Polyester ropes can meet design criteria to extend MODU mooring capability

Comparison between the 10-year criteria and hindcast maximum wave and associated sea state at Lloyd Ridge block 399 during Hurricane Ivan. Note that the directions of wind, wave and current in a 10-year storm are assumed collinear.

The first generation PIM had suction pile anchor/subsea connector/wire buoys. The top of the PIM connected to rig wires when the MODU arrived at the well site. This system was applied successfully at Na Kika and Coulomb fields, in water depths of 1,524-2,438 m (5,000-8,000 ft) even though the rig had 1,372 m (4,500 ft) variable deck load capacity.

The PIM concept used two sets of mooring lines so that one could be used at the rig and the second could be pre-set at the next well site. In this way, the MODU could move to the next site by disconnecting and buoying off PIM lines rather than waiting for the entire mooring system to be retrieved. This also allowed the rig to begin working at the new site within about 24 hours of arrival, reduced idle rig time from 7-8 days down to 2 days or less, and also decreased the exposure to weather related downtime.

The higher initial capex for the two sets of PIM lines was more than offset by opex savings within 24 months, depending on the frequency of moves and the rig day rate.

The PIM system also improves a MODU’s stationkeeping. The semi-taut-leg steel PIM lines improved a rig’s uptime by reducing rig offset compared to a typical catenary system. Suction pile anchors allow higher uplift angle and precise installation for a smaller mooring footprint with more flexibility in dealing with interference to mooring.

The first generation PIM met design targets and set mooring records, but there was room for improvement. The second generation system targeted reduced capex and opex in addition to reduced radius of the system’s watch circle. To that end, lightweight and corrosion-free polyester rope appeared to be a candidate to replace heavy steel wires and submersible buoys. Initial studies also indicated the polyester rope PIM system could have better stationkeeping capability.

At that time, polyester mooring guidelines were under development. There was no proven installation procedure or hardware. Most on-board mooring lines were steel wires which generate torque under tension. Conversely, most polyester rope design was torque-balanced. So, the unknown effects of torque mismatch between rig wire rope and polyester mooring rope was a concern. There also was concern about polyester rope bending-tension fatigue performance. Few polyester rope test results were available. Lastly, the nonlinear properties of polyester rope and no industry experience with polyester mooring line damping made polyester PIM system design performance difficult to confirm.

To maximize the cost savings of a polyester PIM system, Shell designed the second generation system to extend the water depth rating of the TransoceanDeepwater Nautilus and the Marianas.

Working with contractors, rope suppliers, installation contractors, and regulatory agencies, Shell resolved most of the issues. The second generation PIM was installed in the Gulf of Mexico in water depths from 2,286 m (7,500 ft) to 2,774 m (9,100 ft). Since that time, the PIM has been recovered and redeployed at more than a dozen locations.

The system enabledNautilus to operate in twice the water depth of its initial rating. The MODU and PIM survived hurricanes Isidore and Lili. However, the mooring wires failed at the center of Ivan. This was not a surprise since the Category 4 hurricane far exceeded the MODU mooring design criteria. Following that, full-scale testing was done to evaluate the PIM system.

Wire-rope interface

The unknown effect of connecting polyester rope to steel wire was one hurdle. Shell’s earlier studies indicated that the risk of premature failure or strength reduction of polyester rope when connected with a wire could be mitigated by proper rope design. The remaining question was what the magnitude of the wire fatigue life reduction would be and what to do about it.

One proposed solution was to manufacture torque-matched polyester rope. Such a rope did not exist. Shell also determined that due to inherently different material properties, a torque-matched rope could match a wire’s torque only at a fixed load, such as operating tension. In reality, mooring tension in different lines of the same system vary significantly with the line bearing angle and the tension variation increased as sea states worsened. Also, a low-efficiency swivel still would be needed to make handling the steel wires easier. This swivel would make the torque match concept more difficult to implement as it would be difficult to predict when the swivel would lock-up.

So, from the design view, the best way to resolve the wire fatigue problem is not 6- or 8-strand wires in a polyester mooring leg. But, most MODUs use some steel wires for raster handling during rig moves and replacement with torque neutral components such as chains would be cost-prohibitive.

