OTC 15271 The BP Bombax Pipeline Project Design for Construction

Brett Champagne and Derek Smith J P Kenny Inc., Jerry Grass - BP

Copyright 2003, Offshore Technology Conference This paper was prepared for presentation at the 2003 Offshore Technology Conference held in Houston, Texas, U.S.A., 58 May 2003. 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, as presented, have not been reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Offshore Technology Conference or its officers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes 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 where and by whom the paper was presented.

Abstract The BP Bombax Pipeline Project involved designing and installing one of the largest subsea pipelines of this magnitude in the world to support gas expansion in Trinidad and Tobago. The Bombax Project pipelines and manifold are in an area of relatively shallow water, ranging from around 250 feet at the platforms, gently sloping to the shoreline. Such water depths do not generally present significant problems for pipeline design and installation. With a 48-inch pipeline, however, the effects of wave and current are significantly exaggerated and bearing in mind that with such a large diameter, small variations in material thicknesses have a measurable impact on costs, from steel for the line pipe, to concrete for the weight coating. The paper focuses on the design for construction challenges involved with the new 48-inch trunkline and how these challenges were addressed. Specific design topics that will be addressed in this paper include: material grade selection, wall thickness design, and stability design. The paper will also focus on the decision by the project to involve the installation contractors during the design phase. Introduction In response to the increasing demand for energy in the form of natural gas, BP Trinidad and Tobago has started expanding its offshore fields and gas transportation system to supply new LNG trains at Point Fortin on the West side of Trinidad as well as the increased local domestic market. There are two projects currently underway to expand production and transportation of gas from 1.5 bscfd to 3.0 bscfd, the Kapok Project and the Bombax Project. The Kapok Project comprises a new 2.6 BCFPD production platform, Cassia B that is bridge connected to the existing Cassia A platform and a new drilling platform Kapok. The Bombax Pipeline Project includes 63 km of 48-inch offshore pipeline from Cassia B to landfall at Rustville, on the East Coast of Trinidad. From the landfall, the pipeline extends 1.8 km

onshore to the existing Beachfield slug catcher and production facility for onward transportation of gas to the various industries on the island including the LNG facilities on the West Coast. The offshore end of the 48-inch pipeline is connected to the existing 40-inch pipeline via a 20-inch subsea jumper. This jumper facilitates looping of approximately 2/3s of the existing 40-inch pipeline thereby expanding the transportation system capacity. Additional gas production to meet the growing industrial demand will be supported by a new wellhead platform located at the Kapok field along with continued development of existing fields. The Kapok platform is linked to the Cassia B production hub by a 26-inch multi-phase pipeline installed as part of the Bombax Pipeline Project. Due to the development schedule, the Kapok platform will be ready for production before the Cassia B hub topsides are available. To allow Kapok to produce early gas, it is intended to carry out separation on Kapok with the test separator and transport liquids via a new 6-inch line to the existing 12-inch liquids line. The 12-inch liquid line transports liquids to shore from BPs existing platforms originating from the Mahogany platform. Tie-in to the existing 12-inch liquid line is via a preexisting subsea hot-tap tee. The Kapok separated gas is then transported via the 26-inch line to the subsea manifold and into the 48-inch pipeline for transportation to the Beachfield facility via an early gas jumper on the manifold. Figure 1 provides a layout of the new Bombax development and the existing pipelines. To satisfy the requirements for the Cassia B platform safety, a check valve for the 48-inch export pipeline, and an SSIV for the incoming 26-inch line from Kapok are required. Additional the project required the installation of an actuated valve in the 20-inch line looping the 40 and 48-inch pipeline to enable isolation of the large gas inventories in these lines if needed. In light of the complexity required to meet the project requirements, it was decided to accommodate the valving and piping within a single manifold structure. Also in line with the projects objective to maximize the use of local content, the manifold was constructed in Trinidad. Add to this individual 48-inch, 26-inch and 20-inch tie-in spools of up to 300 feet long and 270 tons, collectively Bombax presented an interesting and challenging project.

