NDE of underwater welded steel structures

The International Institute of Welding

Non-Destructive Examination of

Underwater Welded Steel Structures

IIS/IIW Document V-1097-97 (IIS/lIW-1372-97)

V S Davey, 0 Fmli, A Raine and R T Whillock

A B I N G T O N P U B L I S H I N G \\'oodhcad I'uhlishing Limited in association with l'hc \\!clding Institute

Carnhridge En~land

Published by Abington Publishing, Abington Hall, Abington, Cambridge CB21 6AH, England www.wood headpu blishing .com

First published 1999, Woodhead Publishing Limited

0 1999, The International Institute of Welding

Conditions of sale All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher.

No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of product liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein.

British Library Cataloging in Publication Data A catalogue record for this book is available from the British Library

ISBN-1 3 978-1 -85573-427-2 ISBN-1 0 1-85573-427-3

Printed by Victoire Press, Cambridge, England

CONTENTS

Preface

1 1.1 1.2 1.3

2 2.1 2.2

3

4 4.1 4.2 4.3

5 5.1 5.2 5.3

6 6.1 6.2 6.3

7 7.1 7.2 7.3 7.4 7.5 7.6 7.7

Introduction and scope 0 bjectives Scope Further information

Definitions and glossary Definitions Glossary

Components of examination

Personnel, qualifications, and quality assurance Person n el Qualifications Quality assurance

Procedures Principles Content Additional requirements

Pre-inspection activities Briefing Location reference marking Cleaning

Inspection for flaws Magnetic particle inspection Alternating current potential drop Eddy current and alternating current field measurement Flooded member detection Radiographic methods Ultrasonic inspection methods Visual inspection

V

1 1 2 3

4 4 5

9

10 10 10 11

13 13 13 14

15 15 15 15

17 17 20 21 22 23 23 28

8 8.1 8.2

9

10 10.1 10.2

11

12 12.1 12.2 12.3

Inspection for corrosion Cathodic protection potential measurements Wall thickness measurement

Passive NDT methods

Post-inspection activities Recording and retention of data Inspection after repair

Introduction of new equipment

Remote operated vehicle deployment Introduction Present status Conclusions

30 30 32

33

34 34 34

36

38 38 38 40

Appendix A - Further information 41

Appendix B - Summary of underwater NDT qualification schemes 66

Bibliography - Selected reference documents on NDT of offshore construction 68 1 -Government bodies 68 2-Certifying authorities and classification societies 70 3-Advisory bodies and others 73 4-Associated documents 75

iv

This is a revision and replacement of Document IIS/IIW-1033-89, 'Information on Practices for Underwater Non-Destructive Testing' and was prepared by a task force of Working Group 2, 'Inspection of Offshore Welded Constructions' of Commission V Quality Control and Quality Assurance of Welded Products' of the IIW consisting of:

Victor S Davey, (Formerly HSE) UK Olav Ferrli, Det Norske Veritas , Norway Alan Raine, Technical Software Consultants Ltd, UK R Toby Whillock, (Formerly BOMEL) UK

I INTRODUCTION AND SCOPE

This following document has been produced by Working Group 2 of Commission V of the International Institute of Welding and supersedes previous versions.

1 .I Objectives

The objectives of this document are to give technical guidance and information on non-destructive testing (NDT) techniques suitable for use underwater on welded steel constructions.

When properly used, modern underwater NDT techniques are capable of similar performance to comparable onshore NDT equipment. However, as with all technologies, the quality of the results depends on appropriate selection and operation of the equipment. It is accordingly the primary purpose of this document to assist in the production of high-quality reliable results from underwater inspection by promoting what might reasonably be considered good practice at the time of writing ( 1 997).

This document is not a code and neither is it a statutory instrument with legal power, nor does it seek to lay down minimum standards. Nonetheless, a vital constituent of any inspection practice is to accomplish the work with minimal risk to personnel, and it is expected that organisational schedules and procedures will consider this aspect in adequate detail. Notwithstanding the various notes in this document concerned with safety, nothing in this text should be taken as diminishing or removing the onus placed on the individual to adopt safe working practices and to take the final responsibility for the consequences of his own actions.

Furthermore, it is not the intention to produce prescriptive requirements or in any way to restrict underwater inspection by prohibiting certain practices or by inhibiting the introduction of new equipment or techniques. However, where in the opinion of the Working Group, the use of items of equipment in some configurations may give rise to undesirable consequences, this will be pointed out.

1

1.2 Scope

The document is intended for the external inspection of welded steel constructions used offshore such as oil and gas platforms, pipelines and risers, harbour installations, ships' hulls, etc, though the practices discussed should also be suitable for inspection of cast components with appropriate modifications to reflect the different metallurgical properties. Testing of parent material as well as welds is covered.

The document applies not only to methods used manually by divers in water or in a habitat, but also to remotely-controlled or automated methods used from submersibles, atmospheric chambers, or remotely-operated vehicles (ROVs) with inspectors remote from the work site. Only the commonly used NDT methods are described in detail, with less-commonly used methods being given more superficial description.

Since this document is concerned with the external inspection of welded structures underwater, it will not consider inspection in the splash zone or above water, and it does not cover inspection inside pipelines by inspection pigs and suchlike. Similarly, information is not given on diving or operational aspects of underwater inspection activities. It must, however, be reiterated that, as a minimum, proper precautions need to be taken with respect to diver safety and operation, particularly in relation to electrical hazards, implosion of equipment, and the risk of water penetration into equipment. In saturation habitats, attention should also be given to helium penetration into equipment under high external pressure.

The document does not address pre-inspection planning and scheduling, or the selection of areas for inspection. Reference should be made to the appropriate company documentation on these subjects. In addition, the use of inspection results is not covered. Both these activities are strongly dependent on a host of factors such as: construction type and environment, consequences of failure, national regulations, agreement between vendor and purchaser, etc. The choice of the most suitable NDT methods for an application is also dependent on similar factors and only broad guidelines will be given.

2

1.3 Further information

Information in the main part of this document may be expanded upon in the background text given in Appendix A. Reference should also be made to other IIW documents and to reports such as those listed in the Bibliography. Particularly recommended are reports published by the United Kingdom's Marine Technology Directorate Ltd: 'Underwater inspection of steel offshore installations: implementation of a new approach; MTD Ltd Publication 89/104. Publications by the UK Health & Safety Executive (HSE) are also a good source of information for offshore NDT.

3

2

The following terms are defined for use in this document.

2.1 Definitions

Non-des truc five testing fNDT), non-destruc five examination fNDEJ or non-destructive inspection - the carrying out of inspection causing little, if any, damage to the test item, usually by physical test methods in conjunction with visual assessment, either at the worksite by the inspector or via a remote telecommunications link. The words testing, examination and inspection are thoughout this document regarded synonymous, reflecting the common, and somewhat arbitrary terminology used in the offshore inspection industry.

Non-destructive evaluation fNDq - the comprehensive assessment of the results of NDT, including the assessment of any inspection indications, which may comprise formal fracture mechanics analysis and other engineering reliability studies. The overall results and assessment will normally be drawn together in a formal report.

indication - a response from the inspection system suggesting the existence of a flaw.

Flaw - an indication confirmed by use of two or more different inspection techniques, including close visual inspection. A flaw is usually, but not always, a crack. The term is used in preference to 'defect' in accordance with European Standards, the designation 'defect' being considered as connotating 'defective' or 'unfit for purpose'. A flaw is a departure from perfection that does not immediately condemn the article as unacceptable but that nevertheless should be considered in an assessment of reliability. Further information may be found in IIW documents.

Spuious indication or false call - a plausible, but discounted, indication that has, on further investigation, been assessed as not being a flaw. It would normally still be noted in the records of the inspection.

4

Probabfity of detection [POD) - a measure of the performance of an non-destructive testing (NDT) system intended to detect flaws, in terms of the chance of finding a flaw. It is a complicated variable, depending on: NDE technique and equipment: skill of the operator: the inspection conditions: the definition, nature, location, and size of flaw and other factors. It is often presented in the form of a curve as a function of flaw size (usually length). Considerable care should be taken not to mis-apply such curves as their interpretation is far from clear-cut and demands some knowledge of statistical methods.

Accuracy of s/ang lA0.J) - a measure of the performance of NDE equipment in giving an estimate of the size of a flaw. In the important case of surface-breaking flaws, the depth is quoted as being the significant variable for causing brittle fracture. Accuracy of sizing is a complicated concept, being heavily dependent on a number of factors, some ill-defined. The concept of accuracy of sizing is less well established than POD but includes systematic and random errors of size estimates.

2.2 Glossary

Magnetic parfcle inspection fMPIJ or magnetic parfcle testing fMPTj - a combined visual and electromagnetic technique relying on the attraction of minute ferromagnetic particles to magnetic flux diverted from an applied field by a flaw. The particles emphasise the length of the defect rendering it more visible to the inspector. The method is suitable for detecting surface-breaking flaws, but may also have a limited capability for anomalies very near to the surface. The magnetic field may be applied by permanent magnets or by electromagnetic induction using coils, a yoke, or prods. In many ways, MPI can be regarded as a quasi-standard method against which other NDT methods are often compared, though it is known to be very sensitive to the skill of the inspector.

Alternating current potential drop [Acpd) - a contacting electromagnetic technique using the skin effect of the alternating current from which an assessment of crack depth can be made on surface breaking defects. ACPD is a sizing technique rather than a detection method.

5

Alternating current field measurement (ACFM) - a non-contacting electromagnetic inspection technique developed from ACPD (qv) and capable of detecting and sizing surface breaking flaws. A uniform alternating current is induced into the material under test and the disturbances in the surface magnetic field produce information, which is collected using an inspection head, which may be a probe or array. This is fed through an umbilical lead to a computer for data recording and analysis which then produces data on the length and depth of the defect. Because of its ability to produce length and depth data from coated welds it has also been applied using ROVs to deploy the technique.

Eddy current (EC) - a non-contacting electromagnetic inspection technique relying on the perturbation by flaws of eddy currents induced in a metal surface by a probe. Usually the same probe induces the eddy currents and detects the perturbation signalling an indication. EC is principally a detection technique for surface- breaking flaws, although some depth penetration can be gained depending on the frequency of the eddy current field and the metal, and some systems can size indications. The recent trend in EC systems has been broadly similar to that for ACFM, with often a high degree of computerisation allowing data recording and analysis. A computerised system has also been applied with an ROV.

Flooded member detection fFMD) - a generic term covering a variety of approaches to detecting whether a tubular member has been penetrated by a through-thickness crack and has then been flooded with water. Two main types of FMD are used: systems using ultrasonic techniques; and gamma-radiation methods. FMD differs somewhat from the other NDT methods in that it is capable of detecting flaws only once they have gone through-thickness, hence giving only a short residual life for the affected component. Against this, FMD is very fast and so tends to be used as a supplement for other, slower, inspection methods, in order to cover a large amount of the structure. The technique has been adapted for use with ROVs.

Radiography - various forms of this technique for buried flaws are available, principally using X-rays and gamma-radiation sources. The radiation hazard tends to restrict their use offshore, though they have been used for some difficult applications unsuitable for other NDT methods. The capability of detecting and sizing flaws is highly

6

dependent on the skill of the inspector and on favourable flaw orientation. Limited ROV applications have been carried out.

Ultrasonic inspection [UU or ultrasonic testing [UTj - use of ultrasonic waves induced in a probe and transmitted across a narrow gap into the steel. The term is general and may cover a variety of different effects, but it commonly refers to pulse-echo systems using the amplitude information of signals. A large variety of arrangements of probes and systems is available, and they are principally used for characterising (locating and sizing) deep flaws rather than those near the surface and also for wall thickness measurement. See also TOFD and UCW. ROVs have applied ultrasonic wall thickness measurement for corrosion assessment.

firne-of-flight diffraction [TOFDj - a variety of ultrasonic detection systems relying on the time taken for signals diffracted from a flaw to extract information. TOFD systems tend to be heavily computerised and appear to be capable of better quantitative flaw sizing than conventional ultrasonic techniques under opportune circumstances.

Ultrasonic creeping wave [UCWj - a variety of US using a special probe to generate longitudinal surface waves confined to a thin layer of steel just under the steel/water interface and hence is well-suited to detection of surface-breaking flaws. Other than the probe, conventional US equipment can be used.

Close vkual inspection [CVV - detailed examination of a small area from a short distance, usually carried out by a diver. CVI should automatically accompany any other inspection method, for although it is, on its own, of minimal reliability for detecting flaws very useful supplementary information can be obtained by an experienced inspector.

General visual inspection {GVV - the general inspection of the condition of a structure, usually made by a 'flying eyeball' remotely- operated vehicle (ROV), in which the aim is to gain an overall appreciation of the state of the installation with respect to missing or damaged members, the extent of marine growth, debris, and scour, and suchlike.

7

Diving inspector or NDT operator- a diver trained in the application of non-destructive inspection techniques and reporting and sometimes the interpretation of results obtained from the inspection.

Remotely-Operated Vehicle [ROW - an underwater vehicle of various dimensions often used as a replacement for diving inspectors which can be used to deploy non-destructive tools depending on the complexity of the inspection required. Interpretation is carried out by an ROV inspector.

8

3 COMPONENTS OF NON-DESTRUCTIVE EXAMINATION

In general, in order to perform a non-destructive evaluation the following elements are necessary:

0 Personnel 0 Equipment 0 Procedures 0 Acceptance criteria.

Each of these elements must be adequately taken care of in order to perform an examination to proper quality standards for the task in question. The personnel must be sufficiently trained and qualified, the equipment must meet certain requirements on sensitivity, operability, etc, and procedures must be clearly recorded and understood to ensure that the necessary steps are followed.

