OTC 20993

AKPO: The Subsea Production System

Stuart G Nelson, Total Upstream Production Nigeria Ltd., Lagos, Nigeria

Copyright 2010, Offshore Technology Conference

This paper was prepared for presentation at the 2010 Offshore Technology Conference held in Houston, Texas, USA, 36 May 2010.
This paper was selected for presentation by an OTC program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been
reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect any position of the Offshore Technology Conference, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Offshore Technology Conference is prohibited. Permission to
reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of OTC copyright.

Abstract
It has been said All deep water offshore projects are challenging. Nowhere in the world do projects have such
high commercial pressures as those which take on the challenge of Deep Water. Rarely if ever can it be said that
any one field is a repeat of another. Each therefore raises its own issues which each development must solve
anew. Some lessons are the same, and we forget them at our peril. Some are new challenges, and we then call
on our experience of the past to rise to meet them, and hence all our experience is necessary. In the case of
Akpo, many of the issues were totally new and we called on all of our experience of Deep Water fields in the Gulf
of Mexico and West Africa.
The Akpo field in block OML 130, 200 km offshore Nigeria is in 1400 m water depth. It is a gas/condensate field
with high pressures and high temperatures. One of the greatest challenges is to ensure that condensate and gas
in multiphase flow reach the production facilities without being stopped by hydrates and wax and scale deposition.
The technical challenges alone are significant, but when set against the background of increasing oil prices and
high commercial pressure on the suppliers from more than one operator and more than one field, the challenges
take on a new dimension. Added to that for Akpo were the issues of resources of personnel and manufacturing
capacity in a very buoyant market as well as the new challenge of manufacturing in Nigeria.
This paper also addresses the challenges of the Subsea Production System of the AKPO development. It shows
how the conceptual design principles are encapsulated in the simple acronym RAM and how these lead to some
key issues.
The second most important issue facing all Subsea decisions is the fact that the cost of installation whether by
Drilling Rig or by Installation Vessel FAR exceeds (in most cases) the cost of the equipment itself. Added to that
is the cost of the lost production. This cost is effectively tripled if equipment has to be retrieved and then reinstalled. No Subsea Engineer ever wants to see their equipment return to the surface. Nevertheless things do
go wrong even on a single well - and in a system as large as Akpo, the opportunities for something to go wrong
increase. Such is the inevitable nature of large Systems. The ability to recover and install was seen as vital.
Design for installation was a vital strategy in the design process. This lead to design in order to minimize
installation and retrieval costs.
The overall Field Layout, showing the full subsea network and its connection to the FPSO is shown in Figure 1.

OTC 20993

Fig 2 Typical Drill Centre Layout

Fig 1 - Overall Field Layout

AKPO SUBSEA OVERVIEW

The AKPO development is the latest in a series of large, deep offshore developments by TOTAL. However it is the
first such development by TOTAL in Nigeria. A total of 44 subsea wells ( 22 Production , 20 Water injection , 2
gas injection ) will be drilled during the life of the field in a dispersed and clustered subsea architecture.
The development is based on a standalone facility using a spread moored Floating Production Storage and
Offloading (FPSO) moored in 1314m water.
The field is developed using a complex subsea infrastructure with a network of flowlines, umbilicals and production
facilities.
The subsea network comprises 110 km of flowlines in 4 production loops, 1 gas injection line, 4 water injection
lines, 1 gas export line and 65 km of static and dynamic umbilicals. The wells are tied in to the flowlines through 9
production manifolds and 1 gas injection manifold. All of the manifolds are a standard 4 slot design. The water
injection wells are connected using In Line tee pieces in various locations along the water injection flowlines. The
production system also includes 170 subsea horizontal connectors, control modules located on each tree and
manifold, and a Master Control System located on the FPSO and integrated within the FPSO Production Control
System. A typical such Drill Centre, showing 4 Christmas Trees connected to a manifold, and the Subsea
Umbilical Termination with the Flying Leads to each component is shown in Figure 2.

