OTC 17499

Floating Production Mooring Integrity JIP – Key Findings M.G. Brown, Noble Denton Europe Ltd.; T.D. Hall, Welaptega Marine Ltd.; D.G. Marr, Balmoral Marine Ltd.; M. English, U.K. Health and Safety Executive; and R.O. Snell, BP Exploration

Copyright 2005, Offshore Technology Conference This paper was prepared for presentation at the 2005 Offshore Technology Conference held in Houston, TX, U.S.A., 2–5 May 2005. This paper was selected for presentation by an OTC Program Committee following review of information contained in a proposal 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, its officers, or members. Papers presented at OTC are subject to publication review by Sponsor Society Committees of the Offshore Technology Conference. 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 a proposal of not more than 300 words; illustrations may not be copied. The proposal must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, OTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435. Abstract Over the last two years Noble Denton has been undertaking a Joint Industry Project (JIP) to investigate how to improve the integrity of the moorings used by Floating Production Systems (FPSs). The JIP has surveyed the world wide performance of all types of FPS mooring systems including FPSOs, semi submersible production units and Spars. Wide ranging support from 23 sponsoring organizations including operators, floating production contractors, regulatory authorities, equipment suppliers and inspection companies has enabled access to a significant pool of data. This paper utilizes the JIP data to discuss the following:

• Causes of system degradation • Consequences of mooring failure • Key areas to check on a mooring system • Fatigue implications of friction induced bending • Options for in-water inspection • The importance of connector design • Methods to detect line failure • Contingency planning

A few pioneering floating production units have now been on station for many years. Review of inspection data from these units shows that selective repair may be needed to maintain the design specification right up to the end of the operational life. It has been found that wear can be faster on leeside, as opposed to windward lines and that certain weighted chain designs are susceptible to damage. The likelihood of line failure and the implications need to be better appreciated. Following failure, it may well take several months to implement a full repair, due to a lack of spares/procedures and possible non-availability of suitable

vessels. However, it has been found that carefully planned and coordinated inspection operations can detect potential issues early on before more serious deterioration takes place. In general, mooring monitoring/instrumentation and access for in-water inspection seem not to be as advanced as might be expected for a system which is safety critical. Hence good practice recommendations are included which can be applied to both existing and planned future units. Introduction Unlike trading ships, Floating Production Systems (FPS’s), stay at fixed positions year after year without regular dry docking for inspection and repair. Since they cannot move off station, they must withstand whatever weather is thrown at them. Hence at times, depending on their location, their mooring systems need to withstand high storm loadings. Typically during design, mooring systems for harsh environments do not have much reserve capacity above what is required to withstand survival conditions. Therefore deterioration of the lines over time can increase the likelihood of single or multiple line failures. Multiple line failure could conceivably result in a FPS breaking away from the moorings and freely drifting in the middle of an oil field. The Mooring Integrity JIP has been concerned with assessing how mooring systems have performed in the field to identify the level of degradation which has taken place. Hence the project has looked at FPSOs, Semi submersible production units and Spars through out the world. The key objectives have been:

• To feedback operational and inspection experience to the industry and to mooring designers

• To publicize how hard moorings work, their importance and potential vulnerability

• To disseminate best practice guidance From the survey it has become apparent that certain problems have occurred and thus the JIP wishes to publicise these so that they can be taken account of during inspection of existing units and during the design of future units. Taking due account of past experience is particularly important when a design premise or specification is being developed for a new project.

2 OTC 17499

International Survey Significant effort was made to try and ensure that the international survey was as simple and straight forward as possible for respondents. To this end a custom designed spreadsheet based questionnaire with drop down boxes was developed. This spreadsheet was partially completed by Noble Denton, using information in the public domain, before being emailed out for checking and final completion. As well as the questionnaire face to face interviews were carried out with key personnel from different areas of the industry. Conference papers, in-house data and journals were also consulted. Response to the questionnaire was reasonable, but could have been better particularly for non North Sea regions. This perhaps gives some indication of the priority level that at present seems to be associated with mooring systems. Initially it was believed that offshore based staff would be able to complete the questionnaires. However, it became apparent that in some assets there is little in-depth knowledge about the make up and history of their mooring systems. Overall though, in summary, good data was obtained, but not on as many units as had been originally planned. Degradation Mechanisms Intrinsically mooring lines present a fairly simple system for keeping a floating object on station. However, experience from the field has shown that mooring is in fact a particularly difficult dynamic problem. Figure 1 illustrates a number of the degradation mechanisms which a mooring system will be exposed to every day of its operational life. Inevitably the performance of the system will decrease over time. Despite this, at the end of the field life, which in certain circumstances could be in excess of 20 years, the mooring system normally still needs to be capable of withstanding 100 year return period storm conditions. This represents a stern test for any 20 year old mechanical system. It is also logical that the longer a mooring system is out there, the higher is the probability that it will encounter extreme weather conditions. Many of the mooring issues mentioned in this paper refer to chain. This is because chain is normally selected at the two most challenging locations, namely the vessel interface and the sea-bed touch down. Since the loading regime is severe degradation may sometimes occur. However, experience over the years has shown that using wire in these areas does not give a true long term solution. The same would almost certainly apply to the use of fibre ropes.

