DNVGL-RP-F113 Edition November 2016

Transcription

1 RECOMMENDED PRACTICE DNVGL-RP-F113 Edition November 2016 Pipeline subsea repair The electronic pdf version of this document found through is the officially binding version. The documents are available free of charge in PDF format.

2 FOREWORD DNV GL recommended practices contain sound engineering practice and guidance. November 2016 Any comments may be sent by to This service document has been prepared based on available knowledge, technology and/or information at the time of issuance of this document. The use of this document by others than DNV GL is at the user's sole risk. DNV GL does not accept any liability or responsibility for loss or damages resulting from any use of this document.

3 CHANGES CURRENT General This document supersedes DNV-RP-F113, October Text affected by the main changes in this edition is highlighted in red colour. However, if the changes involve a whole chapter, section or sub-section, normally only the title will be in red colour. Changes current Main changes November 2016 Sec.1 Introduction General - harmonized with the 2013 version of DNV-OS-F101 (shall be published as DNVGL-ST-F101 in the future, and is generally used as reference in this edition for practical purposes). Repair in the context of a total integrity management system has been added. Update of damage/failure statistics. Scope and application have been extended to address the repair process, repair and preparedness strategy and lifecycle management. List of references have been updated. Sec.2 Basic philosophy Updated and extended. Sec.3 Pipeline repair activities Pipeline repair process, damage assessment and selection of repair method have been added. Ancillary equipment, surface preparation has been added. Preparedness strategy has been added. Sec.4 Pipeline design basis Lined and clad pipeline have been added. Dimensional tolerances and residual ovality guidelines, installation and reeling have been added. Sour service has been added. Sec.5 Pipeline exposures Design bending moment has been added. Safety factor has been moved to Sec.6 (Design). Sec.6 Fitting design Details regarding gripping and sealing have been moved to this section. Materials selection documentation has been added. Expanded table for load conditions. Safety factors for axial capacity and plug loads has been included. Radial expansion and compression have been clarified. Safety factors; safety factor relating to fatigue harmonized with the latest version of DNV-OS-F101 (2013). Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 3

4 Sec.8 Isolation plugs Isolation plug guidelines have been added. Sec.9 Hot tapping Hot tapping guidelines have been added. Sec.10 Above water tie-in Above water tie-in guidelines have been added. Sec.11 Welding General update and clarification based on input from participants. Weld procedure specification (WPS) (Location for mechanical sampling has been added). Changes current Sec.12 Testing Testing (moved from [C.1]), test requirements have been added. Sec.13 Integrity management of repair installation Life cycle management has been added. Sec.14 Documentation and quality assurance Guidelines on test documentation have been moved from the Appendix to section [14.1.6]. App.B Fitting capacity Appendix has been updated to be more generic (to also include in-line isolation plugs). App.C Typical tests More clarification included of the different pressure tests and test methods that are part of the fitting qualification programme. Leak test holding times and acceptance criteria have been added. New appendices: App.A - Code breaks and design factors. App.D - Stress analyses for fillet welds. App.E - Design resistance; welding on a pipe in operation. App.F - Calculation example - mechanical coupling axial locking capacity. App.G - Pipeline risk assessment and failure statistics. App.H - Guidelines - longevity of polymer seals. App.I - Recommended practice on the fatigue strength of pipes with ring marks in the base material. Editorial corrections In addition to the above stated main changes, editorial corrections may have been made. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 4

5 CONTENTS CHANGES CURRENT... 3 Sec.1 Introduction General Pipeline integrity management system Application of this recommended practice Pipeline failure causes Pipeline failure statistics Pipeline repair systems Scope and application Reference standards Relationship to other DNV GL documents and international standards DNV GL documents Other standards Bibliography Definitions Terms Abbreviations Symbols Verbal forms...22 Sec.2 Basic philosophy Safety philosophy Systematic review General Barrier philosophy pipeline isolation Risk assessments fitting design Limit state design and safety class methodology General Structural design criteria - fittings Alternative design of repair fitting load and resistance factor design method Calculation methods Qualification Philosophy Quality control by testing...28 Sec.3 Pipeline repair activities Damage assessment Pipeline defects and acceptance criteria Temporary and permanent repairs Local repair Pipeline section replacement Hot tapping General Removal of pipeline defects Bypass/branched connections Pipeline isolation Supporting activities General...35 Contents Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 5

6 3.8.2 Pipeline surface preparation Pipeline manipulation Lifting operations Remotely operated vehicle interfaces Instrument piping and components within a fitting assembly Preparedness strategy...36 Sec.4 Pipeline design basis General Pipeline design pressure Pipeline dimensions Pipe material Lined and clad pipelines Dimensional tolerances Welds and surface imperfections Linepipe As installed Extreme maximum and minimum diameter Statistical maximum and minimum diameters Electrical potential Environmentally assisted cracking...45 Sec.5 Pipeline exposures External pipeline forces Maximum axial forces Scenarios Free pipe end end cap (scenario A) Restrained pipeline (scenario B) Expansion loop effects (scenario C) Force boundaries Limiting displacements Design moment Fatigue...53 Sec.6 Fitting design General Failure modes and causes Material properties General Metallic materials Non-metallic materials Fitting strength capacity General Loads Load responses Inner diameter tolerances at the repair location Fitting grip capacity General Safety factors locking capacity Pipe wall utilization activation response General Local radial compression loads...61 Contents Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 6

7 6.6.3 Local radial expansion loads Seal capacity General Seal design capacity Environmental seal...66 Sec.7 Installation and attachment to the pipeline General Entry of fitting First end entry control Seal protection design Water block Second end entry Misalignment limitations Activation Seal test Monitoring and control General Monitoring of pipeline isolation Acceptance criteria...72 Sec.8 Isolation plugs General Design Failure modes Structural integrity of an isolation Testing of in-line isolation tools Installation and retrieval of in-line isolation tools Pigability assessment Pigging and setting In-line isolation period...76 Sec.9 Hot tapping General Hot tap fitting design Failure modes Structural integrity Sealing Testing...80 Sec.10 Above water tie-in General Design Failure modes Structural integrity Testing...83 Sec.11 Welding General Welding concept Hyperbaric welding General...85 Contents Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 7

8 Welding process (informative) Materials Pipe material Auxiliary component material Consumables Welding personnel Equipment and systems General Process monitoring and control Welding installation procedure for the equipment and system Equipment and systems qualification test Welding concept base cases qualification routes Qualification of both equipment and welding procedures Qualification of welding procedures only Preliminary hyperbaric welding procedure specification development Design Failure modes Allowable defect size Welding parameters development Arc stops Small-scale tests vs. full-scale tests Cooling rate Welding atmosphere Restraint Weld cracking Systematic sectioning Maximum defect size Repeatability Monitoring and control Preliminary welding procedure specification Welding procedure qualification Acceptance criteria Validity of welding procedures Production welding requirements General requirements Mobilization Documentation...99 Sec.12 Testing General Pressure testing Pipeline repair spool and pipeline system Pipeline repair fittings Qualification testing repair fitting Sec.13 Integrity management of repair installation Sec.14 Documentation and quality assurance Documentation General General documentation Qualification Contents Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 8

9 Design Manufacturing Testing Storage and transportation Installation Life cycle management Qualification checklist Quality assurance Traceability App. A Code breaks and design factors App. B Fitting capacity App. C Typical tests App. D Stress analysis of fillet weld App. E Design resistance; welding on a pipe in operation App. F Calculation example - mechanical coupling axial locking capacity App. G Pipeline risk assessment and failure statistics App. H Guidelines - longevity of polymer seals App. I Recommended practice on the fatigue strength of pipes with ring marks in the base material App. J Checklist for qualification Contents Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 9

10 SECTION 1 INTRODUCTION 1.1 General Revision history of this recommended practice (RP): this RP was first issued as DNV-RP-F104 Mechanical Pipeline Couplings in 1999, and was then replaced by DNV-RP-F113 Pipeline Subsea Repair in 2007, and DNVGL-RP-F113 Pipeline subsea repair in Pipeline integrity management system Pipeline subsea repair is an element within the overall integrity management (IM) system shown in Figure 1-1 and included in the last quarter of the continuous cyclical Integrity management process ; Plan - Risk assessment and IM planning Do Inspection, monitoring and testing Check Integrity assessment Act Mitigation, intervention and repair. The overall integrity management system and the integrity management process, including a more detailed description of these activities, are covered by DNVGL-RP-F116 Integrity management of submarine pipeline systems. Figure 1-1 Integrity management system, ref. DNVGL-RP-F116 Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 10

11 1.1.2 Application of this recommended practice This RP applies to fittings used to repair and tie-in submarine pipelines. These fittings include: couplings, clamps, T-branch connections and isolation plugs. Mechanical means connect these fittings to the pipeline, but sleeves/couplings and T-branches may also be welded. The RP provides guidance on pipeline repair methods such as pipeline hot tapping and above water tie-in. In addition, recommendations and guidelines on pipeline preparedness strategies, pipeline damage assessments and testing are given in this document. Figure 1-2 gives an overview of typical fittings covered in this document. Two pipe ends Coupling joining pipes Plug Repair coupling with flange adapter Local damage Clamp for local damage or joining pipeline T to connect a branch pipe to the pressurized pipeline Welded split sleeve repair clamp Figure 1-2 Typical installed repair fittings Couplings connect pipes by being directly attached to the pipe walls via mechanical or welded joints. Flange connectors differ from mechanical couplings as flanges join pipes via thick, machined pieces of additional material that are welded or forged onto the pipe ends prior to installation. Clamps are fitted externally to the pipeline to prevent leaks and/or add strength. Hot tap T-branch connections are fitted externally to the pipeline assembly during operation. A pressurized pipeline is machined open to allow fluid flow through the branch, or for the installation/removal of a hot tap penetration deployed isolation plug. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 11

12 Pipeline isolation plugs are pumped with the suitable fluid to the repair site and then activated in order to form an isolating barrier that can resist differential pressure, or the isolation plug can be installed locally via hot tap penetration. The pipe itself represents the key member of the repair assembly, with consequential limitations such as, but not limited to, pipe wall strength, surface irregularities and deviations in shape. Pipeline repair fittings must be installed with caution to reduce the likelihood of damage, e.g. seal damage. The fitting's coupling strength shall be sufficient to resist stresses from all relevant loads, within a factor of safety as defined in [6.5.2]. The section [6.5] on the strength of the mechanical attachments is also applicable to pipeline recovery tools. The given load and resistance factors to be accounted for in the fitting strength capacity are based on pipelines designed and manufactured according to DNVGL-ST-F101. For the repair of pipelines designed according to other standards, the design factors applied must be assessed. Some fittings will be a permanent part of the repaired pipeline, while others (like isolation plugs) are installed temporarily to enable the repair process, see [3.3]. Further, an overview of typical pipeline repair project activities, including the root cause and integrity assessment of the damage, selection of the repair method and pipeline repair strategy (Sec.3), and lifecycle management of repairs (Sec.13 and [14.1.9]), is provided in this RP Pipeline failure causes Pipeline damage that may require repair is typically caused by degradation mechanisms such as internal and external metal loss due to corrosion or hydrogen induced stress cracking (HISC), or by events such as unstable seabed conditions, anchor hooking, trawl gear interference and objects dropped from the surface. The risk of damage depends on the intensity of surface activities such as ship transport and offshore operations as well as the depth, seabed conditions and design and operation of the pipeline itself. The extent of possible damage will vary from insignificant to a fully buckled or severed pipeline. Consequently, the repair and repair preparedness strategy must be based on all these factors Pipeline failure statistics Pipelines are accepted as a safe form of energy transportation and the industry has many years of operational experience. However, pipeline failures do occur. Learning from pipeline failures is important and can help to reduce future failure risks through the implementation of mitigating measures such as a repair and repair preparedness strategy. The failure probability and type vary in different parts of the world, depending on design philosophies and risk exposures, and need to be accounted for when utilizing the data. Reports on public pipeline failure statistics are limited, but some of the available data are presented below: The Pipeline and Riser Loss of Containment (PARLOC) report, /2/ and /3/, prepared and managed by Oil & Gas UK and the Energy Institute (EI), presents the pipeline failures and incidents in the North Sea UK sector reported in the periods up to 2001 in /2/ and for the period in /3/. A summary of the failure statistics from PARLOC /3/ is given in App.G. The CODAM (Corrosion and Damage) database managed by the Norwegian Petroleum Safety Authority (PSA) presents reported incidents and failures in all pipeline systems for the exploitation and transmission of oil and gas on the Norwegian continental shelf, including transmissions from Norway to the UK and the European Continent. The database also covers all types of offshore cables used in the petroleum activities and is continuously updated. DNVGL-RP-F116 Appendix A reports that most of the reported pipeline damage is caused by corrosion, which accounts for 27% of the reported incidents in the North Sea and 40% in the Gulf of Mexico, see Figure 1-3. In the Gulf of Mexico and North Sea respectively, 85% and 45% of the corrosion problems are related to internal corrosion. In addition, fitting, flange and valve failures are a major problem. The pipeline damage statistics vary for different parts of the world, but this RP on pipeline repair is applicable worldwide. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 12

13 Figure 1-3 All reported incidents in percentages for a) the North Sea and b) the Gulf of Mexico. (Incidents connected to fittings and valves in the North Sea, which were app. 30% of the total reported incidents, are not included in these statistics.) Pipeline repair systems Figure 1-4 illustrates the complexity of a subsea pipeline repair. Figure 1-4 Typical support system for a pipeline repair Historically, shallow water repairs have mostly been performed by divers. However, the water pressure limits human hyperbaric intervention to a water depth of a few hundred metres due to the human physiology. National authorities further regulate this type of diving to more shallow depth limits as a means to safeguard the divers. A preparedness strategy should take into account the possibility of restrictions on the allowable diving depth during the design life of the pipeline system. The maximum allowed diving depth varies, depending on governing legislation and company requirements, but is typically in the range of m. This depth limit is small compared to the deep water pipelines. Consequently, pipeline repairs in deeper waters typically have to be carried out based on remotely controlled techniques. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 13

14 A pipeline repair process in general requires a range of planning and investigation work: Investigate the damage, the pipe's condition and the consequences for the pipeline operation, i.e. will any repair be required? Are mitigations such as a restriction on operating pressure and/or flow required? Should pollution counter-measures be started? Should water ingress to the pipe be limited? Plan the uncovering and seabed preparation for the repair, including calculations of the consequent pipeline response. Plan the repair operation based on the state of emergency and the results of the investigations. (Planning, ordering of equipment and support). Perform the seabed preparations, pipeline pressure adjustments and repair. Test to confirm the repair quality and protection of the repaired section, clean up and finish. Develop a recommended plan for the future in-service inspection, verification and correction of the repair site and associated equipment. 1.2 Scope and application This RP is intended to provide criteria and guidelines for the qualification of fittings and systems used for pipeline subsea repair, modifications and tie-ins. It includes aspects relating to the design, manufacture, installation, testing and operation of such fittings and systems. Guidelines on procedures and reference standards for preparedness strategy, damage assessment and lifecycle management of repair are also provided. This RP is intended to be used as a supplement to DNVGL-ST-F101 Submarine pipeline systems, and is therefore also applicable to some types of risers and to topside and onshore parts of the submarine pipeline system. An overview of the systems covered by this RP is shown in Figure 1-5. The latest revision of the referenced standards shall be used. Figure 1-5 Overview of the scope covered by this recommended practice Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 14

15 1.3 Reference standards Relationship to other DNV GL documents and international standards DNV GL documents and international standards is used as a common designation of standards, rules, recommended practices, guidelines etc. DNV GL documents and international standards are all referred to throughout the documents by its document code, e.g. ISO while the bibliographies (papers and reports) are referenced by numbers as listed in 1.6. The documents listed in Table 1-1 and Table 1-2 include references to yet other DNV GL documents and international standards which, through reference in the text constitute either a compulsory or informative reference of this document. Guidance note: Normative references are typically referred to as testing shall be performed in accordance with ISO xxx, while informative references are typically referred to as testing may be performed in accordance with ISO xxx or ISO yyyy, or recommended practice for testing is given in DNV-RP-F xxx. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- In case of conflict between requirements of this document and a referenced DNV GL documents, the requirements of the document with the latest revision date shall prevail. Any conflict is intended to be removed in next revision of that document. Where reference is made to international standards other than DNV GL documents, the valid revision should be taken as the revision which was current at the date of issue of this document. Detailed references to sections or appendices of given DNV GL documents and international standards applies to the revision valid at the issue of this document. DNV GL is in a transition period w.r.t. rebranding of standards. Some of the listed DNV GL documents refer to document codes that will be introduced during 2016/ DNV GL documents Table 1-1 DNV GL documents Document code DNVGL-RP-A203 DNVGL-RP-C203 DNVGL-RP-C208 DNVGL-RP-F101 DNVGL-RP-F105 DNVGL-RP-F110 DNVGL-RP-F111 DNVGL-RP-F116 DNVGL-ST-E273 DNVGL-ST-F101 DNVGL-ST-N001 Title Technology qualification Fatigue design of offshore steel structures Determination of structural capacity by non-linear FE analysis methods Corroded pipelines Free spanning pipelines Global buckling of submarine pipelines Interference between trawl gear and pipelines Integrity management of submarine pipeline systems Portable offshore units Submarine pipeline systems Marine operations and marine warranty Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 15

16 1.5 Other standards Table 1-2 Other standards Document code API RP 2201 EN IMCA D 044 Title Safe Hot Tapping Practices in the Petroleum & Petrochemical Industries Welding. Recommendations for welding of metallic materials. General guidance for arc welding Guidelines for Isolation and Intervention: Diver Access to Subsea Systems IMCA (International Marine Contractors Association ISO Petroleum and natural gas industries Design and operation of subsea production systems - Part 1 General requirements and recommendations (2005) ISO Petroleum and natural gas industries -- Design and operation of subsea production systems - Part 7: Completion/workover riser systems ISO Petroleum and natural gas industries -- Design and operation of subsea production systems - Part 8: Remotely Operated Vehicle (ROV) interfaces on subsea production systems ISO Welding personnel Approval testing of welding operators for fusion welding and of resistance weld setters for fully mechanized and automatic welding of metallic materials ISO Destructive tests on welds in metallic materials Cold cracking tests for weldments Arc welding processes Part 2: Self-restraint tests ISO Non-metallic materials in contact with media related to oil and gas production, Part 1: Thermoplastics; Part 2: Elastomers; Part 3: Thermosets NORSOK M-710 Qualification of non-metallic sealing materials and manufacturers NORSOK U-001 Subsea Production System, October 2015 The latest revision of the DNV GL standards may be found in the publication list at the DNV GL website Amendments and corrections to the DNV GL standards are published bi-annually on These shall be considered as a mandatory part of the above standards. 1.6 Bibliography The following documents are referenced in this RP: /1/ CODAM: Pipeline damages Damages and incidents, Petroleumstilsynet (PSA) Norway /2/ PARLOC 2001: The update and loss of containment data for offshore pipelines, HSE UK /3/ PARLOC 2012: Pipeline and riser loss of containment, , HSE UK [NEW] /4/ Pipeline Defect Assessment Manual (PDAM) /5/ DNV GL report , Rev. 01. PDAM update Assessment of subsea pipeline defects /6/ OREDA - Offshore reliability handbook /7/ Torselletti, Enrico, et al, Submarine Pipeline Installation JIP:Strength and Deformation Capacity of Pipes Passing Over The S-Lay Vessel Stinger. Hamburg: ASME, OMAE /8/ DNV GL Report No An improved CWM platform for modelling welding procedures and their effect on structural behaviour /9/ Research Report 485, Elastomeric seals for rapid gas decompression applications in high pressure services, BHR Group Limited for the Health and Safety Executive 2006 /10/ HSE Research Report, Elastomers for fluid containment in offshore oil and gas production: Guidelines and review. /11/ Cola, M. J., Kiefner, J. F., Fischer, R. D., Jones, D. J., and Bruce, W. A., Development of Simplified Weld Cooling Rate Models for In-Service Gas Pipelines, Project Report No. J7134 to A.G.A. Pipeline Research Committee, Edison Welding Institute, Kiefner and Associates and Battelle Columbus Division, Columbus, OH, July /12/ Bruce, W. A., Li, V., Citterberg, R., Wang, Y.-Y., and Chen, Y., Improved Cooling Rate Model for Welding on In-Service Pipelines, PRCI Contract No. PR , EWI Project No CAP, Edison Welding Institute, Columbus, OH, July /13/ Zecheru, G., Barsan, M. F., Draghici, G., Dumitrescu, A. Factors determining the occurrence of the burn through phenomenon when repairing by welding the hydrocarbons transmission pipelines. Sudura, vol.24, no Pp.17-27, /14/ Karjadi, E.; Smienk, H.; Boyd, H.; Aamlid, O. Extended Reel-Ability of New Aegir Reeling Vessel Based on Reliability Based Assessment and Bending Tests Program, OMAE , Rio de Janeiro. /15/ Karjadi, E.; Boyd, H.; van Rooijen, R.; Demmink, H. and Balder, T. Development on Aegir Reeling Pipeline Analyses by Test Validation, OMAE Nantes. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 16

17 1.7 Definitions Terms Table 1-3 Terms Term barrier clamp confidence coupling critical parameter failure failure mechanism failure mode failure probability fittings function functional specification guidance note limiting parameter locking locking capacity margin (to failure) model parameter performance pipeline fluid Definition physical isolation of pipeline pressure at specified location (e.g. valve or isolation plug) Barriers shall be established that: a) reduce the probability of failures and hazard and accident situations developing, b) limit possible harm and disadvantages. Where more than one barrier is necessary, there shall be sufficient independence between barriers. circumferential structural element, split into two or more parts Examples; connecting two hubs in a mechanical connector or two pipe half-shells for repair purposes. in statistics, confidence is a measure of the reliability of an estimate and can be quantified by standard methods In this recommended practice, the term confidence is used in a broader sense to denote the level of confidence a decision-maker has in a technology. Such confidence is usually measured when decisions are made, e.g. whether to invest in the technology, whether to implement it, and whether to subject it to certain qualification activities. mechanical device to connect two bare pipes to create a structural joint that resists loads and prevents leakage parameter that can lead to an unacceptable level of failure or risk, either alone or in combination with other parameters that have a similar level of risk or failure loss of an item's ability to perform the required (specified) function within the limits set for the item's intended use. This occurs when the margin to failure is negative physical, chemical, temporal or other process that leads or has led to a failure observed manner of failure (on a specified level) probability of failure occurring within a specified time period, or at a specified condition (e.g. when an engine starts) in this publication, the term fittings is limited to: couplings, clamps, hot tap T-branch connections and isolation plugs intended for submarine pipelines, including isolation plugs installed through hot tap penetration, either in-line remote or umbilical controlled purpose for which something is designed or exists performance that a technology has to achieve within the set environment and operational conditions clarification and interpretation note to text given in the paragraph above the guidance note parameter that has a specified minimum and/or maximum qualified value (e.g. misalignment angle, off-set value, design load, contact force or pressure) mechanical or welded firm connection strength of the attachment to the pipe difference between the utilization at failure and the required utilization in the intended use When either the utilization at failure or the required utilization is uncertain, so is the margin. The margin can then be represented by its probability distribution. The performance margin and safety margin are special cases. mathematical description or experimental set-up simulating a function of the technology Models take account of relevant effects of the critical parameters for the modelled function. determining characteristic of the technology s function, use or environment performance of a technology is its ability to provide its specified functions These functions contribute to safety/reliability as well as the output or value generated by the system, equipment or component when in operation. any fluid which may be transported by the pipeline during the commissioning and operational phases, including corrosion inhibitors, flow assurance chemicals and well clean-out fluids Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 17

18 Table 1-3 Terms (Continued) Term qualification programme reliability risk Definition 1) Successive performance of the basic technology qualification process with incremental qualification milestones at increasing levels of detail. 2) Successive performance of the basic technology qualification process for qualification with incremental improvement of the qualification limits. ability of an item to perform a required function under given conditions for a given time interval or at a specified condition. In quantitative terms, it is one (1) minus the failure probability effect of uncertainty on objectives, in a safety context the combination of the probability of harm occurring and the severity of that harm, see App.G Guidance note 1: Further definitions relating to risk may be found in ISO e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 2: Objectives can have different aspects, such as financial, health and safety and environmental goals and can apply at different levels, such as strategic, organization-wide, project, product and process. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- safety safety margin safety factor/ factor of safety sleeve type coupling spool substantiate technology technology qualification state of being safe A technology is safe if it will not fail under foreseeable demands and lead to loss of life, injury, negative environmental impact, or unacceptable economic loss; and if it is unlikely to fail under extraordinary circumstances. difference between capacity and demand, e.g. load effect When either the capacity or demand is uncertain, so is the margin. The margin can then be represented by its probability distribution. factor by which the characteristic value of a variable is modified to give the design value (i.e. a load effect, condition load effect, material resistance or safety class resistance factor), see Sec.6 coupling enclosing the pipe as a sleeve pipe section which is used to connect a pipeline to another subsea structure, e.g. manifold, PLET, tee, or riser to demonstrate for a defined context through evidence and argument way to provide a function (such as by combining methods, techniques, skills, equipment, tools or materials) process of providing the evidence that technology will function within specified limits with an acceptable level of confidence Guidance note: Technology qualification can be seen as the process of substantiating a claim about the provision of a function which is not already covered by validated requirements. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- technology qualification basis technology qualification programme technology qualification plan tee threat uncertainty validate benchmark against which the success of the technology's performance is measured framework in which the technology qualification process is executed qualification activities specified with the purpose of generating qualification evidence and the logical dependencies between the individual pieces of qualification evidence pipeline branch connection potential risk with significant uncertainty about the consequence of failure or likelihood of occurrence that requires further investigation to either quantify it as a risk or remove it from further consideration state of having limited knowledge that makes it impossible to exactly describe the existing state or future outcome(s) to substantiate that something is relevant and complete Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 18

19 1.7.2 Abbreviations Table 1-4 Abbreviations Abbreviation AUT AWTI CP CRA DBB DBM EAC ECA ESDV FAT FEA FKM FL FMEA FMECA FTA GMAW GTAW HAC HAZID HAZOP HICC HISC HNBR HWPQR HWPS IM LAT LRFD MIC NBR NDT NS P&ID PEEK ph phwps POF PTFE PV standard QRA ROV SIMOPS SIT Full text automated ultrasonic testing above water tie-in cathodic protection corrosion resistant alloy double block and bleed double block and monitor environmentally assisted cracking engineering critical assessment emergency shutdown valve factory acceptance test finite element analysis fluoroelastomers fusion lines failure mode and effect analysis failure mode, effect and criticality analysis fault tree analysis gas metal arc welding tungsten, inert gas arc welding hydrogen assisted cracking hazard identification study hazard and operability study hydrogen induced cold cracking hydrogen induced stress cracking hydrogenated nitrile butadiene rubber (polymer seal material) hyperbaric welding procedure qualification record hyperbaric welding procedure specification integrity management lowest astronomic tide load and resistance factor design microbially induced corrosion nitrile rubber or acrylonitrile butadiene rubber (polymer seal material) non-destructive testing Norwegian standard process & instrument diagram polyether ether ketone (thermoplastic seal material) numeric scale used to specify the acidity or alkalinity of an aqueous solution preliminary hyperbaric welding procedure specification probability of failure polytetrafluoroethylene (seal material) pressure vessel standard quantitative risk assessment remotely operated vehicle risk assessment related to simultaneous marine operations system integration testing Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 19

20 Table 1-4 Abbreviations (Continued) Abbreviation SLS SMYS SRA SSC TBA TS ULS WIP YS Full text serviceability limit state specified minimum yield stress structural reliability analysis sulphide stress cracking to be assessed tensile strength: the measured tensile strength ultimate limit state welding installation procedure yield stress: the measured yield tensile stress Symbols Where used in this document, the following symbols are defined as: Table 1-5 Symbols Symbol Definition A e π external pipe cross-section area: 4 D2 A i π internal pipe cross-section area: (D 2 t)2 4 A s pipe steel cross-sectional area: π (D-t) t a misalignment angle (radians) b misalignment between the pipe ends (radians) D outside pipe diameter D c couplings/sleeve bore diameter E modulus of elasticity e diametric clearance (considering constant internal diameter): D c -D e s straightness of the pipe section of concern within the length of the fitting e f change in diameter due to tension force e o out of roundness (OOR, ovality) tolerance, see equation (4.5) e R residual ovality due to bending over lay vessel stinger e l straightness tolerance for the pipe section of concern e m tolerance combination e p change in outer diameter of pipe e c change in internal diameter of coupling/sleeve respectively E t defined in text as either: e tm f y,temp f u,temp f cb f y 1) external diameter tolerance 2) shrink fit shrink fit produces a contact pressure, which generate a fraction of yield stress of pipe de-rating on yield stress to be used in design due to temperature in excess of 50 C for C-Mn and 13Cr steels, and in excess of 25 C for 22Cr and 25Cr steels de-rating on ultimate strength to be used in design due to temperature in excess of 50 C for C-Mn and 13Cr steels, and in excess of 25 C for 22Cr and 25Cr steels characteristic burst material strength yield stress to be used in pipeline design according to DNVGL-ST-F101 =, f u tensile strength to be used in pipeline design according to DNVGL-ST-F101 =, Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 20

21 Table 1-5 Symbols (Continued) Symbol H l Definition pipeline residual lay-tension force length of fitting/sleeve/coupling for: l L < 0.5 k r k d L m M N N D/(14 t); (for calculation of residual ovality) 1 + (D-0.6)/2; (for calculation of residual ovality) defined in text as either: 1) length of linepipe section (normally 12 m) or specified section 2) length of contact surface between sleeve and pipe 3) load effect (in context of Load and Resistance factors). gravity force of pipe with internal fluid and possible concrete per unit length, i.e. combined weight and buoyancy local pipeline design bending moment pipe wall axial force, i.e. the axial force as imposed on the coupling (tension is positive) N (As fy) N' pt N' o N' a N' b N' c Numbers pressure test operation pressure effects only restrained pipe case, either in compression or tension pipe in a curve - moving cases 1, 2, 3 and 4 are related to: expansion forces due to: 1) pressurized and hot, 2) de-pressurized and hot, and contraction forces due to: 3) pressurized and cold, 4) de-pressurized and cold n axial length from the coupling entrance to the end of the same inner diameter. (Length of equal internal diameter) internal pressure external pressure internal pressure difference relative to as-laid, or P i,operation local incidental pressure, see equation (4.1) total probability of pipeline failure P i,depressurized h probability of pipeline failure from natural uncertainties in design loads and load bearing capacities probability of pipeline failure from accidental events probability of pipeline failure from gross errors during design, fabrication, installation, and operation probability of pipeline failure from unknown phenomena R defined in text as either: R s s S S (p i ) S ΔT T o T m t α 1) average bending radius of pipeline curvature 2) resistance (in context of load and resistance factors) point load from the stinger rollers defined in text as either: straightness of a pipe/section specified in % of L. Safety distance (say 0.3 mm) to compensate for deflections and possible protrusions on the pipe end effective axial pipeline force, i.e. forces transferred by soil friction, supports etc. (tension is positive) effective axial force, as function of internal pressure S (As fy) temperature-difference operational temperature for sleeve make up temperature wall thickness temperature-expansion coefficient Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 21

22 Table 1-5 Symbols (Continued) Symbol α fat Definition allowable damage ratio for fatigue α u material strength factor according to DNVGL-ST-F101: 0.96 for normal materials 1.00 for materials to supplementary requirements U β 130 (see [4.6.2]), ε b σ σ eq, nom γ γ 1 γ F γ E γ A γ p γ c γ 2,,, γ m γ mw γ e ν μ x 1 x 2 y 1 y 2 y i total nominal longitudinal strain. The nominal strain is the total engineering strain not accounting for strain concentration factors. bending strain [D/(2 R)]; stress the equivalent stress averaged over the thickness resistance and load- and load effect factors with the following notifications: load factors load effect factor for functional loads load effect factor for environmental loads load effect factor for accidental loads pressure loads factor condition load effect factor resistance (capacity) factors safety class resistance factor, pressure containment safety class resistance factor, local buckling safety class resistance factor, strain material factor weld material factor resistance strain factor usage factor Poisson s ratio friction coefficient simulating lateral soil resistance density of the sea water sleeve eccentricity (offset from centre line) at entrance offset between pipe ends overlap length i.e. degree of sleeve displacement over the pipe(s) at the moment of time considered. Maximum y 1 is the length of the coupling. half coupling length (bridging one pipe end) distance from the coupling entrance to the seal Verbal forms Table 1-6 Verbal forms Term shall should may Definition verbal form used to indicate requirements strictly to be followed in order to conform to the document verbal form used to indicate that among several possibilities one is recommended as particularly suitable, without mentioning or excluding others, or that a certain course of action is preferred but not necessarily required verbal form used to indicate a course of action permissible within the limits of the document Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 22

23 SECTION 2 BASIC PHILOSOPHY 2.1 Safety philosophy The pipeline repair shall comply with the safety philosophy of DNVGL-ST-F101. Safety is ensured by the safety hierarchy shown in Figure 2-1. Figure 2-1 The safety hierarchy stated in DNVGL-ST-F101 The elements of the figure are outlined as follows: Safety objective An overall safety objective shall be established, planned and implemented, covering all phases from conceptual development until abandonment. Systematic review A systematic review shall be carried out in all phases to identify and evaluate threats and the consequences of single failures and series of failures in the pipeline repair system, such that necessary remedial measures can be taken. Safety class methodology/limit state criteria The safety of the pipeline repair system is ensured by the use of a safety class methodology. The pipeline system is classified into one or more safety classes based on failure consequences. QA/QC The safety format requires that gross errors (human errors) shall be controlled by requirements for the organization of the work, competence of persons performing the work, verification of the design, and quality assurance during all relevant phases. This RP complies with the above structure based on the combination of design criteria, i.e. safety class methodology, and tests, i.e. part of the QA/QC. The implementation of this repair philosophy is described below. 2.2 Systematic review General A systematic review shall be carried out in all phases to identify and evaluate threats and the consequences of single failures and series of failures in the repair fitting and pipeline section at the repair location, such that necessary remedial measures can be taken. The extent of the review or analysis shall reflect the criticality of the pipeline and repair system, the criticality of a planned operation, and previous experience with similar systems or operations. The uncertainty in the applied risk review model itself shall also be assessed. Typical methodologies for identifying potential hazards are failure mode and effect analysis (FMEA) and hazard and operability studies (HAZOP). For a general overview of pipeline risk assessment methods, see App.G. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 23