OTC 20993

Fig 2 Typical Drill Centre Layout

GUIDING PRINCIPLES AT THE OUTSET

All equipment must be designed to meet its operating requirements. That much is self evident. However there are
further levels of design, mostly gained through hard experience that were considered at the start of the field layout.
We had to learn from experience because - as George Bernard Shaw said Success does not consist in never
making mistakes, but in never making them a second time.
The guiding philosophies were:

Engineering of the system to ensure installation and production performance in short a subsea layout aimed
at maximizing Availability.
Ensure that due account is taken of the dangers of Hydrates, produced water and wax.
Equipment design to minimize installation costs and retrieval costs.

Designing the System to achieve these goals was how we went into the first stage Basic Engineering. This is
where those principles were put into the specifications and procedures which then produced the drawings and
hardware.
BASIC ENGINEEERING, SYSTEM ENGINEERING AND INTERFACES
Failure of equipment can and does - lead to shut-in of individual wells or even the whole system. Such failure
was unacceptable. The goal of an Oil and Gas company is to produce Oil and Gas and that can only occur when
the wells are Available. Availability - the central parameter of RAM (Reliability, Availability and Maintenance)
analysis was seen as the main focus for the subsea equipment layout and design. It plays into each phase of
the work. In manufacture to get the equipment delivered; in Installation and in production.
The Project Team embarked on a RAM Study at an early stage. Such studies require everyone to focus on the
failure of individual components, and what that means to the overall system. The system was developed to
ensure that availability for the whole gathering system could be optimized. In the Deep Water environment, the
offset of the wells from the FPSO can lead to problems of hydrates, especially during shutdown. In the case of
the Christmas Tree, this showed the need to maintain steel in certain areas to ensure that enough energy was
retained after shut-in to allow the bores to maintain a temperature above the hydrate temperature for as long as it
could take to get the Methanol to each XT. As the Akpo field will include over 44 wells, that time may be quite long
for the last XT. The issues were evaluated carefully through the first year and special emphasis put on preventing
hydrates. Thermal Analysis proved to be one of the most useful tools in designing the equipment. It was found
that maintaining the temperatures was not only a function of the insulation, but also of the heat energy stored
within components. However only when subject to shallow water testing did all of the possible heat loss paths
become evident. Figures 3A and 3B show examples of both theoretical prediction and measurement.

OTC 20993

Fig 3B Thermographic results form Manifold Cooldown

Fig 3A - Thermal Analysis of Christmas Tree

Similar results were found for the manifolds, where, under certain turn-down conditions, the temperatures in valves
which were not in use could cause lowering of the temperatures in flowing valves a disaster in a shut-down. This
in turn led to a discrete valve philosophy for the wells on each manifold.
In like manner is the need to have access to all valves by ROV. All equipment was subject to stack-up tests and
also to shallow water testing to ensure that this did not happen.
INTERFACES
A picture is worth a thousand words. An interface is worth a thousand pictures.
In all of the equipment made for the Akpo field, there are interfaces both internal and external. The interfaces are
also commercial and contractual as well as technical. The scale of projects such as Akpo requires resources from
all over the world, and across many suppliers. Such are the interfaces of the giant developments. One item of less
than total clarity such as a tolerance on a dimension, or the procurement of a vital piece of rigging, or the supply
of an important spare part can lead to failure or at least delays. A major effort was needed to ensure that the
interfaces could be correctly assessed. That effort ensured that many of potential problems were avoided yet
some inevitably slipped through the net.
In summary the System Engineering role cannot be undervalued.
The effect of failures and the cost of
intervention were very revealing. In the departure from the shore to the deep waters of the Gulf of Benin, the fields
become less and less easily accessible. The fields are further from the shore bases. The cost of even the
simplest intervention becomes very expensive, very quickly. The resources to work on these fields are not always
immediately available. Their cost is high. In simple terms, deep water calls for deep pockets. However, as Tom
Peters has said Almost all quality improvement comes via simplification of design, manufacturing... layout,
processes, and procedures.
MINIMIZING INTERVENTION
The studies of the System showed how failures could be EXTREMELY expensive due to the costs of rigs and/or
installation spreads. This aspect of the costs is not always well understood, but to illustrate the problem, consider a
XT. The XT may cost around $7M, but the rig to perform the completion and clean-up may cost $20M. Any
retrieval activities that are required as a result of a failure can double that cost. This understanding generated the
need for design of the equipment as a System which sought to minimize the costs of intervention. This thought
generated the concept of 3 Rs.

