OTC 19637 Kikeh Development: Subsea Equipment Installation Challenges ...offshorelab.org/documents/Kikeh_Subsea_Equipment.pdf · Kikeh Development: Subsea Equipment Installation Challenges

OTC 19637

Kikeh Development: Subsea Equipment Installation Challenges for Malaysia’s First Deepwater Development G. Murray and D. Lowther, Technip Subsea 7 Asia Pacific; A. Ledingham and T. J. Stensgaard, Murphy Sabah Oil Co., Ltd.

Copyright 2008, Offshore Technology Conference This paper was prepared for presentation at the 2008 Offshore Technology Conference held in Houston, Texas, U.S.A., 5–8 May 2008. 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 The Kikeh Development is the first deepwater project in Malaysia located Offshore Eastern Malaysia in the South China Sea with a water depth of 1320 m. Kikeh is produced via a Spar and a subsea water injection and production system back to a Floating Production Storage and Offloading (FPSO) unit (see Figure 1). Kikeh is planned to be developed using 34 wells and initial production commenced in August 2007. The main subsea installation portion of the Kikeh development was executed through an Engineering, Procurement, Con-struction, Installation, and Commissioning Support contract with Technip Marine (Malaysia) Sdn Bhd. The installation scope of work included the following:

5 manifold support piles* 5 subsea manifolds* 5 flexible risers and flowlines with a total

length of over 31 km of flexible pipe 7 riser holdback anchor piles 4 electro-hydraulic control (EHC) umbili-

cals (2 dynamic/static and 2 infield static)*

5 rigid spools c/w vertical connection mod-ule (VCM) clamp connectors (out of 14 in total)*

5 subsea distribution units* 6 hydraulic flying leads (out of 27 in total)* 6 electrical flying leads (out of 48 in total)* Pre-commissioning of the subsea systems

* Free issued from subsea equipment EPCC contractor

The paper will discuss the process of offshore installation vessel selection including the effects on the decision process of utilizing local equipment versus mobilizing deepwater construction equipment from other deepwater basins around the world. In addition, with deepwater Sabah being a greenfield deepwater basin, a number of challenges arose in relation to availability of equipment within the region and logistics to support the deepwater construction operations. Highlights from Phase I, Phase II, and Phase III installation programs are also presented. The Phase II installation program had a late installation vessel change which significantly impacted the methodology and procedures associated with installing the subsea tree to manifold rigid spools, installation of flying leads and performance of