24 2.2.2 Barrier philosophy pipeline isolation DNV GL standards give no specific requirements as to the number of barriers for pipeline systems. However, the safety level of the barrier(s) shall comply with the requirements given in DNVGL-ST-F101, see Table 2-1. The general requirement applied by the industry, in compliance with the above criteria, is to have two barriers during pipeline isolation, where no single failure shall result in the loss of both barriers, unless it can be documented that one barrier alone is as safe as the pipe-wall itself. Single barrier isolation shall be approved by the operator and by the relevant involved contractors exposed to the environment at the isolated pipe end. Monitoring of the pressure between the two barriers is required to verify the performance of both barriers. Each of the barriers shall be demonstrated to retain the full isolation pressure alone. Hence, the requirement of two barriers implies two independent and tested barriers. If two barriers are dependent, a single failure may cause the failure of both - so sufficient redundancy must exist if barriers are dependent on a shared component. Shared components in the isolation barriers can be accepted if they offer the same level of safety and redundancy as permanent pipeline fittings. For each project, each of the barriers shall be tested to a specified differential pressure at the set location as part of the plug installation procedure. The performance of each barrier shall be monitored throughout the isolation period, e.g. by monitoring the annulus, HP (high pressure) and LP (low pressure) locations, i.e. Double Block and Monitor (DBM). This may be provided by e.g.: isolation plug(s) providing two independent and tested barriers a combination of one isolation plug and one isolation valve, e.g. ESDV. Guidance note: For plug trains including a hydrostatic test plug, where failure of the hydrostatic test plug may cause failure of the hydrocarbon barrier plug(s), the hydrostatic test plug should be designed with the same safety class level as the hydro-carbon plug(s). ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- When tethered isolation plugs are used, the void between the plug seals can be vented through the umbilical. The term DBB (Double Block and Bleed) can thus be applied to tethered, umbilical-controlled isolation plugs. However, venting and monitoring through an umbilical increases the monitored volume, and thereby reduces the sensitivity to detect a leak from a measured pressure drop. The pressure monitoring system's acceptable sensitivity for detecting potential leaks through the barrier(s) shall be documented. The stored energy (volume x pressure) in the monitored void should be minimized as this volume is isolated by a single barrier. The term DBM applies to pipeline isolation plugs, as only the isolation status is monitored, while a bleed, as in DBB, would imply tapping into the pipeline between the barriers and bleeding content to a safe zone. In-line isolation using DBM is recognized and accepted by the industry. Hot tap branch installed isolation plugs are mechanically connected to the branch, where the locking failure probability is compliant with governing regulations and recognized pressure vessel standards, and equivalent to a valve. Provided the safety level of the structural connection locking the plug to the branch complies with governing regulations and recognized pressure vessel standards - one plug comprising two independent seals, with pressure monitoring of the annulus and both sides of the plug - the safety level may be accepted as equivalent to a double block and monitor (DBM) barrier. Guidance note: To some applications, such as hot tap valves, DBM can be acceptable for valves. However, erroneous operation of the valve can cause loss of both barriers. For valves installed to provide double barrier, procedures are required to mitigate erroneous operation that results in unintended loss of barriers. This can be obtained by providing two operational barriers; first barrier will be installation of the actuator, and the second barrier is to operate the actuator. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 24

25 2.2.3 Risk assessments fitting design The design envelopes and operational limits for the fitting shall be specified in the design basis. The design and qualification of fittings and other equipment used for pipeline subsea repair shall demonstrate acceptable margins to failure for all potential failure modes, covering all phases (i.e. testing, installation, commissioning and operation). Failure modes of concern shall be identified though a systematic approach. 2.3 Limit state design and safety class methodology General The pipeline's structural integrity shall meet the safety level requirements of DNVGL-ST-F101 during all phases of a pipeline repair. DNVGL-ST-F101 quantifies the consequence of a failure in terms of three safety classes; Low, Medium and High. Each safety class has a specified nominal target failure probability, see Table 2-1. This also applies to repair fittings and the corresponding safety factors are calibrated against these nominal target levels per defined safety class. Table 2-1 Nominal annual target failure probabilities vs. safety classes Safety class Annual POF criteria per pipeline system Low (insignificant risk of human injury): POF 10-3 Medium (low risk of human injury): POF 10-4 High (risk of human injury): POF 10-5 For temporary repair periods of less than one year, the probability of failure during the relevant time period shall be compared to the annual target failure probability. The aggregated failure probability of the considered pipeline section, including the contribution from temporary installed repair equipment such as an isolation plug, shall comply with the criteria given in Table 2-1. Pressure-retaining repair equipment installed for less than one year will only contribute to the failure probability during the installed period, and the annual failure probability contribution from this equipment can be reduced by the fraction of one year for which it is installed. The allowable pipeline utilization factors given in this RP are harmonized with the safety format given in DNVGL-ST-F101. Generally, the safety class Low is sufficient to apply to installations on unpressurized pipes and during testing (provided the content during testing is water). The safety class Medium or High is required during operations involving hydrocarbons, depending on the location being considered. For a temporary repair installed for a limited period, the safety class may be reduced by one level, e.g. from High to Medium, based on a risk assessment. The repair/modification of an oil or gas pipeline may include a temporary opening to the environment or the opening of a pressurized pipeline, e.g. in conjunction with a plug operation and hot tapping. These temporary operational phases should satisfy the safety class Medium criteria when the consequences for the environment (pollution, personnel) of a failure would be comparable to those from a leaking pipeline. For less severe consequences, the temporary phase safety class may be Low. Repair fittings made for contingency purposes for a pipeline system need to comply with the highest safety class level within the pipeline section(s) that the repair system is to cover Structural design criteria - fittings The safety target levels given in Table 2-1 are achieved through the following activities: a systematic review shall be carried out during all phases to identify and evaluate threats, ensuring that all failure modes and mechanisms are identified. the strength of the fittings' pressure-retaining parts shall be according to recognized pressure vessel Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 25

26 design standards or the alternative approach, see [2.3.3] below. Requirements related to materials, fabrication and the mechanical testing of pipeline components given in DNVGL-ST-F101 shall be complied with. design loads for fittings are defined in Sec.5, [6.4.2] and [6.4.3]. functional requirements with respect to locking capacity (gripping) and sealing capacity are covered in [6.5] and [6.7] respectively. safety factors for: temporary and permanent repairs are covered in [3.3], [3.5] and [6.5.2]. Examples are given in App.A. local pipe wall utilization at the interaction with the fitting is covered in [6.6]. Guidance note: All pressure-containing components used in the submarine pipeline system should represent at least the same safety level as the connected riser/pipeline section. This is normally achieved by applying an internationally recognized pressure vessel standard for pressure-retaining parts of the repair fitting. Examples of relevant pressure vessel standards are ASME VIII division 2 Part 5, PD 5500 and EN e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Structural design standards may apply to fitting components whose failure is not related to the pressure containment and functionality of the pressure barrier. The code break between the DNVGL-ST-F101 pipeline standard and the fitting design standards shall be established based on requirements detailed in DNVGL-ST-F Alternative design of repair fitting load and resistance factor design method As an alternative approach to designing and testing the repair fitting according to a recognized pressure vessel standard, the load and resistance design methodology applied in DNVGL-ST-F101 may be used to document the acceptable safety level of the repair fitting (i.e. the same safety level as the connected pipe section), through the following steps: The design load factors given in Table 6-3 that apply to the pipeline may also be applicable for the design of repair fittings, provided an acceptable safety level is documented through a technology qualification process according to DNVGL-RP-A203 or the equivalent (see the guidance note below). The design resistance factors given in Table 6-3 apply to the pipe geometry. Fittings with material specifications, geometry and load paths different to the pipe section may have additional failure mechanisms that need to be reflected in the applied resistance factors and acceptance criteria. All potential failure mechanisms of the repair assembly shall be identified and assessed in order to document that the design resistance of the installed repair fitting exceeds the design resistance of the connected pipe sections for all relevant design load combinations and design lives. The design resistance factors are related to the uncertainties, and the confidence levels are related to the mechanical properties of the material. The mechanical properties specified in the design premises for pipelines are documented through a combination of manufacturing requirements, mill test per linepipe and hydrostatic pressure testing after installation, as specified in DNVGL-ST-F101. For repair fittings, the following requirements apply in order to comply with an equivalent safety philosophy: The manufacturing of the repair fitting shall comply with the manufacturing requirements for pipeline components given in DNVGL-ST-F101. The acceptable safety level of the fitting's sealing and gripping capacity shall be documented by design requirements as specified in [6.5] and [6.7]. The repair fittings shall be tested according to requirements given in Sec.12. To document the acceptable installation of the repair fitting at the repair location, the sealing performance shall be documented by a seal test, or alternatively by hydrostatic pressure testing of the repaired section, see Sec.12. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 26

27 Guidance note: The acceptable safety level can be documented according to methodology given in DNVGL-RP-A203, through a risk assessment process identifying potential threats, failure mechanisms and modes, and then through a qualification programme documenting that the safety level exceeds that of the connected pipe section. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Calculation methods An analytical/numerical approach should be applied as the main tool to enable qualification. This type of approach will establish the individual and combined effects of the relevant parameters. Guidance note 1: Finite element analysis (FEA) may be used for the detailed study of stresses and deflections from symmetrical and asymmetrical loads, including material plastic yielding, friction, contact, collapse and motions, i.e. a combination of a range of non-linear effects. A theoretical model including all parameters and effects will be complex to use, so it is often more practical to apply simpler models for analysing separate parameter effects. Such models can also be studied by FEA, and/or by simplified analysis. Engineering mathematical software is the most convenient tool for handling simplified analysis. The advantage of using mathematical software rather than a spreadsheet is that the method (formulas) is easily documented. Spreadsheets are widely used, but require additional documentation of the formulas which are actively used in the computations. Simplified analysis may be used when the behaviour is understood and the computation model is representative. However, it may be difficult to cover all the relevant combinations and effects of the simplified models with acceptable confidence, and the use of nonlinear FEA may be more appropriate. Elastic - formulas Formulas can be developed either by derivation from textbooks or based on test results. Software with formulas from some textbooks, such as Roark s formulas for stress and strain, is available. These formulas are limited to elastic analysis. Elastic plastic - formulas Formulas for plastic yield can be developed, but normally require calibration by testing and/or FEA. Practical applications would be to establish the possible plasticity of the pipe shell, both through the wall and by hard bodies (seals and grips) forced into the surface. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 2: DNV-RP-C208 provides guidance on procedures for non-linear FE analyses. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Qualification Philosophy The qualification of fittings should, in general, be based on verification of compliance with given functional specifications and the safety margin against possible failure modes. Reference is also made to DNVGL-RP-A203. This publication gives general guidance on the qualification of both new technology and proven technology. The fitting design and/or qualification basis, depending on the level of new technology involved, shall specify the limitations, conditions and design envelopes for the fitting application. The margins to failure on functional requirements, e.g. sealing and gripping, shall be documented in compliance with this RP, covering the qualified envelope of the relevant governing critical parameters, such as structural capacity, deformations, extrusion gaps, relative hardness between the pipe and gripping segments, friction factors and resistance to HISC. This qualification should be based on the following principles: 1) functional requirements shall be quantitative 2) possible failure modes shall be identified (See [6.2], Sec.7, [8.2.1] and Sec.11) 3) theoretical analysis/calculations shall be used as the main tool to document fulfilment of the functional specifications and safety level criteria against failures. The theoretical calculations should be validated by tests 4) for new or unproven technology, the safety factors shall be established based either on recognized standards, or on combinations of all the uncertainties and inaccuracies used in the data, operation, calculations and tests. This applies to loads, strength, sealing and function. (Acceptable failure probabilities versus safety class are defined in Table 2-1) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 27

28 5) measurements and tests shall be used as the main QA/QC tools to document that the manufactured product complies with the functional specifications 6) a systematic approach shall be applied to ensure that all functional specifications are fulfilled for new concepts/applications. This shall be based on a combination of an analytical/numerical approach and type tests 7) experience which is intended to be used as proof of fulfilment of the specifications and safety level criteria against possible failure modes shall be documented 8) basic tests or references to recognized literature shall identify limiting materials and functional parameters 9) alternative methods to those described in this document may be used provided they are supported by equivalent evidence of the suitability of their application 10) to qualify a range of sizes, the following principles shall be followed: potential critical parameters related to scaling shall be identified (e.g. non-linear scaling factors) and addressed to document the acceptable margin to failure the design calculation model shall be validated for the specified range the required type(s) and number of tests required depend on the identified critical parameters and associated risks. Guidance note: Examples where scaling may be non-linear: seal dimension and extrusion gap radial gripping and seal pressure level acting on the pipe wall thermal expansion load from polymer seals. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Quality control by testing The main objective of the quality control by testing is to verify acceptable strength and functionality during all relevant phases of the fitting's design life. The extent of the tests required to ensure compliance with this RP depends on the design type, design standard, confidence in analyses, design assumptions and extent of documented experience. Test requirements are covered in Sec.12, with further guidance given in App.C. A qualification programme shall be established based on the above aspects. This programme shall specify: the tests to be carried out the purpose of the test the parameters to be measured and recorded the acceptance criteria for the relevant failure modes' margins to failure the accuracy of these measurements the type of analysis of the test results to enable correlation with the design analysis and limiting design conditions. Testing shall verify the design envelopes. Testing of the whole assembly is required as part of the qualification programme. Small-scale testing is a recommended approach in order to reduce the extent of the full-scale testing. The following tests are recommended: 1) Basic tests, such as tests of the material strength, seal capacity, seal extrusion gap limits, functionality and performance of the locking between the fitting and pipe, etc. In addition, basic tests can be used to validate the capacity models/calculations used to determine the required functionality and performance of the locking between the fitting and pipe (e.g. small-scale gripping tests). 2) Type tests (qualification tests), which verify the functional requirements of a new type of design with a recognized safety margin. Testing is normally executed on the full repair fitting assembly (e.g. full-scale testing). The functional requirements include documentation of the sealing and gripping capacity for the load combinations and operational envelopes specified in the qualification or design basis. This type test can be combined with the FAT (see below) for the fitting tested. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 28

29 3) Factory acceptance tests (FAT) (i.e. strength pressure test), which verify the manufacturing and assembly of a fitting that is already type tested. The FAT for fittings that are not designed for reuse could be limited to dimensional measurements and checks that the material complies with the design criteria. 4) Final tests (i.e. gross error leak test), which verify the completed installation. For fittings designed to have an axial load capacity provided by a pipeline locking mechanism, the correlation between locking capacity and pipeline indentation marks shall be validated by qualification testing. If this correlation is not qualified, the FAT shall include external loads in addition to pressure. Guidance note 1: The test programme should include the possible deferred use of the fitting, such as installation and activation to pre-installed fittings, such as guide posts and hub connections. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Further details of the typical tests, test pressure levels, holding times and acceptance criteria are given in Sec.12 and App.C. Qualification tests for hyperbaric welds are described in Sec.11. Guidance note 2: 1. The hydrostatic pressure test (i.e. strength test) as referenced in the nominated design standard (e.g. pressure vessel standard) for the fitting is required prior to installation in the pipeline system. This test is often combined with the FAT performed onshore, as the subsea pipeline system is not normally qualified for this pressure level. The hydrostatic pressure test documents a strength resistance level equivalent to the pipe section mill test required for pipeline joints. For testing of a fitting design based on the load and resistance factor design (LRFD) method, see Sec The leak test of a fitting after installation, i.e. seal test, is equivalent to the gross error leak test as specified in DNVGL-ST-F101 for pipelines. The test pressure level depends on the type of seal and potential failure mechanisms, see Sec Pressure and load-retaining components of the fitting used in the hydrostatic pressure test should not be replaced before installation for pipeline repair. Replacing sealing packers will require a new strength test/fat (unless the design is according to [C.4.1] in App.C). ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 29

30 SECTION 3 PIPELINE REPAIR ACTIVITIES 3.1 Damage assessment The severity and root cause of pipeline anomalies or damage should be determined prior to starting the intervention work on the pipeline in order to ensure and document that the required safety level is maintained. Figure 3-1 shows a typical work process from when the pipeline damage is discovered until the corrective actions are identified and assessed. 1) 1) 1) The option of conditional operation and/or temporary repair versus permanent repair depends on the criticality concluded from the integrity assessment and how the safety level develops over time, versus the mobilization time to perform the permanent repair. Figure 3-1 Typical flow chart of activities, from damage indication to completion of corrective actions Guidance note 1: The initial damage will normally be detected during a pipeline internal/external inspection or by a monitoring system. After the damage has been identified, a more detailed inspection should be performed, possibly including non-destructive testing (NDT) to quantify deformations, metal loss, potential cracks, gouges, etc. in the pipeline steel wall as well as measured pipeline configuration. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 30

31 The results of the measured pipeline configuration, close visual inspection and NDT are evaluated further together with the operational data, design data and material data to assess the pipeline's current condition and determine whether or not corrective actions are needed. Corrective actions may include the adjustment of operating parameters, temporary repair or permanent repair. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 2: If close visual inspection is by a person, a risk assessment should be performed to determine the acceptable safety level of the pipeline/environment and whether a reduction in pipeline pressure is required. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Pipeline defects and acceptance criteria Standards providing acceptance criteria for the assessment of relevant pipeline defects are given in Table 3-1. Table 3-1 Typical pipeline defects and acceptance criteria Type of defect Failure mechanism Acceptance criteria Internal metal loss Burst Bending moment capacity Collapse DNVGL-RP-F101 DNVGL-ST-F101 DNVGL-ST-F101 External metal loss Burst Bending moment capacity Collapse DNVGL-RP-F101 DNVGL-ST-F101 DNVGL-ST-F101 Dent 1) Plastic strain Fatigue Local buckling Collapse Pigability DNVGL-RP-F111, PDAM (/4/, /5/) DNVGL-RP-C203, PDAM (/4/, /5/) DNVGL-ST-F101 DNVGL-ST-F101 DNVGL-ST-F101 Gouge Fatigue PDAM DNVGL-RP-C203 Crack Fatigue ECA criterion DNVGL-RP-C203 DNVGL-ST-F101 Local buckle Strain Fatigue Pigability DNVGL-ST-F101, PDAM (/4/, /5/) DNVGL-RP-C203 DNVGL-ST-F101 Global buckle Strain DNVGL-RP-F110, PDAM (/4/, /5/) 1) References to DNV GL standards stipulating acceptance criteria related to dents are given in Appendix [A.2]. Guidance note: The different defect types may occur simultaneously, depending on the root cause of the damage. E.g. an anchor hook scenario may cause global buckling of the pipeline in addition to local pipe wall damage such as dents, gouges and/or dents with gouges. Both the ULS and SLS acceptance criteria should be evaluated and complied with. Managing the risk related to pipeline system threats is essential for maintaining the integrity of the pipeline system. Further details on integrity management guidelines for submarine pipeline systems are given in DNVGL-RP-F e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Failure assessment, acceptance criteria and the repair of pipeline components (e.g. valves, flanges, pig traps, connectors, tees and wyes) are more differentiated and not covered by this RP. Typical component failure mechanisms are seal leaks and a failure to operate, e.g. a valve fails to operate to a closed or open position, or the closing time exceeds the acceptance criteria. However, temporary isolation methods (like in-line isolation tools and plugs deployed through a tee/branch-line) applied to enable intervention and the repair of such pipeline components should follow the relevant guidelines given in this RP. 3.3 Temporary and permanent repairs A permanent pipeline repair shall be qualified for the remaining design life of the pipeline system. Temporary repair solutions shall be qualified for the specified limited in-service time until a permanent solution has been implemented. Neither a permanent nor a temporary repair shall jeopardize the overall safety level of the pipeline system. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 31

32 Guidance note: 1) The initial design life of the pipeline system may be extended after re-qualification. 2) A temporary repair intended for topside use may be exposed to external fire loading and the use of soft seals may therefore not be applicable. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Local repair Pipeline local repair is provided by the installation of a pipeline repair clamp (or equivalent) either to provide structural reinforcement of the pipeline steel wall, or to seal off a potential or existing leak. The basis for a local pipeline repair is that the safety level of the pipeline can be reinstated to be in compliance with the pipeline design standard by local reinforcement and/or sealing. Further, the local repair should mitigate further worsening of the damage (e.g. crack growth and metal loss from external corrosion). To mitigate the further development of internal damage (e.g. local internal metal loss), additional mitigation activities such as cleaning pig runs and the use of inhibitors need to be assessed. Thorough evaluations of the pipe wall's remaining capacity are required to document the acceptable gripping and structural capacity of the clamp in such areas. References to standards providing acceptance criteria for typical pipeline damage are found in Table 3-1. Two main types of clamps are used, either separately or in combination: 1) Leak clamp The repair of local metal loss or crack defects that may develop, or have developed, into a leak, requires a clamp solution that is able to contain the pressure. Depending on the root cause of the leak, the clamp may also be required to provide structural reinforcement to the pipeline steel wall in order to mitigate further degradation or crack growth (e.g. by filling grout/polymer materials into the clamp annulus). 2) Structural clamp Pipeline reinforcement/structural clamps are used to transfer parts or all of the pipeline section forces through the clamp. Structural clamps are used where the damage has reduced, or may develop to reduce, the pipeline section force capacity. Such clamps may also be designed to inject a filler material, e.g. grout/ polymer materials, into the clamp annulus to restrain the fatigue loading of dents and/or to transfer the pipeline section forces from the pipe to the clamp. The design of a repair clamp can be either a temporary or permanent solution. The design of local repair clamps shall, as a minimum, cover the: design conditions design calculations requirements for pipeline preparations filling material qualification and injection procedure, if applicable installation activation testing. Functionality requirements for repair clamps shall as a minimum include the following: for structural clamps, the gripping capacity shall be according to [6.5] the sealing performance shall be according to [6.7] the installation tolerances and monitoring of successful installation shall be according to Sec.7 testing shall comply with relevant tests in Sec.12 for welded split-sleeve repair clamps, the fillet weld between the clamp and pipeline shall comply with DNVGL-ST-101. Guidance note: Pipeline repair clamps installed to seal pin-hole leaks will be exposed to pressure loads (and thermal loads) only, as axial and bending moment loads are taken by the remaining pipeline cross-section. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 32

33 Pipeline repair clamps designed to take axial and bending moment loads transferred from the pipeline should have a gripping capacity designed and qualified as for a pipeline coupling. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Pipeline section replacement Pipeline cut out and replacement involves the removal of the damaged pipe section and installation of a new pre-tested and verified pipeline section/spool piece. Depending on the nature of the damage and its location, the length of the replaced pipeline section can vary from a short piece up to lengths requiring a pipeline installation vessel for the work. The new pipe section/spool piece is joined to the existing pipeline ends by either welding or mechanical couplings. The new pipeline section shall be manufactured, mill-tested and installed according to DNVGL-ST-F101. The repaired location shall be pressure tested after installation unless conditions allow the pressure test to be waived, see Sec.12. Atmospheric welding, including above water tie-in (see Sec.10), and hyperbaric welding by divers shall be according to DNVGL-ST-F101 with respect to welding and NDT. Remote hyperbaric welding is covered by Sec.11. The design of mechanical couplings shall, as a minimum, cover: pipeline preparation requirements design conditions design calculations installation activation testing. Functionality requirements for mechanical couplings shall as a minimum include the following: load conditions as specified in [6.4.2] gripping capacity shall be according to [6.5] sealing performance shall be according to [6.7] installation tolerances and monitoring of successful installation shall be according to Sec.7 testing shall comply with relevant tests in Sec.12. External loads transferred from the pipeline shall be considered when establishing the fitting grip capacity. The fitting grip capacity shall account for the load and resistance safety factors in [6.5.2]. In addition, the relevant limit states to be considered in the fitting design shall generally be based on the nominated pressure vessel standard, typically plastic collapse, local failure and cyclic loading. A calculation example is included in App.F. 3.6 Hot tapping General Pipeline hot tapping is the method of making a connection to an existing pipeline without stopping the pipeline's operation. The hot tap operation will normally include a hot tap tee fitting with an isolation valve at the branch, to be installed on the pipeline (hot tap clamp). Guidelines on the design and installation of a pipeline hot tap system are provided in Sec.9 and cover the following main phases: preparation of the pipeline and seabed at the hot tap location installation and testing of hot tap equipment hot tapping and test of barrier installation and testing of new connection (tie-in) operation. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 33

34 Functionality requirements for the hot tap clamp shall as a minimum include: for structural hot tap clamps; the gripping capacity shall be according to [6.5] the structural clamp's sealing performance shall be according to [6.7] seal welding shall be according to Sec.11 and App.D installation tolerances and monitoring of successful installation shall be according to Sec.7 testing shall comply with Sec.12. Guidance note: API RP 2201 provides information to assist in safely conducting hot tapping operations on onshore pipeline systems; however, the basic process is also applicable to subsea systems. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Removal of pipeline defects Defects in the pipeline steel wall, such as dents and gouges, can be removed using the hot tap process in combination with a branched clamp installed with either a blind flange or valve. The design of the branched tee hot tap clamp shall be according to DNVGL-ST-F101 Section 5F or the equivalent Bypass/branched connections Pipeline bypass is a solution that allows continuous production via a new pipeline segment while the original pipeline is being repaired. The bypass line is usually connected to the original pipeline via branched hot tap clamps. These clamps are fitted with a hot tap valve to ensure a safe environment for hot tapping, allow isolation plugs to be deployed into the pipeline via the branch and seal off the pipeline segment requiring a repair/replacement. The design of the branched tee, hot tap valve and tie-in spool with connectors shall be according to relevant standards referenced in DNVGL-ST-F101, or the equivalent. 3.7 Pipeline isolation Pipeline isolation may be provided by pipeline in-line isolation plugs running inside the pipeline. Internal grip and seal segments expand radially at the desired stop location. When the isolation plug is activated, it isolates the pipeline pressure from the pipeline section to be repaired. Where deployment of an isolation plug that runs inside the pipe is not practicable, pipeline isolation may be provided by an isolation plug that is deployed through the side branch of a hot tap fitting. The required isolation period shall be designed according to and documented by the results of reliability analyses documenting a probability-of-failure-level compliant with DNVGL-ST-F101. Isolation plugs can form a double or single pressure barrier in the pipeline as detailed in Sec.8. The total plugging system shall satisfy the safety class requirements described in Sec.2. It is common to use double barriers where personnel can be affected (i.e. divers for subsea work). For projects where the repaired section shall be subject to a hydrostatic pressure test after repair, one additional plug to provide a barrier to isolate the test pressure may be used. The barrier for hydrostatic testing should form part of the plug isolation train, such that hydrostatic pressure testing of the repaired section can follow on as part of the isolation sequence without removing the isolation plugs, thus maintaining pipeline integrity until testing is successfully completed. The design of a pipeline isolation plug shall cover all scenarios in a typical pipeline isolation process: installation pigging setting testing and qualification un-setting retrieval (including contingency retrieval). Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 34

35 Minimum functionality requirements for pipeline isolation plugs shall include the following: gripping capacity shall be according to [6.5] sealing performance shall be according to [6.7] pigability, both into position and retrieval (considering e.g. wear, bend restrictions, pipeline components) positioning accuracy setting and unsetting pressure monitoring alarms tripped by predefined levels of critical parameters position detection system isolation time contingency systems. 3.8 Supporting activities General An overview of supporting activities for pipeline repair operations is given below Pipeline surface preparation Work on the pipeline shall not compromise the safety level required by the design standard. The qualification of surface preparation tools and planning of each operation shall include the identification and assessment of all potential threats. The safety level acceptance criteria for surface preparation on a pressurized pipe shall comply with the requirements given in DNVGL-ST-F101, while surface preparation work performed on an unpressurized pipeline section may allow for a lower safety level (e.g. safety class Low ) for the temporary unpressurized phase. Typical tools used for pipeline surface preparations are: longitudinal seam weld cap removal tools pipeline coating removal tools pipeline cutting tools pipeline end-preparation tools. Some repair fittings require the pipeline longitudinal weld-seam to be ground flush locally. The removal of the weld cap may expose embedded weld defects. Surface-breaking flaws are in general more critical with respect to fatigue damage than embedded flaws. To ensure an acceptable safety level for possible surfacebreaking flaws exposed after milling the weld cap, an engineering critical assessment (ECA) according to DNVGL-ST-F101 shall be performed to document the acceptable integrity of the pipeline. The size of the flaw to be analysed shall reflect the maximum size not detectable by an NDT performed prior to the milling operation. Guidance note: Typically, a 3mm-high longitudinal weld defect may exist in the seam welds. Because weld flaws close to the surface may be more critical than surface-breaking flaws, a 3mm-high flaw 1.5mm below the surface may be considered as a 4.5mm-high surface-breaking flaw to be considered in the ECA. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- When installing a pipe fitting, the required pipeline preparation for the repair shall be specified, such as: removal of pipeline corrosion and weight coating removal of pipeline longitudinal weld beads requirements for pipeline surface finish, out of roundness, wall thickness and diameter tolerances pipe-cutting tolerances, e.g. pipeline location, angular tolerances, surface roughness, out-of-plane tolerances, bevelling. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 35

36 3.8.3 Pipeline manipulation During a pipeline repair operation, manipulation of the pipeline configuration or re-positioning of the pipeline ends is required in order to do intervention work on the pipeline. Examples of typical equipment used in such operations are: manipulation frames (often called H-frames) buoyancy tanks/lift bags vessel cranes pipeline recovery tools. Changing the pipeline configuration during a repair may introduce additional bending moment loads to the affected pipeline section(s). The repositioning of the pipeline or pipeline ends could be considered as a displacement controlled operation. Analyses of the pipeline re-positioning should take into consideration parameters such as the location of the lifting points, uplift resistance (if partly submerged), etc. in order to establish the maximum loads. Possible dynamic effects should be accounted for in the analyses if relevant. The maximum loads identified shall be taken as a functional load. Control of residual stresses in the new pipeline configuration is required to document acceptable utilization with functional loads after tie-in. The functional load effect factors as specified for a local design check, i.e. only safety factors from combination b in DNV-OS-F101 Sec.4 G303 Table 4-4 Load effect factor combinations, need to be considered. Further, consideration is to be given to the appurtenances near to the repair location (e.g. buckle arrestors) that can be sensitive to fatigue as a result of pipeline manipulation during repairs. Guidance note: DNVGL-ST-E273 Portable offshore units may be used as the design/certification standard for pipeline manipulation frames. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Lifting operations Lifting operations shall be in accordance with recognized international standards. Examples of applicable standards are DNVGL-ST-E273 and DNVGL-ST-N001. A lifting procedure/plan is recommended in order to ensure safe and proper handling of the unit. All materials used in primary structures shall as a minimum be supplied with an inspection certificate of type 3.1 as defined in EN10204 or the equivalent Remotely operated vehicle interfaces Remotely operated vehicle (ROV) interfaces should be designed according to recognized international standards. An example of an applicable standard is ISO Remotely Operated Vehicle (ROV) interfaces on subsea production systems. Accidental loads from interaction between an ROV and fitting equipment (e.g. hot tap branch) shall be considered in the design Instrument piping and components within a fitting assembly Instrument piping (small bore piping) and associated components (e.g. valves, gauges) shall be designed according to recognized international standards. 3.9 Preparedness strategy A pipeline repair preparedness strategy is part of the integrity management process and helps to reduce the downtime for pipeline repair events. A high-level overview of typical activities to be included in a preparedness strategy is: identification of threats to all sections of the pipeline system identification of the likely failure mechanisms caused by the identified threats Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 36

37 evaluation of pipeline restraints on relevant repair operations, e.g.: water depth (e.g. use of divers or remotely controlled equipment, and pipeline collapse restraints) soil conditions pipeline configurations (e.g. curvatures, steep slopes) proximity to other pipelines and subsea assets residual forces in pipeline evaluation of the most suitable repair scenario: temporary or permanent repair local repair pipeline section replacement required de/re-commissioning procedures, equipment and chemicals identification of spare parts, ancillary equipment and long lead items needed for all phases of the repair: pipeline joints, including material certificates tools for pipeline preparation, including a risk assessment of the intervention work (e.g. weld seam removal) the necessary pipeline isolation tools - ensure they are qualified for the specified repair work selection of tools to manipulate the pipeline configuration and assess the condition of the pipeline during a pipeline lift and shifting periods (e.g. temporary long free spans) identification and risk assessment of chemicals required for the chosen repair operation equipment for de-commissioning and re-commissioning the pipeline empty seawater from the pipeline/repair location after the repair is completed - when relevant pressure test of the repair location availability of vessels/barges organization structure and communication lines vessel capabilities local authority s notification and involvement/permits (restricted zones, vessel traffic zones, environmental areas etc.) maintenance of equipment and procedures, including training of personnel having call-off contracts with suppliers of decommissioning/re-commissioning/vessel/repair equipment/ seabed preparation equipment in place. Figure 3-2 and Table 3-2 show an example of a pipeline sectioning and pipeline risk assessment as input to the repair preparedness strategy. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 37

38 Figure 3-2 Example of pipeline sectioning (sections A to E in figure) for different water depths (see Table 3-2) as input to the preparedness strategy For each section of the pipeline system, the relevant threats and associated damage scenario are detailed and the preferred repair solution is proposed. An example of a high-level risk assessment as input to a preparedness strategy is shown in Table 3-2. Table 3-2 Example of a high-level risk assessment as input to the preparedness strategy Section: A B C C D E Water depth Shallow/shore approach Intermediate Deep (no diving) Deep (no diving) Intermediate Main threat Dragged anchors Trawling Corrosion Landslide Dropped objects Type of repair based on damage Probability of damage Permanent repair method/ equipment Temporary repair method/ equipment Local repair/short section replacement Local repair/short section replacement Local repair/short section replacement Long section replacement Local repair/short section replacement Shallow/shore approach Sinking vessels Long section replacement Medium High Medium Medium Medium Low Repair clamp Mechanical couplings Atmospheric or hyperbaric welding (cofferdam) Repair clamp Mechanical couplings Hyperbaric welding (Divers possible) Repair clamp Mechanical couplings Remote welding Re-lay pipeline Remote welding Repair clamp Mechanical couplings Hyperbaric welding Mechanical couplings Atmospheric or hyperbaric welding Re-lay pipeline Repair clamp Repair clamp Repair clamp NA Repair clamp Repair clamp By-pass Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 38

39 SECTION 4 PIPELINE DESIGN BASIS 4.1 General The pipeline design basis shall be specified, and shall as a minimum include: design pressure, fluid temperature and a description of the transported fluid, water depth and sea temperature, external pipe diameter, wall thickness, corrosion allowance and materials specification, a reference standard for manufacturing and dimensional tolerances. Guidance note: The pipeline standards specify most of these tolerances related to pipe fabrication and pipeline installation. Dimensional tolerances of concern to the design are dealt with in [4.6]. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Pipeline design pressure Pipeline design pressure at repair location The pipeline design pressure at a given repair location is the differential pressure between the internal and the external design pressure, defined below. Internal design pressure: The definition of pipeline design pressure is detailed in DNV-OS-F101. In general, the pipeline local incidental pressure shall be used as the design pressure for repair fittings. The local pressure is the internal pressure at a specific point based on the reference pressure adjusted for the fluid column weight due to the difference in elevation. It can be expressed as (ref. DNV-OS-F101): = + (h h ) = + (h h ) = (4.1) (4.2) (4.3) Where: h h is the gravity is the elevation of the reference point (positive upwards) is the elevation of the local pressure point (positive upwards) is the density of the relevant content of the pipeline is the density of the relevant test medium of the pipeline is the local incidental pressure is the incidental reference pressure at the reference elevation is the local system test pressure is the system test reference pressure at the reference elevation is the design pressure at the pressure reference elevation is the incidental to design pressure ratio The incidental pressure depends on the flow characteristics, pressure control and safety system of the considered pipeline system, and is defined as the maximum internal pressure that the pipeline or pipeline section is designed to withstand during any incidental operating situation (100-year value). Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 39