RELIABILITY
REDUNDANCY
RETRIEVABILITY

RELIABILITY is the first and most important key to successful operation.

Lack of reliability can - and eventually

OTC 20993

will - lead to excessive operational costs in deep water fields. Thus ensuring Reliability and having confidence in
the equipment - was a paramount concern. The keys to Reliability are:

Extensive Qualification before manufacture.

Quality Control throughout the manufacturing and testing process.
Qualification Testing
Quality Assurance and Control
Factory Acceptance Testing
Extended (each level of integration) testing
Cool down/ Shallow water / System Integration testing

The Akpo Project embarked on one of the single greatest programs of Qualification testing to date.
REDUNDANCY. All critical equipment must have at least one level of redundancy. This is not always possible for
high-level components such as Christmas Trees. However, system failure is often due to failure of smaller
components such as filters and transducers. For these items, redundancy is relatively simple. For example, each
XT was served by two separate control lines and no two XT were served by the same two control lines.
RETRIEVABILITY. Design components for ease of installation, retrieval, and replacement as modules. (Note In
the event, this approach proved to have huge benefits due to a late discovery which will be discussed later in this
paper).
The Modular approach included the selection of:
Vertical XT so that failure of a component on a XT did not lead to removal of the Tubing Hanger and hence the
completion.
Horizontal connectors which allow disconnection of a jumper and removal of the XT or manifold without
having to disconnect the jumper at each end.
Common Running Tools for all Choke Inserts, XT Control modules and Manifold Control Modules.
Separate (and dual redundant) Electrical Flying Leads, Hydraulic Flying leads and Jumpers were employed so
that only the failed component need be replaced.
Intervention to be maintained as much as possible on site and by FPSO (e.g. Hydrate Remediation System
for the flowline jumpers is initiated by the platform).
IMPLEMENTING THE PHILOSOPHY PACKAGE BY-PACKAGE
In each package, certain key features lead to the design decisions that were made. Some have been alluded to
already. These are now outlined in more detail.

CHRISTMAS TREES AND WELLHEADS

MANIFOLDS

CONTROLS

DRILLING/COMPLETION TOOLS

CONNECTORS AND RUNNING TOOLS

Each of these will now be expanded upon in more detail.

OTC 20993

CHRISTMAS TREES AND WELLHEADS

As was mentioned earlier, the Christmas Tree design selected was a Vertical (Dual Bore - 5 x 2) XT (See Fig 4).
There were various reasons for selecting this design. One in particular was the complexity of the Intelligent Well
Completions (IWC) which intersects multiple reservoirs. In the event of a problem on a XT, the preferred approach
was to leave the Completion intact if the XT should need to be removed.

Fig 4 AKPO Dual Bore Production XT

This led to some constraints on the size and number of penetrations through the tubing hanger. The inclusion of
both Main bore and Annulus within the Tubing Hanger leaves little room for penetrations as can be seen in Figure
5.

Fig 5 Tubing Hanger showing Penetrations

Once again on insulation, the temperature of the flowing product was forecast to reach 115C. This was in excess
of the previous more commonly used GPSU (Polyurethane) pour-in-place insulation methods. An extensive prequalification was performed which led to the choice of a Phenolic Resin based Syntactic Pack-in-place solution.
As later events on other fields proved, this was a good choice. The pack in place, although requiring more manual
effort, ensured that there was no opportunity for through cracks which can occur on the homogenous pour-inplace insulation systems.