Logistically, one PIM design to extend mooring capabilities of more than one rig would maximize flexibility and minimize cost. Shell expected the need to extend mooring ratings of two contracted MODUs in deepwater drilling, Transocean’sNautilus uses 8-strand wires, whereas Marianas uses 6-strand wires. This made it impossible to design one torque-matched PIM system for both rigs.

After theoretical investigations, design studies, and consultations with rope manufacturers, the rig owner and mooring handling contractors, Shell proposed to conduct a full-scale field trial of PIM based on torque-balanced polyester ropes.

Full-scale field trial

The full-scale trial was to test Shell’s rope handling method and design approach. The handling procedures and required hardware directly impact the feasibility of using polyester ropes in a mobile mooring system. The difficulties experienced in one of the early field trials of polyester rope handling procedure by other operators deepened concerns over offshore handling of ropes.

From January to August 2000, a JIP led by Shell successfully prepared and conducted a full-scale field installation and recovery test in 1,890 m (6,200 ft) of water in the Gulf of Mexico onMarianas. The wire/polyester-rope/wire PIMS mooring configuration was used in two lines for about 1 1/2 months.

Key determinations from this field test included the following:

• Existing mooring procedures can be adapted to handle polyester ropes
• Polyester mooring ropes can be safely installed and recovered with existing AHV and equipment
• ROV inspection shows nearly no twist in ropes under operating tension
• Recovered ropes showed virtually no residual twist
• Lightweight polyester mooring extensions clearly showed potential to enable existing MODUs to operate in deeper water.

These findings spurred PIM system development based on torque-balanced polyester ropes.

Shell conducted post-trial extreme tension tests on full size ropes and wires to determine if the wire-rope interface used in the field trial would result in any strength degradation of either wires or ropes. The polyester ropes used in the tests were those recovered from the field trial in the summer of 2000. The same type of wire-rope-wire assembly was tested under extreme tensions. The tensions were cycled over 5,000 times to 50% wire’s MBL, and approximately another 100 times to 60% wire’s MBL. The 50% and 60% MBL tensions are the maximum allowable per API RP 2SK for quasi-static and dynamic analyses, respectively. Additional tests determined the wire twists as a function of tension.

No premature failure of any mooring hardware occurred during testing. After the tests, rope and wire samples were tested for residual break strengths. The residual strength of the ropes were a few percent below the new ropes’ MBL, whereas the residual strength of the wires were slightly above the new wire MBL.

In summary, the full-scale field trial and post-trial tests enabled Shell to select the proper polyester rope construction and termination, and to design the proper line sizes and configuration to use torque-balanced polyester ropes to extend a MODU’s mainly wire mooring lines to work in deeper water.

PIM configuration

The TransoceanNautilus’ main mooring hardware consists of a traction winch, 4,420 m (14,500 ft) long 95 mm (3 3/4-in.) 8-strand wires, 152 m (500 ft) of 4 5/16 in. chain and a drag anchor for each of the eight mooring lines. The rating of the mooring system of the rig was designed up to 1,524 m (5,000 ft) water depth, which was at the leading edge when the rig was designed in the early 1990s.

Based on the positive results of those tests and 8-strand wire properties provided by the supplier, Shell obtained agreement with Transocean, MMS, and USCG to supplement the rig’s on board mooring system with a polyester rope PIM system. The PIM line typically has the following components (below the rig wire and swivel):

• 3 9/16 in. connecting chain with a chaser stopper, one 50 kip inline submersible buoy with 100 ft of 3 9/16 in. R4 chain. The outer-dimension of the buoy measures 8.8 in. x 8.8 in. x 24.8 in. This buoy is added to allow the preset ropes to be buoyed off the seabed and to provide dynamic resistance to twists of the rig wire,
• Polyester ropes (800-ton MBL, 6 9/10 in. with varying length segments to suit for different water depth)
• 6-strand wires (3 3/4 in. x 610-1,067 m (2,000-3,500 ft),
• Suction pile anchors with 15.5 m (51 ft) of 3 3/4 in. forerunner wire and Delmar subsea connector.