40 Ref. DNV-OS-F101, typical and minimum incidental to design pressure ratios are given in Table 4-1. Table 4-1 Incidental to design pressure ratios Condition or pipeline system Typical pipeline system 1.10 Minimum, except for below 1.05 When the design pressure is equal to the full shut-in pressure including dynamic effects 1.00 Other recognized pipeline standards define the pipeline design pressure in addition to the maximum allowable operating pressure (MAOP). For the design of fittings, the design pressure shall be based on the pressure control and safety system, i.e. the incidental pressure shall be used. External pressure: The external pipeline design pressure is: = h (4.4) Where: h is the external design pressure is the density of the sea water is the gravity is the elevation of the local pressure point (positive upwards) In cases where external pressure increases the structural capacity, the external pressure shall not be taken as higher than the water pressure at the considered location corresponding to a low astronomic tide, including any negative storm surges. In cases where the external pressure decreases the capacity, the external pressure shall not be taken as less than the water pressure at the considered location corresponding to a high astronomic tide, including storm surges. 4.3 Pipeline dimensions DNV-OS-F101 details the pipeline dimensions to be used when calculating pipeline loads and utilization. The maximum loads arising from pipeline exposure shall be based on nominal values. For a restrained pipeline, the compressive forces are mainly due to pressure and temperature. For thermal expansion, this means that the larger the steel cross-sectional area is, the larger the restrained force. Pipeline capacity calculations shall be based on a reduced wall thickness arrived at using the established corrosion allowance and tolerances used in the design basis and manufacture. In specific repair projects with a known damaged location, and where the local diameter, out-of-roundness and wall thickness of the actual pipe joint to be repaired are documented by inspection reports (e.g. in-line inspection reports), these dimensional properties can be used in the design of the repair assembly. Guidance note: The maximum loads arising from pipeline exposure should be based on nominal values or average values if dimensional tolerances are such that higher values can be expected. If actual measured pipe wall dimensions at the location is available, these may be used in the pipe wall design check for the repair fitting. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Pipe material Except for the yield stress and ultimate strength, the materials parameters shall be based on the nominal values. The stress-strain curve shall be based on the specified minimum values of yield stress and ultimate strength, f y and f u, as per DNV-OS-F101. The effects of temperature shall be considered for both the pipe material and fitting material, i.e. temperature de-rating. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 40

41 Several types of fittings are designed with gripping segments that grip onto the pipeline in order to provide the required locking capacity. The locking capacity is often based on the ratio between the surface hardness of the gripping segments and the pipeline surface hardness, and this shall be documented by the fitting design. Yield stress (YS) and tensile strength (TS) may also be established based on statistical assessment of mechanical test data from the considered section of the pipeline system, according to requirements given in DNV-OS-F Lined and clad pipelines For internally lined and clad pipelines, the strength contribution from the internal layer may be included in the integrity assessment provided the manufacturing process documents acceptable confidence in the strength and stiffness properties of this layer. Figure 4-1 shows the typical layout of a clad and lined pipeline. CRA girth weld filler A) CRA clad pipeline Pipeline parent metal Clad layer CRA girth weld filler CRA root and hot pass weld B) CRA lined pipeline, with overlay weld to seal the liner at girth welds. Pipeline parent metal CRA liner CRA girth weld filler CRA weld overlay CRA root and hot pass weld C) CRA lined pipeline, with CRA seal weld to seal the liner at girth welds. Pipeline parent metal CRA liner CRA root and hot pass weld CRA seal weld, each side of girth weld Figure 4-1 Typical layout of a CRA clad and lined pipeline at the field joint For the repair of lined and clad pipes, the repair solution shall provide a barrier against the CRA (corrosion resistant alloy) clad or lined layer, preventing pipeline fluid from coming into contact with the parent metal of the pipe section cut-end and the pipe external surface. Deviation from this requirement can be acceptable if the potential metal loss of the exposed parent metal is documented to be acceptable within the design life, e.g. self-limiting corrosion process due to a small volume of aggressive fluid. Pressurization or fluid ingress through the CRA overlay or seal weld into the annulus between the CRA liner and pipeline inner surface is not acceptable. In general, this implies that clad pipes can be cut at any location, whereas lined pipes need to be cut at, or close to, the girth weld to maintain the integrity of the CRA weld overlay between the liner and the pipeline Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 41

42 parent material. Butt welding repairs have no restriction on the cut location of the lined pipe provided this is covered by the qualification of the repair system. The following need to be considered in the design and technology qualification of such repair systems: Material compatibility: pipeline fluid fitting materials locally exposed surfaces of pipeline parent material at cut location. Structural integrity: shear capacity of bonding between parent pipeline material: clad layer, and overlay or seal weld for lined pipe reduced capacity, due to a potential flaws from lack of fusion in the liner weld, for relevant failure mechanisms, such as fatigue, excessive strain levels and local buckling. structural local utilization of liner or clad material when exposed to gripping and sealing interference loads from in-line isolation plugs. An acceptable effect of gripping segment indentations shall be documented. gripping and seal contact pressure load-effect of inner layer, accounting for possible: end seal contact pressure thermal expansion of liner and parent pipe gap between liner and parent pipe. Pipe preparation: surface finish weld bead removal pipe cut skew accuracy preparation of weld cap (field joint and longitudinal weld seam). Installation of fitting: distortion of the CRA liner and end seal weld during installation water entrapment hydraulic lock potential pressurization by steam at operating temperature. Operation after repair: corrosion local buckling of liner, combining operational loads and repair fitting activation loads fatigue failure of seal weld at the end of the liner towards pipeline parent material for weld repairs of lined and clad pipelines, an evaluation of the weld quality required to fulfil the fatigue and fracture limit states. 4.6 Dimensional tolerances Welds and surface imperfections The weld itself can cause local discontinuity on the pipe surface. Surface roughness and discontinuity tolerances are of concern with respect to the seals. The coupling shall be qualified for the pipe either: with the quantified surface imperfection, or after removal of the surface imperfection. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 42

43 4.6.2 Linepipe External diameter tolerance, e t, is mainly derived from the measurement of the circumference and therefore represents an average. Out of roundness (OOR, ovality) tolerance, e o is measured by a gauge, or if available by intelligent pigging including caliper measurement, defined as: = (4.5) Repair equipment manufactured for contingency - where the repair location of a given pipeline system is not known, the pipe section dimensions may be based on project-specific dimensional data from the mill, combined with installation records as given below: Moderate bend strained pipe installation over stinger: The out of roundness can be determined based on measurements from the mill plus the residual ovality, e R, due to bending over the stinger during installation of the pipeline. The residual ovality e R can be estimated from the bending strain and the point load from the stinger rollers as: = 2, , (4.6) where e R = residual ovality e o = out of roundness (OOR, ovality) tolerance from the mill = pipeline bending strain D = nominal outside pipe diameter t = nominal pipe wall thickness R s = point load from the stinger rollers 0.6 = Nominal outside diameter D to be given in metres (4.7), = (4.8) = (4.9), =, (4.10), 101 = characteristic bending strain resistance, according to DNV-OS-F101 This approach is a modification of the approach presented in Ref. /7/, and it should be noted that it is based purely on finite element analyses and is not supported by physical tests. The point load effect may be waived in the case of V-rollers with an angle of at least 30 degrees if the loading of these is symmetric. The reason is that the ovalities caused by each of the two point loads will cancel each other out and the point loads will not act at the bottom, where the maximum bending stress occurs. Guidance note 1: The outer diameter of the pipeline from the mill is typically within +/- 1% of the nominal specified diameter. DNV-OS-F101 Table 7-17 specifies the mill acceptance criteria for outer diameter and out of roundness. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 43

44 Guidance note 2: The design envelope for dimensional tolerances of the pipeline for which the fitting is qualified is specified in the design premise for the fitting for generic projects, where the end user is responsible for the pipeline dimensions at the repair location being within these specified envelopes. Alternatively, the fitting is designed for one pipeline repair project with project-specific dimensional tolerances. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Local out of roundness tolerance, e l, reflects dents and peaking. The straightness of the pipe section is normally measured by a taut string between the ends and shows the greatest distance to the pipe surface. Straightness within the length of the fitting is normally not specified and therefore special considerations shall be made. The straightness of the pipe section of concern, e s, is within the length of the fitting. The following formula applies to a possible S shaped pipe: =( 2 ) 2 /100 (4.12) Where: l = length of fitting for l/l < 0.5 L = length of linepipe section (normally 12 m) or a specified section s = straightness of pipe/section specified as a % of L Ovality of reeled pipes (/14/ and /15/): The following calculation procedure may be used for the residual ovality calculation after several plastic strain cycles during reeling. The residual ovality due to the bend cycle is:, = 1 +,, 1 +,, =, 1 +,,, 1 (4.13) (4.14),, = residual ovality factor at bend cycle. = maximum ovality at bend cycle., = residual ovality at bend cycle., 1 = residual ovality at bend cycle 1, =,0 at 1st bend cycle (= manufacturing ovality)., = DNV-OS-F101 prediction ovality at cycle :, = 0.03 (1 + /120 ) 2, / 2, = bending strain in cycle. The factor, shall be established based on full-scale bend tests and/or finite element analyses and will depend on the pipe diameter, wall thickness, bending strain, back tension, coating, etc. A value in the range of 0.25 to 0.35 for the first cycle and 0.05 to 0.15 for subsequent cycles is suggested for thin coated pipes with a diameter in the range of 300 to 400 mm (12-16 ) and D/t in the range of 15 to As installed The installation procedures can, in particular cases, cause additional flattening (out of roundness) due to the bending of the pipe, see [4.6.2] Extreme maximum and minimum diameter The maximum and minimum internal no touch fitting diameter to cover the tolerance combination e m, which is due to each of the above extreme tolerances excluding the possible flattening effects of the installation, is: = ± ± 0.5 ± + (4.15) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 44

45 Provided the installation effects in [4.6.3] may be neglected, this represents a conservative extreme limit. A less extreme and more realistic limit can be based on procedures described in the next subsection. Guidance note: The effects of the straightness (e s ) should also be dealt with separately for assessing the alignment during installation, see Sec e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Statistical maximum and minimum diameters Statistical evidence shall be used to establish the likely maximum tolerances; if not, the unlikely extreme tolerance combinations presented in [4.6.4] shall be applied. Guidance note: Extreme tolerance combinations are unlikely to occur for most pipeline types. The fitting design is sensitive to the pipeline dimensional tolerance. The specification of an over-conservative tolerance combination could be difficult to meet with one size of fitting. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Electrical potential Subsea fittings are normally protected against corrosion by cathodic protection (CP) systems which cause atomic hydrogen to form at the metal surface and thereby introduce a risk of hydrogen-induced stress corrosion cracking. The potential range of CP using aluminium- or zinc-based anodes is 0.8V to 1.1V, see DNV-RP-B401. This exposure may limit the strength and hardness of carbon steel and stress utilization of high-alloy steels. 4.8 Environmentally assisted cracking Components of subsea pipeline repair systems may be exposed to environments internally or externally which promote environmentally assisted cracking (EAC) related to nascent hydrogen. This includes sulphide stress cracking (SSC) and hydrogen induced stress cracking (HISC), with the production and absorption of hydrogen being related to a corrosion process and cathodic protection (CP), respectively. The presence of hydrogen sulphide (H 2 S) is a prerequisite for SSC and may also enhance HISC. Relevant environmental factors promoting SSC and HISC include ph, H 2 S content, temperature, pressure and, for components exposed to the external environment, also CP. Such parameters shall be defined in the project design basis. The combination of an H 2 S-rich environment and CP may have more detrimental effects on the cracking resistance than CP acting alone, and the 350 HV criterion may be non-conservative for such an environment. Therefore, for exposure to the combination of CP + H 2 S, a qualification test shall be performed to document acceptable robustness against related failure mechanisms. An environmental seal may be applicable to isolate certain components or parts of components from an environment driving SSC or HISC. The environmental seal needs to be qualified for the design life of the repair installation, see also [6.7.3]. For components that are to be directly exposed to an internal environment containing H 2 S, the material selection and fabrication methods affecting susceptibility to SSC shall comply with ISO This may require qualification testing which is to be carried out according to the standard. Testing procedures and results should be reviewed and accepted by the owner of the pipeline. There is no standardized method for HISC testing but some recommendations for SSC testing in ISO are also relevant for the design and execution of HISC testing. HISC testing shall be performed at the most negative potential that may apply for the CP system and at the maximum applicable strain. For certain locations, exposure to H 2 S containing seawater may apply and shall then be included in the test, which shall further include control of the minimum applicable ph. As for SSC qualification, any test procedures for HISC qualification and the test results should be reviewed and accepted by the owner of the pipeline. For components exposed to other chemicals, like chlorides, acids, corrosion inhibitors and biocides, the risk of stress corrosion cracking (SCC) should be assessed. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 45

46 Guidance note: The owner of the pipeline should duly consider any detrimental effects related to the local deformation of linepipe materials including surface deformations, such as those induced by some pipeline repair systems, and the need for qualification testing. It is also important to distinguish between plastic deformation of the material prior to and during service. Plastic deformation of a material while it is exposed to a H 2 S-rich environment will promote hydrogen uptake and may prove detrimental at low deformation levels. Stresses introduced from the repair system on the linepipe material should therefore be considered in order to avoid in-service deformation of the linepipe. The equipment user should determine whether or not the service conditions are such that the ISO series applies or if other test methods, not addressed in ISO 15156, might be required. For repair components in carbon or low-alloy steel, a maximum hardness of 350 HV is generally considered to ensure full resistance to HISC for all practical purposes. For materials exposed to sour service, the maximum allowed hardness is 250 HV. Components in duplex (ferritic-austenitic) stainless steel should be designed for maximum stress/strain according to DNVGL-RP-F112 in order to avoid HISC. Any high-strength alloys on a Fe or Ni basis (typically precipitation hardened) and martensitic stainless steels should be qualified individually if they are to be exposed to CP with high stresses applied. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 46

47 SECTION 5 PIPELINE EXPOSURES 5.1 External pipeline forces An overview of pipeline forces that the different types of repair fittings are exposed to is given in App.A. The local pipeline forces at the repair location should be based on pipeline design loads given in the pipeline design basis, or on pipeline analyses providing design loads for the specific repair location. A pipeline is subject to global (fundamental) pipeline forces resulting from how the pipeline has been installed and how it is operated. During intervention and repair activities, pipeline conditions change and may result in a change in the global pipeline forces. In section replacement repairs, tension and torque forces in the pipeline are removed when the pipeline is cut subsea. The changed pipeline conditions after coupling-installation generate the following forces: 1) soil friction. This force is dependent on the friction coefficient and the force/displacements caused by: a) axial expansion forces due to increased temperature b) axial expansion due to changed pressure c) subsidence of the sea bottom resulting in lateral displacements 2) forces in the pipeline caused by internal and external pressure 3) forces caused by the repair operation and gravity, such as make-up loads, tension/compression, bending moment and torque 4) forces released after the repair operation, such as tension forces in steep slopes 5) possible changes in pipeline support/soil conditions e.g. causing free spans. 6) possible external transverse loads from fishing gear 7) possible hydrodynamic forces caused by current and wave actions 8) accidental loads identified to be of concern e.g. caused by mud slides and dragging anchors 9) forces in the pipeline caused by the loads from an isolation plug for temporary isolation of the pipeline, see Sec.8. Guidance note 1: Torque can be caused by the connecting operation when curved spool pieces are used. Normally, the tension in operation will be of most concern. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Mechanical and welded sleeve-type couplings and pipe branch connections are subject to: bending moments and axial forces transferred from the pipeline forces generated in the coupling by fluid pressure acting on areas protruding from the pipe's internal cross-section. Guidance note 2: The following fundamental assumptions should be considered with respect to force exposure using a mechanical or welded sleevetype coupling or branch connection: 1) repairs are normally planned with well-known seabed conditions and, where necessary, intervention (e.g. rock-dumping) has occurred. Therefore, there should in general be no need to apply factors for an uneven seabed (e.g. from DNV-OS-F101) 2) in accordance with DNV-OS-F101, all loads are to be established for a non-corroded section. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- A load condition of concern to a sleeve-type coupling is tension with no internal pressure. This is a rare case which can occur if: 1) the pipeline, in a hot condition, changes position due to temperature expansion (snaking), and thereafter the fluid transport is stopped. The pipeline then cools off and the pressure is relieved. 2) a free span develops underneath the coupling and the adjacent pipeline, either in an expansion loop or on a long slope. 3) the pipeline is subject to subsidence, mud slides or dragged anchors. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 47

48 4) the connection operation applies large external forces to join the two pipe ends. Item 1 is only of concern to pipelines with elevated temperatures and for some soil conditions. Item 2 is predictable and can be controlled by inspection. Item 3 is only seldom of concern. Item 4 is easily predictable based on the joining tool capacity. 5.2 Maximum axial forces Scenarios The couplings are subject to the forces conveyed from the pipeline (true wall forces), forces generated in the coupling by fluid pressure acting on areas protruding from the pipe's internal cross-section and pretension forces. The following descriptions relate to the true wall axial forces (N). The maximum forces depend on pipeline-soil interactions operating conditions. The following three scenarios represent the limiting conditions: Scenario A: free pipeline, elbow or free end of pipeline, all with internal over-pressure. The axial pipeline forces caused by internal pressure are governed by the pressure and hence the test pressure force dominates. Scenario B: restrained pipeline. The axial pipeline force is governed by the rigidity of the restraint. Scenario C: pipeline on seabed with expansion loops or imperfections. The force is less than half the force determined for a hypothetical completely fixed pipeline, provided the possibility of the pipe being locked (e.g. by sand settling) in an expanded (e.g. by temperature and pressure) configuration is avoided. In general, the above scenarios A, B and C should be included when considering the relevance of the following load cases: 1) pressure test - maximum tension at the manufacturer (hydrostatic pressure test as specified in the fitting's governing design standard) of pipeline (performed after completion of repair, typically 1.05 times the local incidental pressure) 2) pressure test of pipeline - maximum compression 3) operation - (maximum tension) 4) operation - (maximum compression) 5) operation - fatigue 1 (tension) 6) operation - fatigue 2 (compression). Combined load cases with bending moments shall be included for coupling types which are also sensitive to bending moments. Guidance note 1: The stress range at a given repair location may vary for the different operating conditions addressed above through Operating tension and Operating compression conditions, e.g.: VIV response in shut-down (cold condition adding tension in free spans) versus operating (hot adding compression and sagging to free spans) conditions. the fatigue stress range from variations in functional loads at a given repair location within a curved (e.g. global buckled) section of the pipeline may vary depending on the operating condition. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 48

49 Guidance note 2: The implied limits in DNVGL-ST-F101 (0.96SMYS/0.84SMTS) for system pressure testing apply for a large number of joints and are not relevant for the capacity assessment of a single test pipe on which a coupling is mounted for testing. Most fittings will be tested at the manufacturer to a test pressure exceeding the pipeline's local test pressure after installation. Pressurizing to 105% or even 110% of the SMYS for the linepipe material is commonly done. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Free pipe end end cap (scenario A) This load case is typically related to the conditions during factory pressure tests and installations in expansion loops with negligible friction. The normalized force relative to the pipe yield strength is: = (5.1) This maximum axial tensile force will be established as: N a = N pt during pressure test N a = N o at design pressure p i = the internal pressure at the condition considered. Guidance note: The maximum internal seal diameter in the coupling governs the internal pressure term of the axial force. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Restrained pipeline (scenario B) Compression (initially restrained pipe) The maximum obtainable compression forces through the coupling occur if the pipeline at each side of the coupling has been rock dumped before pressurization, or if the pipeline length (i.e. anchor length ) at both ends of the repair location is sufficient for the soil friction to fully restrain the pipe. For restrained pipe, the pipe section at the repair location is not exposed to the end-cap force given by the pipe's inner diameter cross-section. This restricts coupling expansion completely, giving a force relative to the pipe strength of: = (1 2 ) + (5.2) This condition is considered conservative. At elevated temperatures, the adopted design configuration may allow the pipe to buckle. High temperature is of concern for export pipelines close to platforms and flowlines close to wells. The tiein arrangement normally allows for axial pipeline expansion and therefore this force will be smaller than that stated above in most cases. Tension (initially free, then restrained pipe) Shutdown includes pressure release and cooling. Given that the pipeline is initially free to expand longitudinally without any resistance in order to accommodate temperature and pressure effects, then subsequent restraint can be caused by, e.g.: soil penetration beneath an upheaval buckle soil cover on the expansion loop, restricting movement back to the original position the pipeline being rock-dumped whilst in operation. Equation (5.2) also applies to the tension force provided the following definitions are made: Δp i ΔT internal pressure after shut down (pressure at installation) minus the internal pressure before temperature after shut down minus the temperature before. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 49

50 The signs will be changed for the first two terms of equation (5.2): the true axial pipe wall force, for restrained pipe (scenario B), pressurized and cooled down: 3 + (5.3) After pressure relief and the cool down period, i.e. p i is reduced to its minimum and Δp i and ΔT increased to their maximum in equation (5.4), the true axial pipe wall force for restrained pipe (scenario B) is given by equation (5.4). 4 = (1 2 ) + + (5.4) Expansion loop effects (scenario C) The axial forces for scenario C, where compressive axial forces are released by curved configurations such as in-expansion loops or global buckles, are within the limits identified by scenarios A and B. An expansion loop or a pipeline with an initial imperfection may respond to the axial force by deflections of the pipeline curvature, governed by the resistance to this deflection. This is illustrated by Figure 5-1, which shows the effects of lateral soil resistance on axial force. Figure 5-1 The effective force S in the restrained pipe as a function of the pipe expansion in the curvature. This expansion is limited by the lateral soil resistance. Guidance note: Figure 5-1 represents an ideal case which considers: - equal lateral friction coefficients for expanding and contracting motions, and - the pipeline curvature radius, which is not affected by the motion. The maximum pipe tension and maximum positive effective force can only be obtained when the friction coefficient, curvature radius and weight are all relatively high. At the start-up of the pipeline, the friction will first cause compressive forces in the pipeline until the friction resistance capacity is exceeded. The pipeline will then start to move laterally. The lateral resistance corresponds to an axial capacity of S f1, S f2 or S f3. The remaining part of S F causes motion by Pipe/curvature expansion as shown by the figure. This expansion continues until the compressive force is reduced to a level which is equal to the curved pipe's soil friction capacity. Thus, this compressive force remains in the pipe when the expansion stops. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 50

51 When the pipeline is shut down, and thereby cools down and de-pressurizes, it will contract, i.e. be offloaded and subject to tension due to the soil interaction. This remaining tensile force: - will be limited by the soil friction capacity - cannot be larger in magnitude than the compressive force - is created after the initial compressive force is released. These limits are indicated on the upper part of the figure by the two 45 lines. The possible tension effective force is below these limits. Therefore, the maximum tension force is limited to half the possible restrained Effective Force when the end-cap force is neglected. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- The maximum pipe relative tension force in the expansion loop, when conditions enabling scenario B can be neglected, is: 4 = (1 2 ) (5.5) Equation (5.5) is equivalent to equation (5.4) except for the terms expressing the effective pipeline force which has been halved for this depressurized (small p i and large Δp i ) and cold pipe (large ΔT). The contraction of the pipe tends to straighten the pipe's curvature. The maximum pipe relative tension force for scenario C with a pipe under pressure (high p i ) and cooled down is: (5.6) Figure 5-2 shows the actual pipe forces in an expansion loop configuration. Figure 5-2 Simplified calculation of required pullback forces During a pipeline contraction, the maximum tension force relative to the pipe strength is: = + (5.7) The lateral soil resistance ( friction ) coefficient depends on the weight of the pipe, soil type, character of motion and length of motion. Complex methods are available to predict such coefficients. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 51

52 Guidance note: Typical soil resistance coefficients (i.e. residual values after break-out) are given in Table 5-1: Table 5-1 Typical pipe/soil friction coefficients Soil type Axial Lateral Sand 0.5 to to 1.0 Clay 0.2 to to e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Force boundaries The maximum residual tensile forces relative to the pipe's yield strength given by the previous three scenarios are plotted in Figure 5-3 for a typical pipeline. The Medium safety class is considered and no external pressure is included. Figure 5-3 shows the tensile force N' a as a function of changing temperatures for a pipeline free to move during the pressure test, N' b for the extreme case when the pipeline has been free to move and then restrained, and N' c for the pipeline in an expansion loop. Denotations (3) and (4) given in the legend of Figure 5-3 refer to cases with and without internal pressure, respectively. Figure 5-3 Maximum tensile forces in a pipeline for the three scenarios described 5.4 Limiting displacements The internal pipeline pressure should be equal to the ambient local external pressure before the repair operation, including the cutting of the pipe. In general, this implies no residual pipe wall section force and no end displacement of the cut ends. Cutting the pipe close to a curvature may cause residual tensile pipe wall forces from soil friction, leading to separation of the pipe ends when cut. However, for guidance as a limiting case, the maximum possible separation of the cut pipe ends would be caused by cutting the pipe where there is maximum pipe wall tensile stress. After cutting, the effective force in the pipeline is released and reduced to zero. S t is the maximum tension force and μ s is the axial friction coefficient between the pipe and sea bottom (N/m): The axial displacement Δl of the pipe end is: 1 = ( ) 2 2 (5.8) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 52

53 5.5 Design moment The design pipeline section axial load (i.e. true wall axial forces, N) for the repair fitting can be established based on procedures given in section [5.2], the design load specified in the design report for the considered pipeline, or pipeline analyses of specific repair scenario(s). Similarly, the design bending moment, M, for the fitting can be established based on the design moment calculated according to DNVGL-ST-F101 for the considered pipeline dimension and material, the design moment specified in the design report for the considered pipeline, or pipeline analyses of specific repair scenario(s). 5.6 Fatigue Fatigue can be an issue for some types of fittings and load types if they are more sensitive to fatigue loads than the pipeline itself. Typical fatigue loads in the high-frequency range are caused by wave actions transferred from the pipelines to the coupling (via a riser or direct wave actions in shallow waters), or by vortexes in free pipeline spans. These loads normally result in bending loads, for which the sleeve on a mechanical coupling tends to stiffen the pipeline section and make it more resistant to high-frequency loads from such sources. The critical section is often the pipeline itself locally, at the interface with the gripping arrangement, where local deformations and fatigue stresses occur. In the low-frequency load range, the number of pressure cycles for the pipeline is of concern, i.e. the number of full depressurization cycles during the pipeline's lifetime. In general, the fatigue failure mechanism of concern can be similar to that for a pipe, i.e. the development of cracks. But a mechanical fitting failure mechanism can also develop differently. Some types can sustain only a limited number of depressurizations before leaks may be expected, caused by the function of locking and sealing mechanisms. Therefore, only parts of DNVGL-ST-F101 are relevant to fatigue loads, in particular to low-cycle fatigue loads. The term fatigue can therefore be misleading for low-cycle fatigue in this context. The repair operation itself may cause significant fatigue damage to the pipeline at the repair location and at possible fatigue hot spot locations adjacent to the repair location (e.g. at transitions to buckle arrestors) to be assessed. In the case of high-fatigue utilization, the fatigue loading may have to be monitored to enable the accumulated fatigue damage to be calculated. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 53

54 SECTION 6 FITTING DESIGN 6.1 General The design of the fitting shall safeguard against possible failure modes. A failure mode and effect overview shall be established for each fitting type. This section provides general design criteria applicable for pipeline repair fittings that comply with DNVGL- ST-F101. More detailed guidelines for code breaks, loads and applicable load and resistance factors for common fitting types are included in App.A. 6.2 Failure modes and causes The general failure modes for fittings are: 1) failure to install on the pipe 2) installation causes damage to the pipe 3) failure to seal (leak) 4) failure to lock 5) material failure. Conditions for preventing failure modes type 1 and 2 are considered in Sec.7. Design requirements to prevent failure modes 3 and 4 are given in [6.5] and [6.7]. Type 5 failure modes (for metallic materials) are covered by [2.1] and Sec.6. Failure modes related to pipeline isolation tools (other than types 1-5 above) are considered in [8.2.1]. Weld failure modes are further detailed in Sec.11. The corrosion protection of the repair fitting assembly, i.e. repair fitting, mother pipe and, when relevant, branch pipe, goose neck, goose neck clamps and valves, shall comply with DNVGL-RP-B401. As an example, general failure modes type 3 and 4 with possible root causes are identified for fittings below: Table 6-1 Examples of general failure modes and possible causes - sealing and gripping, for failure mode types nos. 3) and 4) defined above 3 Failure to seal (leak) 3.1 Loss of seal compression loads due to the lack of sufficient seal-elasticity to compensate for relaxation caused in some operational conditions by: Local plastic yield of pipe Local plastic yield of seal support structure, back-up rings or metal seal Elasticity of the connection between the two halves of clamps possibly reducing the load on the longitudinal seals Compression set of polymer seal Elasticity of the connection between the two seal contact surfaces Low-temperature stiffening or volume shrinkage of polymer seal 3.2 Load case not considered, e.g.: Compression/expansion load effect of the temperature and additional expansion of polymer seals Effects transferred from the locking mechanism Effect of the swelling of polymer seals Local distribution from unsymmetrical conditions Seal axial loads/displacement/wear: Changing axial loads/displacement Temperature effects. 3.3 Seal micro-performance fails Seal contact force insufficient Seal's ability to fill/seal discontinuities in pipe. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 54

55 Table 6-1 Examples of general failure modes and possible causes - sealing and gripping, for failure mode types nos. 3) and 4) defined above (Continued) Seal/back-up ageing/corrosion Extrusion of polymer seals Explosive decompression of polymer seals Seal welds cracks or pores. 3.4 Seal protection fails Deflections/damage caused by installation Dirt on the sealing surfaces. 3.5 Lack of sufficient seal test pressure. 4 Failure to lock. 4.1 Axial capacity insufficient due to: Lack of friction Mechanical locking fails Pre-tension insufficient Secondary effects of internal pressure Poisson s effects not considered in design Eccentricity Relaxation Corrosion Cracking/rupture of structural weld attachment of sleeve Internal debris/pollution in pipe (scale, wax etc.). 4.2 Micro motions caused by: Uneven axial load transfer distribution between pipe and sleeve. Loads exceeding the limits in parts of the coupling Accumulation of local axial displacements between the coupling and pipe caused by forces/ temperature changes. 4.3 Fatigue (seldom a design case). 4.4 Torque (could occur during the last phase of the installation). 6.3 Material properties General Material selection requirements (metallic materials) for pressure-retaining parts of the repair fitting, manufacturing and mechanical testing shall comply with both the nominated design standard for the fitting and the requirements in DNVGL-ST-F101. A material selection evaluation shall be included in the documentation, e.g. in the form of a material selection report. Such an evaluation should consist of two parts: A. A general description of the function and potential failure mechanisms related to material selection and the exposed environment. The following should be included as a minimum: 1) corrosivity, taking into account specified operating conditions, including start-up and shut-down conditions 2) design life and system availability requirements 3) identification of the considered material's potential failure mechanisms in the exposed environment, and evaluation of acceptable robustness for the specified design life 4) resistance to brittle fracture and HISC 5) inspection and corrosion monitoring 6) access for maintenance and repair. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 55

56 B. A table of the items in the parts list, stating for each item: 1) the material selection (material, grade and surface treatment), including relevant material and fabrication standards 2) exposure (e.g. marine atmosphere, seawater with or without CP, bore fluid, sour service requirements, fully encased, etc.) 3) minimum design/maximum operating temperature (if assessed to be different than for the general conditions) 4) whether the integrity of the item is required for all or only parts of the design life 5) comment column addressing considerations made, etc. Guidance note: The environmental exposure of each of the different components of the repair fitting assembly at the repair location should be assessed. For repairs without a leak, e.g., a grouted clamp on a dent, the compatibility of the grout/polymer/inhibitor fill material with seals should be considered. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Metallic materials The following parameters shall be specified when relevant: 1) material strength for steel, see DNVGL-ST-F101 2) temperature de-rating according to the applied design standard, or DNVGL-ST-F101 section 5D300 3) thermal expansion coefficient and elasticity modulus 4) material toughness (CVN or CTOD) properties at minimum design temperature 5) material chemical composition 6) heat treatment condition 7) friction coefficient 8) galling limit 9) pipeline surface hardness Non-metallic materials The following parameters shall be specified when relevant: 1) material properties shall be documented in accordance with NORSOK M-710 2) thermal expansion coefficient, bulk modulus and elasticity modulus 3) thermal effects on the mechanical properties 4) storage environment condition and shelf life. 6.4 Fitting strength capacity General A fitting used for pipeline repair shall have sufficient strength capacity (resistance) to carry the relevant loads with a safe margin to failure. In general, all fittings are exposed to pressure and installation loads as well as being affected by the thermal effects and loads transferred from the pipeline. The relevant load and load combinations to be considered in the fitting design are given in [6.4.2]. A fitting s capacity to grip to the pipe wall is termed the locking capacity and shall be designed to accommodate the loads transferred from the pipeline. The margin to failure for the locking capacity is defined in [6.5.2] by partial safety factors for both loads and strength (resistance) Loads The fitting's design loads shall at least be equivalent to the pipeline's load capacity or the pressure, axial, bending, torsion and fatigue loads, equivalent to the maximum loads in operation, during installation and testing - as relevant for the fitting. The methods for establishing the maximum axial pipeline operational forces are given in Sec.5. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 56

57 An overview of loads and load combinations shall be established. The main load conditions to be included for pipeline repair fittings are: Table 6-2 Load conditions Load type Internal and external pressure Bending moment Tension, compression Torque Bending fatigue Temperature Installation Seal contact pressure Bolt pre stress Gripping contact pressure Conditions, parameters Pipeline and repair fitting design and test conditions. Seal test pressure. Maximum seal diameters. Pipeline capacity specified or limiting loads. Pipeline capacity specified or limiting loads Pipeline capacity specified or limiting loads Pipeline capacity at the butt weld specified or a specified number of bending cycles related to bending moment. Maximum and minimum related to the above capacities and limits. Cyclic temperature load effects. The seal pressure contribution from thermal expansion of the seal at maximum temperature should not yield the pipe or the seal grove material (i.e. potentially cause reduced seal pressure and a leak at shut-in). Maximum forces limitations for interaction with the pipe and on coupling internals. Upper bound values shall be used to evaluate stresses in the pipeline and fitting. Lower bound values shall be used to evaluate the margin to leakage. Upper bound values shall be used to evaluate stresses in the pipeline and fitting. Lower bound values shall be used to evaluate separation and leakage. Upper bound values shall be used to evaluate stresses in the pipeline and fitting. Lower bound values shall be used to evaluate the margin to slippage/separation. Guidance note 1: The internal pipeline design pressure is the maximum incidental pressure as defined in DNVGL-ST-F101. For hydraulic activation systems, the upper bound value is typically determined by the set pressure for the relief valve or the in-situ pump capacity. For surface pumps, the combined effects of the depth, with respect to both the sea water density and other fluid densities, need to be evaluated. They will vary slightly with depth and with temperature. For assemblies where the activation response is affected by friction, upper and lower bound limits depend on maximum and minimum friction factors. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 2: The design loads for fittings are specified in the fitting's design premise. The end user is responsible for ensuring that the pipeline loads are within the specified envelopes during installation, commissioning and operation. Pipeline intervention, such as rock dumping, may be performed to limit the pipe wall forces at the repair location for compliance with the fitting specification. Alternatively, the fitting is designed for one pipeline repair project with project-specific design loads. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Load responses Fittings installed on the pipeline respond to loads by stresses and deflections. The principles used to predict stresses and related acceptance criteria are presented in [2.3.2]. General methods for calculating stresses and related acceptance criteria are established in the standards referred to in DNVGL-ST-F101 or other recognized pressure vessel standards. In general, these pressure vessel standards specify a limit state design covering: protection against plastic collapse protection against local failure protection against collapse from buckling protection against failure from cycling loading. For pipeline repair fittings, often only the protections against plastic collapse and local failure limit states are relevant, but this shall be assessed case-by-case. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 57