OTC 15271

Figure 1 Bombax Field Layout Material Grade Selection The composition of the gas to be transported via the 48inch Bombax Pipeline does not necessitate the design be suitable for sour service which simplifies the material design. The grades of line pipe considered ranged from API-5L X52 to X100. Manufacturing Considerations. Controlled rolled steels manufactured to API 5L X52, X65 and X70 requirements have been used extensively for offshore oil and gas pipeline construction. The trend towards the use of large diameter high-pressure pipelines has prompted manufacturers to invest in the metallurgical development of high strength line pipe. Submerged Arc Welded API 5L X65 grade line pipe has been deployed in numerous sour and non-sour projects. API 5L X65 is commonly manufactured by controlled rolling and, if necessary, accelerated cooling. The toughness and weldability of X65 is sufficient to meet the requirements of the majority of pipeline codes and construction standards, including ASME B31.8. API 5L X70 grade line pipe has been produced almost exclusively for non-sour service. The extra strength level is attained by using accelerated cooling, modifications to chemistry and advanced controlled rolling technology which all contribute to optimizing microstructure. API 5L X80 line pipe has been used to a very limited extent in oil and gas transmission. Certain mills, particularly in Japan and Germany have the capability of producing this grade. A number of problems concerning chemistry, processing and weldability were overcome in order to make X80 commercially viable. Generally, forming higher strength line pipe is made more difficult due to equipment limitations and strain hardening effects. In addition, the mechanism for strengthening the higher strength grades must involve additional processing and heat treatment. The more conventional strengthening methods of increasing the level of alloying elements reduce the weldability of these steels and are generally avoided. The use of X100 Grade pipe is being considered for onshore pipelines, but to date no major project has endorsed the use of X100 pipe. The concerns with X100 are with the weldability and fracture toughness. In addition there are a

limited number of steel mills capable of producing X100 grade pipe. The selection of the grade of line pipe depended on a number of factors. The main material requirements for the 48inch Bombax Pipeline are that the line pipe is weldable using conventional offshore pipeline welding techniques and that it has adequate fracture resistance to avoid ductile/brittle failure based on the minimum pipeline design temperature. In addition, due to the large wall thickness required for internal/external pressure containment the mills must be able to form large section thicknesses for a given grade and still meet project requirements for mechanical strength. This requirement is often difficult as the average yield and tensile strength of a particular grade decreases as the wall thickness is increased. The technology exists for a number of mills to produce API 5L line pipe up to X80 strength level with good weldability and fracture toughness properties. However, there are limitations for all pipe mills on maximum wall thickness achievable. Generally, as the line pipe grade is increased, the maximum obtainable wall thickness decreases for the same line pipe diameter. Welding Consideration. The weldability of particular line pipe steel depends mainly on its chemistry, grade and heat treatment condition. The welding methods that were considered as potential processes for the Bombax Pipeline Project included: Shielded Metal Arc Welding (SMAW), Gas Metal Arc Welding (GMAW), Gas Tungsten Arc Welding (GTAW), FluxCored Arc Welding (FCAW) and Submerged Arc Welding (SAW). The prime concern with the weldability of line pipe are the properties of the Heat Affected Zone (HAZ) caused during production of the SAW seam weld and pipe-to-pipe girth weld, and to a lesser extent the influence of alloying elements on the weld metal properties due to dilution. The main HAZ properties that require careful control during welding are the hardness and fracture toughness. The fracture toughness requirement for the pipeline was calculated using the BGC calculation method described in ASME B31.8 Section 841 (Ref. 1). From the calculation based on a temperature of 30 C the minimum charpy value was 26 ftlbf with an average value of 33 ft-lbf. These charpy values were difficult to achieve during the manufacture of the heavy wall riser pipe with a wall thickness of 1.29 inches. The fracture toughness requirements for the various steel grades considered, and for the range of wall thicknesses required are shown in Figure 2.

