TECHNOLOGY Sequential piggyback, dual lay used for Irish Sea pipelines

A. Dutta, M. Guinard
McDermott-ETPM West Inc.
Nanterre, France

Based on a presentation to the Offshore Technology Conference, May 1-4, Houston.

Piggyback and dual pipe lay were used sequentially for the first time in 1993 aboard the lay vessel DLB 1601 in the North Morecambe project, Block 110/2a of the Irish Sea (Fig. 1)(111973 bytes). Development was by British Gas Exploration & Production Ltd.

A 3-in. OD pipeline was laid piggybacked onto a 36-in. pipeline for the shore pull operation and then separated in an unconventional transition operation for dual lay for the offshore section.

Detailed engineering studies along with well developed installation procedures resulted in successful pipelaying.

Increasing complexities of laying, trenching, and burying offshore pipelines suggest the techniques used in the North Morecambe project will find further applications.

In fact, McDermott International Inc.-ETPM Services (U.K.) Ltd. (MET), which executed the North Morecambe project, developed the techniques further in 1994 for the Liverpool Bay development of Hamilton Oil Co. In that project, DLB 1601 laid in dual lay two 3-in. pipelines in a bundle with a 20-in. pipeline.

This first of two articles on the North Morecambe project discusses its background and engineering, primarily for the piggyback and dual lay phases. The conclusion covers the innovative transition operation and reports shore-testing results and observations.

Lay techniques

The offshore oil and gas industry has constructed thousands of kilometers of submarine pipeline all over the world. The conventional S-lay method, the single-pipe laying technique, and the piggyback (or bundle) pipelaying technique are both well known.

In the piggyback pipelaying technique, smaller pipelines are generally attached to a comparatively larger pipeline which passes through the tensioners on the laying vessel. Two equally sized pipelines have also been laid as a bundle with both lines passing through the tensioners.

In this technique, different pipelines are fabricated in parallel on the lay vessel. With saddles and straps, these pipelines are attached to form a bundle near the vessels stern. This allows the pipelines to leave the lay vessel as a bundle and be laid together on the seabed.

In such a case, therefore, touchdown-point monitoring of the bundled pipeline holds the same importance as for single-pipe laying.

Simultaneous laying of two (dual lay) or more (multiple lay) unattached pipelines from the same pipelay vessel is rare. This pipelay technique is used only in special cases where the pipelines are to be constructed very close to each other, keeping a specified separation.

Each pipeline must leave the lay vessel separately. The technique necessitates individual laying ramps with tensioning arrangements for each pipeline. Lay-vessel preparation and monitoring of the overall operation associated with such a laying technique therefore need special attention.

Over the last 2 decades, the derrick-lay barge DLB 1601 has performed several pipelaying projects, among them the British Gas North Morecambe development in 1993 for which the DLB 1601 performed a dual-lay operation for the first time.

Irish Sea project

In September 1992, MET received the offshore pipelines construction contract, referred to here as North Morecambe project. The contracts scope of work included the following:

• Installation of a 36 in. (899 mm) OD gas-export pipeline and a 3 in. (89 mm) OD methanol pipeline between tie-in locations (kilometer point-KP 0) at the base of the North Morecambe platform and tie-in points on Walney Island (U.K.) at approximately KP 32.
• Complete trenching for both pipelines except a certain distance near the platform (for installation of spool pieces over anti-scour mattresses).
• Other related works such as surveys, predredging, and backfilling of shore-approach trench, tie-in with risers, and pressure testing.

The pipe W.T. for the 36-in. pipe was 17.5 mm constant throughout the 32-km section. The 3-in. pipe was designed for 7.6 mm W.T. with 12.7 mm W.T. from KP 23.6 to 27.0 for post-lay on-bottom stability.

The 36-in. and 3-in. pipes were made from carbon steel as per API-5L specification,1 Grade X-65 and Grade B, respectively. The anticorrosion coating for both pipelines was 0.65-0.80 mm fusion-bonded epoxy (FBE). Only 36-in. joints were weight coated with high-density concrete (3,050 kg/cu m) with thicknesses of 90 mm, 100 mm, and 120 mm.

The lengths for both pipelines were about 31.47 km, excluding the spool piece and the onshore 670 m to the tie-in point. A further 5 km of estuary crossing and land pipeline to a new terminal was laid under a separate contract.

Fig. 1 (111973 bytes) shows the general route layout (offshore section) of the 36-in. and 3-in. pipelines.

The minimum trench cover requirements (from top of the pipeline) were 0.6 m and 1.0 m for the 36-in. and 3-in. pipelines, respectively. These requirements were specified because of seabed mobility between KP 0 and 24.

High tidal range (about 10 m) in shallow water (

Additionally, seabed roughness also existed in this area.