Figure 1 - Kikeh Deepwater Development Field Schematic

2 OTC 19637

2nd end VCM connections, all which will be discussed in the paper. Pre-commissioning strategy and key project take-aways are also addressed. Introduction The Engineering, Procurement, Construction, Installation and Commissioning (EPCIC) Contract for the Provision of In-field Pipelines and Risers and the Installation of Subsea Equipment was awarded by Murphy Sabah Oil Co., Ltd. to Technip Marine (Malaysia) Sdn Bhd in April 2005. Whilst these types of deepwater subsea developments have been brought into production in other deepwater basins for a number of years, particularly in West Africa and the Gulf of Mexico, the Kikeh Development, as a first for South East Asia, brought about a number of challenges and specific requirements to both Operator and Contractor that necessitated close cooperation in terms of sharing of installation assets across a number of contracts associated with the Kikeh development in order to maintain the overall project schedule. Due to the nature of this type of subsea deepwater project involving the supply and installation of flexible flowlines and risers, the EPCIC contractor elected to utilize a joint venture partnership to execute the project on its behalf. The project team was established in Perth, Western Australia from where all design engineering, procurement activities, and preparation of installation procedures was undertaken. Offshore Vessel Selection Upon contract award the initial priority was to review the critical installation activities with regard to overall vessel capabilities. The primary vessel capacity requirements were crane lift capabilities, flexible pipe lay tensions, and overall deck storage capacity to handle the products to be installed. To assess the installation vessel requirements, some initial engineering was conducted to better understand maximum lift and deployment loads associated with the installation program. The installation program, as defined by the project execution strategy, split the installation scope of work into two distinct phases. Installation of subsea structures, laying of flowlines, laying of static umbilical sections, and pre-installation and wet storing of the dynamic flexible risers and dynamic umbilical sections prior to the arrival and hook up of the FPSO formed the Phase I workscope. Recovery and pull-in of the dynamic flexible risers and umbilical sections into the FPSO turret along with installation of rigid jumpers and flying leads was the scope of work to be performed in Phase II directly following hook-up of the FPSO. Pre-installation of the flexible flowlines, risers, and umbilicals was specified by the client to facilitate a more rapid hookup of the field following FPSO mooring at site and to disassociate the FPSO installation from the flexible pipe and umbilical installation program to reduce schedule risk. Phase III, outside the scope of work of the EPCIC contract, is the period (still ongoing) during which the remaining 10 subsea wells are tied in to the previously installed subsea infrastructure. For Phase I, the largest scope of work involved the installation of the flexible flowlines, risers, and umbilicals. Initial analysis determined the installation loads for the 10-inch water injection flowline and riser would be in the order of 280-300 t. Significant lay tension requirements were primarily driven by the close to 1400 m water depth, large pipe diameter, and laying of the pipe flooded to avoid flexible pipe collapse. The magnitude of the tensions calculated confirmed the requirement for the newly designed flexible lay system - Portable Pipelay System (PPS-01) - which had a capacity of 350 t and, at that time, was under construction in France. Given the overall dimensions and weight of the PPS-01 (17.5 m x 11.8 m x 37.4 m high and 960 t) and the operational power requirements of 1.4 MW, it became apparent that there was no available vessel within the SE Asia / Australia arena which would be able to accommodate the lay system. Therefore identifying and securing an appropriate vessel from another deepwater operating arena would be a priority for the project team. In total there were to be 35 reels of product to be installed on the project; 22 flowline reels, 10 riser reels, and 3 umbilical reels. To accommodate this number of reels, a Heavy Lift transportation Vessel (HLV) was required to transport the reels from Europe and act as a storage vessel from which reels would be transferred to the installation vessel. In addition, in order to maintain an efficient installation program, the installation vessel would be required to accommodate a minimum of 6 reels on its deck, along with the PPS-01 spread, 2 work class ROV systems, and a pre-commissioning spread that would permit flushing/filling of the flowlines with inhibited fluids following lay operations and testing of the umbilicals following installation.

OTC 19637 3

Once the specific equipment criteria had been established, a review of vessels throughout the joint venture fleets identified that the Deep Pioneer (see Figure 2) had the deck capacity to accommodate the required flowline and riser installation spread and also had sufficient additional available power to drive the PPS-01 system and was therefore allocated to the project. Mobilization of the vessel in Europe was driven by the availability of the newly built PPS-01 system which was scheduled to be completed in July 2006. Following mobilization and final commissioning of the PPS-01 system on the Deep Pioneer in Le Trait, France, the vessel was to transit from Europe to Singapore for final project mobilization. Based on the completion of the PPS-01 and the

mobilization activities described herein, a window to complete the Phase I flowline, riser, and umbilical installation scope of work was agreed with Murphy from October 2006 to end of January 2007. For the subsea structure scope of work, analysis determined that the largest dynamic load to be installed was the 10-inch water injection riser hold back suction anchor with a maximum assessed loading of 154 t. Riser holdback piles were utilized to reduce both the overall footprint of the subsea development and the flexible pipe flowline lengths. Given the relatively close proximity of the Kikeh Field to the offshore supply base of Labuan (around 8 hours transit), multiple transits of the installation vessel between the field and supply base to reload equipment was considered not to be cost prohibitive provided that a medium sized construction vessel was used. Evaluation of available vessels based within the Asia Pacific Region determined that the Rockwater 2 (see Figure 3) would be the most appropriate selection, albeit that the crane would have to be upgraded to 200 t offshore lift capacity and the crane wire drum changed out to accommodate 2200 meters of wire to enable the deployment of the heaviest structures to the seabed. Deck layouts confirmed that all 12 suction anchors, 5 manifolds, 5 Subsea Distribution Units (SDUs) and 2 temporary PLETS could be installed with 3 vessel loadouts from Labuan and an overall installation program of circa 25 days. Integrity verification of the flowlines and risers as left at the conclusion of Phase I operations was to be provided by means of a hydrostatic pressure testing program. As all lines were to be flooded with treated water from the Deep Pioneer prior to second end lay down, a means of pressurising each of the flowline and riser systems from their respective manifolds was required to be developed. With the project keen to avoid hydrotesting the flow systems from the pipelay vessel due to the significant schedule and cost implications, alternative integrity pressure testing means were investigated. One system which appeared to have significant merit was the use of a subsea testing unit. However, the systems available consisted of a large seabed based flooding and testing unit with a weight in excess of 15 t and require a vessel with adequate cranage to deploy and recover the unit. This large seabed unit was also deemed cost prohibitive and, therefore, the project team set out to develop a small pumping skid which could be ROV deployed and had the capability of pressurising each of the lines up to the required leak test pressure. The ability to use this small ROV mounted pressurization skid meant a small locally based DP vessel without any cranage was all that would be required to support the Phase I flowline integrity verification workscope. A Malaysian company owned vessel, the Allied Shield, was selected by the project team for this operation.