Figure 1 –Mooring degradation and the key areas to inspect

Historical Incidents Given these degradation mechanisms a search was made of historical records to see what lessons could be learnt from past incidents. This search identified the following incidents which could have implications for present day systems, although particularly for the SALM the failure mechanism was unique to the system concerned:

• Argyll Transworld 58 production semi, complete

break away in 1981 • Fulmar FSU, complete break away in 1988 from the

SALM (Single Anchor Leg Mooring). • A series of semi sub multiple line failures in the

storms of Oct. 1991 and January 1992, see ref 5. • Petrojarl 1, 1994, 2 lines failed at the same time when

hit by a 20 to 25m wave ≈ 10º off port bow. The TW58 and the Fulmar 210,658dwt storage tanker both broke away after 6 years and 7 years on station. These durations tie in surprisingly well with the failure statistics reported later on. The TW58 was designed to and had disconnected its risers before breakaway, but it was still free drifting for 1.5 days in the North Sea before it was possible to attach a tow line to it. The Fulmar FSU did not have propulsion and was drifting for 5 hours before tow lines could be attached. Reference 5 is informative since it gives an idea of how much damage can be inflicted by unusually severe, but not freak, storms. The Petrojarl incident is significant since it shows that if there is a common degradation mechanism multiple line failure may occur virtually at the same time. In this case redesign of the chain guides and up-rating the chain resolved the particular problem. Consequences of Mooring Failure Environmental Impact The design premise of the majority of FPSs is that they should be able to withstand a single mooring line failure without the resulting increased vessel offset causing damage to the risers. Multiple line failure is only likely to occur if a failure has gone un-detected (see later) or if there is general degradation which

Bending & Tension

Highest Tensions

Corrosion

Impact & Abrasion

Wear & fatigue

OTC 17499 3

is affecting all lines in a particular quadrant to approximately the same extent, see

Initial Failure

Systematicdegradation

Failure to detect orreinstate in time

Failure detectedpromptly

Spares/procedures/vessels available to

repair lineMultiple linefailure

Possibility of completesystem failure

Severe weather

Inspect & possiblyrefurbish other lines

System fit forcontinued operations

Figure 2 – Possible Line Failure and Repair Scenarios

In the unlikely event of multiple mooring line failure causing rupture of one or more risers, the extent of hydrocarbon release will be strongly dependant upon whether or not the risers are still pressurized. Typically it is assumed that mooring line failure will be progressive and thus there will be sufficient time to shut down production and depressurize the risers, before the resulting increased vessel offset causes damage. However, the multiple mooring line failure which occurred on Petrojarl 1, when hit by a shock-inducing steep wave, shows that loss of position keeping on a non DP assisted vessel could occur remarkably quickly. This could possibly be in wave heights below survival criteria. Hence, it is recommended that on-board emergency procedures should identify what action should be taken in case of single or multiple riser rupture while the risers are still pressurized. If the risers are depressurized when rupture occurs, the extent of possible hydrocarbon release ranges from 100m3 to 2,500m3. This depends on field specific architecture such as the number of risers and the step out distance of the flowlines. Business Interruption Impact The business interruption cost of a single mooring line failure is not insignificant when the cost of anchor handling tugs, ROV or dive support vessels, new components and deferred production is taken into account. For example the following costs have been estimated for two typical cases.

• ≈ £2M for a 50,000 bpd N. Sea FPSO • ≈ £10M for a 250,000 bpd W. African FPSO

Multiple line failure which does not cause breakaway, but results in shut down for an extended period, would cost much more than the figures outlined above. Causes of System Degradation - Case Studies Corrosion and Wear – North Sea Production Semi A fascinating insight into the possible future performance of modern FPSs is provided by a purpose designed new build

North Sea production unit which has been in continuous operation for coming up to 20 years. During this time the FPS has experienced three mooring failures, plus significant defects have been found on two other lines during inspection. Interestingly all three line failures have been on lines which are defined as leeside lines based on prevailing weather conditions (see Figure 3). Leeside lines are in general under less tension and this seems to result in greater relative rotation/ motion between chain links and thus more wear. On first thought it might be expected that greater wear would be expected on the more heavily loaded windward lines. However, a bar tight line will in fact see less relative rotation between links than a slacker line subject to the same movement of the surface platform.

Figure 3 – Illustration of Windward and Leeward Lines

On this unit the failures have typically been on chain which at the no load equilibrium position is somewhat above the touch down point. Hence, in-water inspection during calm weather should make sure that this area is carefully inspected. Accelerated degradation in this area is highlighted by a more recent ROV inspection which has revealed that a studded chain has shed studs – see Figure 5. This is interesting, since it proves that studded chains can lose studs in situ rather than just during the relatively harsh handling that chain receives during a recovery operation by an anchor handling tug. Figure 4 shows a recovered link which was close to the link which failed in service. The failed link could not be found on the sea-bed. On the photograph it is interesting to note that the area of maximum wear is not at the point of contact between two links under tension, otherwise known as the inter grip area. Instead it is part way down the inner face of one side of the link. Damage was also noted on the crowns of other links. This suggests that some form of dynamic impact/grinding action is occurring which is wearing down the links. Significant inter link motion is thought to have been a factor contributing to the shackle pin failure illustrated on Figure 9. Losing material in this area is significant, since a finite elements analysis of a link will confirm that this is a highly stressed area. This is one of the reasons why it is recommended that tests should be undertaken to determine the actual break strength of worn mooring components.

Figure 2.

4 OTC 17499

Figure 4 – Example of Wear and Corrosion on a Chain Link from

the Sea-bed Touch Down Zone

Based on the original nominal diameter of this chain, which it is appreciated can vary; the combined wear and corrosion rate over its years of use has been estimated to be 0.6mm/year. This wear has occurred around the chain touch down area at the sea-bed, otherwise known as the thrash zone. 0.6mm/year of wear/corrosion is 50% higher than the maximum values found in API’s RP2SK and DNV’s OSE301 (refs. 3 and 4). It is interesting to note that a corrosion rate of 0.3 to 0.88mm/year for uncoated steel has been quoted on a long-term inshore project where sulphate reducing bacteria (SRB) induced corrosion might be experienced.