An essential part of the procedures is to specify clear criteria for the reporting of findings, to ensure that the utmost benefit is derived from the inspection. Inspection is not a clear-cut matter and the act of inspection constantly relies on human judgement and discrimination, and hence is usually highly affected by the ability and experience of the inspection team. Thus, in order to ensure consistency, it is necessary that the procedures address this issue and, in particular, find the correct balance between excessive reporting of inconsequential trivia, which results in slow and ineffective inspection, and the alternative risk of ignoring significant indications. The specification of what constitutes a reportable indication and appropriate follow-up procedures will depend on the policy of the owners of the installation.

In other words, the actual practice of inspection must take place against the background of the Operator's IMR (inspection, maintenance, and repair) philosophy in addition to complying with any local statutory requirements and regulations.

9

4 PERSONNEL, QUALIFICATIONS, AND QUALITY ASSURANCE

4.1 Person ne I

Underwater diving and ROV NDT inspectors should be at least as qualified as their above-water counterparts to ensure adequate inspection. Special quality assurance measures may be needed to compensate for the working environment and possible lack of operator expertise, and will be outlined in Section 4.3.

4.2 Qualifications

Underwater NDT inspectors should be proficient in both diving and NDT. Despite the recent trend in much NDT equipment to de- emphasise the diver’s role in inspection by placing the onus on topsides inspectors linked by telemetry, it is still most important that the diver be competent in his own right as an inspector.

4.2. J Diving qualfica fions

Diver-inspectors should be well-qualified divers and have extensive experience in the types of dive required, in order to allow maximum concentration on the NDT work.

4.2.2 NDT qualfica fions

Underwater diving and ROV NDT operators should be well-trained in the NDT techniques in question and have proved their abilities by examination, either based on a recognised above-water certification scheme with supplementary underwater tests, or a special scheme for underwater N DT.

Many NDT techniques now fall within the coverage of several internationally recognised qualification schemes such as CSWlP (Certification Scheme for Weldment Inspection Personnel) or the equivalent. However, some of the newer NDT methods inevitably fall outside these established schemes and reliance must be placed on training provided by the manufacturer of the NDT equipment or similar competent bodies, together, ideally, with relevant experience.

10

As a minimum, the inspector in such cases must undergo specialist training and be closely supervised by personnel competent in the use of the system under operation.

The theoretical and practical training should be at least as extensive as for an above-water operator for the NDT technique in question, and should reflect the difficulties of working underwater as far as possible. As noted below, tests should be as representative of the actual inspection conditions as possible.

In general, in addition to NDT knowledge, it is highly desirable for the inspector to have some theoretical background in materials engineering, underwater structures, and knowledge of flaws and damage likely to be encountered.

4.2.3 Surveillance personnel

Above-water personnel used to supervise and survey the NDT work should be properly trained and qualified. They should have detailed knowledge of the NDT techniques in question, the structure under survey, and the working conditions, in order to guide and assist the diver-inspector properly.

4.3 Quality assurance

The following measures may help to assure the quality of the inspection.

0 Verification tests

Verification tests are useful to check the adequate functioning of an examination system (personnel, equipment, etc) and are often required by companies engaging inspection firms, government bodies and certifying authorities prior to an inspection job. Verification tests are used more frequently for underwater NDT than is the case above water. More comprehensive verification may be required for novel methods and equipment (see Section 11) . Verification tests should attempt to simulate the conditions to be encountered during inspection as closely as possible, ideally by using test samples of

11

representative geometry, made from similar materials, and containing flaws of the kind expected.

b Surveillance

In addition to its use for quality assurance, surveillance may be required by company procedures in the interests of safety. It may be carried out by inspectors above water, using aids such as underwater TV cameras mounted on ROVs, direct communication lines to underwater NDT inspectors, and/or examination of recording equipment above water.

0 Stringent requirements for reporting and documentation

Inspection performance and results may be recorded by still photography or video recording.

b Audit by re-inspection by a third party

b Tuition and familiarisation.

It is obviously important that an underwater NDT operator is given adequate time to become familiar with new equipment and techniques before their use underwater, particularly if the inspection is to involve unusual applications or areas.

12

5 PROCEDURES

5.1 Principles

It is likely that company procedures will be in place to define the conduct of inspection, but the following will probably be required as a minimum. As with all documentation, clarity of expression is essential so that any ambiguity is avoided. The document should be written in a style capable of being understood by all personnel concerned, with due recognition of any language difficulties. Procedures should define any unusual terminology.

It should be recognised that the NDT procedure will have to comply with the procedure for the associated diving operation as well as forming an integral part of an overall inspection programme. Whilst procedures will cover a variety of functions, they should deal with all relevant aspects of the operation, including: the areas to be inspected, noting unusual and relevant features and hazards; the location and positive identification of inspection sites: a checklist of equipment and operating instructions sufficient for the safe and reliable operation of the NDT systems; the inspection and reporting procedures and terminology to be adopted; the recording of the results; and any emergency instructions. Procedures must be kept up to date and refer to the actual equipment in question.

It is envisaged that the procedures would comply with the company's QA procedures.

5.2 Content

A typical inspection procedure will probably comprise:

Objectives of the inspection Information on the subject of the inspection Reference codes, standards, and other documentation Personnel requirements (numbers, functions, and qualifications) Preparation for inspection Equipment lists and details Calibration of equipment (before and after examination)

13

0

0 Interpretation of findings 0 Recording of results 0 Special quality assurance measures.

Exa mi nu tion (step-by-step work description)

The procedure should be as complete as possible in sufficient detail to allow for the efficient execution of the work, including guidance on dealing with any exceptional circumstances, such as criteria for referral to a higher authority. The intention is to make the best use of expensive resources and tight windows in schedules.

5.3 Add i ti on a I req u i rements

Additional considerations include:

0 Easy means of reliable and speedy reporting requiring a minimum of underwater writing. Tape recording, videotape recording and still photography can be used as aids.

Interaction between underwater diver-inspector and topsides supervisor. This requires good communications and clear definition of responsibilities. As an aid, slave monitors or recording equipment can usually be linked into the primary display, including for use by the diver-inspector.

Special quality assurance measures to ensure reliable and efficient testing.

14

6.1 Briefing

Before the start of inspection, the team members should be properly briefed on the task to be undertaken, using the written procedures as a basis. Particular attention should be paid to resolving any uncertainties and ensuring that any peculiarities of the equipment or the operation are understood, especially when no direct supervision is possible.

Since the description of the findings is often crucial to the success of inspection, the Inspector and any associated personnel should be aware of the terminology to be used, especially any non-standard jargon or terms. Reference may usefully be made to the UK Health and Safety Executive publication: ‘A Handbook for Underwater Inspection’; OTI 88-539; HMSO 1988.

6.2 Location reference marking

On large underwater constructions a good location reference system is extremely important to pinpoint spots for repair or repeated in- service inspection, and suitable aids must be available in order to permit the inspector or ROV pilot to be able to navigate to and identify the correct subjects for the inspection. Flaws and areas examined for corrosion should be marked to allow re-examination on subsequent occasions. When using ROVs it may be required to do a local three-dimensional mapping of the inspection site using suitable probes in order to locate the inspection probes accurately.

6.3 Cleaning

Most underwater NDT methods require some cleaning of the marine growth to be done, the severity of the cleaning depending on capability of the NDT equipment to operate on various types of surface, and on the type of marine growth. If the structure has been given protective coatings this will also be important.

The choice of cleaning method is thus influenced by a number of factors, including the extent of cleaning required, the type of marine

15

growth (soft or hard deposits), and the type of inspection to be carried out.

The most common cleaning techniques are water jetting, grinding, wire brushing, and abrasive blasting. Low-pressure water jetting guns can be used down to 40 m water depth. Hand-held tools such as scrapers and wire brushes are also used, and although they are slow and relatively ineffective by comparison with the power methods, they do find application for inspection of small areas.

Whichever cleaning method is chosen, it must not cause excessive damage to the surface. Care must be taken not to introduce new stress concentrations arising from notches or deep scratches produced during cleaning. The cleaning must not peen over the edges of any cracks, or work-harden the surface as this can affect some NDT methods such as MPI or eddy current examination. Water jetting, wire brushing and abrasive cleaning to bare metal are generally acceptable for MPI or ultrasonic inspection. Grinding should be done only as a last resort to remove hard deposits, and needle guns should be avoided. Reflective metal finishes will make photography difficult and may hinder MPI and so should be avoided. A matt bare metal finish is normally achieved when cleaning is specified to IS0 8501 -1 : 1 988 Grade 2%.

Most eddy current and ACFM systems can be used to carry out weld inspection without having to clean the welds down to bare metal and in some cases they can be applied with certain types of protective coatings in place ( See section 7.3)

Cleaning equipment using power tools or high-pressure water jets is capable of inflicting severe injury or even death in the event of an accident. Supervisory personnel responsible for the cleaning operation shall ensure that all personnel concerned in the inspection have been properly trained and are familiar with the use and characteristics of the equipment, and that the equipment is fully serviceable and complies with relevant safety standards. The cleaning method will be specified in the procedures, and should be approved by the regulatory authority.

Further information on cleaning is given in Appendix A.

16

7 INSPECTION FOR FLAWS

7.1 Magnetic particle inspection

MPI can be regarded as the standard method for locating and characterising the length of a surface-breaking crack, but has to a certain extent been superseded by other methods.

MPI under water is not, in principle, very different from MPI above water. The main differences are the need to apply the magnetic particle suspension through the water, and the turbidity of the water, which may limit visibility significantly. Operation at shallow depths may encounter problems from light levels unless high-visibility inks are used or the inspection is done at night. It is essential to clean the surface thoroughly to bare metal.

The performance of MPI is heavily dependent on the skill and experience of the Inspector, and on the nature of the test surface. Inspectors must be qualified to CSWlP 3.2U or equivalent (see Appendix B), and should ideally have considerable experience in the application or environment under test. Surfaces for inspection should be carefully cleaned back to bare metal, taking care not to peen over the edges of any defects that might be present, as described in Section 6. A careful scrutiny by close visual inspection is essential before and during MPI, as weld undercut, interbead grooves, and other fabrication defects are often mistaken for cracks.

There is no standard specifically written to cover MPI underwater, but BS6072, intended for onshore MPI, is often used in the North Sea. Other Standards may have some application, notably BS4069 for inks, and BS4489 for assessing the ultra-violet light used in MPI. None of these standards is completely satisfactory for offshore use and caution should be exercised in their use.

General information is given in IIS/llW-849-87 'Handbook on the Magnetic Examination of Welds'.

Z 1. 1 General requliemen fs

Magnetic field strengths in the examination area are generally the same as those encountered in above-water testing: peak tangential

17

field strength of 2.4 to 4.0 kA/m is in most cases adequate. The normal component of the field should be kept as small as possible.

Offshore structural steels often have high permeabilities and so lower field strengths can be used, provided the peak tangential magnetic flux density in the steel is at least 0.75T. The relative permeability of offshore structural steels is often quite different from that assumed in the Standards, and will vary from parent plate to weld metal and heat affected zones (HAZs).

The magnetic field may be produced in a number of ways. Alternating or half-wave and full wave rectified currents are recommended, and should be measured by an ammeter, giving rms values for alternating current and rms values referred to half-periods for half-wave rectified current.

The magnetic field strength should be verified using Hall probe instruments or meters of similar accuracy, particularly for complicated geometries such as nodes. Simple flux indicators such as Burmah Castrol strips should be used with caution and basically treated as showing whether a field is present or not: they should not be relied upon for measurement of field strength.

Z 1.2 Magnetisation methods

The most common magnetisation methods are:

8 Coils The coil should be placed 50-1OOmm outside the test area. Only surface-breaking cracks substantially parallel to the coils will be susceptible to detection.

8

8

Parallel conductors Use of parallel conductors will facilitate the detection of cracks parallel to the conductors in the area between them.

AC electromagnetic yoke The AC electromagnetic yoke can be applied to detect surface breaking cracks at 90 degrees to the line between the two poles of the electromagnet.

18

Prods The use of prods is not recommended unless special precautions are taken to avoid damage to the test area (see below). If they are used, prod spacing should be 150-300mm. An rms alternating current of 2.5-4.0 A per mm prod spacing should give the necessary tangential peak field strength.

The prods should be arranged as a yoke to allow one-hand operation, and should be fitted with low-melting-point alloy tips to avoid local heating and burning of the steel surface leading to cracking. Copper or bronze tips should be avoided to remove the chance of creating copper inclusions in the steel surface. Good electrical contact between prods and surface must be made before the current is turned on, to avoid any arcing at the point of contact and local heating.

The use of DC or permanent magnets is possible but not recommended.

Z 1.3 Defecfion system

Fluorescent magnetic particles are normally employed, stored as a suspension in liquid and fed through a hose to the examination area whilst this is being magnetised. After application of the particles, the field should be maintained for a few seconds.

Although fluorescent particles are mainly used offshore, examination near the surface in daylight may require non-fluorescent inks.

Z 1.4 Inks

The concentration of particles should be about 0.5% by volume, with a range of sizes from 0.15 to 25pm, depending on the surface condition of the area - 15pm is a commonly used mean diameter. The liquid suspension should contain a suitable wetting agent such as soap.

The container for the ink should be slightly overpressurized or fitted with a pump to supply the ink to the inspection site, and it is important that the ink should be agitated from time to time to prevent settling-

19

out of the particles. The application system should be periodically cleaned after use to avoid blockages.

7.1.5 Wewing conditions

Ultraviolet light of wavelength between 320 and 400nm is used at an intensity of 1 OW/m* for fluorescent inks. BS6072 indicates that MPI should be performed under very low ambient light levels of less than 10 lux. However, subject to approval by regulator, it may be reasonable to conduct MPI at considerably higher light intensities. A luxmeter should be used to assess the light level. Sufficient time should be allowed for the diver to become acclimatised to the low light levels before inspection is undertaken.

For non-fluorescent inks, the area under examination should be illuminated by natural or artificial light to a level of not less than 500 lux.

A record of the MPI indication may be made by a variety of methods, particularly using still photography or video cameras. Where surface replication is used undue credence should not be placed on the veracity of any replica as this can depend substantially on the conditions.

