Bolted Joints Integrity Management Guidelines

Guidelines for the management of the integrity of bolted joints for pressurised systems

2nd edition

IP An IP Publication

GUIDELINES FOR THE MANAGEMENT OF THE INTEGRITY INTEGRITY OF BOLTED JOINTS FOR PRESSURISED SYSTEMS

GUIDELINES FOR THE MANAGEMENT OF THE INTEGRITY INTEGRITY OF BOLTED JOINTS FOR PRESSURISED SYSTEMS

Page Blank In Original

GUIDELINES FOR THE MANAGEMENT OF THE INTEGRITY OF BOLTED JOINTS FOR PRESSURISED SYSTEMS

June 2007 Second edition

Published by ENERGY INSTITUTE, LONDON The Energy Institute is a professional membership body incorporated by Royal Charter 2003 Registered charity number 1097899 Endorsed by Oil & Gas UK, HSE OSD and the ECITB

The Energy Institute gratefully acknowledges the financial contributions towards the scientific and technical programme from the following companies: BG Group BHP Billiton Limited BP Exploration Operating Co Ltd BP Oil UK Ltd Chevron ConocoPhillips Ltd ENI ExxonMobil International Ltd Kuwait Petroleum International Ltd Maersk Oil North Sea UK Limited

Murco Petroleum Ltd Nexen Saudi Aramco Shell UK Oil Products Limited Shell U.K. Exploration and Production Ltd Statoil (U.K.) Limited Talisman Energy (UK) Ltd Total E&P UK plc Total UK Limited

Copyright © 2007 by the Energy Institute, London: The Energy Institute is a professional membership body incorporated by Royal Charter 2003. Registered charity number 1097899, England All rights reserved No part of this book may be reproduced by any means, or transmitted or translated into a machine language without the written permission of the publisher. The information contained in this publication is provided as guidance only and while every reasonable care has been taken to ensure the accuracy of its contents, the Energy Institute cannot accept any responsibility for any action taken, or not taken, on the basis of this information. The Energy Institute shall not be liable to any person for any loss or damage which may arise from the use of any of the information contained in any of its publications. The above disclaimer is not intended to restrict or exclude liability for death or personal injury caused by own negligence.

ISBN 978 0 85293 461 6 Published by the Energy Institute Further copies can be obtained from Portland Customer Services, Commerce Way, Whitehall Industrial Estate, Colchester CO2 8HP, UK. Tel: +44 (0) 1206 796 351 email: [email protected] Electronic access to EI and IP publications is available via our website, www.energyinstpubs.org.uk. Documents can be purchased online as downloadable pdfs or on an annual subscription for single users and companies. For more information, contact the EI Publications Team, e: [email protected]

CONTENTS Page Foreword Acknowledgements

vi i ix

1

Introduction

1

2

Bolted joint technology and practice

3

2.1 2.2 2.3 2.4 2.5 2.6 2.7

3

3

4

5

Overview Types of bolted joints Bolted pipe joint components Principles of joint assembly and disassembly Controlled tightening of joints Bolted joint reliability Integrity testing

3 4 6 8 10 14

Criticality assessment 3.1 Introduction 3.2 Assessing the risks with bolted joints

17

Training and competence 4.1 Introduction 4.2 Competence management 4.3 Training 4.4 Ongoing competence 4.5 Training in engineering construction skills (TECSkills) 4.6 Vocational qualifications 4.7 Independent accreditation organisations

21

Records, data management and tagging 5.1 Joint identification 5.2 Records and data management 5.3 Review

27

17 17

21 21 21 22 22 22 22

27 28 29

Page

Contents Cont.... 6

7

Management of leaks 6.1 Introduction 6.2 Engineering risk assessment of leaks 6.3 Stages at which leaks occur 6.4 Corrective actions 6.5 Definition and detection of leaks 6.6 Managing leaks and repairs 6.7 Learning from leaks

31

In-service inspection 7.1 Introduction 7.2 Risk assessment 7.3 Degradation mechanisms 7.4 Inspection techniques 7.5 Defect mitigation measures

37

31 31 32 32 32 34 34

37 37 37 37 38

VI

FOREWORD The first edition of this publication was commissioned by the upstream oil and gas industry as part of an HSE/industry drive to reduce the incidence of hydrocarbon releases. Leaking joints have been the main cause of hydrocarbon releases on UK Continental Shelf offshore installations. Similar concerns exist for many onshore facilities handling petroleum and other dangerous substances. In 2005, the UKOOA (now Oil & Gas UK) led Installation Integrity Working Group (IIWG) requested that the Energy Institute manage the review and revision of the UKOOA/IP Guidelines for the management of integrity of bolted pipe joints, first published in June 2002. This project required the formation of a cross-industry Work Group (WG), which included some members that also contributed to the first edition. Other members were drawn from the parent IIWG, consultants and from the industry training organisation, ECITB. This updated publication supports the principles of the IIWG in making available good practice on key integrity issues and is referenced in Oil and Gas UK Hydrocarbon Release Reduction Toolkit, which itself is referenced in the Oil and Gas UK Asset Integrity Toolkit. During the review process, the WG elected to widen the scope to include bolted joints used within pressurised systems and not just pipe joints as is the case for the first edition, and to ensure that the publication is applicable both to onshore facilities as well as offshore oil and gas installations. This publication provides guidance only and is intended to improve knowledge of good practice which should assist operators to develop their own management systems for bolted pipe joints. While every reasonable care has been taken to ensure the accuracy and relevance of its contents, the Energy Institute, its sponsoring companies, section writers and the WG members listed in the Acknowledgements who have contributed to its prep aration, cannot accept any responsibility for any action taken, or not taken, on the basis of this information. The Energy Institute shall not be liable to any person for any loss or damage which may arise from the use of any of the information contained in any of its publications. This publication will be reviewed in the future and it would be of considerable assistance for any subsequent revision if users would send comments or suggestions for improvements to: The Technical Department, Energy Institute, 61 New Cavendish Street, London W1G 7AR e: [email protected]

vii

Page Blank In Original

viii

ACKNOWLEDGEMENTS As Work Group members, the Institute wishes to record its appreciation of the work carried out by the following: Sub Group Champions, who have managed the coordination and compilation of designated sections through leadership of their respective volunteer sub-groups and through providing authorship expertise: Stuart Brooks Rod Corbet Anderson Foster Jim MacRae Robert Noble

BP Exploration Operating Company Ltd. Rotabolt Total E&P UK plc Nexen Petroleum UK Ltd. Hydratight

Sub Group members, who have provided valued input into their designated sections: Blair Barclay Keith Dunnett Bill Eccles Alan Gardner Tim Jervis Gary Milne Phillip Roberts Ravi Sharma Mike Shearer Lawrence Turner Mark Williams Pat Wright

ECITB CNR International Bolt Science (Hytorc) Consultant Shell Exploration & Production Hydratight Shell Exploration & Production HSE Lloyds Register EMEA Shell Exploration & Production Klinger UK Ltd RGB Ltd.

Assistance was also provided by the following other Work Group members: Gwyn Ashby Peter Barker Arunesh Bose Martin Carter Kevin Fraser Norrie Hewie Gavin Smith Roy Smith Jan Webjorn

Mitsui Babcock Marathon Oil Lloyds Register EMEA BHP Billiton IMES Hess Corporation Novus Sealing Hytorc Verax

Liaison with other organisations was provided by: EEMUA IMechE Pressure Systems Group

Andrew Pearson Chris Boocock

Oil & Gas UK (formerly UKOOA)

Bob Kyle

Technical authorship and editing: Phil Smith

ODL

The revision/review project was coordinated and managed by Keith Hart FEI, Energy Institute, Upstream Technical Manager. The Institute also wishes to recognise the contribution made by those who have provided comments on the Draft document which was issued during an industry consultation period.

1 INTRODUCTION A bolted joint is one of many critical components of a pressurised system. Dependent upon the contents and pressure of the system, leakage or failure of a bolted joint can have potentially catastrophic consequences. To meet this challenge, every operator of pressurised systems should have in place a system to positively and actively manage the integrity of bolted joints. It is expected that such a system will be built around the principle of continuous improvement (see Figure 1.1). This document describes the principles and good practice for the establishment of a management system

for bolted joints in pressurised systems. Individually the sections of this document provide details of what is considered good practice in the key areas of ensuring joint integrity. Together they provide the framework for a management system. This document is not intended as a design guide for bolted joints, but as a guide to how to manage joints during construction and commissioning phases and through their operational life. It provides a framework to achieve this based on working with a correctly designed joint.

Figure 1.1: Essential elements of a management system

The following are considered essential elements of a management system which must be applied to ensure that the system is implemented and remains effective: — Ownership There should be an identified owner of the management system, responsible not only for its implementation and ongoing maintenance, but also for communicating its aims and objectives throughout the organisation. The owner should state the expectations for the system and monitor its effectiveness. — Technology and Practice Good practice with regard to selection and control of assembly, tightening and assurance of bolted joints should be applied. Understanding of the theory and practice of bolted joints and development of appropriate procedures should be encouraged throughout the organisation. — Criticality Assessment The range of services, pressures and conditions which bolted joint s experience varies considerably. Each joint should undergo a criticality assessment which will determine the levels of inspection, assembly control, tightening technique, testing, assurance and in-service inspection relevant to the joint. — Training and Competence Everyone with an influence on joint integrity in the organisation should be aware of the management system, its objectives, expectations and effects on project planning and day-to-day working. Good awareness needs to be maintained. Any staff working on bolted joints should be appropriately trained and competent. — Records, Data Management and Tagging The certainty of achieving joint integrity increases if historical data exist on the activities carried out

in the past, ideally from original construction of the joint, linked to the design specification of the joint. Providing and recording traceable data encourages best practice at the time of the activity, and will provide useful planning data for the next time the joint is disturbed. In-service Inspection Learning from both positive performance and incidents is important. A management system should include the means for gathering relevant data on joints which are successful and those that have incidents or leakage issues. These should be collected by everyone involved in bolted joints , and periodically reviewed and analysed to establish trends, issues and improvement opportunities. Management of Leaks The objective of a correctly designed and installed bolted joint is to provide a long-term tight seal and prevent ingress or egress of fluids through the joint. However, leaks can occur and managing the investigation and repair of the leak is essential to avoid recurrence. It can also provide useful data for prevention on other projects. Analysis, Learning and Improvement Analysis of leakage and inspection data coupled with formal reviews of the management system should occur at agreed intervals by the owner and users. Results obtained from commissioning, incident analysis and in-service inspections should be used to generate ideas for continuous improvement. Easily monitored but meaningful performance standards should be put in place at launch to quantify the contribution being made by the management system and evaluate u ser satisfaction. Feedback on good practice in integrity issues and causes and solutions to incidents should be provided both internally and to industry to contribute to continuous improvement.