Figure 3 – Rockwater 2

Figure 2 - Deep Pioneer with PPS01 Lay System

4 OTC 19637

For Phase II, the critical installation activity was the recovery and transfer to the FPSO of the wet stored flexible risers and dynamic umbilical sections. As the Deep Pioneer with the PPS-01 was not going to be available for Phase II operations, an alternative means of lifting the risers and transferring the load to the FPSO pull-in winch wire had to be developed. Recovery and hang off of the riser above the water surface on the recovery vessel was not really a viable option due to the significant loads involved (up to 280 t for the 10” WI riser) therefore a means of completing the transfer of the riser to the FPSO pull-in rigging with the riser termination remaining subsea had to be engineered. It was also apparent that mobilising a heavy DP construction vessel with large crane capacity from another region for a stand alone 20 day riser recovery program in South East Asia was not a realistic option in terms of cost or availability in a market that was already over heated. This led the project team to assess the viability of utilising a large capacity DP Anchor Handling Vessel (AHV) with winch capacity in excess of 400 t as an alternative solution. The Normand Ivan (see Figure 4) was a vessel that Murphy had placed on long term charter to install deep water pre-set moorings, carry out rig moves and provide general operations support to the Ocean Rover MODU within the Kikeh Field. Initial evaluation confirmed that with its 500 t winch and stern roller capacity and twin aft Karm forks rated to 750 t, a workable methodology could be developed for recovery and transfer of all 5 risers and 2 dynamic umbilicals utilising the Normand Ivan. Furthermore, due to the pre-set mooring workscope, the vessel was already fitted out with a work class ROV. A “best for project” decision was made to make the vessel available for this Phase II workscope. Subsequent detailed engineering, however, was to identify a number of additional difficulties that had to be addressed and overcome as is discussed in greater detail later in this paper. Other Phase II activities involved the installation of the 5 rigid spools, 6 hydraulic flying leads (HFLs) and 6 electrical flying leads (EFLs), plus the recovery of the 2nd end of a 6” WI flowline VCM from a temporary pipeline end termination (PLET) and connection to a subsea tree. With deployed loads at the seabed not exceeding 30 t and nothing onerous in terms of deck space requirements, the Rockwater 2 was considered to be the most appropriate vessel from which to undertake this installation workscope with the heave compensated crane considered to be of benefit in relation to landing of the rigid spool VCMs onto the receiving manifold and subsea tree hubs. As for Phase III activities, the plan was to install rigid jumpers and hydraulic flying leads, based on lessons learned from Phase II, from a light intervention vessel – preferably from the newly built Field Support Vessel (FSV) Armada Tuah 100 or, if for any reason this was unavailable, a similarly sized vessel of opportunity. The electrical flying leads were of a size and configuration that made installation possible by a work class ROV on the MODU at opportune moments between well completions support activities. Interface Process Throughout the detailed engineering and procurement phases of the project, considerable cooperation was required between the project team and the subsea equipment manufacturer to develop and close out interfaces in order to ensure the installability of the subsea equipment. The interface process proved to be resource intensive and time consuming with numerous face to face meetings, conference calls, and electronic communication required in order to close out the 177 interfaces raised between the parties. One significant issue was the project team's requirement to have lifting padeyes incorporated within the gooseneck of the flowline VCMs that would permit full load recovery of the flowline from the seabed. The initial VCM/gooseneck design issued by the subsea equipment manufacturer made provision for deck handling padeyes only (10 t SWL). Experience from previous projects provided significant case history that due to 2nd end VCM misalignment issues there was a requirement to have a contingency capability to permit full recovery of the flowline from the seabed back to the surface in order to re-orientate the 2nd end VCM. Additionally future maintenance and repairs, potential field re-configuration and/or field abandonment requirements may also necessitate recovery of the VCMs. Following client review and agreement of this requirement, the VCM gooseneck design was revised to include a padeye enabling full load