7.2 Alternating current potential drop

This technique is purely for crack sizing and is not used for crack detection in the first instance. Both low-frequency (50 Hz) and high- frequency ACPD equipment are available and are used for crack depth measurements underwater on plate material, butt welds and node welds. Low frequency equipment is recommended on rough surfaces, as the larger skin depth offsets the surface irregularities.

Various corrections are required for the effects of crack shape, vicinity of edges, angle between plate and weld cap, etc. Bridging across crack surfaces and angled cracks may lead to significant errors. The technique requires clean surfaces for electrical contact.

20

7.3 Eddy current and ACFM techniques

Although eddy current and ACFM systems are distinct entities, since they work on similar principles and possess broadly comparable performance, they will be presented together. Both techniques are capable of detection and have sizing capabilities, though some equipment can perform only the detection function.

Eddy current NDT has been used offshore for many years, but considerable development has taken place over the last decade. Some systems have been specifically developed with the intention of minimising the amount of cleaning needed and of reducing the workload on the diver and making them suitable for use by ROVs. These systems are highly computerised and are capable of performing extensive signal processing and analysis, together with recording data on floppy disk for subsequent analysis. ACFM possesses similar capabilities.

Interpretation of the signals from the probes is the responsibility of the topside inspector/operator, and the diver's training consists essentially of learning how to deploy the probe correctly. Other systems place more emphasis on the capability of the diver-inspector and slave monitors may be employed to echo the results to both diver and topside operator.

In all these systems, despite the assistance from computer processing, interpretation still requires skill and experience. It is essential that the inspector, whether a diver or a topside operator, has sufficient training to be able to recognise not only the signatures of flaws, but also when the equipment is not being operated properly. Both eddy current and ACFM methods are now included in some national training and qualification programmes like the CSWlP scheme under the electromagnetic inspection of welds, CSWlP-DIV-8-96 and Lloyds Register. Attention is drawn to the comments regarding training (Section 4) and the introduction of new equipment (Section 11).

When operated correctly under good conditions, the performance of these methods is almost as good as MPI, but has the advantage of being very much faster and of usually needing less cleaning. Lift-off of the probe can be a problem, and although these modern systems are not unduly sensitive to surface condition, they do perform better

21

on clean surfaces. Inspection through paint coatings is not only feasible but may be desirable by providing smoother surfaces. This produces a simpler operation than for MPI and makes these techniques more desirable to use and also makes it simpler for them to be deployed by ROVs.

7.4 Flooded member detection

FMD possesses only crack detection capability and that when the crack has penetrated the wall thickness of the immersed member.

Two main systems exist, using ultrasonics or gamma-ray sources. The former works by detecting an echo from the far side of the flooded member, as air-filled members do not transmit the ultrasonic pulse. The gamma-ray method relies on the absorption of gamma radiation by water, with air-filled members giving higher radiation levels on the detector side of the tubular. The gamma source is sited on one side of the member, with the detector on the other.

The ultrasonic system is suitable for deployment by diver, but the gamma-ray system must be carried by ROV. Since radioactive sources are used, similar precautions to radiography must be taken - see Section 7.5.

Although the FMD systems give a high POD of flooding, this is offset by the short remaining life of the member caused by the through- thickness crack responsible for the flooding. Internal corrosion can lead to false readings on some ultrasonic systems owing to the scattering of the ultrasonic signal from the pitted corrosion surface. Partially-flooded members can be difficult to detect, although the implementation of the correct procedures will largely offset this problem by taking a number of readings in different clock positions around the member.

The high speed of use of FMD leads to its main role as a screening system for parts of the structure not inspected by the more detailed, but slower, NDT methods.

22

7.5 Radiographic methods

Radiography possesses both detection and sizing capabilities for embedded flaws.

In general, radiography does not find ready application offshore, owing to obvious problems from the radiation sources. X-ray equipment is normally not suitable for use below water, and gamma- ray equipment is the main variant of radiography employed.

If X-ray equipment is to be used below water, such as in habitats, special precautions are required for the high voltage power supply. Procedures should also pay especial attention to the possibility of accidental exposure to personnel from opened radiation sources, including by failure of protective shielded containers. Some countries may well class such an incident as a notifiable nuclear accident and require it to be reported to the appropriate authorities. There may also be severe restrictions on the transport and storage of gamma radioisotopes.

Gamma radiography is predominantly used for repair weld examination in habitats, but can also be used in water, including for detecting and sizing of pitting corrosion in pipelines.

Precautions must be taken to overcome radiation absorption and scatter in water by exclusion of water from the space between source and object. This can be achieved with gas-filled rubber balloons, light metal cones, etc.

Reference to the following IIW documents may be useful: llS/llW-492- 75,42373,363-71,33569,269-67, 185-65, 183-65, 62-60 as well as to: 'Basic Rules for Radiographic Examination of Metallic Materials by X- and Gamma-Rays', I S 0 5579, and to other relevant I S 0 standards.

7.6 Ultrasonic inspection methods

Ultrasonic testing is used to detect and characterise flaws, and to measure wall thickness, especially for corroded surfaces or to measure ligament thicknesses under defects or following remedial grinding. Wall thickness corrosion measurements are also covered in

23

Section 8.2. A specialised application is in some types of Flooded Member Detection (Section 7.3). Both buried and surface-breaking flaws can be found, though the latter requires favourable orientations and suitable arrangements of probe and detector transducers.

Ultrasonics is generally well suited to underwater use as the water provides good coupling, although since a smooth and clean metal surface is required, cleaning will probably be necessary. Both automated and manual methods may be used, though the large- scale use of manual ultrasonics is operationally difficult. Examination of nodal welds by ultrasonics is often impracticable owing to the restrictions of geometry and the lack of access.

A variety of systems is available, ranging from conventional analogue pulse-echo systems to digital thickness measurement systems and highly-automated and computerised ultrasonic techniques. This section describes commonly used underwater ultrasonic inspection methods and application details, and should not be taken as being an exhaustive description of ultrasonic techniques.

General background may be found in references such as IIS/IIW document: 'Handbook on the Ultrasonic Examination of Welds', IIS/II W-527-76.

Z 6.1 Types of ultrasonic systems

a. Ultrasonic point thickness measurement

Digital thickness meters are used to take wall thickness measurements from specific locations. The purpose way be to confirm the nominal wall thickness of a component, to obtain an indication of whether generalised corrosion is taking place, or for specific tasks such as to measure the ligament of sound material following excavation of a surface- brea king crack by grinding.

Point thickness measurements should not be used to obtain minimum wall thickness measurements for residual life or corrosion rate assessment purposes, or to define localised corrosion in pipelines as revealed by intelligent pigging.

24

b. Fixed point corrosion rate monitoring

Ultrasonic transducers permanently attached to components can be used to monitor corrosion loss rates at specific locations. The transducers may be attached to studs welded into position prior to installation or may be bonded with adhesive prior to, or during, service.

C. Manual ultrasonic pulse-echo weld testing

Manual pulse-echo systems use analogue or digital ultrasonic flaw detectors adapted for subsea use with a slave system located at the dive platform for supervisory and secondary interpretation purposes. Operators must be trained and certified in accordance with recognised standards. This is intended to replicate a conventional manual topsides inspection, but is subject to additional problems of dexterity, distractions, restricted view, working environment and communications. For these reasons alternative forms of inspection should be assessed to identify which is most suited to providing the required information.

d. Semi-auto mated uI trasonic pulse-ec ho examination

In this technique, digitised ultrasonic signals are conveyed to the dive platform by umbilical where they are processed and displayed using an ultrasonic imaging system. Storage of each ultrasonic waveform (A-scan) enables detailed off-line analysis of range and amplitude to be carried out. Through-wall sizing of known cracks underwater has given accuracy of sizing within 1.Omm with this technique. The addition of a positional encoder to the ultrasonic probe or scanning device enables ultrasonic responses to be accurately located with respect to the direction of scan.

e. Automated ultrasonic pulse-echo weld examination

A fully automated mechanised scanner is used to manipulate the probe(s) in the desired scan pattern. Deployment is by diver or ROV, but the inspection operation is controlled entirely from the dive platform. This approach is usually restricted to welds in components which have simple geometry due to the complexity of multi-axis robotics. It provides controlled coverage and enables fully code-

25

compliant inspections to be carried out. Storage of ultrasonic waveforms enables advanced sizing techniques to be performed off- line. Pulse-echo ultrasonic inspection usually requires the use of several beam angles from both sides of a weld. This can lead to complex and time-consuming procedures for scanning equipment and calibration.

f. Automated ultrasonic pulse-echo corrosion mapping

A fully-automated dual-axis robotic scanner is used for corrosion mapping, normally using a single 0" compression probe scanned in a raster pattern over the area of interest. The resolution of inspection is typically a 4mm x 4mm raster, though higher resolutions may be used if better definition is needed.

Both contact and immersion techniques are used and can be carried out with the ultrasonic probe mounted in a carriage, which is referenced off the material surface. It must be noted, however, that wherever the carriage is lifted off the surface through riding over surface roughness such as welds, ultrasonic information will be lost. The use of an immersion probe mounted off the scanner overcomes this effect and enables scanning to be carried out up to weld toes. Immersion probes are normally focused at a point several millimetres (typically 8mm but can vary depending on material thickness) below the scanned surface.

The effects on the ultrasonic inspection of irregular scanned surfaces caused by corrosion or damage can be minimised using an immersion approach with the ultrasonic beam focused at the near surface. Immersion techniques give rise to a dead-zone immediately below the scanned surface owing to reflection at the water/steel interface, and this may necessitate the use of contact probes.

The technique is equally applicable to corrosion mapping and the assessment of hydrogen-induced cracking (HIC). In both cases, special attention needs to be given to the sensitivity of inspection.

g. Semi-automated time-of-flight-dlffraction (TOFD)

The TOFD technique is described in detail in BS7706:1993. TOFD requires the use of an ultrasonic imaging system capable of grey

26

scale imaging. The ultrasonic waveforms are digitised, stored and displayed as grey-scale D-scans in real time. The scan axis may or may not be encoded. TOFD is a fast inspection technique which uses simplistic scanning equipment and the results obtained underwater can be of equal quality to results from onshore. Reproducibility of flaw height measurement underwater has been demonstrated to be in the order of l.Omm. The speed of inspection reduces diver risk and can be advantageous where access is restricted due to the state of currents. Commonly used acceptance criteria can be adapted to apply to TOFD.

h. Ultrasonic creeping wave

Ultrasonic creeping wave is a specialised form of ultrasonics relying on the generation of a surface wave which is restricted to, and which propagates along, the layer immediately under the steel/water interface. The intensity of the wave decreases with travel as energy is lost into the interior owing to mode conversion.

Conventional ultrasonics equipment can be used to drive a special probe, but otherwise the interpretation is similar to normal ultrasonics. The technique is very suitable for detecting surface-breaking flaws, as a surface wave echo is produced which is reflected back to the probe. It can be used with coated steel and is particularly suitable for geometries of awkward access, as it possesses a 'stand-off' capability, the wave reaching ahead of the probe.

As with other ultrasonic techniques, good cleaning of the surface for the probe is necessary.

Z 6.2 humples of suitable appll'ca tions for ultrasonic examhation

a. Depth sizing of known cracks in structures

For sizing of cracks found by surface crack-detection methods, either TOFD or semi-automated pulse-echo using 'backscatter tip diffraction' may be used.

b. Volumetric examination of welds in structures

Flaws revealed by in-service condition assessment surveys or by code-

27

compliant inspection of welds repaired underwater may be inspected by TOFD, automated pulse-echo, or, if in a habitat, by manual pulse-echo. Other suitable applications for these techniques are the monitoring of repaired joints, or for monitoring the stability of known flaws.

C. Pipeline weld assessment and monitoring

Pipelines may be examined for volumetric flaws, such as following criticality assessment, by TOFD, preferably encoded, or by automated pulse-ec ho.

Welds with suspect root/HAZ erosion or cracking can be inspected by TOFD.

d. Pipeline parent material assessment and monitoring

Corrosion at strategic locations identified through criticality assessment may be inspected using automated corrosion mapping, and for corrosion rate assessment at specified locations, fixed point monitoring may apply.

e. Newer rapid scanning ultrasonic inspection techniques

Developments are ongoing to utilise different types of ultrasonic techniques e.g. the resonance technique or use of plate waves and head waves, in inspection systems for rapid scanning of large areas for the detection of corrosion and other flaws. All of these systems utilise modern computing technology.

7.7 Visual inspection

%% I General visual inspection

GVI may be done by a diver but is frequently carried out by ROV. It may be used to give an overall impression of the general state of the structure, including areas of gross damage, coating damage, the extent and type of marine growth, debris build-up and sacrificial anode consumption. It is not suitable for the detection of cracking other than very sizeable cracks. Performance is affected substantially by the turbidity of the water.

28

ZZ2 Close vkual inspection

CVI may be done by diver or by means of an ROV fitted with high- resolution imaging systems. A principal purpose of CVI is to improve the definition of features found during GVI, but on its own, it is not an acceptable method for inspecting for cracks. It is very useful as a supplement to other NDE techniques, allowing the differentiation between weld undercut and a crack, for example. Divers should be encouraged to practise CVI as part of their other inspection activities.

During CVI, a replica technique may be used to obtain a permanent record of a defect suitable for examination elsewhere. Stereo- photography and video recording techniques may also be suitable.

29

INSPECTION FOR CORROSION

8.1 Cathodic protection potential measurements

Electrochemical potential measurements are made to monitor the functioning of cathodic protection systems on underwater parts of steel structures with the main objective to verify adequate functioning of the system. The technique is based on measuring the potential difference between the structure and a reference electrode. Visual examination of the anode consumption will give an indication of the remaining life of the cathodic protection system.

Further information on principles and guidance may be found amongst the documents listed in the Bibliography in particular ' Cathodic Protection Design', RP B 401, Det Norske Veritas, March 1993, and 'Design and Operational Guidance on Cathodic Protection of Offshore Structures, Subsea Installations and Pipelines', MTD Ltd, Publications 90/102.