BOLTED JOINT TECHNOLOGY AND PRACTICE 2.1 OVERVIEW

This section gives a brief outline of how joint s work and provides guidance on the safe and efficient assembly and disassembly of flanged joints and clamps. It also discusses basic proposals for integrity testing. The scope of these Guidelines covers all pressure-containing joints including pipelines, pressure vessels such as reactors and heat exchangers, associated valves and other pressure-containing equipment. Due to operating conditions with heat exchangers and reactors, particularly temperature gradients, different metal joint components and thermal and pressure cycling, a higher level of control and assurance of bolt load is generally required compared to, for example, piping joints subjected to static pressure only. The principles set out are generic in nature and not exclusive to pressure containment applications; they can be applied to bolted joints subjected to other service conditions such as fatigue, vibration and structural loading. The flanged joint is deceptively simple yet, in common with the welded joint, its integrity relies on a number of parameters including the basic design, structural and metallurgical quality of its components and achieving the required design clamp force on assembly. Important to meeting these assembled design objectives is the selection of suitable installation procedures and tools that are applied by competent operators. The importance of planning the joint assembly, preparation of all components, procedures, tooling and ensuring application of the correct methodology is essential.

Pipework and pressure systems are designed to meet varying operational conditions. In order to avoid failure, it is very important that the relevant piping specifications for materials and components are adhered to in full. There are many types of bolted joint and only some of the more commonly used are mentioned here but as mentioned previously, the basic reliability parameters and procedures applied are the same for all.

2.2 TYPES OF BOLTED JOINTS 2.2.1

Flange joints

The most common type of joint is made up of two pipe flanges to ASME B16.5 design code, between which a gasket is compressed by the installed bolting. Similar arrangements are used for other codes such as API 6A, BS 1560 and MSS SP 44. The piping material specification will detail the codes and materials to construct the facility. The principle of a bolted joint is based on the bolting delivering sufficient joint compression and gasket seating stress to withstand maximum service pressure and forces. This is when the bolting is under tensile load as illustrated in Figure 2.1 . For integrity a minimum level of operational gasket seating stress must be maintained throughout joint service, therefore the design bolt load/compression target on installation should allow for creep, relaxation, uncertainty over service loadings and the tolerances of components and tools used.


Possible flange bending or "rotation"

Operating Gasket Stress

Figure 2.1: Working principle of bolted flange joints

2.2.2

Compact flanges

2.2.3

Various types of compact flanges have been developed by specialist manufacturers. Some use gasket arrangements similar to the metallic ring joint whereas others use metal to metal, gasketless contact and the joint becomes a static entity with minimal flange rotation potential. Such compact flanges tend to be characterised by the sealing area being positioned closer to the pipe bore thereby reducing bolt and working load eccentricity and subsequent end load on the bolts. This is a preferred bolted joint design feature and can result in smaller, lighter flange sizes and a reduction in bolt diameter, quantity or strength grade. The design philosophy can vary from type to type so the manufacturer should always be consulted for advice on joint sealing, design bolt tension and installation procedures.

Hub Clamp

Figure 2.2: Clamped connector

Clamped connectors

Clamped connectors (see Figure 2.2) use a split clamp to join the pipe. Hubs at the ends of the pipe have tapered shoulders sloping towards the joint and the clamps have tapered faces, which form a wedging action to close the two hubs together. The hubs have internal sloping faces which bear on taper ring gaskets , causing them to be distorted elastically and form a seal.

2.3 BOLTED PIPE JOINT COMPONENTS 2.3.1

Flanges and clamped connectors

Like pipes, flanges and clamped connectors operate under varying conditions of temperature and pressure. The most critical area on a flange or clamped connector is its sealing face, on which the gasket or seal ring seats to form a pressure retaining seal (see Figure 2.3 on page 7). It is therefore imperative that the sealing face's surface finish complies with the design specification or the manufacturer's recommendations. It must be protected at all times and free from damage, grease and protective coatings. On ASME B16.5 type flanges, the nut seating area at the back of the flange must be clean and of a smooth finish to reduce friction unless stated in the manufacturer's specification. Flanges, blinds and flange facings should be in accordance with the relevant flange code or manufacturer's proprietary requirements. Flanges are marked to identify the size, pressure rating and flange material, as shown in Figure 2.4 on page 7. The pipe schedule used with the flange should

BOLTED JOINT TECHNOLOGY AND PRACTICE

the double jacketed gasket is intended to go against this face; this is an important assembly feature.

also be marked. Corresponding bolts and nuts also carry material identification marking. These should conform to the relevant fastener specification. 2.3.2

Gaskets and seal rings

Correct gasket or seal ring selection and installation are important and only those specified in the piping material specification should be installed. The gasket creates the seal between the two flange faces and contains the internal pressure of the joint. As with flanges, gaskets and seals can be marked to identify principal characteristics, as shown in Figure 2.5 on Page 7. There are three main types of gasket: non-metallic, semi-metallic and metallic. Application selection is dependent on service conditions. 2.3.2.1 Non-metallic These are made from elastomers, cork, compressed fibres, plate minerals and PTFE. Usually the material sheet is cut to the shape of the flange sealing face. They are generally used for low to moderate pressures and temperatures and see wide chemical service including acid and alkaline applications. 2.3.2.2. Semi-metallic These combine a combination of non-metallic filler for compressibility and metal for strength. They are typically used for higher temperature and pressure applications compared to the non-metallic types. Common types include: — Spiral wound These gaskets are constructed with spirally wound metal and soft filler (see Figure 2.6 on page 7). A wide range of metals can be used for the winding strip and support rings as well as various filler materials. On raised face flanges, the gaskets have an outer support ring which locates inside the bolt PCD. They can also be supplied with an inner ring for higher pressure system usage. The inner ring is also used where high process flow rates or abrasive media are found; the inner ring reduces turbulence at the pipe bore. On spigot or recess flanges a simple sealing element gasket is used with no additional support rings. — Metal jacketed These clad gaskets have been traditionally used on heat exchangers. A variety of metals can be used to encase a soft filler material. It should be noted that some heat exchanger flanges have stress raising 'nubbins' on one face and the non-seamed face of

— Kammprofile This is a solid metal ring having a serrated tooth form profile on both faces. A covering layer of graphite or PTFE is applied which compresses into the serrated surface as the gasket is loaded. These are used increasingly for heat exchanger flanges (see Figure 2.5 on page 7). 2.3.2.3 Metallic These are made from one or a combination of metals in a variety of shapes and sizes for high temperature and pressure usage. The metal ring fits into grooves that have been machined into the flange faces. Due to the high application pressures, the seating stresses and corresponding bolt tension are necessarily large to give sufficient deformation to overcome flange surface imperfections and distort against the groove surfaces so as to overcome high service pressures. Oval and octagonal types (see Figure 2.7 on page 7) are commonly used in oil and gas applications under ASME B 16.20 and API 6A. RX rings are perceived to be selfenergising whilst the BX type are designed to fit into a recess that allows metal to metal contact when the flanges are tightened. 2.3.2.4 Specific seal rings These will be found on proprietary equipment manufacturers' joints and should be assembled and tightened in accordance with the manufacturer's specification. 2.3.2.5 All gaskets Gaskets and seal rings should be suitable for their intended operating conditions and be capable of providing a seal under the varying loads imposed by fluctuations in pressure and temperature. Depending upon the application, the main requirements are any or all of the following: — — — — —

Hardness and compressibility. Flexibility. Resistance to heat. Resistance to pressure. Resistance to corrosive action.

Under no circumstances should gasket compound or grease be applied to the gasket or flange faces. Note that for some clamp connectors, the manufacturers recommend that the seal ring be lubricated. Gaskets and seal rings should be: — Stored in their original packing until required.


— Kept horizontal and flat. — Where applicable, left on their individual backing boards until immediately prior to fitting. Specific difficulties can arise with insulating gasket sets and appropriate precautions should be taken if these are to be used. 2.3.3

Bolting

Correct bolt selection, procurement and installation are crucial and only the bolt type as specified in the equipment material specification should be installed. On ASME B16.5 type flanges, for example, the bolts are designed to carry pressure end load at the gasket and also provide the load required to compress the gasket into the flange face in order to effect a seal. Bolt diameters and lengths are specified in the relevant flange code and should also be stated on the fabrication/erection detail drawing. Bolt lengths may have been increased to allow for bolt tensioning equipment, or spades, spacers, drip rings and wafer valves, and the associated extra gaskets. Although the amount of specified bolt protrusion may vary there must be sufficient protrusion for full thread engagement. Many specifications call for a protrusion length of three thread pitches through the nut. Where hydraulic tensioners are used a minimum of one bolt diameter must protrude through the nut to enable safe and effective tensioner operation. The bolt and nut grades and manufacturer's identification should be stamped on both and should be correctly identified before they are used (see Figure 2.8 on page 7). They should both be in compliance with the equipment material specification. The selected fastener material and diameter must provide sufficient elastic or yield strength capacity to safely sustain the design load requirement, service bolt loads and any compensatory overloads needed from the tightening method. Coatings such as hot dipped galvanising and PTFE should also comply with the appropriate coating standard. Bolts with different coatings should not be used on the same flange joint. Bolts, nuts and washers used for joint make-up should be clean, rust free and undamaged. Fasteners can be considered for reuse after considering their service history, operating environment and original risk assessment. Any service coating must be in good condition and still provide 100% fastener surface coverage. This is especially important for PTFE/Organic barrier coatings. Section 7 provides guidance on in-service inspection.

The number of reuses and subsequent life of the bolt should be based on the level of assurance provided by the tightening methodology selected. Greater reusability and longest service life will be provided where the bolt tension requirement is assured by using a load control measurement system with the selected tightening tool. If the bolt is suspected of being overloaded or yielded during a previous installation, it should never be reused.

2.4 PRINCIPLES OF JOINT ASSEMBLY AND DISASSEMBLY 2.4.1

Identification of joint and selection of correct components

Ensure the correct materials are available, matching those detailed in the piping specification, including: — Flanges of correct size, type, material and rating. — Bolts of correct size, material, and length for flange and tightening method. — Nuts of correct grade and size. — Correct thread lubricant. — Correct gasket is available. 2.4.2

Inspect the components and flange faces

Ensure that: — Components and flange faces are clean and undamaged and of the correct surface finish. — Nuts and bolts are clean and free running but not sloppy on threads. — Gaskets are clean and free of damage. 2.4.3

Assemble the components

Components should be assembled in accordance with the procedure relevant to the joint type and specification, and the tightening method to be used. Ensure that: — Bolts are lubricated on threads. — Nuts to be tightened are lubricated on the spot faces. — Bolts are set correctly in the flange to allow for the correct thread protrusion and fitting of tools. — Gasket is centred correctly.