Figure 5 – In water inspection showing a Studded Chain which

has lost its Stud in situ

If the combined wear and corrosion rate is higher than that specified in mooring design codes this may well have significant implications for the true long term integrity of FPS moorings. It is appreciated that the wear rate reported here may well not be appropriate for all regions and platform types/system pre-tensions. However, the fact that this level of wear/corrosion has been experienced does highlight the importance of obtaining more data on wear/corrosion for other long-term moored units. The options for in-water inspection are discussed later on. There are, however, some limitations and hence it would be highly desirable if comprehensive inspection, including dimension checking, could be

undertaken of mooring components whenever a FPS comes off station or has repairs done to its moorings. Mooring line Configuration at the Vessel Interface The design of the vessel interface needs to minimize the potential for wear, corrosion or other forms of degradation. However, experience is demonstrating that this is not always being achieved. This is discussed below. The key points are relevant to mooring systems in general, not just to one particular type. Localised Wear Although there are a number of different turret mooring system designs, including both internal and external turrets, it is possible to categorize them as follows:

a) Non adjustable permanently locked off chains at the turret base

b) Adjustable chains which come up through the turret and are stored in a chain locker.

On Type a) systems the line tensions are not intended to be changed at any time throughout the field life. Type b) systems use a wildcat at the base of the turret similar to that found on a drilling rig running chains. Type b) FPSOs typically adjust their lines lengths and tensions either annually or even monthly. On some designs of spread-moored FPSOs the line lengths are also not intended to be adjusted and the required equipment for adjustment may not normally be present. If the line lengths are never adjusted during the field life this means that the same links in the thrash zone and at the turret interface will need to withstand the majority of the degradation. In addition, inspecting lines in situ is more difficult, since the chain is relatively inaccessible inside the trumpet/chain stopper. It is also much more difficult with such designs to pick up the chain off the sea-bed to make it more accessible for in water inspection. Being able to adjust line lengths can introduce its own perils. During a regular line tension adjustment operation on one North Sea FPSO there was a failure of the lifting and locking mechanism. This was partly due to a late change in chain size and the fact that the tolerances of forged chain links had not been properly taken account of. The failure resulted in the complete line being whipped out of the turret and falling down through the water column to the sea bed. Fortunately no one was hurt and there was no damage to subsea architecture. Modifications to the lifting and locking mechanisms should prevent another incident of this type occurring. It is worth noting that line run-outs are far from unknown on semi-submersible drilling rigs. This incident highlights the importance of reviewing all similar mechanical systems to check that, during the course of a long period of operation, chain/stopper wear or link dimensional variation may not jeopardize the integrity of the mechanism. Wear at Trumpet Welds – Internal and External Turrets On two type a) turret configurations wear has been experienced where the chains have been rubbing against the weld beads where the bell mouth joins with the parallel

OTC 17499 5

trumpet section (see Figure 6). This was first experienced on an early S.E. Asian external turret moored FPSO and more recently on an internal turret moored N. Sea FPSO. For the internal turret a slight shadow was seen on one of the chains during the annual workclass ROV chain inspection programme. To check out this anomaly a test tank mock up of the chain and trumpet assembly was built so that the capability of using a football sized micro-ROV to get in close to the bell mouth could be evaluated. This concept proved to be successful as can be seen from the photograph taken by a micro-ROV in the field, see Figure 7.

Figure 6 - Test Tank Mock-Up of Micro-ROV inspection of Chain

Emerging from Turret “Trumpet”

In the case of the external turret, in air access was such that it was possible to shroud the chains where they were rubbing against the weld beads with a replaceable material (ultra high molecular weight polyethylene sheeting). However, for the submerged trumpets on the North Sea unit a more long-term repair was needed which involved changing out the worn chain at the trumpet with larger diameter chain with a specially applied hardened coating (cobalt chromium) to reduce the severity of any future wear. A special connector (see Figure 15) was developed to allow the new chain to be connected up to standard common link chain. This approach avoided disturbing the wire section of the mooring line on the sea-bed, which is relatively susceptible to damage (birdcage). The original system designer was included in the review process for the repair operation. This represents good practice which, where possible, it is recommended should be followed for any future FPS mooring repair operations. On type a) systems the trumpets are typically pivoted about a single axis so as to minimize chain rotation and wear. Since the rotation is only about one axis and the trumpets are arranged around an approximate circle, the pivoting action cannot eliminate chain rotation for all the lines at the same time. Thus, to minimize wear over a long field life, there may be arguments for selecting a design which can pivot about two axes, although this would be mechanically more complicated.