Figure 4 – Normand Ivan

OTC 19637 5

recovery (250 t SWL for the 10” water injection flowline VCM). Phase I Activities - Manifold/Suction Anchor/SDU Installation Although at times protracted, the investment in the interface process certainly paid dividends culminating in a successful Phase I installation campaign for the Rockwater 2 with all suction anchors achieving the required depth of penetration and manifold mating to anchor supports being achieved with the as-built configuration of all manifolds within the specified tolerance in terms of location, orientation, and levelness. The Rockwater 2 installation program was completed in 23 days, 2 days less than the planned duration of 25 days. The ROV release and locking mechanism for the fold down mudmats of the Subsea Distribution Units (SDUs), something that had displayed specific clearance problems during the onshore system integration testing (SIT) program and which was subsequently corrected, also functioned effectively for all 5 SDUs during the offshore campaign, demonstrating the importance and value of conducting a thorough SIT program. Phase I Activities - Riser/Umbilical Installation The flowline, riser and umbilical installation campaign undertaken from the Deep Pioneer was a major accomplishment for the project with all flowlines, riser and umbilical sections being successfully installed in what was a difficult working envi-ronment in terms of the marginal weather conditions experienced during the northeast monsoon period. However the installa-tion program did take longer than originally scheduled; 80 days overall compared with a planned duration of 63 days. A number of operational factors contributed to the schedule overrun; the offshore team took longer than anticipated to gain suf-ficient operational proficiency on the new flexible pipelay system and ongoing delays were incurred ensuring appropriate squeeze pressures were applied via the tensioners to the different flowline and riser structures. In addition, an error in assess-ing the tensioner squeeze travel resulted in an incorrect selection of tensioner pads for the umbilical installation program. The consequence of this was that a change in the lay sequence was required including an additional trip back to the HLV in order to change out all tensioner pads on both tensioner systems prior to commencing umbilical installation operations. In addition, a number of days were lost having to re-perform 1st and 2nd end connections of the flexible flowlines to subsea manifolds. Flexible pipe manufacturing processes make flexible flowlines susceptible to torsion which can result in a twist being induced in the flowline as it is deployed towards the seabed and increased amounts of tension, via pipe self weight, is put into the flexible pipes. This torsion can create sufficient misalignment in the VCM assembly to prevent the VCM from mating with the receptacle hub on the seabed component (manifold or subsea tree). Historically, this has been a more com-mon occurrence during 2nd end connections of flexible flowlines. However on Kikeh, VCM alignment issues occurred pre-dominantly on 1st end connections which was a surprise, with insulated production flowlines proving to be particularly susceptible. The project team were aware of potential issues arising from 2nd end connections and had taken measures to account for torsion induced misalignment. As a result, following the requirement to recover and re-orientate the VCM of the first 2nd end connection deployed, all other 2nd end connections were successfully completed without further difficulty. Installation procedures for 1st end attachment of the VCM to the flowline specified that the VCM should be orientated with the VCM connector facing the bow of the installation vessel and the recovery padeye facing aft. However, actual offshore experience confirmed that the orientation of the last few meters of flexible flowline had to be determined as it was pulled off the installation reel in terms of residual bend memory from storage on the reel. The VCM then had to be connected to the flowline so it was aligned with the outside of the flexible pipe’s residual bend curvature and hence, during subsea upending, the flexible pipe memory would assist in upending the VCM gooseneck assembly. This could vary by up to 180º from the original orientation specified in the installation procedure. Once this methodology was adopted, ability to achieve 1st end con-nection alignment of the VCM with the associated mated hub on the first attempt was greatly improved. However, the learn-ing curve associated with correctly orientating the VCM gooseneck assembly in accordance with this methodology added 10 days to the overall installation program. Phase I Activities - Pre-commissioning To complete Phase I, in-field verification testing of the as-installed flowlines and risers had to be undertaken and the agreed acceptance criteria was based upon the requirements of API 17J/17B which requires the lines to be pressurized to 1.1 times the design pressure and held for a 24 hour period. At an early stage in the interface process with the subsea equipment sup-plier, testing the flowlines via the high pressure (HP) caps on the manifold VCM hubs had been agreed and 5 caps were modified into testing caps. The project team then elected to develop, in conjunction with its pre-commissioning subcontractor PSL, a stand alone ROV deployed pumping system capable of both pressurizing the lines to the required test pressure (409.5 bar for the Gas Injection flowline/riser system) with both ends of the flexible pipe on the seabed and chemical dosing of the test medium to alleviate any concerns about extended exposure of corrosion resistant alloy material within the manifold to raw seawater. The advantage of this system was that it would permit all testing to be performed from a small, cost effective