OTC 15271

Fracture Toughness Requirements

70.0

Charpy Impact Energy Required (ft-lbf)

60.0

50.0

40.0

30.0

20.0

10.0

0.0 X52 X65 X70 X75 X80 X100

API 5L Steel Grade Lowest Average Value Highest Average Value

Figure 2 Fracture Toughness Requirements The weldability of API 5L line pipe up to grade X65 has been demonstrated during the construction of several large diameter pipeline projects. A smaller number of pipeline projects worldwide have used X70 grade. Several of the mills have performed extensive weldability trials and demonstrated that for X70 and X80 line pipe it is possible to achieve good hardness and fracture toughness properties in the seam and girth welds. Generally, it is recommended that some degree of preheat and control of hydrogen is used during welding of higher strength line pipe, particularly with X70 and X80. The large wall thickness of the line pipe proposed for the Bombax Pipeline creates additional problems related to weldability. As the wall thickness of the line pipe increases, the cooling rate of the weld increases leading to possible problems with hardness, fracture toughness and cold cracking Laboratory data and limited field experience also indicates that line pipe up to API 5L X80 grade should be weldable with wall thicknesses approaching those required for the project. A major benefit of increasing the grade from X65 is that the wall thickness required for internal and external pressure containment will be reduced compared to that required for X65. Therefore, the amount of time to produce a weld will be proportionately reduced. In meetings with potential installation contractors they stated that welding of materials higher than X65 offshore will increase welding requirements. Additional provisions above those normally required, such as mandatory preheat which may affect welding times, may have to be written into the welding specification if the higher strength grades such as X70 or X80 are considered. The installation contractors expressed serious concerns regarding the weldability of X100, and X100 was not considered any further. Effect of Steel Grade on Collapse Resistance. While the higher grade steels result in a thinner required wall thickness, for offshore use this effect is mitigated by the converse effect thin wall has on collapse response and bending strain. While all of the steel grades examined result in D/t ratios outside of those covered by the applicable design codes (D/t greater than 45 for DnV 2000 (Ref. 2), D/t greater than 50 for API 1111 (Ref. 3)), each code has methodologies for applying higher D/t

ratios. The greater the ratio, the more onerous the requirement becomes. For the purposes of the Bombax project, steel grades greater than or equal to X75 result in D/t ratios that would require detailed, dynamic installation analysis using finite element techniques. Further the results would have to be subjected to a significant allowance for a safety factor to account for unknowns such as the actual installation vessel response amplitude, the likely wave state encountered, the velocity and directions of the local currents. Even if correctly calculated the resulting pipeline would not be as robust for installation purposes, and would impart a higher level of risk, as compared to the lower steel grades. Using the methodology of DnV 2000 (Ref. 2) the combined axial tension, bending, and external pressure effects were calculated for the installation conditions of each steel grade for a water depth of 190 ft. This depth was chosen as it represented the maximum water depth common to the thinnest section of each wall thickness schedule. The maximum allowable bending moment for a given axial tension was then calculated for each steel grade and the results plotted on Figure 3.
8.50

X52 8.00 X65 X70

Bending Moment (MMlbf-ft)

7.50

7.00

6.50

X100

6.00

X75

X80

5.50 0 50 100 150 200 250 300 350 400 450 500

Axial Tension (tons)

Figure 3 Maximum Allowable Bending The highest strength steel does not have the greatest collapse resistance, as this is a product not only of steel grade but also of D/t ratios. The DnV methodology is applicable only for D/t ratios less than 45. Based on previous work with DnV (Ref. 4) at D/t values greater than 45 the plastic moment resistance must be modified as follows: D/t 45, Ymp =1.0; 45<D/t 50, Ymp = 0.98; 50<D/t 55, Ymp = 0.96; D/t > 55, Ymp =0.70 (assumed). Where: D = Outside diameter (in) t = Wall thickness (in) Ymp = Plastic Moment Resistance Modifier OFFPIPE software was used to calculate the bending and tension loads for X65 and X70 line pipe in 190 ft of water. A range of applied tensions from 100 kips to 500 kips was used. The results, when combined on Figure 4 with the results from Figure 3 show that while the X70 pipe has a larger allowable bending moment than the X65 for the same tension, the actual stresses in the X70 pipe are higher. This results in a less

OTC 15271

robust design condition. Using the length of sagbend to the touchdown point an estimate of the allowable barge movement was made for each case. For the X65 pipe the allowable movement, assuming an installation tension of 300 kips was 66 ft, for the X70 case the corresponding number was 58 ft. These numbers should not be taken as accurate calculations of the allowable barge movement, as they ignore dynamic effects and localized stresses at the stinger tip, however they can be taken as representative of the relative movement of the two steel grades. The X65 pipe will tolerate more barge movement than the X70 pipe.