Figure 7 - Micro-ROV Photograph of Chain Wear Notches where

Chain Emerges at the Trumpet Bell Mouth

Trumpets or guides are included on type a) FPSO designs to help guide the chain into the chain stopper. The trumpets themselves may include “angle iron” guides to ensure that the chain is in the right orientation when it enters the chain stopper. Once the chains are tensioned the trumpets have no real purpose unless they are required in the future for a new chain pull in operation. Interestingly, the pivoting chain stopper design which was adopted for the Brent Spar buoy did not include trumpets to help guide in the chain. The Brent Spar mooring was a successful design with a 19 year operational life and minimum wear on the chains at the stoppers when they were examined when the Spar was cut up in Norway. There was one failure but this was at a kenter connecting link. Such a failure is not surprising, since standard kenters are known to have low fatigue lives. There are, fortunately, now new designs of kenters with improved fatigue lives, but these still do not at present have classification society approval for long-term mooring. It is significant to note that the chain stopper on type a) designs is typically inboard of the pivot point. This means that the trumpet assembly does not automatically follow the motion of the chain. In fact it is contact between the chain and the outer face of the bell mouth which causes the trumpet to rotate. It is this contact, plus an associated sliding/sawing action, which seems to have led to the chain notches shown on Figure 7. Intrinsically there does not seem to be any reason why the chain stopper should be inboard of the pivot point. If it is outboard of the pivot point movement of the chain should cause movement of the trumpet without the need for chain contact with the bell mouth. This type of arrangement has been adopted on some more recent spread-moored FPSOs. For chain stoppers which are inboard of the pivot points it would appear that long trumpets are not helpful after the completion of the installation process. Thus it is recommended that careful checks should be made on any units which fit this category.

6 OTC 17499

In general achieving compatible chain surface hardness is important for long term integrity, since it affects wear. Unfortunately, at present chain hardness and wear do not seem to be evaluated in any detail. These factors should be taken account of during detailed design, but more work is needed on this area before it becomes part of the standard design process. Friction Induced Bending When a chain is under tension there will be friction and local yielding between the links which will inhibit inter link rotation. It is found that the higher the tension in the line, the greater the frictional forces. This friction can result in out of plane bending on individual links, see Figure 8.

Figure 8 - Illustration of Friction Induced Bending

Thus out of plane bending tends to become more of an issue as water depths and line pre-tensions increase. Over time cyclical out of plane loading can cause fatigue damage. This has been illustrated by a number of fatigue failures which have occurred on a taut moored CALM buoy off West Africa. Historically, mooring line fatigue has not been evaluated, partly due to the complexity, since MODUs work in different geographical locations areas on relatively short assignments. Today, for long term moored units, a fatigue assessment is typically carried out (refs. 3, 4 and 6). Such an analysis is normally in terms of tension loading cycles; it does not consider the combined effects of bending and tension. For long term moored units it is clear that friction induced bending fatigue should be evaluated. This is particularly important for deep water taut moored systems, but will still have some relevance for units in more moderate water depths. Physical testing has been undertaken to evaluate suitable friction coefficients for chain subject to out of plane bending9. In field experience has shown that the orientation of the links where they emerge from the bell mouth can significantly affect fatigue life. Improved fatigue life can be obtained if the “dynamic link” just outboard of the bell mouth is in a vertical plane. In other words the oval face of the link is at 90º to the sea surface. Excursion Limiting Weighted Chain and Mid Line Buoys From a mooring design perspective increasing the chain weight for a section of mooring line in the thrash zone can be a beneficial solution to reduce vessel offsets. This tends to be particularly applicable for moderate water depths in harsh

environments, which represents a particularly taxing mooring problem. There are a number of ways in which this can be achieved. However, from the international survey it is clear that great care is needed to select a robust system if such an approach is adopted. One way of increasing the chain weight, is to hang off short chain lengths from the main mooring chain. This was the solution adopted on one harsh environment FPSO. However, Figure 9 illustrates the damage that has been caused to one of the pins. It is believed that this damage may well have been caused by a dynamic pinching/grinding action of adjacent links.

Figure 9 – Photograph of a Partial Failure of a Hang-Off Shackle

Pin

Another possible approach to increasing the line weight over a certain section is to attach clump weights to the chains. illustrates half of a clump weight from a FPSO mooring line which utilized such a system. In this instance it can be seen that the bolts which kept the two half shells together have failed and the clump weight has thus split open. Again the dynamic loading of the line is thought to have led to the failure of the restraining bolts.

Figure 10 - Chain Clump which has become detached – only one

half of the Clump Weight Visible

OTC 17499 7

Other systems for increasing chain weight locally include a parallel chain system with triplates or using a larger chain size. Both of these systems appear to have worked successfully, although there is a need for careful design of connectors. This is because enhanced wear may be experienced due to an increased rotation resulting from a change in the weight per metre at the connectors. An alternative way of reducing FPS excursions due to mean wind, current and wave drift forces is to add buoys on to the mooring lines. However, problems have been experienced on one FPSO with the buoys becoming disconnected from the lines over time. Interestingly this seems to have been on leeward lines, which indicates that that the increased motion of the less tensioned lines may be contributing to the problem. Connector Failure – Unintended Line Disconnection Careful detailed design of long term mooring connectors is vital to ensure that they are fit for purpose. Figure 11 illustrates an unintended line disconnection on a FSU. This socket was at the transition from wire rope to chain. Hence, there was a weight per metre discontinuity which resulted in extra rotation at the connector. In this instance the socket pin was restrained from rotating by relatively small bolts. The pin wanted to rotate and it eventually sheared the bolts on the end cap which allowed the whole pin to work loose. It is interesting to note the size of the locking-pins which make up the double locking system on the purpose designed connector shown on Figure 15. The substantial size of these pins was based on hand calculations utilizing the expected line loads and an estimated friction factor. In the case of the unintended disconnection, at times, depending on vessel offset, the connectors would have been in the thrash zone. They would have experienced repeated lift up/set down contact with the sea bed.