OTC 20993

Other new developments were the use of a new dual line electrical penetrator design. The Intelligent Well
Completion called for more electrical cables to be run down-hole than had previously been run. The Dual Bore
design however leaves little room for increasing the number of penetrations, and hence it was decided to use a
penetrator with multiple cores. This was a new development by Diamould. As it was new it called for further
Qualification testing and has undergone much evolution, particularly with regard to optimizing its make-up and
completion on an offshore rig immediately prior to deployment.
MANIFOLDS DESIGN FOR OPERATIONS AND INSTALLATION.
During Basic design, the top hole locations were optimized, which then allowed a standard 4 well design to be
employed. By choosing the 4 well design, it further opened up the marketplace for installation vessels.
Standardization of the design was a driving factor as this was to be the first time that Total would build such
complex subsea equipment in Nigeria. The design was by Cameron in Houston. This was the first development
of a manifold to use the new CHC (Cameron Horizontal Connector). In the image below, note the vertical guidance
rails used to ensure the exact location of both the connectors and the doghouses. Also note the topmost section
which is the Hydraulic Distribution Unit which too incorporates guidance as the HDU is also separately retrievable.
The overall layout of each manifold was maintained throughout. The design of the first manifolds in each
production loop was slightly different in that a gas-lift port was provided for later field life. All other aspects
remained identical for the production manifolds. Although the temperatures within the manifold were expected to
be lower than in the XT, it was still marginal for the use of GPSU in the upstream sections of the manifold and
hence the Phenolic resin pack-in-place solution was again chosen.
The initial design respected -the function of the manifold - which included HAZOP and RAM analysis for the
selection of the valves and their failure criteria. Then followed the refinement of the design minimize the number of
lifts.

Fig 6 Manifold, Base Frame and Pile

By careful analysis of the thermal losses and the layout of the valves, each manifold was reduced to just two lifts,
with a maximum of 120 Te lift weight. Judicious analysis of the soil characteristics allowed the use of large
diameter suction piles with thin steel walls. The use of windows to allow for rapid water evacuation on landing,
further allowed the piles to reduce from an original 92 Tonnes to 32 Tonnes each. A tubular base frame also
allowed for installation shock loads and an increased vertical landing speed. The manifold itself (below) shows
how the guidance system was configured to allow for the installation of the spools (flowlines) and jumpers (wells)
alongside each other. Also in the view shown in Figure 6 can be seen the receptacles for the Horizontal
Connectors (dubbed CHC).

OTC 20993

Each manifold contains four Multiphase Flowmeters (MPFM) one for each for each well. Such meters avoid the
need for a separate Well Test line and are therefore highly cost-effective. The selection of the MPFM includes a
water-cut meter so that early indication is available to the operators of the reservoir water production. At very low
water levels, the results can be in error, but always allow a safe approach to be taken to avoid the risk of hydrates.
In following the overall strategy for increasing Availability, the manifolds also incorporate a retrievable Manifold
Control Module, which is almost identical to the XT Control Module in design and which uses the same
accumulators and the same running tool to allow for ease of change out. A further refinement is the separately
retrievable Hydraulic Distribution Unit which means that a manifold with any hydraulic problems can be left in
service and the HDU retrieved individually to implement any repairs.
The first two production manifolds were built in Berwick, Louisiana which was the Cameron facility closest to the
venue for the Cool-down testing, which took place in Mobile, Alabama. The proximity to the design team in
Houston also permitted rapid response to changes as the design evolved during fabrication. The photograph
below shows a manifold being loaded onto a submersible-barge at Mobile where it would be prepared on the
barge with heating lines and data logging cables, then immersed and the Cool-down tests performed. The
experience gained on the earlier tests on the XT in Norway ensured that the manifolds passed the time criterion of
>14 hours from 48C to 23C - on the first attempt.
The remaining 8 production manifolds as well as all 10 suction piles, Final Assembly and System Integration
Testing of the manifolds took place in Warri in Nigeria. It was the first time that the yard had made such complex
deep water equipment, and brought with it the challenges of logistics particularly for the supply of duplex stainless
steel, insulation materials and personnel to perform the demanding welding and testing. It is without doubt one of
the great achievements of this project that it successfully managed to get all of the equipment and materials to
Warri from all over the world in order to build these manifolds and to make them work. The range of skills needed
cannot be underestimated. Maintaining control over the manufacture, welding and testing was perhaps the
greatest challenge and every possible step was taken to ensure the Quality of the work across the range of
activities. Nevertheless, one vital point was missed which is discussed in a later section, but which nevertheless
succeed in proving the validity of our philosophy. Figure 7 shows six manifold pile structure and the manifold base
frames both installed together leaving the site at Warri in early 2008.

Fig 7 - Manifold Piles and Base Frames leaving Globestar Yard, Warri

CONTROLS

Pull the string, and it will follow wherever you wish. Push it, and it will go nowhere at all (Dwight D Eisenhower).
Any system is of no use if not controlled correctly, and the Akpo Field is a huge system to control. There were two
aspects to take into account

OTC 20993

Hardware
Software.