8. I . I Equipmen f

The equipment consists of a reference electrode connected via a voltmeter to the structure to be examined. The reference electrode may be positioned by a diver or an ROV. Electrical contact to the structure may be obtained either topside via a cable or subsea using a spike tip. Special subsea probes have been developed.

A number of reference electrodes are available, the commonest of which is probably the silver/silver chloride cell (Ag/AgCl/seawater), although zinc/sea water, the saturated calomel electrode (SCE), and copper/copper sulphate are also used. Saturated calomel electrodes are normally used for calibration purposes. Each of these cells will have a different voltage and when quoting potential readings, care should be taken to state which electrode has been used. In practice, the Ag/AgCl/seawater scale is very close to the SCE scale, though the other electrodes will require conversion.

The stability of the reference electrode should be better than flmV per 24 hours. Reference electrodes should be calibrated daily during current use and at least once a year opened up to check their condition and potential. The potential of an Ag/AgCI reference

30

electrode depends on the chloride content and the temperature of the water. Calibration should preferably be carried out before and after each dive. In open water, however, the chloride content is normally reasonably constant and little drift should occur from this source. To ensure a long lifetime, reference electrodes should be kept in a chloride solution and Ag/AgCI electrodes should not be exposed to light.

The voltmeter should be of high impedance ( 1 OMS2) with a measuring range between -2000mV and +2000mV. The insulation resistance of the cables needs to be high and any breakdown of insulation may lead to erroneous readings. The overall measuring accuracy should be +1 OmV.

It is important to ensure good electrical contact with the structure, and where this contact is achieved using a metallic tip, this should be made of a seawater-resistant alloy such as titanium or AlSl type 316 stainless steel. The contact area of the metal tip should be less than 2cm2. Some cleaning may be necessary to ensure good electrical contact to underwater structures.

Electrical contact with the structure may be made either above or below the water surface, with the voltmeter sited either topsides or subsea. The reference electrode should be placed as close as possible to the surface of the structure under test at the point to be checked. It is recommended that the reference cell should be within 50mm of the surface for structural CP readings and lOmm for use on sacrificial anodes.

8.1.2 Cathodic protection requkemen ts

Measured cathodic protection potentials depend on a number of parameters such as oxygen content of the water, salinity, temperature, and current. In addition, the location of the measurements will be very significant, the potentials being affected, inter alia, by the geometry of the structure, the nature of the surface (coated or covered in marine growth), and the distance from the anodes. Anode material and geometry will affect the required coverage of the structure.

31

Potential readings for steel structures given in various standards or documents of recommended practice and taken against an Ag/AgCl/seawater reference electrode in aerated seawater, are:

For normal steel structures, more negative than -800mv and for stainless steels, either austenitic or austenitic/ferritic, -550mv.

In all cases, levels more negative than -900mv may cause the generation of free hydrogen in sufficient amounts to cause damage to coatings, hydrogen induced stress cracking or corrosion fatigue, depending on the levels of stress and the material characteristics. (See BS 7361 pt 1,1991,draft pr EN 12495 for offshore structures and NACE Standard RPO 176-94.)

8.2 Wall thickness measurement

Measurement of wall thickness in corroded areas is generally done by ultrasonic examinations, either with digital wall thickness meters or using conventional ultrasonic systems. Systems may be fixed permanently to the area of concern or may be deployed as needed. For further information reference is made to Section 7.6.

Care should be taken to ensure that erroneous measurements do not result from the presence of debris under the probe, the water gap between probe and steel, or any coatings. It is necessary to clean the area well.

A number of readings should be taken and averaged for each spot assessed. Conventional systems may achieve better results using 45" probes than normal (90") probes.

32

PASSIVE NDT METHODS

In addition to the active NDT methods discussed above, there are various passive methods, or monitoring systems. Only stress (strain) measurement systems have found long term application as yet, but it is likely that these passive methods will attain substantial use in the future, particularly for unmanned installations.

Probably the most successful to date in terms of deployment has been leak detection accomplished by devices located in normally dry compartments. Semi-submersible installations and other floating structures have made fairly widespread use of these devices. Other systems include, vibration monitoring to detect changed vibration patterns of installations as a result of cracking, and acoustic emission monitoring to detect the sounds made by fatigue cracking. Remote- sensing systems process the information from sensors and send the results back via hydroacoustic coupling and a telemetry link.

33

10 POST-INSPECTION ACTIVITIES

10.1 Recording and retention of data

Records should be made and kept in accordance with established principles of good documentation and QA, including IS0 9000, where applicable.

Inspection results may be recorded in a variety of formats. Many of the more modern systems allow the signals to be stored on compact disk for record purposes and for re-analysis. Visual inspection information can be recorded on conventional underwater still photography, with close-up pictures of areas of interest. Photogrammetry is becoming increasingly used to give three- dimensional information from which detailed information can be reconstructed. High-definition video recording is also a routine procedure.

Visual images should always give the location and an indication of scale, and ideally should carry the time and date and any other information deemed necessary for QA. Other forms of recording should carry similar data to facilitate QA.

Attention should also be given to ensuring the integrity of data during transport and storage - for example magnetic media should obviously be kept well away from strong magnetic fields.

10.2 Inspection after repair

The repair of flaws lies outside the scope of this document. However, such repairs will obviously be a prime target for follow-up inspection, and, indeed, the design of the repair will probably take into account the capabilities of the NDT methods to be used.

In principle, any suitable NDT technique may be used, though it is recommended that the initial inspections should be very thorough and should employ equipment capable of good sensitivity to any flaws. The materials used in effecting the repair may need to be taken into consideration in selecting the NDT technique.

34

inspectors should be advised to look out for new flaws originating from the repair itself, not merely the re-emergence of the old flaw.

35

I 1 INTRODUCTION OF NEW EQUIPMENT

The rapid development of techniques and equipment for underwater inspection in recent years frequently means that the performance of a new system is not completely established. This lack of information should not necessarily preclude the equipment from being considered for use offshore, though it is obvious that some form of trials will be required to give an indication of performance and operational limitations and to satisfy regulatory authorities. This section will thus address some of the issues involved in the successful introduction of new equipment.

New equipment should undergo an adequate programme of formal testing and evaluation before being used offshore. Whether this is done by the manufacturer, the inspection contractor or by the owner's personnel is largely immaterial provided that the trials are conducted to a high standard. Important factors requiring to be addressed in trials are given in Appendix A.

Following satisfactory evidence of performance in trials, new equipment may then be used offshore on non-critical applications. However, for more significant applications the new system should be used only in conjunction with existing established NDT techniques until it has demonstrated its reliability and performance offshore to the satisfaction of the responsible regulatory authority, whether statutory authority, government agent, or in-house approval. Obviously, should a system have a satisfactory record of use with other Owner/Operators, then many of these requirements might be waived.

Under no circumstances should a new NDT technique be used in isolation to replace an existing method in a structural Survey without prior approval from the regulatory authority.

As official training schemes may not cater for new or specialised methods, the responsibilities for the correct and appropriate use of novel equipment are as follows:

0 the Owner/Operator desiring to use the method/equipment must be satisfied that it is suitable for the work, and that the Inspection Contractor can use the equipment properly;

36

the Inspection Contractor must be familiar with the correct use and deployment of the system to ensure that it is operating as intended in a role for which it is suitable, and must further ensure that the inspection personnel are properly trained and competent to use the equipment in that application and under the conditions applying for the inspection;

the equipment operator/controller/diver must be competent to use the equipment under the appropriate inspection conditions.

37

12 REMOTE OPERATED VEHICLE DEPLOYMENT

12.1 Introduction

Until now the majority of offshore operators have used conventional diving techniques for the inspection of the sub-sea sections of the platforms. It is expected that in future, more work will have to be carried out within the inspection period allocated for each installation, depending on the operating company’s philosophy. As it is not possible to extend the divers’ bottom time, other approaches will have to be explored. One possibility is the greater use of remote operated vehicles (ROV). The cost of using an ROV with a ‘ship of opportunity’ is substantially less than the use of a diving support vessel and the full diving crew, even in air diving conditions, since such ships are more available and less expensive. If ROVs are used to carry out some of the more arduous and repetitive tasks, it will leave more time for divers to carry out the more intricate inspection tasks.

Remote operated vehicles (ROVs) have been in use since the early 1970s. Early trials highlighted the fact that the ROVs available at that time could only carry out a limited number of mechanical tasks and could not cany out any inspection tasks.

A new range of vehicles was developed in the mid-1980s which were able to cany out an increasing number of tasks such as cleaning, visual inspection and flooded member detection. The economic advantages of deploying ROVs from less expensive, non-specific support vessels or directly from offshore platforms was also recognized and these combinations have helped to increase their use during inspection programmes. In some cases the deployment of divers has been reduced to only 10% of the inspection programme applied specifically to weld inspection. In addition new inspection techniques are being developed which may not require accurate weld tracking by the manipulator.

Appendix A reviews the developments in the use of ROVs for inspection, repair and maintenance tasks during the last two decades. The present status of ROV-based inspection is examined and possible future developments are indicated.

12.2 Present status

At the present time it would appear that ROVs are capable of undertaking the following tasks:

38

General visual inspection

Stills photography of marine growth using 35 or 70 mm formats in mono or stereo

Close visual inspection after specific area cleaning utilising fixed stand-off cameras, stereo photography and photogrammetry

Cathodic protection surveys using proximity and/or contact techniques

Cathodic protection surveys up to full current density utilising proprietary CP/CD equipment

Cleaning in specific areas of interest such as butt welds, nodal welds or other anomalous areas using rotary brush cleaning or air entrained grit systems

Cleaning and bulk marine growth removal programmes utilising high pressure water jetting specially developed cleaning carts as required

Flooded member inspection of horizontal and vertical diagonal members using ultrasonic or gamma absorption techniques

Wall thickness measurements of specific areas utilising ultrasonic techniques

Distance measurement in real time utilising videogrammetry techniques involving laser or acoustic technology

Microbiological sampling of marine growth or mud mounds

Mud removal programmes or dredging from specific areas of interest or concern

Bolt tightening or torque testing of subsea fixings utilising ROV powered torque wrenches

Weld inspection and defect sizing.

It can be seen that the range of ROV applications is greatly increased from those of the trials carried out in Loch Linnhe in 1981-1982. This list gives an insight into the range of tasks being undertaken.

39

12.3 Conclusions

The ability of the present ROVs to carry out the majority of subsea tasks has improved during the last ten years due to the development of the vehicles and the increasing sophistication of their multi-function manipulators. They are able to perform visual and close visual inspection, cleaning and maintenance of subsea installations and the majority of mechanical tasks such as lifting, cutting and operating valves.

In terms of inspection tasks, greatest success has been achieved with the radiation-based method of flooded member detection from ROVs, which is now being performed successfully on a regular basis.

The use of ROVs operating from a platform to carry out all of the inspection tasks with the exception of weld crack detection has been shown to be viable and economic, and the combined use of divers and ROVs for crack detection has also been shown to be feasible.

The problem of weld inspection by ROV has been investigated and there are now two possible solutions: magnetic particle inspection using the integrated electromagnet, and alternating current field measurement using the multi-element probe array. These two techniques have now been fully diver-tested and have also been used with ROVs for weld inspection. The ACFM array probes have been used with a number of work class and mid size ROVs.

40

BIBLIOGRAPHY - Selected reference documents on NDT of offshore constructions

INTRODUCTION

This Bibliography contains a reference selection of standards, guidelines, codes, and other documents on NDT of offshore constructions. Because the subject is so vast this selection may not be complete. The documents have been grouped according to the type of issuing body:

1 . Government bodies 2. Verification authorities and classification societies 3. Advisory bodies and others 4. Associated documents

It should be noted that the content of the referred documents may not always agree with the recommendations made in this text.

1 Government bodies

Health and Safety Executive (HSE) UK (was Department of Energy until 1991).

Health and Safety at Work etc. Act 1974, with amendments* along with the Health and Safety at Work etc. Act 1974 (Application outside Great Britain) Order 1995.

Offshore Safety Act 1992 with amendments*.

Offshore Installations (Safety Case) Regulations 1992 (SI 1992/2885), with amendments*.

Offshore Installations and Pipeline Works (Management and Administration) Regulations 1995, (SI 1 995/738) with associated Guidance.

The Pipeline Safety Regulations 1996 (SI 1996/825) with associated Guidance.

Offshore Installations and Wells (Design and Construction etc.) Regulations 1996 (SI 1996/913) with associated Guidance (3 vols.)

68

The offshore installations safety, health and welfare regulations 1 976. Statutory Instrument, 1976 No. 101 9*.

Petroleum Act. 1987

Oil and Gas (Enterprise) Act. Chapter 23.1 982.

Offshore Installations Guidance on design, construction and certification, consolidated 4th edition, 1993 with February 1995 amendment, HSE (may be withdrawn or replaced in the future, i.e. after June 1998).

Manual of Legislative Acts, rules and guidance notes concerning North Sea offshore developments - UK Sector. Compiled by Weston Law Manual Services, Revision 42, May 1996.

A handbook for underwater inspection, OTI 88-539, Her Majesty's Stationery Office, UK, 1992.

Offshore installations: Guidance on design and construction, 5th edition, 1996.( May be withdrawn or replaced in the future.)

'Diving at Work Regulations 1997, SI 1997 No. 2776' ISBN 0.1 1.0651 70.7 HMSO London.

'Commercial diving projects offshore' ,L103; Five Approved Codes of Practice , HSE Books, London, 1997; ISBN 0. 71 76.1 494.8

Note: * amendments have been made to ensure consistency during the transition period from the previous to the new legislation. It is essential to view the legislation listed as a whole in order to ensure that all of the changes required can be accommodated.

Marine Technology Directorate Ltd. UK

Underwater inspection of steel offshore installations: implementation of a new approach. MTD Ltd. Publication, 89/104.1989.

Design and operational guidance on cathodic protection of offshore structures, subsea installation and pipelines. MTD Ltd. Publication 90/102.1 990.