10.000

100 Kips

corrosion rate of 3.937 mils/yr was applied, resulting in a corrosion allowance over the 50-year design life of 0.197-inch. Based on water depths along the route, minimum required wall thicknesses were calculated based on ASME B31.8, API RP 1111, and adjusting for the corrosion allowance. The wall thickness schedule for the 48-inch pipeline was optimized using an iterative spreadsheet and limiting the number of step changes in wall thickness to facilitate construction. The optimized wall thickness is shown on Figure 5. The calculated wall thickness to satisfy code requirements with the corrosion allowance only just slightly fell below the buckle propagation thickness by .003-inches for 3.5 km as shown on Figure 5. Considering the conservative nature of the calculations the risk of buckle propagation was considered to be low, so there were no buckle arrestors used.
Riser Thickness is 1.290 inch Class 3, 301 m long, Thickness is 1.290 inch

9.000 Bending Moment (MMlb-ft)

58 ft of Barge Movement X70

8.000 200 kips

1.140

7.000 66 ft of Barge Movement 6.000 X65

300 Kips

1.120 1.100

1.108 inch

400 Kips 5.000 Barge tension at data point => 300 Kips 4.000 0 50000 100000 150000 200000 250000 300000 350000 400000 Axial Tension (lbf) X70 Allowable X70 Installation X65 Allowable X65 Installation 500 Kips

1.080

Wall Thcikness (inch)

1.060 1.040 1.020 1.000 0.980 0.960 0.940 0.920 0.900 -0.50 0.956 inch 0.936 inch 0.918 inch 3.5 km long (see note) 0.936 inch Offshore

Figure 4 Pipelay Bending vs. Max. Allowable Bending Conclusion Material Grade. Steel grades ranging from X52 to X100 were examined for their impact on project cost, manufacturing, weldability and collapse resistance. X52 steel grade would increase project cost by nearly 2% and yield no significant gain in performance. Steel grades X75 to X100, while offering small potential cost savings in the overall project cost, result in the following problems with regards to performance: Potential welding problems (X80 and X100); Significantly increased risk of collapse or buckle during installation due to higher D/t ratio; Potential to fall outside of the material testing capability of the steel mill for the heavy wall pipe; X70 pipe offered a potential cost savings of $0.3MM. While X70 pipe performs in a similar manner to X65 with regards to manufacturing, X70 will be marginally more susceptible to collapse during installation, and may require additional welding requirements, such as joint preheating, on the laybarge. The potential cost savings was not considered significant enough to overcome the increased risk. API 5L X65 grade was therefore chosen for the Bombax pipeline. Wall Thickness Design The wall thickness determination was performed in several steps. First the required corrosion allowance was determined in conjunction with the BP Upstream Technology Group (UTG). The corrosion allowance was agreed upon with BP UTG based on the performance to date of the existing 40-inch pipeline, and the product composition, temperature, and pressure profiles of the Bombax pipeline system. An inhibited

0.956 inch

9.50

19.50

29.50

39.50

49.50

59.50

Note: Wall thickness is less than 0.003 inch thinner than required to resist buckle propagation . Pipeline is within ASME B31.8 code requirements.

KP

Onshore Pipeline Minimum Required WT to Resist Buckle Propagation Wall Thickness Schedule

Figure 5 KP vs. Wall Thickness Summary The selected wall thicknesses for the 48-inch pipeline are shown in Table 1. Location Offshore(1) 0.000 2.550 30.100 52.100 57.550 Shore Crossing(2) 62.381 KP 2.550 30.100 52.100 57.550 62.381 62.581 W.T. (inch) 0.956 0.936 0.918 0.936 0.956 1.108 Steel Grade API 5L X65 API 5L X65 API 5L X65 API 5L X65 API 5L X65 API 5L X65

*Notes: (1) KP 0 is located at tie-in between the 48-inch pipeline and the spool at the Subsea Manifold near Cassia B Platform. (2) KP 62.581 is the onshore/offshore tie-in point of the 48inch pipeline.

