit the maximum over-matching
in 200 MPa considering the actual welding consumable properties, and based on the yield stress
(Rt0.5) of the materials. All tensile tests taken to comply this requirement shall be fulfilled for as
welded and straining conditions. In case of failure to achieve this requirement, written agreement
from COMPANY shall be required.

3.4.2.5 AUT shall be always carried out for the flaw detection and sizing when an ECA acceptance criteria is
utilized and the AUT shall be carried out according to appendix E of DNV OS-F-101. The detection
ability of an AUT system shall be deemed sufficient if the probability of detection a defect of the
smallest allowable height determined during an Engineering Critical Assessment is 90% at a 95%
confidence level according to item H-301 of appendix E of DNV OS-F-101.

3.4.2.5.1 AUT data base from identical previous projects could be acceptable, if the complete Validation
Reports with all results are opened for Petrobras evaluation. At discretion of Petrobras the AUT
Validation Program may be reduced if the item H-503 of appendix E of DNV OS F 101 is
attempt.

3.4.2.5.2 If the previous results become acceptable for Petrobras a reduced Validation Program shall be
made at the presence of Petrobras on a minimum 2 coupons of the same weld bevel configuration
and same welding processes. The diameter and thickness of the coupons shall be agreed
beforehand. To validate the previous data presented for Petrobras, a minimum of 10
discontinuities shall be selected for macro sectioning. To accommodate smalls positioning and/or
machining errors, at each chosen point of each discontinuity 3 measurements shall be taken: the
first at 1 or 2 mm before the chosen point, the second at the chosen point and the third at 1 or 2
mm after the chosen point. At each discontinuity only the greatest of the 3 height measurements
shall be taken for the statistics.

3.4.2.5.3 A final report shall be prepared and consolidated by a level 3 ultrasonic inspector, and shall have
the following minimum content:
a) Personnel Certification.
b) Calibration Certificates.
- AUT unit;
- Probes and wedges;
- Calibration blocks (Dimensional and velocity).
c) NDT Inspections.
- Manual UT reports;
- RT reports;
- AUT reports.
d) AUT Validation Program Report.
- Previous Qualification Program Results Report;
- Macro sectioning report;
- Comparison Macro sizing x AUT sizing;
AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA
 NO REV.
ESPECIFICAO TCNICA I-ET- 0000.00-6500-210-PSQ-001 0
PROGRAM SHEET:
GENERAL 6 de 10
TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
PIPELINES AND RISERS GIRTH WELDS
- Comparison RT sizing x AUT sizing;
- Resumed data from previous project;
- Data from the resumed validation test;
- Statistics analysis for AUT sizing error and uncertainty
- Final conclusion with a clear indication of the uncertainty of the AUT system.

3.4.3 TEMPERATURE DE-RATING

3.4.3.1 Stress-strain material data are always required at appropriate temperature for parent material and/or
weld metal. Alternatively, for the operational case, it may be done according to figure 2 of section 5
of DNV OS-F-101 pipeline design code for temperatures above room temperature. This de-rating
shall be applied to the whole stress-strain curve. It shall be adopted an upper bound parent material
stress/strain curve as required in item 3.5.5 of this specification, when total nominal longitudinal
strain resulting from installation and operation load effects is equal or larger than 0.4%. For the
installation case, the de-rating is not applicable.

3.4.4 LOAD INPUT DATA

3.4.4.1 Strain cycles shall be established by CONTRACTOR regarding a case by case basis, considering
for instance pipe laying angle. ECA report data shall be according to this installation loads specified.

3.4.4.2 ECA shall take into account all applied stresses, including all fatigue loads during installation and
operation. All the fatigue assessment shall be by means of fracture mechanics approach (Paris law),
according to BS 7910.

3.4.4.3 The fatigue loads during installation shall represent realistically the more critical wave conditions
expected and shall include a minimum of 12 hours holding period. This holding period shall be
representative of maximum expected time of the pipeline holding on the clamps while it will be
laying.

3.4.4.4 VIV shall be considered and taken into account during installation.

3.4.4.5 Operational loads induced from waves motion and currents, as Vortex Induced Vibration (VIV - total
VIV fatigue damage, per year, including damage from the yearly and the storm current profiles), shall
have the induced fatigue damage (per year) established considering the yearly wave and current
statistic distribution from the related Meteoceam Data Report. Operational loads induced from the
Production Unit, such as Vortex Induced Motion (VIM), first and second order environmental loads
and all significant cyclic loads of each specific situation shall be taken into account. If there is a
temperature variation during operation greater than 20C, the cyclic loads induced by this variation
shall be taken into account. Any other significant cyclic load that may imply in a crack growth during
installation or operation shall be taken into account. For risers, it is recommended to use Appendix E
of DNV OS-F-201 for VIV fatigue damage analysis guidance.