69

Norweaian Petroleum Directorate

Acts, regulations and provisions for the petroleum activities. Volumes 1 and 2. Yearly updated compilation of Acts, regulations and guidelines.

US Coast Guard

US Geoloaical Survey

2 Certifying authorities and classification societies

CERTIFYING STANDARDS ORGANISATIONS IN THE UK

As the result of new legislation introduced in 1996, Certification is in transition to a Verification Scheme so that Certifying Authorities for offshore installations will cease to operate from June 1998.

American Bureau of Shiming

Rules for building and classing offshore installations. 1 983.

Rules for building and classing mobile offshore drilling units. 1995. Notice: 1996.

Guidelines for building and classing undersea pipeline systems and risers. 1 99 1 .

Rules for building and classing underwater vehicles, systems and hyperbaric facilities. 1990. Notice 0.1 :1995

Underwater inspection in lieu of dry docking survey. 1996.

Nondestructive inspection of hull welds. 1986.

The certification of offshore mooring chain. 1 986.

Hull condition monitoring systems. 1995.

Certification of firms engaged in thickness measurement of hull structures. 1996.

70

Bureau Veritas

Rules for the construction and classification of offshore platforms. 1975. Amendment No.1 :1982.

Rules and regulations for the classification of mobile offshore drilling units. 1987.

Rules and regulations for the classification of submersibles. 1989.

Det Norske Veritas

Rules for classification of mobile offshore units. Parts 1-7.1 996.

Rules for classification of fixed offshore installations. Parts 1-5.1 996.

Rules for submarine pipeline systems. December 1 996.

Rules for certification of diving systems. 1988.

Rules for classification/certification of submersibles. 1988.

Rules for flexible risers and pipe. October 1994.

Certification of offshore mooring steel wire ropes. Approval Scheme 2.5. May 1995.

Certification of offshore mooring chain. Approval Scheme 2.6. August 1995.

DNV Recommended reporting principles for ultrasonic thickness measurements of hull structures. Guidelines. September 1 993.

Qualification of underwater inspection personnel. RP A402. August 1983.

Cathodic protection design. RP 8401. March 1993.

Underwater welding. RP B604. January 1987.

Underwater non-destructive examination, equipment and procedures. RP 8704. August 1983.

Qualification of ROV for underwater inspection. RP B706. August 1983

71

Germ an isc her Lloyd

Offshore Technology. Part 1 : Underwater technology ( 1 991 -92). Part 2: Offshore installations ( 1 990). Part 3: Mooring and loading installations ( 1 993). Part 4: Subsea pipelines and risers ( 1 995). 1990-95.

Regulations for the inspection of anchor chain cables at dealers' premises. June 1993.

Llovds Register of Shitwing

Rules and regulations for the classification of fixed offshore installations. Parts 1 -8:1989. Notices 1 -4:1995.

Rules and regulations for the classification of mobile offshore units. Parts 1-8: September 1996. Notices 1 - 1 1 :1996. Rules and regulations for the classification of submersibles and underwater systems. 1 989. Notices 1-2: 1 995.

American Petroleum Institute

Recommended practice for planning, designing, and constructing tension leg platforms. API RP 2T. April 1987. Supplement 1992.

Recommended practice for planning, designing, and constructing fixed offshore platforms. API RP 2A.

Recommended practice for in-service inspection of mooring hardware for floating drilling units. API RP 21.

Recommended practice for ultrasonic examination of offshore structural fabrication and guidelines for qualification of ultrasonic technicians. API RP 2X. Second edition. September 1988.

Welding of pipelines and related facilities. ANSI/API Standard 1 104. Eighteenth edition. 1994.

American Weldina Society

Specification for underwater welding. ANSI / AWS D3.6-93. 1993.

72

Canadian General Standards Board Certification Scheme

Standard for certification of nondestructive testing personnel (Level & II underwater inspectors for visual inspection and industrial ultrasonic inspection, magnetic particle inspection and corrosion potential test methods) CPWETDR2 on 8. GPXX 850808. (Draft document)

International Institute of Welding

Information on practices for underwater non-destructive testing. Document V-1097-97( IIS/IIW-1372-97).

National Association of Corrosion Enaineers. USA

NACE RP 0387.1 990 Metallurgical and inspection requirements for cast sacrificial anodes for offshore applications.

NACE RP 0492.1 992 Metallurgical and inspection requirements for offshore pipeline bracelet anodes.

NACE Standard RP 01 76-94; Standard Recommended Practice, Corrosion Control of Steel Fixed Offshore Platforms Associated with Petroleum Production.

NACE Standard RP 0675, Control of External Corrosion on Offshore Steel Pipelines.

Nordsok Standards. Norway

Structural Steel Fabrication, M-CR-1 01, Rev. 2, January 1996

Cathodic Protection, M-CR-503, Rev. 1, December 1994

Corrosion Monitoring Design, M-CR-505, Rev. 1, December 1994

Welding and Inspection of Piping, M-CH-601, Rev. 1, December 1994

3 Advisory bodies and others

Certification Scheme for Weldment InsDection Personnel, UK

Document No. CSWIP-DIV-7-95 - Part 1: Requirements for the Certification of Underwater Diver Inspectors.

73

Part 2: Requirements for the Certification of ROV Inspectors and Underwater Inspection Controllers, 1 st edition, January 1996.

Dansk Inzreniorforeninq. Denmark

Code of practice for pile supported offshore steel structures. Parts 1 and 2, Translation edition, September 1984.

ShiD Structure Committee. USA

Underwater nondestructive testing of ship hull welds, SSC-293. 1979.

Underwater Enqineerinq Group. UK

Handbook of underwater tools, Report UR 18, Second edition. January 1983.

Underwater inspection of offshore installations guidance for designers, Report UR 10.

British Standards Institute

BS 7361 part 1 1991, Cathodic Protection.

pr-EN 12495, Cathodic Protection for Fixed Offshore Structures, 1996, (provisional).

International Marine Contractors Association, London, UK.

Basic level of competence to be met by ROV personnel, IMCA ROO2 Code of practice for the safe use of electricity underwater, AODC 035

Code of practice for the use of high pressure water jetting equipment by divers, AODC 049

Guidance on diving operations in the vicinity of pipelines , IMCA DO06 Rev. 1.

Guidance note on the safe and efficient use of remotely operated vehicles , AODC 006 Rev.l.

Remotely operated vehicle intervention during diving operations, AODC 006 Rev.1.

74

An international code of practice and guidance for diving. To be published April 1998 IMCA London.

4 Associated documents

Additional useful information on underwater inspection can be found in:

American Society of Metals Handbook.

Dover, W. D., and Rudlin, J.R., 1992, ' Results of Probability of Detection Trials (Uncoated Welds) .' IOCE 92 Conference Proceedings, Aberdeen, UK, October.

Dover, W. D. and Rudlin, J.R., 1992 'Crack Sizing Trials.' IOCE 92 Conference Proceedings, Aberdeen, UK, October.

Raine, G. A., Dover, W. D. and Rudlin, J.R.,1992, 'Trials on Coated Nodes.' IOCE 92, Conference Proceedings, Aberdeen, UK, October.

Guidelines for replacing NDE techniques with one another. NT TECHN REPORT 300. Nordtest , Espao, Finland, October 1995.

Davey, V. S., Hydrogen Assisted Cracking of High Strength Steels in the Legs of Jack-up Rigs, Chapter 13 of 'Recent Developments in Jack-up Platforms', Ed. Boswell L F and D'Mello C, Blackwell Scientific Publications, London 1992. ISBN 0 632 03281 2.

Project ICON Final Report, lntercalibration of Offshore Non Destructive Testing; EC, DO XVII, issued by IFREMER on behalf of Programme Thermie, December 1 994; Confidential, available to members only.

Smith, R. L., Metallurgical Conditions Affecting the Reliability of Magnetic Particle Inspection Offshore; Offshore Technology Report OTH 87 275, HMSO London 1987, ISBN 0 114 12888X.

Porter, L. K., Handbook for Underwater Inspectors, Offshore Technology Report OTI 88 539, HSE London 1988, ISBN 071 7 608 484.

Carne, M. M. P., Slater, O., The Integrity of Pipeline Girth Welds, Offshore Technology Report OTH 86 233, HSE London1 986.

75

Underwater Vehicle Trials Loch Linnhe 1981 /1982. Department of Energy OTP 14.

AIM - OETB special programme on Automated Underwater Inspection and Maintenance - Phase 1, final report October 1989.

ROV Development Projects, Havron Energy Systems, May 1990.

Managed programme of research into the Automation of Subsea Tasks -Phase Ill 1990 - 92, Heriot-Watt University.

lntercalibration of Offshore NDT, W D Dover, D A Topp TSC DAT 0221 - February 1990.

Supervisory Control System for Subsea Telemanipulators - An effective way to perform complex IMR teleoperations underwater - Tecnomare 1990, OTC 6356,22nd Annual OTC, May 1990. Part of EUREKA EU191 - An inspection tool kit for an advanced ROV.

Programme to develop the Astable ROV design SID (Structural Inspection Device) revised proposal - Winchester Associates 1 990, Underwater Technology, Volume 15,4 November 1990.

IRM 90 incorporating ROV 90, November 1990.

A new generation of ROV tooling for subsea development tasks, J Bryden, A Walton, Rockwater Offshore Contractors, ROV 90, November 1990 Proceedings.

Advanced ROV for underwater inspection and maintenance, J Mann, P Sieniewicz, Rockwater Offshore Contractors, ROV 90, March 1990 Proceedings.

The development of ROV NDT tooling, N Duncan, M Bowring, Comex UK Limited, ROV, November 1990 Proceedings.

8th Annual Marine Technical Society, ROV 90 Conference, Vancouver, June 1990. New Developments in Subsea Inspection. G.A. Raine and B.A. Jones 58th Autumn Meeting of the IGE 1992.

ROV inspection of welds -a reality. G.A.Raine . M.C.Lugg. U195

ROV deployed inspection using ACFM arrays . M. C. Lugg, D.Cooke , D. Topp IOCE 94, Aberdeen, 1994.

76

Use of robotics for inspection of offshore structures-below water. (From R and D program to offshore operation). F Ricci, T Ingebretsen, U195.

Advances in Manipulator Technology-ATES-An application of the above technology-Espen Moller-Sonsub International-International ROV Forum ‘96 February 1996.

Inspection of subsea nodal welds by the ARM robot manipulator. D.R.Broome, T.J.Larkum, M.S.Hall. Subtech.

DIRECT-Diverless Inspection and Repair using an Electro-Chemical Technique. A JIP proposed by UCL, Edinburgh University and Technical Software Consultants Ltd.

EDICS Evaluation of Diverless IRM of Subsea Completions-Project proposal supported by the EEC. 1994

77

APPENDIX A - Further information

Cleaning - 6.3 There is a strong interaction between the method of cleaning chosen and the nature of the inspection and the characteristics of the NDT equipment, and so it is possible to give only general guidance on the selection and use of cleaning methods.

Whichever NDT technique is used, it is always necessary to remove excess marine growth, both of the soft and hard-shelled types. However, some NDT methods need a bare metal finish over at least part of the surface to be inspected, including: close visual inspection, magnetic particle examination, and ultrasonic inspection. In order to reduce the cost and difficulty of attaining a bare metal surface, two lines of approach have appeared. On one hand, there has been much development of non-contacting NDT methods which are more tolerant of surface finish, including paint and other protective coatings. The various eddy current and related techniques are usually quite capable of operating satisfactorily through a surface layer, and the performance may in fact be improved when operating over a smooth surface such as paint.

The other line of approach has been to improve the efficacy of the cleaning method, and the introduction of new and quick blast- cleaning methods has largely overcome the difficulties of cleaning down to bare metal and has tended to offset the advantages of the more tolerant inspection methods. Attainment of a bare metal finish has several advantages, such as rendering it possible to deploy several examination techniques on the same area, including close visual inspection, which can give a 'second opinion' of the nature of any indication found by the primary detection method. Sometimes there can be an improvement in the performance of even non- contacting methods owing to the reduction in the rocking of probes caused by uneven surfaces.

Where a bare metal surface is required, it may be specified in terms of IS0 8501 -1 : 1988 (see section 6.3) or alternatively Swedish Standard SIS 055900, with SA 2.5 or St 3 being commonly applied. Another possible code is CP3012:1972.

41

The nature and extent of the marine growth and of any protective coating will have a large bearing on the choice of cleaning method. Hand tools can remove soft-tissue marine growth, but tenacious or hard-shelled growth will require power tools or blast cleaning. Where a protective coating has been applied to the steel, it is generally desirable to leave this as intact as possible and to inspect through the coating, since repair of coatings underwater is normally difficult and unsatisfactory. Where an indication occurs under a coating, the coating should normally be removed and confirmation of the indication obtained. Cleaning by hand to a bare metal finish is likely to be ineffective and to produce inconsistent results, with the possible exception of methods such as wire ropes and chains, although these can cause damage to the steel surface. Power tools, whether pneumatic or hydraulic, often require great care in their use, both with regard to safety of personnel and because of their tendency to damage the structure. Needle gunning is generally considered to cause excessive damage, and power wire brushing and some blast cleaning methods may also be undesirable in this regard.

INSPECTION FOR FLAWS - 7

Consideration must be given to the design of equipment for use underwater: the difficulties and dangers of sub-sea working require that special care be taken to make equipment easy to use and safe.

MAGNETIC PARTICLE INSPECTION - 7.1

Under some circumstances, magnetic particle examination is capable of detecting defects lying very near to the surface, but no undue reliance should be placed on this ability.

It should be noted that the relative permeabilities of welds can be markedly lower than those assumed in BS6072, thereby possibly leading to over-magnetisation and increasing the risk of false indications at weld contours. Under-magnetisation under the conditions recommended in the Standard should not be a problem. Under some circumstances, it may be difficult to comply with BS6072 for high-strength steels, owing to the different magnetic properties of these steels.