Selecting a method

Subsea soil conditions and the capacity of post-lay trenching equipment indicated that piggyback laying all along the route would not achieve required depth of burial.

With landfall construction involved at Walney Island, an acceptable solution for the offshore pipelines construction became:

• Piggyback laying of the 3-in. pipeline with the 36-in. pipeline (90 mm coating thickness -- CT) for the shore pull through a predredged trench, subsequently backfilled.

• Following shore pull, separate laying of the 36-in. pipeline (120 mm, 100 mm, and 90 mm CT) and the 3-in. pipeline up to the North Morecambe platform, with a nominal separation of 12 m between the pipelines.

• Outside the predredged trench zone, individual post-trenching for the pipelines in accordance with their respective trench-cover requirements.

• Installation and tie-ins of spool pieces at the base of the platform risers.

DLB 1601 was to conduct offshore pipelaying. To reduce barge time, dual lay was selected instead of separate laying of the pipelines with single pipe laying.

Following completion of the shore pull, pipelaying the 3-in. in dual lay could have been initiated with a deadman anchor, and later a flanged spool piece could have been installed for the tie-in to the piggybacked section of the 3-in. pipeline.

Technical and commercial evaluations, however, did not reveal such a solution (in a zone of mechanical backfill) to be cost-saving and risk free. As a result, the preferred solution was to perform a transition operation from the piggyback mode to the dual lay mode without any intermediate abandonment of the pipelines.

[Editors note: The conclusion of this series deals with this transition operation.]

Engineering

Detailed design and construction specifications for the pipelines were completed by British Gas Exploration & Production which supplied the coated pipe. In accordance with the pipelaying activities under the scope of the vessel DLB 1601, installation engineering was divided into three phases:

Phase 1: Piggyback laying during the shore pull operation (4.3 km length).

Phase 2: Dual lay operation for rest of the route.

Phase 3: Transition operation from piggyback to dual lay.

This order was selected so that the Phase 3 engineering could be guided by the results of other engineering phases, wherever necessary. The draft version of British Standard BS 8010, Part 3,2 was generally followed for the pipelines construction.

According to company specifications, however, the allowable von Mises pipe stresses (for both pipelines) during laying, with only functional loads under consideration, were restricted to overbend pipe stress (0.8 specified minimum yield stress [SMYS] and sagbend pipe stress (0.75 SMYS).

For dynamic pipelay analysis to determine the limiting sea states, the pipe stresses were allowed up to 0.94 SMYS.

Phase 1

Although most of Phase 1 engineering was conventional, certain special activities implemented during the last stage of piggyback laying facilitated the subsequent transition operation. (They are dealt with in the conclusion.) For piggyback laying during the shore pull, the following points are important:

• A 50-m long rigid stinger was used during piggyback laying. Thereafter, it was used to lay the 36-in. pipeline in the dual-lay section.

The stinger was operated under buoyancy control. During any laying condition (piggyback or 36-in. pipelay), the stinger buoyancy moment (in relation to stinger/barge connection hinge) was designed for a constant value of 1,300 metric ton-meters. This was around 80% of the total buoyancy-moment capacity and thus positioned the stinger in a relatively high position.

This value was selected for the shallow water laying and to avoid changes in the stinger setting from time to time. Only when piggyback laying began and for any abandonment (or laydown)/recovery operation, the stinger settings were designed to follow a stinger-tip depth controlled mode.

• A nominal laying tension of 60 metric tons was selected, keeping the pipe stresses well within allowable limits.

• To minimize the pull force during shore pull, 150 buoyancy tanks, each having 4.9-metric tons buoyancy, were installed on the bundle pipeline to reduce its submerged weight.

• Two linear winches were used for the shore pull. With a sheaved skid system, the maximum pulling capacity was 800 metric tons with a four-fold pulling system or 1,200 metric tons with a six-fold pulling system.

Dual lay -- 3 in. concerns

The first task of the dual-lay engineering was to design a suitable laying ramp for the 3-in. pipeline, complete with a compatible tensioner and an abandonment/recovery (A/R) winch.

With pipe storage on the starboard side of the barge, a straight laying ramp for the 3-in. pipeline was impossible.

Fig. 2 (100969 bytes) shows the configuration of the 3-in. pipelaying ramp, which had three sections:

• Initial straight section.

With piggyback laying to be followed by dual lay, the initial ramp section, up to 76 m from the first welding station (WS1), for the 3-in. pipeline was kept the same for both laying modes.This part of the ramp was parallel to the main (36-in.) laying ramp with a separation of about 1.5 m. All welding for 3-in. pipe was performed in this section.

• Curved section.