Figure 2.3: Example of flange face configuration Figure 2.4: Flange identification markings

Figure 2.5: Kammprofile gasket with Ident and class marking

Figure 2.6: Schematic of typical spiral wound gasket

Figure 2.7: Type R octagonal ring type joint

Figure 2.8: Stud point and nut showing identification markings


2.4.4

The following precautions should be taken when breaking a joint:

Alignment

Flanges should align initially in the un-stressed condition without any external forces applied unless stipulated within the design (e.g. cold spring). ASME Piping Code B31.3 (1999 Edition) 335.1.1(c) stipulates that flange faces shall be aligned within 1 mm in 200 mm measured across any diameter, and flange bolt holes should be aligned within 3 mm maximum offset (see Figure 2.9). However, this is considered to be a maximum and best practice is to use half this t olerance, thereby making the alignment tolerance 0,5 mm in 200 mm. In general, because of the many variables involved, company standards should be set as to allowable misalignment, but large forces should be avoided. It is recognised that misalignment greater than that specified here, particularly on pipework connected to non-loadsensitive equipment, may be acceptable. However, pulling the flanges into position could cause unacceptable loads and deflections in other parts of the system, and means that bolt load is being used to pull the flanges together instead of to compress the gasket. If additional force greater than can be applied by a single person is required, where flange misalignment or pulling together is excessive or outside the company standards, or where considerable loads are required to correct the misalignment, then the appointed Technical Authority should be consulted and the outcome recorded. 2.4.5

1.

Ensure beyond all doubt that the line or piece of equipment being worked upon has been correctly isolated and vented to atmospheric pressure, and flushed and purged if appropriate. 2. Ensure that all safety precautions and work permit instructions are in place and are strictly adhered to. 3. Take a position upwind of the flange whenever possible. Never stand in line radially with the flange face. Release the bolt furthest away, allowing any residual pressure of gas or liquid to blow away from you. Do not remove the nut and bolt at this stage. 4. Continue to release the remaining flange nuts, but do not separate them from the bolts until the flange joint has been fully broken. Note: It could be the fifth or sixth bolt to be released before the seal is broken. CAUTION: For pressure energised seals on compact flanges or hub connectors, care must be taken that the join t is released before removing the bolts. Personnel should also be aware of the risk of pipe spring or sudden movement as bolt loads are released.

2.5 CONTROLLED TIGHTENING OF JOINTS

Breakout

Before tightening of the joint is considered, it is necessary to consider breakout. It may be that the joint has already been assembled and tightened before, for example as part of a test programme during construction, or the joint is being opened as part of a maintenance programme after a period in service.

The objective of any tightening is to achieve a correct and uniform clamping force in the joint. The operator needs to know the bolt load or bolt stress value required irrespective of what parameter he will be measuring during the tightening cycle. He also needs to know the tightening methodology selected.

-1 mm 3 mm

1 mm

Angular offset

Centre-line offset

Figure 2.9: Alignment tolerances


The bolt load or stress will have been calculated to be suitable for the joint and its service conditions. These details should be obtained from the record and data management system for the site (see Section 5). Any changes in the flange system such as its size, type and material could change the bolt stress requirement and subsequent selection fastener material/diameter selection. Similarly any gasket change could also change the design bolt load. Any such changes must be checked with a Technical Authority. Hot dip galvanised bolting could change the thread dimensions and this should be considered when selecting the correct tensioner or torque tool. On completion of tightening, the joint should be tagged and details recorded in accordance with the site's records and data management system. The following points are specific to the relevant tightening technique. 2.5.1

Torquing specific considerations

2.5.1.1 Lubricant Regardless of the torque tool used, lubricant has a significant effect on the achieved bolt load or stress for a given torque. A known good quality lubricant, suitable for service and of proven coefficient of friction must be used. It is recommended that where possible sites adopt a single lubricant policy; this avoids the opportunity for confusion. Extra care needs to be taken with high friction surface coatings.

4 Bolt Flange

Lubricant must be properly applied to 'working' surfaces only. This includes the bolt threads and the bearing faces of the nuts. 2.5.1.2 Tighten ing Torque tightening should be carried out sequentially, in stages to 100% of specified full torque, using the crossbolt tightening method. Typically three stages of 30%, 60% and 100% are used. It is important that the flange is brought together evenly to prevent overloading of the gasket at any point and this should be monitored at all times during the process. Once the first 100% level has been achieved a check pass should then be carried out on all bolts using a clockwise pass to ensure all bolts are at the final torque level. If a bolt load assurance system is used then the final tightening cycle or check is measured by bolt load. It is possible that the use of a bolt load assurance method can reduce the number of intermediate, pre-torque cycles. The joint will continue to settle under load and the number of passes at 100% will be influenced by the type of joint and its gasket type. For example, cut gaskets and most ring type joints can be considered as 'soft' joints whereas metallic gaskets such as spiral wound types can be considered as 'hard' joints. A soft joint may require more torque passes to reach the required bolt load in all bolts. Figure 2.10 shows cross bolt torque tightening sequences from ASME PCC1.

8 Bolt Flange

Figure 2.10: Cross bolt torque tightening sequence

16 Bolt Flange


2.5.2

Figure 2.11: Use of multiple torque tools

2.5.1.3 Use of multiple torque tools Multiple torque tools can be used on a joint to help flange faces keep parallel during the tightening process. As with hydraulic tensioners, the use of multiple tools can also reduce the effects of elastic interaction causing variation in the residual bolt load achieved. The use of multiple tools can also increase joint assembly speed. In a typical application four torque tools are connected to a hydraulic pump and arranged evenly spaced around the joint as shown in Figure 2.11. When these bolts are tightened, the tools are then moved to the bolts that lie equidistant between the previous tool positions, should there be an odd number of bolts between the tools. When there is an even number of bolts between the tools, the bolts that are nearest the equidistant location are tightened next. On the first pass, typically 30% of the final torque is applied to the bolts. This first cycle is important in pulling flange faces parallel and achieving satisfactory gasket seating.

Hydraulic tensioning specific considerations

2.5.2.1 Key requirements Hydraulic tensioning involves the use of a number of tensioners simultaneously to tighten a joint. The number of tensioners and passes must be known to determine operating pressures. When tensioning, it is important to ensure that the correct bolt tensioning procedure is used in order to obtain a secure and long-lasting leak-free joint. Usually bolts are tensioned in alternate phases using specified hydraulic pressures, taking into account the load loss factor. In high risk joints where a load control system is used a more streamlined procedure is possible. Flanges should be checked for squareness after each tensioning phase. Confirm the bolt load with a break loose/check pass. Where load control sys tems are used this basic check is not required. Bolt lengths need to be increased by one bolt diameter distance to accommodate the hydraulic jack. Hot dip galvanised bolting could change the thread dimensions and this should be considered when selecting the correct tensioning tool. This should be notified to the tensioning company at an early stage. 2.5.2.2 Tensioning pattern or cover Ideally tensioning should be applied simultaneously to all studs in one operation. Where this is not possible, tensioning should be applied in phases using two different pressures, followed by a break loose/check pass, as shown in Figure 2.12. Where a load assurance system is used the break loose/check pass is not necessary.

2.6 BOLTED JOINT RELIABILITY 2.6.1

Reliability factors

The reliability of a bolted joint is dependent on three key factors:

The tightening procedures are dependent upon the individual supplier of the equipment. An example of a procedure is for 50% of the bolts to be tightened in the first pass followed by a second pass in which all the bolts are tightened to full torque. A third checking pass is then made to ensure that the effects of elastic interaction are minimised. However the methodology may vary for differing vendors and therefore the procedures must be checked with the supplier.

— Joint/flange design and calculated bolt load. — Joint component quality. — Correctly assembled and installed design bolt load. These three factors are critical to joint reliability. Measure and control these factors and bolted joint reliability is assured. Once the bolt design load objective has been established the operator needs to consider the criticality of the joint in terms of operating pressures, process fluids and health and safety. This will determine the level of assurance required on installed bolt load, and

Where space permits and when there are sufficient tools and equipment available, it is possible for all bolts to be tightened simultaneously to their final torque value thereby eliminating the need for intermediate steps. 10


2 n d Pass at Pressure 'B'

1 s t Pass at Pressure 'A'

Figure 2.12: Tension tightening sequence

selection of tightening control methodology to achieve the design objective. The design of the joint is outside the remit of this document; however, it is intended to provide a management system that can gather the correct information from the design specification and apply techniques, procedures and systems, to manage the joint in line with design objectives. The following notes are provided for information on that basis. 2.6.2

capacity of all joint components - bolt, gasket and flange - is also an important assessment to avoid overloading and damage from the tightening forces used in achieving the residual design load and subsequent service loads. Calculation methods based on VDI 2230 (Systematic calculation of high duty bolted joints) take into account these different loading conditions. One such design code, EN 1591 (Flanges and their joints. Design rules for gasketed circular flange connections. Calculation method), is specific to pressure-containing flanged joints but certain gasket performance data are required from the gasket manufacturers for the calculation. Gasket manufacturers also provide design bolt loads for various standard flange ratings based on the gasket performance data.

Bolt load calculations

It is crucial that the design bolt load required to seal the joint has been calculated using an approved method and is known prior to joint installation. The value for each joint and the source of the value should be recorded in the site's record and data management system. This facilitates consistency and traceability and allows conscious decisions to be made regarding bolt load should an issue arise with a joint. The recognised codes generally provide a method for calculation based on operating conditions such as pressure and temperature. The most frequently used code is the ASME Boiler and Pressure Vessel Code. It is relatively simplistic in predicting gasket performance. The latter is an important factor and it has been recognised that more realistic and definitive gasket performance data are required. Both in USA and Europe gasket testing is being conducted, the results of which will be incorporated into an updated ASME code in the future.

2.6.3

Bolt tightening

The purpose of tightening a bolt is to stretch the bolt (like a spring) within its elastic limit such that in trying to return to its original size it imparts a clamping force on the flange. Bolted joints can be tightened by a number of techniques. Torsional based methods range from the simplest low cost spanners through to impact, manual and hydraulic wrenches. These apply a torsional force to generate tensile loading in the bolt. Bolt tensioners are different in that the bolt is loaded by applying a direct axial tensile force with hydraulic jacks to stretch the bolt. Some of this stretch is then captured by the turning down of the permanent nut. A mechanical variation on this method uses torque tightened small diameter screws going through the flange's load bearing nut and reacting against a jacking washer, thereby tensioning the bolt.