58 The fitting shall be designed to accommodate the loading from the connected pipeline section and vice versa, with appropriate safety. If the standard used in the design of a fitting does not take into account forces other than the internal pressure, additional evaluations, e.g. FE analyses according to a recognized pressure vessel standard, are required in order to address the specified design loads that can be transferred to the fitting from the connecting pipeline sections during installation, testing and operation. See also [2.3.3] on the design and pressure testing of repair fittings based on the LRFD method Inner diameter tolerances at the repair location The installation of the pipeline repair fitting assembly may introduce variations to the pipe's inner diameter, e.g. from activation compression forces, a dimensional difference between the spool and pipe, a difference between the inner diameter of the pipe and inner diameter of the repair coupling, or a misalignment or offset between the spool and pipe. Local variations in the inner diameter of the pipeline may affect the pipeline's operability, such as pigability, and the local metal loss from corrosion or erosion. Acceptance criteria or local variations in a pipe's inner diameter are project-specific depending on the type of operation, and should be included as a design consideration. 6.5 Fitting grip capacity General Gripping by balls or teeth penetrating the pipeline surface requires grips with significantly higher hardness than the pipeline, thus ensuring a locking capacity that exceeds the load and load combinations with a safety margin according to safety factors given in Table 6-3. In addition to the failure modes specified in Table 6-1, the following possible failure modes causing a lack of teeth/ball penetration to be considered in the design are: teeth/balls - lack of hardness teeth - lack of sharpness teeth - lack of ductility causing brittle fracture (fragile teeth) teeth/balls - breaking due to cracks caused by stress corrosion/hydrogen embrittlement teeth - shear failure tear-out of pipe material pipe-coating thickness preventing teeth from penetrating the pipe material. The gripping capacity shall be qualified by a combination of calculations and tests. An analytical model to calculate the design gripping capacity shall be established based on theory and the qualification test results. The qualification should also include the loss of gripping segment(s) to document the margin to failure/ slipping. The tests shall be performed using an equivalent pipe with respect to mechanical properties (e.g. yield stress, tensile strength, hardness) and pipe dimensional tolerances. Alternatively, the qualification of the repair fitting covering a specified range of pipeline dimensional tolerances can be based on a combination of analytical model(s) that have been validated by test(s) and testing on one pipeline section within this range. If a range of pipeline diameters and wall thicknesses is specified, the tests shall be carried out on the outer boundaries (conservative) in order to qualify the entire pipeline tolerance range. Established acceptance criteria related to gripping (e.g. indentation marks in the pipe wall from the gripping parts) shall be verified by being measured after the FAT performed on the project-specific differential pressure, hydraulic set pressure, pipe material and inner/outer diameter. Common locking principles showing a cross-section of the pipe wall and the coupling sleeve are illustrated in Figure 6-1. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 58

59 Figure 6-1 Locking principles The locking principles can be divided into two main groups: 1) mechanical attachment between the pipe wall and fitting, as caused by the actions of auxiliary local attachments and/or friction. 2) fillet welds between a sleeve and the pipe. Furthermore, the main mechanical fitting attachment methods are based on the following two principles: 1) external compression of the pipe compression fitting. 2) internal expansion of the pipe expansion fitting. The radial contact forces between the fitting and pipe are based on the initial pre-compression and/or the pipe tension. The latter can be an effect caused by designs using wedges or similar Safety factors locking capacity DNVGL-ST-F101 applies partial safety factors to compensate for submarine pipeline uncertainties. These safety factors related to forces, strength-termed load factors and resistance factors are presented below. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 59

60 The load and resistance factors given in Table 6-3 shall be used to establish the fitting's axial gripping capacity, i.e. capacity to lock onto the pipeline. For the capacity check of the pipe wall utilization at the repair location, design resistance factors are given in [6.6]. The specified design factors apply for both internal and external radial loads on the pipeline, i.e. gripping and sealing loads. The locking capacity, i.e. pipeline gripping, shall resist all relevant loads during testing and the service life. Locking capacity is typically provided by the indentation/penetration of serrated segments or balls gripping onto the pipe wall material or by friction between the plug and pipe wall material. The indentation of serrated segments could fail by either, for example, insufficient indentation, the shearing of pipe material or the shearing of gripping teeth or balls. Locking based on friction shall be documented through an evaluation of relevant parameters and their associated uncertainties (i.e. upper and lower bound values). The safety margin for a frictional lock shall be equivalent to a locking based on serrated segments and shall be determined through a qualification process, see [2.4]. Table 6-3 Partial safety factors Type of factors Ref. to DNVGL-ST-F101 During repair and testing During operation Comments Load factors - - γ 1 the combined load factor Functional loads Includes trawl interference Environmental loads Accidental loads Pressure loads Together with p li (operation) or p lt (testing) Condition load effect DNVGL-ST-F101 specifies a condition load effect factor of 1.07 for uneven seabed. At the repair location, this factor of 1.07 is not required for uneven seabed. Resistance factors 1) - - γ 2 the combined resistance factor 2 Safety class resistance factor local buckling, x 1.14 or 1.26 Material factor x for safety class Low, and all safety classes in particular cases 2) For safety classes Medium or High respectively 1) 1) Weld material factor Resistance strain factor Allowable damage ratio for fatigue, or /0.1 Applies to the fillet weld of a welded sleeve solution For welded sleeves, for safety classes medium and high respectively For safety class medium/high related to crack-type failures. Other types of failure mechanisms must be considered separately 1) The resistance factors are related to failure modes for typical pipelines, such as ductile fractures. Fittings can have different failure modes for the attachment to the pipe and the seals, requiring other partial factors. A brittle-type failure mode for the attachment to the pipe should increase, by 10%. Material factors γ m for soft seal materials should be considered together with the documentation of this material and the lifetime extrapolations based on the qualification tests. 2) The particular case is related to a typical coupling internal displacement load condition, e.g. that the make-up axial preload on the pipe end(s), the abutment load, is reduced without affecting the actual capacity. This is the case for several coupling types and applies to the SLS condition only, for all safety classes. Guidance note 1: The design of the coupling axial capacity may be considered as a local design check (ULS), i.e. only the safety factors combination b in DNVGL-ST-F101 need be considered. Generally, the design of couplings and sleeves is not dominated by pressure containment Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 60

61 but by axial capacity. The wall thickness to be used in establishing the capacity should be the nominal wall thickness (where relevant minus the corrosion allowance), i.e. t 2 in DNVGL-ST-F e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 2: The load factors for functional, environmental and accidental loads (i.e. γ F, γ E and γ A ) are applied on the local pipeline section-load and moment-response to account for related uncertainties, such as pipeline interaction with the seabed. If the relationship between the load and response is linear, the load factor may be applied directly on the load. For free and fully restrained conditions (see [5.2]), the load factor is γ F = 1.0. The definition of the loads is given in DNVGL-ST-F101. A calculation example is included in App.F. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- How to apply the partial safety factors for pipe wall utilization and plug loads is stated in [6.6]. 6.6 Pipe wall utilization activation response General The pipe wall can be exposed to significant radial forces caused by the activation of the fitting. Such high radial forces are beneficial in order to obtain the highest gripping capacity and best sealing performances. This is of particular concern to thin-walled pipelines. For some types of fittings and applications, this can cause plastic yield of the surface only and/or the total pipe wall thickness. Possible failure modes to consider are: uncontrolled extent of yielding fracture caused by excessive tension yield or fatigue loading work-hardening of possible concern to HISC and H 2 S exposure. The concerns and acceptance criteria depend on the following: pipe surface effects a pipeline surface subject to gripping by teeth will normally get shallow indentations from the teeth, representing minor stress concentrations that normally do not affect fatigue resistance. Shallow indentations made by gripping balls are smoother indentations which normally do not affect fatigue resistance. Further, the gripping body introduces compressive stresses that may be beneficial to the fatigue resistance. Guidelines for calculating the fatigue strength of a pipe wall with gripping indentations from teeth are included in App.I. SCC/HISC of the pipe wall due to local strain hardening by a gripping mechanism Local radial compression loads The through-thickness effects of radial compression are normally related to control of the magnitude of yield and in some cases the work-hardening. A radial compressive plastic permanent yield of 2% for the pipe wall membrane is normally acceptable provided: This condition is caused by the make-up and therefore is considered as pre-tension. Further, pipe forces in operation and testing shall not cause further plastic diameter reduction of the pipe. The effects of pipeline axial forces and bending do not cause additional unacceptable accumulation of plastic strain in the area. The acceptance criteria shall be based on the possible degree of pre-tension loss caused by this additional plastic strain and possible reduced material characteristics, see DNVGL- ST-F101. Safety factors given in Table 6-4 can be used. Deviation from this criterion requires assessment according to DNVGL-RP-A203, documenting an acceptable margin to failure for relevant failure modes, see [2.4]. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 61

62 6.6.3 Local radial expansion loads The through-thickness effects of radial pipe expansion have a possible additional failure mode to the effects caused by compression - due to the risk of cracking by excessive plastic tension yield. Therefore, the ultimate capacity for such a connection utilizing the pipe in the plastic range must be based on a combination of plastic FEA, recognized acceptance criteria and testing. The pressure containment (burst) capacity of the pipe with the fitting made up can be based on the burst limit state criterion given in DNVGL-ST-F101, provided possible plastic deformations comply with the functional criteria for the pipeline (e.g. dimensional tolerances on inner diameter variations related to pigability and flow). Pipeline section dimensions Analyses to document acceptable pipe wall utilization when exposed to the plug loads, i.e. radial expansion loads, shall be based on the maximum pipe inner diameter and minimum wall thickness, i.e. measured or specified by accounting for fabrication tolerances. Oval pipes will load the clamps, couplings and plugs unevenly around the circumference. The make-up will tend to reduce the out-of-roundness. This is not expected to change the pipe wall s membrane stress, but will introduce some additional strain into parts of the pipe surfaces. However, ovality needs to be considered in the design of the component in order to allow installation and appropriate activation. Plug loads The plug exerts loads similar to the effects of internal pressure but only locally, limited to just a small length of the pipe. These loads are the radial loads from the seals and gripping segments and the axial load transferred from the plug to the pipeline through the gripping segments. The axial force represents the end cap force caused by the differential pressure. A pipeline with materials in compliance with DNVGL-ST-F101 that is subject to loads from a plug with relatively narrow loading lengths relative to the diameter has an ultimate limit state (ULS) defined by the following equivalent stresses., (6.1) =, 1.15 (6.2) This is provided: the maximum load, pressure and differential pressure combination are considered conservative small friction factors are used for the gripping segments. The application in the non-linear plastic material range is provided: certified true material behaviour is applied in the analysis the limiting stress is de-rated by the material strength factor α U = 0.96 (normally) if the material has not been subject to supplementary requirements the effects on the load of changed geometry (larger diameter) are applied when establishing the ULS condition. The allowable pipeline pressure conditions shall be based on the conditions for the ULS reduced by certain factors. These factors and their background in the partial safety factors are given in Table 6-4. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 62

63 Table 6-4 Partial safety factors for pipes subject to local radial seal and grip loads (plug loads) Property Descriptions and abbreviations Factors Safety class Low Medium High Safety class resistance factor ) , Material resistance factor 1.15 Material strength factor 1) 1.00 Usage factor = 1/(, ) General (linear and non-linear analyses) Allowable usage factors, η At activation load combined with the plug loads at maximum incidental differential pressure. Linear analysis 4) Hoop membrane (mid wall):, 1.15 Membrane equivalent stress criterion:, ) ) ) ) At activation load combined with the plug loads at maximum incidental differential pressure. Equivalent linearized stress, pipe wall surface: Non-linear analysis based on true material stress strain curve 6), 7), a) Total nominal longitudinal strain acceptance criteria (i.e. at pipe surface) Provided ECA is performed for welds: 5) 0.02, 0.02 b) Total nominal longitudinal strain No ECA required acceptance criteria (i.e. at pipe surface) ) For material where the supplementary requirement U has not been specified, 0.96 shall be used, giving a usage factor 4% lower than the numbers given in the table. 2) Effectively 3% higher due to the system test requirements. 3) Due to system pressure test requirements. 4) For areas based on FEM and linearized von Mises stress, i.e. membrane plus bending. Local stress exceeding this level can also be accepted based on further documentation, e.g. tests that show no tendencies towards cracks with a safety margin. 5) ECA is only required for assessments of plastic strains in welds. Plastic strain within a specified criterion is acceptable without an ECA for pipe base material. The ECA is to be performed according to DNVGL-RP-F108. 6) The non-linear FE analysis shall include the following capabilities: a) non-linear plastic material model, based on true stress-strain formulation b) large strains - correctly adjust the geometry (thickness) c) large displacements correctly increase the diameter and corresponding loads. 7) Non-linear analysis is an alternative or supplement to elastic analyses and is only required if the considered assembly and exposed load combinations are significantly affected by non-linear effects, e.g. plastic strains, contact surfaces, friction, large deformations/non-linear geometry effects. 8) Excluding γ FEA, which needs to be assessed based on an evaluation of the finite element model representation of the actual conditions. The FEA model should be validated by strain gauge measurements during qualification testing. Guidance note 1: Residual stress from initial out-of-roundness of the pipe prior to setting the plug is displacement controlled, and may be disregarded provided the OOR is within the 3% criterion stipulated in DNVGL-ST-F101. For a larger OOR, a case-by-case assessment is required. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 2: The yield stress is defined as the stress at which the total strain is 0.5%, corresponding to an elastic strain of approximately 0.2% and a plastic strain of 0.3%. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 63

64 Secondary (i.e. self-limiting) stresses in the pipe wall caused by setting the plug in an oval cross-section may be disregarded in the pipe wall capacity check provided the ovality is within the criterion given in DNVGL-ST-F101. Material true stress-strain curves applied in non-linear elastic-plastic analyses should be based on tensile tests of representative material (i.e. the same material grade and manufacturing method), using the total nominal longitudinal strain at tensile strength (i.e. the strain at the start of tensile strength test necking). Typically, the stress-strain curve may be established based on the applied design pressure vessel standards, such as ASME VIII division 2 Part 5, or by recognized power-law-based approaches such as Ramberg-Osgood. The radial activation load from the plug seal and grips towards the pipe wall should include contributions from both the initial activation and the differential pressure across the plug. The high pressure side of the isolation plug should include the functional load factors given in DNVGL-ST-F101, whereas the low pressure side should reflect the potential lowest pressure without load factors. For scenarios where multiple plugs are used in series to distribute the total differential pressure, the minimum pressure on the low pressure side of each plug should account for possible leaks over one or more of the plugs during the isolation period. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 3: Structural analysis to document acceptable pipe wall utilization should be based on minimum pipe wall thickness, specified minimum material strength and conservative load combinations at pipe set locations. Local pipe wall defects at the repair location, such as reduced wall thickness from corrosion, should be accounted for. Material de-rating may be based on the local temperature reflecting the environment at the plug set location during activation and isolation. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- The principle of applying higher utilization based on plastic analysis compared to that based on elastic analysis shall be verified by strain gauge measurements on the pipe. External sleeves External sleeves or clamps are used to reinforce pipes subject to internal plugging loads when the wall is too thin for plugging at the required pressure. Two methods are used for this, either separately or in combination. They are: 1) pre-compress the pipe radially 2) increase the stiffness (strengthen) the pipe radially. In both cases, the pipe's acceptance criteria shall be met prior to and after the plug has been set. The challenge for item 2 (increasing the stiffness) is to transfer the stiffness to the pipe wall by bridging possible initial clearances between the sleeve/clamp and pipe. It is recommended to perform an FE analysis which includes the following sequences of operation: a) installation and activation of the sleeve/clamp on a pipeline with upper bound internal pressure while the reinforcement clamp is being installed, and b) installation and activation of the plug. Further, it is recommended to perform strain gauge measurements on the pipe's external surface to verify the calculation. The local pipe wall response when subjected to plug loads may be sensitive to the applied load history simulating the installation of the reinforcement clamp and setting of the plug. The representative load history shall be applied in analyses and tests. Friction forces within the fitting's activation mechanism affect the achieved pre-tension and hence the pipe wall stress response, and shall be documented by tests and/or sensitivity studies and conservative assumptions. Pipeline coating layer(s) between the fitting and pipe steel wall may have a significant effect on gripping and sealing (e.g. loss of clamp pre-tension, tangential friction between the pipe and clamp, radial stiffness when setting the plug), and need to be accounted for. 6.7 Seal capacity General A fitting shall have sufficient seal capacity to isolate the specified fluid at a specified differential pressure, temperature and time, with a margin as defined in [2.1]. This applies to operations, pressure testing and after depressurizing the pipeline. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 64

65 Each seal in a series shall be designed for the full differential pressure. The seal system for repair fittings, except for isolation plugs, shall be designed to enable a seal test without requiring pressurization of the pipeline. The main sealing principles for mechanical couplings are illustrated in Figure ), 2) 3) 4) Figure 6-2 Sealing principles 1) pre-compressed soft seals enclosed by anti-extrusion rings, or 2) pre-compressed soft seals strengthened by fibres. 3) metal ribs or corners of grooves in the sleeve. 4) seal welds. Some types of seals can be sensitive to damage if they touch the pipe before seal activation. The seal installation sensitivities are discussed in [7.4] Seal design capacity Typical failure modes, where an acceptable margin to failure needs to be documented for specified design envelopes (examples of failure modes are given in Table 6-1): extrusion stability of seal material/composite anomalies in pipeline surface (e.g. grooves, welds). Calculations or tests of the seal system's response to the load conditions shall be carried out, covering the seal load conditions given in Table 6-5. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 65

66 Table 6-5 Seal load conditions Item of concern Soft seal's clearance to seal Hard seal loads Annulus seal test pressure limit Soft seal volume changes Soft seal fluid migration Calculations/tests Circumferential clearance distribution as a function of the load conditions, including unsymmetrical loads (bending), pipe geometry and manufacturing tolerances. Stresses in back-up rings or strengthening devices and safeguarding against their failure modes (e.g. warping, material plastic yield) Circumferential seal contact load distribution as a function of the load conditions, including: unsymmetrical loads (bending), pipe geometry and pipe surface discontinuities. If there is an annulus seal test feature, calculate the maximum annulus test pressure limit with respect to both pipe failure and seal failure. Calculate the volume changes caused by the fluid in contact and the temperature changes. Calculate the migration rate of the fluids to the seal based on the materials specification, at maximum differential pressure and temperature. The following main design and test criteria for the sealing performance apply: No visible leakage allowed for the required temperature and pressure range through the full service life of the repair fitting. Allowable concentricity and out-of-roundness dimensions shall be considered. Limitations related to out-of-roundness should be specified (e.g. due to local corrosion and/or longitudinal seam welding). A discussion of seals and their application is presented in Appendix [B.2] Environmental seal The use of environmental seals is generally to prevent the ingress of foreign matter to the fitting's internals or the free flow and exchange of oxygenated seawater in order to avoid corrosion. In this context, the environmental seal is not a pressure-containing seal. For fitting designs where an environmental seal is required, the design shall address all relevant failure modes for the operational envelope throughout the required design life. The environmental seal design envelope shall be documented to be compliant with the design premises for the pipeline repair project, based on data sheets from the seal manufacturer or an assessment based on design and/or material characteristics or validated through type or qualification testing. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 66

67 SECTION 7 INSTALLATION AND ATTACHMENT TO THE PIPELINE 7.1 General This section covers recommendations and aspects to consider when installing a typical slip-on sleeve-type fitting (a coupling joining pipes). However, the main elements should also be relevant for other types of fittings (see [1.1]), even though not identical in installation principle, like bolted or welded split-sleeve repair clamps, bolted or welded (hot tap) tees, etc. The limiting installation conditions shall be specified and calculated. An outline installation procedure shall be established. Guidance note: These installation conditions are in particular related to items 6, 7, 8 and 9 of the following operations: 1) Seabed preparations to enable the carrying of heavy frames. 2) Installation of pipe-end manipulating devices (e.g. H-frames) if required. 3) Cutting and removal of damaged pipeline sections. 4) Coating removal if applicable and preparation of pipe ends. 5) Manipulation and aligning of pipe ends or excavations. 6) Subsea measurements and surface adjustments of possible intermediate pipe sections and the fitting. 7) Deployment of the fitting, its installation tool and the intermediate pipe section. 8) Installation and activation of the fitting and possible welding. 9) Testing and inspection of the repair, including possible seal testing. 10) Pressure testing of the pipeline, if required. 11) Deployment of the repaired pipe section to the sea-floor from the lifting frames, if used. 12) Seabed preparations/protection. The pipe ends should be prepared for the coupling installation. Couplings are fitted to the external parts of the pipe and normally require the removal of the pipe coating. Most couplings also require a certain evenness of the pipe end and surface. Therefore, subsea chamfering, grinding or machining may be required. Installation of the coupling onto the pipe ends may require strict control to avoid damage to seals. Therefore, special tools may be required to control the coupling installation, as well as for coupling activation and testing. In most cases, it will be practical to join the pipeline ends using a spool piece (intermediate pipe section). After aligning the pipe ends, the coupling is moved to the correct position and activated. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Entry of fitting The limiting parameters related to the following cases shall be established, and shall include: misalignment angles and offset limiting bending moments, contact forces allowed during installation, and related friction forces to overcome during installation. Such entry cases shall include the following: Case 1 - Entry on pipe end 1. The angular and radial motion of the coupling is normally governed by the rigidity of its suspension system. The pipeline is held in position by the installation system. Misalignment is less than the maximum possible misalignment for the coupling (based on clearance between the pipe and coupling). Final entry is obtained by the coupling s suspension system providing axial positioning of the coupling, within specified radial and angular alignment envelopes. Figure 7-1 Case 1 Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 67

68 Case 2 - Entry on pipe end 1. The misalignment is larger than the maximum misalignment for the coupling based on the clearance to the pipe. Final entry is obtained by the coupling s suspension system providing axial positioning of the coupling, within specified radial and angular alignment envelopes. The risk of jamming is to be considered. Figure 7-2 Case 2 Case 3 - Misalignment is less than the maximum possible misalignment for the coupling (based on the clearance between the pipe and coupling). Entry on pipe end 2 when pipes are misaligned and offset relative to each other. Both pipe ends are held in position by the installation system. The coupling's angular and radial motions are governed by the rigidity of the coupling's suspension system. The pipeline is held in position by the installation system. Figure 7-3 Case 3 Case 4 - Entry on pipe end 2, as case 3, but the alignment tolerances, as governed by clearances, are exceeded. The flexibility of the pipe suspension system, including the pipes themselves, must be considered. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 68

69 Figure 7-4 Case First end entry control Two categories of installation sensitivity are defined: 1) The sensitive type: no interaction between the pipe and coupling allowed prior to activation 2) The less sensitive type: limited interaction forces are allowed. Category 1 requires the strict control of geometric installation parameters, and therefore an accurate monitoring and control system. The limiting combination of in-plane eccentricity x and misalignment angle a (see Figure 7-1 and Figure 7-5) are represented by (Case 1): 2 > (7.1) where, e = diametric clearance (considering constant internal diameter): D c -D D c = coupling bore diameter D = pipe external diameter including tolerances x 1 = eccentricity (offset from centre line) at entrance a = misalignment angle (radians) y 1 = overlap length i.e. degree of sleeve displacement over the pipe(s) at the moment of time considered. Maximum y 1 is the length of the coupling. For installation systems with active control to give the optimum position of the actual offset from centre at entrance, the limit is (Case 2): > 1 These limits also apply to Category 2 couplings, but the degree of control and monitoring can be relaxed. Guidance note: The shape of the coupling can be used to guide the installation e.g. with a funnel to facilitate entry during the initial installation. A practical method of controlling the interaction forces is to provide compliant radial support to the coupling during the installation. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 69

70 Figure 7-5 Misalignment and eccentricity 7.4 Seal protection design The seal is the most sensitive part of a coupling, so it is preferable for it to have no interaction with the pipe prior to activation, i.e. the above Category 1. However, the seal design must document acceptable robustness for the most adverse combination of installation tolerances specified in the technology and/or design premises for the fitting. This may for example be obtained by using a sealing system which is retracted from the inner circumference of the coupling. This system requires an increase in the inner radius of the seal relative to that of the coupling of at least: + (7.2) where, n = axial length from the coupling entrance to the end of the same inner diameter. (Length of equal internal diameter) y i = distance from the coupling entrance to the seal. s = safety distance (say 0.3 mm) to compensate for deflections and possible protrusions on the pipe end. Furthermore, this system requires the seals to remain concentric in the coupling until activation, and that no axial internal friction force inside the coupling can activate the seals. 7.5 Water block Water trapped in cavities which are to be sealed off by the installation can resist further displacements and shall be avoided, unless proven to have no such adverse effects. Further, entrained water in high in-service temperature may cause boiling and pressurization of the trapped volume, potentially damaging the seals. Guidance note: This is of particular concern to designs with several main seals in a series. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Second end entry Installation of a coupling onto the second pipe end requires the careful alignment of the pipes. For plane misalignment of Category 1 couplings with position control during installation (Case 4), then: e > b y 2 x 2, when b y 2 2 > x 2 (7.3) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 70

71 Otherwise e > x 2 (7.4) where, b = angular misalignment between the pipe ends' longitudinal axis (radians) x 2 = offset between pipe ends y 2 = half coupling length (bridging one pipe end) For Category 2 couplings, the misalignment angle b is calculated based on deflections caused by the contact forces inside the coupling. The pipe straightness tolerance shall be included, either as an addition to the pipe diameter, or as part of the misalignment angle b. 7.7 Misalignment limitations The above illustrates in-plane limitations. The global misalignment and offset, i.e. in two planes, must be used to control actual conditions. For this purpose, the root of the sum of squares for conditions in two 90- degree planes can be applied. Example: For second entry, plane v and h : b = 2 + h 2 (7.5) x 2 = h (7.6) 7.8 Activation The bending moment caused by the coupling's activation process shall be calculated. This applies to couplings which bridge two misaligned pipe ends, each with stiff supports. The calculation of this moment shall include: 1) misalignment, 2) pipe straightness, 3) stiffness of pipe ends and their fixation, 4) ability of the coupling to absorb the misalignment without aligning the pipes. The stresses of the internals of the coupling caused by the activation shall be evaluated. This shall include a risk assessment of: 1) over-stressing causing unacceptable deformations or breakage, 2) collapse of the coupling or parts of it, 3) malfunction of mechanisms inside the coupling, 4) uneven seal loads around the circumferences caused by eccentricity between the coupling and pipe. The pipe stresses, deflections and safety factor against collapse during the activation shall be established. Guidance note: When the repair assembly includes misalignment joints (e.g. ball joints for angular misalignment and telescopic joints to adapt to axial tolerances) such that the resulting angular and off-centre misalignments are negligible, the local bending moment on the pipe end from activation may be disregarded. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 71

72 7.9 Seal test The mechanical coupling and clamps shall be designed to allow for a seal test without requiring pressurization of the pipeline. Seal test guidelines are given in [12.2.2]. The radial seal load during the seal pressure test shall be established and compared to the limiting (i.e. minimum and maximum acceptable) seal loads. The limiting seal loads shall be based on tests or documented experience Monitoring and control General The diverless installation of subsea pipeline fittings requires: 1) a system to control the forces and displacements 2) forces to displace and manipulate the fitting 3) a monitoring system to verify that manipulations comply with the limits for the pipe and fitting 4) a monitoring system to verify that the fitting is installed properly. Monitoring of welding shall comply with Sec.11 5) a test and monitoring system to verify the seal's function. Guidance note: The monitoring system may comprise a range of TV cameras, alignment sensors, displacement sensors and force and pressure sensors, etc. The monitoring system should verify that each parameter which can cause a failure is within acceptable limits. A general principle for the monitoring system design is that: - The failure of a monitoring system (sensor) should not stop the operation. - A redundant system or alternative method is required to control and monitor the operation. On this basis, the design should be such that a monitor can display all critical parameters. This could include monitoring without sensors. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Monitoring of pipeline isolation Monitoring of the isolation to verify the plugs' seal integrity (see Sec.8) must be done and approved before the pipeline repair commences. The operating procedures must include a minimum period of stable pressure (the acceptance criterion is typically 4 hours of stable pressure depending on the fluid type and volume) and should take into consideration the extended monitoring time required to verify the seal performance. This should take into account the effects of changes in the ambient temperature and pipeline operating temperature and pressure variations that affect the determination of the seal's condition. The isolation should be monitored at agreed intervals throughout its duration. Risk-based preparedness procedures covering contingency scenarios where acceptance criteria are not met shall be established as part of the mobilization for the pipe-isolation project Acceptance criteria The connection operation shall be planned and conducted in such a manner that the specified functional requirements are met, i.e.: controlled within the established limitations monitored the fulfilment of the functional requirement (e.g. seal test) must be recorded. The records shall serve as documentation of the fulfilment of the requirements. The manufacturer of the fitting shall identify and list the functional criteria to be checked. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 72

73 SECTION 8 ISOLATION PLUGS 8.1 General This section covers the type of fitting referred to in [1.1] as isolation plugs. An isolation plug is an internally installed fitting with the main purpose of isolating parts of a pipeline from the pipeline pressure and content in conjunction with other repair or intervention work on the pipeline. The isolation plug tool is inserted into the pipeline typically through a topside, onshore or subsea pig launcher or through a branch pipe fitting. The plug serves as a temporary barrier to the pipeline pressure. It is brought to a suitable set location, normally by means of a pigging/pumping operation, and is activated by means of an internal activation mechanism. The isolation tools can be operated through either a tether/ umbilical or remote control (via through-wall communication). Where it is not practicable to deploy an isolation plug that runs inside the pipe, then pipeline isolation may be provided by an isolation plug deployed through the side branch of a hot tap fitting. This type of plug typically activates the seal hydraulically and the retaining force is structurally supported mechanically by the branch pipe. Some branch pipe fitting plugs are both hydraulically actuated and held in place by pipeline pressure, with differential pressure over the plug head. The barrier requirements for the isolation plug are given in [2.2.2]. 8.2 Design Failure modes The design of the isolation tool shall demonstrate safeguards against possible failure modes. A failure mode and effect overview shall be established. The method used to demonstrate safeguards against possible failure modes shall be qualified. An in-line isolation tool operation consists of the following main phases: plan the isolation operation and design the isolation plug tool for the operation deploy the tool to the set location set and test the tool monitor isolation unset the tool return and retrieve the tool. Typical failure modes related to these phases should be identified, such as: failure to get the tool to the set location failure to set failure to isolate failure to unset failure to retrieve the tool. Tool design and operation procedures should be reviewed to ensure that relevant elements and contingency scenarios are covered. The main risk related to safety is the loss of a barrier, e.g. a leak through one or both barriers. Examples of main concerns in the other phases are: flow control/pump capacity lack of safety systems (e.g. on pump) procedures training of operators ensuring that a multi-entry pipeline is operated correctly during isolation (pressure control) redundancies, e.g. procedures for unsetting a plug in the case of a loss of communication or power plug(s) get stuck in the pipeline. Typically mitigated by pigability assurance (e.g. all potential obstructions identified, all valves ensured fully open, debris and scaling sufficiently cleaned by pigging) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 73

74 validity of the engineering data/correct pipeline data verification of the set location's condition cleaning/pigging history of the pipeline. Guidance note 1: Failure mode types 2, 3 and 4 are related to the main functionality of the isolation tool and covered by a Failure Mode, Effect and Criticality Analysis (FMECA) and Fault Tree Analysis (FTA) of the tool. Failure mode types 1 and 5 are related to operational activities like tool launching, pigging and retrieving and should be managed by a risk assessment of the operation. Fault Tree Analysis (FTA) should be done to quantify the probability of the tool failing during the different phases of the operation in order to ensure compliance with the requirement in [2.1]. The probability of tool failure during the isolation phase (i.e. failure mode 3 above), while the pipeline is being repaired and prior to the reinstatement of pipeline integrity, should comply with the safety class high criteria for the isolation of hydrocarbons. Lower safety classes may apply for other operational phases, isolation locations and pipeline content, as per DNVGL-ST-F e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 2: Pipeline repair operations involving in-line isolation tools must take into account the following attributes of this isolation method: In-line isolation plugs are designed to isolate the full pipeline pressure. Both the seal and grip may be energized by the pressure differential in combination with the initial hydraulic or mechanical activation. The isolation should be assessed for compliance with the relevant safety classes as per DNVGL-ST-F101, and a generic FTA will quantify the probability of failure (POF). Projects should aim at achieving the lowest POF possible by ensuring sufficient pressure to energize the isolation. These values are affected by the way the isolation is energized, as explained below. Pressure energized isolation: When sufficiently high differential pressure energizes the isolation, the isolation is in a state of self-lock, as it is pressureenergized. Internally activated energized: When differential pressure is below the level of "self-lock", the state can be referred to as hydraulic lock, normally activated by an internal hydraulic system or equivalent activation mechanism in the tool. In-line isolation plug modules are further designed to isolate in one direction only. The pressure regimes throughout the planned isolation operation should be documented, agreed and assessed w.r.t. risks involved. Additional tooling or isolation modules should be considered to support the requirement for pressure testing of the repaired section of the pipeline, as the testing of the repaired section may involve a higher pressure than the pipeline pressure. The isolation tool can as such serve as pipeline isolation during the repair and isolation of the repaired section during pressure testing. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Structural integrity of an isolation The structural integrity of an isolation using in-line isolation tools depends on both the structural integrity of the plug and the integrity of the pipeline wall exposed to the plug loads. The structural integrity of the isolation plug shall comply with the safety philosophy given in [2.2.2] and the design requirements for fittings given in [2.1]. The pipe section where the isolation plug tool is set will be exposed to radial gripping and seal pressure loads from hydraulic activation and response from the differential pressure across the plug. These loads must be calculated to verify utilization of the pipe material capacity according to the acceptance criteria. For criteria governing elastic and elastic-plastic calculations, see [6.5.2] and [6.6]. The locking capacity, i.e. pipeline gripping, shall resist all relevant loads during the isolation period and testing. Locking capacity is typically provided by the indentation/penetration of serrated segments into the pipe wall material or by friction between the plug and pipe wall material. The indentation of serrated segments could fail due to insufficient indentation, the shearing of pipe material or shearing of gripping teeth. The locking capacity shall be determined according to [6.5] in order to comply with the required safety margins of this RP. Locking based on friction shall be centred on an evaluation of relevant parameters and their associated uncertainties (i.e. upper and lower bound values). The safety margin for a frictional lock shall be equivalent to a locking based on serrated segments and shall be determined through a qualification process, ref. [2.4]. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 74