To ensure that the final straight section of the ramp was clear of the Clyde Crane pedestal structure at the barge stern, two directionally opposite horizontal curves were inserted into the ramp. The radius of each curve was 48 m, resulting in about 80% SMYS bending stress in the 3-in. pipeline.

The non-destructive testing (NDT) and weld-repair station was positioned in a short straight section of 6 m between the two curves. A total of 15 supports (DLR1 to DLR15) with combined horizontal and vertical rollers controlled the curved configuration of the 3-in. pipeline in the horizontal plane, 2 m above the barge deck.

By this means, a lateral separation of about 12 m from the center line of the main (36-in.) laying ramp was achieved at the barge stern.

• Final straight section, external stern ramp.

The designs of the final straight section and the external stern ramp were completed from a number of pipelay stress analyses that considered such various laying conditions as different pipewall thicknesses, different water depths, current effects, and transition operation.

Consequently, the required capacities of the tensioner and the A/R winch were also determined. These studies indicated the need for an external stern ramp for laying the 3-in. pipeline. An 18-metric ton capacity tensioner and a 25-metric ton capacity winch also proved adequate.

The contribution of pipe axial stress was considerable in the 3-in. pipe equivalent-stress calculations. With this axial-stress effect, a radius of curvature of 56 m for the external stern ramp was required (by the 3-in. pipe grade) to satisfy the laying-stress criteria.

With no such ramp readily available, a new external stern ramp was designed and fabricated for the North Morecambe project.

This new ramp was, however, designed with future projects in mind to lay 2-4 in. pipelines in dual lay. For 2 and 3-in. pipe, the minimum pipe grade may be Grade B, whereas, the minimum pipe grade for 4-in. pipe must be X-42 to use this ramp.

As shown in Fig. 2,(100969 bytes) the tensioner (Western Gear model LPT 40) was located at the beginning of the final straight section, 29 m from the barge stern. The A/R winch (HBM 25-metric ton, limited torque) was also aligned with this final straight section.

A total of 14 supports (SR1 to SR14) were designed for the combined straight section (six supports) and for the external stern ramp (8 supports). To allow some lateral movement of the 3-in. pipeline over the laying ramp, it was given a "funnel" shape to permit maximum lateral movement of the pipeline of about 0.8 m at the stern ramp tip.

Work stations for anode installation and field joint coating were in this straight section of the ramp.

Dual lay parameters

For dual lay of the 36 and 3-in. pipelines, the individual laying parameters for each were first determined. It is, however, important to note that their nominal laying tensions were consistently verified to ensure the desired position of their touch-down points on the seabed during dual lay.

It was particularly important to reduce variance in the separation (post-lay trenching being involved) between the two pipelines because of transverse currents and/or during laying along the curved sections of the route.

For the 36-in. pipeline, different thicknesses of concrete-coated pipes were to be laid during the dual-lay phase. The water depth along the route also varied 10-40 m. The required nominal laying tensions and the corresponding touch-down point (TDP) distances from the barge stern for different concrete-coated pipes were evaluated for the corresponding water depth.

Before the nominal laying tensions for the 3-in. pipeline were finalized, the minimum and maximum limits of allowable laying tensions (TMIN, TMAX) in different conditions were determined. These limits allowed for flexibility of the TDP during laying.

Evaluating these limits involved imposing a maximum pipe separation, in addition to the allowable pipe stress criteria, over the last ramp support (SR14). A maximum vertical separation of 1 m (to keep the pipeline well within the vertical rollers) was used for determining the maximum allowable tension.

For determining the minimum tension limit, "no separation" was an acceptable condition.

From these allowable laying tension limits, the nominal laying tensions were then carefully selected and the corresponding TDP distances from the barge stern determined.

The behavior of 3-in. pipelay stresses resembled the laying stress behavior (with S-lay method) of large-diameter pipe in deep water from a conventional stinger.3

With the nominal laying tensions and functional loads, the maximum freespan stresses were always found to be less than 15% SMYS. The maximum overbend stresses, on the other hand, were very close to the allowable limit of 80% SMYS in almost all cases.

As usual, high pipe stresses occurred locally when an anode passed over the roller supports. This phenomenon was ignored, however, considering the short duration of such local excessive stresses (which are generally allowed by DnV,814 up to a strain limit of 2%).

Pipelines separation along curves

There were two curved sections along the pipeline route (Fig. 1)(111973 bytes), the first close to shore and the second 850 m before the laydown target.

Because the TDP distances of the 36-in. and 3-in. pipelines were different, the 12-m nominal separation between the lines was expected to change during laying in these curved sections.

Therefore, during laying along a curved route, the TDP distance of the outer pipe was engineered to be shorter than that of the inner pipe to prevent any decrease in separation.

A detailed analysis of the pipelines separation during laying along the second curve only was conducted because the first curve was actually very short.