6 OTC 19637

DP vessel without any requirement for a crane or winch as all components were deployed by the ROV in a single dive. The overall pressurization and testing system comprised:

Under-slung ROV skid consisting of chemical dosing reservoir to treat the seawater, metering valve, flow meter, and pressurization pumps.

ROV fitted with two docking tines (similar to forklift tines) for launch and recovery of the data logging frame. The tines had hydraulic fail-safe latches for safe deployment and recovery.

Separate ROV deployed subsea data logging frames. Three frames were used between the four test systems to enable simultaneous testing.

Each data logging frame was fitted with a three channel data logger and local LED readout to display pressure and tempera-ture to the ROV. An ROV valve panel was fitted to the top of each frame to lock in and bleed off test pressure. The frame was landed out on its associated manifold and connected with a 2-inch ROV hot stab to the modified test cap fitted with hot stab receptacle. It was then connected to the ROV skid by means of a 3m hose on a dual port ROV API hot stab. A laptop computer at the surface controlled the skid pump pressure and flow, whilst chemicals in the dosing reservoir were mixed with the seawater via the pre-set metering valve. Once test pressure had been reached and stabilization achieved, the ROV was able to disconnect the 3 m connection hose. This allowed the ROV and support vessel to transit to another subsea flow system and initiate the pressurization of that system while the intial system’s pressure was stabilizing. Simultaneous testing of the various pressure systems was achieved by utilizing this system capability. Initially, problems were encountered offshore with the ROV servo valves being damaged by pressure spikes from the positive displacement hydraulic supply pumps within the pressurization skid. The issue was resolved by running the skid via a sepa-rate IHPU and large volume reservoir in order to isolate the pressurization pumps from the ROV system and fitting hydraulic accumulators to the pressure and return sides of the hydraulic supply. Once these modifications were put in place the system proved to be robust and all subsea flow systems were successfully pressure tested. Phase II Activities - Riser recovery Having made the decision to utilize the infield AHV Normand Ivan to recover and transfer the pre-laid risers and dynamic umbilicals to the FPSO, detailed engineering uncovered a number of issues and complications. Rotation of 6-strand wire when recovering a flexible component with residual torsion was a known issue for which control measures would be re-quired. Adding a swivel to the system would require de-rating of the wire capacity, with guideline values ranging from 25% up to 40%. A de-rating of 40% combined with an applied FOS of 2.0 would have reduced the maximum recovery load capac-ity down to an unacceptable 156 t. The decision was, therefore, taken to procure new wire specifically for the riser recovery workscope and apply a swivel de-rating of 25%. However, availability and lead time for manufacture left the project with the single option of procuring 1600 m of 90 mm diameter 6-strand wire which, after the aforementioned de-rating, had a maxi-mum SWL of 195 t only - insufficient to recover the 10” WI riser close enough to the surface to permit transfer to the FPSO pull-in wire. An alternative means of recovery was required and the decision was taken to procure an additional 600 m low rotation 104 mm diameter wire pennant with MBL of 915 t. The riser recovery methodology was then developed on the basis of utilizing the 90 mm wire to lift the riser termination to a depth 440 m below the sea surface at which point the riser would be transferred to the 104 mm wire for subsequent recovery to a point within 10 m of the sea surface. With the top of the riser just below the stern roller of the vessel, the FPSO pull-in wire was then connected to the transfer chain on the back deck of the Normand Ivan (see Figures 5 and 6). With the basic methodology finalized, detailed engineering progressed to determine the required recovery rigging arrange-ments to be pre-installed on the riser terminations prior to abandonment on the seabed during Phase I operations. Two sepa-rate chain pennants, one 20 m (lifting chain) and the other 10 m (transfer chain), were shackled to the riser termination. Re-covery of the riser would then be completed by lifting the riser to a point a few meters below the sea surface at the stern of the Normand Ivan whereupon the transfer chain would be passed through the port side Karm fork and the lifting chain passed through the starboard side Karm fork. The FPSO pull-in wire would then be connected to the end of the transfer chain. The 104 mm wire would be paid out until transfer of the load to FPSO pull-in wire was completed. Once completely transferred to the FPSO pull-in wire and with the riser termination suspended just below the bell mouth of the FPSO turret I-tube, the slack section of lifting chain would be released from the riser termination and transferred back to the Normand Ivan.