The Hardware consists of the physical entities which on the seabed are the

Umbilicals (supplied under the UFR contract)

Control Modules (on each manifold and each XT)
Flying leads (Hydraulic and Electrical)
Termination units (see illustration below)
Multiphase Flow Meters

Power and signal share the same conductors in the Flying leads and umbilical. In line with the guiding
philosophies, there is redundancy and additional electrical and hydraulic service lines are built into the umbilical.
In fact the redundancy of the umbilicals and Flying leads can be summarized as follows:

Each well has two redundant lines to provide its services

No two wells are served by the same two lines.

The Stainless Steel Tube Hydraulic Flying leads were made in the UK at DUCO, and an example is shown in
Figure 8.

Fig 8 Hydraulic Flying Lead on reel at DUCO, Newcastle

All data is locally stored and conditioned before being sent back through the umbilical to the FPSO. In addition,
data can be stored prior to connection to the production umbilical by means of the SWIMS (Subsea Well
Interference Module). Using this system, the influence of one well on another remote well can be determined,
thereby providing a better understanding of the communication throughout the reservoir. This is particularly
relevant to the down-hole pressure monitoring in the wells.
To ensure that the demands of any one tree or manifold are not compromised by the demands of others, each
Control Module has hydraulic accumulators storing hydraulic fluid under pressure sufficient for the two operations
of each valve on the tree or manifold which that module services.

10

OTC 20993

Fig 9 Subsea Control Module showing Accumulators

The Subsea Umbilical Terminations (SUTs), shown in Figure 10, allow the umbilical to send the power, signal,
chemicals and hydraulic fluid to each manifold and then to each well. The SUT were all made by Cameron in
Celle, Germany and terminated to each length of the umbilical at the Nexans facility in Norway.

Fig 10 Subsea Umbilical Termination Unit and Bridging Jumper

On the Topsides the hardware consists of the

Umbilical terminations
Chemical Injection pumps and manifolds
Master Subsea Control System
Integrated Process Control System (operator interface)

The Software is the logic of the whole subsea system. This resides in the heart of the Subsea Control System
computer on the FPSO. It is inextricably linked with the needs of the FPSO. It integrates with the Production

OTC 20993

11

Control System. The ICSS covers the process control in addition to Fire & Gas and shutdown systems for the hull,
topsides and subsea equipment. It provides capability for centralized control of the full installation through
workstations in a single control room. It must link with the needs of the production and export of condensate, gas
and injection of water. It must respond to Shutdown requests and control the opening and closing of valves in line
with those needs In addition an Overall Control of Well and Riser (OCWR) module was developed and included
within the field Integrated Control & Safety System (ICSS). The OCWR is a tool to allow, at operator demand, to
change sequences which are repetitive and simple in order to free up panel operators for other tasks. The main
functions are:
Production wells ramp up & optimization
Production risers optimization ( slug, flow balancing)
Topsides optimization ( inlet separator balance , slug control )
Preservation of the filed ( hydrate concerns)
Injection well optimization
IWOCS and RUNNING TOOLS
A vast array of tools is required to support the installation. Many were developed for Akpo as part of the scope of
work with Cameron.
The tools are subject to probably the most demanding environment. Every well must undergo the same
requirements to drill and install the conductors, casing and tubing. Each must therefore use the Running Tools on
many occasions. Items such as the Landing String (shown below during its stack up testing in Leeds) are subject
to enormous forces and items such as hoses, seals and swivel joints undergo countless loadings throughout their
history.
The choice of the Dual Bore design of XT also necessitated the design of the Lower Riser package and
Emergency Disconnect Package for the installation of the XT.
In the case of many of the tools, the main contractor must put the work out to specialist companies, while retaining
much of the scope of supply for the XT within its own boundaries. Since the precision with which such large object
must be set is governed by the smallest of electrical connectors at the Tubing hanger.
Precision is vital, and so too therefore is the machining of the mating components frequently within microns. It is
a major interface to manage incorporating as it does both the very large and the very small and by remotely
operated vehicles. Fig 11 shows the components in the installation process, but this section itself would require a
paper on its own, incorporating as it does all include the many umbilicals, risers, connectors and running tools.

