Pipeline, riser and

subsea engineering

Fundamentals
of pipeline
engineering -
Week one
2

All information contained in this document has been prepared solely to illustrate
engineering principles for a training course, and is not suitable for use for engineering
purposes. Use for any purpose other than general engineering design training constitutes
infringement of copyright and is strictly forbidden. No liability can be accepted for any
loss or damage of whatever nature, for whatever reason, arising from use of this
information for purposes other than general engineering design training.

All rights reserved. No part of this publication may be reproduced or transmitted in any
form or by any means whether electronic, mechanical, photographic or otherwise, or
stored in any retrieval system of any nature without the written permission of the
copyright holder.

Copyright of this book remains the sole property of:

Jee Limited
Hildenbrook House
The Slade
Tonbridge
Kent
TN9 1HR
England

© Jee Limited 2009 *

* This document has been prepared by Jee Limited for Shell under contract
4600004244. The copyright remains with Jee Limited for all materials except the
sections that have been prepared by Shell:
■ Multiphase flow
■ Technical standards
■ Structural integrity
■ Flow assurance field case
 Table of contents
Volume one
FIELD LAYOUTS 9
Expectation 11
Example layouts 12
Pipeline and cable uses 22
Subsea equipment 28
Platforms and floating production systems 35
Riser configurations 40

SURVEY AND ROUTE SELECTION 49

Expectation 51
Survey techniques 53
Soil types 62
Routing of pipeline 69

PIPE MANUFACTURE 83
Expectation 85
Rigid steel pipe 86
Options 86
Seamless 90
High-frequency induction 92
UOE 95
Flexible pipe 102
4 Fundamentals of pipeline engineering

MATERIALS 111
Expectation 113
Selection for strength 114
Selection for corrosion resistance 118
Clad and lined pipes 123
Titanium and composites 127

EXTERNAL COATINGS AND INSULATION 135

Expectation 137
External corrosion protection 138
Cathodic protection 144
On-bottom stability 149
Thermal insulation 154
Pipe-in-pipe systems 167
Insulation 167
Deep-water J-lay and reel-lay 170
Field joints 173
Active heating of lines 176
Additional information 182

PIPE SIZING 185

Expectation 187
Diameter sizing 188
Basic diameter sizing 188
Single phase flow 196
Basic multiphase flow 206
Wall thickness for bursting 210
Wall thickness for hydrostatic collapse 217
Buckles 223
Volume two
MULTIPHASE FLOW 235

DESIGN METHODS 271

Expectation 273
Limit state design 274
Identification of limit states 274
Historical background 278
Derivation of safety factors 283
DNV OS-F101 design 292
HP/HT and HIPPS 297
Fishing interaction 302
Vortex-induced vibration 309

COMMON WORK 317

Expectation 319
Construction survey 320
Route preparation 324
Welding 330
Non-destructive testing (NDT) 346

INSTALLATION METHODS 353

Expectation 355
S-lay 356
J-lay 366
Reel-lay 370
Bundles and towed installations 375
Flexibles and umbilicals 386
6 Fundamentals of pipeline engineering

TECHNICAL STANDARDS 393

CONSTRUCTION SUPPORT 417

Expectation 419
Landfalls 420
Trenching and burial 429
Pre-commissioning 443

TIE-INS, SPOOLS AND RISERS 455

Expectation 457
Tie-ins and spools 458
Rigid steel riser installation 471
Risers fixed to jacket 471
Steel catenary risers 474
Top-tensioned risers 480
Hybrid risers 483
Flexible riser installations 486
Volume three
INTRODUCTION TO INTEGRITY 495
Expectation 497
Failures: frequency and incidents 498
PIMS 507

STRUCTURAL INTEGRITY 517

MODIFICATION AND REPAIR 547

Expectation 549
Isolation 550
Tie-ins 560
Repairs 569
Decommissioning 585

DECOMMISSIONING 589
Expectation 591
Introduction 592
Legislation 597
Decommissioning in-situ 605
Cleaning 606
Product removal 608
Trenching 609
Recovery 612
Re-use 619
Costs 623
8 Fundamentals of pipeline engineering

FLOW ASSURANCE 629

Expectation 631
Operational controls 632
Additives 639
Pigging 642

FLOW ASSURANCE FIELD CASE 655

PIPELINE INSPECTION 671

Expectation 673
Risk-based inspection plan 674
External survey 678
Internal inspection 682
Anomaly assessment 693
Spans 693
Pits and Dents 698
Exposure 702
Remedial works 707
Introduction to
integrity
Introduction to integrity 497

EXPECTATION

EXPECTATION

 Aim to prevent incidents

 Loss of containment
 Lack of operability
 Safe for people, equipment and surroundings
 Extend facility’s life

 Why a PIMS is needed

 What it should include

We will illustrate why a structured approach is required to integrity management and

describe how we avoid accidents.

We aim to keep the number of incidents to a minimum and operate the pipeline safely
for as long as possible.

The best way of doing this is by setting up a pipeline integrity management system
(PIMS). We examine what it should include and how it should be operated.
498 Fundamentals of pipeline engineering

FAILURES: FREQUENCY AND INCIDENTS

BATHTUB CURVE

 Changes in failure rate through life

First phase
Design mistakes
Construction incidents
Number of failures

Inherent defects
Final phase

Component replacement
Wear and tear effects

End of pipeline life

Lack of inspection
Inadequate repairs
Out-of-date or
ignored procedures
Middle phase Operator error
Stable failure rate
Third-party influences

Life of pipeline or component

Pipelines are like any other mechanical or electrical component used in industry.

Initially, there is a high rate of failure. Some of these are due to mistakes made by the
designer or the interpretation by the supplier, fabricator or contractor. These inherent
defects with the pipeline system have not been detected by the assessment or inspection
regime.

Gradually, though, the rate of failure drops to near zero, and for most of the pipeline’s
life it remains reasonably stable. Events are often as a result of third party failures or
damage – perhaps adjacent equipment may suffer a failure and impact on the line or
fishing interaction with a pipeline.

Eventually, however, the equipment wears down or corrodes away through old age.
This even applies to us as our teeth fall out and joints creak.

Even when the defect is found, it may not be easy or convenient to repair. When a
component is replaced, unexpected changes may occur because the part is slightly
different from that which it has replaced. The adage, ‘If it ain’t broke, don’t fix it’ has
some truth to it.
Introduction to integrity 499

Another common source of failure can be traced to operating or inspection procedures

and regimes not being maintained up to date. By this time, the original designer or
writer of a procedure has long since disappeared and his intention may be lost.
Operators tend to become slipshod in adherence, missing out steps or developing their
own fix to events without referring back up for review.

Below are just three incidents showing events at the three stages of the bathtub curve.
We include them to make people aware of some incidents that have happened – after all,
it is better to learn from others than suffer from our own mistakes.

FIRST PHASE – PLUTO

(PIPELINES UNDER THE OCEAN)

 Lead-lined pipeline
 Second world war
 Pressurised laying
 Hydrostatic collapse
 Low through-put
 Increased pumping pressure (re-inflated line)

The PLUTO lines were laid during the second world war to supply petrol (gasoline) to
the allied forces for the re-occupation of France. Two types of different cross-sections
were used: codenamed the HAMEL and HAIS.

The former was rigid steel pipe, laid using floating drums, whilst the latter was the first
use of pipeline reel lay vessels. HAIS line was based on telegraph cable technology but
with the inner core replaced by continuously-cast lead pipe. Trials showed that it needed
to be laid whilst pressurised to offset hydrostatic collapse.

The final specification of the HAIS pipeline shown above was for a flexible pipe
comprising an inner lead pipe of 76 mm (3in) diameter, two layers of prepared paper
tape, 1 layer of bitumen prepared cotton tape, 4 layers of mild steel tape, jute bedding,
steel armour wires and an outermost layer of jute servings.

However, even with internal pressure, the external hydrostatic force caused these to
flatten slightly during installation. This may have been due to inadequate control of
tension, resulting in greater bending at touch-down compared with what could be
achieved today.

Because the through-put was not initially what was expected, the operators increased the
pressure trying to overcome what was thought to be more line friction than had been
estimated. However, this pressure increase fortuitously had the effect of ‘re-inflating’ the
lines over time and so fuel supplies were restored.
500 Fundamentals of pipeline engineering

MIDDLE PHASE –
HURRICANE GEORGES 1998

 Chevron pipeline
 Water depth 33.5 m (110 ft)
 Mudslide initiated by
Hurricane Georges
 Pipeline parted 6.1 m (20 ft)
below mudline
Chandeleur Islands’ lighthouse
 Restart procedure
not followed faithfully
 Leak not identified
 MMS investigation
 OCS report 99-0053

The pictures show the Chandeleur Islands’ lighthouse before and after Hurricane
Georges hit at the end of September 1998. Substantial modification to the seafloor
sediments swept away the barrier islands.

A crude oil pipeline operated by the Chevron Pipe Line Company in the South Pass area,
Block 38 was hit by a mudslide set off by the same event.

It is normal in extraordinary events such as earthquakes, floods and high winds to close
down all operations. A previously planned restart incorporating inspections and tests is
then followed.

However, this procedure was not followed by the operators and they failed to detect that
the line had parted having been covered by 6 m of heavy mud. Some 1306 m³ (8212
bbl) leaked out during restart.

The Minerals Management Service (MMS) investigation concluded that the cause was a
combination of the hurricane and mudslide followed by human error. “The damage to
the pipeline occurred as a result of a natural hazard, specifically, a mudslide that was
precipitated by Hurricane Georges in the latter part of September 1998. The pipeline
was found completely parted 20 feet below the mudline. Deviations from established
other-than-normal startup operating procedures contributed to the failure to identify the
pipeline leak promptly.” Full details are contained in their OCS report 99-0053.

A list of incidents for the USA offshore oil and gas industry can be found at the MMS
website www.mms.gov/incidents. Other countries have similar information such as the
UK at www.hse.gov.uk/offshore/index.htm. In Australia, the www.worksafe.vic.gov.au
website is a good source of general construction operational health and safety (OH&S)
incidents and information. By signing up to their ‘safety soapbox’, they email a weekly
report on safety issues worldwide.

Third party causes include many associated with shrimping nets snagging valves causing
leaks, over-dredging of navigation channels, and jackup legs being dropped onto
pipelines.
Introduction to integrity 501

FINAL PHASE –
GUANABARA BAY PIPELINE LEAK

 Increased throughput
 Operations over-pressured line
 Major environmental damage
 New zig-zag for expansion
 Float and lowered into trench
Satellite image by
Canadian Centre
 Details for Remote Sensing

 11 km (6.8 miles) long

 457.2 mm (18 in) diameter
 52 mm (2 in) concrete

A pipeline leads from Campos Elisios at the north of Guanabara Bay near Rio de Janeiro
to the Ilha do Governador (Governor’s Island), on which the international airport is
sited.

The pipeline was not supplying the required throughput so the operators increased the
pressure causing a rupture near the water’s edge. Guanabara Bay is an enclosed shallow
water bay and the spill caused major environmental damage and loss of public support
for the operators.

The pressure had exceeded the capacity of the line towards the end of its life. The
report identified thermal and pressure cycling (ratchetting) resulting in loss of steel
strength capacity. This was combined with loss of wall thickness, towards the end of its
life.

The pipeline was decommissioned and a new one installed by the SuperPesa Group
(www.superpesa.com.br).

This was floated into position using pontoons and lowered into a pre-prepared trench.
However, the problem of expansion was overcome by forming induction bends in each
pipe-length prior to welding. Expansion buckling problems are thus avoided using this
zig-zag form – each pipe-length has a preferred point of bending avoiding a build-up of
moment at a single point along the pipeline.

The photograph shows the line being installed over the laybarge stinger with rectangular
pontoon floats.
502 Fundamentals of pipeline engineering

CAUSES OF FAILURE

 Failures result from mistakes

 Lack of knowledge transfer
 From designers to installers to operators
 Deliberate risk taking
 Cost savings
 Speed things up
 Lack of maintenance
 Ignoring warnings or procedures
 Combination of a number of minor incidents
 Warning signs not heeded
 Ignored or lack of full understanding

It is usual that failures are caused by mistakes, and in the vast majority of instances, they
can be prevented.

In a lot of cases, the original intent has not been passed on completely to others in the
team. With large undertakings and teams, it is not always possible for everyone to be
fully aware of all the hazards.

Other failures are caused by people making a conscious decision to save money or time.
Maintenance is omitted or let slide, operators or third-parties (shrimpers, dredgers,
jackup operators) deliberately ignore warnings or fail to carry out procedures in full.

In some instances, the failure is caused by a number of smaller incidents which in

themselves do not result in failure (leading people to become complacent) but when
occurring together result in disaster. Perhaps there are unexpected precursor signs,
which require investigation. However, the potential consequences are not fully realised,
or these warnings are simply ignored.
Introduction to integrity 503

PREVENTION

 Verification of calculations and procedures

 Validation of software
 In-house and external reviews
 Full traceability
 Design intent and purchasing
 Inspection of parts
 Quality control
 Training
 Maintenance of equipment
 Testing
 Protective measures
 Hard (rock dump or covers) or soft (inform)

What can we do about it?

We need to ensure that all calculations and procedures are checked. This may be
internally and where appropriate external review by an independent consultant.

Computer software is becoming easier to use these days although we often have a
plethora of rarely used features (bloatware). However, this has two consequences:
engineers often become overconfident in its use and it is often applied to the wrong
problem. It is good practice when conducting an independent check to use a different
package (that you are familiar with).

Every decision on the design intent and purchasing should be traceable. A similar
approach should be made during construction where full records should be available of
linepipe, welding equipment, welders and consumables used at each butt. Full quality
control and testing (where appropriate) is required of every item on the job.

It is essential that everyone should be aware of their sphere of work. This applies from
designers through to operators. During the pipeline life, a maintenance regime should
be adhered to, with appropriate testing of gauges and equipment.

Finally, it is important that where we cannot control events, we provide protection. This
may be in the form of rock dump though shipping channels, dropped object or
overtrawlable covers to pipeline and valvework, or it might be providing information to
fishermen or shipping in the vicinity of the pipeline.
504 Fundamentals of pipeline engineering

COST OF QUALITY

Cost of failure (direct repair,

financial penalties,
environmental damage, delays
Monetary cost $ to schedule and loss of image –
difficult to accurately assess)

Total costs

Cost of inspection,
testing and evaluation

0% Level of quality 100%

The above curve shows how as quality assurance (QA) improves the cost rises steeply as
the curve approaches 100% quality. QA costs include prevention of failures by
inspection and testing of material and equipment used. But it is also important to keep
to the supply programme time restraints to avoid equipment hire overruns or (in the last
instance) client-imposed financial penalties.

If failure does occur, then fines may be imposed by government bodies. The company
image will suffer a loss, which is difficult to quantify financially apart from a rise in
insurance levels but it will have an effect on both future clients and the public. Shell’s
Brent Spar was a notorious example of when a poor public perception caused significant
financial losses to a company.

Although shown as a well-defined line on the above graph, it is difficult to fully assess
exact failure costs. For this reason, it is shown as a dotted line. This is in contrast to
QA and HSE costs, which can be reasonably accurately determined. Nevertheless, by
summing the two, we can get an appreciation of the total quality costs.

By aiming for the minimum point, we can optimise the benefit to the company and
client in terms of profits and customer satisfaction.
Introduction to integrity 505

PIG TRAP INCIDENT – VIDEO

An example of how a number of apparently insignificant and minor changes to

operational procedures can result in a failure is given by the above training video.

The main problems are listed below:

■ Selection of a vertical loading arrangement at the design stage. This caused impact
damage to the release valve and allowed corrosion of the door.
■ Poor choice of material for the trap, based on a cost saving. It was to be always
exposed to the severely corrosive marine spray conditions on the open upper deck
of the platform.
■ Operations placing seawater into the trap to soften the impact of the sphere on
loading. This increased the corrosion problems.
■ The use of a grease to reduce corrosion and erosion at the door seal. This
reduced the friction available to keep the door closed.
■ Lack of appreciation that the door was not staying shut and the belief that such
doors normally need a bit of back pressure to operate.
■ Replacement of the O-ring with a larger one. This solved the immediate problem
but the change was not fed back to the change committee for review.

It was fortunate that no ignition source caused a tragedy. Correct selection of electrical
and other equipment was the final back-up to avoid the fireball.
506 Fundamentals of pipeline engineering

FAILURES – SUMMARY

 Bathtub curve
 Example incidents at three phases of life
 Start of operation – design and construction flaws
 Low level risks – often from third parties
 End of life – equipment wear and tear
 Causes and prevention of failure
 Build-up of small changes causing failure

Any questions?

We have identified the different types of failure and the likelihood of them occurring
throughout a pipeline life.

Some examples of events have been given.

We have covered why they occur and some means of prevention.

The video shows how small out-of-compliances or un-documented changes to

procedures can result in failure.
Introduction to integrity 507

PIMS

PIMS

 Pipeline integrity management system

 During operational phase
 Objectives ensuring safe operation
 No accidents
 No harm to people
 No deterioration of environment
 Integrity assurance cycle Plan
Part most often missed –
people measure but do Learn Do
nothing with data
Measure

A PIMS document codifies good pipeline operating practice. It ensures the three
objectives for safe operation to equipment, people and the environment. By following
the four principle steps of the integrity assurance cycle. The final step of which is seen
by some as the most important. This ensures that the loop is closed out in a report
stating how well a particular operation was undertaken and what should be improved
next time. A later slide expands on this cycle.

That is to say, we should not just follow prepared procedures without looking for better,
safer practices and noting where things could have gone wrong. Nevertheless,
modifications to procedures should only be made following set management change
procedures.

It must be remembered that in operation of pipelines – in particular, those for

hydrocarbon transport – there is great potential for breaching the above objectives.
They are dangerous when guidelines are not followed.

It should be noted that the learning operation (between measuring and planning) is the
bit that is most often omitted. Often nothing is done with the measurements after they
508 Fundamentals of pipeline engineering

are completed and recorded. It needs someone to interpret the results and decide
whether this needs to be incorporated into the next plan.

KEY ASPECTS OF INTEGRITY

 Four distinct activities

 Operations and safety systems
 Structural integrity
 Modifications management
 Flow assurance
 Need to be addressed in parallel
 Complementary to each other
 Cannot be addressed in isolation
 PIMS serves to manage risks
 Procedures, operations/contingency plans & reviews
 Records how risks were removed

The four distinct activities listed above must be considered in a comprehensive integrity
management system.

These aspects should be addressed in parallel and are complementary to each other so
must not be treated in isolation.

We will address each on the following slides.

The activities within the PIMS serve the primary purpose of managing risk. The risk
management process and assessments that have been carried out in the development of
the procedures, plans and reviews that form part of the PIMS are recorded and include
clear statements on the assumptions made, level of risk and actions needed to mitigate
the risk. The assessments address both the threats and the consequences.

The risk assessment process includes periodic reviews and updates as the risk profile
changes with time and experience.

Records are kept of the reviews carried out, the actions taken and how the risks have
been mitigated in design, construction and operation.
Introduction to integrity 509

OPERATIONS AND SAFETY

SYSTEMS

 Pipeline operation
 Kept within designed operating envelope
 Safety systems
 Such as gauges, pressure relief and ESD valves
 Inspected, maintained and tested
 Ensure safe operation
 Optimise performance

The operations and safety systems are concerned with ensuring the pipeline is operating
within the designed operating envelope and that safety systems – for example,
emergency shutdown valves (ESDVs) – are inspected maintained and tested to ensure
safe operation and optimum performance.

The photographs show a 323.8 mm (12¾ in) class 900 ESDV used on an Indian gas
project supplied by Hawa Valves (India) pvt ltd (www.hawavalves.com/spv.htm); and a
cutaway and testing of a pressure relief valve.

STRUCTURAL INTEGRITY

 Monitoring, measuring and prediction

 Internal and external condition of pipeline
 Based on actual condition
 Implementation of controls to maintain line
 Pro-active, risk-based approach
 Threats to pipeline
 Low cathodic protection (CP) readings
 Damage to coating
 Dents
 Joint failure
 Internal corrosion – assessment throughout life
510 Fundamentals of pipeline engineering

The structural integrity activity is concerned with the monitoring, measurement or

prediction of internal and external pipeline condition, the assessment of structural
integrity of the pipeline based on its actual condition, and the implementation of
controls to ensure that the structural integrity is maintained.

This area includes the major activities of pipeline inspection and corrosion assessment.
Integrity can best be assured by taking a pro-active, risk-based approach to find and
assess potential threats.

Pipeline inspection is focused on monitoring and control of the external condition of the
pipeline and the associated protections systems such as coatings and cathodic protection
(CP). Potential threats include low CP, damaged coatings, dents and failure of joints.

Corrosion is the most significant internal degradation mechanism for pipelines and needs
to be assessed, monitored and controlled throughout service life.

The following modules will examine how the structural integrity of the pipeline can be
ensured.

MODIFICATION MANAGEMENT

 Changes to the network

 Additional tie-ins
 Planned servicing or replacement of equipment
 Changes to throughput throughout life
 Outwith envelope
 Alteration of temperatures and pressures
 Line contents composition
 Increase in water cut later in life
 Changes to operational procedures
 Change control committee
 Integrated into updated PIMS

Major modifications and rectification work on the pipeline systems can pose a significant
threat to the ongoing integrity of the pipeline systems if not managed and controlled in
the context of the whole system.

These include replacement parts or changes to throughput when these drift out of the
original specification envelope. The reservoir may not behave as originally envisaged
during the design. For example, a higher water cut at the end of life may result in higher
than expected temperatures. These can increase problems with slugging or pipeline
expansion, resulting in the risk of lateral/upheaval buckling of the pipeline or an increase
in forces acting on the riser guides.

When changes are made to the operational procedures, these should be carefully
scrutinised by a change committee to ensure there is no increase in risk.

Modifications and additions to the pipeline system have to be integrated into the PIMS.
Introduction to integrity 511

FLOW ASSURANCE

 Provide optimum operations processing

 Flow velocity
 Pressures
 Temperatures
 Product composition – chemical treatment/additives
 Flow parameters
 Affect corrosion rates and internal forces
 Control wall loss, hydrates and wax formation
 Water cut, gas-oil ratio and slugging flow
 Identify critical conditions – signal flag
 Need for good record keeping
 Ease of access to interrogate historical data

The purpose of flow assurance is to provide the optimum operating process parameters
(flow velocities, pressures, temperatures, product composition). It should also identify
critical process or flow upset conditions that may threaten integrity.

The flow conditions (such as water cut, slugging flow, changes in temperature) have a
significant influence on corrosion rates and forces produced within the pipeline. These
may change over the service life of the pipeline.

An understanding is needed of past and present history and how conditions may change
in the future.

PRISM – SPAN COMPARISON

 2001
 Exposure
 2002
 Short span
 2003
 Longer span
 2004
 Excessive
 2005
 Rock dump
512 Fundamentals of pipeline engineering

An example integrity analysis package is Prism. The above typical screen dump shows
output for an anonymous pipeline currently being managed by Jee Ltd. [Currently, Jee is
the ‘Pipeline Competent Person’ for 1688 km (1050 miles) of North Sea pipelines (78
subsea lines, 14 umbilical and 6 landlines), defining the requirements and frequency of
inspection for the PIMS.]

The slide shows how a pipeline span increased in length through the years 2001 to 2004.
The final picture shows the pipeline having been rock dumped.

The blue line at top of each year shows lengths of exposed pipeline. Red bars just below
show sections that have started to span. By clicking on the grey bars, a pictorial
representation of the side scan image can be brought up. Cyan indicates sections that
have been stabilised using rock dump.

Other information on the plots are the distances along the pipeline (KP or chainages),
red numbers above the grey bars give anomaly report references, the yellow triangle (left
hand side of grey bar in 2004) shows where the anomaly has exceeded the maximum
allowable value, requiring remedial action.

UN-INVESTIGATED RISER COATING

DAMAGE

The picture shows coating loss from a riser at the lower end of the splash zone. The
thick neoprene corrosion coating has been scraped away by a wire. This may have been
as a result of some unidentified construction activity.

However, records did not reveal this damage for a number of years because no
inspections had been undertaken for the riser. The inspection specification had not
included it. However, when the reports were examined, a section had been included that
said: ‘Riser – no defects noted’.
Introduction to integrity 513

PLAN-DO-MEASURE-LEARN CYCLE

 Plan
 Develop operating procedures and inspection plans
 Hazard identification and assessment (HAZID/HAZOP)
 Set reporting criteria and performance standards
 Do
 Operation and inspection examples coming up
 Measure
 Inspect and test equipment biennially
 Valve closure speed and leak (bypass) rate
 Defect analysis and database
 Learn
 Inspection experience and need for remedial works
 Successful operation – refine inspection plan

The plan-do-measure-learn cycle is vital for good integrity management.

Whether we wish to assure flow rates or to undertake regular inspection, servicing,

maintenance, replacement or modification to the systems, we first have to develop the
appropriate procedures and plans. This may include identification and assessment of
hazards and their risks in formal HAZIDS or HAZOPS. All must be recorded as part of
the PIMS for future review.

We must set standards also for what criteria are deemed out of range and how such
conditions should be reported if they arise.

Standards of performance for each item of equipment must be set. These may be how
quickly a valve closes in normal operation and in an emergency; what amount of bypass
flow is deemed to be acceptable, etc.

We will cover many examples of the ‘Do’ phase in later modules.

Each item of plant and every operation (regular or one-off) in the procedures and plans
developed earlier needs to be compared with the performance standards set down.
Perhaps we need to take a valve out every two years for servicing and performance
testing.

Defects to the pipeline system must all be recorded in the database for future reference.
This must provide an easy and rapid means of comparing how these defects have
developed over time.

The learn phase means that the PIMS procedures can and should be kept up to date with
information whether the operation or inspection proved successful or not. It should
also flag-up when remedial action is needed. If whilst undertaking the work, operators
can identify ways of improving inspections or procedures then this phase may indicate
the need for refinements to make the operation safer or more robust.

For example, in the 2006 survey of the spanning pipeline on the previous slide, it will be
possible to see whether the rock dump succeeded in eliminating the scour and exposure
of the pipeline at the ends of the span.
514 Fundamentals of pipeline engineering

Where:
■ HAZID = hazard identification
■ HAZOP = hazard and operating assessment

PERSONNEL

 Suitably qualified and experienced (SQEP)

 Need for training and familiarisation
 Organogram (organisational chart)
 Relationships between members of team
 Integrity and operations support
 Pipeline design, inspection, corrosion, process and
topsides specialists
 Offshore operations
 Superintendent responsible for assets
 Project team
 Developing operations for maintenance or inspection
 Clearly defined responsibilities
 Single point of overall responsibility

Those operating the PIMS should be SQEP personnel with competence in their own
particular field.

Training or familiarisation may be needed when situations change or when new

equipment is added. Regular refresher meetings provide confidence that the whole team
is familiar with current developments and changes in procedures.

To help the PIMS team operate effectively, it is recommended that an organogram

clearly shows who team members are and how they communicate to each other. This is
particularly so for the larger pipeline networks where personnel are frequently replaced
or move to different areas.

It is necessary to identify the different specialists for support, operations and projects –
as shown in the suggested groupings, above.

It is necessary to appoint one person as a single point of overall responsibility. In the

UK, this is the ‘Pipeline Responsible Person’.
Introduction to integrity 515

PIMS – SUMMARY

 Three objectives for safe operation

 No accidents, no injuries, no environmental damage
 Four parallel and complementary activities
 Operations and safety
 Structural integrity
 Modification management
 Flow assurance
 Integrity assurance cycle
 Plan-do-measure-learn
 Organogram
 Clear roles and responsibilities for team
Any questions?

A PIMS is required to ensure nothing goes wrong whilst operating the pipeline network.

We need to balance the four activities which must be run together. This can be done
using the integrity assurance cycle and providing clear responsibilities for each member
of the team.

INTEGRITY MANAGEMENT –
SUMMARY

 Failure frequencies throughout life

 Bathtub curve
 PIMS
 Means to ensure safe operation
 Extend the life of the facility

Any questions?

Different types of failures occur throughout the life of a pipeline. We have seen how the
frequency of failure and the reasons for them can be seen using the bathtub curve.
516 Fundamentals of pipeline engineering

By setting up a PIMS, it is possible to limit incidents thus ensuring safe operation of the
pipeline. The system should also aim to extend the bottom of the bathtub curve for as
many years as possible.
Structural integrity
Structural integrity 519

Structural Integrity

Pipeline Engineering and

Operations

Ralf Peek
Copyright: SIP C
Structural Integrity
16/10/2009

Shell
Product Brand

Example of Reliability-Based
Limit States Design

• Sakhalin Offshore Pipelines Design for Ice Protection.

• Ice Loading Environment and Effects
• Pipeline Strain-Based Design Requirements
• Probabilistic Modeling

“If may will begin with certainties, he shall end in doubts; but if he will be content to
begin with with doubts, he shall end in certainties.” quoted by Ian Jordaan in “Decisions
under Uncertainty” from Francis Beacon, Advancement of Learning.
520 Fundamentals of pipeline engineering

Sakhalin Offshore Pipelines

Ice Protection Requirements

1.E+00

Distance, s (m) 1.E-01 Max

0 50 100 150 200
0
Burial

Prob of Exceeding
1.E-02
5

Pits Req.
10 1.E-03 (145)
25m
Ice Draft, z (m)

15
Gouges
10
1.E-04
(1086)
20
10

1.E-05
25

30 1.E-06
0.0 0.5 1.0 1.5 2.0 2.5
Deepest Recorded Keel [ASL] Depth (m)

Ice is initially formed as a frozen slab of seawater. Much of it is formed in the “ice-
kitchen” north of Sakhalin around the Amur River mouth and beyond. But the seawater
also freezes further South. Movements of the ice sheet create compression zones where
the ice sheet is broken forming pile of rubble as shown top left. To support such a pile
of rubble more or less istostatically the keel of the pile must be much deeper than the
“sail” is high. The deepest keel recorded by an upward looking sonar is shown bottom
left. The seafloor image from multibeam sonar (top right) shows what happens when
such ice keels scour or gouge the seafloor. There is some argument about whether to
call these “scours” (which seems to be the term preferred by Canadians) or “gouges”.
The former term tends to suggest ice weaker than the seafloor which breaks up against
the seafloor nevertheless leaving some marks, whereas “gouges” tends to suggest
stronger ice which is more dangerous. More than 99% of the time “scour” is the more
appropriate term, but it is not these 99% of events that are of concern to the pipeline, it
is the exceptions, and whether they are still aptly described as scours is open to question.
The bottom right slide shows the probability distribution for the depth of gouges and
the pits that sometimes form at the end of gouges.
Where a keel gets stuck it may become more heavily grounded (e.g. with changing tide)
and than cause more ice to break and pile up on top of it, giving rise to a stamukha.
Structural integrity 521

Piltun Pipeline Route

New Piltun (PA-B) platform

Existing Moliqpak (PA-A)

platform

This slide shows the original design Piltun Route. It involves a total of four 14-inch
pipelines. A gas and and oil line till the existing Moliqpak (Piltun A) platform, and
another pair of pipelines from shore to the Pilun B Platform, laid in the same trench as
the Piltun A pipelines until the bifurcation point about 2km from the Moliqpak (Piltun
A) platform.