Figure 11 - Unintended Line Disconnection due to the Failure of a

Socket Restraining Mechanism

“Dog Leg” or Wavy Mooring Lines on the Seabed During mooring line installation it is important that all lines should be laid straight from the anchor to the fairlead at the no load equilibrium position. This requirement should be emphasized in the installation procedures and reflected in any tug specifications. If “dog legs” or wavy lines do end up being present and they are pulled out by storm loading, this can lead to unbalanced mooring line tensions. In other words a system which was balanced originally with the “dog legs” may no longer be so. If one line takes more of the load coming in from a particular quadrant it is more likely to fail. If this originally taut line fails, the FPS may exceed its allowable riser offset limit if the remaining lines are too slack. At present non straight mooring lines have been noted on two North Sea FPSOs. On these units the initial pre-tensioning operation and the storm loadings which have been experienced have been insufficient to overcome the friction of the lines in the sea-bed mud. But to date, these FPSOs have not yet experienced storm line loadings as severe as the maximum loadings evaluated during the mooring design process. It will be interesting to see if, over the respective field lives, the “dog legs”/wavy lines are pulled straight or not and this should be monitored during annual ROV surveys. If straightening does occur the implications for mooring behaviour should be fully evaluated. Unbalanced Set-Up Pretensions On a long-term moored semi-submersible FPS, offshore personnel doubted the tension readouts on their mooring line winches, since damage was occurring to the wires on the winch drums. In addition, when grappling for certain components on the mooring line they were not found at the expected depth. Therefore, in calm weather, an underwater ROV survey was undertaken of the triplate connectors to obtain their X, Y and Z co-ordinates. From these positions and knowing the submerged weight of the line, it was possible to perform a catenary calculation to estimate the actual line tension. These tensions can then be compared to the winch tension readouts. This process showed that in the worst instance the calculated and the measured tensions were out by 160% ! Tension meters fitted to the base of pull in winches/windlasses can give a poor estimate of the tension in mooring lines, even if properly calibrated, since the amount of friction in the sheaves/fairleads is variable and difficult to quantify. In addition there is a possibility of full or partial seizure of the submerged lower sheaves or wildcats. To check this out, during a period of good weather, a carefully controlled Line Pay-Out/Pull-In test was undertaken. In this test each line was paid out in 2m increments and the line tensions were recorded. The lines were then pulled in again the same amount and the winch tensions noted. If this test is undertaken relatively quickly in calm weather conditions it would be expected that the same line tension would be obtained for the same line payouts. In actual fact this did not prove to be the case for all mooring lines, see for example Figure 12.

8 OTC 17499

Line No11

185.0

186.0

187.0

188.0

189.0

190.0

191.0

192.0

193.0

194.0

195.0

0.0 20.0 40.0 60.0 80.0 100.0 120.0

Tension (te)

Wir

e pa

yout

(m)

Figure 12 – Example of a Pay-Out/Pull-In Test for a Seized Sub

Sea Sheave

Historically semi-sub drilling units have been subject to relatively frequent mooring line failures. The work reported in this section shows that it is possible for a carefully set up Rig to have a seriously unbalanced mooring pattern which the Operators might not be aware of. Further information can be found in ref. 5. It is hoped that Pay-Out/Pull-In tests can be undertaken for other semis to determine how wide ranging or otherwise this occurrence could be. For long-term moored units it is recommended that a ROV should double check the line tension balance by measuring X, Y and Z co-ordinates of known points on the line or the touch down points. This should be done in good conditions and then a back calculation can be done of the line tensions. Recent Multiple Line Failure Incidents Unfortunately serious mooring incidents continue to occur. For example, a December 2004 North Sea storm resulted in a drilling rig losing two of its eight anchor chains. The resulting excessive excursions ruptured the drilling riser. During hurricane Ivan five MODUs broke free from their moorings and were set adrift. One of the units was a fifth generation rig. Fortunately, as far as can be determined, Ivan did not cause damage to the mooring systems on any of the long term moored FPSs in the Gulf of Mexico. Indicative Failure Statistics Based on the limited response obtained during the international survey, it is quite possible that only a fraction of the total number of mooring incidents which have occurred outside the North Sea have been reported. In the North Sea there are statutory requirements for mooring incidents to be

reported to the UK Health and Safety Executive (HSE). Although the North Sea is a hostile climate, units intended for use here are in general designed to a high standard. In addition, a number of units in the North Sea have been around long enough for age related problems to start making an appearance. It thus seems prudent to consider official statistics for this region to be a reasonable indicator of the likelihood of mooring line failure. Based on reference 2 for the period 1980 to 2001 it is reported that a drilling semi-submersible might expect to experience a mooring failure (i.e. anchor dragging, breaking of mooring lines, loss of anchor(s), winch failures) of once every 4.7 operating years, once every 9 years for a production semi submersible and once every 8.8 years for a FPSO. Thus it can be seen that although the failure probability for production units is approximately half that of a semi-submersible drilling unit, the statistics indicate that it would not be totally unexpected for the crew on a FPS to expect a mooring line failure at sometime during a field life which exceeds 9 years. Exactly how these statistics can be related to milder environments is difficult to quantify at present. Good Practice Recommendations In Air-Inspection Mobile Offshore Drilling Units (MODUs) need to recover their mooring lines and anchors on a regular basis when they move from one location to another. This provides periodic opportunities to undertake in-air mooring line inspection when the vessel is in sheltered water. Alternatively a spare line may be bought or rented which can be swapped out with one of the existing lines while the original line is taken to the shore for inspection and possible refurbishment. FPSs spend much longer on location than MODUs. Hence, their mooring lines are normally only recovered when the FPS moves off location. It is possible to recover mooring lines part way through a field life but this has two disadvantages, namely:

1. The lines may be damaged either during recovery or re-

installation 2. The whole operation is expensive since the services of

anchor handling and possibly heading control tugs will be required for a number of days.