Fig 11 Installation of Xmas Tree on Wellhead showing EDP and LRP

12

OTC 20993

It goes without saying that such equipment suffers from considerable wear-and-tear. Without replacements and
back-up, the costs of breakdown become astronomical. The sheer size, number and usage of these tools means
that at some point there will be components out-of-action and that means the potential for high costs and long
delays. The need to have an early and comprehensive strategy and budget for spares and maintenance is
without question a fundamental issue.
CONNECTORS AND RUNNING TOOLS

In the earlier sections of this paper, the modular approach was highlighted as a means to reducing intervention
costs. Total had gained experience with both the standard Vertical connectors and the less common Horizontal
connectors. The experience over the years has tended to show that the horizontal connectors allowed for easier
installation of jumpers and spools by allowing the jumper to be installed and left for another vessel to come to
make the connections at any time thereafter. The Horizontal Connector ( see Figure 12) while embodying many
of the same collets as the Vertical Connector was a first for Cameron. It required specialist Running Tools to be
made to go with it, and also principles for guiding the connectors onto the hubs even when adjacent to each
other.

Fig 12 3D Model of CHC Connector to be mated

A new approach was needed for the running tool and again, new tools had to be developed, and that in turn
required the design of all components using the tools to incorporate the same coarse/fine guidance systems (Fig
13)

Fig 13 CHC Running Tool

OTC 20993

13

In order to ensure that the Cool-down requirements were respected since a huge loss of energy would occur at
the bare connector - it was necessary to design a post-installed insulation cover more commonly referred to as
an Insulation Doghouse. Shortly after the design was commenced, data started to filter through from other fields
which showed the planned design exhibited very large convection-induced heat losses. Some measurements
showed a thermal efficiency as low as 3%. As such another solution had to be found.
The root cause of the heat losses on the original design was down to poor sealing. The conventional doghouses
used a complex arrangement of three opening parts. The Akpo design cut this to two only, and a simple, single
axis seal (see Figure 14).

Fig 14 Two piece Doghouse installed on made up CHC connection

Yet again, a new design i.e. the new Doghouse - called for a new installation tool (Figure 15). This was perhaps
the most complex of all the tools. It was designed and made by Fugro-ImpROV in Aberdeen and proved itself first
time, as did the CHC running tool made by Cameron. Here we must also talk about the interfaces. The jumpers
were made by the installation contractor Saipem to the exact dimensions measured between the corresponding
hubs of manifold and XT. The jumpers had to be welded to these dimensions and only at the very end could the
testing of the jumper and the connectors come together. Back seat testing tubing had to be run through insulation
and to through the connectors to the control panels. In addition, the Doghouse depends on the ability to seal on
the pipe and in this case on the insulation, while avoiding any of the control tubing. This area needed particular
care as it falls exactly at the interface of two, separate and geographically displaced contracts.

Fig 15 Insulation Doghouse Running Tool

14

OTC 20993

In all of the connections in total over 140 connections involving the connectors, doghouses and running tools
have been made to date. There have been no failures or refusals to mate were encountered. Although these were
'firsts' for the engineers it bears testimony to their thorough approach that the tools worked so well. One area we
have yet to test fully will be the replacement of seals and cleaning of the hubs and connectors.
QUALIFICATION AND QUALITY
Equipment qualification began with the development of the field architecture. The layout of the fields and the wells
was the first stage. From the basic layout came the P&IDs for each well and flowline. From these came the
inventory of valves, actuators, meters, sensors, and connectors. Over 115 Qualification tests for AKPO SPS
(Valves, Connectors, pressure transmitters, SCM, HFL and EFL). Key tests performed:

Gate valves and chokes (following API 6A App F PR2)

Tubing Hanger annulus closure device (Gas & Water dynamic cycling pressure tests, Gas body pressure
tests, Debris tolerance function & pressure tests, Sand slurry function & pressure tests)
Subsea Control Module (hyperbaric test, lockdown tests, function test)
Hydraulic Flying Leads (Characterization and maneuverability).