Since the landfall area appeared to be favoured by grey whales, it became necessary to
reroute the lines starting from further South, then heading offshore to a water depth of
30m or more, where ice keels are not longer expected to reach the seafloor, and then to
the North and returning Eastwards to approach the platforms.

Lunskoye Pipeline Route

522 Fundamentals of pipeline engineering

The Lunskoye route is not only straighter but also more straight forward in that it gets
deeper faster, there are fewer dunes and changing seafloor features, and no re-routing
due to grey whales was necessary.
Anectotes emerged that the whales, far from being frightened by the offshore operations
and the subsea noise they generate, appeared to be curious instead. Pieter Swart can no
doubt provide further details.

Design Considerations

• Ice forces are expected to be larger than the pipeline can

withstand -> bury the line
• Even if the pipe is buried below the depth of the deepest
gouge and pit, the ice can produce significant soil movement
below the gouge and pit. The pipeline must be designed to
accommodate such sub-gouge or sub-pit deformations
without failure. I.e. there should be wrinkling due to plastic
deformations or fracture at girth welds
• Collapse e.g. by soil pressure under a Stamukha must also
be considered, but was not found to be critical

Because of the requirements of pipe deformability without wrinkling, a wall thickness

beyond what was required to contain the operating pressure what used, and low yield
strength (X52) line pipe was chosen at an early stage.
The higher wall thickness directly improves the deformability of the pipe without
wrinkling or local buckling. It also helps for the girth welds: perhaps the most
dangerous defects are lack of sidewall fusion defects, which seldom exceed the depth of
1 weld beed (indeed the specification does not allow stacked defects). For a higher wall
thickness, 1 weld beed is a smaller fraction of the wall thickness, and it therefore is easier
to achieve overmatching weld strengths even in the presence of a defect.
Lower X52 grade has the advantage that is is easier to achieve overmatching of the girth
welds, and furthermore in theory a higher yield-to-tensile strength (Y/T) ratio can be
achieved, which is good for the deformability of the pipe without wrinkling.
Structural integrity 523

Basic Approach

• Find burial depth profile B=B(s) such that cost is minimised

subject to:
Burial Depth for Ice Protection
KP

 nX(s) pf(B(s)) ds 
10 -4
Gouge Failure Probability
per gouge crossing
Frequency
(crossings/km-yr) Add
Erosion
Failure Probability For Pipeline Allowance
In lieu of established ice design criteria, it was decided to use this probabilistic approach.
For Sakhalin the failure probability per event pf(.) could reasonably be assumed to be
probabilistically independent of water depth because there was no statistically significant
evidence to the contrary, which implied no basis for reducing the burial depth for
shallower waters. However in the Beufort sea, gouge depths do seems to vary as a
function of water depth. Indeed one should never assume the ice environment in one
part of the world is the same as in another.

Outline

• Gouge Frequency
• Failure Probability Per Event
– Gouge/Pit Depth
• Ice/soil strength model
• adjustment for infill

– Pipe Response to subgouge deformations

• Ice Protection Requirements

• Erosion Allowance / Overall Burial
• Learning Lessons
524 Fundamentals of pipeline engineering

Much of the presentation will focus on ice rather than pipelines, because this is where
the greatest uncertainties were encountered, and these needed to be addressed for the
project.

Gouge Frequency

• Prediction Based on:

– Keel Drafts (from Upward-Looking Sonar)
– Ice Drift Velocities (from Accoustic Doppler Profilers, ACDP)
– Ice Concentrations
– Floe Size Statistics
– Bathymetry
– A number of Assumptions

• Seafloor Record (for verification)

The gouge frequency is the average number of times per kilometre per year a gouge
crosses the pipeline. Perhaps this would be more appropriately called “scour frequency”
because it includes all scours, not just the dangerous ones.
We used 2 different methods to estimate the frequency, one based on counting scours
on the seafloor, and one base on the upward looking sonar data and expected keel drift
patterns.
There is a lot of movement of ice around Sakhalin, and also a lot of near seafloor current
and wave action that tends to obliterate scours on the sandy seafloor quite quickly. This
means that relying only on the record of scours observed on the seafloor could lead to
underestimation of the scour frequency.
Structural integrity 525

Approximation: Rise – Up not dependent on slope

2.5
From Exxon Chayvo 1997 Data
Rise-up Data provided in email by [email protected], 8 Oct 2003.

2 Overall Average Rise-Up

E[ R ] = 0.22m
Rise-Up, R (m)

1.5

Average Rise-Up per

1 slope bin.

0.5

E[ R | s ] = 0.13 (s + 0.1)^0.56

0
0 5 10 15 20 25
Slope, s (m/km)

We had been much less successful than Exxon in performing good survey of the
seafloor (which Exxon achieved in part by helicopter lifting a survey vessel to survey
between ice floes, while the gouges were still a bit fresh), so we decided to exchange the
Exxon gouge survey data at the time form the data on ice loads on the Moliqpak
platform, which is collected by instrumented plates.
The plot above is largely based on the Exxon Chayvo data. “Rise-up” is defined in as
the difference in water depth from where the keel first grounds and the scour starts to
where the scour ends. The rise up data were collected by Tony King and Richard
McKenna, then both with C-Core.
We considered two possible models: one in which the scour length is probabilistically
independent of the slope and one in which the rise-up is probabilistically independent of
slope.
If the length is probabilistically independent of slope, we’d expect average rise-up to
increase proportionally with the slope. On the other hand if the rise up is
probabilistically independent of slope, the average rise up would be constant.
The data above suggests that the reality falls somewhere in between: the rise-up does
increase with slope but not quite proportionally. Nevertheless, after trying both
methods, we decided to use the approach based on rise-up, which implies that the gouge
frequency does not depend on the magnitude of the slope, and only on the orientation
of the slope.
526 Fundamentals of pipeline engineering

Keel Draft Distribution

1.E+03

1.E+02

1.E+01

1.E+00
Number of Features per km^2 of Ice Floe
1.E-01

1.E-02

1.E-03

1.E-04

1.E-05

1.E-06
Keels with Draft > z (Data)
1.E-07

1.E-08 Keels with Draft > z (Weibull Fit)

1.E-09 Floes with Draft > z (Calculated from Weibull Fit)

1.E-10

1.E-11

1.E-12
0 5 10 15 20 25 30 35 40
Keel Draft, z (m)

To establish the keel draft distribution from upward-looking sonar (ULS) data, we
developed the methods of slices, which I believe to be superior to other methods used
because it accounts for the fact that the ULS does not necessarily “see” the deepest part
of the keel as it drifts by.
The above plot shows the density of occurrence of ice keels exceeding a given depth per
square kilometre of ice floe. The method if slices enables this to be estimated statistically
from the one-dimensional data collected by the ULS.
There were several ULS instruments were deployed over several years resulting in a total
of 18 location-years. The total linear length of ice floe covered by records is 45,000km.
It only includes the records from the vicinity of the Piltun and Lunskoye platform
locations, and only observation sites that are at a water depth of 30m or more.
The deepest keels was observed at Lunskoye although it is slightly more to the South. It
was considered that the distribution of keel drafts would not be significantly different
from Piltun and Lunskoye, which enabled the data to be pooled thereby making more
data available to define the above (Weibull) distribution. Nevertheless there are still
insufficient data to define the all important tail of the probability distribution, and no
guarantee that the extension of the Weibull curve is correct.
The draft distribution is especially important in determining the water depth to which
the pipeline needs to be buried.
Structural integrity 527

Predicted Gouge Frequency

1.E+03

• Applies for E-W 1.E+02

oriented pipeline at 1.E+01

Piltun
Lunskoye
a fully exposed 1.E+00
shoal far from

Crossing Frequency
1.E-01
shore
1.E-02

1.E-03

1.E-04

1.E-05

1.E-06

1.E-07
0 5 10 15 20 25 30 35 40
Water Depth (m)

Here the “crossing frequency” is the gouge frequency mentioned earlier. It is the
number of gouges crossing the pipeline per kilometre per year.
The predominant ice drift direction is towards the South. Therefore the East-West
orientation assumed here results in the highest frequency. The frequency is also
dependent on the orientation of the slope, with the highest frequency occurring where
the seafloor rises in the direction of predominant ice drift.

Predicted Gouge Frequency p( v ) E[ v | v ]

0
Piltun
0.14

Correction Factors
337.5 22.5
0.12 Lunskoye
315 0.1 45

0.08

292.5 0.06
67.5
0.04
0.02
1. Bathymetric Shielding 270 0 90

2. Distance from Shore (Landfast Ice, Open Lead) 247.5 112.5

3. Pipeline Orientation 225 135

202.5 157.5
180

1.2
1.0
Piltun near
0.9 Piltun far
1.0 Lunskoye near
0.8 Piltun
Lunskoye far
Lunskoye
0.7 0.8
Correctoion Factor
Correction Factor

0.6

0.5 0.6

0.4
0.4
0.3

0.2
0.2

0.1

0.0 0.0
0 5 10 15 20 0 30 60 90 120 150 180
Distance from Shore (km) Direction of Pipeline (deg.)

The gouge frequency depends not only on water depth, but also on the correction
factors shown on this slide.
528 Fundamentals of pipeline engineering

Bathymetric shielding has 2 components: (a) the orientation of the slope at the location
of the pipeline, considering only uphill gouging, and (b) shielding from nearby dunes that
could exclude gouging from certain directions.
The distance from shore is important because nearshore the one runs into landfast ice
which does not move and therefore also does not gouge.
Finally the pipeline orientation has an effect (bottom right graph) since the predominant
flux of ice is towards the South (top right graph).

Frequency Predictions: Comparison with Seafloor

Record:
Scour Rise-Up Model vs. Survey
Water Depth CSR 1997 Survey Sites CSR 2003 Survey Sites model /
Group (m) Data Ar-Da Ch-B Ch-R CHY LUC LUL PIL REG Grand Total survey
0 - 10 Sum of N_model (1/yr) 15.532 15.532 25%
Sum of N_survey (1/yr) 26.000 12.000 3.000 22.000 63.000
11, 12 Sum of N_model (1/yr) 7.621 29.282 98.593 63.338 198.834 92%
Sum of N_survey (1/yr) 4.000 35.000 41.000 15.000 1.000 109.000 11.000 216.000
13, 14 Sum of N_model (1/yr) 5.435 43.486 5.150 312.908 1.667 122.915 63.529 555.091 102%
Sum of N_survey (1/yr) 2.000 323.000 0.000 49.000 1.000 3.000 140.000 27.000 545.000
15, 16 Sum of N_model (1/yr) 6.018 94.049 11.830 107.546 57.928 42.430 6.839 326.640 123%
Sum of N_survey (1/yr) 3.000 163.000 2.000 21.000 4.000 32.000 41.000 266.000
17, 18 Sum of N_model (1/yr) 5.405 89.177 80.979 22.896 100.215 10.368 309.040 132%
Sum of N_survey (1/yr) 1.000 25.000 0.000 1.000 158.000 50.000 235.000
19, 20 Sum of N_model (1/yr) 3.439 31.812 13.058 4.305 41.068 6.041 99.723 92%
Sum of N_survey (1/yr) 1.000 4.000 1.000 0.000 75.000 27.000 108.000
21, 22 Sum of N_model (1/yr) 1.995 3.191 0.098 0.690 14.133 0.909 21.016 111%
Sum of N_survey (1/yr) 1.000 1.000 0.000 0.000 5.000 12.000 19.000
23, 24 Sum of N_model (1/yr) 0.181 0.265 0.008 0.097 3.030 0.100 3.681 61%
Sum of N_survey (1/yr) 0.000 2.000 0.000 1.000 1.000 2.000 6.000
25, 26 Sum of N_model (1/yr) 0.033 0.026 0.046 0.135 0.175 0.032 0.446 45%
Sum of N_survey (1/yr) 0.000 0.000 0.000 0.000 1.000 0.000 1.000
27, 28 Sum of N_model (1/yr) 0.002 0.000 0.015 0.007 0.008 0.032
Sum of N_survey (1/yr) 0.000 0.000 0.000 0.000 0.000 0.000
29, 30 Sum of N_model (1/yr) 0.000 0.003 0.027 0.001 0.031
Sum of N_survey (1/yr) 0.000 0.000 0.000 0.000 0.000
31, 32 Sum of N_model (1/yr) 0.000 0.008 0.000 0.009
Sum of N_survey (1/yr) 0.000 0.000 0.000 0.000
33, 34 Sum of N_model (1/yr) 0.000 0.000 0.000
Sum of N_survey (1/yr) 0.000 0.000 0.000
35 - over Sum of N_model (1/yr) 0.000 0.000
Sum of N_survey (1/yr) 0.000 0.000
Total Sum of N_model (1/yr) 45.628 262.014 46.262 514.622 1.713 307.576 327.960 24.299 1530.075 105%
Total Sum of N_survey (1/yr) 38.000 553.000 55.000 89.000 1.000 10.000 543.000 170.000 1459.000
model / survey = 120% 47% 84% 578% 171% 3076% 60% 14% 105%

These results show that the gouge frequency figured from the ULS data are more or less
in agreement with that from counting the gouges on the seafloor.
Despite significant attention paid to the frequency issue, it is not actually the most
important factor controlling the burial depth. The depth of gouging when it occurs is
much more important.
In the statistical analysis of data we often assume that the observed results are the
outcomes of probabilistically independent trials. Of course we know this is not the case.
There may be year-to-year variations, and the effect of global warming may play a role.
Nevertheless, without enough information to quantify such trends a model in which we
assume probabilistic independence is still reasonable for estimating failure probabilities.
However when used statistically the same assumption might suggest we have more
confidence in our statistics from the data than we should have. For instance an average
might vary much more than we’d expect by assuming that every value is probabilistically
independent of the next.
Structural integrity 529

Outline 1.E+00

1.E-01

Prob of Exceeding
1.E-02
Pits
1.E-03 (145)
• Gouge Frequency Gouges
1.E-04
(1086)
• Failure Probability Per Event 1.E-05

– Gouge/Pit Depth 1.E-06

0.0 0.5 1.0 1.5 2.0 2.5
• statistics Depth (m)

• ice/soil strength model

• adjustment for infill

– Pipe Response to subgouge deformations

• Ice Protection Requirements

• Erosion Allowance / Overall Burial
• Learning Lessons

This slide shows where we are in the outline presented earlier.

Since the statistics on measured gouge and pit depths are not sufficient to estimate the
probabilities of truly extreme event, Ken Croasdale and Associates, together with Ken
Been from Golder Associates developed a mechanical model that considers the relative
strengths of soil and ice and determines and upper bound of how deep a gouge could go.
Of course the upper bound also is uncertain because of uncertain inputs to the bounding
model, and modeling uncertainty (i.e. that the model may not give the exact result even if
all input parameters were known exactly).

Soil Resistance to Gouging – Existing Model


530 Fundamentals of pipeline engineering

The soil resistance model is based on slip line theory, which models the drained
behaviour of sand. Our tests later showed that dense sands develop much larger
resistance due to gouging because the behaviour is closer to untrained than drained. I.e.
pore water suction generated cannot be dissipated fast enough and causes the sand to
lock up. We believe this the be the main reason why we never saw deep gouges in
dense sands.

Ice Failure Mode:

A bearing failure mode of the ice was considered in which the pressure that causes
failure of the ice is about 5 times the cohesion of the ice.
Structural integrity 531

Ice Strength Test

This shows one of the tests performed to establish the strength the ice rubble that forms
the keels. The consolidated layer near the top is much stronger, and this is cut through,
so that the load applied with jacks goes to the lower ice rubble. To interpret the test
results in terms of an ice strength parameter, Ken Croasdale & Associates considered
two different assumptions:
a) that the ice behaves like a purely cohesive soil (i.e. capable of plastic deformation and
with a shear strength that is independent of the normal pressure on the shearing surface),
and
b) that the ice behaves as a cohesionless material in which the shear strength is
proportional to the normal pressure on the surface on which shearing occurs.
The first approximation gave a better fit to the data and was therefore chosen.
532 Fundamentals of pipeline engineering

Sliding Shear Test

• Determine ice strength

(Performed near Mackinaw
Bridge Michigan?)
• Diagrams from report by Ken
Croasdale & Associates
• Work was done as part of a JIP
effort, with some additional work
to support Sakhalin 2

This slide shows another type of test to establish ice strength. No all test data used
came from Sakhalin, but some did.

Probability Distribution Ice Cohesion

1.E+00

1.E-01 Strength Tests (43)

Weibull Distribution used in [JIPa]

Simulations
Probability of Exceeding

1.E-02

1.E-03   c  30  10.673 
Pe  1  exp    
  39  
 
1.E-04 Strength Test of Grounded
ice (c=55kPa) used as upper
bound to truncate the distribution
1.E-05 used in the simulations of [JIPa].

1.E-06
0 20 40 60 80 100 120
Ice Cohesion c (kPa)

This shows the probability distribution Ice cohesion inferred from the data by Ken
Croasdale and Associates. It also requires extrapolation from the data, that was guided
by a strength measurement from a grounded ice keel.
These strengths are much lower than those used to establish the loads acting on
platforms, because platforms can be affected by the much stronger consolidated layer.
The well-established size effect for platform loading in which the ice pressures that can
develop over a small area are larger than those that can be sustained over a larger area
Structural integrity 533

where not included in the keel strength models. This was regarded a conservative
approximation since the loaded area in the tests is small compared to the loaded area for
a gouge.

Scouring Simulations Results (George Comfort, Fleet

Technology, Brian Wright)

1.E+00
All Scours (1086)

1.E-01
All Pits (145)

1.E-02 JIP Simulations (1,000,000)

Prob of Exceeding

1.E-03

1.E-04

1.E-05

1.E-06

1.E-07
0.00 0.50 1.00 1.50 2.00 2.50
Depth (m)

The probability distribution for the upper bound gouge depth from the mechanical
model (dashed red line) is compared with the statistical data for gouges and pits.
To some extend the mechanical model is more representative of pits because it does not
include pore water suction effects, and in pits there is indeed enough time for the pore
water to dissipate, but the soil parameters were perhaps chosen on the high side (perhaps
in an attempt to get realistic depths for gouges). Further the extreme ice strengths for
the bottom of keels may be underestimated by the probability distribution used in this
model.
These are some of the factors considered in coming up with design probability
distributions for gouge and pit depths.
534 Fundamentals of pipeline engineering

Design Distributions for Gouge & Pit Depths

1.E+00
All Scours (1086)

All Pits (145)

1.E-01
JIP Simulations (1,000,000)

Design Distribution, Scours

1.E-02 Design Distribution, Pits
Prob of Exceeding

1.E-03

1.E-04

1.E-05

1.E-06

1.E-07
0.00 0.50 1.00 1.50 2.00 2.50
Depth (m)

In theory it is possible to use Bayesian statistical methods to combine estimated

probability distributions from different sources of information, but it requires viewing
your particular problem as embedded in a larger situation which also needs to be
characterized in a probability model. This is not an easy concept, and also requires
statements/assumptions to be made about the larger situation that many prefer to avoid.
Here no attempt was made on any Bayesian combination. Instead I simply estimated the
design curves considering how the real situation might vary from all assumptions made
in deriving the curves shown.
Special attention is required in taking the expected value of an uncertainty probability on
a logarithmic scale.
Contradictory to observations in the Beaufort sea that indicate deeper gouges in deeper
water, here the same gouge depth distribution is used for all water depths. In part this
was because of insufficient data to establish a statically significant difference as a
function of water depth.
Also there is the danger that late in the season, when waves are no longer damped by ice
you might get less deep keel grounding on top of a pipeline and wallowing in the waves,
so it seems sensible not to reduce the burial depth in shallower water based on the
possibility of lower gouge depths there, on average.
Structural integrity 535

Pit & Gouge Depth Distribution Combined

(I.e. depth distribution for an ice loading event)

1.E+00

Failure Probability for Buried Line due to Ice-Pipe Contact

1.E-01

1.E-02 Contact due to Gouge

Failure Probability per Event

Contact due to Pit

1.E-03 Contact due to Gouge or Pit

1.E-04

1.E-05

1.E-06

1.E-07

1.E-08
0 0.5 1 1.5 2 2.5 3 3.5
Burial Depth

Having a pit on top of the pipeline is about 1% as likely as having a gouge crossing the
pipeline. Combining the probability distributions for depth of pits and gouges on this
basis gives the above distribution for the depth of an ice loading event affecting the
pipeline.

Gouging Tests in Delft Hydraulics Dredging Flume

(Walter van Kesteren, Yves Friedman)

The pipeline should not only be protected from direct contact with the ice, but also we
needed to make sure that the soil deformations below the gouge would not damage the
pipeline.
536 Fundamentals of pipeline engineering

For this purpose existing gouge tests performed in the centrifuge were supplemented by
a larger (but not full) scale test performed at Delft Hydraulics (now Deltares).
The lower picture shows the “ice keel”, and the upper the dredging flume in which the
tests were performed.
We expected to seem some pore water effects but it was much more than expected, and
although we reduced the target depths a lot the loads were still so high that several of the
thick glass panels shown in the picture on the right were broken.
To complement these large scale tests, a large number of centrifuge tests were also
performed by Hederikus Allersma for the Technical University at Delft.

Subgouge Deformations
Normalised Displacement, Ur/(D+W/6)
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20
0
N orm alised Depth below Scour, (z-D)/(D+W /6)

Relative Density 50% or More.

Displacement is the resultant of Vert. & Hor. Comps
1

D
TUD 1: D=1.50m, 136g, W/D=10, 15deg, Dr=67% TUD 2: D=0.95m, 136g, W/D=16, 15deg, Dr=68%
TUD 3: D=0.54m, 136g, W/D=28, 15deg, Dr=68% TUD 4: D=0.75m, 68g, W/D=10, 15deg, Dr=68%
TUD 5: D=0.48m, 68g, W/D=16, 15deg, Dr=68% TUD 6: D=0.27m, 68g, W/D=28, 15deg, Dr=68%
TUD 7: D=1.50m, 136g, W/D=10, 30deg, Dr=65% TUD 8: D=0.95m, 136g, W/D=16, 30deg, Dr=65%
TUD 9: D=0.54m, 136g, W/D=28, 30deg, Dr=62% TUD 10: D=2.66m, 133g, W/D=3.8, 15deg, Dr=78%
TUD 11: D=2.56m, 256g, W/D=11, 15deg, Dr=65% TUD 13: D=0.95m, 136g, W/D=16, 15deg, Dr=60%, slow
2 TUD 14: D=0.95m, 136g, W/D=16, 15deg, Dr=50%, fast TUD 16: D=1.50m, 136g, W/D= 7, 15deg, Dr 81%
PRSA01: D=4.50m, 150g, W/D=3.3, 30deg, Dr=81% PRSA02: D=2.5m, 75g, W/D= 3.0, 15deg, Dr=90%
PRSA03: D=2.10m, 150g, W/D=7.1, 15deg, Dr=55% RRSA04: D=2.65m, 150g, W/D=5.6, 15deg, Dr=82%
PRSA05: D=2.50m, 150g, W/D=6.0, 15deg, Dr=70% DH12: D=0.19m, 1g, W/D=11.6, 15deg, Dr=60%
DH13: D=0.14m, 1g, W/D=15.7, 15deg, Dr=60% DH14: D=0.14m, 1g, W/D=15.7, 15deg, Dr=65%
DH15: D=0.14m, 1g, W/D=15.7, 15deg, Dr=65% DH17: D=0.09m, 1g, W/D=24.4, 15deg, Dr=65%
DH21: D=0.19m, 1g, W/D=11.6, 30deg, Dr=55% DH22: D=0.19m, 1g, W/D=11.6, 30deg, Dr=55%
DH23: D=0.19m, 1g, W/D=11.6, 30deg, Dr=65% DH24: D=0.11m, 1g, W/D=20, 30deg, Dr=70%
DH31: D=0.19m, 1g, W/D=7.9, 15deg, Dr=65% DH32: D=0.19m, 1g, W/D=7.9, 15deg, Dr=65%
DH33: D=0.14m, 1g, W/D=10.7, 15deg, Dr=65% DH35: D=0.09, 1g, W/D=16.7, 15deg, Dr=70%
DH36: D=0.09, 1g, W/D=16.7, 15deg, Dr=70% DH41: D=0.19m, 1g, W/D=7.9, 45deg, Dr=60%
DH42: D=0.15, 1g, W/D=10.7, 45deg, Dr=60% DH43: D=0.14, 1g, W/D=10.7, 45deg, Dr=60%
DH44: D=0.14m, 1g, W/D=10.7, 45deg, Dr=60% Design Curve
3

Despite all the tests for subouge deformations no very clear picture emerged, and we
decided in end to carry this uncertainty along on the assessment as follows:
a) construct a probabilistic subgouge deformation model that acknowledges uncertainty
in the subgouge deformations, using engineering judgment
b) considering the distribution of gouge depth and width and combining this with the
distribution of subgouge deformations obtain a non-conditional distribution for the
subgouge deformation, i.e. one that is applicable for a randomly selected gouge, rather
than a gouge of given depth and width.
c) find a deterministic effective design curve for subgouge deformation that gives more
or less the same non-conditional distribution for subgouge deformations.
That effective design curve is shown in black above.
There remain disagreements within the industry of how best to handle this issue, but we
needed a way in which we could acknowledge the uncertainty and move forward in the
design accounting for the uncertainty.
Structural integrity 537

Pipe Response Model (C-Core)

Pipe (Elasto-Plastic Beam

model)

Nonlinear Soil Springs

Imposed Sub-Gouge Displacement

By definition the subgouge displacements are the displacements that would occur under
the gouge if the pipe were not present. Since the pipe has considerable stiffness, it will
not completely follow the subgouge deformations. To account for this the subgouge
deformations are imposed on one side of a system of springs that have the pipe attached
to it on the other side.

These analysis approach has been used for some time for fault crossings, in which case
the shearing displacements of the fault rather than the subgouge deformations are
imposed on the soil springs. Guidance for choosing the soil springs properties is then
also available from the fault crossing work in the form of an ASCE publication first
issued around 1984 and updated around 2000.
The type of model has become known as the “uncoupled model” because any soil spring
is not influenced by what is happening to the spring next to it, whereas in real soil such
coupling does exist.

For ice gouging problems the crimes of this model are even more severe because in
essence the soil strength is assumed to exist twice: once resisting the ice gouging forces,
and a 2nd time resisting the displacements the pipe would like to undergo with respect to
the soil because of its flexural stiffness. The actual soil only has its strength once, and
this appears to be the reason why fully coupled models predict lower strains in the
pipeline.
538 Fundamentals of pipeline engineering

Strain Concentrations due to Concrete Coating

3.00

2.80 A
B
2.60 C
Strain Concentration Factor

2.40 D
E
2.20 F
G
2.00

1.80

1.60

1.40

1.20
L = 6.1 m
1.00
0.0% 0.5% 1.0% 1.5% 2.0% 2.5% 3.0%
Global Nominal Bending Strain

When applying strain-based design all possible sources of strain concentration should be
considered. This slide shows the strain concentrations due to concrete coating, which is
not continuous over field joints.

The model used for concrete coating is described in EP 2006-5161. It accounts for
cracking of the concrete in tension, but also for disbonding/slip of the concrete with
respect to the steel pipe according to a model that has been developed by Statoil
engineers and full-scale tested.

Pipe Response Function (Look-up Table, C-CORE)

 = f( D, W, B)
3.0%

Gouge Widths:

2.5% 1
2
5
8
2.0%
Max Tensile Strain

12
16
32
1.5%
64

1.0%

0.5%

0.0%
0 0.5 1 1.5 2 2.5 3 3.5
Clearance, B-D (m)
Structural integrity 539

The uncoupled analysis is done for all relevant values of

D = gouge depth
W = gouge width, and
B = burial depth
A sample of results is plotted above.
A computer program is developed that can interpolate between the results calculated by
finite element analysis (FEA). Thus the value of the applied strain can be rapidly
calculated as part of of Monte Carlo simulation procedure for any values drawn for the
gouge depth and width according to their probability distribution.
It remains to establish a probability distribution for the amount of strain the pipe can
tolerate without wrinkling or fracture at a girth weld.

Monte-Carlo Simulation Results for Pipe Response to

Ice Loading Events of Random Depth and Width
Gives Failure Probability per Event,
pf(B)
1.E+00

1.E-01 Contact
0.5%
1.E-02 1.5%
Failure Probability per Even

2.5%

1.E-03

1.E-04

1.E-05

1.E-06

1.E-07

1.E-08
2.23
0 0.5 1 1.5 2 2.5 3 3.5
Burial Depth

In this plot “failure” is defined as either ice-pipe contact, which is assume to occur if the
gouge or pit is deeper than the burial depth, or the strain in the pipeline reaching a
specified level.
It shows that subgouge deformations are not governing of the pipe has 2.5% strain
capacity or more.
540 Fundamentals of pipeline engineering

Basic Approach

• Find burial depth profile B=B(s) such that cost is minimised

subject to:
Burial Depth for Ice Protection
KP

 nX(s) pf(B(s)) ds 
10 -4
Gouge Failure Probability
per gouge crossing
Frequency
(crossings/km-yr) Add
Erosion
Failure Probability For Pipeline Allowance
Now all in the above equation has been defined. It only remains to for the optimal
burial profile that satisfies the above requirement. A Lagarnge multiplier approach was
used to solve the above constrained optimization problem. For a given value of the
Lagrange multiplier we could calculate the burial depth everywhere, but we then had to
adjust the multiplier (using goal seek in ms excel) until the proper failure probability was
obtained. For simplicity the optimization criterion was based on dredging volume not
considering the MDB.
Initially a uniform risk allocation approach was used, which is based on equal failure
probability per unit length along the length of the line. The difference is results from the
two approaches was not great.
Structural integrity 541

Burial Requirements for Ice Protection

5.00 0

4.50 5
ice
4.00 Water Depth 10

Burial D epth, B (m ) 3.50 15

W ater Depth, Z (m )
3.00 20

2.50 25

2.00 30

1.50 35

1.00 40

0.50 45

0.00 50
0 5 10 15 20 25 30 35 40
KP Location, s (km)

The result of solving the constrained optimization problem.