Table 1 48-Inch Offshore Wall Thickness Schedule For the thinnest wall, this results in D/t ratios greater than 50 (52.3 maximum), which invalidates the combined bending and collapse equations used by API 1111. DnV 2000 Rules

OTC 15271

for Submarine Pipelines collapse equations cannot be used as an alternative as they are limited to D/t values less than 45. DnV 2000 Rules for Submarine Pipelines recommends that for higher D/t ratios finite element (FE) analysis may be used to confirm the collapse design. ASME B31.8, which is the design code for the 48-inch pipeline, requires that the design be examined for collapse and does not specify a method. The increase in wall thickness to meet a D/t ratio of 50, however, would result in an increase in costs of around $2.5 million. To resolve this issue an analytical search was performed for calculating pipeline collapse failure modes and an FE analysis was performed using ABAQUS/Standard Version 5.8 to determine the combination of external pressure and bending loads that would result in collapse of the 48-inch pipeline. Three publications of note have been validated against test results for D/t ratios up to 45, but are not strictly applicable for the 48-inch pipeline (D/t=52.3). However, these three publications were used as a reference to compare against the results of the FE analysis. The AME Joint Industry Project (JIP) (Ref. 5) contains the work performed as part of a Joint Industry Project (JIP) on strain based design of pipelines and is confidential to the JIP sponsors. Extensive Finite Element analyses were used together with large-scale test results to develop expressions for limit state analyses. DNV 2000 OS-F101 (Ref. 2) is widely used for the limit state design of offshore pipelines. It contains load-controlled local buckling limit state expressions which may be used to develop collapse failure envelopes for pipelines under the combined effects of pressure, axial force and bending moment. Hauch and Bai (Ref. 6) report the results of work performed as part of the DEEPIPE JIP on the collapse of pipelines, in particular the effect of outof-roundness and corrosion. The collapse envelopes for the 48-inch pipeline 0.918-inch wall thickness, using the expressions from the above references, are shown in Figure 6. Note that the results shown do not include the effects of axial force, corrosion allowance and pipeline ovality. A safety factor is not applied to these results.
20 AME (JIP) Bai et al DNV 2000 installation

The Finite Element (FE) program ABAQUS/Standard, Version 5.8 was used to develop a model of the pipe section for collapse analyses. The pipe section was subjected to external pressure, axial force and bending leading to collapse, which results in a failure due to a combination of ovalization, yielding and local buckling. The non-linearity of the collapse phenomenon can be correctly modeled with the FE analysis. In the FE model, element type S4R was used as it has the capability to model large displacements and rotations as well as membrane finite strain formulation. This element type is a four-node shell element where each node has six degrees of freedom. Nine integration points were used through the thickness of the shell element. The symmetry of the problem with respect to plane of bending was used to optimize the FE model. Half the pipe circumference was modeled and symmetric boundary conditions applied along the longitudinal edges of the model. A short section of the pipeline (2 m) was modeled and pipeline continuity was modeled using the multi-point constraint (MPC) subroutine in ABAQUS. The fine mesh density, 20 elements lengthwise and 18 elements along the half circumference, was used. The FE mesh is shown in Figure 7 and the collapsed model is shown in Figure 8.