OTC 19637 7

Figure 5 – Normand Ivan Recovering Kikeh Risers Figure 6 – Riser Recovery Methodology

In the field, the recovery, transfer, I-tube pull-in and hang off of all 5 risers and 2 umbilicals was successfully completed without incident. The issues encountered with using rotating 6-strand wire to recover flexible risers from the seabed with the lack of definitive industry de-rating guidelines when introducing a swivel to the system turned what at first appeared to be a relatively straightforward riser recovery methodology selection into a significantly more complex and marginal work scope for the selected vessel with considerable additional components having to be procured to ensure a safe recovery of the riser systems. Project Recovery from Phase II Vessel No Show As described earlier in the paper, the regionally based Rockwater 2 (RW2) was selected by the project to undertake the Phase II rigid spool and flying lead installation post FPSO hook up at the Kikeh site. The large open deck and heave compensated crane with ability to reach the seabed made the vessel ideally suited to perform the tie-in installation activities. A further ac-tivity added to the RW2 Phase II workscope was the disconnection of a 6” WI flowline from the PLET upon which it had been temporarily stored, the crosshauling of the flowline and connection to a nearby subsea tree by means of a 2nd end con-nection of the VCM. All deck layout arrangements and installation procedures were developed, reviewed and approved and arrangements for mobilization of the RW2 in Singapore were in an advanced state of preparation when the project team learned that the vessel had sustained a thruster failure whilst working in New Zealand and would be out of commission for a number of weeks. The uncertainty over the time to complete both the thruster repair and the outstanding workscope, which by then was in the New Zealand winter and subject to the additional risk of increased weather standby, was considered to be an unacceptable risk to the project’s first oil date and the decision was taken to source an alternative vessel to undertake this Phase II workscope which, potentially, could become the critical path for first oil. The Normand Ivan had already been used in an opportune manner to install a number of hydraulic and electrical flying leads in the field, therefore, transfer of the remaining flying leads scope to this vessel was a relatively straightforward decision with the necessary equipment and procedures already in place. Additionally, completion of the 2nd end 6” flowline transfer from the temporary PLET to an adjacent subsea wellhead was considered to be operationally achievable utilizing the Normand Ivan by adding a second winch which could provide the additional lifting point required for attachment to the pre-installed bridle arrangement in order to lift the flowline clear of the seabed 30 m back from the VCM. Installation of the rigid spools, however, was a more complex issue. The 16 m spools, with VCMs at either end, had to be transported to the field and low-ered into the water by means of a crane. The A-frame at the stern of the Normand Ivan did not have sufficient travel to lift and over-board the spools and spreader beam. With inquiries and investigations to source an alternative vessel with adequate crane capacity proving unsuccessful, the option of utilizing the crane on the MODU Ocean Rover was evaluated. Whilst sub-sea construction simultaneous operations (SIMOPS) with the MODU had hoped to be avoided to minimize potential impact to the drilling and completions program, there was no other viable option available and an agreement was reached to utilize cranage on both the MODU and the tender assist drilling (TAD) rig, West Setia which were both working at the Kikeh site. Having 2 drilling rigs in the field concurrently provided increased crane availability to offload the rigid spools without impact to ongoing rig operations. With access to the rig cranes confirmed, the methodology for rigid spool deployment was final-