Given that even in-air inspection will not necessarily detect all possible cracks and defects which may be present; there is an understandable interest among operators to undertake in-water inspection. However, there will still be times when anomalies are identified which can only be resolved with true confidence by undertaking in-air inspection. One definite advantage of in water inspection is that it is easy to identify which parts of the chain have been in the thrash zone and at the fairlead. This is more difficult to determine for long lengths of chain lying on a quayside. In-Water Inspection To date chain mooring components have been the subject of the greatest effort to develop in-water inspection methods.

OTC 17499 9

This is because they are typically used in the sections of moorings subject to the greatest deteriorative forces, particularly at the seabed touchdown (thrash zone) and at the vessel interface. Both windward and leeward lines should be inspected, but a particular check for wear should be undertaken on the leeward lines, see Figure 3. Care is needed when inspecting the touchdown zone, since potential hazards such as rocks or debris on the sea-bed can cause mooring line abrasion. These hazards may be partially obscured by the sea bed/mooring line and thus good visibility with powerful lighting is required. In-Water Chain Measurement A number of in water mooring chain measurement systems have been developed with varying success, ranging from simple diver-deployed manual calipers to a prototype stand-alone robotic system and ROV deployed systems. Diver inspections are not a favoured option. Mooring chains are highly dynamic and therefore are potentially dangerous when divers are in close proximity. Also diver inspection has proven to generate inconsistent results and has inherent depth limitations, for example, when checking the thrash zone. A stand-alone robotic system has been developed, but so far this has proven too large and cumbersome for practical offshore operations. In addition, it does not appear able to inspect the vital seabed touchdown or get in close to the fairleads. ROV-deployed systems include both mechanical caliper and ‘optical caliper’ systems. Mechanical calipers have met with limited success, primarily because during deployment onto chain they have the potential to be knocked out of ‘true’ and consequently may well have to be recalibrated between successive measurements. The most established ROV-deployable chain measurement system is effectively an ‘optical caliper’7, comprised of multiple high resolution video cameras and lights on deployment frame, which is equipped with scale bars in pre-assigned orientations and at set distances from each other and the cameras (Figure 13). The system measures the chain parameters by calibrating from the tool scale bars and resolving dimensions and optical distortions using offline image analysis software. This type of system has no depth limitation, requires no physical recalibration and can be configured to measure not only chain components at the seabed, but also in difficult to access regions such as the vessel interface. It can also be configured to measure other types of mooring ‘jewelry’ such as connectors, shackles and kenter links. The ‘optical caliper’ chain measurement technology is used extensively by offshore operators and is accepted by a number of offshore certification authorities. In this respect in at least one instance it has been used as the basis for an extension of

the prescribed recertification period for an in-service FPS facility.

Deployment guide Underwater

light

Camera block

Figure 13 - Illustration of ROV deployed ‘optical caliper’ measurement system7

Loose Stud Detection In studded chain, loose studs have been implicated in crack propagation and fatigue. Accordingly studded chain inspection and recertification protocols require the assessment of the numbers of loose studs and degree of ‘looseness.’ However, there is no consensual industry opinion with respect to loose stud reject criteria. Traditionally chains have had to be recovered for detailed loose stud determinations and have relied on a manual test, either moving the stud by hand or using a hammer to hit the studs. The resulting resonance (a ‘ping’ or ‘thud’) is used to assess whether a stud is loose or not. Recently an ROV-deployable loose stud detection system has become commercially available7. The system uses an electronically activated hammer to impact the stud and uses a hydrophone and a micro-accelerometer as sensors. A software program is used to distinguish between ‘loose’ and ‘tight’ responses. Cross checks can be carried out in that very loose studs can be detected using a ROV manipulator or a ROV deployed high pressure water jet. Component Condition Assessment As well as chain dimension checking it is also important to assess link integrity and condition. The overall, or general, condition of mooring components often gives insights into the types of deteriorative processes that are at play. For example surface pitting may be indicative of pitting corrosion, ‘scalloping’ or indentations of wear, fretting corrosion, or ‘anvil’ flattening, and unusual geometry may indicate friction bending, or plastic deformation (e.g. stretch).

10 OTC 17499

Underwater visual condition assessment by ROV is particularly difficult because of the inherent ‘flatness’ of video images from standard 2D inspection cameras. With 2D cameras it is very difficult to distinguish whether a visual artifact on a surface is merely a mark, or a region from which material has been lost (e.g. a pit). The shortcomings of 2D video can be addressed by using 3D visualization, a long-time goal in the underwater inspection sector. Over the last two decades a number of 3D visualization systems have been implemented but, until recently, with limited success due to problems with user comfort and impractical and cumbersome viewing systems. Advances in 3D camera design and the development of user-friendly viewing systems have led to the introduction of a new generation of 3D video systems7. These cameras come in a range of configurations, sizes and depth ranges and have proven very effective for the assessment of the surface condition and general geometry of mooring components. Improvements have also been made in video asset management, so that it is now easier to access data without trawling through hours and hours of video footage7. Marine Growth Removal A key challenge of conducting in-water inspection is getting access to the component(s) to be inspected. Materials which have been in sea water for extended periods accumulate varying levels of marine growth which can be heavy, depending on geography, water depth and season10, (see Figure 14). This growth needs to be removed so that the underlying mooring components can be inspected.