Fig. 3 (54320 bytes) shows the derived variation in separation with stepwise barge movement (two double joint/step), taking into account the variation in water depth and possible adjustments in the 3-in. pipelaying tension. These results did not include the effects of any change in 36-in. pipelaying tension or any transverse current.

With the nominal laying tensions of 36-in. and 3-in. pipelines in the maximum water depth of 40 m, typical variations in separation with the barge steps (marked as "nominal") have been shown in Fig. 3 (54320 bytes) for reference.

As indicated, the increase in separation during laying along the second curve was expected to be 17-46%. The increase in separation strongly depended on the difference in TDP distance of the two pipelines. Moreover, with barge movement, the separation was found to increase nonlinearly and then become constant up to the end of the curve laying.

Transverse currents

The effect of transverse currents on the 36-in. pipelaying was assumed to be marginal because of the high laying tension, stiffness, and weight of the pipeline.

On the other hand, the behavior of 3-in. pipelaying was anticipated to be quite problematic because of these currents, particularly regarding as-laid shape on the seabed. Certain qualitative studies therefore estimated the probable variation in pipelines separation due to currents.

The maximum expected tidal currents during installation were about 0.75 m/sec at the sea surface and 0.5 m/sec near the seabed. The direction was nearly perpendicular to the pipelaying plane over almost the entire route.

For the determination of 3-in. pipelay configuration under currents in maximum water depth, each pipewall thickness case with its TMAX, TMIN values was analyzed. Different current intensities (25%, 50%, 75%, and 100% of the design value) were considered.

Fig. 4 (95549 bytes) shows the variations in pipeline lateral displacements at different locations with current intensities. Comparatively large pipeline displacements in the freespan were confirmed from these results, although the 3-in. pipelay stresses were not a problem.

As shown in Fig. 4,(95549 bytes) the lateral shift of the touch-down point was found to be relatively small because of the resistance from soil frictional forces. One can therefore understand that the current effect produced no immediate total shift of the touch-down point but does so progressively related to the barge movement until the pipe moment at the touch-down point becomes 0.

As a result, the pipelines separation gets progressively reduced or increased depending on the current direction. Fortunately, the current intensity and direction over a few hours (several barge movements) do not remain constant.

Studies considering time-dependent current intensity and barge movement revealed that the maximum variation in the pipelines separation would be about (3 m through the straight route section. During laying along the second curve, the separation between the 36-in. and 3-in. pipelines was expected to be 9-21 m including both the route-curvature and current effects.

Limiting sea states

For dual lay, the following method was used to evaluate the limiting sea states, assuming that the barge pitch is the most critical motion for overstressing the pipelines.

Step 1:Determine the limiting sea states for the 36-in. pipelaying to avoid overstressing of the pipeline, barge or stinger "bottoming-out" in shallow water, and pipeline "lifting-out of stinger," whichever is critical.

Step 2:List the allowable barge-pitch motions in different water depths corresponding to the limiting sea states determined in Step 1.

Step-3:Verify the 3-in. pipeline stresses for these barge pitch motions. If the stresses fall within the allowable limits, consider the limiting sea states for 36-in. pipelaying as applicable for both pipelines.

Otherwise, evaluate the acceptable barge-pitch magnitudes for the 3-in. pipeline and determine the corresponding limiting sea states which will be treated as valid for both pipelines.

This method was justified by installation procedures which involved simultaneous abandonment of both pipelines in case of adverse sea states and subsequent recovery using the same sequence in reverse order.

The barge-pitch magnitudes corresponding to the limiting sea states for 36-in. pipelaying were taken for the 3-in. pipelay stress verification in terms of imposed barge draughts at bow and stern.

In all cases, the 3-in. pipeline stresses were found well within acceptable limits. It was thus decided to consider the 36-in. pipelaying limiting sea states as the overall limiting sea states for dual lay.

Permissible significant wave heights were found to vary 2.2-8.4 m depending on the laying tension, water depth, wave incidence angle, and zero-up crossing period.

Acknowledgment

The authors thank British Gas Exploration & Production Ltd. and McDermott-ETPM West Inc. for permission to publish this article and personnel in the North Morecambe project for their contributions.

References

1."Specification for Line Pipe," American Petroleum Institute Specification 5L, 38th Edition, May 1, 1990.

2."Code of Practice for Pipelines -- Part 3. Pipelines subsea: design, construction and installation," British Standard 8010, 1993.

3.Bruschi, R., et. al., "Laying Large Diameter Pipelines in Deep Waters," presented at the 13th International Conference on Offshore Mechanics and Arctic Engineering (ASME), 1994.

4."Rules for Submarine Pipeline Systems," Det norske Veritas, 1981.

The Authors