There are other service loads acting upon the joint which can be just as significant as the internal pressure. Transverse vibration, axial cyclic fatigue and structural loading all come into play. The joint can also suffer relaxation or increase in compression dependent on component materials and temperature. The strength

None of these systems directly measures the achieved bolt load. However steps can be taken to 11

improve correlation between actual residual bolt load achieved and the tightening system's power input of torque or initial hydraulic pressure. Robust procedures, well maintained, calibrated tooling and the use of competent operating personnel help improve the correlation. Totally uncontrolled tightening with spanners is not a preferred option for tightening any size of bolt. However, where a risk assessment identifies a significant risk and where a superior tightening method is not possible, e.g. in a space too restricted for torque or tensioning equipment, spanners can be used with a load control system. 2.6.3.1 Torque tightening Torque control methods such as impact wrenches have far less load control than hydraulic wrenches. For the smaller bolts (< 1", M24) calibrated and maintained hand torque wrenches will generally provide good practice for controlled torque tightening. The variation in a torque reading and the resultant bolt load is dependent on many factors e.g:

— Friction in the fastener mating interfaces. — Fastener quality e.g. nicks, thread laps, general damage etc. — Tolerances in bolt, nut and flange dimensions. — Tolerances in bolt, nut and flange material and mechanical properties. — Operator competence. — Accuracy of the torque application device. — Bolt diameter. — Surface coatings and lubrication.

system. Torque values for particular bolt sizes can be found within specific operators' standards or, in the case of proprietary manufacturers' connectors, from their catalogue or from approved bolting service providers. When selecting values great care must be taken to ensure that the same lubricant or anti-seize compound is used as stated in the data sheet from the management system. The actual lubricant friction factor must be recorded. Many sites find it advantageous to specify one lubricant for all bolt torquing operations. Elastic interactions in the joint can significantly affect the residual bolt load achieved through torque tightening. These effects can be reduced by simultaneously tightening a number of bolts in the joint with multiple torque tools similar to hydraulic tensioning methodology. This procedure is detailed in 2.5.1.3. 2.6.3.3 Hydraulic tensioners When joint conditions are favourable and all bolts in a joint are tightened simultaneously, hydrau lic tensioners can provide a consistent bolt tension. Whilst the bolt tension, or preload, is known through the hydraulic pressure applied, the residual bolt load at the end of the tightening cycle is subject to the amount of relaxation that occurs on load transfer. The latter depends on a number of factors, some joint related, some tool related and others 'fitter' related, e.g:

— Tolerances in bolt, nut, flange and gasket material properties. — Tolerances in bolt, nut, flange and gasket dimensions. — Operator skill and control of technique. — Load loss factors during the process. — Calibration of pressure gauges. — Correctly maintained tensioning system.

Great care has to be taken in evaluating the frictional conditions and resultant friction factor used in the torque tension equation to improve the reliability in correlation between torque and bolt load. The choice of lubricant, surface coating and fastener quality can improve the torque/ bolt load variation. One new hydraulic torque system uses a hardened washer introduced under the load bearing nut such that its design provides system reaction and reduces bending stresses associated with traditional torque reaction against the adjacent bolt or joint structure. The washer has a specially prepared bearing surface that is intended to improve friction consistency, in the nut bearing interface face, and bolt load variation.

Two specific types of load loss factors to be considered when calculating the required level of compensatory hydraulic overload pressurisation are Tool Load Loss Factor (TLLF) and Flange Load Loss Factor (FLLF). TLLF occurs in all tensioning cases, whereas FLLF does not occur in 100% tensioning. —

Tool Load Loss Factor

When the load is applied to the tensioner it stretches the bolt and lifts the permanent nut clear of the surface. Whilst the load is held by the tensioner the nut is then turned back against the flange surface. When the tensioner pressure is released the load transfers from the tensioner to the threads of the nut. In taking up the load the threads deflect resulting in a loss of load. This factor is allowed for in the calculation of applied load.

2.6.3.2 Torquing process It is vital to ensure that the correct bolt torque figures are available prior to making up a flange joint. These should be stored along with the source of the bolt load calculation in the site's record and data management

12


Note: This means that with hydraulic tensioning a higher load than the residual design load should always be applied.

and for ensuring that tools are used by personnel competent and trained in their use. Such procedures should be specific to the equipment employed.

— Flange Load Loss Factor

2.6.5

Flange Load Loss Factor only occurs when less than 100% tensioner coverage is used. When using only 50% cover (e.g. eight tensioners on a 16 bolt flange) when the second pass is applied, the gasket undergoes further compression, effectively relieving some of the load in the bolts tightened by the first pass. By tightening the first pass to a higher load, i.e. adding on FLL F, the need for more than one pass at the second pass pressure can be avoided. It should be noted that when two passes are used the combination of FLLF and TLLF may mean that the applied bolt stress is greater than the yield stress of the bolt. An alternative technique such as multiple passes at the second pass or pass B pressures may then be required. Careful use of load factor curves to predict the above factors and realistic selection of the system for short, medium and long grip length joints can improve the correlation between compensatory overload pressurisation and the residual design load target. As indicated above, the number of jacks selected for the tightening can improve the load transfer relaxation situation, particularly with respect to joint elasticity effects.

The selection of control of installed bolt load through torque, hydraulic pressure or direct through a load control system, should be dependent on the risk assessment. Assured bolt load provides assured joint reliability assuming the design and component quality and assembly are also assured. Selection of an appropriate tightening methodology with bolt load assurance will provide the minimum risk. Risk increases if bolt load assurance is not provided. It is recommended that any load control system is 100% load test calibrated to ensure all bolts tightened in the joint are loaded correctly and to the system's assured accuracy tolerance. Several techniques are commercially available to control and assure bolt load, as set out below. 2.6.5.1 Direct length measurement This method uses mechanical extensometry to measure the bolt extension. Accuracy is dependent on the level of physical load test calibration carried out. A readily available technique is the indicating rod bolt type. A rod is inserted into a drill hole in the bolt that runs the fastener's complete length. The rod is anchored at the opposite end to where the measurement takes place. At the measuring end a precise datum face is machined leaving the rod end flush with the bolt face. Relative displacement of the rod compared to the bolt face is measured and calibrated against bolt load by physical load test.

2.6.3.4 Tensioning process The hydraulic tensioning values needed to achieve the residual design load derived from 2.6.2 should be obtained from the record and data management system. Tool pressures must be specific to the tool used. The bolt tensioning operation must be carried out in accordance with the tension equipment manufacturer's specified procedure and the load loss factors should be recorded. Ideally tensioning should be applied simultaneously to all studs in one operation. Where this is not possible, tensioning should be applied in phases using two different pressures as described 2.5.2.2.

2.6.4

Load control systems - Assured bolt load

2.6.5.2 Ultrasonic direct length measurement This method determines the stress by measuring the time of flight of an acoustic pulse travelling from one end of the stud or bolt to the other. The time will vary depending on the extension and the stress in the stud or bolt. The monitored time is proportional to the bolt extension and stress and can be converted to provide an output as a bolt tension or stress as required. The pulse is generated by a hand-held processing unit and is independent of the tightening method. Accuracy is dependent on precise datum faces machined at each end of the fastener, the level of physical bolt load/extension load testing carried out and operator skill. It is recommended that only skilled operatives are used to carry out this technique. 100% load test calibration can provide accuracy results similar to mechanical methods. Calibration by calculation only provides the least degree of accuracy.

Equipment and tools

In order to improve flange integrity and safety in operation, it is important that pneumatic and hydraulic torque/tensioning equipment meets the required specification and is maintained and calibrated as a minimum on an annual basis or more often if circumstances warrant it. Gauges should be calibrated prior to extended use. There should be clearly defined procedures stating who is responsible for ensuring that tools are calibrated 13

2.6.5.3 Load monitoring sensors There are several load monitoring sensors commercially available. These include capacitance, fibre optic and strain gauge techniques that take the form of sensor inserts placed into a converted bolt. Another type is the compression load cell that fits like a washer under the nut or bolt head. One load cell monitors any change in the nut face stress using an amorphous material. Other types use strain gauges in the cell structure. Signals from all types of sensors can be read by a hand-held device or hard-wired logging systems; they have future potential for remote signal monitoring. The sensors are particularly useful where there is a need to continuously monitor bolt load in service. 2.6.5.4 Mechanical load indicating bolts These comprise standard bolts converted to monitor bolt load. The bolt has a pin with a rotor attached, anchored in an axial drill hole. The rotor air gap is set to rotate freely until a specified bolt load is achieved. The indicator is enclosed in a protective cap. Simple finger feel of this cap determines bolt load status. Tension is indicated at make-up and throughout the life of the joint. Variations of this technique include a dual indicating maximum/minimum load range system as well as a visual indication system.

Pressure testing should be carried out to a documented procedure which complies with the HSE Guidance Note GS4 'Safety in pressure testing'. Additional guidance can be found in the OCA 'Guidance Notes of Good Contracting Practice - Pressure Testing'. 2.7.1.1 Standard pressure (strength) test On newly constructed or installed pipework and pressure equipment, company standards will normally conform to a relevant design code such as ASME B31.3. The objective of a strength test is to prove the quality of materials and construction of the equipment before it enters service or re-enters service following significant repair. This test is carried out at a specified pressure above the design pressure - detailed within the relevant design code. Pressures are typically 1,25 to 1,5 times the design pressure for hydrostatic tests or 1,1 times for pneumatic testing. This is a strength test of the system and whilst it will indicate some issues with joints it does not provide assurance of the integrity or in-service reliability of the bolted joint. 2.7.1.2 Leak test Leak testing may be carried out on equipment prior to strength testing. In this case, testing should be limited to a pressure not exceeding: — 10% of design pressure.

2.7 INTEGRITY TESTING The combination of the procedures and processes recommended in this document together with appropriate testing prior to going on line and in-service inspection programmes described in Section 7 will provide the highest level of assurance. Testing is not a substitute for correct assembly and controlled tightening. It should be standard practice to assemble and control-tighten joints correctly the first time to eliminate rework and minimise downtime.

Leak testing is normally carried out on equipment in order to prove the integrity of joints disturbed after a strength test has been successfully completed or during subsequent maintenance work. In this case, testing should be limited to a pressure not exceeding:

2.7.1

NB - on older equipment, the strength test is likely to have been carried out several years earlier.

Levels of pressure testing

Once the joint has been tightened and certified, and details recorded in the record and data management system, the joint should be subject to an integrity test prior to going into service. The level of testing is determined by the operator and will normally comprise one or more of the following: — — — —

— 110% of design pressure, or — 90% of relief valve set pressure if still in place and un-gagged.

Standard pressure (strength) test. Leak test. Service test. Functional test.