42

There are currently several means of assessing the flux density being generated, none of them completely satisfactory. The so-called 'disc indicators' or test strips provide only a crude guide and should be employed merely as a simple confirmation that a magnetic field is in fact being generated during inspection. Nearly all flux meters available commercially measure the applied field strength and thence calculate the flux density from assumed values of relative permeability. They are thus susceptible to the same errors as BS6072.

The attainment of the magnetic field may be by several methods. The preferred route in many instances is by coil magnetisation, in which several turns of a current-carrying coil are looped around the tubular member being inspected. In order to get a uniform field, it is important to ensure the correct spacing of the coil from the area being inspected, and that the separate loops are parallel and contiguous round the circumference of the member.

In some instances, this procedure will not be convenient or possible, in which case a hand-held magnetic yoke may be used. Attention should be paid to the spacing of the legs of the yoke, and to making good contact with the surface being tested. The orientation of the yoke with the weld can be very important, and BS6072 indicates that the yoke legs should be perpendicular to the weld line. There is some evidence that better performance may be attained by reducing this angle to a more acute orientation with the weld.

Magnetisation by permanent magnets is not considered to be as satisfactory as magnetisation by the previous methods, but may sometimes be acceptable. Magnetisation by prods is not generally to be encouraged owing to the risk of local hardening caused by arc strikes following accidental contact with the steel.

For methods using electrical magnetisation, attention should be given to the nature and design of the power supply equipment, which should be of approved design and should not be too cumbersome or difficult for the diver to use. Similar remarks also apply to the design of the ultra-violet lamp and ink injection equipment.

Light levels should be checked by use of a luxmeter, of which there are several available commercially, some of them combined with flux-measuring devices. At present, BS6072 calls for the use of low

43

ambient light levels not exceeding 10 lux. This restriction can cause marked difficulties in the scheduling of inspection by MPI, as near- surface underwater inspection has to be done on moonless nights and in calm conditions. There is evidence to suggest that the 10 lux restriction is too conservative and that MPI can be carried out satisfactorily at higher levels of illumination. Procedures have been approved by Lloyds and DNV to permit the use of MPI at levels up to 140 lux, and some operators have successfully used MPI at very considerably higher levels. The effect of higher light levels on the probability of detection performance of MPI has not been ascertained, and therefore the recommendations of BS6072 should be exceeded only with caution.

However, if a significant enhancement to the amount of information gathered by an inspection in terms of being able to inspect in addition, an area which would otherwise not be possible within the time schedule, or if the risk to the inspector is decreased, then the use of MPI at high ambient light levels may be justified and should be given serious consideration. Some stipulations may have to be considered before the technique using these light levels are used. When MPI at high light levels is employed as the only crack-detection method, it would be wise to reduce the expected detection performance significantly, and not to place excessive weight on the findings. If MPI is being used to confirm the presence of a defect detected by another method, such as an eddy current method, then a less conservative approach may be in order. In all cases, the Owner/Operator should ascertain the suitability and effectiveness of the technique under the envisaged conditions before its formal use in a Survey, and must obtain the approval of the Regulatory Authority for the procedure to be used.

It is important that the correct grade and type of magnetic ink is used, as its composition, make-up, and nature can have a considerable effect on the sensitivity and reliability of the MPI technique. Where applicable, the ink should normally be freshly made-up in the correct proportions, and in any case it should be agitated thoroughly just before application. Many designs of test equipment perform this mixing and agitation automatically. It is advisable not to use an excess of ink, as improved detectability is not gained, and gentle agitation of the water near the test site should remove surplus ink.

44

CLOSE VISUAL INSPECTION - 7.7.2

CVI by an experienced diver can provide invaluable information and diver/inspectors should be reminded of the need to use their eyes in conjunction with other NDT methods. It is obviously suitable only for surface-breaking cracks, but features such as weld undercut and other fabrication defects, which might confuse supposedly more sophisticated methods, can be readily assessed. Apart from some limited investigations done around 1980 there appears to have been no research on the POD and AOS capabilities of CVI on cracks in steel.

Various proprietary kits are available for the replica recording of defects, but the accuracy of reproduction is often questionable, and full reliance should not be placed on this method of characterising a defect. This procedure is generally more applicable for dents and other damage rather than for cracks. It is sometimes possible to use computer processing of stereo-photography or video images to yield information about the size of a defect such as a dent in order to assist onshore engineering assessments.

INTRODUCTION OF NEW EQUIPMENT - I 1

Trials

As with verification trials, it is most important that the trials should be made as representative of the actual offshore conditions as possible and that they should be as statistically valid as possible. Whilst it is not possible to give a comprehensive list of the important variables, the following should be given serious consideration:

Trials must be blind, with the trials inspector at all times being unaware of the true location and nature of the flaws. This implies that:

> trials specimens must not be uniquely identifiable by the inspector, to prevent 'learning' from previous trial runs only the controller or director of the trials should know the true locations of the flaws in each sample

>

45

> where it is necessary to test various geometries, multiple samples of each geometry should be prepared some form of concealed identification will have to be used to allow unambiguous identification by the trials director

> some specimens should not contain any flaws in order to reduce the expectation of finding flaws and to simulate the effects of boredom.

>

Trials specimens should be made of steel to the appropriate grade and fabricated using welds typical of those found offshore.

The geometry of trials specimens should be representative of typical offshore applications.

Flaws used in the specimens should be representative of those expected offshore - this can be a major problem and will be discussed further below.

The trials specimens should contain a range of flaw sizes if meaningful information on overall performance is to be obtained.

The flaws should be well-characterised before the trial by a variety of NDT techniques. If possible, after the trial, at least some of the specimens should be destructively sectioned and examined optically to establish the true flaw characteristics.

The inspector should be generally experienced and qualified on other NDT equipment, and should have undergone training by the manufacturer on the equipment under trial, including how to recognise when the NDT equipment is malfunctioning.

In general, although manufacturer’s representatives may be present during the trial, they should take no part in it once the initial training is over and the formal trial has begun.

The results must be properly documented, with good QA.

46

In addition, at least some of the trials should be conducted in front of reliable independent witnesses.

The manufacture of specimens deliberately containing flaws can be fraught with difficulty. Trials ideally should be carried out on representative flaws produced by similar processes to those occurring in the real structure. However, the creation of such flaws in the laboratory environment can be difficult and expensive and it may be necessary to use artificial flaws. This must, however, be justified for the NDT technique in question prior to a trial. The performance of certain NDT techniques depends strongly on real flaw characteristics such as the surface roughness and crack opening. The most common flaw sought in offshore installations is the surface-breaking fatigue crack, usually around the weld toe of tubular joints. The reproduction of tubular specimens containing such flaws is not simple and it may be acceptable to use artificial defects such as spark-machined notches or copper strips welded into the weldment. A significant problem is arranging for sufficient numbers of different sizes of flaw to constitute statistically-valid trials - it can be shown that 29 specimens in each size class are required to allow the establishment of a 90% probability of detection with 95% confidence. Further information is available in references given at the end of the Bibliography.

Measures of performance

The performance of NDE equipment is difficult to define, but considerable progress has been made in the formal measurement of performance using a number of parameters, principally the probability of detection (POD) and the accuracy of sizing (AOS) - see Definitions in 2.1 - depending on the nature of the equipment in question. Typically, POD is presented as curves showing the chance of detection plotted against the most significant defect size, which is often depth, but in some cases length or a combination of the two. It is vital that such curves be generated only from statistically significant data by persons familiar with the underlying concepts. Reference may usefully be made to documents such as those recommended in the Bibliography.

Since the definition of the size of defect is itself not simple, attention should be paid to the basis for the construction of POD or AOS curves. It is recommended that such curves should be used chiefly as a guide

47

to performance, as they may have been derived under experimental arrangements not exactly simulating offshore conditions, and it is likely that the true performance of inspection equipment offshore will be somewhat degraded. In any event, Owner/Operators should ensure that appropriate procedures are developed and tested for their particular circumstances and installations.

The performance of an NDT system is also measured by the number and distribution of the spurious indications. Some of these plausible, but false, indications arise for no obvious reason, and must be considered to be part of the statistical fluctuations accompanying any process of measurement.

Indications may also arise from causes other than cracks. Defects in the parent steel, such as laps, can be detected, but welding can result in features, which although not cracks, can give crack-like indications in some circumstances. Examples include: copper traces; weld strikes; and weld undercut.

Most NDT techniques are incapable of fully characterising an indication. It is normal practice to check an indication with another NDT technique, not only to confirm that the indication is not spurious, but also to acquire supplementary information to assist in the assessment of the flaw. For example, MPI is often deployed to check an indication produced by an eddy current system. Depending on the technique and the application, use of a second technique to check an indication may help to reduce the waste of time resulting from the number of spurious indications. However, it is not possible to give detailed guidance, as the evaluation depends heavily on the reliability of the NDT techniques, the consequences of incorrect detection, and on the Owner's IMR strategy.

REMOTE OPERATED VEHICLE DEPLOYMENT -12

Historical Development

ROV Trials

UNDER WA JER VEHICLE TRIALS LOCH LINNHE, 1981-1982 The then Department of Energy carried out a series of underwater vehicle trials during the winter of 1981-1 982. They used three vehicles

48

and the trials were carried out at a depth of 330 feet. Nine tasks were carried out, all of them of a mechanical nature. The only one, which could be classed as being associated with inspection, was the cleaning of a plate. The mechanical tasks carried out were the cutting of a length of steel tube, the deployment of an explosive cutting charge, the cutting of wire rope, the attachment of a lifting line to items of debris, the removing, replacing and tightening of nuts to a set torque, the turning of a valve wheel, the disconnecting and connecting of a hydraulic coupling and the establishment of a guide wire. The vehicles used were not identified but they had various manipulators with between 3 and 7 functions.

All the tasks were completed during the trials, but the variation in times taken between vehicles was substantial. It took 22 minutes for one vehicle to carry out one particular task whereas another vehicle took 1 10 minutes. In the case where a down line was being attached, one vehicle took 22 minutes where another took 460 minutes Improvements were recommended in pilot training, camera positioning, manipulator control, navigation information systems and the performance of power supplies.

The report on the trials concluded that underwater vehicles could complete the tasks set, given the necessary preparation and tooling and that they generally showed the potential to offer such services to the offshore gas and oil industry. It was suggested that the future role of underwater vehicles was for dedicated tasks using specialized tooling and that the role of the vehicle would be to supply this tooling to the installation, to deploy it and then to form the power and control link to the surface for that particular tool. Such tooling would have to be developed by the remote vehicle operator industry, given the commercial incentive from the oil and gas companies who would operate and use the vehicles.

The report did not address the role of the ROV as an inspection tool.

REMOTE OPERA TED VEHICLE CAPABILITIES STUDY 1986 A study of the ROV capabilities was commissioned from Winchester Associates in 1986. The object of this study was to assess the capabilities of available and future vehicle systems with regard to their suitability for inspection or combined inspection/cleaning operations on offshore structures, and to assess their capabilities with regard to pipeline operations and repair techniques. A total of 46 vehicles were assessed and a final list of 20 drawn up. There were over 70 different ROVs in the market place at the time.

49

The study confirmed how little progress had been made in platform cleaning and inspection vehicle design from 1981 to 1986 and indeed the surprising amount of repetition that there was. The vehicles were split into two categories, namely small inspection vehicles with limited NDT capability and larger vehicles capable of undertaking certain cleaning procedures but with a similar level of NDT capability to that of the small vehicles. The main trend was towards large systems produced by modifying standard production vehicles to produce hybrids. Only two dedicated systems had been built. The small vehicles, on the other hand, were becoming more and more capable and very much cheaper, and designers had started to take advantage of the miniaturisation of some of the essential components such as video and photographic cameras. There was still no question at that time of these small vehicles being able to undertake more than a limited inspection role but, as shown later, the role of these vehicles was becoming more important as they were able to take lighter, more portable inspection tools to the inspection site.

The conclusion of this review was that it was not possible to perform routine cleaning and inspection tasks wholly with a single vehicle or indeed, with a combination of vehicles.

A problem highlighted in the report was that the cost of a vehicle was typically 22.5 million and it would require a very high day rate to recover this capital cost. It was suggested that if the vehicle cost could be reduced to 20.75m, a commercial rate to put a system into operation for cleaning and inspection would be in the order of 23- 4 , 0 0 0 per day. This would include the use of 6-7 personnel, with the ROV being deployed from a fixed structure. In comparison air diving from a support vessel could cost around 212-14,000 a day, so substantial savings could be made if the system could be made at the lower capital cost.

Prior to 1986 the known market for cleaning and inspection using ROVs amounted to only two contracts both in the Norwegian sector. Some operators did use the vehicles for visual inspection and cathodic protection measurement, but the amount of work represented only a relatively small amount, of the order of days. One of the problems identified was that the technology had not been developed to make the complete package of cleaning and inspection available. It was stated that whereas it might be possible to deploy an MPI system underwater to carry out inspection, there was still no means of reporting the findings in terms of a defect location, even based on an elementary clock system. An MPI system mounted

50

on an ROV had been tested in 1985, but the system was based on parallel conductors and was very difficult to deploy.

It was suggested that eddy current systems could be used for crack detection instead of magnetic particle inspection. This would need a highly sophisticated manipulator which could track the weld very accurately to allow the eddy current device to follow the toe of the weld in order to accurately detect linear defects. One major operator had suggested a further development using an underwater moulding material which could be applied to the area that was to be inspected with MPI. This material once set could then be peeled off and taken to the surface where the magnetic particle indications could be clearly viewed. This would then get over the problem of recording any ambiguities produced by the magnetic particle inspection system.

CLEANING AND INSPEC77ON TRIALS I988 In 1988 an offshore inspection company used an underwater electro- magnet and carried out trials with an offshore operator in Norway. Initial trials at Stavanger using wire brush cleaning were eventually abandoned in April 1988. The operator then moved to an offshore field in July and changed its cleaning technique from wire brushing to low pressure grit blasting. (i.e. 25-40 psi above ambient).