Figure 7 Finite Element Mesh used in Analyses

Allowable Bending Moment (MNm)

15

10

Figure 8 Finite Element Model after Collapse

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

External Pressure (Nm m -2)

Figure 6
Failure Envelopes for 48-inch Pipeline (WT = 0.918 inch

The non-linear material properties of the pipe steel were used in the analyses. The stress-strain data were developed using the Ramberg-Osgood relationship and the steel properties, as follows:

OTC 15271

= + r o E E o
where:

Wall Thickn 0.918

External Pressure 0.64

Bending Moment Capacity (MNm) FE DNV Bai-et-al AME

= total strain = stress E =Young's Modulus r = Ramberg-Osgood coefficient o= nominal yield stress N = Ramberg-Osgood exponent

14.25 14.58 14.22 14.25 Table 2 Summary of Bending Moment Capacity Calculations

E E log u 1 log o 1 u o N= log u o E o

r =
where:

The FE results show good comparison with the analytical solutions although these were developed and validated for pipes with D/t ratios only up to 45. A further check was performed against the force-moment combinations obtained from the OFFPIPE pipelay analysis against the results of the FE analysis as further validation. The results of the FE analysis are superimposed with the results of the pipelay analysis on Figure 9. The force-moment results from the pipelay analysis fall below the collapse envelope. Based on the results of the FE analysis and the comparison against the pipelay analysis the recommended wall thicknesses were considered sufficient to resist collapse during the anticipated installation conditions and over the pipeline operational life.
20 FE - no ovality FE - 0.5 % ovality FE - 1.0 % 0vality installation

= minimum specified yield strength o = 448 Nmm-2 = strain at o o = 0.5 % = minimum specified ultimate tensile strength u = 530 Nmm-2 = strain at ultimate strength u = 21 % The FE analyses were performed by applying the loads in the following sequence: external pressure axial force bending In the last load step, bending was applied until collapse occurred. The RIKS method was used for the solution for the analysis to continue beyond the point of maximum moment. The analyses were initially run for a perfect pipe, i.e. no ovality and subsequent runs performed for 0.5 % ovality and 1.0 % ovality. Pipe ovality was defined as:

Bending Moment (MNm)

15

10

0 -25 -20 -15 -10 -5 0 5 10 15 20 25

Axial Force (MN)

Figure 9 Collapse Failure Assessment Diagram (WT=0.918 inch) Stability Design Stability design of the 48-inch pipeline was a challenge due to the combined wave and current velocities ranging along the pipeline from 3.5 ft/s to 9.2 ft/s coupled with the industry limitations of concrete weight coating of 6-inches. During the Define Stage of the project a limited amount of geotechnical and metocean data was available to assess the stability of the pipeline. Fugro was contracted to conduct a geotechnical investigation along the proposed 48-inch pipeline route and collect current data for 3 months. The geotechnical investigation included a combination of drop cores offshore and borings nearshore. The current data was collected from 3 ADCP current meters and from this data hindcasts were made for 1, 10 and 100 year. From this work the project had access to more accurate geotechnical and near bottom currents to better refine the trenching and weight coating requirements. The data collected from Fugro was sent to a third party to be run in a complex refraction and shoaling model. The data was used as input to a program model to compute the wave

f =

(Dmax Dmin ) (Dmax + Dmin )

where: f = ovality Dmax = maximum diameter Dmin = minimum diameter The FE results are compared with the analytical solutions for the case of no axial force and no ovality in Table 2.

OTC 15271

height, current, and directionality for pre-determined points of interest along the proposed pipeline route. Upon receipt of the new metocean data and geotechnical investigation, the stability analysis for the Execute Phase was conducted. The tool used to analyze stability of the 48-inch Bombax pipeline was Level 2 of the AGA Stability program. Level 2 of the AGA Stability program makes use of the linear wave theory to assess the hydrodynamic parameters of each calculation. The remaining wave theories were analyzed using an alternative method to AGA. This involved calculating the hydrodynamic forces acting on the pipeline utilizing the applicable wave theory. Once the hydrodynamic forces were solved, calculations were performed to assess whether the soil can restrain the pipeline both vertically and laterally. Stability of the pipeline was achieved by a combination of submerged weight, application of concrete weight coating, trenching of sections of the pipeline, backfilling of the trenched section, and naturally occurring embedment. The stability analysis was performed with Level 2 of the AGA Stability program for three conditions to derive stabilization requirements for the offshore pipelines: Stability under installation conditions: Stabilization requirements are assessed for a oneyear return period wave combined with a oneyear return period steady state current. During installation the pipeline is assumed empty. Stability under hydrotest conditions: Stabilization requirements are assessed for a one-year return period wave combined with a one-year return period steady state current. During hydrotest the pipeline is assumed seawater filled. The installation stability requirements are more onerous than the hydrotest stability requirements; hence no analysis was performed for this case. Stability under operating conditions: Stabilization requirements are assessed for a hundred-year return period wave with a ten-year return period steady current and ten-year return period wave with a hundred-year return period steady current. The more onerous requirements are used to define the operational stabilization requirements, which in this case was the hundredyear return period wave with a ten-year return period steady current. The concrete weight coating and trenching requirements from the stability design are provided in Table 3. In order to meet the design requirements Allseas elected to pre-trench from KP 59.9 to KP 62.5 using a cutter suction dredger and post trench from KP 52.10 to KP 59.9 with the Digging Donald a mechanical cutter/jet sled.