Figure 14 – Illustration of Marine Growth on Long Term Deployed

Chain

Cleaning options include manual brushing by divers, rotary brushing with wire or synthetic fibre brushes and ROV deployed high-pressure water or grit-entrained high pressure water. Each system has its own pros and cons. Once marine growth is removed it is possible to conduct various levels of inspection including general visual

inspection, dimensional measurement and assessment of mechanical fitness. Unfortunately cleaning off marine growth and scaling by high pressure water jetting may accelerate corrosion by exposing fresh steel to the corrosive effects of salt water. At present there are currently no in-water inspection methods for mooring components that do not require the prior removal of marine growth. This represents a technology gap which warrants further investigation. The time required to remove marine growth depends largely on the cleaning option chosen and in light of the cost of ROV vessels, can be a substantial component of the cost of an inspection program. Consequently it is essential that the planning stage of mooring inspection campaigns should consider the most suitable cleaning options for the expected conditions. Line Status Monitoring and Failure Detection Given the safety critical nature of mooring lines one might imagine that they would be heavily instrumented with automatic alarms which would go off in case of line failure. In practice many FPSs are not provided with such instrumentation/alarms – see indicative statistics below. On type a) turrets in which the chains are permanently locked off under the hull it is particularly difficult to monitor these lines in a reliable manner. For example, how do you readily distinguish between mooring line and instrumentation failure, without direct intervention ? Another factor which makes it difficult to be 100% sure of the condition of a set of mooring lines is that line breaks do occur along the sea-bed or in the thrash zone. If this happens the line will drag through the mud until the friction exerted by the soil surrounding the chain matches the tension in the chain at its sea bed touchdown point. Experience has shown that high line pulls are required to drag large diameter chain through the sea-bed. The following indicative statistics, based on data from the majority of North Sea based FPSOs, give an indication that instrumentation is not as prevalent as might be expected for such a heavily regulated region:

• 50% of units cannot adjust line lengths, • 50% of units cannot monitor line tensions in real

time, • 33% of units cannot measure offsets from the no-load

equilibrium position, • 78% of units do not have line failure alarms, • 67% of units do not have mooring line spares

available. The present position of the U.K. Health and Safety Executive is that Operators should have in place suitable performance standards for the time taken to detect a mooring line failure. This is particularly important as common mode failure mechanisms such as fatigue or wear are likely to be prevalent on more than one mooring line and early detection of a line

OTC 17499 11

failure with appropriate mitigation strategies could prevent system failure. Depending on the inherent redundancy of the mooring spread, the time taken to detect a failure could range from virtually instantaneous detection to detection in a matter of days. It is clearly not appropriate to rely on annual ROV inspection to check if a mooring line has failed. Monitoring the excursion of a FPS, particularly using differential GPS is inexpensive and will provide mariners with a feel for the mooring integrity. But without real time monitoring of the environment it is unlikely to indicate a line failure in anything but storm conditions, unless in deep water. Satellite drift is also a potential factor affecting the reliability of offset monitoring. New methodologies to detect a mooring line failure typically feature acoustic transponders deployed through the turret, attached to the hull of the FPSO, or installed on the seabed to provide an indication of the catenary’s profile. Such systems should be trialed in the near future in the North Sea. Another option may be a response learning system which takes into account the expected performance in measured weather conditions. The response will be different if a line fails due to a resulting change in the mooring system stiffness. Such an approach requires further development work. But if the concept proves successful this could prove to be a relatively simple and inexpensive retrofit. Contingency Planning - Spares and Procedures Based on the indicative failure statistics reported earlier it is quite conceivable that a FPS may lose a line during its operational life. There is likely to be a several month lead time to procure components such as large diameter chain, wire/fibre rope or purpose built connectors, see for example Figure 15. Hence, to minimize FPS safety and business exposure in case of line failure, it is believed to be well worthwhile to have spare lines, connectors and procedures available for immediate use if required. For deep water projects the procedures should ideally be developed which are based on a generic anchor handling vessel rather than a high specification installation vessel. Installation/construction vessels are unlikely to be readily available at short notice and tend to be expensive. If a line does fail and no spares are available it may be possible to “mix and match” making use of available equipment from the established marine supply and rental companies. However, the impact of introducing non standard elements into a mooring system is best considered before a failure occurs. Long term mooring (LTM) shackles should ideally be used as the connectors, but virtually any type of shackle including alloy shackles would do in the short term. Repairs of this nature should give time for the procurement of the correct equipment, which may take around six months depending on industry demand. Because the mooring system has been damaged and then modified, it may be necessary to obtain concessions from the relevant Classification Society/Independent Competent Person (ICP). A reduced operating envelope may have to be accepted during the period that the temporary repairs are effective.