75 8.3 Testing of in-line isolation tools The fitting testing philosophy is covered in [2.4.2] and detailed in Sec.12. Sec.12 and App.C describe typical tests (basic tests, type tests, FAT tests and final tests) that also apply to in-line isolation tools. Some additional testing may be relevant for the FAT in order to test in-line isolation tools in the intended environment. Such testing can be undertaken to verify actual performance or explore the performance limits in the actual environment, such as: pigging testing, including testing the friction capacity, pigging, flipping and reversing pressures in the pigging arrangement testing the performance of the tool-tracking system testing/demonstrating contingency unsetting systems system integration testing (SIT). This may include installing the tool in the real or a replica pipeline section and a specific pipeline medium to simulate an operation external gas pressure test, to document that a possible leak in the hydraulic system is not affected by the intrusion of external environment overpressure battery capacity performance setting and unsetting the plug tooth embedment trials strain gauge verification of the FEA parking of plug - lifting/loading and unloading the plug train into the pig trap isolation tests at the full range of expected pressures The relationship between a potential pressure drop caused by a leak and an equivalent volume leak rate per time should be calculated for considered systems/scenarios. This is to make any risk assessment easier to quantify in terms of pressure loss compared to the volume of liquid/gas. after testing, a risk assessment shall be performed to highlight any remaining risks, based on ALARP criteria an analytical model predicting the annulus pressure response to possible temperature fluctuations and variations in isolation pressure during the FAT and barrier testing (after setting the plug) should be established, to document if measured variations are caused by an actual leak or environmental variations. 8.4 Installation and retrieval of in-line isolation tools The installation of in-line isolation plugs is divided into five distinct phases: launching and pigging setting isolation unsetting return pigging and retrieval. Operating procedures shall cover each of these phases, including contingency procedures for the relevant failure modes Pigability assessment Tools running inside the pipeline system, such as for pipeline isolation, shall be verified for their ability to traverse all relevant pipeline sections and components between the launch/retrieval point and the set location. Such verification may be referred to as a pigability assessment, and is a theoretical assessment that includes all relevant pipeline components. The pigability assessment documents the margins against a stuck tool. A vital part of the quality of this process is the timely provision and correct use of pipeline data, as the Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 75

76 evaluation of tool designs and configuration depends on this. This includes but is not limited to a thorough review of: pipeline isometric drawings the pipeline's design, fabrication and installation résumé (DFI) detail drawings of pipeline components: a) valves b) tees, wyes c) flanges d) bends e) isolation joints f) flexible joints, etc. operational service records to assess the need to use cleaning pigs in-line inspection calliper data reports gauge pig data reports. In some cases, the results of intelligent inspections, the pigging history and UT measurements may be required to clarify aspects related to the tool's pigability Pigging and setting The pigging and setting of pipeline isolation tools require a well-planned operation and procedures for controlling the: metering of pumping flow, pumping pressure, total pumped volume and tracking of isolation tool limit on operational parameters, maximum flow, requirements for flow control, maximum pigging pressure differential and relief valves on pump spreads contingency plans for pigging, setting, isolation, unsetting and retrieval of tools, including plans for retrieving the plug if it becomes stuck in the pipeline during pigging capacity for bleeding off gas in order to return the tool to the launcher/receiver. The pigging and setting phase of the operation is a joint operation requiring well planned and reviewed operations procedures involving the pipeline operator and the contractors for the isolation and pumping services. As the isolation depends on the pipe interface at the actual set location, the acceptance criteria for the approved barrier shall be clearly defined. Guidance note: Typical procedures required for setting and unsetting: project communication lines, roles and responsibilities metering of pumping flow, pumping pressure, total pumped volume and tracking of isolation tool limit on operational parameters, maximum flow, requirements for flow control, maximum pigging pressure differential, relief valves on pump spreads contingency plans (in particular for the phases involving the setting, isolation and unsetting of tools, and for the retrieval of stuck plugs). The contingency plan should also cover minor deviation events from the procedures to prevent delay and increase the clarity of procedure step end conditions. aligned operating procedures and the coordination of a joint operation are key to risk reduction. Contingency system(s) for unsetting/releasing the tool are required and must be activated by external means to take account of a potential loss of communication with the tool. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e In-line isolation period Typical in-line isolation tools are tested and operated to provide temporary isolation, with durations ranging from a few hours up to 1 year. The lifetime of the isolation (and individual elements of the tool) should be evaluated when considering longer isolation durations. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 76

77 The main concerns are in general damage to the pipeline or isolation tool. First step: agree with the industry stakeholders on the definition of the term permanent or define the required length of isolation based on realistic scenarios. Examples made only for the purpose of listing concerns: short-term isolations/temporary isolation (<1 year) intermediate isolations (1-10 years) permanent isolations (until the end of the pipeline service life). Second step: define requirements and risks during the isolation period. Typical questions and specifications to be stated in relation to isolation tools: Is monitoring of the isolation required throughout the isolation period? What are the consequences of reduced/lost isolation? Is it possible to retrieve the isolation tool for maintenance/replacement? Operating conditions/production upsets that may affect the isolation (like higher or lower pressure, min/ max temperature, etc.). Pipeline content causing risks of future phases of an isolation (hydrate formations, water/wax or other operating conditions). Third step: identify gaps between the current (design lifetime) limits of isolation tools and the desired isolation period: list gaps and identify solutions determine what must be replaced on the isolation tool determine what tests can be done to qualify an extension to the lifetime. For all isolation categories, the following areas (below) and the results of the gaps (above) should be studied, and testing performed to qualify/verify compliance with the requirements: long-term effect on the isolated pipeline integrity, preservation of the pipeline to prevent damage like galvanic corrosion dead legs need to be considered w.r.t. corrosion threats, such as microbiologically influenced corrosion (MIC) structural elements - factors affecting integrity like isolation/preservation medium, galvanic corrosion, cyclic loading (temperature and pressure) main seals - factors affecting the integrity/lifetime of an activated seal: pressure, temperature isolation/ preservation medium, cyclic loading (temperature and pressure) long-term degradation of elastomer seals/fluid compatibility monitoring frequency/requirements if monitoring through tool a) onboard battery capacity b) assumption that all means of communication outside the pipeline can be replaced/maintained for retrieval after long-term static storage, consider the performance of: a) the hydraulic system (general robustness, seal performance, hydraulic fluid performance) b) the unsetting functions (mechanical, springs and packer retracting/unsetting the modules) c) the pigging discs' performance (after years in-situ). For permanent in-line isolations (i.e. not meant to be removed), it may be assumed that after the initial verification period, further monitoring, unsetting or return pigging is not required. As such, the replacement of batteries and maintenance of the hydraulic system, mechanisms and soft components will not be required for retrieving the tool. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 77

78 SECTION 9 HOT TAPPING 9.1 General Hot tapping is the technique of attaching a branch fitting - welded or a bolted mechanical fitting - to a pipeline in service, and then creating an opening in that piping or equipment by cutting a branch outlet within the attached fitting. 9.2 Hot tap fitting design Failure modes The design of the hot tap clamp and connected goose neck shall demonstrate safeguards against possible failure modes. A failure mode and effect overview shall be established. The method used to demonstrate acceptable safety factor against possible failure modes shall be qualified. The following main phases should be covered and, where applicable, include relevant contingency scenarios: a FAT and SIT of the hot tap clamp, hot tap ball valve, hot tap cutter tool and goose neck assembly pipeline preparations, e.g. seabed interventions, coating removal and the removal of the pipeline longitudinal weld cap installation of a hot tap fitting and branch isolation valve integrity of the pipeline during hot tap installation test of the hot tap fitting and valve (to verify barrier integrity) make-up of hot tap cutter tool leak test of hot tap clamp and cutter assembly activation of cutting tool, control and monitoring of qualified cutting rate verification of successful cutting and retrieval of coupon re-verify barrier integrity removal of cutting tool and blinding of hot tap branch tie-in loads during spool installation design conditions new connection test condition new connection. The main failure modes in these main phases of the operation are: performance of the cutting tool excessive structural utilization of the pipeline and/or hot tap clamp relative displacement between the pipeline and hot tap clamp, i.e. failure to grip, and between the pipeline and goose neck spool clamp, causing loading on the hot tap branch and relative displacement at seals not covered by the design premises seal failure between the pipeline and hot tap clamp coupon dropped/unable to retrieve coupon after cutting Structural integrity The design of a hot tap clamp, tool and operation shall be risk-based and comply with the safety level criteria given in DNVGL-ST-F101 and this RP. The design requirements related to hot tapping are given below: All design envelopes for the hot tap fitting shall be specified in the design premises, e.g. governing design standards and acceptance criteria, design life, pipeline design loads at the hot tap location, pipeline section dimensions and tolerances, material specifications, spool tie-in loads, pipe wall force envelopes in the mother pipe at the hot tap location during hot tap clamp installation and in-service, and design loads during operation. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 78

79 The design and qualification of the hot tap clamp shall comply with the safety philosophy, see [2.1]. The installation procedures for the hot tap clamp, hot tap tool and tie-in spool shall be used as a basis for defining the load history applied in the design calculations. The design documentation shall include a failure mode, effect and criticality analysis (FMECA) that identifies all the main threats related to all the relevant phases, including a qualitative probability and consequence assessment, mitigations and documentation of acceptable margins to failure. The free span of the mother pipe at the hot tap installation, with the hot tap clamp installed, shall comply with DNVGL-RP-F105. The structural integrity of the mother pipe with the loads from the hot tap installation and the in-service design loads shall comply with the design criteria stated in DNVGL-ST-F101 and with [6.6]. The pressure containment and the structural integrity of the hot tap clamp, for relevant phases, e.g. lifting, installation, make-up, testing, cutting, in-service, accidental loads, shall comply with the requirements given in this RP, see [2.1], Sec.6, Sec.9 and code break guidelines in Appendix [A.8]). The required pre-tensioning level of the hot tap clamp shall account for the local pressure level in the pipeline during installation and the possible load variations during in-service life. A reduction in pipeline pressure after a hot tap installation will reduce the clamp pre-tension level, affecting the gripping and sealing capacity. An increase in pipeline pressure after a hot tap clamp installation will increase the utilization of the clamp assembly. The gripping or friction capacity in relation to the mother pipe shall be documented to resist the branch pipe and mother pipe design loads (static and cyclic) derived from the design factors given in Table 6-3. The seal design and qualification of the hot tap clamp shall comply with the requirements given in [6.7] for relevant phases (e.g. in-service, cutting, accidental loads). The strength and pressure containment of the hot tap fitting shall be documented by the test specified in Sec.12. The structural integrity and sealing performance of the hot tap valve shall comply with the requirements given in DNVGL-ST-F101 for pipeline components The corrosion protection of the hot tap assembly, i.e. hot tap fitting, mother pipe, branch pipe, goose neck and goose neck clamps, shall comply with DNVGL-RP-B401. The following features shall be covered by relevant design sensitivity studies and functional performance and integration tests in order to document acceptable performance: interface with the hot tap ball valve, hot tap cutter tool and goose neck spool interface with make-up tools and the ROV panel evacuation of seawater, pressurization of the hot tap clamp and cutter tool retrieval of coupon cut from mother pipe environmental control for welding if relevant hot tap cutter performance on specified branch size, mother pipe diameter, wall thickness and material specification. Further potential contingency scenarios: recovery due to failure events during installation cutter, pilot and coupon retainer failure hot tap tool stall unable to break free from stall unable to recover tapping machine control system failure hydraulic failure communication failure seal failure valve failure. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 79

80 Reference is also made to API RP 2201, which covers the safety aspects to be considered before and during hot tapping on in-service piping Sealing See [6.7] Testing See Sec.12. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 80

81 SECTION 10 ABOVE WATER TIE-IN 10.1 General Pipeline repair by above water tie-in (AWTI) includes lifting one section of the pipeline above water using a vessel in order to allow an atmospheric weld connection when replacing the damaged section. AWTI can be used as a pipeline repair method for a short damaged section (e.g. damage which involves the substitution of a length of pipe in the range of one or two joints) or for a long damaged section where the AWTI is performed at each end of the new long pipeline section. Repairing a long damaged section involves laying a new length of pipe and then one or two AWTIs are performed to complete the repair. The above water tie-in method is well proven through several offshore pipeline construction projects. AWTI as a repair method is mainly limited by the minimum required vessel draft and the maximum water depth to control acceptable stresses in the pipe (i.e. shallow water depths). An alternative to AWTI is the surface welded flange method: The surface welded flange method consists of lifting the cut pipeline end to above the sea surface by means of vessel davits and/or anchor winch lines in a J-shape configuration (with the aid of buoyancy tanks if required to reduce the lifting weight and/or the pipe stresses), and lowering it back down to the seabed for a subsea tie-in (i.e. with a flanged spool). The surface welded flange method does not require the pipe end section to be horizontal, giving a more favourable condition in terms of lifting loads and, in particular, of pipe stress for a specified water depth. The method is therefore typically applicable to greater water depths than the AWTI. In addition, as the repaired pipeline configuration does not have an S-lay shape with horizontal configuration at the top, the method is applicable to scenarios with specific constraints (e.g. many lines running parallel to each other at a distance which is less than the water depth, post-trenched pipelines in hard soils, etc.). The typical approach is to cut the pipeline subsea and use end plugs installed at the pipe ends (pipeline recovery tools or isolation plugs). Cutting the pipeline after lifting would entail some specific requirements: The capacity of the damaged pipe to withstand the lifting loads. The capacity of the vessel to lift the pipe in the non-cut condition. This scenario could be applicable mainly to small/medium-diameter pipelines in very shallow waters Design Failure modes The following main phases should be covered and where applicable include relevant contingency scenarios: planning and engineering of the AWTI operation pipeline preparation for AWTI, e.g. seabed interventions, coating removal, installation of pipeline lifting points mobilization and preparation of the vessel, e.g. anchor positioning capacity, pipeline lifting and restraining arrangement, welding equipment and procedures pipeline isolation using isolation plugs AWTI operation: lifting restraining of pipeline cutting, removal of damaged section, weld preparation welding NDT lay-down Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 81

82 commissioning operation. The main failure modes for these main phases of the operation are: excessive utilization of the pipeline quality of the tie-in weld not compliant with the requirements given in DNVGL-ST-F Structural integrity The planning of the AWTI operation shall include a pipeline FEA to document that the pipeline section affected by all phases of the AWTI operation complies with the design criteria given in DNVGL-ST-F101. These analyses shall be based on project-specific parameters, such as: water depth, soil properties, pipeline section dimensions, material, pipeline coating thickness, pipeline conditions, vessel size and lifting capacity, safety class for the operation, sea state, lifting components (i.e. buoyancy tanks). The analyses simulating the AWTI operation shall include sensitivity analyses to account for davit pay-in, pay-out error due to inaccuracy and the possible loss of a buoyancy tank. When planning the AWTI operations part of the preparedness strategy, the water depth range suitable for AWTI repair should be established by analyses. For above water tie-in, the following conditions shall be assessed: Preparedness plans shall be established, including aborting the operation in the case of unplanned events (e.g. if the weather window is exceeded). Before cutting off the pipeline, the isolated section shall be depressurized (e.g. using small-bore hot tap equipment). The hot tap branch in combination with additional isolation plugs installed with the highpressure side towards the weld enables local hydrostatic pressure testing of the above water tie-in weld and the pipeline sections affected by the lifting. The reaction forces and moments at each support and lift point shall be established by analyses to document the acceptable safety level of the supporting structures and vessel handling and positioning system for all phases of the operation. The pipeline shall be adequately aligned and restrained during the pipeline cutting operation to ensure safe operations. The vessel performing the lift shall be adequately moored to control its position during lifting and lowering. Anchors may be used to position the vessel during the operation. The pipeline integrity shall be documented to comply with the design requirements given in DNVGL-ST- F101, both during the lifting operation to retrieve the pipeline and during laying after welding. The lifting is typically performed from a vessel fitted with a number of davit cranes, which are connected to the pipeline sections by lifting clamps (installed either subsea or on board the vessel performing the laydown prior to the AWTI). To aid the pipeline lifting and reduce pipe stresses, a number of buoyancy tanks may be used. The two pipeline sections to be connected need to be prepared with an adequate overlap in length suitable for the lifting height to ensure there is enough pipe that can be cut back, when the pipe is restrained and aligned at the vessel side, for the final cut and preparation for welding. The vessel will be required to manage the lowering of the connected pipelines from multiple lifting points simultaneously. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 82

83 The lowering process shall take into account any additional length of pipe to be laid on the seabed. The repaired pipeline is typically lowered to the seabed with a lateral movement of the vessel to allow for the additional length of the omega shaped configuration that is obtained following the repair (this is achieved with a sequence of small and controlled vertical and lateral movements). The welding shall be performed in compliance with DNVGL-ST-F101 Appendix C, and NDT according to DNVGL-ST-F101 Appendix D. Fatigue damage to the affected pipeline sections from the above water tie-in operation shall be documented. Guidance note: 1) The pipeline deflected shape should be monitored and controlled in order to maintain the pipe stresses within the pre-defined/ allowable limits. 2) The actual measured pipeline configuration should be compared to the static and dynamic AWTI calculations previously performed during the engineering phase; in this way, unexpected situations and/or any discrepancies may be immediately evaluated (and, if relevant, compared to the sensitivity analyses, such as those performed to account for possible buoyancy tank loss and/or error in the pay-in/out of cable). 3) In order to maintain pipeline stresses within acceptable limits, all key parameters should be monitored while the AWTI is being performed, such as the davit loads, angles, pipeline end section angles and pipeline deflected shape. 4) There is a risk of lifting clamp slippage on the pipeline's external concrete coating: if this occurs, correct clamp closure and/or anti-slipping collars together with reduced davit-lifting angles are essential. 5) The gripping capacity on concrete coating may have been reduced due to e.g. marine growth and the degradation of the coating's mechanical properties. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Testing In general, hydrostatic pressure testing of the pipeline sections affected by the AWTI repair operation is required to document pressure containment after the incurred load exposure from the above water tie-in operation. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 83

84 SECTION 11 WELDING 11.1 General This section covers subsea hyperbaric dry welding by remote operation, i.e. automated welding without personnel attendance in the habitat. Remote hyperbaric dry welding comprises fillet or butt welding used as a primary strength member or for sealing purposes, and may be utilized in connection with pipeline repair, modification, hot tap and tie-in. Diver-assisted hyperbaric welding for pipeline repair and tie-in (dry-habitat welding, mechanized and manual welding) is covered by Appendix C of DNVGL-ST-F101. The requirements below are based on the principles of those requirements and have been extended to cover remote welding operations in and exceeding the water depth that can be reached by divers. Thus, this document represents a supplement to the requirements specified in DNVGL-ST-F101. Figure 11-1 shows a typical fillet weld when welding starts and a macro section of a completed GMAW with a large number of passes. It is intended to be used for the deep-water remotely operated welding of a sleeve to a pipeline. Figure 11-1 GMAW - welding setup and completed fillet weld 11.2 Welding concept A welding concept shall ensure that welding is repeatable and results in welds with consistent properties and an absence of injurious flaws. This means that: a qualified welding procedure shall be followed essential variables shall be established, adhered to and monitored non-destructive testing (NDT) shall be performed to ensure that weld defects are within defined maximum acceptable limits or, if NDT is not performed, then welding shall be performed by systems qualified for defect control through process parameter monitoring, and visual monitoring shall ensure that geometrical deviations are within defined maximum acceptable limits. A welding concept shall be established in order to achieve the required characteristics for remotely operated hyperbaric welds. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 84

85 A welding concept is defined by the following main parameters: joint configuration (butt weld, fillet weld or other) intended weld geometry extent of NDT essential welding and environmental variables for monitoring. Welding concept base cases This document describes general principles and in particular two welding concept base cases, with associated qualification routes. The base cases are: a) qualification of both equipment and welding procedures b) qualification of welding procedures for a particular application using already qualified equipment. Further details of the qualification routes for the welding concept base cases are given in [11.8] Hyperbaric welding General Welding shall, as a minimum, conform to the definition mechanized welding in DNVGL-ST-F101, Appendix C: Welding where the welding parameters and torch guidance are fully controlled mechanically or electronically but may be manually varied during welding to maintain the required welding conditions Welding process (informative) The following aspects should be considered when selecting the welding process and consumables for hyperbaric welding: Operating tolerances arc stability for the relevant habitat pressure, including sensitivity to residual magnetism metal transfer characteristics bead stability cooling rate: preheat and interpass temperature requirements welding environment (dew point and atmospheric composition) Weld robustness weld metal strength and toughness environmental oxygen level (risk of weld metal oxygen pick-up) and potential risk of loss of weld metal toughness hydrogen level (risk of hydrogen entrainment from welding environment) and potential risk of hydrogen-induced cracking (cold cracking) fusion characteristics (susceptibility to common welding defects) Productivity deposition rate maintenance requirements (e.g. grinding) consistency and repeatability of acceptable quality welding for extended periods of time. The possible incidence of welding defects and other failure mechanisms should be considered when selecting the welding process and material combination, and the development of welding parameters should be considered when planning. The current range of experience of automated welding processes suitable for remote operation is limited to gas metal arc welding (GMAW) and gas tungsten arc welding (GTAW). Hence, the relevant characteristics of these processes are given below. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 85

86 GMAW The major advantage of GMAW for hyperbaric dry welding is its ability to maintain a stable arc across a wide range of pressures and deposition rates and its flexible filling capability. However, this necessitates special control techniques which modify the static and dynamic characteristics of the power supply according to the demands of the welding arc. Contact tip wear may limit the arc-on-time so that contact tip and wire are essential for large welds. Depending on the welding parameters, excessive levels of weld spatter may cause clogging of the contact tip GTAW Very low levels of impurities in the weld may be expected when GTAW is used. Wear of the tungsten electrode and associated arc instability, particularly at higher pressures, are limiting factors for remotely operated GTAW. The Marangoni effect, which is the surface tension and weld pool flow effect on the bead shape and penetration, is stronger for hyperbaric GTAW than GMAW and is affected by the pressure and sulphur (S) and phosphor (P) contents of consumable and base materials. Hence, at high pressures, the control of the weld pool may be less predictable when using GTAW than when using GMAW Other welding processes If welding methods other than GMAW or GTAW are considered for remotely operated hyperbaric welding, the technology of the considered solution should be qualified according to DNVGL-RP-A203. Plasma welding is a proven process of high-pressure capability and operation and may be well suited to mechanization too, but is also subject to tungsten electrode and orifice wear. Several-hundred-volt arc voltages may be required to carry out plasma welding, necessitating the use of special welding power sources Materials Pipe material The following pipe material data shall be assessed: chemical composition; carbon equivalent (weldability) and inclusion shape control (risk of laminations) dimensional tolerances diameter ovality dents/flat spots peaking weld reinforcement, height of longitudinal weld seams for pipe body and pipe ends as relevant lamination control performed when relevant NDT type and extent surface cleanliness and condition for welding. If such data are unavailable or uncertain, they shall be collected as part of a pre-survey Auxiliary component material The material to be used for the hyperbaric weld joint shall be compatible with the pipe material. The material shall be either tubular in accordance with the specification for linepipe in DNVGL-ST-F101 or forged in accordance with DNVGL-ST-F101 Sec.8. Tubular material shall be subject to NDT as required by DNVGL- ST-F101 Sec.7. Forged material shall be subject to NDT as required by DNVGL-ST-F101 Sec Consumables All welding consumables and gases shall be in accordance with DNVGL-ST-F101, Appendix C, and the following additional requirements: Filler wire The filler wire used during production welding shall be from the same batch as that used during the qualification of the hyperbaric welding procedure specification (HWPS). Alternatively, if a new batch of welding consumables is introduced for production welding, batch testing in accordance with DNVGL-ST- F101, Appendix C, C400 is acceptable. Mechanical testing shall be performed in line with that done during Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 86

87 qualification, but limited to tests that are only relevant for documenting weld metal properties if tests specified in DNVGL-ST-F101, Appendix C, C405 are not relevant Tungsten electrodes For GTAW, it shall be possible during production welding to monitor the tungsten electrodes' tip geometry. If required, it shall be possible to replace the electrodes directly or using another qualified method. The effect of wear/blunting of the electrode tip shall be assessed during qualification Shielding gas Shielding shall be provided by the use of an inert gas with qualified purity, including a moisture limit. The gas purity and composition in all containers shall be certified and traceable to the gas storage containers. The gas purity and moisture content shall be verified after purging the gas supply system prior to the start of welding. The moisture content of the shielding gas shall be monitored at/near the torch during the welding operation. This is also applicable for cases where the hyperbaric chamber/habitat has an inert atmosphere. Guidance note: Knowledge of the water content in the gas at the working depth/pressure is essential. The acceptance level for the water content in the shield or welding chamber gas should be specified in terms of the dew point ( C). The dew-point level should be measured directly during pre-qualification and qualification hyperbaric welding and a safety margin applied for the HWPS maximum acceptable torch/chamber gas dew-point temperature. This is of particular concern for minimizing the potential risk of hydrogen assisted cracking (HAC) or hydrogen induced cold cracking (HICC). During the offshore phase, the dew-point level in the torch shielding gas line (GTAW) or welding chamber gas (GMAW) should be measured directly to ensure compliance with the established acceptable level. This measurement should be made at the ambient water depth/pressure. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- The maximum allowable water content in the shield gas used in the actual welding is governed by the moisture content of the gas used during the qualification welding, with a safety margin Welding personnel Personnel involved in the welding operation (the welding coordinator and welding operators) shall be qualified. The welding operation includes the execution of the work as well as the maintenance, preparation, control and monitoring of the key equipment. The key equipment comprises the welding control software, welding control system, habitat, welding equipment, consumable handling system, gas handling system, power system, heating/drying systems, and monitoring and recording systems, both subsea and on the support vessel. The responsible welding coordinator shall be qualified by experience and training in accordance with DNVGL-ST-F101, Appendix C, and shall be present when the welding is being qualified and carried out. Welding operators shall be qualified to ISO by performing a test using the actual equipment under simulated/realistic field conditions and hyperbaric pressure, e.g. in an onshore welding facility. Test piece(s) representing the actual weld configuration (butt weld or fillet weld) and size shall be welded by each welding operator. However, for large pipeline repair fillet welds, one specimen can have multiple welding operators involved provided it is known which areas are welded by which operator and there is enough material for the mechanical testing of each operator's weld region. The test pieces may be weld sections provided the size is sufficient to obtain the test specimens required by DNVGL-ST-F101, Appendix C. For fillet welds, the test pieces shall be subjected to macro examination and non-destructive surface testing. For butt welds, the test pieces shall be subjected to macro examination and volumetric non-destructive testing. The acceptance criteria for the testing shall be that acceptable bead build-up has been obtained and that there are no defects so large that they do not meet the relevant hyperbaric welding procedure specification. The qualification is only valid for the welding equipment used during qualification welding and the actual weld configuration used, and within a variation of ½ to 2 times the load-bearing material thickness. A training programme for all welding operation personnel according to DNVGL-ST-F101, Appendix C shall be established. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 87

88 11.6 Equipment and systems General All welding equipment shall be in accordance with DNVGL-ST-F101, Appendix C. The suitability of all equipment used (including NDT equipment if applicable, ref. DNVGL-ST-F101, Appendix D) shall be documented prior to qualification welding. This may be based on previous experience or an equipment qualification test. The documentation shall include all items listed in the section on the equipment and systems qualification test below. All equipment shall be properly maintained according to a documented procedure Process monitoring and control General monitoring and control requirements are given in [7.10]. The process monitoring and control shall assure a sufficient degree of continuous monitoring to enable confirmation that the welding parameters and related parameters stay within the defined safe parameter range (programmed range plus combined system accuracy). Further, the process monitoring and control shall give warning of deviations outside the essential variables range, i.e. safe parameter limits. The sampling frequency of the monitoring signals shall be sufficient to enable an assessment of the effect of possible short-time parameter deviations. The amount of data recorded may be less than the monitored amount provided the data are processed prior to recording. This processing shall include conclusions on parameter performance. In particular, the effect of short-time parameter deviations shall be concluded with respect to the weld quality, i.e. whether or not the weld is outside specification. Algorithms for such conclusions shall be qualified. All process monitoring shall be based on calibrated feedback signals, not input or demand signals Welding installation procedure for the equipment and system A welding installation procedure (WIP) for the welding equipment and system shall be established and considered qualified when the tests stated below have been successfully completed and the specified requirements for the equipment are fulfilled. The welding installation procedure qualification test can be divided into a: surface test shallow-water test deep-water test simulating site depth. The selection of tests for the respective areas depends on the sensitivity to water and depth of the item tested. Further, some parameters may be simulated by changing other parameters. Therefore, the conservatism in conducting the test at sites, e.g. on the surface, other than the actual site shall be documented. This is most practically done by: defining test procedures that specify the objective of each test, the test method to be used and the acceptance criteria for each test documenting the conservatism of each test when performed under conditions that do not simulate the site depth. Thus, the WIP can be qualified by performing the following tests: WIP - surface Describe and perform tests confirming those capacities and tolerances that can be based on the surface test. WIP - shallow water Describe and perform tests confirming those capacities and tolerances that can be based on the shallowwater test. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 88

89 WIP - site Describe and perform tests confirming those capacities and tolerances that can be based on the deep-water test only. In addition to all the checks in the equipment and systems qualification test given in [11.7], the following shall be tested: installation at maximum inclinations/misalignment alignment/clamping system for the items to be joined by the weld locking to the pipeline cleaning within the best tolerance limit, and cutting and grinding to the best tolerance limit Equipment and systems qualification test An equipment and systems qualification test shall be performed to verify adequate functioning for test welding under actual or simulated field conditions. The purpose of the checks listed below is to give assurance that the equipment provides specified tolerances and boundary conditions to allow test welding to be performed under repeatable and optimum conditions. The test shall be performed according to a documented procedure and as a minimum address the following: Mechanical systems: 1) tightness/leak rates of temporary sealing systems for compliance with specified leak tolerances 2) the total motion envelope of the equipment to be used in the habitat for the actual dimensions of pipe and weldments 3) accuracy control of the welding torch, wire and electrode motions for compliance with the tolerance requirements (also includes control of travel speed) 4) accuracy control of consumable feeding for compliance with the tolerance requirements (wire feed rate) 5) accuracy control of the other robots used to handle cameras, grinders and other tools 6) alert system to notify motions outside the control tolerances. Power system: 7) electrical insulation resistance at high voltage 8) electrical power at maximum consumption 9) hydraulic power piping systems' sealing performance at the maximum test pressure 10) hydraulic power at maximum consumptions 11) power alarm systems for electricity and hydraulics. Gas and moisture: 12) gas supply capacity at the maximum estimated (to be specified) leak rate. 13) gas cleaning capacity at the maximum gas contamination level (to be specified) 14) gas cleanness and moisture monitoring 15) gas cleanness and moisture alarm. Temperature: 16) pre-heating or post-heating capacity to obtain the maximum temperature of the workpiece heat input 17) pre-heating control tolerances (number, positioning, attachment method and calibration of thermocouples or pyrometers) 18) related temperature alarm 19) cooling capacity to obtain maximum cooling 20) cooling control tolerances 21) related temperature alarm. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 89

90 Electricity for welding: 22) voltage, current and pulse frequency at the welding arc for maximum power 23) minimum tolerance limits for these parameters 24) system to notify deviations from the qualified tolerances (alarm system). Control system: 25) execution of the control commands with resulting actions within the qualified tolerances 26) monitoring signals to comply with the accuracy tolerance specification for the relevant habitat environment 27) the visibility and resolution of TV monitors in the relevant habitat environment with respect to atmospheric contaminations, temperature, humidity and motion characteristics 28) signal sampling frequency compliance with qualified sampling rates 29) batch processing of signals enabling identification and correct actions from short-time parameter deviations from the qualified tolerances 30) recording of signals directly or via a pre-processor to verify and document the current weld quality 31) display systems 32) display system ergonomics for compliance with the personnel s capabilities in controlling the weld and inspection of the weld (perform quality assurance of the weld) 33) display system resolution 34) functioning Welding concept base cases qualification routes Qualification of both equipment and welding procedures In the qualification routes that do not include an NDT of the final weld, the absence of defects shall be ensured by a qualification programme such that the level of confidence in the weld integrity is equivalent or higher than if an NDT had been performed. Means to ensure the quality of weldments to compensate for the absence of NDT shall include the relevant welding tests to the tolerance limits, defined process monitoring and control limits as described in this document Butt weld subjected to non-destructive testing Welding differs from DNVGL-ST-F101 in that no personnel are available in the habitat for visual inspection and the preparation/rigging of NDT equipment. Hence, the qualification programme should be as described in DNVGL-ST-F101 and modified in accordance with [11.9]. The following differences related to NDT shall at least be covered: consequence of incorrect rigging of welding equipment surface NDT method capabilities to detect weld surface irregularities consequence of incorrect rigging of NDT equipment Fillet weld subjected to non-destructive testing An inherent feature of fillet welds is the root defect, which is in general impossible to characterize using automated NDT equipment such as automated ultrasonic testing (AUT). NDT of fillet welds to detect other volumetric and planar defects will in some cases be possible depending on the weld size and access for inspection. The consequence of the presence and detectable size of an inherent root defect and other defects shall be evaluated and the probability of detection shall be assessed by pre-qualification testing along the lines stated in [11.9] Fillet weld without non-destructive testing The absence of NDT requires additional measures to be taken to ensure weld integrity, with process control and monitoring as the recommended method. Welding parameters shall be developed as outlined in [11.9] to ensure weld integrity. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 90

91 Guidance note: In principle, a large number of passes is recommended due to the common relationship between the weld pass size and the maximum weld defect size, so that this will reduce the effects of possible defects. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Qualification of welding procedures only For this base case (using the already qualified equipment), an existing HWPS cannot be used for a specific application within the limits and ranges of the variables originally qualified. Hence, a new phwps covering the intended application shall be prepared and qualified as required in Table 11-1 and [11.10] for applications with and without NDT Preliminary hyperbaric welding procedure specification development The phwps shall specify the ranges for all relevant parameters. The effect of the parameters' variation on weld quality, including the accuracy and tolerances of the monitoring equipment in terms of both mechanical properties and defect level, shall be quantified. The activities listed in Table 11-1 shall be included in the development of the phwps relevant for the selected qualification route. The following are further details to be clarified: Design The design shall generally be in accordance with this publication Failure modes All possible failure modes shall be identified and assessed. Fillet welds are susceptible to fatigue failure due to high stress concentration at their root (defect). Hence, a qualification scheme to verify a margin to fatigue failure may be relevant (see [5.6]) Allowable defect size An engineering critical assessment (ECA) shall be performed when required by and in accordance with Appendix A of DNVGL-ST-F101 and for relevant load cases (including cyclic loads). Allowable defect sizes shall be calculated based on a realistic range of fracture toughness values Welding parameters development All welding parameters shall be identified. The effect of the welding parameters' variation on the mechanical properties and defect level shall be established. Parameter sensitivity tests shall be used to determine the limits that still result in acceptable mechanical properties and the absence of flaws exceeding the allowable defect sizes. Confirmation of acceptable parameter limits and ranges shall be based on welding tests where the relevant parameter or set of parameters varies (maximum and/or minimum) sufficiently to be able to operate with a safe margin to failure, as illustrated in Figure Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 91

92 Table 11-1 Overview of phwps development activities Phases Activities 1. Definition phase a) Define the boundary conditions for the weld connection with respect to forces to be transferred, its environment and the welding environment b) Define the welding concept; weld type/geometry 2. Pre-qualification phase (iteration process) Based on the welding tests, variation tolerances shall be established for each welding parameter or group of welding parameters. The following Figure 11-2 illustrates: c) Design of the weld connection, including strength calculation with the effects of gross defects and misalignment d) Identify the possible failure modes and mechanisms and their respective criticalities e) Determine the allowable defect sizes (including ECA) f) Identify and rank welding parameters that may affect weld quality g) Define the preliminary parameter variation range and include this in the pre-qualification welding test programme h) Define the size and boundary conditions for test pieces for qualification testing. Document conservatism i) Perform test welding j) Mechanical and restraint testing, and other relevant testing, as required for the assessment of failure modes and mechanisms k) NDT to locate flaws, followed by systematic sectioning to determine flaw sizes (height and length) l) NDT for confirmation of weld acceptance m) Define the final phwps, including the ranges of all essential parameters that can affect the weld and the margins between operational limits and test qualification limits a) the upper and lower parameter limit as illustrated by an assumed probability distribution (dotted curve) b) upper and lower safe limits given by vertical lines c) safety margins for: the material and test method (dark grey) control and monitoring tolerances (light grey) the set point variation (white) d) results of the welding test; diamonds illustrate test results, acceptable (white) or unacceptable (black). Figure 11-2 Parameter variation and tolerances Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 92