They found this was more efficient and easier to apply and cleaned three nodes. They also attempted to carry out magnetic particle inspection. However, problems with the ultra-violet lamp that they had hired led to the trial being abandoned.

In November 1989, the operator again mobilised for trials at Dusavik in Norway. Three nodes containing various diameter concrete or epoxy coated tubulars were available for cleaning and inspection. In addition, a longitudinal fillet weld and an 8 ' tubular sample (prepared by Det Norkse Veritas to approve diving inspection procedures) was also inspected. The DNV samples provided contained 1 1 defects. These included small defects, holes and longitudinal defects up to 15 mm in length.

All the nodes were satisfactorily cleaned using the low pressure grit system. Inspection was carried out using an electro-magnetic yoke, which had a built-in ink dispenser and was carried by one of the two manipulators (a second manipulator carried an ultra-violet lamp and a camera). Defects in the nodes and the longitudinal fillet welds were all detected using the AC yoke technique and a variety of inks. There

51

was a problem, however, in recording the indications. The camera was used to record the indications but because of the poor illumination produced by the ultra-violet lamp, although the indications could be seen clearly, the background weld could not be, and the position of these indications around the circumference of the samples could not be defined. This was a common problem with inspection equipment of this type.

AUTOMA TED UNDER WA TER INSPECTION AND MAINTENANCE fAIM} PROJECT I989 In 1989 the UK Department of Energy and its OETB reviewed the status of ROVs for inspection and maintenance. Over the years ROVs had progressively replaced the diver in sub-sea tasks such as drilling support, pipeline surveys and even the maintenance of sub-sea production systems. In each of these areas ROVs gave cost savings through higher productivity than divers did and provided much improved operational data. ROVs also made a valuable contribution to health and safety offshore by reducing or eliminating the need for divers in the water. The Offshore Energy Technology Board (OETB) thought that there was a broad consensus in the industry that full development of the use of ROVs in underwater structural inspection with maintenance and repair was appropriate. Against this background the OETB agreed to a new initiative to look at the possibility, commencing with a study report (Phase 1 ).

The report was mainly aimed at the use of ROVs to replace saturation diving for the inspection of deep water platforms and suggested there was a significant cost saving to be made by such replacement. It gave an estimate for one UK continental shelf operator with a spread of platforms and depths, which indicated an annual saving well in excess of 23 million against the then annual cost of E6 million. It listed other advantages which result from the AIM project which included a major reduction in the number of personnel deployed offshore in inspection duties and less weather dependency (extending the time window for ship mounted operations and reducing the incidence of 'waiting on weather' periods). More consistent NDT standards and improved inspection records would also be achieved. ROV supply vessels would replace the more expensive diving support vessels, and for operators with multi-role support vessels, savings could still be achieved by using an inspection ROV in lieu of mixed gas divers.

52

Developments in ROV technology

ROV DEVELOPMENTPROJECTS In 1990 an analysis was carried out by Havron Energy Systems on a number of ROV projects that were being prepared by various organisations for sponsorship. A simple cost benefit analysis was carried out for a Southern Basin North Sea field based on an annual diving inspection programme of 200 days and a budget of 22.85 x 106. The air diving rate was estimated at Sl4K/day and the ROV operation rate at EIOK/day. If the programme could be divided into 60% air diving and 40% ROV operations, then a cost saving could be made of 10% of the annual budget, i.e. E285,000/year. This value increased by to 75% for deep water situations using an ROV operated from a platform. One UK operator quoted E l .7 x 106 for a combined diver plus MPI inspection from a diving support vessel, compared to S.4 x 106 for an ROV platform mounted operation. The benefits identified were improved efficiency, improved safety since there are less operations involving divers and improved operations, which could be applied to shallow, deep or hazardous diving operational areas, (i.e. NW Atlantic, Australia, Canada).

Nine ROV related projects were identified including the AIM project. Four of these projects were prepared by ROV companies each using its own ROV. All had the same basic objective i.e. the development of NDT equipment suitable for existing ROVs. All suggested the application of either magnetic particle or eddy current equipment, which could be attached to a preferred manipulator.

Heriot Watt University proposed a project to develop ROV control and ROV data processing and interpretation. University College London proposed a project (ICON) which quantitatively assessed NDT techniques that could be applied to ROV use.

Two further projects proposed by Tecnomare and Winchester Associates, had the objectives of producing new underwater vehicles. The Tecnomare project was part of an EEC funded 'Eureka' project to produce two vehicles. These were a free-swimming 6 ton, 20 foot long vehicle and a work and inspection robot vehicle. The vehicle's design was based on the success of the Tecnomare TV track meter and supervisory controlled underwater tele-manipulation system.

The Winchester Associates project involved the production of a novel astable vehicle. This is the most advanced ROV project to be proposed in the last 10 years and the vehicle could have a major

53

impact on the introduction of ROVs for complete inspection and maintenance. Winchester identified two areas of weakness that they needed to eliminate. The first was the production of a manipulator which would produce spatial resolution and repeatability of better than 2 mm. The second was the identification of the correct inspection equipment. The two are very much complementary. If an inspection system is identified that does not require accurate weld tracking the need for an accurate manipulator is not so critical. Winchester studied the offshore manipulators. They also examined an automobile manipulator which although very accurate, was too heavy to be used on ROVs. Winchester suggested the use of a coarse manipulator for cleaning operations and a more sophisticated manipulator for inspection. Their proposed vehicle is of a novel design incorporating technology accepted in other fields and transferred to this project. The concept is of a highly manceuvrable small vehicle having a solid docking system. It removes the majority of vehicle access and manipulative maneuverability problems. The problems highlighted by manipulators and inspection techniques could be solved once new inspection techniques became available. The vehicle proposed would meet the majority of the requirements of the AIMS project.

These projects led to the conclusion that, in order to progress the use of remote operated vehicles for inspection and maintenance, two projects were of importance, namely the ICON quantitative assessment programme using ROV inspection techniques and the Winchester Associates astable vehicle programme.

ROV DEVELOPMENTS

During the early 1990s several papers were presented at conferences worldwide to highlight the improvements in ROV deployment and application. Since 1980 there has been a great number of developments in the use and design of ROVs such as the development of ROV NDT tooling, advanced ROVs for underwater maintenance, advanced high performance units for inspection repair and maintenance, the implications of ROV hydro-dynamic design, and a new generation of ROV tooling for subsea development tasks. This section reviews some of the published documents available that report these developments.

In the 1970s the ROVs that were available were unreliable and limited the number of tasks that they could perform. The 1980s saw the development of more advanced tooling and the ROVs being used for more sophisticated tasks interfaced with subsea assemblies. These

54

projects were mainly funded by the larger operators and proved that large scale components and modules could be transported subsea. The paper concluded that ROVs would be developed to a stage where they could be used in primary roles in deep water development and become more cost effective for use at diver operating depths but that the problem of weld inspection still had not been solved.

A paper was produced by an ROV operator who described the tasks carried out by divers as being general and close visual inspection, mud sampling, cleaning, cathodic protection surveys, ultrasonic thickness measurements, magnetic particle inspection and flooded member detection. Initially only a small number of these tasks could be accomplished by an ROV but in the 1990s it was suggested that at least four vehicles were capable of undertaking these tasks. This paper also presented a cost analysis showing that the cost of using a dive support vessel with a diver operation is 8 times more than using a platform-based ROV.

Papers have been produced describing the problems of weld inspection using magnetic particle inspection and eddy current devices. A single pole electro magnet had been produced which could be applied by a single manipulator. Also a single manipulator could apply an eddy current device which had been developed. The problem with both of these techniques was that very accurate location of the manipulator was required and only small areas of the weld could be inspected for each application of the manipulator.

One further important development in the use of remote operated vehicles was reported. This was the use of ROV flooded member detection using gamma transmission techniques. In 1990 one supplier of this equipment used both large and small ROVs to inspect 19 platforms for flooding detection. During that period of time they inspected 1,802 members for flooding using the gamma transmission technique. The success of this project suggests that ROVs can now be used quite economically and safely for flooded member detection.

Seven papers updating the situation regarding dedicated manipulation, diver ROV interface, ROV for maintenance and the use of telemanipulation and 3 dimensional control were presented at one conference.

The need for the use of ROVs in deep water (i.e. >300 metres) had been identified in the latter part of the 1980s when exploration had led to the development of the deeper sections of the Gulf of Mexico

55

and off the coast of South America. Together with the need for the use of ROVs the associated requirement was identified for a manipulator that would operate reliably at these depths. Workers at Woods Hole developed a dedicated manipulator and proved its reliability to operate at 700 metres during trials in the Mediterranean. The manipulator used was designed to operate at depths up to 6000 metres fully equipped with forward and side scanning sonar, still and video cameras and a manipulator. Whereas the majority of manipulators produced in the past had been designed for high strength so that they could be used for turning valves, cleaning and attaching wire guides, the manipulator described was more sophisticated and had less strength but could be finely controlled. This manipulator also had to operate from a hovering ROV rather than from a fixed base. The combination of high performance electric motors and low friction, zero back lash reducers achieved the requirement of a scientific manipulator that was able to collect artifacts from a shipwreck without any damage.

Two complementary papers were presented by representatives from a Norwegian operator and an inspection company describing the use of a combination of an ROV and a diver for the inspection of platforms. It was calculated that the combination of Diver Support Vessel (DSV) and diver had resulted in an average of 32% downtime, including waiting on weather (WOW) and of the 68% available time only 10% had actually been used for inspection. In 1989 they decided upon a policy of introducing ROVs with a dedicated DSV together with divers to reduce the downtime. The 1989 structure and riser programme consisted of the inspection of 37 steel structures and 51 risers. All of the structures and risers were subject to general visual inspection, including cathodic protection measurement. 43 welds were cleaned for further magnetic particle inspection by divers.

50 areas were cleaned for flooded member detection and a number of areas were cleaned for close visual inspection. Two work class and one eyeball ROV were used together with divers and the whole operation was undertaken in seven weeks including mobilisation. Downtime, including WOW, was reduced to 17.2% and the use of divers was reduced to 10% of the inspection tasks.

Once again the weakness of the ROV not being able to carry out the magnetic particle inspection was highlighted, but the diver worked with the ROV to carry out these and other tasks, using the ROV as a work station. This philosophy was further progressed with the development of a modified ROV to produce theDIVEROV. The use of the DIVEROV with the diver increased the efficiency of the inspection

56

programme; the diver and ROV worked together or independently on the same or different sites. The weather window for the DSV was increased and the number of support crew members was reduced, (the six DIVEROV personnel being part of the existing DSV crew).

This combination of diver plus ROV is acceptable for shallow and medium depth applications and is proving to be efficient and economical. However, the installation of platforms at greater depths produces severe limitations for the diver/ROV combination.

Some diving inspection companies have always been promoters of the atmospheric diving system (ADS) in which a 'diver' can operate at much greater depths. These ADS can be propelled by the divers (e.9. the JIM suit) or motor driven (e.g. the WASP). Although they can operate at great depths because of their construction, mobility is often limited and the work tasks performed have been mainly of a mechanical nature, e.g. lifting grout bags or replacing nipples. Nineteen examples have been published of this type of application.

One further answer to the problem of inspecting and maintaining deep water platforms and subsea assemblies is the use of dedicated remote operated maintenance vehicles (ROMV). These are vehicles produced to service and maintain subsea assemblies in a repetitive manner. They are deployed from low cost supply vessels fitted with heave compensation systems so that they can operate in severe weather. The system developed for a Norwegian operator swims to a subsea assembly, locks on to a track system and pulls itself to the subsea installation where it carries out programmed mechanical tasks before returning to the surface. This eliminates the costly exercise of a special DSV and a saturation diving spread.

Three-dimensional imaging systems have also been developed and the main application has been to combine the system with a computer-controlled telemanipulator to guide and control manipulators for special applications. The subsea part of the system includes the manipulator and the TV cameras. The topside section contains the surface unit together with the supervisory computer, the surface arm controller and the interface system. The aim of the system is to aid the manipulator in tracking accurately around the weld geometry on a node. The geometry is constructed by the computer before the task starts and the TV cameras produce and update the X, Y and Z coordinates of any point on the geometry of the node. This information is then fed to the supervisory computer which compares the results with the computer-generated model and automatically tracks the front of the manipulator. A man-machine-interface is used

57

through the three axis controllers to traverse the manipulator. Trials were performed to carry out mechanical tests such as valve actuating, cutting of wire ropes and connection of couplings. All were carried out satisfactorily, but the error in location was 5-20 mm. This would not be adequate for performing weld inspection, as most weld scanning techniques require better accuracy.

The same system was used to aid accurate navigation around platforms and was very effective. It was possible to simulate added mass, extra drag coefficients and actuation delays, and still produce good tracking performance. An offshore inspection company published a table of comparative costs for DSV inspection and platform-based ROV inspection in 1991 describing the use of a tether management system. This is a platform mounted, bad weather, ROV deployment system. It uses a clump weight, guide wire and heave compensation and allows the ROV to be deployed in weather conditions up to Beaufort Scale 10. A comparison was carried out in 1990 which showed the capabilities of the platform deployed ROV for 1990 compared to 1985, all but weld inspection was claimed to be possible. Using a workclass ROV and a six man team it was possible to increase on line operations from 48.8% to 90.5% with only 0.5% waiting on weather compared to 36.5% in 1987.

A Norwegian inspection company in 1991, using a similar platform deployment system inspected three platforms in four weeks commencing in mid September. One complex was inspected from the central platform, the ROV swimming 150 metres in water to reach the furthest points of the other two platforms. Flooded member detection using the gamma radiation technique was used, deployed with an eyeball ROV.

The use of the atmospheric diving suit has been extended with the more adaptable NEWT suit. It has been used with both an eddy current system and the ACFM system for crack detection. This ADS is lighter in weight and easier to use than the original JIM suits and is more adaptable to subsea inspection.

ROV weld inspection

The problem of weld inspection using the ROV is complex and can be separated into various areas. The four main components are the selection of the ROV, the selection of the manipulator, attachment of the ROV onto the structure and the inspection tool.