3.4.4.6 For reference, the applied loads to be considered are in the table below:

AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA
 NO REV.
ESPECIFICAO TCNICA I-ET- 0000.00-6500-210-PSQ-001 0
PROGRAM SHEET:
GENERAL 7 de 10
TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
PIPELINES AND RISERS GIRTH WELDS

3.4.4.7 The effect on the crack growth of hydrostatic test or any other situation of maximum stress shall be
assessed.

3.4.4.8 The maximum expected holding period between the pipeline laying and the beginning of the
hydrotest shall be regarded, as far as the consumed fatigue life is concerned, from the dynamic
stresses induced by environmental load - for instance, Vortex Induced Vibration (VIV).

3.4.4.9 Separate reports of all fatigue loads shall be submitted to COMPANY for approval before the
submittal of the ECA report. Pipelaying Analysis Report clearly defining the strain cycle for each
pipeline to be laid shall be included.

3.4.5 METHODOLOGY BASED IN FRACTURE MECHANICS APPROACH

3.4.5.1 For the calculation of the maximum acceptable flaw size before installation, CONTRACTOR shall
take into account, step by step, every crack growth promoted by all loads during installation and
operation, in such a way that after all operational life the maximum crack size is still allowable.

3.4.5.2 Even if contractors option is not to carry out the full ECA, ECA Fatigue of sub-section F of
appendix A of DNV OS-F-101 shall be carried out.

3.4.6 METHODOLOGY FOR ECA IN PIPELINE FOR SOUR SERVICE CONDITIONS

3.4.6.1 When the pipeline or riser are specified for sour service conditions, hydrogen charged fracture
toughness test are required in order to determine the fracture toughness at weld metal and HAZ
according to item 3.6.1.3, for the operational case.

3.4.6.2 For assessment of the operational case in sour service conditions, the significant typical stress
ranges from operational cyclic loads may be limited in such a way that the resultant range of the
applied stress intensity factor (KI) will be limited in levels close to the fatigue threshold in order to
reduce the influence of the hydrogen in the fatigue behavior.

3.4.6.3 The CONTRACTOR may alternatively carry out fatigue tests at the expected stress ranges, in an
environment representative of the project conditions, if it is expected to have values of K greater
than the threshold stress intensity factor range (Kth), in order to determine the constants A and m.
The experimental data shall be statistically treated with two standard deviations (as per BS 7910).

3.4.6.4 Other alternative, is to carry out the fatigue analysis utilizing a FCGR including a knock up factor in
the in air curve of BS-7910 or other FCGR, in such a way that final FCGR will be conservative
comparing to experimental results of fatigue crack growth in H2S environment, from well recognized
institution and/or publication, statistically treated with two standard deviations (as per BS 7910) and
in conditions not less critical than the project conditions.

3.4.6.5 The whole linepipe thickness has to be considered for sour service conditions. It means that both
superficial and embedded defects shall consider H2S effects.

3.4.7 FINAL CRITERIA LIMITS

3.4.7.1 The final criteria for the production girth welds shall take into account the AUT uncertainty, as in
item 3.5.3 of this specification. These final criteria for the production welds shall have a maximum
height limited to 4 mm for pipelines with thickness greater than 12.7 mm and they shall have a
maximum height limited to t/4 (where t is the wall thickness) for pipelines with thickness up to 12.7
mm. This limited height is to guarantee a good weld quality, according to good industrial practices.
AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA
 NO REV.
ESPECIFICAO TCNICA I-ET- 0000.00-6500-210-PSQ-001 0
PROGRAM SHEET:
GENERAL 8 de 10
TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
PIPELINES AND RISERS GIRTH WELDS
Also the maximum acceptable flaw length, taken into account the AUT uncertainty, shall be limited
to 200 mm.

3.4.8 CRACKWISE FILES

3.4.8.1 Contractor shall use preferentially Crackwise software for calculations and provide all the necessary
Crackwise files used in ECA to COMPANY for the purpose of ECA approval. If CONTRACTOR
decides not to use Crackwise software, the employed software with its license shall be temporarily
provided for COMPANY for the same purpose.