Evidence of Changing Seabed (Need Erosion

Allowance, Allan Terrill)

We enlisted the help of Romke Bijker, who estimated from data such as the above a
“Morphological Design Basis” (MDB). This represents the greatest water depth
expected at any point along the pipeline considering seafloor erosion and deposition.
The ice protection requirement is added to the MDB to define the depth to which the
pipeline needed to be lowered.
542 Fundamentals of pipeline engineering

Resulting Burial Requirements

5.00 0

4.50 total 5
ice
4.00 Water Depth 10

Burial Depth, B (m ) 3.50 15

W ater Depth, Z (m)

3.00 20

2.50 25

2.00 30

1.50 35

1.00 40

0.50 45

0.00 50
0 5 10 15 20 25 30 35 40
KP Location, s (km)

The plot was used to estimate dredging volumes. The actual burial specification was
based on a specified depth with respect to a reference sea level, rather than a depth with
respect to a changing seafloor.

Effect of Clay Seams

Li

D
E
Ice, cohesion ci Hs

F C

A B

Clay Seam, cohesion, cc

Lc

Upon surveying the new whale-friendly route a number of clay seams were found that
could grossly invalidate the assumptions about subgouge deformations. By looking at
the relative strengths of the ice along failure plane ED, and the soils along failure surface
FABC the and considering the probability distributions for ice cohesion and geometric
Structural integrity 543

parameters, the probability of slip along the clay seam was calculated. The resistance
from the stiffness of the pipe was also included in the calculation.

Calculated Probabilities of Slip at Clay Seams

1.E+00
Ice Contact due to Gouge or Pit
1.E-01 Su=5kPa
Su=10 kPa
1.E-02 Su=20 kPa
Failure Probability per Event

Su=30 kPa
1.E-03 Su=40 kPa
Su=50kPa
1.E-04 Su=60kPa

1.E-05

1.E-06

1.E-07

1.E-08
R equired
Burial D epth

Minimum Seam
D epth if Su=20kPa

0 1 2 3 4 5 6
Burial Depth or Seam Depth (m)

The flatter lines give the probability of slip along the clay seam as a function of the depth
and strength of the clay seam. This is compared with the probability of failure pipeline
failure in absence of clay seams as a function of pipeline burial depth. The plot can be
used to determine the minimum depth and strength of the clay seam to ensure that the
probability of failure is not controlled by slip on the clay seam, as indicated by the dotted
black lines.
544 Fundamentals of pipeline engineering

TEAM
• SEIC
– Yves Friedmann
– Allan Terrill, survey, Erosion Allowance
– Jan de Waal survey
– Patrick Flynn, Offshore Pipelines Manager
• SIEP
– Heedo Yun
– Gawain Langford
• C-CORE: FE analyses
• Canadian Seabed Research (CSR): survey interpretation.
• Delft Hydraulics, Walter van Kesteren, Dredging Flume Tests
• Technical University of Delft, Henderikus Allersam, Centrifuge Tests
• Ken Croasdale: JIP
• George Comfort (Fleet Technology), Brian Wright: ice-seabed interaction
simulations

This slide summarizes the various parties that contributed to the ice design of the
pipeline.

Lessons Learnt

• Importance loading data collection (ice and scour data in this

case)
• Must deal with uncertainty – probability theory provides a
good way of doing this and communicating with
management
• Insufficient statistical data is a reality, failure probabilities are
as much an estimation as a reality. There is a real danger of
quantifying uncertainties where statistical data are available
and making assumptions where they are not. This results in
conditional failure probabilities, rather than the probabilities
that management should base decisions on

This presentation has focused mainly on the issues that are unique to ice design, and
where the largest uncertainties are. This is where efforts should concentrate to reduce
those uncertainties if possible. This is not to belittle considerable effort that went into
ensuring that the pipe had the needed deformation capacity, by selecting a deformable
grade of pipe, and ensure that the girth welds would not fracture despite the possibility
of defects.
Structural integrity 545

The reduction of the largest uncertainties is indeed the best place to concentrated efforts.
There is also a real danger in projects that effort will be concentrated around
uncertainties covered by the expertise of people already involved.

Some Contact References

• EP Projects (Rijswijk)
– Ralf Peek
– Spencer Wilmshurst (stresses & strains)
– Paul Verlaan (ice)
– Bareld Hospers (geotechnical)

• EP Projects (Houston)
– Frans Kopp
– Kevin Ouyang

• Global Solutions (Aberdeen)

– Ian Stanley

• EPE
– George W. Brown
– Arash Nobabar (pipe – ice – soils)
Modification and
repair
Modification and repair 549

EXPECTATION

EXPECTATION

 System needs to be upgraded over time

 Identify the methods of pipeline isolation,
hot tap and tie-in
 Major repairs and pipeline replacement
 Know what to consider when planning the
decommissioning of a pipeline

The requirements of a pipeline system changes over time.

We may wish to add a new branch to the network. This involves a tie-in with a tee or
wye (shaped like a T or Y) to existing pipe. Where this is not a hot tap, it is necessary to
first isolate the section to be cut open.

An overview is given for the methods that can be used for pipeline isolation and tie-ins.

The ‘Integrity Management’ module covered minor remedial works intended to stabilise
the damaged section. Where a deep dent or major damage has occurred, it may be
necessary to replace a section of the line or even the whole pipeline or riser. We will
look at methods to carry these out.

Finally, the considerations for planning the decommissioning of pipelines are discussed.
550 Fundamentals of pipeline engineering

ISOLATION

ISOLATION

 Isolate before working on a pipeline

 Repair, replacement or tie-in
 Make the pipeline safe to work on
 Procedure
 De-pressurise
 Remove hydrocarbons and toxic products
 Fill with inert medium
 Suggested basic procedure
 Decommission pipeline and flood
 Why may this be undesirable?

When we are not hot-tapping a connection, in order to effect a repair or to install

facilities for a third party tie-in to the line, we need to isolate a section of the pipeline.
This is necessary to make the pipeline safe. This effectively means removing the internal
pressure and hydrocarbons or toxic contents from the pipeline.

One way of doing this might be to shut down, depressurise and water-flush the entire
pipeline.

In most cases, this is undesirable from an operational point of view. For example, in a
gas trunk line, the depressurisation would involve flaring a lot of gas (lost inventory), and
flooding would lead to a requirement to dewater and vacuum dry, which could put the
pipeline out of service for many months.
Modification and repair 551

LOCAL ISOLATION

 Isolating just the worksite from the pipeline

contents

Work site

Hydrocarbon
Inert fluid/gas

Isolation plug

The alternative to depressurising and flooding the entire line is to perform a local
isolation. This introduces an internal barrier between the product and the worksite.
Therefore, only a short section of the pipeline is flooded and the time taken to flood,
dewater and dry is greatly reduced.

Depending on the isolation system used, pipeline pressure may also be resisted, avoiding
the need to depressurise the system.

ISOLATION METHODS

 High friction pigs

 Tethered or remote set isolation plugs
 Pipe freezing
 Hot tap and stopple

There are a range of isolation techniques. The main ones are listed above and are
described in the following slides.
552 Fundamentals of pipeline engineering

HIGH FRICTION PIGS

 Close to platform
Outer pipe wall
 Head 3 bar (44 psi)
Direction of flow
 Trains of pigs
 5 to 6 (maximum 9)
Flange
Differential diameter
Pressure acting Main body
over the seal of pig

Differential
pressure

Friction Wall force Friction force

Wall force

High friction pigs are bi-directional pigs with oversized polymeric discs giving a high seal
with the pipewall. They are available from a range of pig manufacturers.

The principle they use is that increasing the differential pressure acting over the pig seal
will result in an increase in the force applied on the wall. This increased wall force then
causes an increase in the frictional force resisting the pigs movement. Increases in
frictional force result in an increase in the differential pressure. The point at which in
the pig becomes trapped in the pipeline is determined by controlling the differential
pressure applied over the pig.

They can generally hold differential pressures of about 3 bar (44 psi) and are therefore
used for isolation where the pipeline is depressurised. The use of trains enables a greater
total differential to be held. They are pigged into place. Because the high seal discs will
wear during this placement operation, they are generally limited to use within about 2 km
(1.2 miles) of the pig trap.

The design of high friction pigs is critical to their functioning. Consideration should be
made to ensuring the flanges are capable of holding the seals against the high drag forces
and ensuring they do not pull out. Care should be taken to ensure the seals will not
buckle. Compression set may become a problem, where the seal material relaxes and
does not provide the necessary sealing resistance. High-friction pigs should therefore be
used with caution.
Modification and repair 553

ISOLATION PLUGS

 Tethered close to platform 150 bar (2.2 ksi)

 Remote away from platform 210 bar (3.0 ksi)
 Deployed through pig launcher
 Upstream pipeline pressure locks plug
against pipe wall

Courtesy: ITAS – isolation plugs Isolation plug locking mechanism

Isolation plugs are pigged to the required location, have brake shoes which set against
the pipe wall and hold them in position even against full-line pressure, and elastomeric
seals which are inflated to effect a complete seal against the pipe wall.

The ITAS plug shown (www.itas.biz) is typical of such units, with a conical taper
mechanism able to lock the pipe onto the pipe wall when required and release the plug
after the operation. The sealing system is separate and there are discs to allow the
pipeline fluids to move the plug into the position.

Tethered isolation plugs are suitable for use near to a platform and have been used
extensively for functions such as change-out of platform emergency shutdown valves.
The plug is pigged into position and receives power and control through an umbilical
which is run down the inside of the line. There is a limit to how far the plug can be
pigged from the platform because the plug has to tow the umbilical behind it. Tethered
plugs are capable of withstanding 150 bar (2180 psi) differential pressure.

Remote set isolation pigs are similar in principle to the tether plugs, but do not utilise an
umbilical. Power and control is provided onboard and is remotely operated. This means
that there is no limit to where in the line the isolation pig can be used. Remote set plugs
are capable of withstanding of the order of 210 bar (3000 psi) differential pressure.
554 Fundamentals of pipeline engineering

ISOLATION PLUG

Single-module HydroPlug schematic

This is another manufacturer’s isolation plug showing the details of the slightly different
locking mechanism.

ST FERGUS – VIDEO

There are two parallel gas lines coming ashore, each with a vertical loop and tee leading,
to the slug catcher and gas/condensate separator (not shown).

One of these looped lines is clearly shown in above, with the second just on the left edge
of the photograph. The end of this second loop leads to the pig trap in the foreground
with the sealing equipment ready for insertion. The trap was used to insert a pair of
Modification and repair 555

isolation plugs, which safely allowed planned maintenance involving the replacement of
valves.

Previous to this operation, it had been necessary to ‘blow down’ the line, flaring off (and
losing) all the valuable gas. However, recent UK legislation prohibits flaring.

Tethered plugs require the valves to be closed and a temporary pig trap door with holes
for strippers to be fitted. Two plugs are used to provide double blocking. These were
pushed against the normal gas flow using water until they had passed up over the vertical
loop beyond the valves requiring replacement for refurbishment.

Although successful, subsequent operations now use remote-set (rather than tethered)
plugs.

REMOTE ISOLATION PIG

The figure above illustrates the method of remotely activating a SmartPlug isolator.

The vessel sends extreme low frequency (ELF) signals to a seabed array. As the plugs
arrive, the ELF communication link (ECL) activates the remote activation system (RAS)
which locks the plugs in place.

For a section of pipeline to be de-activated, it is necessary to send a pair of plugs through

the line in the same operation with sufficient separation.
556 Fundamentals of pipeline engineering

PIPE FREEZING

 Introduce plug by freezing liquid/gel within

the pipeline
 Water-based gel
 Maintain at -20 °C to -40 °C (-4 °F to -40 °F)
 Assess low temperature on steel – no impact damage
To surface vessel

Coolant Coolant High seal pigs

Freeze
jacket Vent

Gel

Insulation Cut-out area Product gas

Pipe freezing produces a plug by freezing a slug of water or gel within the pipeline. The
gel or water is introduced within a train of high seal pigs.

Liquid nitrogen is used to chill the coolant on the vessel. This coolant is then pumped
through a freeze jacket installed around the pipe. The pipe is maintained at a
temperature of -20C to -40C (-4F to 40F).

There may be an effect of the low temperatures on the strength of the pipeline’s steel.
There must be no impact damage on the section to be frozen.

PIPE FREEZING

 Form a solid frozen plug in pipeline

 Can withstand pressures >270 bar (3916 psi)
 Avoids need to drain down and refill systems

Pipe freezing a 323.9 mm (12¾ in) carbon Internal view of freeze plug
steel oil line Courtesy: Cyril Bishop Courtesy: BJ Process and Pipeline Services
Modification and repair 557

Pipe freezing involves the controlled formation of a solid frozen plug inside the pipeline
using specialist equipment and techniques. Once formed, the plugs provide isolation of
the line while pipework modifications are carried out.

The pictures show the process on a landline. However, the technique has frequently
been successfully used subsea.

HOT TAP AND STOPPLE

 Can be used at any

location
 Holds up to 150 bar
(2200 psi)
 Only single block
 Sequence
 Install split tee
 Bolted or welded (shown)
 Hot-tap – drill
 Bypass (optional)
 Insert stopple

The final isolation method to be considered is hot-tap and stopple.

■ A split tee is installed around the pipeline. This can either be welded or bolted to
the pipeline.
■ A valve and cutting head are attached to the tee and the hot-tap is made.
■ The valve is closed, the cutter unit is removed and a stopple unit is attached.
■ The stopple is inserted into the pipeline to isolate a section of line.
■ If desired a bypass line can be used, but this is rarely done subsea.
■ The isolated section of line can be purged and worked on.
■ When the repairs are complete the stopple units can be removed from the valves
and a seal disk can be inserted to allow the valves to be removed, leaving only a
blank flange bolted to the tee.

However, the system only provides a single block to the flow. Some safety requirements
require a double block.
558 Fundamentals of pipeline engineering

HOT TAP AND STOPPLE – VIDEO

This animation shows the hot tap and stopple operation.

This can withstand up to 150 bar (2200 psi).

ISOLATION – SUMMARY

 Make the pipeline safe to work on

 Range of methods dependent on
location and pressure
 High friction pigs
 Isolation plugs
 Pipe freezing
 Hot tap and stopple

Any questions?

The simplest method of isolation is to insert a high friction pig to block the line. The
use of this is limited by both pressure differential and distance from pig inserter.

Isolation plugs lock against the side of the pipe wall, forming a barrier to prevent
product flow in the isolated section. These plugs can be either remotely operated or
tethered, depending on the location of the isolated section.
Modification and repair 559

Pipe freezing is another method of isolating a section of the pipeline, involving the
formation of solid frozen plugs capable of withstanding high pressures.

Line stopping involves hot tapping the pipeline, to insert a block head, which prevents
flow of product through a section.
560 Fundamentals of pipeline engineering

TIE-INS

TIE-INS

 Expansion of pipeline network

 Providing an entrance for a third party
 Repair / replacement of pipe section
 Diversion of pipeline leg
 Methods
 Existing flange, tee or wye
 Hot tie-in
 Hot tap – use of valved flange
 Cold tie-in
 Isolate section and purge
with inert gas

Repairing damage is not the only reason for needing to work on the pipeline.

Other common reasons are expansion of the network as new fields are brought on
stream or the requirement to tie in a third party pipeline. Because of decommissioning
of some unmanned platforms, a new bypass diversion was inserted into the Frigg line
(2004 and 2005 seasons).

Where facilities such as a tee or wye have not already been provided, it is necessary to
add them whilst the pipeline is in service.

An alternative solution is to use a single hot tap (without the stopple). This can be
installed with hyperbaric welding or clamp as before, and provides a valved flange to the
new branch.

For a cold tie in, following isolation using one of the methods already described, a
section of the line is purged with inert gas.
Modification and repair 561

STATS GROUP HOT TAP TEE

REPAIR CLAMP

 Tee
 1168.4 mm by
609.6 mm
(46 in by 24 in)
 Contingency
bolted clamp
 144 bar
(2.1 ksi)
MAOP trunk
pipeline

The photograph shows a hot tap tee repair clamp supplied by Stats Group.

It is a contingency repair system for a 1168 mm (46 in) 144 bar (2.1 ksi) trunk pipeline.

TIE-IN PROCEDURE

Isolation Inspection for:

Gas test  Diameter and ovality

 Pipe material
Remove coating
 Wall thickness
Clean weld area  Corrosion

Cut / bevel pipe  Laminations and

inclusions
Jointing / welding  Weld seams in vicinity

Radiography / ultrasonic tests  Deposits in pipe

The procedure for a tie-in follows the flow diagram to the left.

Once the coating is removed, full inspection is required prior to ensure the area is
acceptable prior to cutting into the pipe itself.
562 Fundamentals of pipeline engineering

ADD A NEW TEE

 Hot tap to avoid shut-down

Installation of horizontal tee

A new tee can be added to the pipeline without the need to shut down the existing
pipeline, using the hot tap procedure described previously.

The horizontal tee is installed with the cutter unit and clamp bolts attached. In the Gulf
of Mexico, it is more common to use a vertical tee, so the slinging arrangement is
different.

GROUTED TEE TIE-IN CONNECTION

 Eliminates welding on pipeline

 Maintain production capacity during tie-in
 No pipe ovality
problems
 Metallurgical
independence
 Reduces safety
hazards

For smaller lower pressure lines, grouted tees may be used. This avoids welding.
Modification and repair 563

The grouting sleeve can be dimensioned to allow for any pipe ovality and provides an
electrical insulation between the new and existing lines.

TDW SHALLOW WATER REPAIR

USING HOT-TAP – VIDEO

This video illustrates the hot tapping process used to perform a repair to a 323.9 mm
(12¾ in) flowline in the Yellow sea off the coast of northern China. This flowline
supplied 40 million m³ (1.4 billion ft³) of gas daily to Tianjing. Due to the vital
importance of the pipeline to Tianjing city, the repair work has to be carried out without
interruption to normal supply of gas to Tianjing and especially to the main power plant
in the area.

CNOOC had developed the fields in the west of Gulf of Bo Hai, Yellow Sea in northern
China. In the late spring of 2000, the pipeline was damaged, possibly due to impact
from a sunken ship. The damaged pipeline had a dent of 280 mm by 130 mm by 12 mm
(11 in by 5 in by ½ in) and a rupture of 65 mm (2½ in) long. Initially, CNOOC installed
a make-shift sleeve pipe for the damaged pipe section. However, due to the severity of
the damage, it was unlikely that the pipeline could be pigged in the future and therefore a
permanent repair was required.

Armed with the hot-tapping technology from TDW and Oil States Hydro-Tech, repair
work to the pipeline was carried out in October 2001. CNOOC were in charge of the
project, while COOEC carried out the repair work. The hot tapping operation allowed a
bypass to be installed, enabling the continuous supply of gas to Tianjing while the
damaged pipe was replaced by a pre-fabricated section.

Where
■ CNOOC = China National Offshore Oil Corporation
■ COOEC = China Offshore Oil Engineering Corporation
■ TDW = TD Williamson
564 Fundamentals of pipeline engineering

PREFITTED BLANK FLANGE

 Avoid shut-down of existing pipeline

 Requires knowledge of future requirements
 Repair sites cannot be foreseen
Tappable
 Added cost to pipelay blank flange

Double block and bleed

valve provision

Tappable flange

A recent innovation is the tappable blank flange, which provides a location at which a
valve can be installed and hot-tapped into should it be required at a later time.

This avoids the need for fitting a pipe clamp or hyperbaric welding. Compared with
these alternatives, the tappable blank flange is relatively inexpensive.

However, the possible tie-in locations need to be foreseen. Future damage locations
cannot be determined. How many of these flanges should be added to pipelines and
where is a matter of judgement. The added cost must be allowed for – especially if it is
for a third-party.

Good practice dictates the use of double block valves with a bleed system between. This
means that safe operation can be guaranteed.
Modification and repair 565

VALVING

 Provide suitable facilities and valving for:

 Precommissioning new line
 Operational pigging
 Future entrants without shutting down existing
facilities
 Principles
 Double block and bleed arrangement
 Provides safe isolation
 Drains at low points in piping
 Vents at high points

Consideration should be given to the requirements of pre-commissioning, operational

pigging and future tie-in facilities.

To achieve isolation, a double block and bleed valve arrangement is required. This
means that two valves are used with a bleed-tapping between. This provides redundancy
and a means of monitoring for leakage.

In piping, provision of drains and vents will enable dewatering and drying of pipework.

PIGGING FACILITIES

Pig trap
Branch line flow

New
Existing

Main line flow

566 Fundamentals of pipeline engineering

Once the mainline tee has been fitted with double block and bleed valves, the branch
flow can be attached. It too has a double block and bleed termination.

The valve arrangement provides full isolation both from the main line and the branch
line ensuring a safe working environment for divers or future operators of the lines.

This illustrates how pigging facilities can then be provided. Temporary or permanent
pigging facilities can be attached.

If the pig trap were then removed, this arrangement could be used for provision of
future additional tie-ins. Again, an extra double block and bleed valve set arrangement
would provide the means of isolating the two-branch lines independently.

ADD A NEW WYE

 Wye piece
 Same diameter as the existing
 Allows pigging of the new line into the existing
 Longer pigs
 Different size lines
 Dual diameter pig Main
pipeline Pigged in
this direction

Branch line

Where the new branch is of the same diameter as the existing line, the use of wyes rather
than tees means that both the lines are piggable.

Longer pigs are normally used to ensure that they pass the junction. In gas lines, the
pressure on the branch should be adjusted to ensure easy passage. If the pressure is too
high, the pig will stop before the wye. If it is too low, as the pig passes, the flow will
divert back up the branch.

Where the branch line is smaller than the main line, a dual diameter pig is used. If the
branch is not to be pigged, then bars are sometimes provided at the opening of the wye
to ensure easy passage of the pig.

Although normally pigs are sent in only one direction, an arrangement like railway points
is available that ensures pigs can be sent in the reverse direction to either branch.
Modification and repair 567

NEW WYE

 Shut down
existing pipeline
 Isolate and
cut-out section
 Install new wye
Abaqus FE stress model
entrant

New wye on skid

New

Existing pipeline

The new valves, pipework and wye piece would normally be mounted onto a skid with a
protection structure over.

Valves are normally added to enable shut down of either branch. Again, double block
and bleed system would be provided.

The pipeline would be shut down and the section isolated as before. The new wye
would be connected up using bends.

The FE model of a reinforced wye piece recently carried out by Jee shows the high
stresses (yellow) during hydrotesting in the ‘crotch’ area. The wye can be manufactured
from sections of a 30 bend and a straight pup piece. The three stiffener plates help to
prevent the widest section from bellying out.
568 Fundamentals of pipeline engineering

TIE-INS – SUMMARY

 Need for tie-ins

 Tie-in procedure
 Thorough inspection of existing line condition
 Grouted, bolted clamps and prefitted flange
 Valving
 Double block and bleed
 Pigging facilities
 Comparison of tee and wye

Any questions?

With the continuing expansion of pipeline networks, it will often be necessary to connect
(or tie-in) new pipelines with existing pipelines to provide services to new locations.

Tie-ins can also be used to bypass sections of pipe that require replacement.

We have looked at the steps in undertaking a tie-in with careful inspection of the existing
line to ensure that it is in a suitable condition.

A number of methods have been described including bolted or grouted clamps and a
pre-fitted blank flange - useful when the need had been foreseen.

The main principles of valving and fitting of pig traps has been described.

Tees and wyes have been compared with regard to pigging.

Modification and repair 569

REPAIRS

REPLACEMENT SPOOL

 May be needed if pipeline is severely

dented, buckled or ruptured

 Basic sequence on following slide

 Two options:
 Hyperbaric weld
 Mechanical connector

In cases where the pipeline damage is too extensive to be repaired by a clamp, it will be
necessary to cut a section out and insert a new replacement spool.

The techniques for doing this are initially to isolate the pipe and then to insert the new
section with either a hyperbaric weld or a repair connector. These issues are addressed
in the following slides.
570 Fundamentals of pipeline engineering

SPOOL REPLACEMENT

 Locate damage  Remove and clean

coatings
 Isolate
 Cut-out
 Excavate
 Install new spool

Weld or connect?

Once the location of a leak or damage has been detected, the pipeline on either side is
isolated using one of the methods described earlier.

The seabed beneath the site needs excavating to gain access for the equipment needed.

The damaged pipework is cut out and a new section of spool inserted.

We have two options to make the ends up: welding or jointing.

HYPERBARIC WELD

 Dry weld within chamber

 Diver performs weld
 Multiple qualifications
 Diving, welding, NDT
 Procedures to account
for pressure effects:
 Arc voltage
 Arc stability
 Chemistry
 Heliox mix is needed
even in air diving depths
Modification and repair 571

During the initial development of the North Sea oil and gas fields in the late 60s and
early 70s, it became apparent that the wet welding techniques used in salvage and civil
engineering applications would not be adequate for these deeper water conditions. This
led to the development of hyperbaric welding techniques which have been utilised for
the tie-in and repair of subsea pipelines.

A hyperbaric weld is carried out in a dry chamber, known as a habitat, which is placed
over the pipeline and the seawater is expelled with a helium/oxygen breathing gas.
Diver-welders enter the habitat and perform the welding operation in a dry environment,
working at the ambient seabed pressure. Prior to the installation of the habitat, the pipe
ends are aligned using pipe handling frames located on the pipeline.

Hyperbaric welding was initially developed using manual procedures, such as gas
tungsten arc welding (GTAW) and shielded metal arc welding (SMAW). GTAW (due to
its high quality but slow deposition rate) has been used mainly for the root and hot pass,
and SMAW, which is a quicker but a less controlled process, used for the weld cap.

The quality of manual welding is directly related to the performance of the welder, which
can vary from welder to welder, and is also be dependent on the water depth, as human
performance can be impaired due to the effects of hydrostatic pressure.

In recent years, mechanised welding systems have been developed which have improved
the overall quality and repeatability of hyperbaric welding, and also allow welding to be
carried out at greater depths. Special welding procedures need to be prepared to account
for the different pressure at depth.

However, the operator needs to maintain full qualifications in diving, end preparation,
welding of a number of possible diameters, wall thicknesses and pipes material as well as
many types of NDT operations.

HYPERBARIC SPREAD

Welding habitat Pipe alignment frame

 Frigg diversion 2004/5

 Large pipe, 813 mm (32 in); thin wall, 19 mm (¾ in)
 Total of 6 pups for line diversion
 Platforms to be removed

These figures show a welding habitat and associated pipe alignment frame.

Subsea 7 have undertaken a number of hyperbaric welds for Total on the Frigg lines’
TP1 / MCP-01 bypasses as part of an EPIC contract.
572 Fundamentals of pipeline engineering

This required hyperbaric welding on 813 mm and 610 mm (32in and 24in) pipelines to
re-route them around two platforms in 90 m to 110 m (300 ft to 360 ft) of water.

Hyperbaric welding was selected because they are large diameter with a relatively thin
wall, 19 mm (¾in).

In the first season 4 pups (8 welds) were welded with a further 4 welds during the second
season.

Once the Frigg diversions were completed, the steel platforms will be removed in their
entirety (similar to the procedure at Maureen) but only the topsides will be removed
from the concrete gravity base structures.

This contract follows Subsea 7’s successful completion of hyperbaric welding work in
Australia.

PROS AND CONS OF WELD

 Diving time generally greater than

connectors
 Water depth limitation
 Usual manual SMAW – 200 m (660 ft) deep
 Exceptional trials much deeper by Comex and Global
 GTAW deeper – in excess of diving depths
 Statoil’s use of robotic equipment
 Pipeline returned to original condition
 ‘Golden weld’ testing

A hyperbaric weld will generally take considerably longer than is required to make up a
mechanical connection. There are water depth limitations for hyperbaric welds, although
welds are generally feasible within diver depths.

Exceptional diving depth trials have been undertaken by Comex to 501 m (1644 ft) in
the Mediterranean, whilst a successful working dive at 328 m (1075 ft) has been carried
out by Global in the Gulf of Mexico. Robotic equipment at very great depths could use
tungsten welding equipment.

The advantage of a hyperbaric weld is that the pipeline is returned to its original
condition with no subsea equipment remaining.

It is also possible to avoid hydrotesting of the repair by use of the ‘golden weld’, a
combination of non-destructive testing methods.
Modification and repair 573

GOLDEN WELD

 Full NDT of repair welds

 Radiographic, ultrasonic and magnetic particle
inspection
 Test results used as evidence for weld
acceptability
 Repaired section does not require
hydrotest or further testing
 Spools or pup pieces have been yard-tested

A ‘golden weld’ will undergo a thorough testing with a number of different non-
destructive testing (NDT) methods. The results are used to verify the acceptability of
the weld.

This then eliminates the need to conduct a full hydrotest of the entire pipeline containing
the repaired section.

Note that new sections or pups being used to replace defective sections will have already
undergone a full hydrotest in the yard or on the vessel prior to their incorporation into
the existing system

MECHANICAL CONNECTORS

 A number of systems available

 Install connector on cut ends of pipeline
 Seal on outside of pipe – provides flanged ends

Flexiforge
roller tool
574 Fundamentals of pipeline engineering

The alternative to a hyperbaric weld is the use of a mechanical connector. There are a
variety of connectors available, but all make a seal with the outside of the pipe, which
first needs to be cleaned of all coatings.

The main types are described in the following slides.

The diver in the picture is inserting the Flexiforge tool into the connector. See next
slide.

FLEXIFORGE CONNECTOR

 Pipe cut and flange sleeved onto end

 External pipe coating cleaned off
 Exposes bare steel
 Internal tool rollers
plastically expand pipe ‘Anvil’
 Connector ‘anvil’
remains elastic
 Grips and seals
around outside
of pipe end

The Flexiforge system is available from Big Inch.

The system involves an end fitting incorporating a standard flange, which slips onto the
cut pipe end and is swaged. This is a cold-forging process performed using an internal
expansion tool. The pipe is plastically expanded into the connector. The connector
incorporates a system of rings and grooves that ensure a metal-to-metal seal.

Because the thicker-walled Flexiforge fitting is elastically expanded whilst the pipe is
plastically expanded, the fitting becomes pretensioned in the hoop direction once the
forging tool is removed. This ensures a high axial load capacity.

Once the two flanges are fitted, a short spool can be connected across the gap.
Modification and repair 575

MORGRIP CONNECTOR

Gripping segments Pipe end External seal

(two per pipe shown) abutment test port

Undamaged section of Ball Twin Replacement pipe

pipeline cage seals spool

The Morgrip connector is available from Hydratight.

The connector is slid on to the cut pipe, positioned and then activated by tensioning
longitudinal bolts. The Morgrip contains graphite activated metal sealing rings (shown in
black in the picture).

It has an attachment mechanism, based on ball bearings, which indent the external
surface of pipe. These act to wedge the connector onto the pipe, so that the harder the
pressure tries to push it off, the more the ball bearings dig in. Again, it is important that
all coatings have been removed from the pipeline.