58

The initial philosophy of ROV weld inspection was to have the ROV copy the diver application as illustrated previously and the tools that were available suited this purpose. With the initial introduction of electromagnetic techniques this still required weld toe tracking which necessitated the manipulator to be able to track the weld toe in the same manner as the diver but with the disadvantage of the ROV's instability. This would require that the ROV should be rigidly attached to the node weld area using one of the methods suggested above. A sophisticated manipulator would then traverse the inspection tool along both weld toes and in the case of wide caps make an extra scan along the centre of the weld cap. The manipulator should have the manoeuvrability to be able to scan all the weld with a limited number of ROV placements. Such manipulators are being developed.

The second philosophy is to use a macro inspection technique that does not require accurate weld tracking or placement of the inspection tool. One technique that was applied earlier, but suffered from access problems, was the application of the electromagnet. In that case two manipulators were required to apply the technique, but now an integrated subsea electromagnet has been produced. This includes a module attached between the poles of the magnet which contains two fluorescent light sources, a miniature colour camera and an ink jet supply. The initial system was produced for diver application and the ink supply was manually activated and the camera had a fixed focus, requiring diver manipulation when recording the defects. This did not make it conducive to ROV use. Since this initial development a remote focus and ink control has been attached, thus the electromagnet can now be applied with one manipulator. As long as the poles of the magnet are across the weld the magnetic field can be applied and the weld inspected, the ink being sprayed and the weld being viewed remotely and the defect if present being recorded. Thus the ROV can now be used to clean the weld areas and from the same tool skid, deploy the electromagnet to locate and detect surface breaking cracks. The length of weld being inspected in one application would be 150mm.

Although the technique can then be used to detect the defect, it cannot be used in isolation, as it will not be able to determine the severity of the crack. The length will only be able to be measured if there is some form of measuring device on the weld and there will be no defect depth information available. Thus a diver would be required to determine these values or another ROV tool would have

59

to be deployed. If the defect should be located beyond the diving limit, then an additional technique would have to be used.

An alternative compared with magnetic particle inspection is the alternating current field measurement (ACFM) technique, which is a non-contacting electromagnetic technique used for subsea weld inspection. This has been further developed as a multi element array system for ROV deployment. The technique allows the detection of surface breaking defects without the necessity for cleaning the surface to be inspected. The technique, as well as detecting the defects, also sizes them in terms of their length and depth. The diver system has undergone successful POD/POS trials and trials through 2mm of epoxy coatings.

An original system was developed consisting of an array of four rows of twelve elements mounted in a flexible matrix which covered an area of 150mm x 50mm. This system could then be deployed by a manipulator and as long as the array covered both toes of the weld, then the weld toes and cap would be inspected. Using this technique the weld could be inspected using a single manipulator and any defect detected could be sized without removal of the coating in order to evaluate its severity. The data was presented as two rows of information , one presenting the Bx and Bz data and one the Butterfly data, for each row of sensors, from which the sizing data could be obtained. This mode of deployment is particularly well suited to ROV operation. Following this development a commercial system was produced for a North Sea operator which had to be integrated with an ROV and could operate in water depths of up to 300ms.

ROV DEPLOYED ACFM WELD INSPECTION SYSTEM

To enable an ROV deployed inspection system to inspect welds, it must be able to satisfy certain parameters.

1 ) The inspection system must be able to detect the minimum defect size specified by the platform operator, in either mild steel or duplex steel and if required through up to 5mm of coating.

2) Probes must be available to meet the geometry requirements, which may be 90 degree, flat butt welds or 30 degree welds.

60

3) The system must be compatible with any typical work class ROV without the need for deployment with sophisticated manipulators and be able to operate in water depths up to 300 metres.

4) The data should be in a form that could be audited and stored for reappraisal.

One system produced consists of three array probes as specified above which can be mounted on an ROV skid. The skid also contained a subsea electronics unit, a suction system and a cable management system. The topside unit consisted of a topside communications interface and a PC running the data collection and an a lysis software.

It was essential that the array probes should be able to hold in position whilst collecting data and that their position should be known on the weld being inspected. In order to do this a suction system was produced for the two original probes and these also included inclinometers and position transducers. The suction system and the flexible holders on the arrays accommodated some movement by the ROV. The probes also included induction solenoids to induce the uniform electric current into the weld to be inspected, the multiplexer circuitry and the electric motor to drive the array. These array probes consist of 5 rows of 12 pairs of sensors, each capable of measuring two components of the magnetic field around the weld area and their variation if a crack is present. The sensors are flexibly mounted in order that they may follow the contours of the weld profile and these are all multiplexed together. In order to have a high degree of defect detection, the array is moved within the static inspection head housing along the weld length, so that information can be collected every 2mm by the sensors. This information is also required for defect sizing. The deployment of the array, the movement of the sensors and the collection of the data takes 20 seconds, with the data capture taking only 5 seconds of that operation. The sensors collect them as five separate over-layered coloured traces on topside PC giving data presentation.

The depth and length of the defect can then be determined. The length measurements are simpler to determine because the distance scale is fixed for a given sensor spacing. If the defects are longer than the sensor length, then adjacent scans are combined to simulate a longer scan. Because of the flexibility of the software package, it is also possible to have a view of all of the defects from a given ROV

61

location and this gives the location of defects along the weld length. The software also allows the inputs from the inclinometer and the position transducer to be viewed. The fixed array for the inspection of the 30 degree weld geometry produces data in the same form but the array needs only to be in position for 3 seconds to collect the data. This probe is different from those previously described, in that the sensors are 6mm diameter and are more closely spaced than the flexible arrays.

The technology associated with arrays has been expanded even more with the capability of detecting and sizing volumetric defects. One such application is the detection and sizing of corrosion pitting in hazardous areas. An ACFM array has been produced that can detect and assess corrosion damage. This can be deployed by a manipulator and thus can operate in areas of high risk such as nuclear reactors or could be used with an ROV for corrosion mapping. The array sensors are mounted in an inspection head and the head is placed in position by a manipulator. The array is then moved within the static head producing a linear scan to provide complete coverage of an area. The area covered is 127mm x 127mm and the system is capable of detecting pits 0.75mm diameter x 0.75mm deep in both carbon and stainless steels.

NEW DEVELOPMENTS IN WELD INSPECTION

In 1990 a joint project between two Norwegian offshore operators and an inspection company commenced the development of advanced technology for underwater inspection. The system was used in offshore operations in 1994. The original project lasted 24 months and was completed in November 1992 and demonstrated through wet and dry tests, that all types of NDT could be deployed from an ROV.

The system was redesigned to incorporate all of the inspection tools into a package that was in front of the ROV. It was decided that a new ROV was required and this was based on a vehicle around which the inspection tool package was built. This package could accommodate 9 tool positions. A computer aided telemanipulator (CAT) was used with an environment acquisition system to deploy the tools. In 1994 the system was deployed with the following results. The original system had a standard eddy current probe, but this proved vulnerable to mechanical wear and damage and the probes were

62

housed in more robust stainless steel housings. It was also decided to use an array scan eddy current tool during offshore trials. During the offshore trials the eddy current operation was too dependent on the environment acquisition equipment and a semi manual function was preferred. An MPI system was also deployed and although it worked well, a change in the location of the ink nozzles was necessary as the strong currents washed the ink away from the premagnetised area.

Another project aimed at the reduction of underwater inspection by employing ROVs used 3D images together with an interactive window, which inserts a video image into a graphic display, reconstructed by using stereo images. A supervisory control system is then used to control a hydraulic manipulator which deploys the tools and carries out dedicated tasks. A trial was carried out using an ACFM standard probe with the system in a tank in 1995. The probe was able to follow the weld toe in a dynamic situation and detect defects

A further project had the objective of producing a new subsea system for cleaning and inspection of nodal welds. This was funded by the UK government and industry together to produce a new manipulator and a computer control system. The supervisory controller provides full control over the manipulator including manual and automated robotic task execution. A 3D video representation of the ROV, manipulator and worksite is used to monitor the manipulator during its operation. Electromagnetic position sensors are used to ensure that the inspection probe follows the weld toe. ACFM standard and array sensors have been deployed by the system during trials.

The ACFM system has been used during an industry project with the Norwegian system when both the standard and array probes were deployed by ROV. The standard probe was used to inspect node geometries and the array was used to inspect flat plate welds. The ACFM array probe has also been deployed by an ROV by an offshore operator to inspect node geometries on an offshore structure.

Ultrasonic inspection has also been done by ROVs deployed with scanning frames to carry out wall thickness measurements and ultrasonic annulus measurements in the UK and Australia. The deployment is being further developed to include corrosion mapping and node inspections.

63

Access problems

Although it is now possible to inspect welds and detect defects and size them, the main problem is actually getting the sensor to the weld area. With the ACFM technique it has been proven that the technique is independent, being able to inspect, detect and size if the ROV manipulator combination can deploy the array to the work site. In the majority of cases it is not possible to predict if a particular combination of sensor, ROV and manipulator, will be able to complete the inspection required until the ROV reaches the worksite.

A software package has been developed that can simulate inspection and other intervention tasks, on subsea nodal welds. This has a number of manipulator and ROVs in its library, which it can combine and deploy on models created from platform drawings.

The system has the manual as well as the ACFM array inspection tools within its library, so that access and coverage can be modelled prior to inspection being carried out. This then gives the inspection company and the operator an expectation of the number of welds that will be inspected by any ROV manipulator/NDT sensor combination. The system also has magnetic particle inspection, flooded member detection, cathodic protection tools as well as cameras within its library.

The modelling will give reasons as to why some welds cannot be accessed , give the best probable direction for approach and collision warning. It also has the facility of virtual viewing simulating the aspect from the ROV at the work area and can model the best position to dock and the optimum number of sticky feet to deploy to hold the ROV in position.

With the cost of offshore operations being so high the software package can play an important role in planning an inspection programme.

Defect rectification, removal and repair

Once the defect has been detected and sized the structural engineers will have to make some decision as to how to proceed next. From their calculations they will be able to decide whether the defect can remain in place, whether it requires removal by grinding or a major repair is required. The decision will not only depend on the

64

size of the defect but also on the location of the weld within the structure, the fatigue sensitivity of the weld, the stress concentrations around the joint and how the defect will affect the structural integrity of the welded joint. Once a decision has been made the problem then arises as to how the solution can be implemented. In standard operations divers would be used to carry out subsea rectification and repair work, either by grinding to remove the defect or by the repair of the weld using a dry habitat or wet welding techniques. If the aim of the offshore operator is to use diverless intervention, then the ROV will be expected to carry out rectification as well as inspection.

At the present time there is not a proven technique that can be deployed by ROVs, but there are projects which are investigating the possibilities.

One project is aimed at investigating the various inspection and repair techniques that could be applied by ROVs. At the present time there are several inspection techniques being deployed by ROVs. The effectiveness of selected techniques will be demonstrated during the project. There are also several metal removal techniques being developed that could possibly be deployed by ROVs. These include electro-chemical machining. It is intended to collect data from the inspection to give the parameters of the defect and to use these to control the metal removal technique, thus ensuring efficient removal. Existing simulation software will be adapted for planning, inspection and repair, the electro-chemical cutting tool will be adapted for subsea ROV deployment and a special ACFM tool will be produced specifically for re-inspection tasks. The effectiveness of the defect detection and removal will be validated by continued fatigue testing of repaired weld joints.

A second project is to evaluate diverless inspection repair and maintenance (IRM) for subsea completions and deep sea structures. The project will review possible IRM activities using ROVs and provide a database for ROV tools available for IRM tasks. It will assess the use of task simulators for procedure development and evaluate two ROV systems, for metal removal and a maintenance task.

65

APPENDIX B - Summary of underwater NDT qualification schemes

B 1. Introduction

At present only one well-developed, special scheme for qualifying underwater examination personnel is in extensive use, the CSWIP scheme in Great Britain.

Other schemes have been developed and used, but have been withdrawn or are non-operational. In some countries the CSWIP scheme is relied upon. In other countries the qualification of diver- inspectors is assured by a combination of traditional NDT operator certification according to national or international schemes and standards, diving qualification and special training for the inspection in question.

B 2. Great Britain

The Certification Scheme for Weldment Inspection Personnel (CSWIP) is managed by a board representing a range of industries and professional institutes. A working group called 'CSWIP Phase 7 Underwater Inspection Panel' prepares certification requirements and documents for underwater inspectors. The present defining documents are entitled CSWIP-DIV-7-95 1st Edition Part 1 and 2, and are dated January1 996.

The CSWIP Scheme is divided into 4 categories of personnel:

3.1 .U. Visual, photographic, video, cathodic protection, ultrasonic

3.2.U. Above skills plus MPI, ultrasonic A-scan thickness operation 3.3.U. Pilot/Observer Inspector 3.4.U. Underwater Inspection Controller

digital thickness operation.

In addition there is now a category, 'Personnel Engaged in the Electromagnetic Inspection of Welds-CSWIP-DIV-8-96'

For further information enquiries can be made to:

T.G. Jessop, Secretary to CSWIP, Abington Hall, Abington, Cambridge CB1 6AL, England

66

B 3. Other countries

Det Norske Veritas has developed a scheme for qualification of underwater inspection personnel as described in the document 'Qualification of Underwater Inspection Personnel', RP A402 dated August 1983. The document has not been withdrawn, but the scheme has not been maintained, and must be regarded as not operational.

An earlier French scheme (described in Bureau Veritas Guidance Note N.I. 182 'Procedure to Obtain the Diver-Inspector Aptitude Certificate' dated May 1984) is no longer in effect, and in France qualification of underwater inspection personnel is covered by the general certification scheme for NDT personnel.

A draft standard for training was prepared in the mid-1980s under the Canadian General Standards Board (CGSB) Certification System. The draft, entitled GPWETDR2 on 8 and dated 8-8-1 985 was prepared by a joint task force including the Canadian Society for NDT and the Canadian Association of Diving Contractors. The scheme has, however, not become operational.

67