2.7.1.3 Service test A service test is one which is normally carried out on a joint where it has not been possible or practicable to carry out a leak test first. Service tests are carried out with the pressure system in service, normally during start-up. The test is normally carried out (but not necessarily always) at maximum normal operating pressure using the process fluid as the test medium, supplemented by water or inert gas from an external source if necessary. The scope of service testing is to demonstrate joint integrity for any joints where leak

14


testing is not reasonably practicable, i.e. witness joints. 2.7.1.4 Functional test This test is normally carried out at the working press ure using a suitable test medium. Its objective is to ensure that the equipment and its components function properly e.g. valve cycling. 2.7.1.5 Testing mediums Hydraulic test mediums (incompressible fluids) are commonly treated water, glycol or diesel. These have low stored energy; however, there can be material compatibility issues which require consideration e.g. chlorides on stainless steel. Pneumatic test mediums (compressible fluids) are commonly nitrogen with a helium trace, air or steam. Safety Note Strength testing is almost always carried out using liquids (hydrostatic or hydraulic testing). Although pressure testing using a liquid is not without risk, it is by far the safer method and should be used wherever practicable. Pressure testing using air, steam or gas (pneumatic testing) is more dangerous because of the higher energy levels involved. The energy released during a total failure of equipment containing compressed air can be up to 200 times the energy released by the same test if water was used as the test medium. Pneumatic strength testing should never be carried out using flammable gas. Pneumatic leak testing to 10% of design pressure can be used to find small but significant leaks in equipment which will contain flammable gases and/or liquids. Caution should also be taken when carrying out hydrostatic testing at low ambient temperatures (<7ºC) to avoid the risk of brittle fracture. Refer to the HSE Guidance Note GS4 'Safety in pressure testing' and the associated research report for further details.

configuration of the system, and the hazards associated with the introduction of high pressure testing equipment would be greater than the hazards associated with service testing. Where this method is proposed it should only be carried out in accordance with a company procedure for service testing and a written justification must be recorded and a risk assessment carried out. 2.7.2

Test recording

The type of test, specification and acceptable leakage rate criteria should be determined and documented by the operator based on the criticality assessment already carried out on the joint to determine the assembly and tightening assurance specification. Results of tests should be recorded in the record and data management system. 2.7.3

Witnessed joints and reverse integrity testing

Where joints have no means of isolation to allow leak testing of the installed joint, such as the last connection on an open flare line, or where a large number of joints makes it impracticable or unreliable to conduct a leak test, the operator should regard this as a higher risk joint in his criticality assessment and therefore consider a number of additional steps including: — Witnessing assembly of the joint. — Witnessing controlled tightening of the joint. — Applying a load assurance system to assure the required bolt tension has been achieved. — Using a reverse integrity test using a proprietary gasket. This is based on the principle of pressurising the annular space above and below the seal ring using a test gas, usually nitrogen. Witnessed joints should be highlighted in the record and data management system, including the results of any tests or witness inspections.

2.7.1.6 Testing using process fluid or gas For process hydrocarbons systems, although it is not the preferred means of testing, under certain conditions it may be considered appropriate to carry out testing with the service fluid rather than with water, nitrogen or some other medium. This should only be considered where it can be clearly demonstrated that it is impractical to carry out leak testing due to the

2.7.4

Joint failure during integrity testing

Where a joint fails an integrity test, then applying more bolt load alone is not the answer. Investigation and analysis in accordance with the measures described in Section 6 should be carried out.

15

16

CRITICALITY ASSESSMENT safety and environmental aspects of the local and distant environment. For onshore, this will often be part of the Control of Major Accident Hazard (COMAH) assessment for the site. For offshore, Safety Case, PFEER and Pipeline Safety Regulations will apply. The UK Health and Safety Executive OIR/12 database contains useful information to enable offshore industry operators to develop their risk assessment. Risk may also occur with joints containing harmless fluids e.g. water, which would damage building fabric or product, or risk interaction with electrical installations if they leaked. There are a number of areas which will affect the criticality of the joint. These can be grouped as follows:

3.1 INTRODUCTION There is a variety of bolted joints involved in pressurised systems, ranging from low pressure joints containing water or compressed air to high pressure joints containing steam, hydrocarbons or explosive or poisonous gases. Although every joint should be designed and installed to safely contain the pressure and contents specified, it is logical that joints at higher pressure or with hazardous contents will require additional vigilance due to the potential consequences of failure. The criticality of a joint may have an effect on a number of areas addressed in the management system including: — Choice of tightening method. — Choice of personnel assembling and tightening the joint. — Level of bolt load assurance. — Level of records and data stored against the joint. — Level of inspection and testing prior to entering into service. — Level of testing and inspection in service.

3.2.1

Leak potential

One method of determining the criticality of a join t is to consider the potential for a leak. The potential for a leak will increase with: — — — — — — — — —

3.2 ASSESSING THE RISKS WITH BOLTED JOINTS The level of risk will primarily be based on the service conditions the bolted joint is exposed to , along with the impact any release would have on the operational,

17

Process and test pressures. Cyclical load. Vibration load. Low temperature. High process temperature. Structural load. Corrosive environment. Aggressive environment. Unknown conditions of any sort.


3.2.2

3.2.4

Service fluid

Local factors

The contents of the pressure system have a major effect on the criticality of the joint and should be considered in determining the level of inspection, control and testing applied to the joint. A joint's criticality will increase if the contained service is:

Local factors must always be considered when assessing a joi nt' s criticality. Table 3.1 describes some of the factors which may occur at individual joint level.

— — — — — — — —

The criticality of the joint is shown in Table 3.2. The criticality level can be determined by considering all of the factors identified in 3.2.1 to 3.2.4. The operator should use the level of criticality to set standards and specifications for:

Hydrocarbon. Corrosive. Explosive. Poisonous or noxious. Radioactive. High temperature. Environmental contaminant. Expensive.

Such joints would often be viewed as at least of medium criticality. 3.2.3

Loss potential

The criticality may also increase if loss of the service would render the plant inoperable. For example a fire service line, although having safe contents, would cause a plant shutdown if inoperable. Similarly a cooling water system for a computer plant could be highly critical. The loss potential may also increase with pipe size and the area through which it runs.

3.2.5

Joint criticality rating

— Joints which will be included in the management system. — The level of inspection and assurance at assembly stage. — The level of personnel who will control tighten the joints. — The control tightening method. — The level and method of bolt load assurance. — The level of inspection during the controlled tightening stage. — The type and level of integrity test prior to entering into service. — The type and level of in-service inspection.

Table 3.1: Local factors Factor

Problem

Vibration or slug flow

If severe may cause joint to loosen

Cyclic temperature

If severe may cause joint to loosen

Substitute materials to those in Piping Specification

Compatibility not guaranteed

Local joint history

If this individual joint is misaligned or difficult to close, or if this type of joint is problematic on this site

Untested joints

Cannot be leak tested prior to start-up (e.g. tie-in points)

Vendor package joints

Often assembled and tightened to vendor's system, outside of asset system

Exception on joint

Flange face marked, piping load, history of leakage with root cause unidentified

18

CRITICALITY ASSESSMENT

Table 3.2: Joint criticality - Examples of criteria used and controls applied Joint Criticality

Low

3.2.6

Controls — Joint identified and recorded in database — Assembly not witnessed but carried out to a procedure by trained and competent contractor — Bolt loads taken from database — Controlled tightening applied by use of hand torque wrench or torque wrench — Tightening carried out by competent personnel (see Section 4) — Integrity test by local arrangement — In-service testing includes visual inspection

Medium

— Joint identified and recorded in database — Assembly witnessed or a sample of joints witnessed and carried out to a procedure by trained and competent personnel — Bolt loads taken from database — Controlled tightening applied by use of hand or hydraulic torque wrench or tensioner by competent personnel — A sample of joints witnessed by specialist personnel — Integrity test may include nitrogen helium or similar — In-service testing in accordance with the techniques described in Section 7 — Consider use of load assurance

High

— Joints uniquely identified in database and identified as High criticality — Assembly by specialist contractor or witnessed by specialist contractor — Controlled tightening using hydraulic tensioner or hydraulic wrench with load assurance system by specialist personnel — Integrity test using nitrogen helium or similar prior to entering into service — In-service inspection at higher level in accordance with the techniques described in Section 7

effective procedure than tensioning. The shorter grip length joint also makes the tensioner less reliable as a control system. By thoroughly researching the friction factor for the preferred lubricant and taking into account the surface coating and bolt quality, torque tension variations may be reduced.

Sample risk assessment selections

Assured design bolt load on installation by measuring with a load control system provides assured joint reliability or minimum risk with any tightening technique. Under the same joint conditions reliability will be less assured and risk will increase by using only the tightening technique. Selection is down to the operator's risk assessment, past history of the joint and associated life cost of the techniques available to him. These examples are not intended to be prescriptive but show possible methodology selection subject to an operator's individual situation.

ANSI B.16.5 600LB 10 INCH; HAZARDOUS FLUID. 1,1/4 in. Hydraulic tensioner tightening Whilst the 1.1/4 in dia bolt could be tightened using a hydraulic wrench, it may have insufficient control to provide a reliable level of bolt load. The service conditions in terms of pressure, temperature and contained fluid provided intermediate risk. The bolt diameter and grip length were such that the hydraulic tensioner could provide sufficient bolt load for joint reliability under service conditions. We could have this same 600 lb flange but service temperatures could be high (350°C plus) and/or cycling. This could present an increased risk such that assurance

ANSI B.16.5 150 LB 5 INCH; HAZARDOUS FLUID. ¾ in. bolt Torque tightening; torque control; known low friction lubricant for friction factor control. The operator may decide this application does not warrant the use of a load control system. The smaller diameter means that torque tightening is a more cost

19


is needed on installed bolt load. For this a load control system is used with hydraulic tensioner tightening to minimise the risk of a leak.

3.2.7

Record the criticality assessment

The joint risk criticality should be recorded in the records and data management system (see Section 5). Before work on any joint (e.g. design, modification or maintenance) the risk criticality should be identified and recorded. If the risk criticality has not already been identified and recorded, a criticality assessment should be performed and recorded in the records and data management system.

ANSI B.16.5 900LB 16IN, HAZARDOUS FLUID. 1,5/8 in. Hydraulic tensioner tightening; load control system. Some operators link tightening method selection to bolt diameter. For example, hydraulic tensioners are usually specified for diameters 1,1/8 or 1,1/4 diameter and above. The larger the diameter, the more effective tensioners become compared to torque in terms of providing tightening power with variation in bolt load. The higher pressure, pipe diameter and process gas in this situation results in the operator regarding risk as 'high'. Therefore assurance on installed bolt load is necessary and a load control system is required to ensure design objectives are achieved on installation. It would be quite feasible however to select a hydraulic torque wrench with a load control system for this application.