Figure 15 – Purpose designed connector for common link to common link chain allowing some compliance in two planes

Maximum Sea State for Continued Production Following Line Failure Once a mooring line fails it is believed to be no longer appropriate to apply the lower damaged system line safety factors. This is because, in most instances, the reason for the line failure will not be immediately apparent. Thus with the increase in loading in the remaining lines there is an increased chance of a further line failure. Hence, it is recommended that the higher intact system line safety factors should be applied. Meeting the intact line safety factors with a degraded system will typically result in a reduction of the maximum allowable sea state. Data on the reduction in the maximum operational sea state in case of line failure should be readily available on all units. The international survey indicates at the present time this data is not generally available either with the designers or on the units offshore. Conclusions Moorings on FPSs are category 1 safety critical systems. Multiple mooring line failure could put lives at risk both on the drifting unit and on surrounding installations. There is also a potential pollution risk. Research to date indicates that there is an imbalance between the critical nature of mooring systems and the attention which they receive. On many FPSs there is an important need to improve the knowledge base of offshore personnel on the intricacies of their mooring systems and their potential vulnerability. This will help to ensure that mooring systems receive the amount of attention they deserve, particularly during inspection operations. The interface between the surface vessel and the mooring line requires particular attention for all types of FPS. Carefully planned innovative inspection making use of all possible tools has been demonstrated to be able to detect problems relatively early on before they become a potential source of failure. The use of micro-ROVs to gain access to restricted areas not accessible by conventional ROVs and divers has been part of the key to this success. The inspection which has been

12 OTC 17499

undertaken has shown the importance of achieving compatible surface hardness since it affects wear. Unfortunately, at present chain hardness and wear do not seem to be considered in any detail. In situ in-water inspection techniques are continuing to improve, but further developments are needed to provide dimensional data on links away from the inter-grip area and to improve the marine growth cleaning off speed. At present no in-water techniques exist to check for possible fatigue cracks and the development of such technology should be encouraged. Inspection access needs to be improved and design briefs should assign a higher priority to designing systems which are easier to inspect. On one long term deployed North Sea unit chain wear and corrosion in the thrash zone has been found to be significantly higher than what is specified by most mooring design codes. This wear seems to be more pronounced on less heavily loaded leeward lines compared to the more loaded windward lines. Hence, it appears that more interlink rotation is occurring on the leeward lines. More data is needed to find out if this is a general finding which could have long term implications for other FPSs in the North Sea and elsewhere. At present there is little data available which indicates how the break strength of long term deployed mooring components will be reduced by wear, corrosion including pitting and the possible development of small fatigue cracks. Thus to assess long term integrity with any confidence it is recommended that break tests on a statistically representative sample number of worn components should be undertaken. Recovered lines from the thrash zone and from the fairleads/chain stopper area would be ideal for testing. Such material is likely to be available whenever a FPS comes off station or has repairs done to its moorings. As well as break tests, MPI, photographs and comprehensive dimension measurements should be undertaken. It is important that this data should be fed back to the industry. Certain North Sea Operators have shown a willingness to make this data available. Offset monitoring has limitations in quickly detecting line failure unless a FPS is in deep water. However, it is cheap and easily installed. Hence it should be installed as standard on all units. In addition, all units should know the maximum sea state in which they can continue to produce in case one line fails. On board emergency procedures should identify what action should be taken in case of riser rupture while the risers are still pressurized, although the likelihood of this happening is low. A possible contributory mechanism for the relatively high failure line failure rate among drilling semi-submersibles has been identified. This is believed to be due to rigs thinking they have set up balanced pre-tensions when in fact this has not been achieved. Hence, it is recommended that Pay-In/Pay-Out tests should be undertaken to check whether the line tension readings can be relied upon,

Finally a general lack of suitable spare lines, connectors and repair procedures has been noted. Given the substantial procurement lead-time associated with these items it is recommended that Operators should review their assets to see how they could deal in the short term with one or more failed lines. The reported statistics show that line failures have been higher than might normally be expected for custom designed systems which are not regularly recovered and redeployed. Thus the business interruption potential due to mooring problems should not be underestimated. Acknowledgements The crucial support to this project provided by the following supporting organizations is gratefully acknowledged: B.P., Chevron Texaco, ENI, Norsk Hydro, PetroCanada, Statoil, Bluewater, SBM, Maersk Contractors/North Sea Production Company, Wood Group/Amerada Hess, Bureau Veritas, ABS, Lloyd’s Register, U.K. Health and Safety Executive (HSE), Craig Group/IMS, Vicinay Cadenas, Ansell Jones/Oceanside, MARIN, OIL/Zhengmao, Welaptega Marine, Balmoral Marine, BMT/SMS, National Oilwell-Hydralift/BLM, Hamanaka Chains and in particular to Williams Marine Enterprises. The project Steering Committee itself has been exceptionally strong and it is hoped that it will be possible for the committee to continue to meet during future FPSO Forum/JIP Weeks. This will provide a continuing reporting/recording mechanism as more data becomes available. New participants to this committee will be welcome. References 1. “FPS Mooring Integrity JIP Report”, A4163, 2005, Noble Denton

Europe Limited, Aberdeen. 2. “Analysis of Accident Statistics for Floating Monohull and Fixed

Installations” HSE Research Report 047, 2003. 3. “Recommended Practice for Design and Analysis of Station-

keeping Systems for Floating Structures”, API RP 2SK, 1997. 4. “Position Mooring,” DNV Offshore Standard OS E301, June 2001 5. “Design and Integrity Management of Mobile Installation

Moorings,” HSE Research report 219, 2004 6. “Station-keeping systems for floating offshore structures and

mobile offshore units,” ISO Draft International Standard, ISO/DIS 19901-7, Part 7, 2004

7. “Cost Effective Mooring Integrity Inspection Methods,” Hall, A.D., OTC 2005, May 2-5, Houston, paper 17498

8. “Review of Mooring Incidents in the Storms of October 1991 and January 1992,” HSE Offshore Technology Report – OTO 92 013.

9. “Failure of Chains by Bending on Deepwater Mooring Systems,” Philippe, J., OTC 2005, paper17238.

10. “Marine Bio-deterioration : an interdisciplinary Study,” Costlow, J.D., and Tipper R.C. (Eds.), pp. 384, Naval Institute Press, Annapolis, Maryland, 1988.