93 The variation tolerances shall include: safety margins to cover all uncertainties the inaccuracies and tolerances of the monitoring and control equipment. The resulting parameter window shall be sufficient to allow each parameter or group of parameters to be set within the ranges required. There will thus be two windows for parameter variations: one ultimate variation tolerance window, and a smaller variation tolerance window for the planned weld operation Arc stops Arc stops shall be simulated and any resulting defect size determined and evaluated for possible removal Small-scale tests vs. full-scale tests The size of the pressure chambers used for qualification welding as well as practicalities may limit the size and fixture of test pieces. Additional means to control weld temperature and cooling rates as well as restraint conditions may be necessary. Hence, it shall be demonstrated/documented that the effect of a possible reduction in the size of the test pieces used for qualification welding will represent the actual or conservative conditions with respect to: restraint cooling Cooling rate The effect on the weld cooling rate of the external and internal pipeline environment (the pressurized circulated atmosphere, water on the outside and on the inside, or possibly other fluids inside the pipeline) shall be taken into account. The cooling rate identified by numerical analysis or measurements shall be simulated during pre-qualification welding. If weld properties are significantly affected by the cooling, the amount of cooling shall also be conservatively applied during qualification Welding atmosphere The qualification of the welding atmosphere shall reflect the actual production atmosphere. Hydrogen pick-up and possible weld contamination from residual oxygen or fume gases for the welding parameters used shall be assessed based on testing at conservative humidity conditions in the shielding gas at the given pressure, see [11.4.3] Restraint The effect of residual stresses caused by weld solidification and thermal shrinkage shall be taken into account. Possible adverse effects are caused by: weld/bead dimensions and shape material properties content of diffusible hydrogen rate of cooling Weld cracking The sensitivity to weld cracking shall be assessed by tests such as the Tekken-type self-restraint test according to ISO or modified restraint tests with documented conservatism. Weld cracking of the fillet welds shall also be assessed by way of a mock-up test welded under hyperbaric conditions and with an internal cooling environment equivalent to actual conditions (e.g. provided by the internal water flow rate and temperature). The severity can be adjusted by changing the root gap in accordance with the root gap tolerance of the HWPS. The number of macro sections taken from the fillet welds shall be sufficient for: statistically valid sampling Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 93

94 comparison of detected root flaws with the ECA derived from BS7910 Option 2 or 3D non-linear analyses Systematic sectioning The welds performed in order to determine tolerances for parameters identified as critical for the weld defect level shall be subject to destructive examinations by systematic sectioning. The maximum defect size shall be determined by the systematic (macro) sectioning of test welds. Systematic sectioning is also useful to verify any applicable NDT systems. The systematic macro sectioning shall be based on volumetric NDT to determine the indications that will be subject to sectioning. The systematic macro sectioning shall determine the type, height and length of the indications from the volumetric NDT Maximum defect size The maximum defect size determined by systematic sectioning shall be compared to allowable defect sizes obtained from the ECA. The extent of the systematic macro sectioning shall be sufficient to determine that the probability of defects exceeding the critical defect size established by the ECA is 90% at a 95% confidence level for the established parameter range. A possible NDT method intended to replace or reduce the amount of macro sections shall be qualified to obtain an equivalent confidence level Repeatability When acceptable parameter ranges are achieved in welding trials, a series of test welds shall be carried out with the same parameters and mechanically tested to verify the repeatability and consistency of the test results. The number of test welds is governed by the variation in the obtained results and the strategy for defining safety margins Monitoring and control The parameter variation used in the pre-qualification forms the basis for specifying the monitoring and control to be applied to the actual operation. Inaccuracy tolerances in monitoring and control shall form part of the input for establishing the safe margins, ref. Figure Preliminary welding procedure specification A phwps based on the results of the development work shall be prepared in accordance with DNVGL-ST- F101, Appendix C. The phwps shall include tables for each weld operation, with parameter windows for the essential welding parameters as described above Welding procedure qualification When a phwps has been defined, based either on the development scheme outlined in [11.9] or on a previous HWPS, qualification shall proceed as outlined in Table Qualification welding of welding procedures Qualification welding shall be performed in accordance with DNVGL-ST-F101, Appendix C and the defined phwps. The tests shall be carried out at the upper and/or lower safe range of parameter variation determined during pre-qualification, see the range limited by the vertical lines in Figure Test welding Test welds, including relevant results from phwps development, shall confirm that acceptable results are obtained when the critical parameters are varied within the established safe range. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 94

95 Table 11-2 Overview of HWPS qualification activities a. Define the final phwps, including ranges for essential parameters that can affect the weld and the margins between operational limits and test qualification limits b. Test welding in a pressure chamber with relevant environmental conditions, equipment and specimen sizes/ fixtures, at parameter limits with margins c. NDT of all test samples d. Mechanical testing - an example of specimen location/orientation is given in Figure 11-3 e. Systematic sectioning f. Issue of HWPQR g. Issue of qualified HWPS The mechanical testing of fillet welds as listed in Table 11-3 shall be performed in accordance with DNVGL- ST-F101 Appendix B. The size of test weld samples shall be equivalent to the fitting's actual fillet weld assembly. If the repair fitting fillet weld is too small in size to fabricate mechanical test coupons and it is not possible to increase the weld size and still achieve representative material properties for weld qualification testing, the weld qualification may be based on coupons made from butt weld joints formed from the same materials and equivalent welding procedures. Figure 11-3 Example of specimen location of fillet welds - for mechanical testing Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 95

96 Acceptance criteria As a minimum, the tests given in Table 11-3 shall be performed. Welds shall meet the acceptance criteria for strength and toughness required by the weld design and the defect level determined during the phwps development, with a safety level consistent with the requirements in DNVGL-ST-F101 Sec.2. The maximum hardness shall be in accordance DNVGL-ST-F101, Appendix C for the relevant material type, unless otherwise qualified. Table 11-3 Type and number of tests for the qualification of welding procedures Required tests No. of tests for butt welds No. of tests for fillet welds 1) Transverse weld tensile - Longitudinal all weld tensile 2 Bending - Charpy V-notch impact testing sets Macro sections and hardness As per DNVGL-ST-F101 Appendix C See note 1) 3 Notes The specimens shall be taken from each test weld Specimens shall be taken from each test. The number of specimens shall be determined based on the weld size and geometry. As a minimum, the test shall include the weld metal and fusion lines (FL) and FL+2 mm and FL+5 mm in both base materials. Additional samples shall always be taken when a sufficient weld cross-sectional area allows for such testing. The specimens shall be taken from the start, end and middle of each test weld. The macro sections shall be documented by photographs (magnification to give sufficient resolution). Microstructure See note 1) Examination of corrosion resistant alloys only Fracture toughness test See note 1) Non-destructive testing See note 1) Systematic sectioning See note 1) See note 1) General note: The specimens shall be taken from each test. As a minimum, the test shall include the weld metal and fusion lines between buttering and pipe material. For CTOD testing of the fillet weld, a simulated 90 o butt weld with buttering on one side of the weld groove shall be prepared. The CTOD values shall be used to establish ECA-based flaw acceptance criteria for the fillet welds. To be determined based on the weld size and geometry. If possible, volumetric testing shall be performed. In the case of no NDT in the actual operation. The number is to be established based on the principles stated above. 1) For fillet welds that are too small to be sampled for any of the tests given in this table, measures shall be taken to enable testing. Such measures may include the welding of a wide-angle butt weld (i.e. 90 o butt weld with buttering on one side of the weld preparation by first building up the weld groove on one side of the bevel to achieve a heat transfer more representative of fillet welding) with the relevant welding parameters or increased size of the fillet weld (requires any tempering effect to be considered when specimens are sampled) Validity of welding procedures A qualified welding procedure remains valid as long as all variables are kept within the qualified range (e.g. a depth/pressure range) and are established essential variables. If one or more variations outside the qualified range occur, the welding procedure shall be considered invalid and be re-specified in a new phwps and qualified. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 96

97 11.11 Production welding requirements General requirements All production welding shall be performed according to a qualified hyperbaric welding procedure specification (HWPS) and accepted welding consumables handling procedure HWPS confirmation test A test weld according to the qualified HWPS shall be performed on board the support vessel at the site, shortly prior to the production welding, using the actual equipment to be used for the work. The test pieces for the confirmation test shall be of a practicable shape that challenges the welding system in a way similar to the actually intended welding. The welding control and monitoring system shall perform as if qualified, but taking into account the topside pressure Welding equipment The welding equipment used for the work shall be similar in function to the equipment used during welding procedure qualification and equipment qualification testing. The type, cross-section area and length of electrical cables are defined as part of the welding equipment, as well as the power source, welding control system and software make and model, tungsten electrode and contact tube (as applicable) Habitat environment All gas supply lines with connections and cavities/chambers shall be leak tested and flushed by the shielding gas intended for the welding prior to use. The last part of the flushing shall include measurement of purity and moisture. The gas environment shall be monitored at acceptable intervals during the welding to ensure that all weld parameters are within the qualified envelopes. If it has been identified during pre-qualification tests that particular fume gases are of concern to the weld quality, the monitoring of these gases needs to be included to ensure a safe welding environment Material check The following checks of the sleeve/pup-piece shall as a minimum be carried out before deployment to the site depth: Dimensions (diameter by gauge, wall thickness and length) measured at four (4) equidistant points on the pipe circumference. Bevel details (if applicable). The root face thickness shall be accurately measured for each clock hour position around the pipe circumference. Laminations on the joint faces, using ultrasonic testing at a minimum distance of 100mm from the edges and magnetic particle/dye penetrant testing of the pipe edges/bevels when relevant Filler wire The filler wire used during production welding shall be from the same batch as that used during qualification of the HWPS. In general, it is recommended that the same batch of consumables is used throughout the process ranging from qualification to the full-scale mock-up test of the repair and the offshore campaign. Alternatively, if a new batch of welding consumables is introduced for production welding, batch testing in accordance with DNVGL-ST-F101, Appendix C, C400 is acceptable. Mechanical testing shall be performed in line with that which was done during qualification, but limited to only tests that are relevant for documenting weld metal properties if tests specified in DNVGL-ST-F101, Appendix C, C405 are not relevant. Filler wire which shows any sign of damage or deterioration, or cannot be properly traced and identified, shall be discarded Pipe surface/bevel preparation, alignment and lamination check The pipe surface shall be checked before welding for tears, scale, rust, paint, grease, moisture or other foreign matter on the fusion faces that may adversely affect the weld quality to an extent not included in the HWPS qualification. The pipe dimensions, surface/end cut, bevel dimensions, root gap around the circumference and alignment shall meet the dimensional tolerance and surface appearance specification (see [7.1], [ ] and [11.6.3] WIP-site) to be established as part of the qualification scheme. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 97

98 The pipe end material properties can be affected by the cutting method. Acceptable properties will normally be obtained e.g. from cutting by mechanical means, such as by diamond wire or water-jetting/grit and possible additional grinding. A lamination check using ultrasonic testing in accordance with DNVGL-ST-F101, Appendix D shall cover 100% of the area to be welded and in addition 100 mm upstream and downstream of that area. Acceptance criteria for possible lamination shall be established as a part of the qualification scheme. Compliance with the specifications shall be verified by mechanical and/or ultrasonic means prior to the relevant non-reversible operations, e.g. the mobilization, the cutting of the pipe and the welding operation Cleaning of the weld Upon completion of each welding pass, the weld shall be inspected (camera) and cleaned if found necessary Inspection during welding Inspection during welding shall be executed from the surface weld control room and/or an inspection room. Inspection shall as a minimum include the following: a) A camera in the welding habitat. Inspection by the welding operator or the welding coordinator and a video recording, all continuous to the extent qualified. b) Monitoring, recording and display of the habitat's environmental parameters (temperature, humidity, pressure, atmosphere composition). Alarm for critical parameters to be included. c) Photo/video, recording and display of pass identifications used for welding. d) Monitoring, recording and display of welding current, arc voltage, filler wire speed, welding speed and shielding gas flow. Alarm for critical parameters to be included. Weld starts and stops shall be performed in compliance with the weld qualification tests. Guidance note: A normal procedure is to start and stop the weld at places so that these locations do not coincide in adjacent passes. At least 4 passes should be made before the same start or stop position is used. The passes should be deposited in a balanced sequence around the pipe circumference in order to minimize residual strain and distortion. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Inspection and testing after welding After completion, the weld shall be subject to NDT to the extent that this is included in the qualification. This shall include a visual examination. The visual examination shall include a 100% visual camera inspection by the welding operator or welding coordinator Out-of-specification weld passes Weld passes that, based on the inspection during welding or the inspection and testing after welding, are to be considered as out of specification shall be removed Hydrostatic testing System pressure testing (hydrostatic testing) and a leakage test of the repaired/modified pipeline section shall comply with the requirements stated in DNVGL-ST-F Interruption of welding If the welding is interrupted, e.g. due to equipment failure or weather limitations, the appropriate course of action for all foreseeable extents of welding completion and equipment status shall be described in a contingency procedure such that the integrity of the pipeline is ensured. Consequently, this contingency shall be a part of the qualification Repair welding Repair welding shall be qualified. Local grinding due to local excessive spatter or poor bead shape may be performed for the current/last welding pass. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 98

99 11.12 Mobilization Due to the complexity involved in the use of the remote hyperbaric welding system, the likely sporadic use of the system and possible major consequences of welding interruption or failure, specific training should be performed as close in time as practically possible ahead of the planned repair welding. In order to maintain their qualified status, the equipment, systems and welding consumables shall be: stored under conditions according to a qualified procedure tested at intervals according to a qualified procedure, including calibration and recording. Personnel qualifications may be maintained by regular training, including using the equipment Documentation General documentation requirements are given in Sec.14. These are further detailed in DNVGL-ST-F101 Sec.12. The as-built material documentation shall include the following: welding procedures (WPS) and welding procedure qualification records (WPQR) welding and NDT personnel qualification records NDT and visual inspection reports for pup-pieces/sleeves/lamination control and hyperbaric welds (if applicable) material certificates for base materials and welding consumables records of all essential welding parameters. See [ ], subsection Inspection during welding. Where no NDT is applied on the hyperbaric weld, 100% documentation of all relevant welding parameters and weld pass positioning shall be included in the as-built documentation records of habitat/chamber atmosphere and shield gas purity. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 99

100 SECTION 12 TESTING 12.1 General The design of pipeline repair fittings applies a risk-based approach according to which the qualification process shall establish a safe margin to failure for all failure modes of concern. For some failure modes (or a combination of such), an adequate safety margin can only be documented by testing. Other failure modes can be assessed through analytical and numerical models which have been validated by testing. Testing of either parts of the fitting design or the whole fitting assembly may be required by the design standards used. Typical mandatory tests are pressure tests, such as a hydrostatic strength test of the fitting (whole assembly), mill test of new pipeline joints, system pressure test of the pipeline spool or fire tests for topside applications, etc. In addition, DNV GL certification/approval schemes require specific tests such as fitting material mechanical tests, pressure testing including external loads, cyclic load tests, etc Pressure testing Pipeline repair spool and pipeline system Pipeline repair spools made up of new or old spare pipe joints shall comply with the requirements given in DNVGL-ST-F101. The repair spool shall consist of pipe joints where each joint has documented strength confirmed by the mill pressure test, see [14.1.6]. The pipeline repair spool shall be subjected to a system pressure test after installation according to DNVGL-ST-F101 Section 5 B200. Guidance note 1: The pressure testing of the pipeline after repair (i.e. pipeline system pressure test) is to verify the repair location and not the entire pipeline. The test pressure is therefore determined in accordance with DNVGL-ST-F101 Sec. 5 B202, i.e. to exceed the local incidental pressure at the repair location by a factor of typically For pipeline section replacement, this test will also verify that no gross error exists in the new pipeline segment installed. Waiving of the pipeline system test after repair should be evaluated on a case-bycase basis and follow the principles outlined in DNVGL-ST-F101 Table 5-1. Pipeline isolation plugs may be used to limit the part of the pipeline exposed to pressure testing. DNVGL-ST-F101 specifies that pipeline system pressure testing after repair may be required: when the original mill pressure test or system pressure test does not satisfy the requirements of the design standard in the case of e.g. a new design pressure (i.e. re-qualification) when a significant part of the pipeline has not been system pressure tested, e.g. new pipeline sections as part of a modification or repair campaign when repair operations expose parts of the pipeline or a new replacement section to loads with possibly large uncertainties (e.g. above water tie-in repairs, or the replacement of long pipeline sections) as an alternative in order to document the current condition of the pipeline system if general inspection techniques cannot be utilized to inspect the pipeline's internal or external condition. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 2: When establishing the test pressure level after installation of a repair fitting, all potential failure mechanisms to the parts of the pipeline system exposed to the test pressure should be established, such as capacity of bolted connections, HISC and as-is condition of the pipeline system. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- In general, pipeline repair by welding or repair fittings shall provide the safety level compliant with the DNVGL-ST-F101 acceptance criteria without requiring a pipeline system pressure test afterwards. To waive the requirement of a hydrostatic pressure test of welded connections after the repair, the hyperbaric weld connections need to comply with the golden weld requirements in DNVGL-ST-F101, i.e. critical welds such as tie-in welds that will not be subject to pressure testing. Repair fittings that deliver leak barriers shall provide double sealing with test ports and a seal test shall be performed after installation to document acceptable installation. A weld qualification programme according to DNVGL-RP-A203, documenting the acceptable repeatability of each governing parameter and representative mock-up test(s) in an equivalent hyperbaric environment followed by macro examinations of the weld, can be used to document a golden weld condition. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 100

101 Currently, there is no option to waive the pressure testing of a mechanical connection. The pressure test requirements for a pipeline repair fitting are detailed in [12.2.2] Pipeline repair fittings The safety level of the repair fitting assembly shall be documented to be at least the same as that of the connected pipeline by an appropriate combination of design requirements and tests, derived based on a risk assessment (ref. Sec.2). The pressure test requirements for the repair fitting are summarized in Table Table 12-1 Pressure test requirements for a repair fitting Phase Qualification (type test) Type of test Strength test ([2.1], Sec.6 and App.C) Test load level Designed based on PV standard ([2.1]) See PV standard, typically Note 2 =1.43 ( ) Designed based on LRFD ([2.3.2]). = γ, ( ) =, =, FAT Installation of fitting Pipeline re-commissioning Seal test ([6.7] and App.C) Locking capacity ([6.5] and App.C) Note 1 Note 2 Cyclic loading Note 5 Hydrostatic pressure test Seal test ([6.7] and App.C) 1. Seal test, or alternatively 2. System pressure test Note 1 Note 4 (isolation plugs) Note 1 Same as strength test above =1.43 ( ) = γ, ( ) Note 3 = γ, 0.96 ( 2 ) 3 Note 1: The seal test pressure level depends on potential failure mechanisms related to the installation and activation operation and the pressure level that is required to approve the performance of the seal(s) after installation with respect to the potential leak failure mechanisms. Typically, the back-seal test pressure level for: seal welds is equal to the local incidental pressure: = and for polymer, graphite and metal seals is equal to 1.1 to 1.5 times the local incidental design pressure, depending on the seal robustness level qualified from the type tests and the potential failure mechanisms related to the installation (pre-tension level, surface roughness and local unevenness, damage to the seal and seal-surface during installation, and seal robustness documented during the qualification phase). A seal test through a test port between the primary and secondary seal will in general separate the pipe and fitting, contributing to a reduction in the seal contact pressure, and is hence more conservative than pressurization inside the pipe section. Guidance note: The seal test pressure may be applied to an annulus external to the pipe, and could therefore be lower than the pipeline test pressures. This is because the internal pressure normally improves the sealing capability due to the pipe expansion compared to external pressure, which compresses the pipe. However, an external differential water pressure due to the depressurization of a gas pipeline will have the opposite effect and must also be considered. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Note 2: External load and moment according to the applied pressure vessel standard, or as specified in this table for the LRFD method. The minimum holding times for pressure leak tests are given in Appendix [C.6]. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 101

102 The acceptance criteria are as follows: No visible leak. No pressure drop within minimum specified holding times given in Appendix [C.6]. No unacceptable permanent deformation. Indentation marks from a body gripping onto the pipe wall shall not reveal any slippage or shear failure. Note 3: The test pressure level,, is limited by a maximum of 96% of the pipe SMYS utilization if the fitting is tested on a pipe section equivalent to the pipe section to be repaired. Note 4: See [7.10.2]; the isolation pressure provided by isolation plugs shall be monitored for a specified minimum period of time (typically four hours depending on the volume, type of fluid and temperature) to document the acceptable performance of the barrier(s) in the set condition. Note 5: For repair fitting assemblies that are exposed to cyclic loading during one or more of the testing, installation or in-service phases, potential, related failure mechanisms shall be assessed. Typical sources of cyclic loading are: Variations in pipeline functional loads (i.e. pressure and temperature), causing variations in pipe wall axial stresses and/or global bending stresses in the pipe sections with curved parts, and introducing external bending moments to tee branches. Clamp bolt make-up loads during functional testing that lead to accumulated fatigue damage in the clamp and bolt assembly. Typical failure mechanisms caused by cyclic loading are: local yielding of the in-seal contact area when the assembly is exposed to the design bending moment, causing reduced seal contact pressure when exposed to a reversed bending moment accumulation of slippage between the pipe wall and fitting locking segments fatigue development of viscous-elastic deformation of the polymer seal extrusion of polymer and graphite seals crushing of grout filling material ratcheting. Repair fitting assemblies where potential failure mechanisms caused by cyclic loading are identified shall be tested for cyclic loading to document acceptable robustness against an identified failure mechanism Qualification testing repair fitting The typical tests recommended for the qualification of pipeline repair fittings are based on the qualification philosophy described in [2.4.2]. The recommended tests are basic tests, type/qualification tests, FAT and final tests. An overview of these tests is summarized in Table 12-2 and further guidance is detailed in App.C. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 102

103 Table 12-2 Overview of tests according to the DNVGL-RP-F113 test philosophy, detailed in App.C Basic tests (design and manufacturing phase) Type/qualification tests 2) (after assembly) FAT Final tests (after installation) Material/mechanical tests 3) Installation and lifting tests Activation test Sealing test Combined effects: Testing of seal design envelope: Activation tests Verification of successful pipeline grip by load Pipeline system test - Extrusion gap tests soft seals testing or measurements/ (see above) calculation - Metal seal tests Testing of gripping design envelope: - Friction factor tests - Galling tests - Pipeline gripping tests - Corrosion tests (e.g. sour service) Graphite seals Polymer seal test according to NORSOK M-710 or ISO Material and material combination related to subsea use Strength tests/leakage tests: Pressure test - Pressure test - Fitting tensile test, no pressure - Fitting compression test, no pressure - Torque test - Bending test - Fatigue test - Temperature test (short/long term) - External pressure test - Combined loads test Testing of seal assembly 1) - Full-scale tests - Small-scale tests Testing of gripping assembly 1) - Full-scale tests - Small-scale tests Seal test Deactivation test (if required) Integration and subsea tests - Installation and lifting tests - Activation test - Pressure test - Contingency retrieval test 1) May also be done as part of the basic testing in order to validate the calculation model's FEA models. 2) For the design of one type of fitting, it will likely be more practical to combine the Type testing and FAT. 3) The acceptance criteria and extent of testing are detailed in DNVGL-ST-F101 Guidance note 1: The amount of testing will depend on the relevant failure modes and the confidence level in the analytical model(s) for documenting compliance with the safety philosophy as detailed in [2.1]. An example of a fitting qualification process showing the different types of tests is outlined in the figure below. For novel/unproven technology, the amount of testing may be more extensive and DNVGL- RP-A203 is recommended to be used as guidance. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 103

104 ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Guidance note 2: The analytical model for calculating the pipe wall local response to the gripping and sealing activation load should be validated by strain gauge measurements as part of the Type and/or FAT test programme. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 104

105 SECTION 13 INTEGRITY MANAGEMENT OF REPAIR INSTALLATION The focus of life cycle management is to continuously monitor the integrity of the pipeline system, including the repair installation, to ensure the required safety level and functionality throughout the required design life. DNVGL-RP-F116 gives guidelines for the integrity management of the pipeline system. The life cycle management of the repair fittings should be based on a structured risk assessment that considers degradation mechanisms, event damage scenarios and potential future threats. Typical items to be assessed in such a risk assessment are: degradation mechanisms of non-metallic components: ageing chemical composition of the pipeline medium and inhibitors movement of static seals wear of seal and seal contact surfaces metal loss from corrosion and erosion of metallic components, relevant for both pipeline and components change in operational conditions: e.g. reversed flow, change in temperature, increased number of shut-downs deposits of/contamination from particles incidents change in 3 rd party interference design: e.g. increased size and weight of trawl equipment. The life cycle management of pipeline repair fittings, tools, equipment and spares (e.g. spools and pipeline sections) covers both the contingency storage phase and the in-service phase after repair. The life cycle management documentation typically includes: technology qualification reports on the fitting, fitting design premises and reports, FMECA report, fitting qualification test and FAT test reports, material certificates, lifting and installation procedures welding procedures for the design/installation procedures for the re-assessment of fittings and spare parts preservation and storage condition requirements for contingency parts (i.e. tools, equipment, spools and repair fittings) procedures for commissioning the contingency parts (e.g. inspection, NDT, functional and pressure testing) procedures for in-service integrity management activities or maintenance requirements for the repair assembly. The output from the risk assessment will be input to the governing inspection, monitoring and test requirements for the pipeline system to ensure that the repair is fit for its designated service throughout the design life. Documentation requirements are given in [14.1.9]. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 105

106 SECTION 14 DOCUMENTATION AND QUALITY ASSURANCE 14.1 Documentation General General documentation principles are presented in DNVGL-ST-F101 and this RP. The documentation should be available for assessment. Sec.11 describes details related to hyperbaric welds. The specification of pipeline subsea repair fittings shall include a list of all limiting parameters and relevant parameter combinations for installation and operation. Furthermore, it shall describe the minimum requirements (main specifications) for tools which are required to enable coupling installation within safe limits. Such documentation requirements are detailed below General documentation 1) description, including the background, objectives and overall description of solution 2) installation principles, including hyperbaric welding when relevant 3) main specifications and limitations 4) arrangement drawing with position numbers Qualification Documentation shall include: 1) calculations and related dimensional drawings and materials, as well as tests related to the design and installation principles 2) identification of possible failure modes and documentation of a reasonable safety margin against these failures 3) interpolation/extrapolation methods to be applied to the actual designs 4) materials specification 5) manufacturing and quality control principles, including main principles of FAT procedures 6) limitations, assumptions and requirements for installation tools and installation procedures Design Documentation shall include: 1) specifications and limitations 2) an assembly drawing with parts list and material specifications 3) detailed dimensional drawings 4) identification of materials 5) design analysis 6) outline procedures intended for tests with the objective of documenting design features 7) general material selection requirements. A material selection evaluation is included in the documentation, see [6.3.2] Manufacturing Documentation shall include: 1) material certificates 2) manufacturing records for bolt pre-tension, welding procedures, including qualification records, welding operator qualifications and NDT personnel qualification 3) dimensional measurement report on key dimensions 4) test reports 5) unique identification (for traceability of fittings and their main components) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 106

107 6) manufacturer's design, fabrication and installation resume (DFI). The DFI resume shall identify possible inspection and maintenance requirements and provide guidance on possible repair/retrieval 7) manufacturer's Certificate of Conformity, with specified criteria Testing Documentation shall include: Pipes The pipe manufacturing method shall be specified. The detailed pipe dimensions shall be measured and documented by a dimensional sketch, including information on: 1) straightness in two planes (90 degrees apart) within the attachment length of the fitting to the pipeline. The straightness shall be recorded as deviations from the straight line at maximum intervals of 1/10 the coupling length 2) the accurate diameters shall be measured at sections: at each end of the fitting's attachment to the pipeline, where seals interact at the middle of the attachment to the pipeline at maximum and minimum straightness deviation 3) each cross-section for diameter measurements shall be measured at four diameter positions equally spaced around the circumference 4) local imperfections (welds, undercuts, artificial imperfections). The sketch shall show depth (height), length, shape and curvatures. Photographs and plastic replica can be used to supplement the sketch 5) the wall thickness shall be measured in eight places equally spaced around the circumference at the attachment to the pipe 6) end-cut evenness or chamfer. The pipes shall be marked to identify the measurement positions and show the intended axial and angular location of the coupling. Test certificates valid for the particular pipe shall document actual material properties. Hardness (Brinell or Vickers) shall refer to measurements in weld areas. Details of pressure test levels applied in the mill test shall be properly documented in the test certificates and comply with DNVGL-ST-F101 Sec.7 E100. Fitting Drawings with dimensional tolerances shall be available. The actual dimensions of the critical parts, such as the minimum internal diameter, shall be recorded with an accuracy of at least ±0.1mm measured at, or transformed to, a 20 C ambient temperature. Material test certificates shall document the actual material properties of both metals and seals Storage and transportation Documentation shall include: Pipes The storage of pipe joints shall be properly documented. For temporary conditions, the requirements in DNVGL-ST-F101 Sec.8 F400 shall apply. In addition to the temporary storage requirements, permanent storage shall also consider: documentation of pipe inspections and checking for loose material, debris and other contamination documentation of pipe cleaning, both internally and at pipe ends the pipeline ends shall be protected against ingress of dust, water or any other material after cleaning and prior to being stored the individual pipes of pipe strings shall be marked in accordance with the established pipe tracking system using a suitable marine paint. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 107

108 Fittings The storage of fittings shall be properly documented for the intended storage period. Documentation shall as a minimum cover the following: storage/preservation environment critical components, e.g. storage of polymer materials maintenance/testing frequencies. Guidance note: For repair fittings and equipment that are stored for contingency, procedures should be prepared to ensure that the equipment is fit for service when required, covering preservation and maintenance within specified shelf and design life, and procedures for testing before use. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Installation Documentation shall include: 1) lay-out drawing of the installation 2) dimensions, including tolerances and material identification of the pipes to be connected 3) pipe and fitting manipulation documentation for compliance with both pipeline and fitting design requirements 4) inspection records of pipe end-cut geometry, pipe surface roughness and cleanliness and alignments of pipe ends 5) confirmation of the fitting make-up within prescribed limitations and the quality of any hyperbaric welds 6) leak test report with a process & instrument diagram (P&ID) for the leak test system 7) final inspection documentation 8) installation contractor s DFI resume 9) for hyberbaric welded items, see supplements in [11.13] Life cycle management Documentation shall include: 1) design documents 2) material certificates 3) test reports 4) as-built drawings 5) installation documents 6) pipeline intervention documentation 7) pipeline surface preparation documentation 8) seabed intervention documentation 9) cathodic protection and coating design reports 10) vendor recommendations for maintenance and testing activities Qualification checklist Methods used for qualification depend on the type of fitting. App.J presents a checklist for use in the qualification. The list is split into three main parts: A for input parameters, B for qualification parameters and C for documentation. The qualification part, the B list, is furthermore split into two main columns, one for analysis and theory and another for tests. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 108

109 14.3 Quality assurance The manufacturer and installation contractor shall: perform design, manufacturing and installation work according to generally recognized quality assurance procedures follow recognized standards/acceptance criteria. Guidance note: A method for documenting the quality of the coupling is described in: DNV-OSS-301 Certification and verification of submarine pipeline systems ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e Traceability The installation records (documentation), manufacturing records and qualification documentation of each installation shall be traceable. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 109

110 APPENDIX A CODE BREAKS AND DESIGN FACTORS A.1 Introduction The design loads and associated design (load and resistance) factors applicable for the different types of repair fittings depend on the load interaction between the pipeline and fitting, considering the complete operational life of the repair. Guidelines on code breaks and the load and resistance factors applicable for the different parts of the considered pipeline repair fittings are given in the following sub-sections. The guidelines given in this appendix are based on the assumption that the repair fitting is designed and pressure tested according to a recognized pressure vessel (PV) standard. Alternatively, the repair fitting may be designed and tested based on the LRFD methodology, as described in [2.1]. A.2 Reinforcement clamp Typically, the reinforcement clamp provides local reinforcement of the pipe wall to: mitigate fatigue loading at local pipe wall damage, such as dents, gouges, i.e. cyclic local bending of the pipe wall when exposed to variations in the functional loads, in order to stop fatigue crack growth. Gouges and/or cracks (e.g. initiated by excessive tensile strains) at the dent may reduce the fatigue strength significantly reinforce the pipe wall at local metal loss locations, and/or mitigate external corrosion in the splash zone and above water provide local temporary reinforcement of the pipe wall at the pipeline isolation set location. Schematic guidelines on code breaks and design factors are given in the figure and table below. Figure A-1 Reinforcement clamps Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 110

111 Table A-1 Example of code break and load and resistance factors Part Standard Loads Load factors Resistance factor Clamp body PV standard Internal design pressure (P li ) PV standard Allowable stress in applied Internal test pressure (P FAT ) PV standard design standard for relevant limit states External pressure (P e ) PV standard External loads (N, M) 1) ref. Table 6-3 Bolting PV standard Loads as specified for clamp body above Pre-tension Filler RP-F113 Creep or expansion from curing, and interaction loads between pipe and clamp. Pipe wall RP-F113 Internal design pressure (P li ) Internal test pressure (P lt ) External pressure (P e ) External loads (N, M) 1) PV standard To be established through qualification Allowable stress in applied design standard for relevant limit states To be established through qualification ref. Table 6-3 ref. Table 6-4,,,, Dent size ST-F101 Section 5 D1300 2) Fatigue ST-F101 Loads from: - variation in operational pressure - variation in external loads (N, M) Recommended tests: Filler coupon tests Creep/expansion tests Mock-up installation test (n x Δσ)*DFF 3) ref. Table 6-4,, Cube test (according to e.g. EN or equivalent) due to higher stiffness of clamp body) mechanical properties Testing of the filler material's volumetric properties Qualification of installation and grouting procedures and training of personnel. 1) Pipe wall force (N) at clamp. The stiffness transition between the pipe and clamp section may reduce the pipe wall's local buckling resistance and must be checked when relevant. 2) General criterion for the allowable dent size without repair (i.e. dents without gouges or cracks); the impact frequency for dents to be repaired is assumed to be low: allowable usage factor η = 0.7, i.e. the acceptable plastic dent depth versus pipe steel outer diameter is Hp/D 0.05 η = Larger dents may be accepted, provided the following is documented: - functional requirements (flow, pigability) - collapse resistance (DNVGL-ST-F101 Sec.5) - local buckling (DNVGL-ST-F101 Sec.5) - fatigue (DNVGL-ST-F101 Sec.5 and DNVGL-RP-C203) 3) Design factor fatigue (DFF)= 6;10 (Medium & High) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 111