3.2.8

Risks to personnel

It is important to note that assembly of flanged connections involving the use of high pressure hydraulic tools and systems will present a level of inherent risk to the operator which if not assessed, controlled and if possible mitigated, may result in a serious incident. For all flange assembly operations the risks during assembly should be fully and formally assessed, the selection of methods and tooling reviewed, hazards identified and where possible, the risks mitigated on the basis of the ALARP (As Low As Reasonably Practical) principle. All personnel involved should be made fully aware of the potential dangers of accidental leakage of high pressure hydraulic fluid from the tools and systems deployed. During training of personnel, it should be emphasised that the risks from high pressure fluid systems are constantly present during the tightening/ loosening procedures. The need for constant observation and inspection of the equipment throughout the whole operation should also be stressed.

Tightening method selection based on bolt diameter; whilst satisfactory for general 'rule of thumb' on low and some intermediate risk standard ANSI flanges, the policy could be problematic for non standard joints especially those where the bolt diameter to clamp length ratio is relatively small (less than four to one for example). Where one would normally nominate tensioning for a larger bolt diameter, the latter situation could result in the target bolt load being practically unreachable due to joint elasticity. The higher compensatory hydraulic overload may be outside the elastic capacity of the bolt or even the capacity of the hydraulic jack itself.

20

TRAINING AND COMPETENCE maintain joints, or to supervise or assess such work. Includes a process to assure that third party vendors and contractors can demonstrate that their personnel are managed using equivalent systems to equivalent competence standards.

4.1 INTRODUCTION All personnel carrying out work on bolted join ts should be trained and competent to a level appropriate to the required technical skills and failure risks of the joint involved. Similarly, supervisory personnel and assessors should also be trained and competent to ensure they are aware of the issues involved in achieving a leak-free joint.

4.3 TRAINING The skill levels that individual companies use will depend on a number of variables. For example, a company with a large number of personnel may decide on a number of skill levels appropriate to the type of work an individual may perform. Other companies may decide to train all their personnel to a higher level as a matter of course. This approach is particularly relevant to remote sites where it is imperative to have personnel with the necessary skills available at all times. As such, the training specifications for the following Engineering Construction Industry Training Board (ECITB) TECSkills units have become the benchmark standards for the UK offshore oil and gas industry:

4.2 COMPETENCE MANAGEMENT Control of the competence of people working on bolted pipe joints is a critical factor in achieving joint integrity. Hydrocarbon release incident data for the UK offshore oil and gas industry indicate that poor bolted pipe joint make-up is a major cause of leaks, and a review of historical causes confirms that the skills and practices used have not given leak-free joints. Therefore an important element of a management system is to ensure that any person working on a given joint has been trained and assessed as competent to perform the task. Fundamental to the demonstration of personnel competence is the provision of a documented competence management system that:

PF010 PF015*

— Contains clear standards for recruitment, training, development and ongoing competence assessment. — Is based upon, equivalent to or better than a nationally or industry-recognised technical standard. — Provides demonstrable capability for all staff personnel who might be expected to make, break or

PF018 PF019 *

Jointing Pipework using Flanged Joints (Hand Torque Tightening). Assembling and Tightening Bolted Flanged Connections. Assembl ing and Tensi oning Bolted Connections. Assembling and Tightening Bolted Connections (Hydraulic Torque Tightening).

superseded by PF018 and PF019

21


Schemes operated by individual companies should be aligned with these or equivalent specifications. Such schemes should also address those individuals used during turnarounds and periods of high activity whose core function is not assembly and tightening of bolted joints.

training. It is rooted in the Engineering Competence Standards (ECS) based on national occupational standards for the engineering construction industry. Successful completion of TECSkills On-the-Job Performance Units or equivalent units from other Independent Accreditation Organisations (IAOs) (see 4.7) can contribute to the evidence requirements of vocational qualifications. An occupationally competent coach and IAO representative support the learner in the attainment of new skills and knowledge when undertaking training or performing these units. In response to the UK oil and gas industry, the ECITB developed training and performance units PF010, PF018 and PF019 for assembling and tightening bolted flanged connections. These units form part of the TECSkills training programme for training pipe and mechanical fitters, hence the PF title.

4.4 ONGOING COMPETENCE Successful completion of an appropriate training course is only the first step towards gaining and demonstrating competence. The course should be followed up by an agreed training and assessment plan between the coach and learner, which will establish whether the training has been effective and identify gaps in the learner's skills and knowledge. Together with a logged record of experience and a site assessment, this can lead to a recognised qualification such as the ECITB's TECSkills units PF010, PF018 and PF019 (see 4.5). These units, with supporting material, may contribute as evidence towards obtaining a vocational qualification unit (see 4.6).

4.6 VOCATIONAL QUALIFICATIONS A vocational qualification (e.g. National or Scottish Vocational Qualification - N/SVQ) is effectively a portfolio-based validation process that will include onsite assessment by an occupationally competent assessor. No training is necessarily required to take a vocational qualification. The qualification is based on evidence of competence by a variety of techniques, including documentary evidence, questioning, site observation and testimonials. A competent assessor will easily identify weak candidates. The standard of candidate able to pass a vocational qualification is controlled by the awarding body in line with national guidelines.

An example of the competence requirements for authorised bolt tightening personnel is given in Table 4.1. To assist in demonstrating ongoing competence, a record should be maintained of each individual's mechanical jointing performance. This should comprise details of the types of joints the individual has worked on (including evidence that a representative sample of joints have been made up in the presence of a competent assessor), whether the joints have performed satisfactorily, and details of any further training required. It is the responsibility of the individual to maintain this certified history and to have it formally validated by an approved assessor. If there is no record of successful past work within a 12-month period it is recommended that an assessment is performed to identify any re-training requirements.

4.7 INDEPENDENT ACCREDITATION ORGANISATIONS Examples of bodies who can be contacted for advice are given below. There are many other agencies and individual companies which are available to provide training. However, it is essential to ensure that the training they provide is to a recognised standard.

An example of a mechanical jointing performance record is shown in Figure 4.1.

4.5 TRAINING IN ENGINEERING CONSTRUCTION SKILLS (TECSKILLS)

ECITB (Engineering Construction Industry Training Board) SEMTA (Science, Engineering and Manufacturing Technologies Alliance) API (American Petroleum Institute)

The ECITB's Training in Engineering Construction Skills training programme (TECSkills) is an example of a flexible training scheme for craft and other site operatives to cater for both initial and skill enhancement

22

TRAINING AND COMPETENCE

Table 4.1: Competence requirements for authorised bolt tightening personnel Key Requirement Training provided should include knowledge of the specific joint types employed at the worksite. Operators should ensure that any training carried out on their behalf meets with this requirement. Knowledge Base Awareness of: — Health and safety precautions — Pressure, temperature and hostile environmental factors (such as corrosion and vibration) on the degradation of bolted assemblies — Factors which result in bolt load variation — Applied and residual loads — The effect of different lubricants on friction losses — The relative accuracy of different methods of tightening — The techniques for application of tensioned bolt loading — Joint assembly methods and tightening procedures — The need to check gaskets, nuts and stud bolts against specification — Safety precautions when handling and removing Compressed Asbestos Fibre (CAF) gaskets — The requirement to tag and complete records for assembled joints — The need to: Check the compatibility of the selected torque tools and equipment capacity prior to use Top up oil levels in hydraulic pumps Clean and protect tools and equipment from corrosion Understanding of: — The principles of joint component sealing action — The principles of bolt elongation and tensile stress — The function of gasket or seal types — The importance of correct bolt loading — The effect on bolt load and seal compression using different methods of tightening — The importance of using the correct lubricant — The importance of the correct selection of joint components to comply with the design specification — The correct sequence and number of tightening passes required for torque and tensioned bolts — The principles and techniques used for direct bolt length measurement — The need for and using reporting procedures when defects or faults in bolt tightening equipment or its assembly are identified — The principles of preparing bolted joint connections for assembly — The need for seal face cleanliness and for nuts to be free-running — The effect of joint alignment and gap uniformity on residual bolt loading — The importance of gasket storage, handling, preparation and installation — Good installation practice for bolting, washers and nut orientation for tightening method and equipment to be used — The need to report variances from design specifications and tightening procedures — The principles and requirements for the safe selection, calibration, installation and use of hydraulic torque and bolt tensioning equipment — The principles of carrying out bolt de-tensioning and joint breakout safely and correctly — The importance of attending product-specific training and following the manufacturer's procedures for proprietary joint types — Why mixing components from different equipment manufacturers is prohibited — The principles of inspection after tightening and the procedures and techniques to be used such as 'break loose' tests (check passes) and bolt tightness 'tap-test' — The requirements for the storage, preparation, maintenance and calibration of torque tools and bolt tensioning equipment for its safe use

23


Table 4.1: Competence requirements for authorised bolt tightening personnel (cont'd)

Ability to: — Recognise and rectify faults with torque or tensioning equipment — Interpret joint or flange manufacturer identifying marks — Identify defects, distortion and surface irregularities on flange sealing faces and threads

Demonstrated Application of Knowledge Demonstrate ability in: — Preparation of all joint components — Correct selection and assembly of joint components — Diagnosis and rectification of problems with hydraulic equipment — Selection and correct installation of hydraulic torque or tensioning equipment — Correct application of the various tightening techniques — Carrying out specified tightening sequence and subsequent tightening passes to ensure axial alignment and squareness of joint assembly — Carrying out joint breakout safely and correctly — Carrying out bolt 'break loose' tests (check passes) to check integrity of assembled tensioned assembly — Completion of a joint record sheet — Integrity and inspection checks of completed joint assembly — Maintaining a personal portfolio of joint assembly Demonstrate awareness of: — The health and safety precautions at the worksite — Hot bolting and live plant procedures and risk assessments

24

TRAINING AND COMPETENCE

JOINTING PERFORMANCE RECORD

Name

ID No

Installation

Date

This is to certify that the Technician named above has produced satisfactory leak-free mechanical joints of the types indicated below within the past 12 months.

Joint Type

Satisfactory Performance

Requires Training

Date

Signature

Comments

RTJ Raised face Insulating gasket Compact Clamp connector Taper-Lok Kidney Other (installationspecific) Note: 1. This record does not replace a recognised NVQ but certifies a Technician's ongoing competence in making a specified mechanical joint. 2. It is recognised that certain installations do not have all types of joints. Supervisor (Position)

Verified (Position)

Name:

Name:

Signature:

Signature:

Date:

Date:

Figure 4.1: Example jointing performance record

25

26

RECORDS, DATA MANAGEMENT AND TAGGING The certainty of a successful joint being made up increases if data are controlled and historical data exist on the activities carried out in the past. Recording traceable data encourages best practice at the time of the activity, and will provide useful planning data for the next time the joint is disturbed. Learning from incidents is important. A management system should include the means for gathering relevant data, which should be collected by everyone involved in bolted joints and periodically reviewed to establish trends, performance and improvements. This can be achieved if records and data are kept for each joint as part of a management control process.

should be securely attached to the join t and may hold no other data than the unique tag number. In selecting a permanent tag, consideration needs to be given to the attachment method, the temperature of the flange and tag and security device material, the permanence of the tag markings, and avoidance of corrosion spots due to dissimilar metals or water traps. 5.1.2

Temporary tags

The purpose of a temporary tag is to uniquely identify a joint during a work scope and to indicate the status of the joint during the work scope. The tag will normally hold a unique ID number for the joint which is traceable; it may also hold a small amount of information such as tightening method and date, person who assembled the joint, person who tightened the join t and person who tested the joint. A common method is to use multipart tags where the status is indicated by the colour of the portions remaining on the tag. Common status conditions are:

5.1 JOINT IDENTIFICATION

In order to record data and plan activities - each joint needs to be clearly and uniquely identified. This requires the joint to be physically tagged so that its identity is clear and visible at the joint location including a unique Joint ID number in order that it can be recognised in a joint database or other record system. All joints should be tagged, there are also advantages to having both permanent and temporary tags assigned to joints.