Two sets of seals are incorporated to allow a leak test to be performed between them.
576 Fundamentals of pipeline engineering

MORGRIP

 Generally
diver-installed
 ROV version
available
 Tensioning
bolts
 Activates
gripping system
and metal seals

Both diver and ROV installable versions are available. The Morgrip connector has been
used for both repairs and new-build tie-in applications.

PROS AND CONS OF CONNECTORS

 Connectors quicker than hyperbaric weld

 Availability of diverless systems means no
depth limitations
 Some systems able to accommodate poor
axial tolerances
 Need to perform leak test
 Back pressure between the seals
 Time to procure
 Emergency repairs
 Smaller sizes of pipe

The main advantages of mechanical connector systems are the fast make-up time and,
because diverless systems are available, no depth limitation.

Mechanical connectors do need to be leak tested which, for some connectors, requires a
full system hydrotest.
Modification and repair 577

This full system test can be avoided if the connector includes a seal test port, as does
Morgrip. This tests as a back pressure between the seals.

Some types of connectors are not made for larger diameter pipelines, and specials may
be difficult to procure in time for emergency repair operations.

Depending upon the risk assessment, it is normal to have either standby connectors or a
fully trained hyperbaric welding team.

ROVS USED FOR DIVERLESS

REPAIRS

A-frames or
moonpool

In shallow water, ‘Cursor’

use a direct umbilical Strong umbilicals launch
supplying control, system
power and support
Workclass Tether
or eyeball ROV management
system (TMS)
Sea current
or ‘Top Hat’

Workclass ROV carries

Heavy equipment tool pack or equipment
and tools lowered Lightweight slung beneath
from surface on frame tether
or in basket Steerable garage
unit with thrusters
Secondary
eyeball ROV

The figure shows three procedures for operating ROVs, two of which are specifically
designed for deep-water applications. When operating in deep-water environments, one
of the main concerns is the time taken to lower the ROV to the seabed (this can be
several hours). As the ROV requires a power supply cable there is a problem in that the
umbilicals become both heavy (due to their length and strength requirements) and are
subjected to large loads due to sea currents. Operators have developed two main
systems where a powered unit with separate thrusters carries the workclass ROV down
to the work site. This unit can be sized to withstand the loads from the main umbilical.
When in position, it then releases the ROV on a lightweight umbilical or tether. This is
normally up to a few hundred metres long, but can be made up to 1 km (3280 ft).

The figure shows two different methods of deep-water ROV installation. One involves
lowering the ROV in a steerable garage. Any heavy equipment or selections of tooling
can then be lowered to the seabed on a separate frame. This method may also include a
secondary eyeball class ROV slung beneath, which can be used to oversee the operations
of the workclass ROV or other tasks.

The second method involves lowering the ROV on a device known as a TMS (tether
management system) or Top Hat which releases the ROV at the worksite. The ROV in
this case grasps a separate tool unit beneath. This may be a trencher, burial device,
flowline connection module, suction anchor installation, mining or military.

Launch using a ‘Cursor’ enables the almost neutrally-buoyant ROV to be pushed safely
through the surface zone (where the thrusters have difficulty operating) into the deeper
578 Fundamentals of pipeline engineering

water beneath the vessel. The cursor can run down a set of rails or wires, and it holds
the TMS or garage.

Typically, workclass ROVs locate themselves by the left arm grasper, and manipulate
using the right arm. When following a pipeline, the ROV may fly above the route, run
on tracks or grip the line using wheels.

DIVERLESS REPAIRS

 Diverless repair techniques:

 During construction:
 Cut and remove damaged section, install A&R head to
pull pipeline back up and continue laying
 Recover buckled or dropped pipeline to surface to effect repair
 In operation:
 Repair pipe section in place with mechanical connectors or
grouted sleeve
 Flexible risers:
 Normally replaced
 Leaking annulus can be repaired by flushing with inhibited
water
 Risk reduction plan

ROV-operated systems have been developed for pipeline repair in deep water. That is,
below diver depth.

During installation, the buckled section may be cut off and removed, and a temporary
pulling head fixed onto the end of the undamaged section to enable the line to be
dewatered, thus restoring its buoyancy. The recovery wire is then pulled up to the
laybarge and laying resumes.

Alternatively, the buckled line may be used as a recovery system to pull the undamaged
section back in order to effect a repair on the barge.

We have just seen how a pipe can be repaired with mechanical connectors once in
operation.

Flexible risers are normally replaced. However, some repairs of impact damage to the
outer layer may be effected using clamps. The annulus is then filled with inhibited water.

In practice, most deep-water systems are built to perform a specific repair as part of a
risk reduction plan. Fortunately, they are rarely used.
Modification and repair 579

CASE STUDY 1:
BUCKLE REPAIR CUTTER

 Sonsub-Saipem
 Bluestream wet buckle repair system

Diamond
cutter unit

This cutter can be used to repair a wet buckle at depths of up to 2200 m (7220 ft). It
was developed for use on the Bluestream Project in the Black Sea.

The intention is to cut the pipe below the buckle and insert a recovery head. The
evacuated pipe can then be brought back to the surface and laying continued.

BLUESTREAM HEAD

 Bluestream depth 2150 m (7050 ft) of water

 ROV-operated equipment
 Use of cutter to produce ‘square end’
 Pressure to force pipe onto anvil
 Pipeline plastically deformed to form seal
 Single use unit
Pipeline Anvil Deformed pipe wall

Pressurised
volume

The second tool designed to be used by the ROV sealed the end of the pipe, enabling
water to be removed and allowing the more buoyant pipe to be lifted.
580 Fundamentals of pipeline engineering

A thin-walled can was inserted inside the pipe and expanded. This plastically deformed
the end of the pipe onto an anvil. The latter deformed elastically.

When the pressure was released, the anvil recovered (elastically) and held the
permanently deformed pipe wall using friction.

Valves (not shown) permitted the pipe to be purged. This reduced the weight enabling
the pipeline to be recovered onto the laybarge, where the unit could be cut off.

Fortunately, it was a contingency item only. Although proven technology, it was not
used in anger.

CASE STUDY 2: RISER REPAIR ON

FOINAVEN – VIDEO

Diverless riser repair is shown in the above video.

The main points are:

■ The deep water – 500 m (1640 ft) – flexible risers had external sheath damage
during installation, and this caused flooding of the armour layers.
■ The task was to displace the seawater with inhibitor in order to prevent
deterioration of the armour wires.
■ This was carried out by remotely-operated vehicles (ROVs), which fitted clamps
over the damaged sections and drilled vent holes to facilitate flooding of the
annulus with inhibitor.
Modification and repair 581

CASE STUDY 3: BALLAST WATER

SEALINE IN BAHAMAS

 Loading facility
 Crude sent from shore storage tanks to ‘sea island’
 Ballast water line discharge from ship to shore
 Low-pressure line 914.4 mm (36 in) by 1200 m (3940 ft)
 Inspection using RTD tethered crawler
 Ultrasonic pipeline
inspection technology
 Detector stand-off at
nominal bore 75 mm (3 in)
 Transmission medium
 Water or oil
 Zig-zag coverage of wall
Spider PIT

The ballast water line is used to transfer contaminated water from the tankers moored at
the terminal when they are receiving crude oil from the onshore storage tanks. The site-
applied internal lining at the field joints failed due to quality control (QC) problems
during construction, resulting in severe internal corrosion.

A single pipeline was used to transport oil offloaded from tankers at a sea island
structure to the tank-farm onshore. Because it was a single line, it was not possible to
regularly pig the line after each discharge.

Röntgen Technische Dienst bv (RTD Quality Services) of Rotterdam provided an

ultrasonic tethered crawler inspection Spider PIT to detect loss of wall thickness in the
line. The detectors had a stand-off from the wall of 75 mm (3 in) and were passed in a
zig-zag pattern over the whole inner surface of the line. It is necessary to use a liquid
medium such as water or oil between such detectors and the inside surface of the steel.
582 Fundamentals of pipeline engineering

CORROSION OF SEALINE IN
BAHAMAS

 Severe pitting corrosion

 Particularly bottom of line
 12.7 mm (½ in) down to 3 mm (⅛ in) wall thickness
 Corroded areas over 0.5 m (18 in) long

The RTD crawler proved that severe corrosion had occurred, particularly to the bottom
of the line. The nature of the corrosion took the form of deep pits: in a number of
places, the wall had been reduced from 12.7 mm (½ in) down to just 3 mm (⅛ in); in
some areas, the corrosion affected over 500 mm (20 in) length mainly at the 6 o’clock
position (bottom of pipeline). In three exceptional pits, there was just 2 mm (0.078 in)
of wall left.

It was suspected that the intermittent discharge of oil (every few days or so) allowed the
small amount of water to drop out and collect at the bottom of the pipeline and then
travel back down to the lower offshore end of the pipeline.

The photographs show sections of the line that were subsequently removed during the
repair operations. The first shows the corroded wall with a core sample removed for
testing (at lower right) and the pits covered with corrosion product (rust). The second
shows a cleaned surface prior to repair, demonstrating the depth of the pits encountered.
Modification and repair 583

GRP RELINING OF SEALINE IN

BAHAMAS

 Glass-reinforced plastic liner

 Simple butt connection with GRP overlap
 Strings pushed into steel pipe avoiding buckling
 Pressure test liner
 Grouting of annulus
 300 m (1000 ft) sections
 Start at lower end
 Monitoring at ports
 Grout release ensures
full filling of annulus

Land and Marine Engineering relined the sealine using a glass reinforced plastic (GRP)
liner. This was purchased in 15 m (49.2 ft) lengths and assembled into 200 m (656 ft)
strings using simple GRP overwrapping of the square-butt ends.

The strings were pushed down into the damaged steel line using winches on the beach
attached to a beam at the rear of the string. Sets of rollers on top, bottom and sides of
the pipe were required to avoid strut buckling of each string. These had to be released as
the beam moved forward.

The photograph shows the pressure testing of the liner. Once this was completed, the
annulus was filled with grout to provide fixity to the liner within the pipe during
operation. This commenced at the offshore end with displaced water being released
from ports drilled into the damaged steel line every 300 m (1000 ft). The ports were
monitored for arrival of the grout so that operations could move to the next section
towards the shore.
584 Fundamentals of pipeline engineering

REPAIRS – SUMMARY

 Operations for repairing ruptured pipelines

 Hyperbaric weld
 Diver qualifications and welding equipment
 Golden weld testing
 Connectors
 Typical examples – Morgrip and Flexiforge
 Benefits of using welds or connectors
 Diverless repair
 Installation and operation – risk-reduction systems
 Relining weakened line
Any questions?

In the Pipeline inspection module, we looked at repair clamps that can be used as a
reinforcement for minor defects such as dents.

However, when a section of the pipeline has become significantly damaged, then it may
be necessary to replace that section of the line with a replacement spool.

Two types of connection can be made to tie-in the spool. One is to perform a
hyperbaric weld, this will be relatively expensive. A ‘golden weld’ means that the system
does not require a pressure test.

The other tie-in method is to fit mechanical connectors to the ends of the spool and the
existing pipeline. A range of mechanical connectors are available, only two of which
have been shown here. The advantage is they do not all require a diver to make the tie-
in and allow remote operation in deep water.

In deep water, we must use diverless techniques to connect the pipelines to the risers or
well heads. We also need diverless methods for repair to lines should damage occur.
The tools used for undertaking such work usually are attached to ROVs. We have
examined how these tools are operated.

Methods of pipeline and flexible riser repair have been shown, both during installation
and operation.

Repair systems must be fully tested and available for emergency operations. Fortunately,
they are rarely needed.

Where a leak has not yet occurred on a short length of sealine, it is possible to provide
strength using a GRP lining.
Modification and repair 585

DECOMMISSIONING

DECOMMISSIONING

 OSPAR convention 1992

 Portugal agreement 1999:
 Platforms under 10 000 tonnes (11 000 US tons)
removed
 34 over that on case-by-case
basis
 Maureen platform – removed
 Frigg – workscope though to 2012
 Topsides to go with tubular steel
supports
 Concrete gravity bases remain
 Pipelines on case-by-case
 Emotive issue: Brent Spar

What to do in order to decommission pipelines and platforms has been debated for
decades. The focus of the debate has been platforms, and pipelines have received little
attention.

The centre for the discussions has been the OSPAR (Oslo/Paris) Convention. This met
in Portugal in 1999 and agreed that in the North Sea, platforms under 10 000 tonnes (11
000 US tons) should be removed. Platforms over that weight, along with pipelines,
should be considered on a case-by-case basis.

The Maureen platform has already been removed from the Northern North Sea.

The plans to remove the topsides to the Frigg platforms have been drawn up. The
tubular steel support structure will also be removed but those with a concrete gravity
base are to stay. The alternative to totally removing tubular supports is to cut them off
at a level safe for shipping. This might be preferred when they are fixed to the seabed
with difficult-to-detach piles. The Frigg removal work is to be completed by the end of
2012.
586 Fundamentals of pipeline engineering

We should not be complacent about decommissioning. The public storm over the deep-
sea disposal of Brent Spar is a case in point. No matter what the carefully evaluated
scientific and environmental solution might be, a lot of damage can be done to a
company’s reputation and sales if the public relations are not handled correctly.

DECOMMISSIONING TO DATE

 Common in Gulf of Mexico

 Not as common in North Sea
 Most rigid pipelines put into ‘protective storage’
 Reuse – perhaps for CO2 injection
 Deferred decommissioning
 No precedent for other lines
 Accountants cost savings
 Many flexibles pulled up for reuse
 Especially in Brazil
 Brazil – used as reef (with topsides removed)
 Nursery for fish

Although a common practice in GoM, there have been relatively few pipelines
abandoned to date in the North Sea.

Most of them are cleaned and then sealed, so a decision on their long-term future can be
made at a later date. They might be reused for development of smaller reserves or for
injection of CO2 back into reservoirs. This helps with extraction and may gain carbon
credits in the future.

By deferring the decommissioning, it has the advantage of not setting a precedent. The
costs of the work can be postponed into some future year’s accounts.

However, the trend elsewhere appears to be towards leaving buried pipelines in place
and removing unburied lines.

Flexibles have a good record for re-use, particularly in Brazil where they are routinely
retrieved, refurbished and re-laid. This is not so in Australian waters where flexibles
tend to be recovered to the shore for disposal.

Incidentally, the Brazilians have also recently placed a disused structure in a fish
spawning ground to act as an artificial reef. This has been done with the backing of
government fisheries scientists and has been shown to be successful in terms of
providing a safe haven in which fish can breed.

This lead might be followed elsewhere in the world. It raises the question of whether
pipelines on the seabed are beneficial to fish (and even to fishermen) in acting as
breeding sites.
Modification and repair 587

DECOMMISSIONING – SUMMARY

 Limited subsea pipeline decommissioning

 Mothballed pipelines
 Disused but protected
 Flexibles reused
 Topsides removed
 Support structures either removed or cut off

Any questions?

Decommissioning involves the removal of subsea equipment and components at the end
of the service life.

In the North Sea, there has been limited decommissioning of pipelines. Many are
mothballed or abandoned in a state of protected storage after the subsea manifolds,
wellheads or similar structures have been removed.

Where possible, flexibles will be reused.

Platform topsides often require removal due to their visible nature, but how they are
ultimately disposed of will be determined on a case-by-case basis.

The supporting structure may be removed entirely or cut-off at a level safe for shipping.
588 Fundamentals of pipeline engineering

MODIFICATION AND REPAIR –

SUMMARY

 Identify the methods of pipeline isolation,

tie-ins and repair
 Know what to consider when planning the
decommissioning of a pipeline

Any questions?

An overview was given for the methods of isolation, tie-ins and repairs that may be
needed during the life of a pipeline.

Finally, the considerations for planning the decommissioning of pipelines have been
described.
Decommissioning
Decommissioning 591

EXPECTATION

EXPECTATION

 Pipeline decommissioning
 Principal factors influencing selection of method
 Legislation and current thinking
 Differences worldwide
 Options for a disused pipeline
 Operations for decommissioning and leaving in-situ
 Recovery of pipelines
 Potential for re-use
 Hazard/safety and financial costs
 Expected boom in North Sea 2010 to 2020

An overview is given of the processes for decommissioning pipelines and other offshore
components.

The principal factors that influence the methods of decommissioning are identified:
these being environmental and safety concerns, public opinion, political needs and finally
cost-effectiveness. The latest legislation and current thinking regarding the correct
decommissioning strategy is discussed.

Decommissioning of pipelines in-situ is examined in detail and the required operations

are detailed. Also, the methods available for the possible recovery or re-use of certain
pipeline systems is discussed.

An estimation is provided of the likely financial and safety costs and benefits for removal
of pipelines.

It is predicted that there will be a sharp increase in decommissioning operations in the

North Sea within the next few years.
592 Fundamentals of pipeline engineering

INTRODUCTION

INTRODUCTION

 Need to decommission
 Increasingly environmentally-
conscious world
 Pollutants and toxins
 Effects of seabed debris
 Fishing and dredging interests
 Congestion of seabed
 Now a legal requirement
 Determine methodology for
removal at design phase
 Involve stakeholders early on
 Avoid ‘decide and defend’ approach

At the end of the operational life of a pipeline, there is a need to address the future
condition and status of the pipeline, so that it never presents a risk of pollution or
interference with the activities of other users of the sea.

It is important that all stakeholders be involved early on. We must avoid making a
decision and then trying to defend it. This can result in a huge increase in additional
costs.

The upper picture shows 12 shore-end pipe connections exposed at low tide, at the
Thorness Bay SOLO pipeline terminal (part of the PLUTO pipeline system) on the Isle
of Wight. They have survived over half a century of battering by the sea.

The lower photograph shows the effect of a century of oil spillages that are currently
being cleaned up in Baku, Azerbaijan on the shores of the Caspian Sea. The historical
remediation work is a prerequisite condition for new abstraction concessions.
Decommissioning 593

ENVIRONMENTAL AND SAFETY

INFLUENCES

 Environment
 Look for environmental benefits
 Least-impact option
 Assess environment hazards
and injury to personnel
 ‘Sterilisation’ of seabed for
future pipelines
 Return shoreline to original
 Safety
 Safety – nuisance on seabed
 Snagging of trawler nets Shell Brent Spar
 If cannot present a good safety case – leave as is!
 Consider all risks during removal and disposal

The decision as to whether the pipeline is abandoned in-situ or recovered to land for
disposal or recycling, is influenced by the above issues. The considerations include:

Environmental
■ Would the removal represent a benefit to the environment or would resources
required be better spent in other directions? It is common for coastal and local
authorities to demand the removal of lines at the landfall, allowing the sea to
erode beaches and cliffs naturally (for decades to come)
■ Contamination from unclean lines
■ Determine best possible environmental option. Greenpeace demonstrated about
the decommissioning of the Shell Brent Spar. Finally, they admitted the original
solution would have been a cleaner option overall

Safety
■ Hazards relating to subsea pipelines
■ Snagging of trawl equipment
■ Nuisance to future seabed construction
594 Fundamentals of pipeline engineering

IMAGE AND PUBLIC PERCEPTION

 Political need
 International guidelines and common approach
 Installation reviewed for best solution
 Public opinion
 Pressure groups
 Media sensationalism
 Local politicians
 Exxon Valdez, Alaska
 Erica and Prestige,
Biscay and Spain
 Fishing and tourism
Prestige

Political need/public opinion

■ Legislation and guidelines
■ Each installation to be viewed on its own merits
■ Operators being persuaded to take action
■ Public now more aware of issues. However, pressure groups do not always reach
a considered opinion (for example, the Brent Spar situation). They are sometimes
heavily influenced by the press or local politicians

It was right that huge amounts of money were spent cleaning up long lengths of the
formerly pristine Alaskan coastline following the leaks from the Valdez in 1988.

The sinking of the tankers Erica and Prestige in the Atlantic off south-western Europe in
1999 and 2002 caused an outrage. Local holiday and fishing industries had just
recovered from the first incident when the Prestige sank with some of her cargo
remaining sealed on board.

Though there was no financial gain to be made for heavy oil remaining in the tanks,
public outcry demanded that it be recovered from a depth of 3800 m (12 500 ft) to
prevent it gradually seeping for decades to come. It is unclear how much of a threat this
might have been, given the viscosity of the heavy oil and the low rate of corrosion at that
depth. The fuel had to be pumped out of holes drilled into the hold through a 150 mm
(6 in) bore hose.
Decommissioning 595

BENEFITS AND COST

EFFECTIVENESS

 Benefits and cost-effectiveness

 Sale of recovered materials is negligible
 Hazard has been removed with any liability
 Minimum maintenance on empty line
 Can accountants delay for another fiscal year?
 Sell on the facility?
 Was money left in budget (or included in sale)?
 Pipelines are ‘out-of-sight and out-of-mind’ of public
 Reuse for another field or for CO2 disposal
 International agreement reached February 2007
 Have they sold the liability with facility?
 Sale of platforms to wind generator companies
 Trunk pipelines used for power lines

Benefits and cost effectiveness

■ Total removal of a hazard
■ Eliminates future monitoring
■ Sale of recycled materials generates little income
■ Have the operators budgetted for pipeline removal?

It is in the interests of the company to delay removal of facilities. They can undertake
minimum survey and maintenance for a number of years whilst the line is empty.

Perhaps it is possible to find a new use for the pipeline. Perhaps further smaller fields
can be discovered and developed. Or we may find in the future that carbon dioxide can
be disposed of in reservoirs, gaining carbon credits. An international agreement was
reached in February 2007 on the use of hydrocarbon reservoirs for carbon sequestration.

One benefit to us is the fact that offshore pipelines are hidden from view of the press
and public. Providing they remain inert and safe, they can often be left on the seabed
with little concern. This contrasts with the landfall and offshore jackets and other
topside structures.

One important aspect to note is whether the current owner of the subsea facilities has a
budget to de-commission them. It is now common for the original owner to have sold
them on to smaller oil companies. Problems may arise in the future should these small
companies go into liquidation with no assets for removal.

Proposals have been made to sell platforms to wind generation companies. The
pipelines provide a conduit for power lines to shore.
596 Fundamentals of pipeline engineering

OPTIONS FOR ABANDONMENT,

DECOMMISSIONING AND REMOVAL

 Leave pipeline on seabed

 Bury pipeline below seabed
 Recover pipeline to shore

 In all options:
 Remove all end and crossings structures
 Manifolds, SSIVs, wellheads, drilling templates and mattresses
 Make safe
 Clean, seal and water-fill all pipelines
left on or buried beneath seabed

These are the decommissioning options that are addressed later in this section.
Decommissioning 597

LEGISLATION

CONVENTIONS ON OFFSHORE
STRUCTURES AND PIPELINES

 International
 Geneva Convention on the Continental Shelf 1958
 London Dumping Convention 1972
 UN Convention on Law of the Sea (UNCLOS) 1982
 International Maritime Organisation Guidelines 1989
 IMO is part of UN – sets MARPOL, SOLAS and ISPS standards
 European
 Oslo Convention 1972
 Oslo Commission Guidelines 1991
 OSPAR (Oslo-Paris) Convention 1992 and 1998
 USA
 Minerals Management Service (MMS) legislation

There is a plethora of conventions relating in some way to the removal of installations

from the seabed. Most of them have been aimed at shaping what should happen to
structures and platforms when decommissioned. By comparison, pipelines have
received far less attention.

In Europe (the European Union and signatory countries, together with Norway, Iceland
and Switzerland), the OSPAR convention holds.

The International Maritime Organisation is part of the United Nations and is

headquartered in London. IMO sets international maritime standards, such as
MARPOL (prevention of MARine POLlution), SOLAS (Safety Of Life At Sea) and ISPS
(International Ship and Port Security). These standards are not law but are binding for
signatories, which includes all the major trading nations as well as many minor ones.

The United States of America is not a signatory to OSPAR. The MMS legislation –
which has similar aims – is applicable under the auspices of the US Department of the
Interior. MMS together with the US Department of Transportation (DOT) Research
and Special Programs Administration Office of Pipeline Safety (RSPA/OPS) have
jurisdiction in US waters.
598 Fundamentals of pipeline engineering

OTHER NATIONAL REGULATIONS

 Netherlands
 Mining Act 2002 and Mining Decree (Mijnbouwbesluit)
 Norway
 Petroleum Act 72 1996
 UK
 Petroleum Act 1998

In addition to the OSPAR convention, other national regulations apply.

Guidance from these national bodies further interprets the international agreements.

PRAGMATIC APPROACH TO
PIPELINE DISPOSAL

 Comparative assessment
 A balanced judgement Technical
Feasibility
Complexity
(variable burial
depths)
Press Social Environmental
and Sea users Energy usage
media Community Air
Stakeholders Sea
Land

Operating Economic Safety

company’s Cost/benefit Risk assessment
reputation Uncertainty (diving operations)
Legacy/liability Implementation

Both the MMS and DTI take a pragmatic approach to the requirements of the
legislation. This makes a balanced judgement of all factors including the environmental,
safety, cost and feasibility. The underlying criteria for any pipeline left on the seabed
should involve its cleanliness, stability, burial and the interests of other users.
Decommissioning 599

Some aspects of pipeline removal may involve danger to the diving team charged with
removal of equipment. For example, the lifting of mattresses may involve the use of
hooks/lifting lugs well outside their certification date. With the sandwaves in the
southern North Sea and elsewhere, and the soft mud of the GoM, West Africa and
Brazil, unburial equipment may no longer be able to reach the depth they are now
covered to. The weight of soil and destruction of the seabed during removal operations
effectively prevents removal.

Finally, the operating company’s own reputation should be included. The press and
other media can quickly distort the real picture – a case of ‘give a dog a bad name’ –
particularly when dealing with the oil industry.

EUROPEAN AND US APPROACHES

‘EVERYTHING TO BE REMOVED’

 OSPAR July 1999

 Signed at ministerial meeting in Sintra, Portugal
 Presumption of total removal
 Concentrates on offshore rigs and wellheads
 Larger structures may be ‘derogated’
 Pipelines likely to be assessed case-by-case
 Flowlines probably removed along with wellheads
 Trunk lines left in place – except at shoreline
 MMS removal studies – Regional Supervisor
 Same approach but slightly different interpretation
 All large lines ‘deemed to be an obstruction’ to be removed
 Some deepwater infield flowlines may be left (buried in mud)

The OSPAR meeting decreed that the approach to be followed must minimise damage
to the environment.

The basis of the treaty is that everything will be removed apart from those items that are
‘derogated’ or exempted, after an impact and risk assessment study. Smaller platform
structures must be removed. Larger structures may be partly demolished and removed,
apart from sections that would be difficult or dangerous to take away. Pipelines may also
be dealt with on a case-by-case basis.

The UK DTI guidance means that it is expected that wellheads and associated flowlines
will be removed, whilst the large diameter trunk lines to shore will be capped and left in
a safe condition, apart from at the shoreline, where they will be removed.

In the USA, the same basic assumption is used, in that everything should be removed.
However, it is expected that MMS will allow small flowlines to be left buried in the mud,
whereas the larger diameter trunk lines must all eventually be removed. The decision of
the MMS Regional Supervisor will determine whether a line is an obstruction and thus
require its removal, under code 30 CFR 250.1754. Other applicable regulations are 30
CFR 250.1750 to 1754 and 30 CFR 250.1006. The MMS grants a 61 m (200 ft) ROW
corridor for trunk lines and their license requires removal within 1 year of cessation of
use: nevertheless, there is a waiver granted in 95% of cases. Only pipelines that are an
obstruction have been removed.
600 Fundamentals of pipeline engineering

An MMS report examines the best options for this work: Report N° 32.701.001/R1 16th
September 2004, An assessment of safety, risks and costs associated with subsea pipeline
disposals.

See http://www.mms.gov/tarprojects/480/ScanPower%20Final%20Report.pdf and

addenda.

ACTUAL PRACTICE FOR REMOVAL

 Driven by political expediency

 Not science or logic
 Hydrocarbons still being found & recovered
 Possible future use for redundant pipelines
 Remain as a company asset
 Most disused rigid lines are still left in situ
 Many removed in Norway and Gulf of Mexico
 Most flexibles removed and re-used
 Spares stored in Brazil and new ends fitted for reuse
 Practice proscribed in Australia

The disposal of pipeline assets is not necessarily logical, but instead is dependent upon
politics. It might be foolish to remove lines when hydrocarbons are still being developed
– and will continue to be until around 2030 or more. Future use could therefore be
made of these lines.

Therefore, to date, most rigid lines have been decommissioned by cleaning and leaving
them in place.

Nevertheless, a number of pipelines have been removed in Norway and Gulf of Mexico.

Most flexibles have been removed and re-used, particularly in Brazil where many spare
lines are stored underwater in a sheltered bay. However, in Australia, such reuse is
deemed to be too high a risk.
Decommissioning 601

UK DTI GUIDANCE NOTES

 Legislation set out in DTI Guidance Notes

 Based on The Petroleum Act 1998
 Converts convention into good practice
 Gives guidance on pipelines
 Approach to be taken
 Consider all options and effects
 Future consequences of corrosion
 Lines that can be left in place
 Buried and long trunk lines
 Lines that should be removed
 Small diameter and untrenched lines
 Monitoring
 Unlimited time period !

The current status of legislation is set out in DTI Guidance Notes based on The
Petroleum Act 1998, and covers the areas shown above.

Approach to be taken:
■ Based on individual circumstances
■ All feasible options to be considered
■ Removal to have no effect on environment
■ If left in place (decommissioned in-situ), decision based on rate of deterioration
and possible future effect on marine environment
■ Consider other users of the sea

Lines can be left in place:

■ If adequately trenched
■ If likely to self-bury
■ If exposed sections retrenched
■ If trunk lines

Lines that should be removed:

■ Small diameter of up to 323.8 mm (<12¾ inch)
■ Rigid lines and flexibles that are not trenched or buried

Monitoring:
■ If decommissioned in-situ, monitoring programme to be established on case-by-
case basis
602 Fundamentals of pipeline engineering

CURRENT STATUS FOR

PLATFORMS

 Platforms
 Structures under 10 000 tonnes removed
 Remaining 34 structures case-by-case – derogation?
 112 000 tonne Maureen removed 2001
 Gravity structure with base storage tanks
 Refloated and towed to fjord
 No buyer found
 Now scrapped

The OSPAR meeting agreed that the smaller platforms should be removed, and that
larger structures should be evaluated.

In 2001, the 112 000 tonne Maureen gravity platform with a height of 241 m (790 ft) was
refloated and towed to a Norwegian fjord for disposal.

Originally, it was hoped to resell the unit intact, but there were no buyers. It has been
scrapped.