3.5 SAFETY FACTORS / INPUT DATA / NDT UNCERTAINTY

3.5.1 The ECA shall not use global safety factors for fracture assessment according to annex K of BS 7910.
The ECA shall be based in conservative input data.

3.5.2 Conservative input data are required and they shall be according to DNV OS-F-101.

3.5.3 Regarding the use of NDT conservative input data, for the purpose of the use of ECA criteria, the
NDT inspection shall be based in AUT system. AUT probability of detection, sizing accuracy and
undersizing error tolerances shall be according to items H-301 and H-302 of appendix E of DNV OS-
F-101. Both height and length shall be considered.

3.5.4 Regarding the use of fracture toughness conservative input data, they shall be based on experimental
tests and the reliability shall be according to table K.3 of BS 7910 or better. If different thickness
exist and only one is to be tested, the thicker wall shall be used for the fracture toughness test with the
highest Pcm. In case the highest Pcm does not correspond to the thicker material, the thinner material
shall also be tested and the lower fracture toughness results from both thicknesses shall be utilized for
the ECA. When J-R curve arises, at least the average of six points minus one standard deviation shall
be used to represent the Lower Bound Curve.

3.5.5 When the design condition is strain-based, according to item G-204 of appendix A of DNV OS-F-
101, the stress/strain curve used in the ECA shall be representative of upper bound tensile
properties/curves, as in item 3.4.3.1 of this specification. It is recommended to use the parent material
data plus two standard deviations or otherwise the stress-strain curve data shall be submitted to
COMPANY for approval.

3.5.6 The allowable damage ratio for fatigue shall be according to table 5-9 of DNV OS-F-101 in case of
PIPELINES and according to table 5-9 of DNV OS-F-201 in case of RISERS. The safety factors for
the different stages shall be compared and then the highest safety factor shall be used for all the
fatigue assessments (both installation and operation). For the fatigue crack growth analysis, the ECA
shall be based in fracture mechanics (Paris Law) and shall use a safety factor not less than half the
allowable damage ratio (if complying with item H-300, appendix E of DNV-OS-F-101) and not less
than any other value established in the contractual design specifications. The total number of cycles
used as cyclic stress input data shall comply with this final fatigue safety factors, and it shall provide
a total fatigue life complying with SN design curves and contractual requirements. A validation shall
be made, checking if the equivalent resultant SN design curve calculated based in fracture mechanics
(or any other downgraded) regarding DNV RP-C203, BS-7608, or any other contractual specification
used in ECA shall will be the one to be used for the free spanning calculations (according to DNV
RP-F-105 or other equivalent required contractual document).

3.5.7 STRESS CONCENTRATION FACTOR

AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA
 NO REV.
ESPECIFICAO TCNICA I-ET- 0000.00-6500-210-PSQ-001 0
PROGRAM SHEET:
GENERAL 9 de 10
TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
PIPELINES AND RISERS GIRTH WELDS
3.5.7.1 SCF formulas of item E-206 of appendix A of DNV OS-F-101are acceptable. Other different
mathematical formulas used for the calculation of the stress concentration factor (SCF) regarding the
weld misalignment shall be clearly stated in the ECA Methodology document or in the ECA report
and submitted for approval.

3.6 TESTS FOR CHARACTERIZATION OF THE MATERIALS FRACTURE RESISTANCE

3.6.1 FRACTURE TOUGHNESS TESTS

3.6.1.1 All fracture toughness tests shall be according to all parts of BS 7448 or to DNV RP F-108. For the
use of a full ECA acceptance criteria, at minimum the tests required for the ECA stated at table A-10
of appendix A of DNV OS-F-101 shall be carried out. SENB tests are required for operational case.
SENB tests with reduced notch length as per item A-908 of DNV OS F-101 are not acceptable.

3.6.1.2 For the purpose to determine the materials fracture resistance in the weld joints, both weld metal and
HAZ of all welding procedures, which ECA criteria is applicable, have to be tested before
establishing the final flaw acceptance criteria according to DNV OS-F-101 at item A-802 of appendix
B. Testing of fusion line/HAZ shall be conducted on surface notched specimens of square sections (B
x B) with orientation corresponding to NQ in the notation of BS 7448. The notch and the fatigue pre-
crack of the HAZ specimens shall start from the weld face, except when the weld joint preparation
allows that a notch and the fatigue pre-crack that comes from the weld root side and reach the fusion
line will not grow through the weld metal during the test. At least three specimens with the crack in
HAZ and three specimens with the crack in the weld metal shall be tested as required in DNV OS-F-
101, and it is recommendable to have at least six samples prepared for the HAZ in order to get at least
three valid. No fracture toughness value may be adopted as lower bound without carrying out
fracture toughness tests, as required by DNV OS-F-101. Fracture toughness values from correlations
with Charpy V notch or any other impact value from BS 7910 or any other code will be not
accepted.