— — — —

Joint to be broken out/Joint broken out. Joint to be assembled/ Joint assembled. Joint to be controlled tightened/Joint tightened. Joint to be tested./Joint tested.

Joint tagging can bring a number of benefits: 5.1.1

Permanent tag

— — — —

The purpose of a permanent tag is to uniquely identify a joint throughout its life cycle, enabling all activities and data on that joint to be recorded. Permanent tags

27

Control competence. Assist in the preparation of work permits. Provide cross-shift communication of job status. Assist job completion confirmation.


— Aid leak and seep searches by identifying disturbed items (which have a higher probability of leakin g). — Support a record and data management system. 5.1.3

Example tagging procedure

The following is an example tagging procedure. Individual operators' schemes will depend upon a variety of issues including the number of joints to be tagged and the size of local organisation. All joints that are to be made or disturbed during construction or maintenance work should be identified, recorded and tagged. The tag should be fitted using a suitable tie and in a position adjacent to or on the joint. The person who breaks the joint(s) should mark up the tag identification numbers on a copy of the relevant Isometric or P&ID and its corresponding register. The Isometric or P&ID and register should be controlled by the relevant designated person. These records complement the leak test certificate and provide an audit trail. At the completion of each stage of the job (inspection, assembly, tightening and testing) the responsible person should record their name against that stage. This could be done directly on the tag or the relevant task could be crossed off on the tag and the name recorded in the work pack. Once testing has been satisfactorily completed, the removable insert of the tag should be returned to the jo b co-ordinator. The task completion should be recorded in the work pack when all joint tag bodies are returned, indicating that all work has been completed. This can be checked against the permissions required for restarting the plant. Plant start-up should be prevented until all the tag inserts are signed off and returned to allow sign-off of the job. After start-up and while the root of the tag remains attached, search teams should patrol the disturbed area and inspect tagged joints for leaks and seeps. Any leaks or seeps should be reported to a nominated supervisor. The root of the tag should be left on the joint until the operation is satisfied that the joint is not likely to leak (normally 48 hours after start-up.) During that period the tag makes leak searches more effective. An example of a multipart tag is shown in Figure 5.1.

Figure 5.1: Example of a multi part tag 5.2 RECORDS AND DATA MANAGEMENT

A successful record and data management system will aid and provide information during the work planning and execution process. Once unique joint IDs have been established then useful and essential data can be recorded against them. As work is carried out and recorded the status of the disturbed joints should be updated to reflect the status of all joints including temporary blinds. This process should be carefully controlled and reported as laid down by the management process. The preparation and collection of data by competent personnel will assist in ensuring all joints are assembled, tightened during construction or reinstated during maintenance and ready for leak-free service. Additionally, as the status of all pipework has been carefully monitored, it should not be possible to introduce pressure into any joint before all joints have been reinstated.

28

RECORDS, DATA MANAGEMENT AND TAGGING

5.2.1

Recommended data

The following data are recommended as a minimum for bolted joints on critical services: 5.2.1.1 Joint details — Identity of joint. — Joint location. — Drawing references. — Size, type, class. — Flange and bolt material. — Gasket specification. — Approved bolt stress and source. — Approved tightening method and settings/tools to achieve approved bolt stress. — Lubricant used.

and maintenance of the asset. — Records of any modification, exceptions or deviation from standards with the joint. Whilst this data can be kept in hard copy format, a database system is recommended due to the high volume of data required and the ease of searching and retrieving data that computerised systems offer. 5.2.1.4 Joints included in the database It is recommended that the operator as a minimum keeps records for all joints on critical services and those on other services which have a history of leakage, or potential to leak, and that this is kept with any relevant procedures for monitoring the specific joint. To minimise the possibility of the problem resurfacing, methods for countering the leak should be included within the individual joint records. There are however benefits in maintaining a system for all joints, in terms of safety, efficiency and traceability. Statistics show that using a system to control joint integrity will reduce the effort required to achieve a successful integrity test.

5.2.1.2 Additional data Additional data can be recorded to make the system more user friendly and effective as a planning tool, such as:

— Status of the joint. — Any exceptions or anomalies regarding the joint. — Location description of the joint.

5.3 REVIEW 5.2.1.3 Joint history — Starting at the construction stage: Records of assembly, break out, reassembly, inspections and controlled tightening. Including personnel involved, equipment and procedures used. Results and measurements taken where appropriate. — Records of inspection and testing of the joint. — Records of subsequent disassembly, inspection, assembly, tightening and testing during operation

The entire process and outcome should be reviewed by managers and members of the work team. Identified improvements to the activities, work scopes and procedures should be recorded and retained for when the work is next repeated. Information of performance and good practice should be shared with industry.

29


30

6 MANAGEMENT OF LEAKS However, there are a significant number of industries where good business practice or regulatory requirements make it essential to formally assess loss of containment events and determine root cause and measures to prevent recurrence. When a leak or incident occurs, a common approach to manage such situations is an engineering risk assessment which utilises the collective skills within the organisation to address three fundamental questions:

6.1 INTRODUCTION

The objective of a correctly designed and installed bolted joint is to provide a long-term tight seal and prevent ingress or egress of fluids through the joint. However, leaks can occur and the Duty Holder or operator has overall responsibility to manage this situation. This section introduces some important features that may be required of the management system for pressurised systems after joint make-up. These include:

— Safety impact to ongoing operations - is it safe to continue to operate the plant? — Environmental impact - what is the environmental impact of continued operations? — Economic cost - what is the business cost?

— Management of leaks and releases and the appropriate engineering risk assessments that might be required. — Definition of leaks. — Integrity testing of joints as an assurance measure of joint tightness. — Potential options for repair or replacement of leaking joint s.

An engineering risk assessment should be performed to establish whether it is acceptable to continue operations. The assessment should also identify control or hazard mitigation measures required such as increased surveillance or plant de-rating. Alternatively the outcome of the assessment might indicate that there is significant hazard with continued operation and that immediate shutdown, repair or replacement is required. It should be noted that most companies have some form of environmental policy which requires recording emissions from process systems. In the UK statutory rules require reporting of leaks and emissions depending upon the fluid and the magnitude of the leak.

6.2 ENGINEERING RISK ASSESSMENT OF LEAKS

The degree of review or assessment of leaks will depend upon the industry, the nature of the process fluid and pressure and temperature conditions; all factors affecting the criticality of the joint (see Section 3). In some situations it may be acceptable for joints to leak.

31


6.3 STAGES AT WHICH LEAKS OCCUR

6.4 CORRECTIVE ACTIONS

Leaks often occur when the joint is under test, as described below. However, regardless of when a leak occurs, it should always be investigated and recorded in the records and data management system (see Section 5). This will not only store useful data against the joint and assist in preventing the same issue arising on the same joint in future, but will also enable trend analysis to prevent leakage on other joints. A leak decision and analysis tree, such as the example illustrated in Figure 6.1, can assist in determining an appropriate course of action and also provides input to the data management system.

Where a release is identified, a corrective action must be carried out to secure a tight joint. Some measures include:

6.3.1

— Depressurise the system, and check load on bolts. — Identify root cause of the problem (and notify appropriate authorities). — Depressurise and completely remake the joint after component inspection. Other measures such as hot bolting and tightening of live joints are not recommended.

Standard pressure (strength) test 6.5 DEFINITION AND DETECTION OF LEAKS

A leak occurring during a strength test is an indication that there is a major issue in the installation of the bolted joint. Although the joint is subject to a higher than working pressure at this stage, it is not subject to temperature or cyclic loading and therefore leakage during this activity suggests poor assembly or applied bolt load. Identifying and correcting the cause is essential for reliable operation of the plant. 6.3.2

Leak test

Service testing

Where the service test identifies a slight leak, there will be a temptation to apply more bolt load to seal the leak. This may be successful, but if the load was correct in the first place then consideration must be given to identifying why the joint leaked (see also 6.3.2 above). 6.3.4

A release of hydrocarbon or other hazardous fluid should be recorded as a leak when the release rate is equal to or greater than: — Liquid leaks: A release rate of one drop per 15 seconds (four drops per minute).

A leak test is not a replacement for correct joint makeup and tightening; rather, it is merely part of the assurance process. Where a joint is just failing a leak test it is tempting to increase bolt load. However if the bolt load was correctly applied in the first instance, increasing the bolt load could be hiding a problem, e.g. a nipped gasket or grit on the gasket, which will manifest as a service leak later. 6.3.3

The following definition of a leak is widely used in the upstream oil and gas industries:

Leaks occurring operation

during

start-up

or

These leaks potentially have the greatest impact not only for safety but also commercially as they will either delay start-up or stop production.

— Gas leaks: A release that will cause a hand-held gas detector 10cm 'downwind' of the release source to indicate 20% Lower Explosive Limit (LEL). The most likely method of detecting a leaking bolted joint is observation by operations and maintenance personnel or inspection personnel during routine operation in the plant. There is no substitute for 'line walking' as most leaks are of relatively small magnitude. The more significant leaks may also be detected by plant safety systems such as gas detectors or, in extreme cases, by the process control system. All leaks should be tagged and entered in the maintenance system for repair and the record and data management system as soon as is reasonably practical. It may also be reportable. An emission from a joint with a lower release rate than a leak is described as a seep. These too should be tagged and periodically checked to ensure they h ave not worsened, and be entered into the maintenance system for repair at the next scheduled service for that item. It should also be recorded in the records and data management system.

32

MANAGEMENT OF LEAKS

START

PASS OR FAIL?

PASS

FAIL Joint failure leakage rate above the target acceptance criteria

Is this a single failure or are there numerous failures in the system?

SINGLE

NUMEROUS

TIME OUT

YES

Technical review required

Is there a common style of joint that is failing?

NO

YES

Is the leakage rate above the maximum acceptance criteria?

NO

NO

Update workpack with new torque/tension figures

Is the technical authority willing to approve the leakage result?