112 A.3 Leak clamp The leak clamp is installed to seal off an identified potential or existing local leak. It is not designed to transfer any axial pipe wall forces or bending moments; and is used on the condition that the local pipe section resists the axial and bending moment design loads. Leak clamps are used for either temporary repair until the permanent repair solution is installed, or as a permanent repair. Leak clamps should be designed with double sealing to allow a back-seal test after installation. Schematic guidelines on code breaks and design factors are given in the figure and table below. Pressurized volumes are shaded. Figure A-2 Leak clamps Table A-2 Example of code break and load and resistance factors Part Standard Loads Load factors Resistance factors Clamp body PV standard Internal design pressure ( ) PV standard Allowable stress from Internal test pressure (P FAT ) 3) applied design standard for relevant External pressure ( ) limit states. Seal contact pressure ( ) 2) Make-up leak test ( ) Bolting PV standard Loads as specified for clamp body above Pre-tension Pipe wall RP-F113 Internal design pressure ( ) Internal test pressure ( ) External pressure ( ) Seal contact pressure ( ) 1) 2) Make-up leak test ( ) PV standard Allowable stress from applied design standard for relevant limit states. ref. Table 6-3 ref. Table 6-4,,,, Recommended tests Seal test ( ) RP-F113 Seal test pressure, see Table 12-1, Note 1. Test duration and acceptance criteria are given in Appendix [C.6]. FAT PV code FAT pressure; duration and acceptance criteria are given in the PV standard 1) Lower bound for seal functionality, upper bound for structural integrity, as defined in [6.4.2]. 2) Seal contact pressure includes the effects of make-up, differential pressure, test pressure, swelling and thermal expansion. Load and resistance factors to be established through qualification and testing. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 112

113 A.4 Structural clamp The structural clamp is installed to provide reinforcement and transfer some or all of the local pipeline axial loads and bending moments in case the identified damage should develop through the cross-section during the pipeline's service life. Structural clamps are in general combined with an integrated leak clamp, but can also be designed without a seal to stop leaks. The clamp's seal system should be designed with double sealing to allow a back-seal test after installation. Schematic guidelines on code breaks and design factors are given in the figure and table below. Figure A-3 Structural clamps Table A-3 Example of code break and load and resistance factors Part Standard Loads Load factors Resistance factors Clamp body PV standard Internal design pressure ( ) Internal test pressure (P FAT ) Allowable stress External pressure ( ) from applied design Loads from activation of gripping 4) PV standard standard for Seal contact pressure ( ) 1) 3) relevant limit states. Make-up leak test ( ) Bolting PV standard Loads as specified for clamp body above Pre-tension Gripping/ locking capacity PV standard RP-F113 Steel wall force (N) 2) Ref. Table 6-3,, Pipe wall RP-F113 Internal design pressure ( ) Internal test pressure ( ) External pressure ( ) Local pipeline external loads (N, M) Loads from activation of gripping 4) Seal contact pressure ( ) 1) 3) Make-up leak test ( ) (also consider: ) Allowable stress from applied design standard for relevant limit states. Ref. Table 6-4., Ref. Table 6-3 Ref. Table 6-4,,., Recommended tests Seal Test RP-F113 Seal test pressure, see Table 12-1, Note 1. Test duration and acceptance criteria are given in Appendix [C.6]. 1) Lower bound for seal functionality and gripping capacity, upper bound for structural integrity. 2) Steel wall force N is detailed in Sec.5. 3) Seal contact pressure includes effects of make-up, differential pressure, test pressure, swelling and thermal expansion. Load and resistance safety factors to be established through qualification and testing. 4) Load and resistance safety factors for gripping loads to be established through qualification and testing. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 113

114 A.5 Sleeve repair Sleeve repair can be used as an alternative to mechanical couplings and butt weld repair in pipeline section replacement repairs. Design procedures for the sleeve repair fillet weld are included in App.D. Figure A-4 Typical code break and design factors for sleeve repair Table A-4 Example of code break and load and resistance factors Part Standard Loads Load factors (ref. Table 6-3) Sleeve RP-F113 Internal design pressure ( ) Internal test pressure (P lt ), External pressure ( ) (also consider: ) Local pipeline loads (N, M) 1) Weld deposit RP-F113 Pipeline loads (N, M) 1), Pipe wall RP-F113 Internal design pressure ( ) Internal test pressure (P lt ) External pressure ( ) Local pipeline loads (N, M) 1) Assembly, after repair ST-F101 System pressure test (also consider: ) (also consider:) Resistance factors (ref. Table 6-4),., Recommended tests FAT (TQ) RP-F113 Burst and axial load capacity > pipe capacity - or project-specific design load criteria 1) Steel wall force N is detailed in Sec.5.,,.,,., Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 114

115 A.6 Repair coupling and flange adapter Figure A-5 Repair coupling (left) and flange adapter (right) Table A-5 Example of code break and load and resistance factors Part Standard Loads Load factors Resistance factors Coupling, PV standard Internal design pressure ( ) PV standard Allowable stress from Flange adapter, Internal test pressure (P FAT ) applied design flange body Local pipeline loads (N, M) 2) standard for relevant limit states Make-up leak test ( ) External pressure ( ) Loads from activation of gripping 4) Seal contact pressure ( ) 1) 3) Bolting PV standard Loads as specified for coupling body above Pre-tension Gripping/ locking capacity See above PV standard RP-F113 Steel wall force (N) 2) Ref. Table 6-3 Pipe sections RP-F113 Internal design pressure ( ) Internal test pressure ( ) Local pipeline loads (N, M) 2) Make-up leak test ( ) External pressure ( ) Loads from activation of gripping 4) Seal contact pressure ( ) 1) 3) Allowable stress from applied design standard for relevant limit states Ref. Table 6-4 (also consider: ) Ref. Table 6-3 Ref. Table 6-4 Recommended tests Seal test RP-F113 Seal test pressure, see Table 12-1, Note 1. Test duration and acceptance criteria are given in Appendix [C.6]. 1) Lower bound for seal functionality and gripping capacity, upper bound for structural integrity (see [6.4.2]). 2) Steel wall force N is detailed in Sec.5. 3) Seal contact pressure includes effects of make-up, differential pressure, test pressure, swelling and thermal expansion. Load and resistance safety factors to be established through qualification and testing. 4) Load and resistance safety factors for gripping loads to be established through qualification and testing.,,,.,,., Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 115

116 A.7 Isolation plug Schematic guidelines on code breaks and design factors for isolation plugs are given in the figure and table below. Figure A-6 Pipeline in-line isolation tool Table A-6 Example of code break and load and resistance factors Part Standard Loads Load factors Resistance factors Plug module PV-code Hydraulic activation Max. diff. pressure design ( ) Max. diff. pressure test ( ) Contact pressure from seals Loads from activation of gripping Pipeline (local) RP-F113 Min/max. diff. pressure design ( ) Min/max. diff. pressure test ( ) Contact pressure from seals 2) Contact pressure from gripping 2) Locking capacity/ gripping PV standard Allowable stress from applied design standard for relevant limit states ref. Table 6-3 ref. Table 6-4 RP-F113 Min/max. diff. pressure design ( ) 1) ref. Table 6-3 ref. Table 6-4 Max. diff. pressure test ( ),., Local pipeline loads (N, M) Recommended tests FAT PV standard Typically 1.43 x p li No leak Seal test installed RP-F113 See [7.10.2] No leak 1) The gripping capacity shall be documented for two conditions: hydraulic activated only, at maximum differential pressure not providing self-locked condition of the plug hydraulic activated plus self-locked by differential pressure, at maximum allowed differential pressure. 2) Seal and gripping loads:,, seal contact pressure includes the effects of make-up, differential pressure, test pressure, swelling and thermal expansion. Load and resistance safety factors to be established through qualification and testing load and resistance safety factors for gripping loads to be established through qualification and testing. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 116

117 A.8 Hot tap installed isolation plug Schematic guidelines on code breaks and design factors for a hot tap installed isolation plug are given in the figure and table below. Figure A-7 Hot tap installed plug Table A-7 Example of code break and load and resistance factors Part Standard Loads Load factors Resistance factors HT clamp and branch w/bolts PV standard Make-up bolt pre-load 1) Internal design pressure ( ) External pressure ( ) Seal contact pressure ( ) 1) 3) Branch loads (N, M) 2) Pressure test, FAT ( ) Make-up leak test ( ) Locking PV standard Internal design pressure ( ) External pressure ( ) Branch loads (N, M) 2) PV standard PV standard Allowable stress from applied design code for relevant limit states PV standard Pipe wall (local) RP-F113 Make-up bolt pre-load Internal design pressure ( ) External pressure ( ) Seal contact pressure ( ) 1) 3) Branch and pipeline loads (N, M) 2) Pressure test, ( ) Ref. Table 6-3, Ref. Table 6-4,., Make-up leak test ( ) Recommended tests FAT PV standard Typically 1.43 x p li PV standard Seal test RP-F113 p li Seal test pressure, see Table 12-1, Note 1 Test duration and acceptance criteria are given in Appendix [C.6] 1) Lower bound for seal functionality and gripping capacity, upper bound for structural integrity (see [6.4.2]). 2) Branch load and moment is mainly the plug retaining load. Steel wall force N is detailed insec.5. 3) Seal contact pressure includes effects of make-up, differential pressure, test pressure, swelling and thermal expansion. Load and resistance safety factors may need to be established through qualification and testing. In addition, accidental loads such as the ROV impact on the branch shall be evaluated in risk assessments. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 117

118 A.9 Hot tap tee Schematic guidelines on code breaks and design factors for a hot tap tee are given in the figure and table below. Figure A-8 Hot tap tee and clamp Table A-8 Example of code break and load and resistance factors Part Standard Loads Load factors Resistance factors HT clamp and branch w/bolts PV standard Make-up bolt pre-load 1) Internal design pressure ( ) External pressure ( ) Seal contact pressure ( ) 1) 3) Branch loads (N, M) 2) Pressure test, FAT ( ) Make-up leak test ( ) PV standard Allowable stress from applied design code for relevant limit states Gripping RP-F113 Internal design pressure ( ) Ref. Table 6-3 Ref. Table 6-4 External pressure ( ),,., Branch loads (N, M) 2) Pipe wall (local) RP-F113 Make-up bolt pre-load Internal design pressure ( ) External pressure ( ) Seal contact pressure ( ) 1) 3) Branch and pipeline loads (N, M) 2) Pressure test, ( ) Make-up leak test ( ) Ref. Table 6-3 Ref. Table 6-4 Recommended tests FAT PV standard Typically 1.43 x p li PV standard Seal test RP-F113 p li Seal test pressure, see Table 12-1, Note 1 Test duration and acceptance criteria are given in Appendix [C.6] 1) Lower bound for seal functionality and gripping capacity, upper bound for structural integrity (see [6.4.2]). 2) Steel wall force N is detailed in Sec.5. 3) Seal contact pressure includes effects of make-up, differential pressure, test pressure, swelling and thermal expansion. Load and resistance safety factors may need to be established through qualification and testing.,,., In addition, accidental loads such as the ROV impact on the branch shall be evaluated in risk assessments. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 118

119 APPENDIX B FITTING CAPACITY B.1 Fitting strength capacity B.1.1 General The locking capacity depends on the method of attachment to the pipe: the attachment is nearly uniform along and around the pipe and employs many balls or teeth, similar to a friction-based connection the locking capacity depends on local attachments, such as edges, teeth or balls, penetrating the pipe surface. The structural strength of the fitting parts and the locking capacity of a fitting shall be sufficient to convey the pipeline forces. The following parameters of concern are discussed for the various fitting groups: Bending The bending strength of the sleeve is governing, together with the pipe's ability to convey the bending moment and transverse shear forces. A sleeve will, however, increase the pipeline stiffness locally. Contact forces The radial contact forces between the pipe and sleeve govern the capacity to transfer axial and torque forces in combination with the locking coefficient (the efficiency of the locking). This radial contact force is generated by the pre-tension and internal pipe pressure. This force is further increased by pipe tension for couplings with a wedging effect. The radial compressive contact force is limited by either the collapse strength of the pipe or the radial stiffness of the sleeve. Radial expansion contact forces imposed by e.g. isolation plugs are limited by the available stroke length activating the plug or by the pipe-bursting capacity. Pressure Internal pressure will expand the pipe and hence may improve the locking capacity based on external gripping. This load type may therefore not be dimensioning for the coupling or structural clamp. For isolation plugs, the effect of internal pressure adds to the radial expansion forces conveyed from the plug gripping and sealing segments. This may be governing for the plug tension capacity and allowable pipeline strains and stresses in a plugging operation. Only relatively small tension capacities can be verified by a pressure test alone. An external differential pressure can occur in gas pipelines after depressurization. This load condition tends to contract the pipeline diameter and may therefore reduce the coupling's tension capacity. Tension/compression Pipe tension tends to contract the pipeline diameter due to the Poisson effect. Thus, the radial contact forces may be reduced for couplings which have no wedging effects, resulting in a slightly reduced tension capacity. Couplings with wedging effects may increase the radial contact forces through pipe tension. Increased radial contact forces cause pipe contraction and may lead to an axial displacement of the pipe inside the coupling. Axial pipe compression may reduce the contact forces and cause the pipe to slide inside the coupling. Likewise, the capacity will be reduced for axial compressive pipe loads unless the pipe ends meet a recess in the coupling or the other pipe end. Torsion Significant torsion capacity is seldom required. The torsion capacity is related to the contact forces multiplied by the friction coefficient and contact radius. Local gripping by balls etc., which prevents the pipe from rotating, improves the torsion capacity. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 119

120 Temperature Normally, the pipe is warmer than the sleeve. This causes some increased contact force between the coupling and pipe, depending on the design. Different thermal expansion coefficients of sleeve and pipe will, however, also affect this contact force. Isolation plugs are exposed to the pipeline medium and temperature effects may increase the compressive loads from the plug seals. Fatigue Significant fatigue capacity is seldom required. FEA/testing can be applied to demonstrate fatigue capacity. Some aspects of the coupling types are discussed below. B.1.2 Symbols The following symbols are employed in the formulas that are derived in the sections below. Note that, in the formulas, subscripts of s and p are used for sleeve and pipe respectively. See also [6.4]. Table B-1 Symbols used below D = Outer diameter of the pipe. It may be assumed that the difference (e t ) between the sleeve's inner diameter and the pipe's outer diameter is negligible compared to the diameter of the pipe. Therefore, the pipe's outer diameter may be taken to be equal to the sleeve's inner diameter L = Length of contact surface between sleeve and pipe e f = Change in diameter due to tension force e t = Shrink fit - difference between the sleeve's inner diameter and the pipe's outer diameter e tm = Shrink fit produces a contact pressure, which generates a fraction of the pipe's yield stress e p = Change in the pipe's outer diameter e s = Change in the sleeve's inner diameter t = Thickness T m = Make-up temperature T o = Operational temperature for sleeve ΔT = Temperature difference between pipe and sleeve B.1.3 General compression fit Fittings designed to transfer axial pipeline loads through a forged mechanical connection between the pipe and sleeve are dependent on initial high compressive forces between the pipe and sleeve. Figure B-1 illustrates a general compression fit between cylinders. ts L D tp Figure B-1 Illustration of compression fit ( Shrink fit ) e t = e p + e c Where: e t = shrink fit - difference between the sleeve's inner diameter and the pipe's outer diameter e p = change in the pipe's outer diameter e c = change in the sleeve's inner diameter. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 120

121 B.1.4 Expanded sleeves One type of forging process expands the pipe to obtain a compressive radial load between an outer sleeve and the pipe. The forging sequences are: 1) Expand the inner pipe until yield stress 2) Continue the expansion by yielding the inner pipe within the limit of: a) an acceptable permanent deformation b) an acceptable stress of the sleeve 3) relieve the internal forging force. This causes the sleeve to elastically shrink whilst the pipe has experienced a permanent (plastic) deformation. The remaining compressive force between the pipe and sleeve must be sufficient to: assure a locking in the axial direction seal. The seal is best achieved by a local surface yield occurring circumferentially between the two surfaces during the forging process. Internal ribs in the sleeve are beneficial for this purpose. Internal ribs also improve mechanical locking in the axial direction and thus improve the axial force capacity. B.1.5 Pipe collapse The contact forces during/after make-up are limited by: type 1: type 2: the pipe's collapse strength for uncontrollable radial deformations. the pipe's yield strength if the radial deformation is controlled and equal all around. Some fittings can cause the above types of pipe collapse under the following conditions: friction (e.g. shrink-fit connections) type 1 during make-up grip, balls (wedged) type 2 during the make-up and tension of the pipe flanged type 2 during make-up B.1.6 Locking friction factors Several types of isolation plugs, structural clamps and couplings partly depend on friction. The friction coefficient depends on a range of factors: 1) static or dynamic 2) surface finish 3) material combinations 4) possible lubricants. There are no distinct differences between mechanical locking and friction. Very rough surfaces tend to increase the locking capacity. Commonly used static friction coefficients for steel/steel surfaces range from 0.1 to 0.6. Sliding friction can be less. Note, however, that NS (Eurocode 3) specifies a slip factor (friction coefficient) to be used in friction connections varying from 0.2 to 0.5, depending on the surface treatment applied. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 121

122 Table B-2 Friction coefficients Slip factor 0.5 Surface Surfaces blasted with shot or grit, with any loose rust removed, no pitting Surfaces blasted with shot or grit and spray-metalled with aluminium Surfaces blasted with shot or grit and spray-metalled with zinc-based coating proven to provide a slip factor of not less than Surfaces blasted with shot or grit and painted with a zinc silicate paint to produce a coating thickness of micron 0.3 Surfaces cleaned by wire brushing or flame cleaning, with any loose rust removed 0.2 Untreated urfaces B.1.7 Geometric locking External local force, where balls provide the lock The point loads from the balls are to be distributed. The size of the balls is the key parameter for determining the number of balls to be used, and this is limited by geometrical conditions and local deformations. The minimum ball diameters will therefore be determined based on: 1) clearance to the bridge between the pipe and sleeve 2) deformation of the pipe 3) deformation of the sleeve 4) strength. The locking is based on a local plastic yield of the pipe caused by the radial force from each ball. Local buckling of the pipe wall, instead of the required local plastic yield, is avoided by applying a sufficient number of balls around the circumference. Thereafter, the ball diameter must be optimized to obtain sufficient indentation and the number of ball rows must be optimized to achieve sufficient holding capacity. External grip from teeth on wedge Locking is obtained by an axial load generated by bolts which force the wedges into the pipe. During activation, teeth on the wedges penetrate the pipe surface and cause locking, shown in Figure B-2. The contact pressure between the pipe and wedge depends on the axial activation force, friction coefficient and magnitude of the wedge's taper angle. Sleeve K Wedge 1 R Wedge 2 Grip teeth Pipe Figure B-2 Configuration of a simple connection providing locking by wedges Pipe expansion into grooves in sleeve Analysis and tests of a typical coupling indicate for the following load conditions: 1) Simple tension: the pipe waves in the grooves will be very slightly smoothed, but this Poisson effect will not affect the tensile capacity. The contact pressure at the edges due to tension will increase for the coupling because of the increase in the axial component. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 122

123 2) Axial compression: the pipe waves in the grooves will be slightly deeper but this will not affect the axial capacity. 3) Tensile load capacity: the few sharp edges on the sleeve penetrate the pipe surface. Pipe tension causes high stresses around the pipe circumference, thereby limiting the tension capacity when only a few edges carry the load. The pressure will not improve the tension capacity. B.1.8 Longitudinal force distribution Connections will have a longitudinal shear force distribution. This depends on the thickness of pipe and sleeve, elasticity modulus, and the length and type of joint between the pipe and sleeve. Generally, the radial load is expected to be higher at the sleeve ends due to the effects of the undisturbed pipe. Thus, a friction shear capacity will be higher close to the sleeve ends. This is an effect which, to some extent, may compensate for the higher shear stress at the coupling entrance caused by external forces. There will be some shear stress due to a temperature gradient. The change in shear stress due to this temperature gradient will in most cases be small compared to that caused by tension. The impact of temperature on shear stress decreases for couplings with larger lengths. The effect of fluctuations of temperature will in most cases be small. B.1.9 Micro motions Each temperature cycle may cause an internal relative longitudinal displacement between the pipe and sleeve. When this motion is combined with tension, a fully friction based junction may experience a small resultant longitudinal sliding for each temperature and load cycle. For the majority of applications, however, the resultant effects are not considered to be of concern to the long-term locking capacity. B.2 Seal Capacity B.2.1 Discussion Soft seals Seal manufacturers normally recommend limitations on the use of the seal, including pre-tension for pressure activated seals as well as limits for extrusion gaps as a function of pressure, temperature, time and load type. These recommendations shall be documented; however, as the application of soft seals for fittings is often outside normal use, further qualification may also be required. Metal seals The gasket's make-up pressure must significantly exceed the material yield strength of the seal material (or pipe material). Otherwise, the seal material will not flow into the discontinuities and a seal cannot be obtained. These requirements can be relaxed if all discontinuities are removed completely, but this is seldom practical for pipe surfaces. Consequently, wide metal gaskets will be impractical as this would require unacceptable high radial loads on the pipeline. Such high loads could cause a pipeline collapse due to the resultant high hoop stresses. Therefore, radial metal seals for pipelines have thin sealing areas, often obtained by the local fitting geometry penetrating the pipeline. B.2.2 Compressive loads The sealing action depends on a compressive load between the seal and its sealing surface. The contact pressure must exceed the pressure of the fluid to seal. A safety factor must be applied to ensure this condition. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 123

124 B.2.3 Uniform loading Circumferential seal loads Some areas of concern are: 1) Loads closer than 0.2 diameters to the end cause significantly more inward deflections of the pipe wall than if they were further away. Elastic deflections at the end can exceed 4 times the deflections in a midsection of the pipe. High loads close to the ends are more likely to cause plastic deflections. 2) A distributed circumferential load on a pipe length of less than 0.1 diameter gives equal pipe shell global responses similar to that from a line load of the same force (the contact stress is reduced proportional to the length). 3) Plasticity of the pipe wall can be caused by high seal loads. This starts with a yield related to plate bending (meridional bending) before hoop yield. Formulas for the plastic behaviour of the pipe wall can be developed based on the pipe wall's plastic capacity and calibration. 4) Local plastic yield of the pipe surface is required for metal seals. Formulas for the penetration depth can be developed based on the theory related to Vickers hardness measurements and calibrations. B.2.4 Thermal effects Polymers The thermal expansion of rubber in an enclosed space can be a matter of concern, as the thermal expansion coefficient of polymer materials can be more than 10 times that of steel. Typical conditions for a polymer seal tightly enclosed within a steel boundary are indicated below, assuming: 1) equal temperature in the steel and seal 2) a thermal expansion coefficient that is 10 times that of steel 3) an incompressible seal Steel stress magnitude: σ = E α Δt = 25 Δt Thus, unstressed steel with a yield strength of 350 MPa will yield at a temperature 14ºC higher than at make-up. The steel will, however, be pre-stressed, and hence will yield at a lower temperature. The magnitude of the permanent relaxation will be as follows, assuming: 1) plastic yield in one direction 2) typical polymer seal thickness l in this direction: 50 mm 3) temperature increase from make-up: 50 C 4) pre-stressed steel to yield. Relaxation magnitude: α Δt l = 0.3 mm. Consequently, the effect of different thermal expansion coefficients must be considered in the design, i.e. there must be sufficient space for this expansion to avoid such effects. Fluids The expansion coefficients for fluids trapped in cavities must be considered in the design: 1) hydrocarbon gases (mainly methane) initially at 205 bar and +4 C, then heated to 60ºC typically give an increase to approximately 2.1 times the initial pressure 2) the thermal expansion coefficient of water depends on the temperature, pressure and salinity Fresh water has a thermal expansion coefficient of 0 at +4ºC. 3) oil typically has a thermal volume expansion coefficient of 0.001/ C. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 124

125 B.2.5 Swelling Polymer materials tend to expand due to the absorption of fluids, and this is to be considered in the design. B.2.6 Eccentricity Flexibility in the systems design must compensate for the possible eccentricity between the seal and pipe. This eccentricity may be caused by: 1) external forces acting on the pipe ends during activation 2) the function of the locking mechanism and positioning of the fitting 3) pipe deviations from straightness 4) pipe deviations from roundness. B.2.7 Axial effects Load effects Elastic deflections of the pipe and sleeve due to the axial pipe forces can, for some designs, cause a relative axial displacement between the pipe and seal. This can be of concern for long-term use and should therefore be considered in the qualification plans. The concern is related to: 1) seal displacements over local discontinuities of the pipe surface 2) wear. Axial load effects are of most concern in relation to thin metal seals. Thermal effects The seal is often located at some distance from the locking of the pipe. The pipe section between the locking and seal will expand by temperature, whilst the sleeve external to the pipe will expand less due to the cooling effects of the water. Therefore, effects similar to those above must be considered in the design. B.2.8 Installation Water block Water trapped in cavities during make-up prevents further action. This is of particular concern to the seals made by the pipe expanding. The pipe is intended to expand into grooves of the sleeve, but this can be prevented by water located in the groove. Several designs therefore apply a resin filled with gas bubbles to reduce the water block effect. Such resins must, however, be qualified for the water depth in which they are installed. Deep water requires relatively compact resins to avoid the collapse of the gas bubbles inside. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 125

126 APPENDIX C TYPICAL TESTS C.1 Introduction The test requirements related to repair fittings are given in [2.4.2] and Sec.12. Guidelines for the test requirements are given below. C.2 Basic tests C.2.1 General Basic tests can be used to either establish limiting parameters which are not established by analysis, or validate numerical and analytical models. Basic tests can be used to reduce the extent of type tests (qualification tests) required in combination with analysis. The output from basic tests may be used as input to the analytical and numerical design models to reduce the model parameter uncertainties. Basic tests are also recommended to be applied in the qualification process for the various technologies/ functions/components to validate the calculation/analysis method(s) used. Basic testing can be performed both as small-scale testing (selected parts of the design) and full-scale testing (whole assembly). Examples of typical basic tests are given below. C.2.2 Materials Typical tests are related to material properties and are well regulated by international and recognized standards. Information can be obtained from literature and manufacturers, as indicated below: 1) properties of metallic alloys are easily obtained for commonly used metallic materials 2) relevant properties of non-metallic sealing compounds (rubbers, plastic, carbon, etc.) are difficult to obtain. Tests related to resistance against the various types of corrosion mechanisms are dealt with as for the pipeline itself, and additional corrosion mechanisms relevant or imposed by the repair fitting (e.g. large local pipe wall strains combined with sour service and crevice corrosion). Ageing tests of polymer seal materials are used to predict the lifetime of a seal in specified environments and are therefore time consuming. The test time can be reduced by an increase in the test temperature, e.g. based on the procedures given in ISO Appendices D and E. Ageing tests must be supplemented by detailed documentation of the materials. Care should be taken so that the increase in temperature does not introduce other degradation mechanisms, see also NORSOK M-710 or ISO C.2.3 Combined effects Some combinations of design and material parameters require separate tests to establish limiting parameter values and validate the numerical/analytical model for the relevant pipeline application range. Normally, at least three tests with separate specimens should be used to indicate the possible spread of results. C Seal tests Extrusion gap test of soft seals This establishes the relationship between the: 1) size of the clearance gap 2) pressure 3) friction 4) temperature, and 5) time. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 126

127 The seal manufacturers often give recommendations (limitations) based on documented testing. However, the intended seal applications can be outside such recommendations and will therefore require separate tests. Two types of gaps are of concern: The first gap of concern is related to clearances to the seal. The size of the tested clearance gap must be determined accurately, and can be affected by both the pressure and temperature. The gap is either preset/fixed or measured during the test. The second gap of concern is related to sealing against discontinuities on the pipe. The allowable operating limits for pipeline surface discontinuities (gouges, weld seams, scratches etc.) shall be established through the qualification process. The pressure may be applied either via a test fluid or, for larger compact soft seals, directly as a compressive force causing the intended internal pressure in the seal compound. The term extrusion shall be defined in relation to the seal's failure mode. For a soft seal as well as backup rings, this could address: 1) the permanent deformation into the gap as a ratio of the gap size 2) plastic deformation of internal strengthening members, such as metal springs in the seal 3) the relative amount of fibre reinforcement fracture in the seal 4) loss of seal pressure. Metal seal tests Metal seals shall seal against the pipe surface, including defined surface discontinuities. Therefore, sufficient plastic yield of the pipe surface and/or the seal must be obtained. Important test parameters are: 1) material hardness of the seal 2) material hardness of the pipe 3) shape of the seal 4) load applied to the seal 5) defined discontinuity of the pipe 6) for seals that can be marked during installation: defined discontinuity of the seal. The determining parameter could either be: a leak test, or microscope investigations of the specimens being forced together, combined with a later full-scale test including defined discontinuity. C Friction factor tests Most types of couplings and in-line isolation tools are affected by friction either during installation/make-up or in operation. Friction coefficients which are critical shall be documented by tests. Friction factors from internationally recognized handbooks may be applied in the design documentation provided: the considered assembly complies with the conditions and assumptions presented in the reference literature, including material specifications, surface roughness, material hardness and contact pressure level and distribution sensitivity studies are included in the design documentation, documenting an acceptable friction-factor envelope, and the applied friction factor is within this envelope with an acceptable margin and confidence. type tests confirm the expected performance of the interacting parts of the assembly. Important test parameters to be qualified are: 1) material combinations 2) surface roughness Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 127

128 3) specific compression load 4) velocity (dynamic friction) 5) possible in-between fluids or contamination. Forces which are required for movement and compression are monitored in order to establish the friction coefficient. C Pipeline gripping tests The pipeline gripping mechanism must be tested in order to verify analytical (or numerical) design assumptions made for pipeline gripping mechanisms in couplings and in-line isolation plugs (and some types of repair clamps). The test establishes the relationship between the grip indentation and axial load capacity based on parameters such as: pipeline differential pressure and external loads pipeline dimensions pipeline material properties (yield strength, surface hardness) material properties of the gripping segment (e.g. surface hardness). The grip design is often based on a radial compression/expansion load caused by an axial movement. The axial movement is transferred to a radial compression/expansion load by wedges. Testing shall verify the axial grip capacity related to the axial movement and the effects on the pipeline. Relaxation of the pipeline gripping and/or seal mechanism due to movement of the fitting's internal components or relaxation of the hydraulic/mechanical pre-tension is a relevant failure mode which can only be assessed by testing the complete assembly. If this is not covered by the basic testing, it shall be covered by the type testing. C.2.4 Galling test Galling causes damage to the surface finish as well as high-friction coefficients. The galling limits are determined in the same way as friction coefficients but combined by a microscope survey of the surfaces. C.2.5 Polymer decompression limits (explosive decompression) Seals in gas systems can be damaged by high decompression rates. Gas which was dissolved in the material at high pressure can form bubbles when the pressure is reduced, and this can result in seal damage. Important test parameters are: 1) material type 2) size, shape of material and gas pressure in the exposed area 3) gas type, either the actual or a type which exhibits similar effects 4) pressure 5) saturation time 6) decompression rate 7) temperature 8) method for detecting blisters. C.2.6 Materials and material combinations related to subsea use Materials and material combinations with possible failure modes related to subsea use shall be qualified by basic tests. Such features could be: 1) the volume elasticity and water absorption properties of materials that fill voids 2) swelling 3) electrical isolation 4) hydraulic systems' pressure-compensation systems. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 128

129 C.2.7 Examinations The tested parts and assemblies shall be thoroughly examined after the test. This examination shall include: 1) measurements of dimensions, relative to before testing 2) permanent deformations and marks. C.3 Type tests C.3.1 Introduction A type test (qualification test of the type) verifies, in combination with analysis, the functional requirements and safe operational limits of the fitting type. The number and extent of type tests depend on: 1) the extent of documented experiences 2) the extent of the analysis performed 3) the accuracy and conservatism of the analytical approach 4) the extent of the basic tests performed. A type test is succeeded by a factory acceptance test (FAT). The type test could also be combined with the FAT. This combined testing is efficient when only one fitting of the type is made. The optimum qualification scenario is an analysis qualified by a practical combination of basic tests and a type test. The type tests include the extreme tolerance combinations of dimensions, pressure, temperature, fluids, operation and installation for which the analysis is either incomplete or indicates a particular risk of failure. In addition, type tests are used to verify analyses. This involves the accurate measurement of sufficient parameters for comparison with those in the analysis. The measurement and monitoring of accuracy shall be documented. Examples of typical type tests are given below. C.3.2 Test specimens Pipes The pipes selected for type tests should represent the extreme dimensional tolerance combinations, surface discontinuities and material properties, unless the effects of these are sufficiently covered by the basic tests and analysis. Such pipes will, however, be difficult to obtain as pipes will normally have only some adverse combinations. The effects of other combinations which are not available on the test specimens must be covered otherwise, e.g. by analytical and/or FEA sensitivity studies combined with necessary tests verifying the validity of these analytical or FEA models. The method used to manufacture the pipe shall be specified. The detailed pipe dimensions shall be measured and documented by a dimensional sketch as specified in [14.1.6]. The pipes shall be marked to identify the measurement positions and show the intended axial and angular location of the coupling. Test documentation valid for the particular pipe shall document actual material properties. Hardness (Brinell or Vickers) shall be measured in weld areas. Fitting The fitting geometry, dimensional tolerances and material properties shall be documented as specified in [14.1.6]. Material test certificates shall document the actual material properties of both metals and seals. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 129

130 C.3.3 Installation Installation and retrieval tests of piggable isolation plugs, including pigging to location, tracking and retrieval, shall be performed to identify critical parameters, e.g. the minimum pipeline bend radius, limitations on the pipeline's inner diameter and out of roundness. The installation test for fittings other than piggable isolation plugs shall simulate the exaggerated actual installation, i.e. design conditions with a margin. This shall include e.g. the coupling or clamp maximum design limits with respect to the pipe minimum end chamfer (if any), maximum misalignment (and pipe straightness deviation), and eccentricity between the pipes and fitting, including a margin. Furthermore, the stiffness of the pipe supports, the support of the fitting, as well as the effects of gravity/ buoyancy shall comply with the fitting limiting specifications. Thus, the limiting forces and critical seal interactions can be simulated during the installation test when the fitting's position is adjusted while the fitting is resting on the pipes. The displacement shall be performed with actual maximum specified velocities. The basis for the test procedure/test rig set-up is: 1) an overview of the critical tolerance combinations for installation 2) applied safety factors on tested tolerance combinations 3) stiffness of pipe ends including subsea fixation, if the installation causes pipe deflection that may have an adverse effect 4) stiffness of the fitting support if it may have any adverse effect 5) weights dry and submerged 6) displacement velocities 7) possible different effects of a dry test versus a submerged test. After installation, the fitting shall be removed from the correct position and the internals shall be inspected for interactions with the pipe. The seal area is of main importance. The installation test shall be repeated to cover all critical tolerance combinations, and at least three tests shall be performed. All the parameters mentioned above, including the seal's visual appearance, shall be recorded and compared with the acceptance criteria. Photographs shall also document interaction marks. C.3.4 Activation The activation test shall simulate the most adverse design conditions with a margin. Activation and de-activation of isolation plugs, including communication, shall be performed to identify critical parameters (e.g. limitations on the pipeline inner diameter and out of roundness, maximum steel wall and coating thickness with respect to communication, battery life and contingency properties in case communication is lost, etc.). In general, the most critical tolerance combination for activation should be selected. This will normally be the thinnest pipe-wall thickness, largest clearance between fitting and pipe, and largest misalignment (and pipe straightness deviation) combined with largest stiffness. Furthermore, the fitting shall be positioned at its maximum specified deviation from its intended position, e.g. one pipe end for a coupling. This deviation shall be in the most critical direction. For a coupling, this will result in a shorter distance between the coupling seal and pipe end. The basis for the test procedure/rig set-up is: 1) an overview of critical tolerance combinations for activation 2) a margin applied to these tolerance combinations for determining test combinations 3) stiffness of pipe at the fitting location, e.g. the ends, including the effects of subsea fixation to alignment frames, if the activation causes internal bending moments inside the fitting 4) weights dry and submerged Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 130