3.6.1.3 For the purpose to determine the materials fracture resistance for the operational case, when sour
service condition is a requirement, hydrogen charged SENB specimens, prepared and conducted as
described at 3.6.1.2 shall be carried out. Some cares shall be taken to carry out this tests, as described
bellow:

3.6.1.3.1 After the fatigue pre-crack preparation the crack tip must be protected from the environment to
avoid any corrosion crack tip blunting, typically with a thin strip of tape or mastic.

3.6.1.3.2 The bath used to charge the specimens shall be the same of solution B of NACE TM 0284 (same
PH and H2S content). The same cares to prepare the solution for NACE TM 0284 tests shall be
taken, like how to wash the specimens shall be taken.

3.6.1.3.3 Total time of the specimens hydrogen charging shall be at least the minimum enough to saturate
the specimens at 1 bar. For instance, an exposure of a 25mm thick specimen for about a week is
required to give the maximum atomic hydrogen levels. Alternatively, hydrogen charging may be
done reaching a steady state representative of the maximum H2S partial pressure expected during
pipeline life if design may anticipate this"

3.6.1.3.4 The specimens shall be washed after been taken out of the bath. The time shall be the minimum
as possible. After washed they shall be chilled and stored at low temperature in a recipient with
liquid nitrogen to prevent the hydrogen diffusing out prior to testing.

AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA
 NO REV.
ESPECIFICAO TCNICA I-ET- 0000.00-6500-210-PSQ-001 0
PROGRAM SHEET:
GENERAL 10 de 10
TITLE:
ALTERNATIVE FLAW ACCEPTANCE CRITERIA OF SUBMARINE RIGID
PIPELINES AND RISERS GIRTH WELDS
3.6.1.3.5 Tests shall be carried out as soon as possible.

3.6.1.3.6 The time within been taken out of the recipient containing liquid nitrogen and the beginning of
the test shall be as minimum as possible.

3.6.1.3.7 Test temperature shall be as required by DNV OS-F-101 for operational case with +/- 2C
tolerance. Specimens shall be warmed up to test temperature to begin the test. Test shall begin as
soon as test temperature is stabilized.

3.6.1.3.8 Specimens shall be loaded at a very slow strain rate in order that the hydrogen can diffuse to the
crack tip during the test. The strain rate shall be at least an order of magnitude lower than that
conventionally used in fracture mechanics tests, and limited to maximum 0.008 mm / minute.

3.6.1.4 For the purpose of determination of the materials fracture resistance, for reeling conditions and any
other method that introduces large plastic strain, a J-R curve or CTOD-R curve shall be determined
for installation assessment. SENT specimens shall be tested for installation assessment according to
DNV RP-F108 and SENB specimens shall be tested for in service assessment. The segment
specimen testing shall be carried out according to DNV RP-F-108 at room temperature. Temperature
for installation fracture toughness tests shall be not greater than minimum installation design
temperature and temperature for in service assessment fracture toughness tests shall be not greater
than minimum in service design temperature.

3.6.1.5 No fracture toughness tests may be carried out without the presence of the COMPANY
representative, unless with written release. All the fracture toughness tests reports shall be presented
for COMPANY. All fracture toughness tests reports shall include: sample identification; welding
procedure specification number; test coupon identification; all measured test data to calculate fracture
toughness value; pictures of the pos-metallographic validation when required by BS 7448; all force x
load-line displacement, force x CMOD or equivalent graphs; characterization of the CTOD type if
applicable according to BS 7448 (CTODu, CTODm and CTODc); graph of J-R/CTOD-R curve if
required; all other information required in the test report according to BS 7448.

AS INFORMAES DESTE DOCUMENTO SO PROPRIEDADE DA PETROBRAS, SENDO PROIBIDA A UTILIZAO FORA DA SUA FINALIDADE.
FORMULRIO PERTENCENTE NORMA PETROBRAS N-381 REV G. A IMPRESSO OU REPRODUO DESTE DOCUMENTO TORNA A CPIA NO CONTROLADA