There are currently plans to remove a number of steel jackets from the North Sea,
whereas, for some of the early concrete structures this is extremely difficult and they are
being left in place.
Decommissioning 603

ALLSEAS PIETER SCHELTE

 New build for end 2010

 48 000 tonne topside
 25 000 tonne jacket
 S-lay capability

Allseas is constructing a new-build heavy lift vessel to remove topsides and jacket legs in
two operations (as on right). It will be commissioned towards the end of 2010 to meet
the expected boom in North Sea decommissioning work. It may be thought of as a
linked catamaran that can dock on either side of the platform. It is designed to handle
topsides lifts up to 48 000 tonnef (105 822 kip) and jacket lifts over half this, operating in
significant wave heights of 3 m (10 ft).

The photograph is of a model showing the original concept of adapting a pair of tankers.
The new-build has a strengthened central section which limits the transit speed to 11
knots. The slot and eight lifting beams are retained. The tilting beams are designed to
accommodate any wind and wave movement.

However, it can also accommodate a full curve S-lay stinger to lay pipe vertically. Its
1361 tonnef tensioning capacity will far outstrip anything currently available in the global
fleet. The 170 m (558 ft) stinger and seven welding stations will be able to lay up to 1524
mm (60 in) pipe.

Vessel length x breadth : 360 m (1181 ft) x 117 m (384 ft)

Slot width x length : 52 m (171 ft) x 122.5 m (402 ft)
Dynamic positioning : DP with twelve thrusters of 86 MW (115 330 hp) total

Other heavy lift vessels are available, such as Saipem’s S7000 and Heerema’s Thialf and
Balder. They may be used in combination with submersible transport vessels such as the
Mighty Servant 3.
604 Fundamentals of pipeline engineering

LEGISLATION - SUMMARY

 Governments provide legislation to give

guidance on decommissioning
 Most structures assessed uniquely
 Not all equipment must be removed
 Decommission in-situ if no detrimental effects to
environment
 May require monitoring program

Any questions?

Governments have introduced legislation to cover the decommissioning of offshore

structures and subsea pipelines. The legislation gives guidance on considerations for
decommissioning, but most substantial structures will be assessed individually for their
unique requirements.

We need to be aware that not all offshore equipment requires removing from the
installed location. Decommissioning in-situ may be acceptable if no detrimental affect to
the environment can be proven. To ensure this, a monitoring program may need to be
implemented to ensure there is no long term damage to the environment.
Decommissioning 605

DECOMMISSIONING IN-SITU

DECOMMISSIONING /
ABANDONMENT IN-SITU

 Cleaned, sealed and left in-situ

 Do not disturb!
 Leave in a safe state
 Likely to be the preferred option for:
 Pipelines that are already trenched/buried
 Trunk lines in North Sea
 It may involve the following operations
 Cleaning
 Product removal
 Trenching

Decommissioning or abandonment in-situ is the preferred option for pipelines that are
already trenched or buried and North Sea trunk lines. The terms differ in UK and US
with ‘abandonment’ being preferred in America.

There is less disturbance of the seabed by this method, especially where the seabed takes
many decades to repair itself. In the soft seabeds of the Gulf of Mexico, this is less of a
concern.

There are three main operations for decommissioning in-situ, these are cleaning, product
removal and trenching. These operations do not necessarily have to occur in that order.
606 Fundamentals of pipeline engineering

Cleaning

CLEANING

 Cleaning required to remove deposits from

pipe wall
 Wax, gas dust, scale, asphaltenes
 Internal LSA (NORM) scale
 Cleaning by running through trains of pigs
 Trains consist of foam, batch, brush and scraper pigs
as required
 Most likely done before water filling (avoid
problems of contaminated water disposal)
 Pigging programme during last months of production

Cleaning is required for the removal of wax, gas dust and scale from the pipe wall.
These may contain heavy metals and other toxins. This operation is achieved by driving
pig trains through the line. The pig trains will include brush pigs or scraper pigs.

LSA (low specific activity) scale forms in well tubulars and contains a relatively low level
of radioactivity per unit mass, which however, is in excess of the background level. LSA
scale is a term used in Europe; elsewhere, it may be referred to as NORM (naturally-
occurring radioactive materials) scale.

Note that on older pipelines with external coal tar epoxy, there is sufficient radioactivity
to be detected should the coating break off and float to shore.
Decommissioning 607

BRUSH PIGS

 Removal of scale, rust, dust (gas) and wax

Brush pigs are used for removing scale, rust, dust and wax from the pipe wall. By-pass
ports are used to produce flow in front of the pig. This flow helps prevent the build-up
of debris or wax in-front of the pig. By-pass ports are typically threaded holes with
plugs. The operator can therefore adjust how much by-pass occurs.

SCRAPER PIGS

 Polyurethane or steel blades or ploughs

 Wax removal
 Disposal of wax-loosening chemicals

Scraper pigs are used for the removal of heavy wax deposits from the pipe wall.

If chemicals are used to loosen the wax first, then there must be tanks to separate and
recover them for safe disposal.
608 Fundamentals of pipeline engineering

Product removal

PRODUCT REMOVAL

 Remaining product displaced by water

 Pig train provides separation
Water Oily water Oil

 Batch pigs
 20 m to 30 m (60 ft to 100 ft)
apart
 Depends on diameter
 Number depends on line
length and wall corrosion

An essential step of decommissioning is to remove the product from the pipeline. This
is achieved by driving the product out with water. A train of batch pigs provides the
separation between the water and the product.

The batch pigs are arranged such that they are pushed by water from behind, displacing
the product ahead of them. They are spaced to allow for some mixing of the liquids:
typically, this is a function of the diameter of the pipeline. The number of pigs in the
train depends on how much will pass each of the seals – a function of the length of the
pipeline and the pitting on the pipe wall.
Decommissioning 609

INHIBITION

 Water contains high doses of

 Corrosion inhibitor
 Oxygen scavenger
 Biocide
 Pipe left in safe condition
 Minimise any further deterioration or rusting

The water contains a number of inhibitors to ensure the continuing integrity of the
pipeline for an indefinite period. Doses are much higher than would be expected for
pre-commissioning, which lasts only a few weeks.

Corrosion inhibitors and oxygen scavengers prevent aerobic and acidic corrosion. The
biocide prevents the growth of sulphate-reducing bacteria (SRB), which in turn will
produce anaerobic corrosion.

Trenching

TRENCHING

 Lower pipeline below surface of seabed

 Backfill
 Methods covered in detail elsewhere
 Why bury now when
not at installation?
 Open trench
 Scour of seabed
 Is wall corroded?
 Reduced strength
 But no longer pressurised
 Spalled concrete in nets
 Crossings at live pipelines
610 Fundamentals of pipeline engineering

The pipe may need to be both lowered into a trench and then buried to leave a smooth
seabed. This is as discussed in the other modules.

However, we need to ask how it is possible to bury the pipeline now when it was not
undertaken earlier during installation. Perhaps the pipe was only trenched and not
buried or some scour has since occurred.

What if there is corrosion of the wall and the steel strength is reduced.? It may not be a
problem since it does not have to withstand operating pressure anymore.

Will deteriorated concrete spall off and end up in fishermen’s nets?

Also, where the pipeline crosses another live line, detailed analysis is required. Can the
line be lowered sufficiently or is it necessary to cut the line?

TRENCHING ISSUES

 Remaining strength of pipeline

 Will it withstand trenching operation?
 Must avoid leakage - corrosion defects and welds
 No longer has high internal pressure
 Loss of concrete weight coating
 Especially at field joints
 Must avoid debris
 Loss of coal-tar epoxy
 Radioactivity
 Jetter arm/machine burial
 Lower stress than plough
 Slower in stiff clays

We need to assess how the pipeline will behave during the lowering process.

Perhaps a less efficient option may be chosen, such as jetting. Ploughs may cause higher
loads to be applied to the pipeline.

Although coal-tar epoxy is not used nowadays for corrosion coating, there are many lines
that used it historically. It may break off during the recovery of a corroded pipeline.
Not only is coal-tar carcinogenic, but it contains natural radioactivity, which can cause
concern should it be washed to shore.
Decommissioning 611

DECOMMISSIONING IN-SITU -
SUMMARY

 Preferred option for trunk lines and

buried/trenched pipelines
 Need to remove product and clean
 Exposed lines may require trenching
 Ensure pipe strength is sufficient to withstand
trenching operation
 Avoid concrete break-up

Any questions?

Decommissioning in-situ is often the preferred method for trunk lines and pipelines that
are already buried or trenched. To decommission in-situ, it will be necessary to ensure
that all product is removed from the pipeline and the pipeline interior cleaned of debris
and harmful deposits. Cleaning can be done with specialist cleaning pigs and batches of
cleaning chemicals during the final months of producing. The final removal of the
product itself is done with pig trains, pushed through by water.

There may be a requirement that the decommissioned pipeline is buried to ensure it does
not become a future obstacle on the seabed. It may be necessary to ensure the strength
of pipe is sufficient to withstand the trenching operation. Care should also be taken to
ensure the concrete coating does not break-up and become debris on the seabed.
612 Fundamentals of pipeline engineering

RECOVERY

RECOVERY

 Recovery preferred option in UK for:

 Small diameter rigid and flexible lines that are not
trenched or buried
 Up to 323.8 mm or even 406.4 mm (12¾ in or 16 in)
 Short flowlines rather than long, large diameter trunklines
 Bundles with integral towhead manifolds
 Recovery presumed in US for all lines
 MMS may permit derogation based on safety case
 In all cases certain lines must be removed:
 Contaminated pipelines (mercury or cadmium)
 Landfalls and inshore lines, subject to local
authority control

Smaller unburied lines are easiest to remove. Even 406.4 mm (16 in) lines (larger than
required by legislation) may be disposed of in this way.

It is becoming the policy to remove all the small diameter and short flowlines within
fields, leaving the main length of large diameter, long trunk (export) lines in place.

Bundles are large diameter structures that sit on the seabed. These are unburied and may
be towed away on-bottom. However, the East Frigg bundle in Norway only had the
manifold ends removed.

Contaminated lines and the landfall section of other lines may require removal by the
appropriate regulatory authorities. For example, the Dutch P15 lines recover oil with
significant amounts of mercury. The developers were given permission on the
understanding that all these lines – some 60 km (37 miles) of both flow and trunk lines –
must be removed as the fields become depleted.

At landfalls, there is often a local authority requirement for the beach section (only) to be
removed. This prevents concerns regarding far-future (50 years plus) erosion of the
shoreline and any protective cliffs or sand dunes.
Decommissioning 613

RECOVERY

 Recovery methods:
 Small diameter rigid or flexible flowlines – reel barge
 Thin walled downhole or flowline tubing - reeled
 Larger diameter pipelines - laybarge
 Pipeline and bundle recovery - cut and lift process or
float and tow method
 Bundle recovery - off-bottom towing method

A number of different approaches have been used, as listed above.

Where the wall and particularly the welds can withstand the stresses, small lines can be
removed by reel barge.

A simple laybarge can be used to dispose of larger diameters.

An alternative used for larger lines in poor condition (unable to withstand stresses) is to
cut into short sections and lift vertically.

If buoyancy can be provided to bundles, then these could be towed to shore using an
off-bottom method. This may be accomplished either with individual buoys or use of
gel to flush the water from the annulus. A similar process may be used for pipelines.
614 Fundamentals of pipeline engineering

RECOVERY BY REEL BARGE

 Reverse lay operation for un-concreted pipe

 Reassessment of weld stresses
 Loss of section – risk if rupture occurs on reel
 Already well established for small diameter
lines and flexibles (Phillips in GoM in 2001)

Phillips recovered a line using reeling techniques. However, care needs to be taken with
the stresses at welds, bearing in mind that the line may have already been yielded during
the lay and now with thinner wall it will be further damaged.

The tensioner has to work in reverse to keep tension on the reel and yet be able to lift
the line up from the seabed.

Ideally, it needs to be undertaken on lines that were initially laid from a reel barge.
However, it may be possible to permit damage on recovery of lines that were laid by
other methods, providing that they are not concrete coated.

The consequence of breakage on the reel can be catastrophic. The welds in particular
may be subject to fatigue fracture.

The pipelines can then be brought ashore for disposal.

This method has been used for a number of lines in the Gulf of Mexico and one in the
North Sea.

Flexibles do not suffer from yielding on recovery and can be spooled up safely.
Decommissioning 615

RECOVERY BY LAYBARGE

 Simpler (cheaper) than normal lay vessel

 J-lay does not require high tension or overbend

This method is intended for the recovery of pipelines that are either exposed on the
seabed or have been unburied with specialist equipment.

Either an existing laybarge or converted workbarge could be used. The J-lay approach is
preferred because it requires less tension on the (possibly corroded) steel pipeline and
there is no overbend on a stinger, which requires the steel to normally be taken to
around 90% of yield during laying operations.

Outline Procedure:
■ Deploy retrieval clamp and recovery wire to pipeline end
■ Recover wire through the A&R winch and secure pipeline with tensioners
■ Move barge with pipeline being recovered through tensioners
■ Cut pipe into pre-determined lengths and store on barge deck
■ Transfer pipe lengths onto supply vessel for disposal/recycling on shore or re-use
after refurbishment

ADVANTAGES
■ Utilisation of existing equipment
■ High speed operation for long lengths

DISADVANTAGES
■ High operating costs
616 Fundamentals of pipeline engineering

CUT AND LIFT

 Removal of three BNFL outfalls in 2005/6

 Demanded by local authority as part of planning
 Severely corroded line breakup
 Coal-tar radioactivity on beach

Lifting stillages onto

transport barge

This method is intended for the recovery of ‘short’ lengths of pipeline or bundle that are
either exposed on the seabed or have been unburied with specialist equipment. A
shallow-draught first-generation workbarge could be utilised.

Outline Procedure:
■ Pipeline cut on seabed into pre-determined lengths
■ Recovery grabs deployed to lift sections out of water onto barge deck
■ Transfer pipe sections onto supply vessel for disposal/recycling onshore or re-use
after refurbishment

ADVANTAGES
■ The pipeline does not require high degree of structural integrity
■ Work could be undertaken inshore in shallow water

DISADVANTAGES
■ Subsea cutting equipment for large diameter lines would need further
development
■ Debris from cutting released into the marine environment
■ Barge operations slow and weather-dependent

A small barge was used by Land & Marine to recover three rain water outfalls, which
were potentially contaminated with low level radioactivity, in 2005. Their removal was a
requirement of original planning consent. Seabed stillages (storage racks) were used to
store the sections prior to lifting onto the barge. Of particular concern was the
condition of one line, which was so severely corroded that there were fist-size holes in
the wall. Additionally, pieces of coal-tar enamel breaking off and floating ashore (with
its natural inherent radioactivity) could give a false reading of unapproved releases.
Decommissioning 617

REFLOAT PIPELINE AND TOW TO

SHORE

 Use of divers to attach buoyancy to line

 In deeper water, pull back onto barge and
relaunch
 Bottom tow or CDT to shore
Converted recovery vessel Pipe-lift davits
Hold-back winch
Sea level
Abandoned pipeline Stingers Flotation bags
Seabed fore and aft
Unburied pipeline
Cut and install end cap Offshore tow tug

Flotation bags
Tow head
Seabed

An alternative would be to attach buoys to long sections of the pipeline – perhaps up to

4 km (2½ miles) and then tow it to shore, where it can be cut up.

In shallower water, divers could undertake this. An alternative would be to pull it up

onto a specially-designed barge, where buoyancy could be added. A pair of stingers
would be required to be fitted to the specially-designed laybarge for support. It is likely
that a flat-bottomed anchor barge could be adapted.

Once ashore, the steel and concrete could be recovered.

BUNDLE RECOVERY

 Bundle recovery by off-bottom towing

 Use ballast to raise bundle off the seabed
 Fill bundle with air
 Use ballast chains to stabilise
 Tow to an onshore facility
 Advantage
 No offshore cutting of pipeline
 Disadvantages
 Depends on structural integrity of bundle
 Extensive preparatory work
618 Fundamentals of pipeline engineering

This method is intended to recover continuous lengths of bundles that are lying on the
seabed with minimal rock dump protection.

The bundle and towheads are prepared and raised clear of the seabed for towing to an
onshore bundle fabrication area in the on or off-bottom mode.

Outline Procedure:
■ Remove flowline spools and fit pig launchers and receivers
■ Fit carrier pipe clamps with fill and drain valves at suitable points
■ Check integrity of ballast chains and replace as necessary
■ Purge carrier with air to displace inhibited water
■ Purge flowlines to raise bundle clear of seabed (fit any supplementary buoyancy)
■ Connect tow and trail tugs
■ Shorten rigging to raise towheads clear of seabed
■ Tow to fabrication area along pre-surveyed route with protected pipeline crossing
points
■ Recover on fabrication area for disposal/recycling or re-use after refurbishment

ADVANTAGES
■ No offshore cutting of pipeline
■ Less dependency on specialist vessels

DISADVANTAGES
■ Extent of preparatory work before tow can commence
■ Probable use of divers
■ Dependent upon structural integrity of bundle
■ Need to protect third party crossings

RECOVERY - SUMMARY

 Recover the following

 Small diameter rigid and flexible pipes
 Downhole and flowline tubing
 Bundles and contaminated pipelines
 Landfalls if required by local authority
 Methods of recovery
 Reeling or lay-barge recovery (pipe lay in reverse)
 Cut and lift or refloat and tow
 Off-bottom tow for bundles
Any questions?

Recovery of pipelines is a viable options for the above listed pipeline types. The
methods of recovery are detailed above and are specific to the type of pipeline being
recovered. The method available will be dependent on the type of pipeline and we have
examined the appropriate methods for different pipeline types.
Decommissioning 619

RE-USE

RE-USE

 Re-use feasible for:

 Flexibles
 Thin-walled downhole or flowline tubing
 Bundles
 Rigid flowlines
 Consider methods in following slides

 Although feasible, it is unlikely that reuse

of rigid lines will be commercially viable

Some pipeline re-use is possible. We consider how this may be done in the following
slides.

However, with corrosion issues on rigid lines, and the difficulty of recovering without
overstressing, reuse of these is unlikely to be economic.
620 Fundamentals of pipeline engineering

FLEXIBLES

 Re-use of flexibles
 Common offshore Brazil
 End lifted and re-reeled
 Cleaned, inspected and overhauled prior to
re-use
 End fittings usually replaced
 Australian QA / insurance recertification concerns

Any damage tends to occur near the end fittings so that is why these are replaced. The
whole length is fully inspected and any repairs carried out.

Brazil has a stockpile of flexibles, stored in a lake (out of sunlight), that are ready for
reuse.

Nevertheless, in Australia, reused flexibles cannot get insurance.

BUNDLES

 Bundle dewatered
 Some chains released by ROV/diver
 Towed to new location using off-bottom
method
 Reflooded and weight mattresses added

 Goosander field, north of Kittiwake

 Installed by Subsea 7 for Venture
 Reusable bundle 12.3 km (7½ miles) in two lengths
 Production flowline, water injection, gas lift
with chemical injection and control lines

There is the possibility of a bundle being reused. However, this would require a very
similar configuration (both length and flowline cross-section) for the second field.
Decommissioning 621

It would need to be refloated by purging the flowlines with gas. The annulus could be
dewatered using a gel slug, which could pass the pipeline spacers.

If additional weight needs removal, then an ROV or divers could cut the straps and
release some of the chains.

When it has been towed to the new location in the on-bottom configuration, the weight
would be restored by reflooding and the addition of weight blocks or concrete
mattresses.

At least one reusable bundle has recently been installed in the North Sea. The
Goosander field, some 12 km (7½ miles) northwest of Kittiwake (in block 21/12 of UK
sector of the North Sea) is being developed by Venture using a reusable pipeline bundle,
fabricated in two main sections, 4.9 km (3 miles) and 7.4 km (4½ miles) long, both of
which were engineered and installed by Subsea 7.

The bundles each comprise a 219.1 mm (8⅝ in) production flowline, a 168.3 mm (6⅝
in) water injection flowline, a 101.6 mm (4 in) gas lift line, and a chemical
injection/control line. The system has also been designed to allow other fields to flow
through the infrastructures.

Currently, the bundle is still producing at the original field.

RIGID FLOWLINES

 Bottom tow to near-by location

 Re-reel
 Issue of total plastic strain
 Weld fatigue
 Serious issues with weld quality after internal corrosion
 JIP currently underway at The Welding Institute
 Unlikely to be financially viable

Bottom tow can also be used to relocate rigid pipelines at a new route nearby.

Although flowlines can be recovered onto a reel barge, there are some concerns with the
condition of the welds when subject to additional cumulative yielding again during
recovery and reinstallation. This is particularly the case when considerations of
corrosion are allowed for.

Neither is likely to be economic. Which operator wants to start production with all the
damage from a previous operation?
622 Fundamentals of pipeline engineering

RE-USE - SUMMARY

 Some pipeline types can be re-used

 Flexibles, tubing, bundles, rigid flowlines
 May need to carry out some or all of following
operations
 De-water, clean, inspect, recondition, replace end
fittings
 Relocating method depends on pipeline type
 Lift, bottom tow, off-bottom tow, re-reel

Any questions?

Some pipeline types can be re-used at other locations. Re-use may have been an initial
requirement for which the pipeline was initially designed or the pipeline is found to be in
a suitable condition for re-use at the end of its service life.

If re-using the pipeline, then the above operations may need to be carried out before the
pipeline can be moved and reinstalled. The required operations for the type of pipeline
have been detailed in this section. Also detailed have been the methods available for
each pipeline type.
Decommissioning 623

COSTS

ENVIRONMENTAL AND FINANCIAL

COSTS

 In all cases, cleaning is required

 Burial contracts placed ‘at-cost’
 Limited data to date for HAZOP/SAFOP
 Qualitative rather than quantitative assessments
 All lift and recovery has safety issues
 Outweighs leave in-situ by a factor of ten
 Risks to divers
 Steel and concrete reclamation
not cost-effective
 Anodes depleted

For all options, it is necessary to first clean the pipeline.

Burial contracts for those pipelines that require reburial or deeper burial have been let in
Norway to CTC. These have been bid on the basis that work will only be carried out if
the vessels and trenching equipment have no other work. The rates are quoted per
kilometre buried, and are effectively charged at minimal costs to keep the equipment
busy in off seasons. If other commitments do not allow any work in a particular year, it
can be passed over to the following quiet period. Pipelines up to 1524 mm (60 in) are
trenched on a one-pass ‘reasonable endeavours’ basis.

Because there have been very limited removal operations to date, the hazard or safety
assessments are necessarily of a qualitative nature.

The salvage costs for the materials do not nearly make the operations cost-effective.
With the exception of removal of anodes – which can cause metal pollution of the
seabed – there are little environmental benefits either.

Where:
■ HAZOP = hazard and operability (analysis study)
624 Fundamentals of pipeline engineering

■ SAFOP = safe operations (analysis)

LINE REMOVAL CHECKSHEET

 Environmental issues
 Soil disturbance – seabed modification issues
 Coral – primary or secondary growth (on pipe)
 Disposal of steel
 Disposal of concrete and other coatings
 Disposal/storage of residual contents
 Health and safety issues
 Divers hours at depth
 Handling of pipeline offshore and onshore
 Radioactivity of residual product
 Proximity of live adjacent assets or adjacent pipelines

This and the following slides show some of the issues that must be considered when
carrying out an assessment and selecting the best removal method.

When using this list to determine weightings, remember it all depends on your
perspective! Thus, it is recommended that wide advance stakeholder engagement helps.

Cost estimates should be undertaken for each option.

Decommissioning 625

LINE REMOVAL CHECKSHEET

 Pipeline criteria
 Burial condition – additional costs for unburial
 Diameter – larger lines more expensive
 Overcrossings – cut 30 m (100 ft) either side
 Cannot reverse lay – requires cut-and-lift options
 Heavy corrosion or build-up prevents cleaning
 Injury to personnel hot cutting or spillages to sea
 Piggablity – may require fitting of temporary pig traps
 Line integrity for removal – strong enough to lift
 Water depth – use of ROVs or divers to recover
 Removal of mattresses and crossings
 Re-qualification of lifting points

LINE REMOVAL CHECKSHEET

 Future liability
 Fishing – snagging of nets on disturbed spoil
 Tourism – diving on coral areas
 Other operators – future use of area
 Leaks from uncleaned lines left in place
 More cuts required for cut-and-lift rather than reverse lay
 Reputation / stakeholder views – rubbish left behind
 Possible benefit to fishing – providing shelter
 Future management/monitoring – none, if removed
 Remaining life – remove whole infrastructure at EOFL
 Future lines – require more crossings
 Use of local contractors – may need new expertise
 Precedence – may set standard for line removal

Where:
■ EOFL = end of field life
626 Fundamentals of pipeline engineering

MMS COST ESTIMATES

 Gulf of Mexico – local equipment

 Total of over 4320 km (2683 miles)
 62% in range 101.6 mm to 323.8 mm (4 in to 12¾ in)
 3.2% over 762.0 mm (30 in)
 Shallow water depths, 60 m to 150 m (200 ft to 500 ft)
 Initially concentrating on nearshore, and
areas where with shrimp trawler interaction is likely
 Typical pipeline length 6.5 km (4 miles)
 Does not include onshore costs
 Crossings and other difficult sections excluded

The above cost estimates are taken from MMS paper 32.701.001/R1 of 2004.

They refer to the Gulf of Mexico and assume that a suitable barge could be adapted
within two days mobilisation/demobilisation. For other areas of the US – fields such as
Alaska or California, where there are much shorter lengths of pipeline (>2% of total) to
be recovered – sailing costs become significant for specialist equipment and expertise.

In order to limit the study, the nearshore and shallow water sections have been costed,
where third party injury are the most likely.

The assessment has been limited to lengths of pipelines allowing a continuity of work for
the vessel, averaging typical flowline and export line lengths.

None of the onshore disposal costs have been included and sections have also been
excluded where recovery conditions are difficult.
Decommissioning 627

COSTS

 Budget costs for trenching

 $31 000/km ($50 000/mile)
 Vessel costs for removal
 $90 000 to $190 000 per day for shallow water
 Inshore out to 60 m (200 ft)
 $190 000 to $210 000 further offshore
 Depths over 150 m (500 ft)
 Removal costs over range of diameters
 $33 000/km ($53 000/mile) <101.6 mm (<4 in)
 $72 000/km ($116 000/mile) >762.0 mm (>30 in)
 Costs may reduce once knowledge is gained

The costs may reduce as expertise and specialist equipment becomes available.

COSTS - SUMMARY

 Checklists
 Environmental, health and safety
 Pipeline criteria and future liabilities
 Cost estimates for GoM
 Burial
 Removal
 Safety issues greater with removal than for
abandonment in situ

Any questions?

Some idea of the cost of removal of pipelines is provided by MMS for the GoM region.

In all cases, it is more hazardous to remove pipelines than to abandon in situ.

628 Fundamentals of pipeline engineering

DECOMMISSIONING AND
ABANDONMENT - SUMMARY

 Principal factors influencing available

method of decommissioning
 Latest legislation and current thinking
 Different approach between Europe and US
 Operations required for decommissioning
pipelines in-situ
 Methods available for pipeline recovery
 Potential for re-using some pipeline systems
 Checklist of issues and costs
Any questions?

An overview has been given of the processes for decommissioning pipelines and other
offshore components. The principal factors that influence the methods of
decommissioning were identified: these being, environmental and safety concerns, public
opinion, political needs and finally cost effectiveness.

The latest legislation and current thinking regarding the correct decommissioning
strategy was discussed. Decommissioning of pipelines in-situ is examined in detail and
the required operations were detailed.

Also, the methods available for the possible recovery or re-use of certain pipeline
systems were discussed.
Flow assurance
Flow assurance 631

EXPECTATION

EXPECTATION

 Ensure pipeline operates, and continues to

operate, in the intended manner
 Daily, weekly and monthly operations
 Optimise flow throughput rate
 Minimise internal corrosion and erosion
 Tasks to ensure safe working
 Maintaining flow within design envelope
 Pigging
 Removal of water, sand, hydrates and wax
 Different types of pigs and their functions
 Additives to pipelines to enhance operations

Flow assurance is the skill of optimising the throughput of oil and gas through the
pipeline, whilst reducing as much as possible the loss of wall thickness through corrosion
and erosion.

This is done by injecting additives and pigging on a very frequent regular basis to remove
unwanted deposits in the line. Controlling the flow in a pipeline within the safe working
design envelope also helps prevent internal damage to the pipe walls.

We will introduce the activities required for the safe operation of subsea pipelines. By
this, we mean work carried out on a frequent regular basis (rather than annual
inspections).

Some of these regular operations will involve additives to the product in order to
enhance the flow. This may be done as a continuous process or in a batch.

A description is given for the various types of pigs and their functions.

The Inspection module covers the use of intelligent pigs.

632 Fundamentals of pipeline engineering

OPERATIONAL CONTROLS

CONTROL THROUGHPUT AND

CONDITION

 Maintain designed flow

 Flow characteristics (slugging)
 Fluid properties
(minimum/maximum temperatures and pressures)
 Avoiding transients
 Whole life of pipeline (conditions at start and end)
 Controlling
 Sand and wax deposition or hydrate formation
 Viscosity, foaming or emulsions
 Minimise
 Corrosion and erosion

There is a range of conditions that must be controlled.

These are particularly associated with the flow conditions and rates. As discussed later,
problems can arise in the pipeline if the flow conditions are not carefully controlled.

The resulting problems that may arise are due to sand, wax or hydrates. These can block
the line. If temperatures fall, the fluid may be too viscous to pump efficiently. Other
problems are the gas-oil mixture may become foamy or the product mix may form
inseparable emulsions.

Corrosion and erosion are also controlled by the way the pipeline is operated.
Flow assurance 633

CRITICAL POINTS IN SYSTEM –

MAIN FLOW ASSURANCE ISSUES

 Hydrates – 100% of projects

L
 Maintain heat, additives / depressurise S

 Liquid hold-up/slugs – 90% A

 Keep in stable region and choking A

 Wax – 50% (oil) W

 Maintain heat and pig

L

H
Choke I

 Scale – 30% (naphthenate) H H

Safety
 Injection / acid valve H
L L

 Asphaltenes – 10% to 20% (oil) W

S
 Chemical inhibition A

 Ice – <5% S
S L
 Inhibition A

Produced products can cause blockages at certain points in the system. These include
dips and bends in the line and valves, such as the SCSSV and choke at the wellhead tree.
The initial letters indicate where risks are greatest.

Hydrate blockages can form due to cooling at valves or as the flow passes through water
gathered at dips. This is the main driver for almost all projects. It can be prevented by
maintaining heat within the oil phase and injecting chemicals. Once formed, the system
should be depressurised to allow the hydrate to melt.