Each component was subjected to a thorough analysis of its operating conditions. The following lists are not
exhaustive but illustrate the point.
Hydraulic component review:
z External pressure
z Internal Pressure
z Pressure differential
z Hydraulic operating pressure
z Thermal conditions
z Life of field conditions
z Frequency of operation
z Total number of cycles anticipated
Electrical component review:
z Operating voltage and current
z External pressure sealing
z Internal cavity pressure differential
z Thermal conditions
z Life of field conditions
z Frequency of make and break (connectors)
z Vibration and shock
Using this concise approach, each component was assessed from the vendors available data. In most cases, the
vendors had carried out testing to industry standards. There are good guidelines available. Under the API,
guidelines for example are API 6A and API 17D. These list in detail the standards of pressure tests required. API
17D also takes account of the environmental conditions. Furthermore, it calls for extended cycle testing of valves.
The Akpo Qualification program taken as the forerunner of other developments in the same block undertook
115 additional Qualification tests to ensure that the equipment covered the full design envelope.
The equipment is now operating and operating well. There are always new conditions that were not foreseen at
the outset, but much has been gained towards the goal of increased Availability by ensuring that the three stages
of Qualification, Quality Control and Testing were followed.
QUALITY ASSURANCE AND CONTROL.
There is an axiom: You get what you Inspect, not what you Expect. The approach to QA/QC for Akpo was very
focused and cost effective. The equipment was qualified for use generically. This is akin to testing under
laboratory conditions. Real equipment is manufactured and used under real conditions. The high capital cost of
the equipment and high cost of remedial intervention for equipment failure or malfunction warrants extra attention
to detail during manufacturing and fabrication. This basic philosophy determined the QA/QC approach. Even so
things will slip through. Therefore the approach can be seen as minimising losses rather than prevention

OTC 20993

15

altogether although that should be the goal.

QA/QC PHILOSOPHY
With this premise the project, team developed the Quality Plans and procedures for Quality Assurance.
Unfortunately, this premise is somewhat archaic in some modern project cultures. However, the experiences on
Akpo support this approach to Quality Assurance. The primary objectives of the Quality Assurance program were
to get it right the first time.
Additionally, all manufacturers were required to produce a detailed manufacturing plan or inspection and test plan
(ITP). The development of the manufacturing plans was finalized with face-to-face meeting and discussions to
ensure both parties understood all the requirements and limitations of the plan procedures. The process helped
build understanding, trust, and cooperation between project team members and the supplier personnel. The
process involved learning on both sides.
One example of this is that we, on the project side, observed that many times the manufacturer would default to
standard procedures and processes. This was not always an appropriate approach. There were specific
procedures to be followed which were not the standard methods used by the contractor. As a result, some of the
requirements in the equipment specifications were unwittingly excluded in the manufacturing and ITP plans. In
most cases, it was simple case of not reading the specifications.
The model for implementing the ITP was delegated to the working level.

Service personnel and wherever possible (although subcontracts and internal commercial pressure often
prevent this) must be involved in the Quality Plan.
Anticipate Murphy's Law applies which applies no matter what one does to take it out of the equation.
Be rigorous almost ruthless - in adherence to the plan and in maintaining regular visits, audits and witnessing
especially in subcontractor venues.

ASSEMBLY TESTING
The Akpo subsea architecture is very complex, as discussed in accompanying papers and as seen in the attached
field schematic.
At least 30 major vendors and contractors in North America Europe, the Far East and Africa were involved in the
supply of the primary components and fabricated assemblies. Because of the project schedule and the diversity
of locations, it proved impossible to assemble all components in a single location for an all-encompassing
Systems Integration Test. At the same time, it was concluded early in the project that it would be necessary to
conduct extensive Integration testing of the assembled components. This further level was the System Integration
Testing (SIT) generally refers to an extensive series of tests performed on all of the subsea control system
equipment. SIT was intended to pick up where Factory Acceptance Testing (FAT) ended. Two or more adjacent
subsea components were brought together to confirm that they would
z
z

Fit together, physically

Perform their functions as designed.