8 OTC 19637

ized: a second work class ROV system would be mobilized on the portside of the Normand Ivan, the rigid spools would be loaded out onto the deck of the Normand Ivan and transported to the field whereupon the vessel would be positioned under the crane of either the MODU or TAD. The spool would be lifted clear of the vessel and lowered below the sea surface to a point where transfer back to the Normand Ivan winch wire (deployed over the A-frame) could be undertaken. With transfer of the rigid spool to the Normand Ivan’s winch wire complete, the vessel would then transit on DP mode 2-4 km to the tie-in location with the spool in the mid water column and the ROV monitoring the spool configuration and providing restraint against spool rotation as necessary. Deployment of the spool would then be undertaken in accordance with the original meth-odology with the exception that there was no active heave compensation system to assist landing of spool VCMs onto the receiving manifold and subsea tree hubs. Excellent cooperation was received from the operations teams from both the MODU and the TAD which greatly assisted the timely completion of all SIMOP and risk assessment reviews and approval of the revised procedures. Offshore installation of the 5 rigid spools was successfully completed and integrity of VCM connections verified by either back seal test or internal pressure test. Both lifting and transfer of the spools at the drilling rig location and subsequent initial land-out onto receiving manifold/subsea tree hubs were, however, extremely weather sensitive operations requiring sea con-ditions of 1 meter significant wave height or less. One of the 5 spools did not fully land out on the tree hub end of the spool – it appeared that the jumper spool was fabricated slightly short of the required length even though taut wire measurements were taken at the Kikeh site and the measurements were replicated at the onshore fabrication location. However, slight mis-measurements and/or small fabrication tolerances were planned for and one of the contingency tools supplied along with the rigid spools by Aker Kvaerner was an ROV deployable hydraulic jacking tool – capable of either pulling a VCM connector down on to its hub or, if operating in the reverse direction, jacking the VCM off the hub. This simple tool, operated by dual port hot stab, successfully “pulled down” the VCM and rigid spool and landed out the assembly with no further delay. The overall time to complete the 2nd end connection of the 6” flowline plus installation of the HFLs/EFLs (6 of each) and the 5 rigid spools was just over 33 days compared to an estimated 18 days for the Rockwater 2. Although considerably more inefficient, the use of the Normand Ivan to complete this Phase II workscope was an essential change which was effectively managed and contributed to achieving the first oil target date. HFL/EFL Installation

Experience gained with the removal of the HFLs from their storage crates during the SIT program in Port Klang, Malaysia was that, due to their length, weight and stiffness, it required several people, cranage and a lot of space to manipulate them into place. This confirmed to the project team it was not practical to transport the HFLs offshore in the storage crates and lift them out and overboard them on the deck of an installation vessel. Therefore, as there was a total of 48 HFLs to install, in order to ensure a safe and efficient handling and overboarding process throughout the offshore phase, the decision was taken to mobilize an 8.6 m diameter installation reel (ex Kikeh umbilical storage reel). HFLs were to be transpooled onto the reel at Murphy’s supply base in Labuan and then the reel and drive system would be loaded onto the Normand Ivan AHV (see Figure 7). Again, due to the weight and stiffness of the HFLs, it was not possible (despite the terminology) for the Work Class ROVs to “fly” the HFLs into place. Consequently, procedures followed similar processes used for installation of flexible jumpers. The 1st end cobra head was connected to a clump weight and the HFL was overboarded with as-sistance of the Normand Ivan’s A-frame and run vertically to the seabed on the vessel’s winch wire. Once the clump weight was landed adjacent to the 1st end stabplate, the ROV, tooled up with a flying lead orientation tool (FLOT) and Class 1-4 torque tool, docked into the 1st end cobra head and made up the connection. The vessel then pro-ceeded in DP mode to lay the HFL along the proposed route. The 2nd end cobra head, which was secured into a purpose made support frame prior to overboarding, was placed adjacent to the 2nd end stabplate. The frame ensured that any tendency for the tubes in the HFL to make the 2nd end cobra head twist was countered, thereby ensuring easy docking of the ROV tooling. Figure 7 – HFL Installation Reel on Normand Ivan