Is the torque/tension applied to the joint correct for this application? (Check procedure and all assumptions made)

NO

YES

YES Correctly apply the torque/tension in the field

NO

Has the torque/tension been correctly applied in the field?

YES Break the joint and investigate failure mode. Apply rigorous checks. Reassemble and retighten joint to approved method.

Figure 6.1: Example leak decision and analysis process

33

SUCCESS


achieved, the only option may be to shut down the plant to carry out the repair. An example of such a situation is given in Figure 6.2 which shows a flange with gasket seating face pitting. Corrosion at the bottom of the pipe has caused metal loss of the entire gasket seating width and resulted in a leak. In this case a unit shutdown was undertaken to replace a pipe spool.

Leaks may pose special risks in confined spaces such as pits, trenches, buildings and modules. Examples include concentration of toxic or poisonous gases or heavier than air asphyxiates such as argon or carbon dioxide gas. Fugitive emissions as described in the IPPC regulatio ns are beyond the scope of this section. It is assumed that issues such as Best Available Techniques (BAT) for sealing have been addressed by the specifiers and designers of joint systems.

Carry out on-line repair On-line repairs to live plant have been carried out using techniques such as encasement clamps and various forms of glass and carbon fibre wraps. These are considered to be engineering repairs and need appropriate technical skills and installation competences. Specialist service companies can provide this type of product on a world-wide basis. Some high level guidelines on safety considerations are presented in EEMUA publication 199. Guidelines on requirements and qualification of repairs to corroded or damaged piping using composite wraps are presented in ISO document ISO/PDTS 24817. A typical clamp type repair is illustrated in Figure 6.3.

6.6 MANAGING LEAKS AND REPAIRS An engineering risk assessment will provide a technical basis for reviewing repair options which can range from shutdown and repair/replace to continued operation with no intervention. The choice of options may be further restricted by Company policy depending upon the type of facility. In many industries, unplanned shutdown for repairs is normally avoided wherever possible. Whatever the planned course of action, it should be formally documented before work begins and carried out in a safe, managed and controlled manner. No matter what the circumstances, the temptation to tighten up the joint beyond design parameters should be resisted. There are a number of candidate repair and replacement strategies. These include:

Detail design considerations include: Definition of the expected design/operating life of the repair. Impact of fluid on bolting, e.g. exposure of some bolting to fluids with H 2S or chlorides. Pressure end cap forces. Site constraints such as insufficient space to install a clamp.

— Continued service accepting the joint leakage: It may be acceptable to permit continued leakage from the joint based upon the engineering risk assessment and environmental impact until a planned shutdown.

6.7 LEARNING FROM LEAKS — Continued service operating with de-rating: It may be acceptable to permit continued joint service by imposing a control measure such as derated duty point or downgrade condition. This may be appropriate where the leak rate is pressure or temperature activated.

In order to prevent future leaks lessons should be learnt from past incidents. Operators should develop a process to capture data in a form that can be readily reviewed and analysed. The process should aim to:

— Isolate and repair the leaking joint (line isolation): There may be sufficient valves to isolate the leaking flange and allow maintenance to be carried out. This may prove difficult in some plants where the valves do not provide tight shut-off or the required level of isolation as required by the company's safety policy cannot be achieved.

— Improve the quality of information gathered on join t leaks. — Identify and better understand the causes of failure(s). — Provide data for the hydrocarbon leaks database. — Provide data for long-term learning on leak occurrence. — Ensure periodic review and learning.

— Shutdown unit, isolate and repair joint: In some instances, where safe isolation cannot be

These details should be recorded in a data management system.

34

MANAGEMENT OF LEAKS

Figure 6.2: Flange with gasket seating face pitting

Figure 6.3: Encasement clamp repair on a 24 in. seawater line to stop leak on stub of composite flange

35


36

7 IN-SERVICE INSPECTION 7.1 INTRODUCTION

In-service inspection of bolted joints is an integral activity to ensure the continued integrity of the joints and as such should be built in to all relevant inspection programmes. This section looks at the possible damage that can occur, the inspection methods available for detection of defects and the mitigation measures that can be put in place to minimise such degradation. A summary of the key issues addressed in this section is included in Table 7.1.

— Flange orientation, particularly blind flanges installed horizontally, allowing water to collect in the holes. Stud bolts in blind flanges in firewater mains have suffered this form of damage.

7.3 DEGRADATION MECHANISMS

A number of degradation mechanisms can contribute to the failure of a bolted joint , most of which are corrosion related. Figures 7.1 - 7.11 (see pages 39-41) illustrate the most common problems found.

7.2 RISK ASSESSMENT 7.4 INSPECTION TECHNIQUES

For high-risk join ts it is recommended that methods are implemented for monitoring bolt stress to ensure that pre-load is maintained. Generally, a risk assessment should be carried out to determine the inspection requirements. The following factors should be considered for specific joints:

7.4.1

Non-destructive testing

The most common method of in-service non-destructive inspection is visual inspection, normally carried out as part of general visual inspections of pipework or structures as opposed to specific bolt inspections. The limitations of this method are that only the external parts of the joint are visible which will detect loose bolts and corrosion; however the extent of surface degradation on the strength of the joint is difficult to measure. Where bolt threads or nuts show the effects of significant corrosion then further investigation should be undertaken to ensure that the joint is still fit for purpose. Some Operators use a Performance Standard to quantify the extent of bolt degradation. An example of a PS for low alloy steel bolting material is shown below: Bolts shall be visually examined for evidence of corrosion and other defects (mechanical damage or

— High temperature pipework may cause bolts to creep and cause leakage. — Large numbers of temperature cycles can cause the bolts to loosen. — Mechanical vibration or shock loading may cause the flange bolts to loosen. — Areas of high external corrosion may cause the bolts to lose integrity. Susceptible locations include insulated pipework, bolts open to harsh environments or those subject to deluge system tests. — Internal corrosion can cause flange faces to lose material.

37

Table 7.1: Summary of key issues Damage type

Typical conditions

Inspection technique

Mitigation measures

General corrosion

Exposed areas

Visual, sample removal, Cylindrical Guided Wave Technique

Material selection, thread protectors, coatings

Galvanic corrosion

Dissimilar metals flanges, bolts, gaskets

Visual

Material selection, gaskets, bolt/flange insulating kits, weld overlay

Localised bolt corrosion

Dissimilar metals, exposed areas

Sample removal, visual, Phased Array UT, CGWT

Material selection

Crevice corrosion

Exposed areas

Visual, sample removal

Material selection

Fatigue

Joints subject to vibration, cyclic loading

Visual, Phased Array UT

Pipework design

Creep

High temperature applications

Time-of-flight UT

Material selection, ASME SA-453

Stress corrosion cracking

A combination of a chloride-containing environment, susceptible material and tensile stress

Phased Array UT

Material selection

Hydrogen embrittlement

Hydrogen can form on surface during manufacture or be caused by Cathodic Protection

Visual, highlighting any corroded High Strength Fasteners for replacement

Material selection - most common in High Strength Steels e.g. ASTM Standards A345 Gr BD, A490 and A547 Thread protectors and coatings

Flange face corrosion

Pipework containing a corrosive medium, dissimilar materials

Intrusive visual, UT

Material selection, gasket selection, weld overlay

— Ultrasonic inspection of flange faces using shear wave transducers - detects flange face corrosion and erosion. — Black light NDT of threads and body on bolts that are to be re-used on high critical joints - detects stress cracking.

cracking). Bolting showing signs of mechanical damage to plain shanks or threaded portions with in the stressed portion or any cracking shall be replaced with new bolting. Bolts, studs and screwed fasteners that have corroded such that the diameter of the smooth shank or the major thread diameter is less than 90% (i.e. 10% loss in diameter) of the nominal size, after removal of the corrosion product, shall be replaced with new bolting. A number of more specialised techniques are available which can be used to check for specific conditions; these include:

7.4.2

Destructive testing

Where degradation is thought to have occurred and assessment is not possible through non-destructive techniques, sample removal of bolts for destructive testing can be carried out to estimate if joints are still fit for purpose. Finite element analysis has also been used to model the effects of progressive removal of layers of bolt material.

—

Phased Array Ultrasonics - detects thread wear and cracking from the bottom of the threads, as illustrated in Figure 7.11 (see page 41). — Time of Flight (TOF) UT - measures bolt elongation. — Cylindrically Guided Wave Technique (CGWT) detects corrosion wastage.

7.5 DEFECT MITIGATION MEASURES In-service inspection requirements can be greatly reduced by designing the bolted joint to include 38

IN-SERVICE INSPECTION

measures which will reduce the risk of degradation due to mechanical damage and corrosion. The following are some commonly used measures:

Coatings Bolts can be supplied with a variety of life extending surface treatments such as hot dip spun galvanising, which research shows offers the best long term protection. Zinc, PFTE or electroless nickel are also used.

— Material selection Corrosion-resistant alloys e.g. stainless steel, duplex and cupro-nickel alloys are used. However, they can suffer specific rapid failure mechanisms such as stress corrosion cracking. In addition, the high costs of these materials restricts widespread use.

Cathodic protection Used for underwater applications. However, there is usually a need to apply coating to the pipework and flange joints to minimise the risk of hydrogen embrittlement.

— Thread protectors Neoprene, polyethene and aluminium are common. However, the potential for loss of thread engagement strength needs to be assessed.

Flange protection Gaskets (material selection is important to avoid galvanic corrosion), flange protectors, coatings.

Figure 7.1: General corrosion (General corrosion of flanges and bolts.)

Figure 7.2: Galvanic corrosion (1) (Galvanic corrosion where dissimilar materials have been used for the bolts and one flange.)

Figure 7.3: Galvanic corrosion (2) (Galvanic corrosion where dissimilar materials of bolts only has been used.)

Figure 7.4: Localised corrosion (Severe localised corrosion of bolt body.)

39


Figure 7.6: Fatigue (Failed bolt displaying typical fatigue failure characteristics.) Figure 7.5: Crevice corrosion (Advanced crevice corrosion of a stainless steel bolt.) Crack Initiation Site

(6.7X)

Crack Initiation Site

Figure 7.7: Stress corrosion cracking (Failure surface of bolt which had been subject to the combined influence of tensile stress and a corrosive environment - a typical example would be austenitic stainless steel in high chloride conditions.)

Figure 7.9: Flange face corrosion (Flange face corrosion in seawater pipework.)

Figure 7.8: Hydrogen embrittlement (Fracture surface of a bolt that resulted from hydrogen embrittlement cracking.)

Figure 7.10: Galvanic corrosion (Galvanic corrosion of clamped assembly seal ring.)

40

IN-SERVICE INSPECTION

Figure 7.11: Phased array ultrasonics

41


42

Bolted Joints Integrity Management Guidelines

Recommend Documents