The rig wire payout typically varies from 427 m to 671 m (1,400 ft to 2,200 ft) depending on water depth at the anchor and the mooring line scope or other constraints on the sea floor.

The polyester PIM system is the practical mooring solution that allowsNautilus or a similar MODU to drill in up to 10,000 ft of water without waiting for the torque-matched rope to be developed. For a MODU with marginal payload capacity in ultra deepwater, such as Marianas, the polyester PIM would not require costly upgrades to the rig’s existing payload capacity. So far, the polyester PIM system has been deployed successfully at more than 12 sites since its first deployment in August 2001.

Hurricane experience

In September-October 2002,Nautilus was moored using the polyester PIM system at Shell’s Mensa field (MC 730, water depth 1,646 m [5,400 ft]) when two hurricanes, Isidore and Lili, moved through the Gulf. Based on satellite photos and hurricane wind force diagrams, the rig experienced strong tropical storm wind forces when the center of Hurricane Lili passed at about 209 km (130 mi) away from the rig. The platform and the polyester mooring system survived the hurricanes without any visible damage.

After more than two years of service, the used samples of polyester rope inserts and rig wires were tested. The used rope sample was tested for stiffness at three different mean loads and load ranges according to API RP 2SM. The results showed that the rope had stiffened at lower load levels while the stiffness at high-load level was almost unchanged. This indicated that the ropes have been bedded-in, and their mooring performance has improved.

To maximize the information from the test rope sample, no destructive test was performed at rope level. The sample was dismantled carefully to examine and test the subropes. The used subrope break strengths were compared with the new subrope strengths. Using the correlation formula between subropes and full-size rope provided by the rope maker, it was found that after over two years of service, the polyester insert’s strength was about 4-5% lower than the new ropes’ average actual breaking strength, but was still about 2% higher than the rope’s specified design MBL. These results confirmed that the performance of polyester ropes in the PIM system met design expectation.

In addition to polyester rope tests, a sample of the rig’s 8-strand wire was tested. The residual wire strength at 740 tons was higher than wire design MBL of 680 tons. Test results also revealed that the 8-strand rig wire’s torque is significantly higher than any earlier available estimate. At approximately 20% MBL load, the 8-strand wire’s torque was about 70% higher than that of an equal size 6-strand wire. Its twist at the same load level is about twice that of an equal size 6-strand wire. The wire’s torque-tension relationship is nonlinear in tension ranging from 550% MBL. Based on these findings, a section of 152 m (500 ft) of 95.25 mm (3 3/4 in.) 6-strand wire (the same type as the bottom wire) was added between the top of a pre-set mooring leg and theNautilus rig wire swivel to provide protection.

On Sept. 15, 2004, Hurricane Ivan, a strong Category 4 storm with maximum sustained wind speed of 140 mph and gust of 175 mph, hitNautilus directly at Lloyd Ridge block 399. The hindcast wind data clearly showed that the eye wall of Ivan swept the rig twice as the storm passed.

The mooring design criteria forNautilus prior to Ivan was the 10-year storm (that is the maximum MODU mooring design criteria per API RP 2SK, 2nd-Edition 1996).

Hindcast data indicated that the sea state at LR 399 during Ivan far exceeded the existing 10-year hurricane criteria. The results of a post-Ivan mooring analysis found that the rig’s wires would fail (when tension safety factor is reduced to 1) in a storm with return period of 65-year to 85-year, depending on the heading (assuming the wind, wave, and current of the storm are collinear for simplicity).

Each mooring line had 1,067 m (3,500 ft) of bottom wire, 2,286 m (7,500 ft) of polyester ropes (2 x 914 m [3,000 ft] and 457 m [1,500ft]), a 50 kip buoy, and 152 m (500 ft) of 6 x 3-in.-strand wire connected to approximately 610 m (2,000 ft) of 8 x 3-in.-strand rig wire. Lines 3, 4, 7, and 8 each had one 13.75-m- (45-ft)-long polyester insert. The nearly equal line separation angle is typical for a MODU mooring design, and is assumed to give more uniform offset radius in operating conditions regardless of weather headings. However, this mooring pattern may need to be optimized for survival because a MODU’s total mean load coefficients could vary more than 50% with storm heading.