KP 0.00 2.55 2.55 30.10 30.10 52.10 52.10 57.55 57.55 58.00 58.00 59.55 59.55 60.41 60.41 62.55

48-inch Concrete Coating & Stabilization Method 4.50inches; On-Bottom 3.75inches; On-Bottom 4.00inches; On-Bottom

5.00inches; Post Trenched, minimum 1/1 OD 4.50inches; Post Trenched, minimum 1/1 OD 4.50inches; Post Trenched, minimum 1/1 OD 5.00inches; Post Trenched, minimum 1/1 OD

5.00inches; Pre Trenched and Backfilled Table 3 Stabilization Recommendations for 48-inch Pipeline

Liaison with Contractors The resulting pipeline configuration of high D/t ratio and concrete coating thicknesses described earlier presented a high potential for buckling during lay. As a consequence, the project recognized early on that without input from the installation contractors, designing a 48-inch pipeline based on design theory still required confirmation that it could be installed safely. Pre-bid meetings were held and provided the dual function of establishing lay ability and to pre-qualify suitable contractors. The initial meetings were held to present the contractors with the project schedule, proposed route, its bathymetry, soils conditions and expected sea states. They were tasked with using this data to develop preliminary lay configurations for their proposed vessels and to show the project that their equipment was capable of the lay, with any modifications required identified and any impact on schedule clearly noted. All contractors gave full support to these endeavors and responded enthusiastically, recognizing that they are not normally given this opportunity. The two main technical areas raised by the project team were wall thickness, and reinforcement design in the concrete weight coating to withstand the potential for cracking and spalling off. From these early discussions, it was clear that the pipeline was capable of being installed. The focus of a second set of meetings was to provide the contractors with an update on design issues, discuss the current schedule, and resolve bid clarifications. Conclusions The 48-inch Bombax Pipeline was successfully installed in April of 2002 without any major incidents by Allseas vessel the Solitaire. The 48-inch pipeline included a range of wall thicknesses with a maximum D/t ratio of 53 and up to 5-inches of concrete weight coating which was added to provide onbottom stability. The X65 linepipe was produced using the DSAW technique by Sumitomo and coated by Bredero Price. Acknowledgments J P Kenny Inc. would like to thank BP for the opportunity to work on the Bombax Pipeline Project and publish this article.

OTC 15271

References: 1. ASME B31.8 Gas Transmission and Distribution Piping Systems, 1999. 2. Det Norske Veritas, Submarine Pipeline Systems, Offshore Standard OS-F101, 2000. 3. API RP 1111, Design, Construction, Operation, and Maintenance of Offshore Hydrocarbon Pipelines (Limit State Design), of 3rd Ed., 1999. 4. JP Kenny Ltd., GEP Project Design Manual, GEP Wall Thickness And Buckle Arrestor Confirmation Report, MET Document No. A32703-JPK-ING5112, Rev. D03, 12/12/97 5. AME, Establishment of Strain-Based Criteria and Analysis for the Assessment of Subsea Pipelines, Final Report, Joint Industry Project, January 1997. 6. Hauch S and Bai Y, Bending Moment Capacity of Pipes, Offshore Mechanical and Arctic Engineering, July 1999.