Liquid slugs problems can arise where the flow changes direction, such as at the base of
the upright section of the well, at dips in the flowline or at the base and top of the riser.
Condensates can also drop out as the pressure changes, such as after the choke. They
are an issue for most projects. Pressure variations at the riser due to slugging may affect
the well. Slugs can be controlled by operating in the stable region and by choking as
required. A slug suppression system (S³) can be fitted. This uses fast-acting valves on an
extra separator to cut out severe slugging.

Wax may be deposited after valves in oil wells as well as along the flowline. About half
of developments require control of wax. It is not normally an issue with gas production.
The solution is to maintain heat and pig regularly to avoid a major build-up.

Scale (BaSO4) is an issue with both oil and gas lines. Scale is an issue in about a third of
developments. Its control is by chemical injection. Once formed, it requires acid
treatment to remove and recover the flow. Injected water can block the pores of the
reservoir with scale. A similar approach is needed for naphthenate soap scale, which
forms at the oil-water interface.

Asphaltenes may be deposited out of the oil in vertical sections of the tubular and riser.
Asphaltenes must be controlled in around 10% to 20% of developments. It is not a
problem with gas production. Continuous chemical injection will limit its deposition.

Ice may form in the flowline in the coldest of seas. For example, the seabed temperature
at Ormen Lange is around -2 ºC (28 ºF). It can also form as hydrate plugs are
depressurised, releasing the methane. Water is initially produced from vapour within the
gas phase (typically, around 3% of total), then from the formation water (between 30%
634 Fundamentals of pipeline engineering

and 90%), and then from any water-injection system that has been installed. Continuous
chemical inhibition is very reliable. It is very difficult to recover once formed.

Where:
■ BaSO4 = barium sulphate
■ SCSSV = surface-controlled sub-surface safety valve

HYDRATES

 Methane and water compounds

 Resembles snow or ice in gas lines
 High pressure and low temperatures
 Forms plugs downstream of valves – hysteresis
 Joule-Thomson cooling
120
Hydrate formation – Hyde gas (SNS)
1.5
100
Pressure bara (ksi)

Hydrates
80
1.0
60
tion
ma
40 For
n 0.5
iatio
ssoc Safe operation
20 Disa (no hydrates)
0 0
32 41 50 59 68
0 5 10 15 20
Temperature C (°F)

Hydrates are a problem in multi-phase flow and wet gas flow, where water is present.
They are formed under conditions of low temperature and high pressure.

They are subject to hysteresis, which means in practice that the temperature at which the
hydrate melts (disassociates) is higher than that at which it forms. Their formation
depends upon the conditions they have been subjected to upstream. Hydrates have a
tendency to form downstream of valves because of the Joule-Thomson cooling effect at
a pressure reduction. It should be noted that hydrate plugs can form in minutes but may
take weeks to reduce the pressure and allow them to slowly disassociate at the ambient
water temperature.

A hydrate plug can block the line. This presents both an operational and a safety
problem. The pressure differential across that plug will increase. It may then shift at
high pressure and travel along the line at high velocity. It can therefore cause damage
when it reaches a bend or some equipment.

Control of hydrate formation is by the control of the operating pressure and temperature
of the pipeline, and by injection of a hydrate inhibitor, which shifts the formation curve
to the left.

The graph shows the safe operating zone to the right of the dotted line for the Hyde gas
line in the Southern North Sea. The solid line shows the temperature of formation but it
will take a higher temperature (dotted) for the hydrates to melt. Inhibitors move the
critical formation and melting lines to the left.
Flow assurance 635

WAX

 Long-chain paraffins
 Condenses on pipe wall – restricting flow
 Keep up flow
rate and heat
 Use of wax
suppressants

Wax deposition can be a problem for oil lines. As the oil cools the long-chain paraffins
in the oil can form a waxy deposit on the pipeline walls. This restricts the flow, increases
the pressure drop down the line and can be very costly. Wax formation is restricted by
maintaining a higher fluid temperature, which can be assisted by keeping up the flow-
rate.

CORROSION

 Corrosion inhibition
 Corrosion inhibitor carried in liquid phase
 Stratified flow – only bottom of pipe is protected
 Pitting corrosion in upper part of pipe becomes a problem
 Water phase drop-out
 Low flow velocity – water drops out Gas
 Water can be highly acidic
 Increased corrosion at
Oil
bottom of pipe – tramlines
 Shut-downs can lead to
water accumulation at low points Brine

As corrosion inhibitor is carried in the liquid phase, we may not get adequate protection
in stratified flow.
636 Fundamentals of pipeline engineering

Water drop out in stratified flow can cause localised corrosion. The water can be highly
acidic, leading to rapid corrosion. At the interface between oil and the water, selective
attack can lead to the formation of tramline grooves in the wall.

EROSION

 Multiphase flowlines
 Fluid contains sand and water
 Erosion at bends and valves – sandblasting
 Oil trunk lines
 Cavitation collapse
Flow
Weld Corrosion
bead pit Erosion
Erosion

 Erosional velocity 122 100

Ve  Ve 
 
 For oil  = 800 kg/m3, (50 lb/ft3), Ve = 4.3 m/s (14.1 ft/s)

It is perhaps easy to understand that flowlines containing multiphase fluids direct from
the well may contain sand and water, in addition to the oil and gas. This can result in a
sandblasting effect on the pipe wall, causing erosion of the steel as the flow changes
direction at bends.

However, erosion can also occur in oil trunk lines to shore, where there are no abrasive
elements in the flow. If the velocity of oil itself is too great, then any small bump or
cavity in the wall may cause cavitation bubbles to form. When the bubbles collapse or
implode, the resulting shockwaves can also erode the wall.

This effect is often seen downstream of a girth weld, where the bead or root protrudes
out of the boundary layer of fluid attached to the wall, causing disruption to the flow.

The photograph shows the effect of a field weld root bead and the lines of erosion it
causes downstream.

API RP 14E gives the above formula for the velocity at which erosion may start to
occur.

Where:
■ Ve = erosional velocity
■  = density of liquid

Different formula are needed for SI and metric because of the units in the numerator
constant.

Where corrosion has left a pit, then this can also disrupt the flow and erode the wall on
the downstream side, increasing the length of the defect.
Flow assurance 637

TURN-DOWN – OPERATIONAL
ENVELOPE

 Reservoir becomes depleted

 Flow parameters change throughout life
 Water cut increases
 Temperature in gas lines may increase
 Axial expansive forces – at end or uplift/lateral buckling
 Temperature in oil lines may decrease
 Insulation designed for end of life – may have crushed
 Wax problems – more frequent pigging
 Flow rates decrease – colder delivery
 Gas/oil ratio alters
 Operational changes
 Chemical additives may be altered – new supplier
 Water and gas injection

Because we are abstracting oil and gas from the reservoir, the flow parameters through
the pipeline will alter throughout life.

As the field depletes, the amount of produced water may increase significantly. The
thermal capacity of water is often much greater than the product – particularly for gas
lines. This water can be very hot, and as these slugs pass through the pipeline, the
expansion of the pipeline may increase – this could cause uplift or lateral buckling
problems where there was no risk of this before.

In oil lines, the temperatures tend to decrease over time causing waxing which may
require more frequent pigging.

In general, flow rates will decrease causing additional cooling along the length of the
pipeline.

For mixed lines, the ratio of gas to oil may alter. This is dependent on the depth of the
oil well within the reservoir.

Operationally, the chemicals added to the line may alter as suppliers are changed, or the
regime for injection of water or gas may be varied over time to improve the reservoir
recovery.

It is necessary to account for all of these changes whilst maintaining the line within the
operational envelope.
638 Fundamentals of pipeline engineering

OPERATIONAL CONTROLS –
SUMMARY

 Maintain flow within envelope

 Temperature and pressure
 Deleterious effects
 Slugs, surge
 Wax, hydrate blockages
 Increased viscosity, emulsions
 Corrosion, erosion
 Turn-down
 Methods to maintain flow within envelope
Any questions?

We have looked at what we mean by routine operations.

They are the means of ensuring that the flow is maintained within the design envelope.
This generally means controlling the temperature and pressure in the line.

If the product strays out of this, then some of the effects may be the formation of slugs
or pressure surges.

We have looked at some of the effects when temperatures fall – such as the formation of
wax and hydrates, the increase in friction due to high viscosity or formation of
emulsions, which cannot be separated.

The pipeline itself can be affected by corrosion and erosion.

Flow assurance 639

ADDITIVES

ADDITIVE APPLICATION

 Continuous  Batching
 Hydrate inhibitor  Corrosion inhibitor
 Corrosion inhibitor  Periodic coating of pipe wall
(say 3 monthly)
 Erosion/corrosion
 Used in wet gas lines or from
control unmanned minimum facilities
 Wax suppressants  Gives good covering over all
 Drag reducing of wall
agents  Methods
 Biocide  Introduce slug of additive
between pigs
 Spray pig

A range of chemicals can be added to the flow as part of the pipeline operation. There
are two main methods of introducing additives to the pipeline – continuous injection
and batching.

Continuous injection involves, as the name implies, the continuous pumping of an

additive into the product stream. The types of additive commonly used in this way are
listed above.

The alternative means of introducing additives to the pipeline is to periodically send

slugs of additive into the line between pigs. This method is used for corrosion inhibitors
only, because they are able to coat the pipe wall rather than modify the behaviour of the
fluid.
640 Fundamentals of pipeline engineering

BATCH INJECTION OF INHIBITOR

SLUG

 Logistics
 Inhibitor and carrier fluid
 Pigs, launchers, storage tanks and pumps
 Personnel
 Supply of carrier fluid (diesel)
 Disrupt normal operations
 Receiving slug
 Ensure large enough separator and slug catcher

Leading pig Slug of carrier fluid and corrosion inhibitor Trail pig

For batch injection of corrosion inhibitor, there is a number of requirements. These

include provision of the pigs to contain the slug of inhibitor, the pumps and personnel
to carry out the work. The means of supply of the carrier fluid to the end of the pipeline
must be considered.

This contrasts sharply with the automatic continuous injection method where only the
inhibitor itself is needed.

Normal product delivery operations must be disrupted during the batch treatment
process.

The slug must be caught at the far end of the pipeline and the carrier fluid decanted out
from the product.
Flow assurance 641

INHIBITOR SPRAY PIG

 Inhibitor carried within pig

 Sprayed onto walls at pressure
 Proprietary system
 Therefore not ‘off-the-shelf’

The alternative method of batch inhibiting is the use of the inhibitor spray pig. This is a
proprietary product and would have to be designed and built for a specific application.

ADDITIVES – SUMMARY

 Additives used to improve flow and reduce

corrosion
 Inject continuously or in batch
for corrosion prevention

Any questions?

We can introduce additives into the pipeline to improve the flow characteristics and
reduce the possibility of corrosion. To improve flow, additives can reduce drag and
suppress wax formation. To reduce corrosion, there are inhibiting chemicals.

The injection of additives can be either a continuous process or anti-corrosion agents

can be delivered in batched slugs.
642 Fundamentals of pipeline engineering

PIGGING

WHAT ARE PIGS?

 Pigs are devices driven by the pipeline fluid

 Low differential pressure
 Bleeding past rubbers
 Surge and stopping in gas lines
 Bypass used to slow passage
Metal or plastic body

Bypass
Flow

Pipe wall
Suspension rubber cups

Pigs are devices driven through the pipeline by the pipeline fluid. They come in all
shapes and sizes and perform a number of functions, as explained below.

In essence, they contain the components shown in the diagram above. There are cups to
seal against the pipeline wall, giving the device its driving force, and a body upon which
the cups are mounted, which contains brushes, gauge plates or other devices that give
the pig its function.

Only a small pressure differential is usually needed to propel a simple pig, around 0.5 bar
(7.5 psi), depending on the diameter and condition of the line.

Normally there is some bleeding of the line contents past the rubber disks in either
direction as they deform to follow the contours of the wall.

In gas lines, pigs will tend to repeatedly surge forward and stop if they are not carefully
controlled.

Some pigs are built with a bypass tube to slow them down and flush debris out within
the upstream product.
Flow assurance 643

PIG LAUNCHER

Vent valve D
Mainline trap
and gauge
Door valve B Pig signaller
Pig
Launch Flow
Trap Mainline
tray kicker bypass
valve C valve A

Flow
Under normal operation, valves A, B and C are left open and the pig launcher door is
kept closed. When a pig is to be launched, the valves B and C are closed and the vent
valve D is used to release gas pressure. The door is opened and the pig pushed into the
trap. Valve D is shut again. The door is resealed and valve C cracked open again until
the trap pressure equalises with that of the pipeline. Valve C is closed and valve B
opened. The pig can be launched by opening valve C again and then gradually closing
valve A. A pig signaller indicates passage of the pig. Once the pig is in the main
pipeline, valves A, B and C are fully opened again for normal operation.

For pig receipt, a similar unit is used except that the pig signaller is on the other side of
valve B to indicate that the pig has been caught.

The photographs show a typical landline pig launcher and a subsea unit supplied by
Pipeline Engineering (www.pipelineengineering.co.uk) to installation contractor Subsea
7. This operates at a depth of 130 m (426 ft) in Esso’s Jotun field in the North Sea
(Norwegian sector).

It is a 168.3 mm to 273.1 mm (6⅝ in to 10¾ in) subsea class 1500 vertical pig launcher
with receiver facilities. The unit can launch or receive both conventional and intelligent
pigs. It has full subsea capabilities including a soft landing system and ROV operations
compatibility.

The unit consists of three sections: a manifold interface, the protection head and pig
launcher. The manifold section is bolted to the subsea manifold and includes three
pedestals for the Soft Landing System. When the launcher/receiver is not in use, the
protection head is used to protect the manifold from damage and corrosion. It is fitted
with hydraulic quick connect/disconnect collet connectors and a control panel to allow
removal and connection of the head by ROV.

The pig launcher assembly, which is kept either onshore or on the barge until required, is
fitted with three pig release fingers and three baskets capable of launching and receiving
three conventional pigs or one intelligent pig respectively. The pig launcher and receiver
is designed in accordance with PD 5500, permanent pipework to Det Norske Veritas
(DNV), and the launcher/receiver structure to DNV and NORSOK.
644 Fundamentals of pipeline engineering

JOBS DONE BY PIGS

 Cleaning wax and solids

 Sweeping out liquids
 Corrosion inhibition
 Proving bore is clear

Pig designed for very fine

cleaning of aviation fuel line
Note use of sailcloth disks
between wire bristles

Most pipelines are pigged at least occasionally, if not routinely, for the reasons listed on
the slide above.

■ Cleaning out waxes and solids may be necessary on an oil pipeline where wax is
deposited on the wall. Deposition of wax can dramatically increase the pressure
necessary to get the flow through the pipeline. Where it is not possible to insulate
the pipeline sufficiently to avoid wax deposition, regular pigging to remove the
wax is often the solution. This would apply particularly to oil trunk lines.
■ Sweeping out liquids tends to be done by spheres. These are usually polyurethane
balls pressurised with glycol to a diameter a few percent above the pipeline
internal diameter. Taking a gas trunk line as an example, it might accumulate
condensates that would need to be swept out on a regular basis. This would keep
the level of liquids in the line under control and would avoid the occasional arrival
of a very large slug of liquids at the terminal, an event that might cause process
problems.
■ Corrosion inhibitor pigs can be used to introduce a slug of inhibitor into a line
with the objective of coating the entire pipeline inside wall with corrosion
inhibitor.
■ Proving the bore is clear is carried out during hydrotesting at the pre-
commissioning stage. It may also be used when a dent is suspected.
Flow assurance 645

SPHERES

 Solid (smaller diameters) or inflated

 Sweeping liquids from lines
 Spreads inhibitor onto wall circumference
 Automatic launch for frequent basis

Spheres are frequently used in wet gas and multi-phase lines on a regular basis. This
might even be more frequent than once a day.

They are used to sweep liquids from lines to prevent the build up of slugs. They can
spread a corrosion inhibitor around the full circumference of the wall to prevent it
collecting at the bottom of the pipe.

Some platforms have an automatic launch system with a cartridge containing half a
dozen spheres. These can be individually launched with little or no manual input.

They are usually slightly oversized (2%) to give a good seal with the inside bore of the
pipe.
646 Fundamentals of pipeline engineering

FOAM PIGS

 Polyurethane foam body

 Usually polyurethane coated
 May incorporate ribs or brush bands

Picture courtesy PII Kershaw

Foam pigs are generally bullet-shaped, moulded from open cell polyurethane foam and
usually with an external PU coating. They come in a variety of shapes, sizes and colours.

They have no independent sealing elements but are compressed axially by the pressure
differential, which gives sufficient radial expansion to form a seal between the pig body
and pipe wall.

FOAM PIGS

 Sweeping liquids and solids

 Light cleaning duties
 Flexible but not too durable
 Can get heavy duty versions
 If stuck, can sometimes be broken up with
increased pressure – but don’t rely on it !

Foam pigs are a general-purpose but lightweight pig suitable for sweeping liquids and
solids from a pipeline. Specific applications can be ‘built-in’ at the moulding stage where
Flow assurance 647

gritted bands, brushes, jetting holes, magnet inserts (for tracking) or studs (for fixing of
gauge plates or scrapers) can be added.

They have the advantage of being tolerant of tight bends and bore restrictions. They will
also deform or even break-up if they become stuck. This makes them useful for an
initial pig run prior to a more robust brush or scraper pig being used.

Their disadvantage is that they wear out quickly, although heavy duty versions exist for
single pass usage on long lines.

GEL PIGS

 Uses
 Pre-commission of barred tees and unpiggable lines
 Separator or batching
 Debris pick-up or condensate removal
 Biocide laydown or corrosion inhibition
 Dehydrating, dewatering or drying
 Paraffin solvent
 Isolation or stuck pig removal
 Structure PPSA gel
 Water, hydrocarbon or chemical mixtures as base
 Pseudoplastic, cohesive, self-healing and shear thinning
 Disposal
 Chemical breaker to destroy visco-elastic structure
Gel pigs provide additional tools to the operator.

The gel pigs are formed from highly-viscous gelled liquids: their complex rheology
(originally derived from downhole uses) gives them many useful flow characteristics.
Their first use was in Canada in 1971 to remove water from oil flowlines that could not
be pigged conventionally. Thereafter, they have been extensively around the world for
lines that were unsuitable for or were unequipped for rigid pigs.

Their basis is usually water, although hydrocarbons, chemicals, solvents, acids or a

combination of fluids have been used. Their non-Newtonian behaviour, such as their
cohesion and shear thinning, allows self-healing and only a very small amount of bypass
– typically 1%. This is far less than for conventional pigs.

They are typically much longer, with lengths from a few metres to 100 m (330 ft) in
length, depending on usage. In some instances, they are strong enough to be a direct
replacement for a foam pig – though perhaps being transported with a temporary outer
steel sleeve for handling, before being inserted into the pig trap. Other uses require long
slugs of gel pushed forward in front of or between a pair of ordinary disk pigs.

A chemical breaker can be injected into the pig at the far end of the line for disposal.
This causes the visco-elastic structure to be destroyed.

The picture shows a sample of gel from the Pigging Products and Services Association
(www.ppsa-online.com).
648 Fundamentals of pipeline engineering

BATCH PIGS

 Sweeping liquids from lines

 Inhibitor batching
 Sweeping spheres and pigs
entering from side branches
 Gel pigging
 Multiphase lines
 Sand gathered
into gel
 Removal of
stuck pigs

Batch pigs can be solid molded or metal bodied.

Metal bodied pigs utilise polyurethane or rubber seal discs or cups. The material
selection is dependent on the specific application. Metal bodied pigs offer the ability to
add a range of attachments that may be used to perform a range of functions.

They are commonly used for removal of liquids and inserting a batch slug of corrosion
inhibitor.

If smaller diameter spheres or pigs have been used to clean a smaller diameter branch
pipeline, once it enters the main line the larger bore means that there is no differential
pressure to move it. A batch pig can be sent down the main line to sweep it up.

Gel pigging may be carried out to de-sand a multiphase line. The gel picks up solid
debris and removes it from the line. The gel also lubricates the pig and gives improved
drive.

With care, such pigs may be used to remove other pigs that have become stuck in the
line.
Flow assurance 649

BRUSH PIGS

 Removal of solid
debris and wax
 Bypass ports
 Jet of fluid
 Debris suspended
in front
 Swept out
 Slows pig

Brush pigs are used for cleaning the bore.

By-pass ports are used to produce flow in front of the pig. This flow helps prevent the
build-up of debris or wax in-front of the pig. By-pass ports are typically threaded holes
with plugs.

The operator can therefore adjust how much by-pass occurs prior to entry.

SCRAPER PIGS

 Polyurethane or steel
blades or ploughs
 Wax removal

Scraper pigs are used for the removal of wax deposits from the pipe wall.
650 Fundamentals of pipeline engineering

The spring joint at the middle of this example helps it negotiate tight bends.

SPECIALIST PIGS

 Pin-wheel pig
 Magnetic cleaning pig

Picture courtesy PII Kershaw

In addition to the basic cleaning and sweeping operations, pig bodies can be configured
for specialist functions.

The picture above shows a magnetic cleaning pig in use in Germany, where magnets
have been mounted on the body of the pig and have pulled a veritable bird’s nest of
welding rods from the pipeline. Note also the bi-directional sealing discs on the pig and
a gauging disc, which can just be seen behind the welding rods.

Approximately 3000 rods were extracted on the first passage of the pig. This dropped to
1000 rods at the next pass and down to zero after five runs.

These are only two examples of specialist pigs. You can imagine the myriad of shapes,
sizes and functions that have been adapted over the years.
Flow assurance 651

A CAUTIONARY TALE

 Slug formation uphill

Direction of flow

Slugs formed up-hill Dribbling over crest

High head loss No recovery of head

 Small pig for first run but is lost in system

 Big pig purges slugs but gets stuck ¾ along

 Splashes into condensate and water puddles

 Formation of hydrate plug
 Pressure differential melts hydrates
 Pig surges forward – almost hits end of pipe

Jee Limited was approached to resolve a problem with an infield line that was normally
pigged on a regular monthly basis. A pig had become stuck in the launcher and it was
decided to suspend pigging for a number of months.

In the meantime, the pressure needed to maintain flow rose more than tenfold. The
reason for this was traced to the undulating nature of the pipeline: the high pressure loss
was due to slugs of product which needed to be forced uphill. On the downhill slope,
the liquid dribbled over the crests of the hills and very little pressure head was recovered.

The system pressure was brought down and the stuck pig removed from the launcher.

When the operations were resumed, it was decided to use a smaller than normal pig for
the first run. This got lost in the system as the gas flowed past the pig.

The full size pig which followed had the effect of removing some of the slugs, but
puddles of condensate overlying water had formed in a hollow. As the pig hit these, a
spray of water and condensate splashed out and formed a hydrate plug. This effectively
blocked the passage of the pig about three-quarters of the way down the length of the
pipeline.

The pressure behind the pig was increased to 90 bar (1300 psi) and that in front of the
pig was gradually reduced by bleeding off. When the differential reached 50 bar (725
psi), the hydrates became unstable and melted.

This occurred just as the Jee engineer arrived on site. The pig surged forward and
backward like a spring until the pressures equalised on either side.

When their calculations were checked, even though the volume of the pig catcher had
been forgotten, there was just enough distance to prevent the pig hitting the end of the
line.
652 Fundamentals of pipeline engineering

PIGGING – SUMMARY

 Pigs and spheres are devices driven by fluid

 They carry out
 Cleaning
 Gauging
 Batching and
 Specialist functions
 May be used daily, weekly or monthly
 Cost-benefit analysis

Any questions?

Pigs are small components that are driven through the pipeline by the flow of the fluid.
They can perform a wide range of activities to ensure the correct operation of the
pipeline, these activities are summarised above.

For routine operations, these may be used very frequently.

The exact frequency of use will be determined by a cost-benefit analysis.

FLOW ASSURANCE – SUMMARY

 Operations
 Maintain temperature and pressure within envelope
 Chemical injection
 Wax, hydrate and corrosion control
 Improved flow rate
 Continuous or batch
 Pigging

Any questions?

We have introduced routine operations to maintain the pipeline flow within the designed
envelope.
Flow assurance 653

Some improvement to the regime can be achieved by adding corrosion inhibitors or

other chemicals to prevent the formation of wax or hydrates.

Most routine operations are undertaken using pigs or spheres.

Flow assurance field
case
Flow assurance field case 657

Flow Assurance Field Case:

Brazil BC-10
Loek Vreenegoor (GSNL GSUF)

Flow Assurance Field Case – P241 Course

BC-10 Block, Heavy Oil Development

BC-1 0 A ssessment A rea

5km
B M C -25 B E S-100 B ES-200

E SS -116 1 -SHEL-9 -ESS

D iscovery O -n
O orth
-n E-north

B C -10
(C achalote)
1 -SHEL-11 -ESS L
O -w est
L
B -w est
B C -60
E SS -100 C
D iscovery O-south
1-SHEL-12 -ESS
(Jubarte) 1 -SHEL-7 -ESS 1 -SHEL-1 -ESS
C-east
R elinq uished B M C -5
1 -SHEL-6-ESS (2002)
D isc overies 1 -SHEL-2 -ESS

Tertiary
A-w est
M aastr ich tian

S anton ian R elinq uished

C enom an ian
(2002)

“N am orado” 1 -SHEL3 -ESS

B C -60 B M C -4

Flow Assurance Field Case – P241 Course

The BC-10 Block is located in the Campos Basin approximately 120 km SW of the coast
at the city of Victoria in Brazil. Fields are located at roughly 2km seawater depth and
there is a lack of gas and oil transport infrastructure in the area. The BC-10 license was
signed on 6th August 1998, with Shell as the operator for the License with a 35% interest
(Petrobras 35%, ExxonMobil 30%). Partners got re-shuffled at a later stage, resulting
into a 50%-35%-15% split between Shell, Petrobras and ONGC.The discoveries all have
the following general characteristics:

Deepwater setting;
658 Fundamentals of pipeline engineering

Poorer crude properties (lower API, higher viscosity, high total acid number - TAN);
Relatively low reservoir pressures and temperatures;
Strong emulsion tendency and possible complex rheology;
Artificial lift is required to achieve economic production.

The planned BC-10 development is a phased cluster development based on a Floating

Production Storage and Offloading (FPSO) concept, with the vessel anticipated to be
located halfway between C and O-North (ON), targeted in phase 1. A-West (AW) would
follow in phase 2. The phasing was changed at a later stage, where C, B-West and A-
West would be developed in phase 1 and ON would follow in phase 2. First oil is
expected in 2010.

The initial concept of the BC-10 heavy oil deepwater development off the coast of Brazil
included gas lift to lift the viscous crude up a 2km riser onto an FPSO. Flow assurance
studies indicated that the required gas lift boost could not be achieved and the project
embarked on a concept relying on subsea pumps to lift the viscous fluids to the surface.
We discuss the following critical flow assurance issues that had to be addressed for this
concept:
Fluid properties and rheology,
Steady state thermal hydraulics,
Hydrates and methanol dosage rates,
Wax and asphaltenes,
Scale,
Naphthenates,
Chemical systems engineering,
Transient thermal hydraulics: slugging and hot oiling.

BC-10 FEAST and Rheology

C reservoir: Fluid 1-SHEL-11-ESS, sample 2.07

ON reservoir: Fluid 3-SHEL-16HP-ESS, sample 2.25
Additional Lab Tests
Dry O il Viscosity ON Emulsion Viscosity ON
100000 250000
3000 psia, Sample
50F
2000 psia, Sample 70F

10000 1000 psia, Sample 110F

Visco sity (cP) -

200000
100 psia, Sample
14.5 psia, Sample
1000
3000 psia, Model
(cP)

150000
Viscosity(cP)

2000 psia, Model

Viscosity

100 1000 psia, Model

100 psia, Model 100000

14.5 psia, Model

10
0.0030 0.0032 0.0034 0. 0036 0.0038
50000
T emperature (1/K )

0
0 10 20 30 40 50 60 70 80
Water Cut(%)
Water Cut (%)

Flow Assurance Field Case – P241 Course

Coordinated through FEAST (Fluid Evaluation and Stability Testing), fluid samples of
the various fields were collected during the exploration phase. Basic fluid property
analyses were performed, described and documented in associated PVT reports.
Comparing the results of the various PVT reports the following samples were concluded
to be the most representative for the C and ON reservoirs:

C: Fluid 1-SHEL-11-ESS, sample 2.07, API 24;

Flow assurance field case 659

ON: Fluid 3-SHEL-16HP-ESS, sample 2.25, API 16.

Note that the previous slide does not show the location of ON (well 3-SHEL-16HP-
ESS).

PipeSim and PVTSim viscosities are observed to give an underprediction at low

temperature and required further attention before embarking on steady state (PipeSim)
and transient (OLGA, using PVTSim properties) simulations. Live oil viscosities were
measured as a function of pressure, temperature and water cut to enable reliable tuning
of the PVT models. Measured and modeled dry oil viscosities are plotted in left hand
graph; emulsion viscosity as a function of water cut is presented in the right hand graph.
Note that live oil emulsion viscosity data is typically not part of standard PVT reports.
Both C and ON turned out to form stable emulsions with Newtonian fluid behavior (no
dependence on shear rate) with free water present above 60% water cuts, giving rise to
the viscosity decrease observed in the right hand graph. With the development phases
changed it was decided to model B-West as an ON analogue with a lower GOR (200
scf/bbl) and assume AW to be a light fluid with a GOR of about 600 scf/bbl.

BC-10 Concept 1 – Gas Lift

FPSO a nd Sub Sea W ells

Flowline
B-West
Water Injection Line
Control Umbilical
Power Umbilical
Gas export line

O-North and O-South

A-West

BC –10 Standalone FPSO + Subsea

C
• FPSO between O-North and C
• Sub sea or well lift and water injector at
O-North and O-South
• Sub sea or well lift or gas lift for C
• Gas exported

Figure1

Flow Assurance Field Case – P241 Course

The initial concept of the BC-10 development was based on gas lift in the wells or at the
foot of the riser to lift the fluids up the 2km riser onto an FPSO. The concept is
schematically displayed in this slide showing the FPSO halfway between C and ON
where the gas is separated from the oil and exported through a subsea tie-in with the
Jubarte network 27km away from the BC-10 block. Oil is offloaded to a tanker. Oil
recovery is enhanced through water injection (water flooding). The initial concept looked
promising until a detailed analysis demonstrated that gas lift could not give the pressure
drop reduction (or boost) required to make the development economical.

Alternative artificial lift options were evaluated and eventually led to the final concept
based on so-called caisson ESP’s (Electrical Submersible Pump). In simple terms, a
caisson is nothing more than a 100m deep hole of 1m diameter containing an ESP.
Drilled into the seafloor, the caisson receives production from upstream manifolds and
the ESP pumps the hydrocarbons up the 2km riser onto the FPSO. Caissons have the
option of operating in a separating or non-separating mode. In the separating mode, gas
collected at the top of the caisson is allowed to flow freely to the FPSO through a
660 Fundamentals of pipeline engineering

separate gas riser. The next slide presents a snapshot of the caisson concept during the
development path of the BC-10 project..