131 5) possible different effects of a dry test versus a submerged test 6) activation procedure 7) monitoring and measurement procedures. Other tests normally follow on from activation tests. Fittings which enable repeated activation shall be subject to at least three activation tests. The deactivations shall be monitored in a similar way to the activations. C.3.5 Strength/leakage Type tests of the fitting strength/tightness shall normally be carried out for the hydrostatic test condition as specified in the design standard for the fitting (see Appendix [C.4]). Alternatively, the fitting can be tested to failure or according to the alternative design approach as detailed in [2.1]. The basis for the test procedure is: 1) design capacity specification for separate loads and combined loads 2) a margin between design conditions and test conditions, see Sec.12 3) the actual dimensions and yield strength of the pipe for the test 4) the activation condition which gives the least strength capacity 5) measurements/monitoring of longitudinal (and rotational) displacements between the fitting and pipe as a function of the load 6) possible strain gauge measurements for verifying the analysis, supplementing the analysis and determining loads 7) leak detection measurements. Leaks and unacceptable permanent deformations and displacements are rejection criteria. Leak test holding times are given in [C.6]. Pressure test The basis for the test procedure is: 1) test pipes with end caps 2) pressure causing a defined hoop stress utilization depending on the application, e.g of the actual yield strength of the test pipe. Structural clamp and coupling tensile test without pressure For most pipelines, the tension capacity of the repair fitting does not need to equal that of the pipeline. The basis for the test procedure is: 1) a pipe tension, without significant internal pressure, as a fraction of the yield capacity of the actual testpipe 2) a small insignificant internal (or annulus) pressure necessary to check the seal tightness. Structural clamp and coupling compression test without pressure A test can document the compression capacity in couplings where: the pipe ends do not meet each other the pipe ends do not meet a recess, and an axial pipe displacement inside the coupling can cause negative effects. The basis for the test procedure is: 1) a pipe compressive force as a specified fraction of the yield capacity of the actual test-pipe 2) a small insignificant internal (or annulus) pressure to check the seal tightness. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 131

132 Torque test A significant fitting (i.e. structural clamp or coupling, or hot tap tee) torque capacity is needed for only a few pipeline applications. The basis for the test procedure is: 1) a pipe torque as a specified fraction of the torque yields the capacity of the actual test-pipe 2) an insignificant internal (or annulus) pressure to check the seal tightness. Bending test Bending moments can be introduced to structural clamps and couplings by the activation test described in [C.3.4]. Bending moments introduced after activation will in general be of less concern to repair fittings. Normally the pipe, and in some cases the connection to the coupling, will represent the limit. An insignificant internal (or annulus) pressure should be applied as necessary to check the seal tightness. Pipe bending moments from branch connections are of concern to T joints (such as hot tap tees). Angular misalignment joint fittings shall be tested to document acceptable resistance to specified external loads. Pipe bending moments can also cause additional plastic deformation of the pipe wall when already subject to strain in the plastic region caused by high loads from e.g. seals in couplings, clamps and plugs. Fatigue test Only a few pipelines are subject to alternating bending loads of concern. The pipe itself, at its connection to the repair fitting, will in most cases represent the weakest point of the connection with respect to fatigue. The general limiting fatigue resistance of the pipeline is at pipeline butt welds. Pipe fatigue criteria are described in DNVGL-ST-F101. The basis for the test procedure is: 1) a specified maximum alternating bending moment as a fracture of the bending yield capacity of the pipe tested 2) number of cycles with this load 3) distribution of the magnitude of bending moments and higher number of alternating loads 4) an insignificant internal (or annulus) pressure to check the seal tightness. Temperature test The main objective of the temperature test is to document acceptable functional performance for a specified temperature envelope. The need to perform temperature test(s) depends on uncertainties and potential related failure modes in the analytical/analysis model(s). Typical threats and failure modes concerning temperature loads on polymer seals are related to the thermal expansion factor being 10 times larger than that of steel. For solid polymer seal cross-sections filling the seal groove, thermal expansion may cause yielding of the interfacing steel material, resulting in a leak at shut-down. Further, polymer materials may have non-linear de-rating, giving less resistance to extrusion failures at elevated temperatures for exposed duration. The analysis shall show whether the temperature reduces the strength/seal capacity more than that related to the material strength reduction. From this analysis, important combinations with any of the above tests, with part-loads, can be established and will form the basis for test procedures to assure the strength capacity at elevated temperatures. Relevant temperatures are: 1) maximum and minimum fluid temperatures 2) water temperature 3) resulting temperatures of the pipe and fitting 4) transient temperature distribution within the fitting during start-up and shutdown. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 132

133 External pressure test Deep subsea gas pipelines can have a particular load case if the pressure is relieved from the pipeline. The external pressure tends to reduce the interaction forces between the fitting and pipe due to additional contraction of the pipe. The basis for the test procedure is: the external differential pressure resulting from the water depth and the remaining internal pressure. The limit is the pipe's collapse pressure. Combined loads The analysis shall show whether any of the specified combined design load cases gives a smaller safety margin against failure than the separate cases described above. Such cases shall form the basis for testing combined loads. Possible combined loads for couplings are: internal pressure causing a hoop stress of the pipe equal to 80% of the actual yield strength combined with tension (simulating a free span) and/or compression bending moments. C.3.6 Seal tests C General Seal tests are partly included in the Basic tests described above. In addition, the basis for seal test procedures includes: 1) facility used to confirm the integrity of the connection after make-up If no basic test: 2) test of the relationship between the seal compressive load and pressure leakage limit 3) test to confirm sealing at defined pipe surface irregularities 4) gas seal leak test 5) gas migration test. This includes detection using helium combined with gas circulation outside the seal, via a detector, to find helium atoms in the gas stream. 6) test to confirm the seal function if there is a defined eccentricity between the pipe and coupling. 7) the seal test pressure confirming the integrity of the installation should in general be times the design pressure on the seal provided the maximum pipe stress is less than 0.96 of the specified minimum yield stress or the resulting pipe strain is within acceptable limits. Strain causing local plasticity of the pipe needs to be qualified for the specified load envelopes, including activation, pressure and temperature. For an annular seal test, the qualification test pressure may also be limited by the collapse pressure limit for the exposed short annulus pipe section. A seal test through a test port between the primary and secondary seal will in general separate the pipe and fitting and may exclude seal activation from differential pressure across the seal, contributing to a reduction in the seal contact pressure, and is hence more conservative than pressurization inside the pipe section. Basic tests and type tests are described above. For the sealing function, these may be performed as fullscale or small-scale tests as described below. C Full-scale tests A pressure test alone would be appropriate as a FAT of a seal system which has been qualified. However, the pressure test will not normally give sufficient assurance for a seal that has not been qualified. There are several ways of ensuring that the sealing of a gap has sufficient strength in maximum adverse conditions. ISO 10423, Annex F.1.11, which applies to wellhead equipment as well as engineering Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 133

134 judgement, forms the background for the following tests and acceptance criteria. Two approaches appear feasible: 1) cyclic test between the lowest and highest temperature and pressure for the number of maximum operating conditions 2) cyclic test between the lowest and highest temperature and pressure for the number of cycles required to show seal pressure stabilization. The latter item 2) seal pressure stabilization can be verified by stabilized extrusion, stabilized leak pressure (increase the seal test pressure until a leak occurs provided this does not harm the seal) or direct measurement of the seal pressure. The cycle test time as well as the number of cycles can thereby be established. Conservatism can be included by having: a maximum temperature increase (say an increase of 10 C) and a maximum pressure (say by a factor of 1.1 of the maximum seal pressure) a minimum pressure (say by a factor of 0.9 of the sealing pressure) and cycles (say 1.5 times the actual numbers). The first cycle with a high temperature should preferably have a longer duration to assure the stability, say three times the period required to indicate stabilized conditions. The following cycles could then be limited to one-third the duration of the first. The same exposure times used for the higher temperature limit exposure should be used for the subsequent lower temperature limit exposure. The number of test cycles can be terminated when three consecutive cycles do not show any changes, i.e. when stabilized conditions are confirmed. The former approach 1) is more uncertain with respect to the holding time per cycle; say three hours for the first cycle followed by one hour for the following high as well as low temperature exposure. The total number of test cycles should be equal to the predicted number of pipeline depressurizations multiplied by 1.5. C Small-scale tests It can be difficult and costly to perform the full-scale test for the conservative limit of all the contributing parameters. In that case, small-scale tests establishing limiting parameters can replace parts of the fullscale test. (Small-scale tests are in principle part of the basic tests). Small-scale tests can be used to establish: a correlation between the extrusion gap, seal pressure, seal strength and elasticity, temperature and time possible swelling caused by fluid exposure chemicals or a mixture of chemicals affecting aging seal material characteristics, such as a thermal expansion coefficient and volumetric (bulk) elastic modulus a seal friction coefficient. The setup for the small-scale tests can differ from that for the full scale tests, but must reflect the failure mechanism considered in a conservative manner, i.e. the failure must be allowed to develop in a similar way to the actual fittingin a conservative manner. It is an advantage to establish an analytical numerical model for the seal pressure which is validated by use of the results of the small-scale tests. This analytical model should calculate the seal pressure changes caused by the parameter variations. An actual extrusion will most probably take place only in a limited part of this circumference, reducing the pressure in this part. Ideally, such an analytical model could include the effects of pressure variations around the seal circumference and possible seal mass redistribution due to this. It is, however, assumed that, even if sophisticated, such an analytical model will have a wide spread of results compared to small-scale test results. Consequently, the conservative approach is advised: calculate the seal pressure reduction at the place of extrusion without including possible circumferential seal mass redistribution. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 134

135 The use of small-scale test results and an analytical model requires some higher safety factors compared to the full-scale test for the worst parameter combinations. A simpler analytical model could be applied to fully enclosed seals, since there would be no seal extrusion. The main concern would be to estimate possible volume changes caused by the seal thermal expansion and check that the enclosing materials have sufficient elasticity, thereby avoiding plastic deformations of the weaker parts of the seal support, and/or including this plastic deformation. This volume change would be caused by the elasticity/plasticity of the pipe itself and/or the anti-extrusion rings. In general, a full-scale test should be used to calibrate analytical tools. C.3.7 Locking tests The mechanical locking capacity as described in Appendix [B.1.6] shall be tested to document an acceptable margin to failure for relevant load combinations. Typically, the locking capacity should be tested for the make-up condition and, when relevant, if the locking is affected by the pipeline pressure. The evidence of achieved acceptable locking, such as grip marks on the pipe wall, activation displacement and/or load, shall be established and qualified by tests. C.3.8 Integration and subsea A subsea test shall be carried out. The test shall cover features with different effects (possible failure modes) from dry on-land testing, e.g. a possible water block. Alternatively, a relevant and representative track record may be used. The tests shall, as a minimum, include: installation activation a pressure test monitoring systems. Materials and material combinations with possible failure modes related to the subsea use shall be qualified by basic tests. Such features could be: 1) volume elasticity and water absorption properties of materials filling voids 2) swelling 3) electrical isolation 4) hydraulic systems' pressure-compensation systems. C.3.9 Examinations The fitting and pipe shall be thoroughly examined after the tests. This examination shall include: 1) examination for marks and measurement for permanent deformation of the pipe 2) examination of the fitting's internals, in particular the seals, if possible before disassembly (e.g. may not be possible for fittings designed for one activation only). 3) measurement of the fitting's critical dimensions 4) disassembly of the fitting and measurement of critical dimensions. The measurements shall be carried out with the same accuracy as indicated above. C.4 Factory acceptance tests C.4.1 Introduction The factory acceptance tests (FAT) checks the manufacturing of the fitting. It is, in principle, a spot check of only some aspects of the fitting. The aspects of concern are those related to possible errors in manufacturing. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 135

136 Some types of fittings are designed for only one activation, i.e. only for the actual pipeline connection. Testing of such single activation fittings is therefore limited to the basic tests, type test and testing after installation. The following information regarding FAT applies in general for fittings designed for possible reuse. C.4.2 Manufacturing The manufacturer s quality control shall verify that the materials, dimensional tolerances and make-up forces comply with the design documentation before assembly. Critical dimensional tolerances and the surface finish shall be measured with an accuracy of at least 1/10 of the prescribed tolerance band. Where relevant, the magnitude of bolt pre-tension etc. shall be recorded. C.4.3 Extent of factory acceptance test Fittings capable of activation, de-activation and re-activation shall be tested. The test shall follow a procedure with defined acceptance criteria. The acceptance criteria shall be documented by the qualification work. Factory acceptance tests can be carried out for nominal conditions with respect to dimensional tolerances, pressures and time. Typical tests for the connection of pipes are described below. C.4.4 Activation test The fitting shall be installed on pipes which are similar to the actual pipes with which the fitting is intended to mate. The key parameters shall be identified and recorded during activation and be within the prescribed limits. C.4.5 Pressure test The fitting installed on the test pipes with end closures shall be subject to a pressure test equal to the test pressure intended for the pipeline in addition to test requirements specified by the design standard. Pressure vessel standards require hydrostatic pressure testing, typically 1.43 times the design pressure, to verify the integrity. This test may be performed as part of the FAT or separately and pipeline stresses shall not exceed the criteria used in the mill test. Acceptance criteria shall be according to the nominated design standard and the holding time shall be the larger of the requirements given in the governing fitting design standard and [C.6]. Alternatively, the fitting can be tested to failure or according to the alternative design approach as detailed in [2.1]. C.4.6 Seal test The seals shall be subject to a seal leakage test via the annulus or installation tool. The seal test pressure level depends on potential failure mechanisms that are related to the installation and activation operation and the pressure level that is required to demonstrate acceptable performance of the seal(s) after installation with respect to the potential leak failure mechanisms. Typically, the seal test pressure for: seal welds is equal to the local incidental pressure,, and for polymer, graphite and metal seals is 1.1 to 1.5 times the local incidental pressure depending on the potential failure mechanisms related to the installation (pre-tension level, surface roughness and local unevenness, seal and seal surface damage during installation, seal robustness documented during the qualification phase). Leakage is not acceptable. Minimum holding times for pressure tests as a function of test volumes are given in [C.6]. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 136

137 C.4.7 Deactivation test The fitting shall be deactivated after testing. The key parameters shall be identified and recorded during the deactivation and shall be within prescribed limits. C.4.8 Examinations The dismounted fitting and the test pipe shall be examined to check that the appearances/tolerances are within the acceptance criteria, including: seals grips (connection area to pipe) marks on the pipe surface from seals and grips dimensional measurements for possible plastic deformations of sensitive coupling internals dimensional measurements for possible plastic deformations of the pipes. Any possible need for the replacement of parts, e.g. seals, following activation/deactivation shall be recorded. C.4.9 Insufficient type tests The FAT can be combined with type tests in the case of incomplete separate type tests. Where such a combination is applied, the FAT must be extended to include the requirements relevant for the incomplete type test(s). C.5 Installation verification tests C.5.1 Introduction Final tests shall verify that the completed installation complies with the prescribed criteria. In some cases, final testing consists only of a leak tightness test. However, the verification of the completed installation often also comprises the monitoring and recording of parameters which are important for assurance of the prescribed criteria. TV or sensors are required as applicable to perform such monitoring. The following describes typical installation testing. C.5.2 Measurements, monitoring and recording The measurement or monitoring of the limiting parameters can provide assurance that the fitting is installed within its limits. These limits are normally: 1) pipe conditions with respect to surface conditions and, if applicable, the end-cut 2) pipe alignment and alignment of the coupling relative to the pipe ends prior to installation 3) pipe gap between ends for couplings 4) contamination monitoring and control to avoid seal and locking failure 5) displacement control of the fitting during installation and control to avoid excessive forces 6) the position of the fitting relative to its intended and possible limiting position on the pipe, e.g. pipe ends 7) the monitoring/control of activation displacements/forces to assure activation within limits. C.5.3 Testing Sealing shall be tested to the qualified pressure and holding time, with testing either at the annulus or via an installation tool, see [C.4.6] for more details. The time depends on the stabilization period due to the size and length of the pipe work connected with the test. Normally, much less time than that allocated to the pressure test of the pipeline itself can be allowed, due to the small volume which is pressure-tested. Minimum holding times for pressure tests as a function of test volumes are given in [C.6]. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 137

138 Seals which are not tested after installation shall be qualified for this purpose, i.e. to have a sufficient small risk of leakage. Further, they shall be checked for leakage in a pipeline leak test. C.5.4 Dismounting 1) Temporary connections for control and monitoring shall be sealed off after disconnection. The sealingoff integrity shall be verified by appropriate means, depending on the consequences of a leak through the seal-off. The verification method shall be part of the qualification. 2) Forces applied to the connection after make-up and testing due to the final pipe manipulation shall be controlled within the connection limitations. C.6 Holding time (minimum) for leak testing Leak tightness tests are to be performed with the applicable test pressure determined by the type of part and design pressure and held for a specified time to determine that the part is free from leaks. The rate at which the pressure of a leaking chamber decreases is related to the volume of the test container. The recommended holding times are therefore categorized as follows according to test volume. The pipeline shall be free from leaks and the pressure variation shall be within 0.2% of the test pressure. Pipelines with a volume greater than 5,000 m 3 should be leak tested using water and shall be held for a minimum of 24 hours after a stabilization period. Correcting for pressure-test volumes, the recommended minimum holding times for the other categories are given in Table C-1 for water and Table C-2 for gas. Table C-1 Recommended minimum leak holding times for water as a test medium Category Volume Holding time Pipelines m 3 24 hrs Short pipelines m 3 8 hrs Local repair section m 3 2 hrs Assemblies 1-10m 3 15 mins Components 0.1-1m 3 15 mins Back-seal tests m 3 15 mins Table C-2 Recommended minimum leak holding times for gas as a test medium Category Volume Holding time Pipelines m 3 - Short pipelines m 3 - Local repair section m 3 - Assemblies 1-10m 3 8 hrs Components 0.1-1m 3 2 hrs Back-seal tests m 3 15 mins Note: for unsupported seals retained against axial displacement by friction against the pipe wall and fitting surface (e.g. inflated- and tongue seal, without any groove or equivalent), any additional holding time to document acceptable resistance against slippage needs to be assessed. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 138

139 APPENDIX D STRESS ANALYSIS OF FILLET WELD D.1 Assumptions The same equations are used for the design of fillet welds in the NORSOK Standard N-004 Design of Steel Structures and DNVGL-ST-C101. Some additional guidance on the design of welded connections is given in DNVGL-ST-C101, section C600 Direct calculation of weld connections : 601 The distribution of forces in a welded connection may be calculated on the assumption of either elastic or plastic behaviour. 602 Residual stress and stresses not participating in the transfer of load need not be included when checking the resistance of a weld. This applies especially to the normal stress parallel to the axis of the weld. 603 Welded connections shall be designed to have adequate deformation capacity. 604 In joints where plastic hinges may form, the welds shall be designed to provide at least the same design resistance as the weakest of the connected parts. 605 In other joints where deformation capacity for joint rotation is required due to the possibility of excessive straining, the welds require sufficient strength so that they do not rupture before general yielding in the adjacent parent material. D.2 Sectional forces Below, it is assumed that the design of the fillet weld is based on the results of an axisymmetric finite element analysis. The stress distribution at a section through the weld that is normal to the weld surface is considered (see Figure D-1). C σ ae τ e C D a B M s Q s F s e M B p A σ be F p Figure D-1 Elastic stress distribution at a critical section The calculated circumferential stress in the weld is neglected in accordance with [D.1]. The axial force in the weld normal to the considered throat section can be derived as: N = a / 2 σ ( x) dx n a / 2 (D.1) The bending moment over the throat section can be derived as: M s = a / 2 σ ( x) n a / 2 x dx (D.2) The shear force acting on a section through the considered weld can be obtained as: Q = a / 2 τ ( x) dx e a / 2 (D.3) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 139

140 D.3 Design of the fillet weld Reference is made to Figure D-2 for a design check of the fillet weld. C σ τ ap p C b c b B D M s Q s F s e M B p A σ bp F p Figure D-2 Stress distribution at a critical section for a capacity check of the weld The throat section B-C is considered. As a first approximation: σ bp = f d (D.4) Then the moment capacity is calculated as: σ b ( a b) = bp M s (D.5) From this equation, the distance b required for moment capacity is determined. Then the shear stress is calculated as: Q τ p = b + c Then an improved value of b may be derived from the moment capacity where the maximum allowable axial stress is corrected according to von Mises using a mean shear stress: (D.6) σ bp f b ( a b) = M 2 d 3τ 2 p s (D.7) And an improved shear stress may also be obtained as a second iteration from the equation. Then the axial stress is obtained as: σ ap = N a 2b (D.8) The final check of the capacity for region c is made by σ 2 + τ ) ap 3 2 p f d (D.9) Similar design checks are performed along sections A-B and B-D. The design allowable stress is given by: =, (D.10) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 140

141 APPENDIX E DESIGN RESISTANCE; WELDING ON A PIPE IN OPERATION General Welding onto an in-service pipeline is used to facilitate a repair, such as the installation of split sleeve repair clamps, or to install a branch connection using the hot tapping technique described in Sec.9. The main risks related to welding on in-service pipelines that must be considered are: the risk of burn-through or blowout as a function of heat input, covered by this appendix the risk of hydrogen-induced cracking due to high HAZ hardness, hydrogen content and root notches, covered by Sec.11. if welding on a leaking pipe with a leak clamp temporarily installed, there is the risk of a gas leak into the habitat. This needs to be controlled by using a gas containment barrier and inert gas purging or double bleed seal leak monitoring. A burn-through, or blowout as it is sometimes referred to, will occur when welding onto a pressurized pipe if the unmelted area beneath the weld pool has insufficient strength to contain the internal pressure. A burnthrough typically results in a small pin-hole leak at the bottom of the weld pool. The risk of burn-through will increase as the pipe wall thickness decreases and the weld penetration increases. The burn-through resistance at the weld melt pool when welding on a pressurized pipeline can be documented by different methods, and the required level of advanced analysis depends on the confidence in governing factors such as structural utilization, heat input and material specifications. The two primary methods used by the industry today to document acceptable burn-through resistance at the weld melt pool are: the Battelle model /11/ the PRCI thermal analysis model /12/. These computer model methods predict inside surface temperatures as a function of the welding parameters (current, voltage, and travel speed), geometric parameters (wall thickness, etc.) and the pipeline operating conditions (contents, pressure, flow rate, etc.). The risk of burn-through for a given application can be evaluated based on the various contributing factors /13/. Below is given a screening approach applicable for all C-Mn and CRA pipeline materials, excluding any copper alloys, to document acceptable burn-through resistance at a weld melt pool when welding on an inservice pipeline. Screening criterion approach The DNV GL Report No An improved CWM platform for modelling welding procedures and their effect on structural behaviour, /8/, provides an analytical algorithm for calculating the weld cooling time during in-service welding operations and procedures for documenting the acceptable margin against burst when welding on a pressurized pipeline in operation. The study /8/ documents that the local burst capacity resistance at the weld melt pool is intact/acceptable for in-service welding on pipelines provided the weld heat flow is in a three-dimensional (3D) condition, as described below. In a 3D weld heat flow condition, the weld heat transport (i.e. weld cooling ratio) is not affected by the pipeline's internal fluid flow, as illustrated in Figure E-1. The weld cooling time Δt T1/T2 in a 3D weld heat flow condition is analytically calculated by Rosenthal and Rykalin s 3D solution: 3D condition: 1 / 2 = (E.1) Where: Δt T1/T2 = weld cooling time between the temperatures T 1 and T 2, [s] Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 141

142 QW = η U I / v = weld heat input, [kj/mm] k = thermal conductivity, W/m 2 C η = efficiency factor (η = 0.8 for all arc weld processes except submerged arc welding, where η = 1.0) T 0 = temperature before start of welding (T 0 = 10 C is used in Figure E-2 below) T 1 = temperature at which phase transformation during cooling starts (typically 800 C for carbon steel) T 2 = temperature at which phase transformation during cooling ends (typically 500 C for carbon steel) Figure E-1 Cross-section view of a weld heat flux in a semi-infinite thick solid plate (3D heat flow condition) where the base material is the pressure-containing pipe wall In the case of in-service welding where the weld cooling time Δt T1/T2 (i.e. weld cooling ratio) is affected by the pipeline s internal fluid flow (illustrated in Figure E-2); the pipeline s burst capacity at the weld melt pool is reduced, and the acceptable burst resistance depends on the local pipe wall heat transport (thermal boundary layer) by the internal fluid flow. Figure E-2 Cross-section view of the weld heat flux transfer from a plate (solid base material) into a flow of fluid where the base material is the pressure-containing pipe wall. The pipe wall thickness δ 3D is the minimum pipe wall thickness where the weld cooling rate is not affected by the heat transport (i.e. weld cooling ratio) from the pipeline s internal fluid flow, according to equation (E.2) (E.2) Where δ 3D = pipe wall thickness for the considered 3D condition, [m] Q W = η U I / v = weld heat input, [kj/mm] η = efficiency factor (η = 0.8 for all arc weld processes except submerged arc welding, where η = 1.0) U = welding input voltage, [V] Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 142

143 I v = welding input current, [A] = weld head travel velocity, [m/s] T 0 = temperature before start of welding (T0 = 10 C used in Figure E-3) T 1 = temperature at which phase transformation during cooling starts (typically 800 C for carbon steel) T 2 = temperature at which phase transformation during cooling ends (typically 500 C for carbon steel) ρ = pipe material density, [kg/m 3 ] c = pipe material specific heat, [J/kg C] Note: Δt T1 / T2 is the time it takes for the weld seam and adjacent heat-affected zone to cool from 800 C to 500 C. This is the interval in which the most important structural changes occur in the steel. If the period is very short, ferritic, perlitic, bainitic and martensitic structures can form. There is also the risk of coarse grain formation and this has negative effects on the steel's mechanical properties. If the period is very long, only ferrite and perlite can form from the austenite, and this reduces the hardness. Equation (E.2) is plotted in Figure E-3: conservatively assuming no internal fluid flow heat transport using a reference temperature of 10 C based on carbon steel density of 7850 kg/m 3 and a material specific heat capacity of 460 J/ C. By calculating the weld heat input Q w (defined above as η U I / v), the minimum acceptable pipe wall thickness with respect to burst capacity for the 3D heat flow condition is found. Wall thickness above the plotted curve for a given heat input is acceptable. Figure E-3 The diagram indicates when it is appropriate to use a 3D heat flow approximation (i.e. above the plotted curve) during the in-service welding of a steel pipeline at 10 C (ρ = 7850 kg/m3 ; c = 460 J/ C), as demonstrated by equation (E.2) above. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 143

144 APPENDIX F CALCULATION EXAMPLE - MECHANICAL COUPLING AXIAL LOCKING CAPACITY DNVGL-ST-F101 Section 5 part D specifies limit states to be used in the pipeline system design. For pipeline repair couplings, the following limit states should be considered for the locking capacity. ULS (ultimate limit state) SLS (serviceability limit state) FLS (fatigue limit state) Pressure containment Relative displacement Cyclic loading Pressure containment Relative displacement Cyclic loading/fatigue One of the main failure modes for pipeline repair couplings and clamps is separation due to tension loads induced by the internal pressure. Separation due to pressure induced loads may be considered in a similar manner as pressure containment for submarine pipelines. However, the pressure containment criterion in DNVGL-ST-F101 supposes system effects, i.e. the entire pipeline system is considered. The design of pipeline repair couplings/clamps is a local design, where system effects are not relevant. Therefore, the load factors detailed in this RP are based on the load effect factor combination in DNVGL-ST-F101 Section 4 G300;, &. For a pipeline repair coupling, failure of the ULS condition would mean that the pipe end is pulled out of the coupling's gripping and sealing components, leading to loss of containment. The ULS condition shall in general be the base case for the design and the locking capacity (i.e. gripping capacity) shall exceed the ULS design axial load by, Some couplings are designed for the SLS condition, which is defined as loss of the abutment of pipe(s) end, leading to the potential relative movement of the pipe under the seals, but with no loss of containment. In an SLS condition, the seals need to be qualified for this movement. For couplings designed with a pipe-end abutment within the coupling (see Figure F- 2 below), the capacity shall exceed the SLS design axial load by, Fatigue is normally not an issue for repair couplings, but needs to be assessed on a caseby-case basis. The following figures show: details of a typical pipe-coupling abutment details of a pipe/pipe-end abutment within a coupling details of a coupling without a pipe-end abutment. Figure F-1 Details of a typical pipe-coupling abutment (over seal activation ) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 144

145 Figure F-2 Details of a pipe/pipe-end abutment within a coupling Figure F-3 Details of a coupling without a pipe-end abutment Design loads (axial) for fitting grip capacity: Tensile axial load: generally the highest tensile loads are given for an unrestrained pipe section and this case shall always be considered. Under certain circumstances, larger axial tensile loads can also occur during the cool-down of partially restrained pipelines that have globally deflected or buckled. Compressive axial load: pressure and in particular temperature lead to compressive loads for restrained pipes. The highest compressive loads are obtained for a fully restrained pipeline. Nevertheless, compressive loads for hot pipelines are generally limited by global buckling to lower levels than given by full restraint and these need to be obtained through a global analysis of the relevant section of the pipeline. Design equations for axial loads are derived from DNVGL-ST-F101. The relevant force acting on the coupling which the gripping capacity shall be designed for is the axial force in the pipeline. This axial force is often called the wall or true wall force (as opposed to the effective axial force). = ( ) + The design equation for the axial load in a pipe wall (N) in the unrestrained condition is: For the restrained pipeline condition, the effective axial force is: = (1 2 ) Guidance note: The equation for the restrained pipeline condition gives the highest and most conservative estimate of the compressive design load. This equation may be used if information on compressive loads is not detailed or if the global buckling load has not been established. For most couplings, compressive design cases may not need to be considered provided the coupling is at least as strong as the pipeline and the pipe abutment is within the body of the coupling. ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- The installation of mechanical couplings supposes that the pipeline has been cut and the damaged pipe section removed, giving zero effective axial force during the repair. The external pressure,, due to water Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 145

146 depth will then apply both internally and externally and this represents the initial state for testing and operational loads: = 0 = ( ) 0 = Due to the split between the pipe ends within a coupling (even if they are abutted), the internal pressure acts on the pipe ends and creates an axial load which shall be included in the coupling gripping design load. The axial load is given by ( ), where is the area up to the outer steel diameter (or up to the seal groove located on the coupling). For the unrestrained condition, FEA analysis shall take into account the end cap based on (Pi Ai) in addition to internal pressure applied to the coupling/pipeline internal area. Equations to calculate the repair coupling design loads are given below: Internal pressure load to be carried by the coupling gripping: Internal pressure load to be applied to an FE model: ( ) ( ) The design axial pipe wall force for the unrestrained condition is: = The terms, including S F, S E and S A (i.e. effective axial pipeline force contribution from the functional, environmental and accidental loads), can be disregarded in most cases and the following design load applies: = ( ), for operation = ( ), for installation/test Where and are the local incidental pressures as detailed in DNVGL-ST-F101 and include safety factor. For the operational condition, is 1.0, while for the testing this load safety factor is The unrestrained condition, as defined in [5.2], will often represent the highest tensile load to be designed for, giving a characteristic pressure induced axial load of: = ( ) Combining the characteristic axial load with the load safety factors given in Table?63 gives the gripping capacity design load, L d, for the coupling: Where: Type of factors Ref. to DNVGL-ST-F101 = During repair and testing During operation Comments Functional loads γ F Includes trawl interference Environmental loads γ E Pressure loads γ p Together with p li (operation) or p lt (testing) Condition load effect γ c DNVGL-ST-F101 specifies γ C =1.07 for uneven seabed. However, for pipeline repair, no additional factor of 1.07 is to be applied for uneven seabed. Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 146

147 Required design capacity: Pipeline repair components that will be integrated should normally be designed according to the ULS condition. In addition, pipeline mechanical couplings designed with pipeline abutment should be designed and qualified for the SLS condition, documenting the acceptable robustness of the seals for the specified cyclic pipeline movement. Resistance factors for the design condition. Safety class Limit state Safety factor (ULS) Low Medium High ULS and SLS ULS, SLS , The required resistance, R d, is given by the characteristic resistance divided by the material- and safetyclass safety factors given in Table 6-3: = (, ) Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 147

148 APPENDIX G PIPELINE RISK ASSESSMENT AND FAILURE STATISTICS G.1 Introduction This appendix presents an overview of analysis methods for pipeline risk assessment and their relationship to pipeline failure statistics that could be used as a basis for the pipeline operator establishing the preparedness strategy and spare part and equipment philosophy. G.2 Risk and probability of pipeline failure A pipeline failure has occurred when a structure or component can no longer safely continue to meet the criteria for which it was designed. This includes one or more of the following effects: loss of component or system function deterioration of functional capability to such an extent that the safety of the installation, personnel or environment is significantly reduced disruption of normal use due to structural damage. A structural failure may, for instance, be defined as a loss of serviceability affecting operability or a major loss of containment. The definition of failure in a pipeline risk analysis can generally be classified into two categories: Loss of integrity: an undesirable event which will compromise the pipeline integrity although not necessarily imply imminent structural failure. Failure criteria in design standards are in this category. These are typically formulated as a violation of a theoretical limit state (e.g. exceedance of calculated structural capacity), implying a loss of structural integrity. Loss of containment: failure defined as a loss of containment (leakage or rupture) is used in risk analysis where the focus is on the consequence of such an event. The failure data used in quantitative risk assessment (QRA) is based on this definition of failure. Figure G-1 Chain of events and definitions of failure The relationship between the definitions of failure can be illustrated by the chain of events shown in Figure G-1. Due to some initial cause, the pipeline may be exposed to a load that results in loss of integrity, e.g. significant deformations of the cross-section. The criteria used in design standards interpret this event as a failure, although the pipeline may still be fit for continued operation. The pipeline is here in an unwanted (damaged) condition, so that the probability of loss of containment increases. Guidance note: The loss of containment definition is typically less stringent than any loss of integrity criterion, in the sense that loss of integrity will occur before (or simultaneously with) loss of containment. The definition of failure in design standards is usually in the form of loss of integrity, as such criteria can be established based on theoretical models, e.g. by structural analysis of loads and load effects. It is not always practical to determine the exact conditions for loss of containment using this approach. Pipeline failure databases generally consider only loss of containment failures, which is why assessments relying on historical data make use of this definition of failure. Moreover, whereas loss of containment is clearly defined, there is a wide variety of loss of integrity criteria. Many of these make use of quantities that cannot be detected or observed (e.g. forces). ---e-n-d---of---g-u-i-d-a-n-c-e---n-o-t-e--- Recommended practice, DNVGL-RP-F113 Edition November 2016 Page 148