The lay pattern of 7 broken lines indicated that the first failure occurred either in line 4 or line 5, and line 3 failed after lines 4 and 5. After that, the remaining 5 lines also failed. The failures of mooring lines 1 to 7 occurred in the rig wire near the fairleads, where the wire is subject to bending-tension loads. The post-storm inspection of wire failures found signs of wire overloading. Line 8 failed in the suction pile anchor’s forerunner wire, and was dragged by the drifting rig for over 113 km (70 mi). The weight of steel hardware in lines 1-7 caused broken line to fall to the seafloor. The pressure at over 2,743 m (9,000 ft) depth crushed submersible buoys. However, there was no failure in any part of the polyester ropes in the PIM system.

After the mooring ropes were recovered from the seafloor, Shell conducted an investigation of the polyester ropes’ condition. Test methods and matrices were designed to allow comparison with test results obtained prior to Hurricane Ivan.

In survival draft, it was found that when the wind speed reaches 98.5 kts, the mean load of wind alone would exceed the windward rig wire’s MBS of 1,500 kips. In view of the rapid increase of wind, wave, and current of the hurricane and inspection of damaged wires, it is deduced that the rig wire failures were due mostly to overload rather than bending-tension fatigue. Therefore, the maximum load in lines 1-7 before their failures were estimated at above 90% of polyester ropes’ design MBS of 800 tons. This assessment is supported by the resultant damage of the buoy’s steel frame from the chaser stoppers pulled by the ropes’ recoil force after overload failure of rig wires.

One sample cut from the bottom segment of 914 m (3000 ft) rope in line 8 showed residual strength of at least 90% new rope MBS; whereas another sample cut at the lower end of the mid 914-m (3,000-ft) segment showed residual strength of 95% of new rope MBS. Both samples showed clear signs of being dragged on the seafloor by the drifting MODU.

The load-bearing cores of the ropes were found not affected by seafloor mud. In fact, no visible mud or sand was found to reach the core of any rope samples even though all 6 samples examined were exposed to seafloor in 2,743 m (9,000 ft) water depth, or were dragged on the seafloor.

The fact that the rope structure maintained its integrity in such extreme condition indicates that the rope specification has met our design expectations. More importantly, these results suggest that pre-laying the same type of polyester ropes with the built-in filter on seafloors in a controlled process during installation should not adversely affect ropes’ performance. This finding has the potential to simplify the current rule of “no contact” with seafloor for polyester ropes, thus allowing the industry to improve polyester mooring installation cost and efficiency.

Note: This is an excerpt of a paper presented by the authors at the Offshore Technology Conference in Houston.

Conclusions

Shell’s experience in developing the polyester PIM system and the evaluation test results since the 1990s had shown the following:

1. It is feasible to use torque-balanced polyester ropes to extend an existing MODU’s mooring capability even if the rig uses 6-strand rig wires. The rig wire’s as-built relationships should be determined before the mooring system configuration is finalized. This practical solution is cost-effective and does not adversely affect the rig mooring’s ultimate survivability.

2. Optimization of a MODU’s mooring pattern may improve its survivability in hurricanes of higher return periods.

3. The test results of polyester ropes recovered from the seafloor after Hurricane Ivan showed that the ropes’ filter met design expectation of protecting load bearing core from seafloor mud and other debris during frequent handling. This suggests that polyester ropes could be pre-laid on the seafloor in a controlled process without detrimental effects.

4. Short polyester test inserts do not necessarily represent long rope’s condition in the same system. A flawed test insert will be the weak link of the mooring system, and it could not be detected effectively by non-destructive testing. For a permanent mooring system, recovering the test inserts is costly and inherently risky.

5. More cost-effective methods are needed for in-situ monitoring of synthetic mooring lines. If test inserts have to be used for monitoring purposes, longer rope length would offer better test samples while reducing the aforementioned shortcomings.

H. Shu, D.A. Loeb
Shell Intl. E&P Inc.