In addition to test programs performed on the subsea control equipment, an extensive series of tests were
conducted around the subsea production Christmas trees and all associated components and accessories. The
test series was designated Extended Factory Acceptance Test (EFAT). Most of the testing was carried out by
the Akpo Project Team (See Figures 16A and 16B). One of the great benefits of the test program was in finding
the physical clashes and conflicts, and instances of sub-optimal performance were identified. These
shortcomings were assessed and corrective action taken prior to the start of the offshore installation campaign. It
is difficult to assess the cost of rectifying these shortcomings if they had not been discovered during SIT. They
would have been found during installation. Offshore spread rates ranged between $250,000 and $1,00,000 per
day. Intangible benefits also resulted by familiarizing personnel with the equipment. This undoubtedly resulted in
optimization of procedures and further reductions in the duration of installation and minimization of risk. All of the
subsea control system equipment has now been successfully installed on the seabed with minimal difficulties.

16

OTC 20993

Intervention
Umbilical

Landing String EFAT in Leeds

Fig 16 A and B showing Stack Up Testing of IWOCS Equipment in Leeds

TEST OF THE APPROACH MANIFOLD LEAKS

Earlier in this paper we alluded to the decisions to use a modular approach, and to make installation and
retrievability as efficient as possible. The test of this philosophy came in January 2009. The Manifolds had been
installed by the Saipem FDS, a large purpose built installation vessel. All of the spools had been connected and
the commissioning of the subsea system was under way. On the first production loop to be tested however, it was
found that the flowline could not hold pressure above 240b (requirement was for 340 b). The medium for testing
was water and so an eco-friendly, water soluble dye was deployed with the test water. Leakage could be seen as
a cloudy mass coming from under the insulation on the main headers in two manifolds. The leaks were within the
manifold itself. The ROV visuals could not pinpoint the exact location of the leaks. The leaks had to be found and
repaired and that meant retrieval of the manifold. At this time, all spools connecting the manifold to the flowline
were in place, as well as all the jumpers to wells and flying leads. Also the FDS had already left the site. A much
smaller vessel had to be used.
The manifolds had been designed for installation by as wide as possible a range of vessels. In this case the
SAIPEM S3000 was available, and its heave compensated crane capacity was ideally suited to lift the manifold
and with enough deck space to store the manifold
Approximately one week after discovering the leak, all of the necessary rigging and spreader bars had been
assembled on the S3000. The connectors were all removed at the manifold only and the flying leads hydraulic
and electrical - disconnected and parked. Again the only disconnections necessary were at the manifold. The
jumpers remained in place on the seabed.
The manifold was lifted and after a pressure test on deck and judicious removal of small inspection areas of the
insulation, it was found that some bolts holding the Pressure Temperature Transmitters (PTT) in place were loose.
The insulation was removed, the gaskets and bolts replaced and the insulation reapplied. The manifold was
retested and redeployed within 72 hours of its removal from the seabed.

OTC 20993

17

Other examples were found on a total of 5 manifolds, and the team became adept at retrieval, repair and reinstallation until the critical path became the curing time of the insulation resin.
We were concerned as the manifolds had all held pressure during surface tests. However subsequent laboratory
trials showed that under atmospheric conditions, flanges with bolts made up with only 10% of the requisite torque
would still hold pressure safely. The problems were found when such low torque bolts went under water. Then
as a result of the hydrostatic pressure and the flanges not being in contact, the load was further removed from the
flanges which resulted in the 'un-seating- of the gasket. Once so displaced, the gaskets could not hold any
pressure.
This one simple point had been a missed during a routine removal of an instrument done to prevent damage during
welding. The need to maintain vigilance over all aspects of all pressure retaining equipment cannot be
overstressed, however the relative efficiency of the repairs proved the point regarding retrievability.

CONCLUSIONS
The Akpo Development brought together the largest subsea production system to date in Nigeria deep water. It
was a feat which saw the design fabrication, integration and testing of over 4000 items of equipment. The
principles of the 3 R's Reliability, Redundancy and Retrievabilty were used in the effort towards greater
Availability and thereby the profitability of the field. There were many 'firsts' along the way, in the search towards
achieving these goals. The vertical XT, the horizontal connectors, dual-redundant controls to each XT and
individual flying leads were all new developments in the project and due to unforeseeable circumstances, proved
the concept of the modular approach. Akpo is now in production and the equipment working well. It has provided a
huge and significant benchmark for future subsea developments.