OTC 19637 9

Phase III Activities Due to the late Phase II vessel change which required the Normand Ivan to perform all of the rigid spool and flying lead work scope in Phase II, all Phase II tie-in procedures were suitable for immediate use in Phase III operations where both Normand Ivan (back to a single ROV configuration) and the FSV were planned to be utilized. The VCM guide rail system, designed specifically for capturing the 1st end of the rigid spool at the manifold prior to land out to allow a single ROV to relocate to the 2nd end to confirm alignment prior to full land out, has now been tried and tested with 3 jumpers being successfully in-stalled using only a single work class ROV to monitor the operation. Overboarding of hydraulic flying leads continues to be the most difficult aspect of the installation due the sheer size of the umbilical reel and under roller equipment, and the associated labor intensive load outs. Consequently, since having the use of a vessel with an A-frame cannot be guaranteed, another installation method is soon to be adopted using a more compact car-ousel unit with an overboarding chute. Electrical flying leads are now routinely installed by the MODU ROV, where they are deployed in a figure of eight configuration around two posts added to the ROV deployment cage. Take Aways / Conclusions Deep water projects, which now have a significant track record over a number of years and are becoming more globally widespread, remain extremely challenging and can be unforgiving with very limited margin for error. Attention to detail is the key, with what may appear to be small omissions potentially having significant impacts if not identified and corrected prior to commencing offshore operations. Consequently, thorough SITs are a critical part of any successful deepwater development program. Large complex deep water development projects are often tendered, negotiated and awarded in a relatively short time frame. It is critical that at project commencement a thorough evaluation of selected vessels and major installation equipment items, such as flowline and riser lay systems, is conducted to determine if they are adequate when all contributing factors such as dynamic loading, contingency requirements and spare equipment are taken into account. Large construction vessels primarily used to install pipelines, risers and mooring systems are generally extensively booked up a number of years in advance and have limited windows of availability. The requirement for such a vessel, or major item of equipment such as a large capacity heave compensated winch, if not identified and confirmed at the tender stage, needs to be confirmed and ordered very early in the project life cycle as restricted availability or excessive lead times may well end up impacting the project schedule. The value of a large capacity Anchor Handling/Supply Support Vessel, ideally with a crane or A-frame and a work class ROV, on a major subsea development in a remote greenfield location can not be underestimated. The Normand Ivan, which was chartered to essentially install pre-set moorings for the MODU and provide general rig moves and supply runs, ended up becoming a key construction vessel. This would not have been possible, however, without the Kikeh Project focus on inter-face management and being prepared to take “best for project” decisions to optimize available assets across functional boundaries, in order to preserve the project schedule and first oil target. Installation of pile guidance mud mats and stabbing of driven piles for the Spar mooring system, installation of flying leads and rigid spool piece jumpers, assistance during hook up of the GAP system between the Spar and FPSO, and recovery and transfer to the FPSO of flexible risers and dynamic um-bilicals summarizes the range of construction activities performed by the Normand Ivan throughout the Kikeh field installa-tion program. Non-appearance of key installation assets is a genuine risk for a remote greenfield development in what is an extensively over booked and under supplied market. Maintaining a vessel in field that can support both drilling and construc-tion activities throughout the offshore installation and start-up program can be a significant risk mitigator for any remote de-velopment such as Kikeh. VCMs are key components in any deep water development and, when used with flexible flowlines present installation chal-lenges, in particular the effect of flowline torsion on final alignment of end connectors at subsea structures. Whilst this is a known phenomenon within the industry it has again been encountered on the Kikeh project and highlights the fact that it is critical when combining these technologies to look at all factors governing the orientation of the end of the flowline/VCM interface and to design out, where possible, costly offshore learning curves and develop installation aides to assist with orien-tation of the connector attachment. Furthermore, since installation contingency planning will almost always have to consider complete flowline recovery to the surface, it is essential to design the end connections, whether VCMs or another design, to take the full recovery loads. Finally, continuous communication and cooperation between company and contractor is essential to execute a deepwater greenfield development. It is a given that unanticipated and unforeseen issues and problems will arise on a project develop-ment the size and complexity of Kikeh therefore communication, contingency planning and the teamwork exhibited by company and contractor will form key ingredients to the success of the development project for both parties.