BC-10 Concept 2 – Caisson ESP Artificial

Lift Phase 1
1 x 12” Oil Flowlines with 10” Riser
1 x 8” Gas Flowline
1 x 6” Service Flowline
2 x 8” Oil Flowlines
1 x 6” Gas Export
3 x Control and Power Umbilicals

(Ostra)
(Argonauta)
C Field
B West Field
6” Oil Flowlines
PM1
PM2 (Abalone)
A West

Gas Export to 16 K
M
Jubarte (27Km)

Phase 2
5

Flow Assurance Field Case – P241 Course

C wells produce into production manifolds PM1 and PM2, and Phase 2 has the A-West
wells producing into PM2 as well. At a later stage of the project A-West was made part
of Phase 1 with wells producing into PM1. Note that the final names of the fields are
shown as well: Ostra for C, Argonauta for B-West and Abalone for A-West. Green lines
represent multiphase flow, red lines gas production and the orange lines umbilicals and
power supply. Distance from PM1 to the FPSO is about 9km; from A-West to PM2 is
about 16km. Caissons are indicated by yellow boxes, operating in a separating mode at
500 psig and 120F on the C side and in a non-separating mode on the B-West side due
to the low GOR and high viscosity. FPSO arrival pressure is 300 psig. The next slide
gives a schematic drawing and picture of a caisson ESP
Flow assurance field case 661

Caisson ESP

Flow Assurance Field Case – P241 Course

Concept 2: C to Host Oil Line

Capacity and Arrival Temperature – Steady State

Flow Assurance Field Case – P241 Course

Steady state thermal hydraulics needed to be assessed for the various flowlines that make
part of the concept. We’ll briefly discuss the 9km 12” C to host oil line as an example.
Critical elements are capacity and arrival temperature. The line needs to be able to deliver
planned production rates at temperatures above the hydrate formation temperature from
Early Life (EL, 0% WC) to Mid Life (ML, 50% WC) to Late Life (LL, 85% WC).
Anticipated caisson pressures of 500 psig may turn out to be higher, giving a higher
pump suction pressure which needs to be checked as well. And two riser options had to
be considered: a Steel Lazy Wave Riser (SLWR) versus a Flexible Catenary Riser (FCR).
The two graphs in this slide give the flowline inlet pressure and arrival temperature as a
662 Fundamentals of pipeline engineering

function of liquid production rate, covering the various sensitivities discussed above.
Pump curves included in left hand graph demonstrate that 3 pumps of 1500 HP can
roughly deliver 80 mbpd (where the curves cross-sect) at a 115 F arrival temperature.

Note that the arrival temperature needs to be sufficiently high to guarantee temperatures
above the hydrate formation temperature after 12 hours of cool down after shut-in.

Concept 2: C Fluid Hydrate Curves

Reservoir fluid and oil/gas after separation in Caisson
ESP Temperature [C]
-1.1 1.7 4.4 7.2 10.0 12.8 15.6 18.3 21.1 23.9 26.7
5000 345

4500 310

4000 276

3500 241

3000 207

Pressure [bar]
Pressure [psi]

2500 172

2000 138

1500 103

1000 69

500 34

0 0
30 35 40 45 50 55 60 65 70 75 80
Temperature [F]

Overall Gas-Base Case Liquid-Base Case

Flow Assurance Field Case – P241 Course

Heat retention will be used as the main hydrate inhibition strategy for the multiphase
lines on the C and B-West side of the BC-10 development. Dead oil displacement will be
used to evacuate the live oil content after shut-in, where sufficient cool down time will
be critical to allow appropriate decision taking and robust operation. Hot oiling will be
applied to heat up the flowlines before restart. So far a standard approach with a proven
track record. Note, however, that subsea separation introduces a new challenge.

C and A-West fluids will be separated in the caisson. As a result, the single hydrate curve
for the associated reservoir fluids will be split into two hydrate curves for the gas and
liquid after separation. This slide shows the hydrate curves for the three C fluids:
reservoir fluid and gas and liquid separated at base case conditions of 500 psig and 120F.
As a result of the low heat capacity of the gas, heat retention will not work and
continuous inhibition with methanol will be required to prevent hydrate formation in the
gas line downstream of the caisson separator. The actual split of the hydrate curve is
strongly dependent on the pressure and temperature at which the caisson separator will
operate. In the design basis the separator has been assumed to be highly efficient in
terms of gas/liquid separation: 99.95%. Thermal hydraulic studies demonstrated that a
somewhat lower separation efficiency would lead to an increased liquid hold-up in the
gas line resulting into a higher hydrostatic head giving a higher operating pressure of the
caisson separator. The higher operating pressure has a strong impact especially on the
hydrate curve for the separated liquid, possibly shifting it to the right by 5F and thus
reducing cool down times. Decreased separation efficiency also increases the amount of
water carried over into the gas line, requiring increased methanol dosage rates thus
further increasing the liquid hold-up in the gas line. As a result, detailed sensitivity
studies were performed, varying separation temperature, pressure and separation
efficiency to determine methanol requirements as a function of pressure and water
content. See results in next slide.
Flow assurance field case 663

Note that liquid carry-over into the gas line is a very critical issue for BC-10. Hydrostatic
head build-up can choke the gas line, which would lead to an immediate shut-in of C and
A-West production, followed by restart complexities as a result of a liquid filled gas line
that would need to be emptied first.

Concept 2: C Gas Methanol Dosage Rates

C gas after separation in Caisson ESP
1.8

1.6
Methanol Dosage Rate [bbls / MMscf]

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0
300 350 400 450 500 550 600 650 700 750 800
Pressure [psi]

0.50 bbls water/MMscf 1.24 bbls water/MMscf 2.35 bbls water/MMscf

3.49 bbls water/MMscf 6.50 bbls water/MMscf

Flow Assurance Field Case – P241 Course

Concept 2: Wax and Asphalthenes

SARA Stability Screen – Fluids Stable for
Asphalthenes
2.5

1-SHEL-1-ESS B-West
1-SHEL-11-ESS C
2
1-SHEL-11-ESS C
1-SHEL-9-ESS O-North
Saturate / Aromatic

1-SHEL-7-ESS O-South

1.5

UNSTABLE
0.5
STABLE

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Asphaltene / Resin

10

Flow Assurance Field Case – P241 Course

Wax and asphaltenes risks were quickly identified as being very low. BC-10 fluids on the
ON side showed a high level of biodegradation, leading to low proportions of the high
molecular weight components required for wax deposition. Although not as biodegraded
664 Fundamentals of pipeline engineering

as the ON and B-West fluids, C fluids turned out to be low on these components as
well. The critical exception in this assessment is the A-West fluid. Due to the absence of
samples it had to be anticipated that the A-West fluid could possibly have significantly
different wax and asphaltene properties compared to the other fluids. The SARA screen
plotted in the graph on this slide shows all BC-10 samples on the stable side of the plot,
indicating low risks of asphaltene precipitation and deposition.

Concept 2: Scale
Scaling Tendency Parameter ST – Scaling unlikely
BaSO4 Scaling Tendency
B-West Midlife, 50%, 2 Well
Scaling Tendency
key

3300 ST

0.8 No scale
1
Scale
3
unlikely
Pressures

2300 5
Scale
6.5
possible
8
Scale
16
very likely

1300

300
90 100 110 120 130
Temperatures
Production P, T, 2 Well

11

Flow Assurance Field Case – P241 Course

Calcite and barite scaling risks have been assessed and showed no precipitation from
reservoir to caisson and caisson to topsides for the entire lifecycle of production
calculated. Risks are calculated in term of the ST (Scaling Tendency) parameter, plotted
for B-West water in the graph on this slide. As a result, no scale inhibition is
recommended. However, surveillance programs should always be in place and if possible
production water samples should be collected to check on any changes in the water
composition. Collection of gas samples and checking for carryover of water is desirable
as well. Also, pressure drops should be monitored especially around the riser base, as
that is the only point in BC-10 C production where the scaling tendency turned out to be
>1.5.
Flow assurance field case 665

Concept 2: Naphthenate Scale

Risk Low
Nap
GOR/ Viscosity -
Fluid Reservoir Depth T P CO2 acid Ca2+ Na+ HCO3
Prospect Well Name GWR @Tres & API (°) Sat Aro Res Asph TAN NAN pH
type Formation (MD, m) (°F) (psia) mole% conc. (ppm) (ppm) (ppm)
(scf/bbl) Pres (cP)
(wt. %)
oil 2,927 116 4,309 0.12 154 164.4 15.1 32.3 46.5 17.7 3.45 4.2 1.51 1.93
B-West 1-SHEL-1-ESS Paleocene
water 3,125 132 4,602 6.5 2.6 6.5 6,665 62,075 252
oil 2,850 132 4,245 0.10 278 8.5 23.5 60.5 31.9 7.4 0.22 2.5 1.314 1.85
1-SHEL-11-ESS oil 2,957 147 4,371 0.07 268 7.9 23.6 61.0 31.1 7.6 0.35 2.6 n.d. n.d.
C water Maastrichtian 3,037 149 4,965 1.02 4.4 6.1 4,789 60,904 108
oil 2,931 131 4,325 0.01 248 9.0 24.0 60.6 31.3 8.1 0.08 2.2 n.d. n.d.
3-SHEL-13-ESS
oil 2,969 124 4,370 0.01 232 n.d. 23.8 61.9 32.8 5.3 0.05 2.1 n.d. n.d.
A-West 1-SHEL-2-ESS oil Cenomanian 4490-4549 n.d. n.d. n.d. n.d. n.d. 25-40* No fluid sampled low** No water sampled
oil Eocene 2,931 129 4,302 0.6-0.9 338 34.7 16.1 34.2 45.7 16.8 3.32 3.4 1.45 1.55
O-North 1-SHEL-9-ESS
water*** Eocene 3,077 149 4,965 2.06 3.8 6.3 3,593 35,882 157
oil Eocene-1 2,920 135 4,555 0.12 202 97.7 14.2 37.6 44.2 15.9 2.3 4.8 n.d. n.d.
O-South 1-SHEL-7-ESS No water sampled
oil Eocene-2 3,091 127 4,309 1.20 410 13.5 17.6 32.8 46.2 18.7 2.3 4.5 n.d. n.d.
* Density estimation from cuttings' extracts
** Expected low because no signs of biodegradation were found in cuttings' extracts
*** Contaminated with ca 60% WBM
Parameters Impact Driving force
Naphthenic Acid
direct Higher concentration is usually considered to enhance naphthenate formation
Number (NAN)
-
HCO3 / pH formation
direct Higher pH enhances naphthenate formation
water
CO2 gas evolved from Sudden pressure drops in the system volatilise CO2 , which increases the pH of formation brine. Typically: the more CO2 the larger the rise in pH can become. CaCO3 formation will ‘compete’ with
direct
oil naphthenate formation..
Formation water
direct Higher concentration of cations (Ca+2 ) results in larger concentration of naphthenates (provided that naphthenic acids are present).
composition
Molecular Weight
direct Higher MW enhances precipitation of naphthenates
carboxylic acids
Total Acid Number
indirect Higher TAN could be indicative of higher concentration of naphthenic acids
(TAN)
API indirect Higher API crude seems more prone to sodium naphthenate formation. Lower API crude seems more prone to calcium naphthenate formation.
Total Dissolved Solids
indirect Higher concentration TDS enhances naphthenate precipitating out
(TDS) formation water

12

Flow Assurance Field Case – P241 Course

Naphthenate scales are a relatively new solid not as well understood as the default scales
discussed above. Often high TAN crudes are immediately associated with naphthenate
formation issues. The overall picture is bit more complicated though, requiring other
parameters to be evaluated as well, but where it has to be stated that naphthenate risks
are more based on a qualitative assessment by a specialist than being the outcome of a
straightforward calculation. The table in this slide shows the various oils and parameters
based on which the overall naphthenate risk for BC-10 has been assessed as low.

Concept 2: Chemical Systems Engineering

Chemicals, Dosage, Storage, Compatibility

13

Flow Assurance Field Case – P241 Course

Chemical systems engineering covers the base chemical dosages, the required chemical
injection rates and ranges and the FPSO topside chemical storage requirements. In
666 Fundamentals of pipeline engineering

addition, the chemical compatibility issues and work plan, including chemicals to
chemicals, chemical to produced fluids, chemical to metallic/non-metallic materials,
should be handled. The table in this slide provides a high level overview of the BC-10
flow assurance risks and applied strategies based on which the chemical systems
engineering details mentioned above have been defined and determined.

Concept 2: Transient Operation

Slugging A-West line

Elevation Profile A-West to C to Host - SLWR

Host at 300 psig
1900
1700 A-West to C Curved to Host - SLWR
1500
1300
Elevation (m)

1100
900
700
500
10 mmscfd A-West @ 15 mbpd 15 mmscfd from C
300
Gas Line
100
-100 Oil Drain Valve
0 4000 8000 12000 16000 20000 24000 28000

Horizontal Distance (m)

14

Flow Assurance Field Case – P241 Course

As already mentioned a reliable operation of the C and A-West fields is dependent on

minimum liquid carry-over from the caisson separator into the gas line. Too high liquid
hold-up can choke the gas line, forcing the caisson into a non-separating mode with gas
volume fractions beyond the specs of the ESP thus leading to an immediate shut-in of
the producing fields. The elevation profile of the 16km flowline from A-West to C
combined with the higher GOR of the A-West fluid introduced a slugging risk which
could affect stable operation of the caisson separator with a possible negative impact on
separation efficiency. The elevation profile plotted in the graph on this slide that shows
the location of the caisson and the SLWR riser used for the gas line. A-West is
producing at 15 mbpd, giving an associated gas production of about 10 mmscfd. The gas
line will also be receiving gas from C at roughly 15 mmscfd. Topsides arrival conditions
are frozen at 300 psig. Transient simulations were performed to evaluate the slugging
tendency of the A-West line and its impact on level oscillations of the caisson separator.
As a sensitivity the simulation was repeated for an A-West GOR of about 1000 scf/bbl.
Results are presented in the next slide.
Flow assurance field case 667

Concept 2: Transient Operation

Slugging A-West Line
A-West to C Caisson to Host - Caisson Pressure A-West to C Caisson to Host - Liq. InFlow Sep.
(Base Case: Curved, SLWR, 15mbpd, 15mmscfd C, 627 scf/bbl, 0%WC) (Base Case: Curved, SLWR, 15mbpd, 15mmscfd C, 627 scf/bbl, 0%WC)
625 40000
Base Case
35000
Worst Case: GOR 1038 scf/bbl

Liquid Flow (bbl/d)

600 30000

Pressure (psia)
25000

575 20000
Base Case
15000
Worst Case: GOR 1038 scf/bbl
550 10000

5000

525 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6
Time (hrs) Time (hrs)

A-West to C Caisson to Host - Caisson Level A-West to C Caisson to Host - Liq. OutFlow Sep.
(Base Case: Curved, SLWR, 15mbpd, 15mmscfd C, 627 scf/bbl, 0%WC) (Base Case: Curved, SLWR, 15mbpd, 15mmscfd C, 627 scf/bbl, 0%WC)
45 40000
Base Case
35000
Worst Case: GOR 1038 scf/bbl

Liquid Flow (bbl/d)

40 30000
Liquid Level (m)

25000

35 20000

15000

30 10000
Base Case
5000
Worst Case: GOR 1038 scf/bbl
25 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6
Time (hrs) Time (hrs)

15

Flow Assurance Field Case – P241 Course

The blue base case curves in the top right graph indeed shows a slugging inflow into the
caisson separator. However, separator level is observed to be stable with maximum
oscillations in the order of 1 to 2m as a result of the small size and high frequency of the
hydrodynamic slugs. The opposite is observed for the higher GOR case plotted in red.
Level variations in the order of 10m occur as a result of larger slug sizes entering the
separator at a lower frequency with a high risk of reducing the separation efficiency.
Recent samples collected for A-West revealed a highly waxy crude with a GOR of
around 4000 scf/bbl. As a result wax deposition and slugging of the A-West flowline has
become a top priority for the BC-10 project.

Concept 2: Transient Operation

Hotoiling of A-West Loop

Elevation Profile
Host to PM1 to A-West to PM1 to Host - SLWR
1900
6in Service Line - Host to PM1
1700 6in Production Line - PM1 to A-West to PM1
1500 12in Production Line - PM1 to Host
PM1 Up
1300
Elevation (m)

A-West
1100 PM1 Down
900
700
500
300
100
-100
0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 55000

Horizontal Distance (m)

16

Flow Assurance Field Case – P241 Course

668 Fundamentals of pipeline engineering

An other critical mode of transient operation is hotoiling. Before startup after an

extended shut-in the flowlines need to be heated up with hot dead oil to prevent hydrate
formation. Note that production fluids flowing from the manifolds are not inhibited and
would possibly form an immediate hydrate blockage if produced into a cold system at
ambient seabed conditions of 4C. The most challenging hotoil loop includes A-West at
25km distance from the FPSO. The elevation profile is plotted in the graph on this slide,
where the legend shows the various flowlines and diameters. Heated oil is pumped from
the FPSO through the 6” service line to PM1, from PM1 to A-West, from A-West back
to PM1 and from PM1 up the 12” oil line back to the FPSO, thus traveling a distance of
about 50km. Hotoiling needs to be continued sufficiently long to ensure a 12 hour cool
down time in case of a sudden shut-in; worded differently, minimum temperatures need
to be sufficiently above the Hydrate Dissociation Temperature (HDT). Note that we
have to consider two hydrate curves at two locations of low temperature:

The PM1 temperature on the return leg will need to be sufficiently above the A-West
fluid HDT.
The oil line riser base temperature will need to be sufficiently above the HDT of the
separated oil.

Concept 2: Transient Operation

Hotoiling of A-West Loop

Hotoil Host-PM1-Aw-PM1-Host

25000 120

20000 96
Hotoil Rate (bbl/d)

Time (hrs)
15000 72

10000 48
Hotoil Rate
Time T_PM1>HDT_fluid
5000 24
Time T_RB>HDT_oil
Time to 99% of Rate
0 0
3000 3500 4000 4500 5000 5500

Hotoil Pressure (psia)

17

Flow Assurance Field Case – P241 Course

The riser base is selected as the location of highest pressure and lowest temperature, thus
forming worst case conditions for hydrate formation. Topsides pumping pressure is
designed to be in the range of 3500 to 5000 psia. Transient simulations were performed
to determine what hotoil rates would be achieved at this pressure range and how long it
would take to get above the HDT’s at PM1 and the riser base, starting from ambient
seabed temperature conditions. Results are presented in the graph on this slide. At a
maximum pumping pressure of 5000 psia a hotoil rate of about 20 mbpd is achieved.
Note that this rate is not immediately established: it takes almost 2 days to get to 99% of
the steady state hotoil rate, indicated by the light blue curve. Coincidently the curves for
the temperature at PM1 to get above the A-West HDT (green curve) and the riser base
temperature to get above the HDT of the separated oil (blue curve) coincide above 4000
psia hotoil pressure. At a 5000 psia hotoil pressure temperatures at both locations are
observed to take roughly over a day to get above the HDT’s of the respective fluids.
Results demonstrate that hotoiling can be a slow and tedious process where
Flow assurance field case 669

temperatures will need to be carefully monitored to ensure a smooth start-up of the

system.

Thanks for your

attention.
Shell Global Solutions GSUF Contacts:
Multiphase Flow Systems: Ruud Henkes (NL), Greg Bethke (KL), Leonid Dykhno
(US)
Hydrates: Ulfert Klomp (NL), Greg Hatton (US)
Wax, Asphaltenes, Naphthenates: Mark Grutters (NL), George Broze (US)
Scale, Emulsions, Rheology, Foam: Menno van Dijk (NL), Cor Kuijvenhoven
(US)
EP Projects Contacts:
Keith Stevens (KL), Wade Schoppa (US), Susan Lorimer (EPE)
Further Learning - SOU:
Flow Assurance Course, PRFA 18
Subsea Systems Integration Course, P228
Flow Assurance Field Case – P241 Course
Pipeline inspection
Pipeline inspection 673

EXPECTATION

EXPECTATION

 Need for pipeline inspections

 Basis of inspection
 Methods used
 Both internal and external inspection
 What anomalies to look for
 During routine inspections and maintenance
 Discuss methods of assessing anomalies
 How to correct anomalies that are found

We will introduce the activities required for the operation of subsea pipelines. The first
requirement is to understand the inspections that are carried out. Why do we do them
and how do we decide what to look for?

Various methods of conducting internal and external inspections of pipelines are

examined. A description is given for the various types of anomalies that may be found
during an inspection.

An overview is given for the methods of assessing the various different anomalies and
finally the methods of correcting those anomalies are discussed.
674 Fundamentals of pipeline inspection

RISK-BASED INSPECTION PLAN

WHY INSPECT DURING

OPERATION?

 Damage
 Debris
 Anode removal or
early wastage
 Exposure
 Spans
 Leaks
 Rock cover

In the ‘Installation methods’ module, we mentioned that surveys were carried out to
characterise the seabed, and that further surveys were required during operation. These
typically look for anything that may have gone wrong with the pipeline:
■ Damage – impact from trawler activity or objects dropped from platforms or
supply boats
■ Debris – near to or draped over the pipeline (large boulders are dragged and lifted
by trawlers)
■ Anode wastage – usually anodes are knocked off by fishing but they can be
ablated away following coating damage
■ Exposure of a previously-buried line
■ Spans – scour of soil from beneath the pipeline
■ Leaks – usually from flanges
■ Loss of rock cover
Pipeline inspection 675

HOW OFTEN TO INSPECT

 Of old (1970 to 1990) once a year by ROV

 Now, responsibility of Operator
 Inspection plan

There is a variety of techniques to survey the pipeline, and the operator is also presented
with a choice of when it is necessary to survey. In the prescriptive regime of the 1980s,
an annual survey of the entire length of the pipe was required. In the goal-setting regime
of today, in UK waters, it is the responsibility of the operator to determine a safe
inspection interval as part of his inspection plan. The result of this is that most
inspections in the early part of the pipeline life are carried out annually, and they are then
spaced further apart if the results are benign.

INSPECTION PLAN

Identify risks

Write an
inspection plan

Inspect pipeline
and report

Interpret results

Decide what Decide when next to

to fix inspect and scope
of inspection
676 Fundamentals of pipeline inspection

The flowchart above shows the processes involved in pipeline integrity management. It
is an iterative process in which the results of inspections are fed back in to the inspection
plan and future inspections modified accordingly.

RISK-BASED INSPECTION

 Hazards identified
 Inspections targeted according to
 Risk (probability x consequence)
 Value of inspection
Highest value
of inspection
Consequence
of failure

n
io
ect ity
l
Not worth sp a
In itic
inspecting cr
Probability of failure
for

Risk-based inspections use probabilities, consequences and the usefulness of inspections

to arrive at a suitable inspection regime. The various hazards facing the pipeline are
identified and the risk is evaluated. The value of inspecting is then assessed for that
particular hazard. For example, inspection can tell you if a span is developing to an
unacceptable length, but it cannot tell you if somebody is going to drop an anchor on
your pipeline. All this data is used to give an overall value for the inspection, and the
inspections can be prioritised accordingly.
Pipeline inspection 677

RISK-BASED INSPECTION PLAN –

SUMMARY

 Range of hazards and anomalies

 Variety of inspection techniques available
 Inspections driven by risk and value

Any questions?

A subsea pipeline faces many hazards during its lifetime. Inspections must be done to
ensure that the pipeline continues to operate safely. Several tools are available to
externally inspect the pipeline. The external inspection strategy will be driven by an
assessment of the risks due to the anomalies and the usefulness of the chosen inspection
technique in countering those risks. The inspection plan will evolve over the life of the
pipeline using feedback from the results of the inspections.
678 Fundamentals of pipeline inspection

EXTERNAL SURVEY

EXTERNAL INSPECTION METHODS

 Initially, side-scan sonar, ROTV or AUVs

 Rapid review of pipeline route
 Spans and exposure
 ROV
 Slower investigation
work of defects
already uncovered
 CP system
 Diver
 Shallow water
investigations at
landfalls and risers

The year-to-year maintenance of the pipeline involves external inspection surveys

followed by remedial works to correct any problem areas. The pipeline may be surveyed
by side-scan sonar, remotely-operated towed vehicles (ROTVs) held some 20 m (66 ft)
above the seabed, or autonomous underwater vehicles (AUVs). ROTVs can view at an
angle of 30° from the horizontal, so this height will cover the pipeline and the immediate
adjacent area.

The initial surveys may be carried out by ROV (as shown in the picture), especially if
detailed information is needed. However, side-scan sonar is far more common for the
routine survey of pipelines.

AUVs are more suitable for inspection of long lengths of major trunk lines

Modern side-scan sonar is quite capable of picking up any physical objects of concern to
the pipeline and of estimating span heights and lengths. Being considerably faster than
ROV surveys (perhaps by as much as a factor of 10), and therefore lower cost, it is in
widespread use.

However, ROVs do have some specialist advantages:

Pipeline inspection 679

■ They are able to carry pipe detectors, as shown in the picture, which will detect
buried pipelines and will determine the level of rock dump on top of them.
■ They can take cathodic protection measurements.
■ They can make a thorough investigation of spans, both in terms of touchdown
point and vibration characteristics. These measurements can resolve whether or
not it is truly necessary to take remedial action on a span.

Divers are limited to shallow water investigation work and near platforms. Just below
the surface, ROVs are unable to operate due to cavitation of their thrusters, and also
because they are destabilised by wave and current action.

SIDE-SCAN SONAR

 Towed fish or ship-mounted

 Monitor pipeline profile,
spans, burial, seabed
features, lateral buckling
Track of vessel

Exposure due to scour

172 m (564 ft)

Pipeline span

Side-scan sonar techniques use a towed fish such as that shown in the picture above.
They are based on sonar, whereby the device emits a sound pulse and listens for the
echo. It interprets the strength, time and direction of the echo to give a picture of the
pipeline and seabed in sufficient detail to gauge pipeline features such as embedment and
spans. Seabed features can be distinguished such as sand waves, debris, trawl scars, etc.

The sonar printout is from an integrity management contract currently being undertaken
at Jee. The seabed mega-ripples are indicative of a mobile sandy bottom and the
exposed section of pipeline is clearly visible together with the region either side of the
span where the mega-ripple pattern has been disrupted by scour. The shadow at the
centre shows where scour has developed sufficiently to cause an unsupported length of
pipeline to span freely. This may be further investigated by ROV to determine the
support end points.
680 Fundamentals of pipeline inspection

ROV VISUAL SURVEY

Typical screen shots from an ROV survey show disrupted weight coating at an anode
beneath substantial marine growth. The second view is of a spanning pipeline at a field
joint.

Both screens show the date and position (Eastings and Northings) and chainage along
the pipeline with other essential ROV camera view data.

EXTERNAL INSPECTION – SUMMARY

 Initial pass – side-scan sonar, ROTV or AUV

 Further investigation by ROV or diver
 Exact length of span requires closer inspection
 Each technology has pros and cons

Any questions?

The external inspection of the pipeline can be carried out using the following
technologies:
■ Side-scan sonar
Pipeline inspection 681

■ ROTV
■ AUV
■ ROV
■ Diver

Side-scan sonar using fish or ROTV is the cheapest and fastest method but cannot pick
up fine details – especially those beneath the pipe.

ROV is a more flexible method than side-scan sonar, and the ROV can carry a range of
extra instrumentation.

AUVs are a relatively new technology but much development is going on. They have
potential for being one of the most flexible and economic methods.
682 Fundamentals of pipeline inspection

INTERNAL INSPECTION

WHAT IS INTELLIGENT PIGGING?

 Internal survey for corrosion and cracks

 Typically every 5 or 10 years

In addition to external inspection, there are internal inspection techniques available for
pipelines. These range from the simple gauge plate check for dents and debris, through
to sophisticated pig checks for cracks and corrosion.

In this section, we will look at three types of intelligent pigs – the magnetic flux,
ultrasonic and eddy current pigs, describing their function and their uses.

Before an intelligent pig is sent down the line, a full pig cleaning operation would
normally be carried out to ensure the intelligent pig remains undamaged and that the
data obtained from the pig run is the highest quality possible.
Pipeline inspection 683

INSPECTION TECHNOLOGIES

 Magnetic flux
leakage
 Axial or
transverse
defects
 Ultrasonic
thickness
 Radial wall
thickness loss
 Axial or
transverse
cracks
There are two main technologies: magnetic flux leakage (MFL) and ultrasonic thickness
measurement (UT).

Each has started by finding wall thickness loss (general corrosion defects, both inside
and outside). The alignment of the sensors allows different direction of defects to be
detected.

For example, the ultrasonic detectors can be used at a 45º angle to the surface of the
pipe wall to find weld cracks. If they are aligned axially with the pipeline then they can
scan the girth welds, or by aligning in the circumferential direction, the seam welds can
be surveyed.

The following slides contain examples of each of these pigs.

684 Fundamentals of pipeline inspection

PII INSPECTION PIGS – VIDEO

The inspection tool is inserted into the pipeline using pigtrap launchers (shown above)
and is propelled along the inspection route by the pressurised oil or gas. The tool uses
magnetic flux leakage (MFL) or ultrasonic sensors to detect defects in the pipe wall and a
digital signal processing data logger to record up to 1000 pieces of data per second. The
ultrasonic pigs require a slug of liquid to make contact with the wall. In gas lines, the
liquid needs to be loaded between separator pigs with the intelligent pig at the centre of
the slug.

These sensors can be aligned either around the circumference or axially to find cracks in
the seam or girth welds. Loss of steel (pitting, grooves or cracks) may be generally
around the pipe or may be localised at the bottom, sides or top of the pipeline.

Above-ground sensors track the position of the tool before logging the information
using GPS. On retrieval from the pipeline, the data is uploaded from the tool and then
sent to the lab for processing and interpretation.
Pipeline inspection 685

MAGNETIC FLUX PIGS

The picture above shows the PII magnetic flux pig. In the picture you can see the
magnetic brushes and the finger-like arrays of magnetic flux detectors. The rest of the
pig contains power and data storage facilities. It is used to detect internal and external
corrosion defects in oil and gas pipelines. Variants are available to detect both axially-
oriented and hoopwise-oriented cracks.

Typical speeds of intelligent pigs are from 0.3 m/s to 5 m/s (1 ft/s to 16 ft/s). If the
product flow is faster than this, it is normal to include a bypass system to permit the pig
to travel slower than the oil. Head losses for pigs are typically less than 1 bar (15 psi).

FLUX LEAKAGE DETECTION

No defect present

Brush Brush
Sensor
N Magnet Magnet S

Pipe wall
Defect Defect present section

Brush Brush
Sensor
N Magnet Magnet S

Leakage field
686 Fundamentals of pipeline inspection

The above diagram illustrates how the magnetic flux pig detects corrosion defects.

In the non-corroded condition, the two brushes form a magnetic circuit and the flux
passes through the pipeline wall with little passing either side. However, in the lower
diagram with the defect present, there is less metal through which the flux can pass, and
some of it leaks outside the pipe wall and is detected by the sensor.

By passing the pig through a test line with known defects, a comparison can be made
with the signals detected on the sensor during the survey run.

ALL-DEFECTS SUMMARY –
DATA ASSESSMENT PRESENTATION

100
Internal corrosion located in 8.74 mm (0.344 in) wall thickness pipe
90
Reported peak depth as %ge of wall

80

70

60

50

40 1.5 x MAOP
(Simplified RSTRENG)
30

20 This area
contains
10 416 341 internal 1.5 x MAOP
corrosion features (ANSI/ASME B31.G)
0
4 8 12 16 20 24 28 32
0 100 200 300 400 500 600 700 800
Reported axial length mm (in) Courtesy of PII Group

The above diagram shows an example of results obtained from an MFL pig. It shows
over 450 000 internal metal loss features, and there were also over 3500 external metal
loss features identified. This analysis looks for any defects where the dimensions exceed
those tolerable at 1.5 x MAOP, or where the peak depth exceeds 80%.

In this case there were 19 defects identified as possibly needing repair. A detailed
assessment was also carried out in each case, measuring the effective depth and length of
the defect, which reduced the number of defects requiring repair down to just one.

The RSTRENG and the ANSI/ASME B31.G methods are two alternative approaches
for assessment.
Pipeline inspection 687

ULTRASONIC PIG

The ultrasonic pig is used to detect corrosion defects in liquid lines. The liquid is crucial
in acting as a couplant for the sound emitted by the ultrasonic probes. For these pigs to
function in a gas line, they must be run within a slug of liquid. Alternatively pigs are
available that use wheels running along the pipe wall to transmit the ultrasonic vibrations
into the pipe wall.

More modern ultrasonic pigs are also able to detect cracks of may different orientations
in the pipe wall, including longitudinal cracks.

PRINCIPLE OF OPERATION

 Sound waves transmitted into pipe wall

 Echoes from front and back of pipe wall
picked up by detector

Transmitter/ Requires liquid couplant

receiver Transmitter Receiver

Pipeline inner wall

Lamination
Weld crack
Pipeline outer wall
688 Fundamentals of pipeline inspection

The probes on the ultrasonic pig work in exactly the same way as hand held ultrasonic
thickness probes. They emit a sound wave and detect the front and back pulses reflected
from the specimen. By calibrating the speed of sound in the pipe steel (using a sample
of known thickness), it is possible to interpret the sound time delay in terms of metal
thickness. It is also possible to interpret, from the time of the first reflection, whether
the corrosion defect is on the inside or the outside of the pipe wall.

By angling a pair of such piezo-electric sensors, it is also possible to detect radially-

aligned welding cracks.

EXAMPLE ULTRASONIC PIG

OUTPUT

 More intuitive output

 Colour-coded depths Normal (no defect) Plate inclusions

Corrosion at
helical weld
Acid attack pitting

The output from the ultrasonic pig is a pixellated scan of the entire internal surface of
the pipe.

In the picture above, any normal reading has been colour-coded white and corrosion
defects given a colour code depending on their depth. One can see on the left-hand side
a spiral weld with preferential corrosion. Across the middle of the scan, there are some
corrosion pits characteristic of acid attack and one can also see some red inclusions
running parallel to the plate-rolling direction.

The output of the ultrasonic pig is far more intuitive than the magnetic flux pig and a
general picture of the state of the pipeline can be obtained rapidly at the end of the pig
run.
Pipeline inspection 689

OTHER PIGS

 Gyroscopic pig
 Camera and tethered inspection
 Crawlers
 Wax and scale assessment
 Calliper pigs
 Gyroscopic pigs

A variety of other pigs have been developed for pipeline inspections.

Gyroscopic pigs can be used to survey a pipelines shape and can also detect spans by the
vibration of the pipeline as the pig passes through the pipeline span. They are also used
to check for out-of-straightness of pipelines in trenches prior to backfilling, in order to
reduce the risk of upheaval buckling.

It is possible to survey the sections of pipeline close to the pig trap using tethered pigs.

Crawler pigs use on-board power to propel themselves against the fluid flow. Their
range is limited by the availability of on-board power. There is also a range of crawler
pigs that can be introduced at the landfall end of a subsea pipeline and crawl up to 17 km
(10½ miles) inside the line, towing an umbilical to give an on-site evaluation of the line’s
condition.

Pigs have also been developed to measure the wax and scale coating the inside of a
pipeline.
690 Fundamentals of pipeline inspection

EDDY CURRENT PIG FOR

FLEXIBLES

Flexible pipes present a different set of inspection requirements. The primary mode of
failure for which it is necessary to inspect is that of cracks in the hoop and armour
windings (rather than corrosion defects). The eddy current pig, pictured above, is being
developed to detect such cracks. It functions by inducing an eddy current in the
windings and detecting the difference in the resulting electromagnetic field caused by a
crack.

EDDY CURRENT MEASUREMENT

 Detect defects in metallic layers

The above picture is an output from the device showing a crack in one of the armour
wires.
Pipeline inspection 691

INTEGRITY OF FLEXIBLE RISERS

 External inspection Vent gas

monitor on
of flexibles flange

 Similar to rigid pipe plus

additional monitoring
 Polymer coupons
 Removed from port for testing
 Vent gas monitoring
 Continuous automatic testing
 Fibre optic cables
 Laid within the armour layers
 Continuous length assessment

Flexibles can be externally inspected but the scope for internal inspection is limited. The
integrity of the flexible riser liner can be monitored using coupons or vent gas
monitoring. Coupons are small samples of liner sitting within the flow of oil or gas that
can be removed during inspections for testing.

The condition of the polymer liners in flexibles can be assessed by monitoring of the gas
that diffuses into the windings layers. This photograph illustrates the vent gas
monitoring system.

The flexible pipe manufacturers are developing fibre optic cable systems that are laid
into the armour windings. These can monitor any potentially damaging stretching of the
armour wires.
692 Fundamentals of pipeline inspection

INTERNAL INSPECTION – SUMMARY

 Comprehensive internal survey of steel

pipelines for corrosion and crack detection
 Two main technologies, two orientations
 Pigs available for other tasks
 Flexible inspection under development

Any questions?

Intelligent pigs enable comprehensive internal inspections of the pipeline to check for
corrosion or cracks in the pipe wall. There are two main technologies used by intelligent
pigs for internal inspections, these are magnetic flux leakage (MFL) and ultrasonic
thickness (UT). A third technology of eddy currents is under development to enable the
inspection of flexibles.
Pipeline inspection 693

ANOMALY ASSESSMENT

Spans

SPANS

 Section of unsupported pipeline

 Uneven seabed Pipeline crosses seabed depression

 Rock dump
 Sandwaves
 Scour
Pipeline crosses seabed
 Rocks with changes in slope

 Coral outcrop
 Pipe crossing
Pipeline crosses seabed rock outcrop

A pipeline span is simply a section of the line that is not in contact with the seabed. This
can be due to a variety of reasons, the most common of which is an uneven seabed on
the selected route. Pipelines submerged in seawater form quite efficient beams, resulting
in a relatively high bending stiffness and a tendency to span over seabed undulations.

Rock dump can cause spans, in that the rock berm is designed to be stable and resist
dissipation due to environmental loads. The seabed around the rockdump may not be as
stable, and scour of the seabed may result in a pipeline suspended between periodic
mounds of rockdump.

Sandwaves are a feature of many soft seabeds, including the southern North Sea. The
sandwaves tend to propagate, resulting in continuously moving pipeline spans unless the
pipeline is lowered to below the trough level.

Seabed scour can be regional or localised. Regional scour is the general lowering of the
seabed, which tends to destroy pipeline trenches and create spans. Localised scour is
694 Fundamentals of pipeline inspection

caused by the presence of the pipeline. This can be caused by seabed currents, fish
digging themselves in under the pipe, or variations in seabed sediment.

SPANS IN UK SECTOR OF NORTH

SEA

 33000 spans on 7800 km (4900 miles) pipe

 Only 800 > 0.5 m (1.6 ft) high
with 260 > 0.8 m (2.6 ft) high
 19000 < 10 m (33 ft) long
around 100 > 60 m (197 ft) long
Span Height Log (F) = 1.6632 - 2.9169 Log (H) S p a n L e n g th L o g (F ) = 4 . 2 0 6 2 - 0 . 0 3 8 0 L

0.1 1 Span Height (m) 10 10000

10000

Frequency
100

100
Frequency

1
0 50 100 150 200
1 S p a n L e n g th (m )

Spans are very common. In the North Sea UK sector alone, there are in excess of 33
000 spans in 7800 km (4900 miles) of pipelines. That is an average of one every 230 m
(750 ft). Of these, most are short and low, maybe only a few inches high at most. Only
a few hundred are of any concern, either due to the integrity of the span or for their
potential to cause hooking of fishing gear. First-pass span analysis is principally about
identifying which spans present a problem and require further evaluation.

Typical spans have been described statistically. The distribution of span height is log-
log; while the distribution of span length is log-linear. There is no significant correlation
between span height and length.
Pipeline inspection 695

SPAN ASSESSMENT

 Determine critical span lengths –

during pipeline design
 First-pass
 Installation case
 Operation case
 What spans are likely to occur

 Assess fitness-for-service of span found

following construction or during operation
 Detailed assessment of stress and fatigue

There are a number of approaches to span assessment, which vary depending on when
they are carried out in the life of a pipeline.

During design, a first pass spanning assessment will often be performed. The purpose of
this assessment is to determine limits on allowable span length for the installation
contractor to work to. The assessment of these span limits is normally based on
conservative criteria, which ensure that no short or long term damage of the pipe will
result.

Also during design, an assessment of the seabed profile along the proposed route may be
performed to identify whether pipeline spans are expected to occur, and if so where and
how long. This assessment of the route would be based on survey data and would use
finite element analysis (using a general FEA package such as ABAQUS, or a specialist
pipeline package such as Orcaflex or Sage Profile) to ‘lay’ the pipeline over the
anticipated seabed profile. This analysis would give predictions of the numbers and sizes
of expected spans and therefore allow an assessment of the route preparation or span
remedial work that will be required. This is obviously important to allow assessments of
cost to be made.

Subsequent span analysis is performed during operation of the pipeline to address any
anomalous spans identified. The analysis is therefore to determine the acceptability of a
known span length, and would entail a detailed assessment of stresses and fatigue.
696 Fundamentals of pipeline inspection

WHY CORRECT SPANS?

 Yield and plastic hinges

 Vortex-induced vibration
 In-line or cross-flow motion induced by currents
Problem
bending
and
fatigue

End
Current and supports
Axial tension and
wave action
compression

Gap and trench

shielding

The section of a pipeline that spans is subject to its own self-weight, fluid loading and
potentially third-party loads from fishing gear. This could cause it to yield and to fail in
bending with plastic hinges. If this mechanism could occur, then it is necessary to
stabilise the span and give it additional support.

A second mode of failure for spans is a fatigue failure due to vortex-induced vibrations
(VIV). These are vibrations induced in the span due to the passage of currents (and
waves) perpendicular to the pipe. These cause the pipe to oscillate at its natural
frequency which, over a period of time, can lead to fatigue failure. Again, should fatigue
failure due to VIV be predicted for a particular span, it would need to be supported to
prevent this happening.

Major span lengths can prevent internal pipeline inspections because of the risk of
overstressing failure when the weight of an intelligent pig is passed through.
Pipeline inspection 697

SPAN CONCERNS

 Peak stresses – static and dynamic loads

 Global buckling – axial operational loads
 Fatigue caused by strong currents
 Vortex-induced vibrations (VIV)
 Plus wave and tidal oscillatory loads
 Particularly in shallow water
 Code for assessment
 DNV RP-F105

Span analysis considers three main aspects:

■ The potential failure due to excessive stresses from a long span.
■ The buckling of the span through excessive local bending.
■ The span failing due to column buckling caused by thermal expansive axial forces.
■ The likelihood of vortex-induced vibrations (VIV) occurring and hence the
potential for fatigue failure.

These aspects are considered during both the pre-construction analysis and the
operational analysis.

SPANS – SUMMARY

 Spans are unsupported sections of pipeline

 Result of uneven seabed terrain
 A common occurrence (33 000 in UK North Sea)
 Subsequent problems
 Bending and yield, VIV and fishing gear snagging
 Design assessment
 Determine maximum allowable span length
 DNV out-of-straightness/bottom roughness

Any questions?
698 Fundamentals of pipeline inspection

Pipeline spans are a common occurrence where the seabed terrain is uneven. The main
problem with assessing ‘real’ spans is determining the actual length and height of the
span. The difficulties in determining accurately the span height and length arise from;
■ The variable end conditions for different seabed soil types.
■ The potential for any mid-span touchdown.
■ The spans may move and change shape over time.

For the design of a pipeline, it is important to assess the maximum allowable span
length. This should be done for both the installation and operational cases. For
maximum allowable span lengths, it will be necessary to establish the peak stresses in the
pipeline for both static and dynamic loading of the span, consider the potential for
pipeline buckling due to the combined axial and bending loads within the span. Also,
possible fatigue due to VIV and interference with trawl gear may need to be assessed.

The DNV approach makes use of finite element methods to assess interplay between
adjacent spans.

Pits and Dents

PIPELINE WALL DEFECT TYPES

 Internal corrosion pit

 General wastage of wall
 Dent

Having found a defect during the internal survey, the next step is to decide whether it is
safe to continue operation or whether it is necessary to make a repair. We will consider
the approaches taken in evaluating the internal corrosion and dents.

Internal corrosion is rarely so simple as an isolated pit. The picture above shows some
general corrosion, some erosion, some preferential attack of the weld and some isolated
pits. The issue is how do we evaluate whether the pipe is safe?
Pipeline inspection 699

ASSESSMENT OF CORROSION

 Long defect fails as rupture

 Short defect fails as leak
 Based on
 Axial length
 Remaining wall thickness
 Proximity of other defects
 Codes
 ASME B31.G
 RSTRENG
 DNV RP-F101
 EPRG (dents)
Using finite element analysis and burst tests, corrosion defects have been assessed and
formulae developed to predict the safe operating pressure for a given defect or set of
combined defects. The assessment method is based primarily on the axial length of the
defect (or its equivalent axial length if there are a number of defects together) and the
remaining wall thickness.

A long defect will fail as a rupture while a short defect will fail as a leak. It is therefore
crucial to distinguish whether groups of pits are close enough to act as a single defect or
whether they will all act as isolated pits.

Cookbook formulae are given for the above analysis in ASME B31.G and more recently
(and more comprehensively) in DNV RP-F101, Corroded Pipelines, 2004. (Note: this
should not be confused with DNV OS-F101). The European Pipeline Group (EPRG)
has an assessment method for dent defects.
700 Fundamentals of pipeline inspection

ASSESSMENT OF IMPACT DENTS

 Gouges or cracks could fail by rupture

 Plain dents could fatigue or obstruct pigs
 Based on
 Unpressurised dent depth
 Gouges, cracks, sharp folds
CATS pipeline dent

The assessment of dented pipelines is usually based on the unpressurised dent depths. It
is crucially dependent on whether there are any localised defects such as gouges, cracks
or sharp folds within the dent. Essentially plain dents (without gouges, cracks or sharp
folds) fail at the same pressure as undented pipe. Consequently the problems that they
cause are centred on fatigue and the obstruction of pigs. The latter is due to the fact that
the dent reduces the diameter of the pipeline locally and could cause the pig to jam. The
fatigue aspect is due to the fact that there are stress concentrations at the dent, which will
flex as the internal pressure varies. This is covered in the next slide.

The photograph shows the diver inspection of an anchor impact dent on the Central
Area Transmission System (CATS) pipeline in the North Sea, after the concrete and
bitumen coating had been removed.

Should there be any gouges or cracks, then the dent could fail rapidly and
catastrophically due to a time-dependent creep in the material. To avoid this, the
pressure must be held below 85% of what it was when the dent was formed.
Pipeline inspection 701

FATIGUE IN DENTS

The above chart shows stress concentration factor (in this case the stress divided by the
internal pressure) versus D/t ratio. For offshore pipelines we are typically in the D/t
ratios of 15 to 25. The experimental results in the graphs show that the deeper the dent,
the higher the stress concentration factor.

It can be seen that a 7% dent in a pipe with a D/t ratio of 25 induces a stress 10 times
higher than that for a plain undented pipe. Given that fatigue is proportional to stress
cubed, this dent therefore reduces the fatigue life by a factor of 10³ or 1 000.

In summary, known defects may be acceptable provided that they are are not too severe.

PITS AND DENTS – SUMMARY

 Internal corrosion
 Determine risk of rupture or leakage
 Assess corrosion length, remaining wall thickness
and proximity of other defects
 Dents
 Determine risk of rupture, interference with pigging
 Reduced fatigue life on gas lines
 Assess dent depth and presence of other defects

Any questions?
702 Fundamentals of pipeline inspection

For the assessment of internal corrosion defects, the objective will be to determine the
risk of pipeline rupture or leakage. The assessment should consider the length of the
defect, the remaining pipe wall thickness and the proximity of other defects to establish
if they are significant.

Exposure

PIPELINE EXPOSURE

 Pipelines buried for

 Protection
 Stability
 Insulation
 DNV code
for impact
assessment
 RP-F107

Pipelines are often buried for reasons of protection, thermal insulation or stability. If a
section of pipeline that was previously buried is found to be exposed on the surface then
this needs to be assessed. Depending on the reasons why it was originally buried, it may
be acceptable for a short section of line to be left exposed.

Note that rock dump cannot be used to restore thermal insulation because of the free
flow of water between the rock, so mattresses or trench and burial are needed.

DNV RP-F107 provides a method of assessing the risk to an exposed pipeline with
appropriate methods of protection.
Pipeline inspection 703

EXPOSURE ANOMALY
ASSESSMENT

 Pipeline may be acceptable with no cover

 Exposure risks
 Upheaval buckling
 Loss of stability in storms
 Lateral buckling or coating damage
 Potential for third-party interference
 Impact damage – coating or anode removal
 Cooler arrival temperatures for thermal insulation
 Risks of emulsion, wax or hydrate formation
 Assess potential for deterioration
 Develop into span with bending and VIV risks

Exposed sections of pipeline may still be acceptable, particularly for short lengths.
However, assessment is required.

Where cover was provided to prevent pipe movement, the line may be at risk of
upheaval or lateral buckling. In shallower waters, winter storms may move the pipeline
and initiate buckling or damage coating.

Trawler gear may impact the pipeline and remove coating or anodes.

Where the cover was needed to maintain the temperature of the pipeline, the arrival
properties of the product may become out of specification.

Assessment is undertaken to establish if the exposure is critical and so requires remedial

action, or if the defect is not critical and the pipeline may continue to operate as normal.

For the assessment of spans, the potential problems are bending, buckling and vortex-
induced vibrations.
704 Fundamentals of pipeline inspection

LOSS OF ROCK-DUMP PROTECTION

 Summer survey revealed gas line exposure

 Shallow water of shore approach
 Possible causes
 Insufficient original design – OK until 100 yr storm
 New analysis methods to combine current and wave
 Adjacent pipeline rock dump modified flow pattern
 Settlement of foundation
 Broken armour stone Natural Armour layer
seabed
 Surveys & storm records Under layer

 Identify cause
 Restored before winter

Whilst scour is normally a problem for mobile sandy or silty seabeds exposed to strong
currents, Jee was involved with an exposure of a rock-dumped pipeline.

The summer survey of 2006 of a natural gas export pipeline for a major operator
revealed that the armour layer had been lost in the shallows of the shore approach – in
water depths of 7 m to 9 m (23 ft to 30 ft). The survey showed some 100 m (330 ft) of
exposed pipeline and a short 9 m (30 ft) spanning section. This deterioration had been
worsening over the last three surveys, despite the pipeline protection having been stable
for the previous 17 years. An adjacent pipeline had been rock dumped at about this time
and may have modified the current and wave regime.

However, there are a number of other possibilities. The original size of armour stone
may have been too small. It is common to design for 1 in 100 year return wave and
currents. Such a severe storm may not have occurred until just before the adjacent
pipeline needed additional protection (both lines were affected at the same time). Recent
developments in rock sizing recommend adding algebraically the shear due to waves and
the shear on the rock slope due to current : earlier analysis methods added their
velocities algebraically and then applied them to the slope. The natural seabed used as a
foundation may have settled or scoured away. The armour rock may have been damaged
and broken in storms over the years, leaving a stone size insufficient to provide stability
– many shore protection works fail gradually in this way.

A study of the recent storm data and last five years of annual video and side-scan surveys
is likely to indicate why the armour stone has been lost.

However, due to the length of exposed pipeline and the proximity of winter storms, it is
necessary to order more rock dump immediately to restore the protection for the final 10
to 15 years of the pipeline’s life, and avoid further spans forming.
Pipeline inspection 705

EXPOSURES – SUMMARY

 Exposure caused by
 Scour of sand or silty seabed
 Damage to rock dump
 Cover needed for
 Protection – impact from trawler or dropped object
 Stability – uplift or lateral buckling
 Insulation – flow assurance (wax / hydrate formation)
 Assessment may show no action to be taken
 Monitor in following surveys
Any questions?

Exposed lengths of pipeline can be due to scour of sand or soft sediments or by damage
to rock armour layers.

The assessment needs to consider why the pipeline was covered in the first place.

It may show that no action need be taken immediately, but that the situation requires
monitoring.

ANOMALY ASSESSMENT –
SUMMARY

 Pipeline span
 Assess risk and consequences of buckling and VIV
 Internal corrosion and dent
 Determine risk of rupture or leakage, interference
with pigging and reduced fatigue life
 Exposure
 Cover needed for thermal or impact protection
 Risk of further deterioration, buckling or damage
 Could be precursor to span
 Assessment codes – ASME and DNV
Any questions?

Spans are not necessarily a cause for concern. There are many short spans on pipelines.
However, in the case of spans higher than 0.7 m (2 ft), these become a risk to trawler
706 Fundamentals of pipeline inspection

men. Yielding damage can also occur by buckling or bending under self weight. Fatigue
damage may be caused by VIV in strong currents.

For the assessment of internal corrosion defects, the objective will be to determine the
risk of pipeline rupture or leakage. The assessment should consider the length of the
defect, the remaining pipe wall thickness and the proximity of other defects to establish
if they are significant.

For the assessment of dents in the pipe wall, the objective will be to determine the risk
of pipeline rupture, reduced fatigue life or if the dent depth is sufficient enough to
prevent the passage of pigs. The assessment should consider the unpressurised dent
depth and the presence of other defects, such as gouges, cracks or sharp folds.

Sections of pipelines that become exposed may be acceptable, but the situation must be
assessed and monitored because scour can develop into spans.

ASME and DNV codes can be used to assess anomalies found during the inspection.
Pipeline inspection 707

REMEDIAL WORKS

REMEDIAL WORKS

 Retrofit anodes
 Span correction
 Pipeline stabilisation
 Anti-scour fronds
 Grout bags
 Mattresses
 Rock dump
 Clamp
 Sealant application

Having conducted a survey, it may be necessary to carry out some maintenance tasks
(remedial works). These are detailed in the following slides.

Retrofit anodes tend to be a sled full of zinc placed next to the depleted anode and
electrically connected to the pipe. Before fitting, it would be normal and wise to
establish the cause of the anode depletion.

Span correction is applied where the span is too long and may be prone to overstress or
fatigue due to vortex-induced vibration. The correction takes a number of forms. It
could be the placement of sand or grout bags at mid-span to provide support.
Alternatively mattresses could be placed below and above the span, or the span could be
rockdumped.

If the pipeline has been dented or there is a corrosion defect then a clamp may be placed
around the pipe – either to seal any leaks or to support the dent and prevent fatigue.

Clamps are dealt with below, but other options are covered in the Modifications and
Repairs module.
708 Fundamentals of pipeline inspection

STABILISATION

 Grout bags
 Anti-scour fronds
 Mattresses
 Rock dump

Where the pipeline has deburied and is unstable and moving around on the seabed, pipe
stabilisation is necessary. Were it not corrected, this could lead to a fatigue failure of the
pipe. A number of techniques are available. These include;
■ Anti-scour fronds, as shown in the picture. When placed over the pipeline, these
will trap sediment from the water and build a sand berm which stabilises the
pipeline.
■ Concrete or bitumen mattresses laid over the pipe.
■ Grout bags placed over the pipe.
■ Rock dumped onto the pipe.

STABILISATION MATTRESSES
Pipeline inspection 709

Dense mattresses are placed over the pipeline to stabilise it. These mattresses may be of
bitumen and rope construction, or of concrete blocks on a rope matrix (shown above).
Both, when placed over the pipeline, will conform to the shape of the pipeline and
seabed.

PRINCIPLE OF BOLTED CLAMP

First we will consider clamp repairs, using a landline clamp to show the principles. The
picture shows the clamp being installed around a pipe. Its flanges will be bolted together
on both sides. There are elastomeric seals around each end and down both sides so that
if any fluid does leak out into the annulus, it is contained by the clamp.

SUBSEA HIGH PRESSURE REPAIR

CLAMP

 Sturdy
 Rated up to
200 bar
(3000 psi)
 Hinges open
for installation
 Pipe must be
able to take
weight
710 Fundamentals of pipeline inspection

When the pipeline clamp is scaled up to accommodate large diameter and high-pressure
lines, it can become a very heavy and robust device. The clamp shown above weighs 40
tonnes (44 US tons) and has a steel thickness of 406 mm (16 in). There are hydraulic
clamps on the top to hinge it open and closed. Each bolt is 76 mm (3 in) in diameter, is
over 1 m (40 in) long and needs buoyancy attached to help the diver lift it into place.

From the viewpoint of the pipeline engineer, one crucial calculation to carry out is to
check that the damaged pipeline still has sufficient strength to withstand the self-weight
of the clamp being attached.

CLAMP AS A PRESSURE VESSEL

 Clamp becomes part of pipeline

 Permanent or temporary repair

Damage to pipeline (hole)

Pressure-
retaining
seals

There are two ways to use pipeline repair clamps: as pressure vessels or as structural
supports.

The pressure vessel mode is shown in the diagram above. The clamp forms a sealing
chamber around a leaking pipe. In this mode the clamp is truly pressure containing and
forms part of the pipeline system.
Pipeline inspection 711

CLAMP AS STRUCTURAL SUPPORT

 Clamp not
pressure-
containing
Seal
 Permanent repair
Damage to pipeline BP CATS pipeline
bolted repair clamp
Simple (dent or bend)
seals

Epoxy or grout

The second mode is for structural support.

The clamp is attached to the outside of a dented (but non-leaking) pipe. The annulus
between the clamp and pipe is filled with grout and this is allowed to set. When the
pipeline is repressurised, the hoop stresses from the pipe are transmitted out to the
clamp. The clamp thereby gives structural support and stops the dent flexing, hence
returning the fatigue life of the damaged section back to that of the undamaged pipe.
The clamp is therefore a permanent structural repair but is not pressure containing.

The clamp shown above reinforced the CATS pipeline after it was accidentally dented by
an anchor. The rubber seal at the end is to allow the grout to set, but it is not designed
to withstand the pipeline’s high pressure. The grout needed to reach a strength of 10
N/mm² (1450 psi) in an ambient measured water temperature of 10 ºC (50 ºF) before
the line could be returned from its shut-in pressure of 50 bar (725 psi) to its previous
operating level of 121 bar (1755 psi). The grout reached this strength within 17 hours:
its 28 day characteristic strength was 69 N/mm² (10 000 psi). It took 70 days from the
damage being reported by the harbour master to full operating service being restored
with the bespoke clamp.
712 Fundamentals of pipeline inspection

LEAK REPAIR

 www.Seal-Tite.com
 Pressure-activated sealant for small leaks
 Can be delivered using a batch pig train

One company, Seal-Tite, provides a new leak repair concept without the need for
external intervention. It is claimed that Seal-Tite is able to cure leaks with only a brief
off-line period.

The leak repair sealant is deployed inside the pipeline: for subsea pipeline applications it
can be delivered to the leak site in a train of pigs.

The pressure differential across the leak polymerises the liquid sealant and plugs the leak.
This system has been used successfully offshore and in a wide range of other
applications.

POLYMERISING SEALANT PROCESS

Safety valve Safety valve

Pressure P increase
across leak site as sealant
builds at leak
Flow

starts
polymerization site
Seal element Seal element

1. Fluid escaping through leak site 2. Sealant bridging across leak site

Safety valve

Sealant
flexible
Seal Element polymer seal
at leak site

3. Leak sealed
Pipeline inspection 713

The pictures above illustrate the process of sealing a leak adjacent to a safety valve.

Initially the fluid escapes through the hole. The pressure differential across the leak
starts the polymerisation process. The sealant starts to solidify at the edges of the leak
and the hole is gradually plugged.

REMEDIAL WORKS – SUMMARY

 Unacceptable anomalies must be rectified

 Anodes can be retro-fitted to a pipeline
 There are a range of techniques for
rectifying spans
 A clamp can repair dents or corrosion
 Severe damage will necessitate replacement
of the damaged section
 Leak-sealing technology is available

Any questions?

If anomalies are found to be unacceptable then rectification of the anomaly must be

performed. This could take the form of;
■ Fitting an anode sled to a pipeline.
■ Removing or diminishing a span by using fronds, mattresses, rock dump or
support.
■ Fitting a clamp to a damaged section of pipeline.
■ Using a sealing solution to plug a small leak.

In cases of severe damage the affected section of pipeline will need to be replaced.
714 Fundamentals of pipeline inspection

PIPELINE INSPECTION – SUMMARY

 Need for and basis of pipeline inspections

 Methods used for inspection
 Anomalies to look for in routine inspections
and maintenance
 Methods of assessing anomalies
 Methods available to correct the anomalies
that are found

Any questions?

We have introduced the main concepts for the integrity management of subsea pipelines.

Inspections are necessary to ensure the continued safe operation of the pipeline.
Anomalies are identified and corrected before they are severe enough to cause problems.
The nature and frequency of inspections is determined by a risk-based inspection plan.

External inspections can be done using ROV, side-scan sonar or AUV. Internal
inspections are done using intelligent pigs.

Following an inspection, any anomalies are first identified and then assessed. Those that
are judged to be unacceptable are corrected.