HSE

Health & Safety

Executive
Floating production system
JIP FPS mooring integrity
Prepared by Noble Denton Europe Limited
for the Health and Safety Executive 2006
RESEARCH REPORT 444
Crown copyright 2006
First published 2006
All rights reserved. No part of this publication may be
reproduced, stored in a retrieval system, or transmitted in
any form or by any means (electronic, mechanical,
photocopying, recording or otherwise) without the prior
written permission of the copyright owner.
Applications for reproduction should be made in writing to:
Licensing Division, Her Majesty's Stationery Office,
St Clements House, 2-16 Colegate, Norwich NR3 1BQ
or by e-mail to [email protected]
HSE
Health & Safety
Executive
Floating production system
JIP FPS mooring integrity
Noble Denton Europe Limited
No 1 The Exchange
62 Market Street
Aberdeen
AB11 5PJ
The main objective of this report is to improve the integrity of the mooring systems on Floating Production
Systems (FPSs). It is intended to be read and understood by non mooring specialists such as FPS
Operational staff - so that the people who live and work on FPSs will be better able to become more
involved in the vital task of looking after their own mooring systems. Meanwhile the included feedback on
the actual performance of mooring systems in the field should assist designers and manufacturers to
improve future mooring designs. Hence, the report attempts to identify gaps in the existing knowledge of
mooring behaviour and components to provide a road map for future work. Appendix C includes a paper
presented at the 2005 Offshore Technology Conference (OTC) which represents a stand alone summary
of the key points of the JIP.
This report and the work it describes were funded by the Health and Safety Executive (HSE). Its contents,
including any opinions and/or conclusions expressed, are those of the authors alone and do not
necessarily reflect HSE policy.
HSE BOOKS
A4163-01
CONTENTS
SECTION PAGE NO.
1 EXECUTIVE SUMMARY 11
2 INTRODUCTION AND SCOPE 17
3 MOORINGS OVERVIEW 22
3.1 Mooring Basics 22
3.2 Mooring Line Constituents 39
3.3 Determination of Minimum Break Load (MBL) & Maximum Stresses 58
4 CONTEXT SETTING - HISTORICAL INCIDENTS AND THEIR
SIGNIFICANCE 65
4.1 Long-Term Degradation Mechanisms 65
4.2 Multiple Line Failure Incidents 74
4.3 Petrojarl 1 Multiple Lines Failure (1994) 76
5 CONSEQUENCES OF MOORING LINE FAILURE 77
5.1 Single Line Failure 77
5.2 Multiple Line Failure 79
5.3 Danger of Hydrocarbon Release 81
5.4 Business Interruption Consequences - Two Case Studies 82
6 HANDLING, TRANSPORTATION/TRANSFER AND INSTALLATION 85
6.1 Transportation/Transfer 85
6.2 Installation of Mooring Lines and Connectors 86
6.3 Installation Watch Points from a Mooring Integrity Standpoint 93
7 CORROSION, FATIGUE AND WEAR (CASE STUDIES) 99
7.1 The Balmoral FPV An Industry Benchmark 99
7.2 Corrosion and Wear Allowance Discussion of Code Requirements 101
7.3 North Sea FPSO Apparent Corrosion and Wear Data 108
7.4 Sulphate Reducing Bacteria (SRB) Induced Pitting Corrosion 112
7.5 Stress Corrosion Fatigue 113
7.6 Wear Analysis (Shoup and Mueller Work) 118
8 UNBALANCED LINE PRE-TENSIONS (CASE STUDIES) 122
8.1 North Sea Semi-Submersible FPS 122
8.2 Line Payout/Pull-In Test 123
8.3 North Sea FPSO 124
9 MOORING BEHAVIOUR AT THE VESSEL INTERFACE (CASE STUDIES) 126
9.1 Permanently Stoppered Off Versus Adjustable Lines 126
9.2 Wear at Trumpet Welds Internal and External Turrets 130
9.3 Use of Bending Shoes 143
10 FURTHER MOORING CASE STUDIES 145
10.1 Wire Rope Systems 145
10.2 Unintended Line Disconnection 146
10.3 Excursion Limiting Weighted Chain Problems 151
10.4 Line Run Outs and Quick Releases 155
10.5 Windlass Failures 158
11 SPARS AND OFFLOADING BUOYS (CASE STUDIES) 160
11.1 Brent Spar Buoy 160
11.2 Floating Loading Platform (FLP) 163
12 TURRET MECHANICAL IMPLICATIONS FOR MOORING INTEGRITY 165
12.1 Introduction to Turrets and Failure Modes 165
2
A4163-01
12.2 Implications of Mechanical Repairs 168
13 GENERAL TRENDS AND STATISTICS 169
13.1 Questionnaire Process 169
13.2 Summary Statistics for Unit Type and Geographical Area 171
13.3 HSE UK Sector and Norwegian Statistics 174
14 CONNECTORS AND TERMINATIONS 177
14.1 Background 177
14.2 What Type of Connectors Can be Considered for Long Term Mooring (LTM) 177
14.3 Terminations General 182
14.4 Connector/Termination Design Flow Chart 186
14.5 Detailed Design Guidance 189
14.6 Proof Load Testing and Its Impact on Fatigue 193
15 OUT OF PLANE BENDING CHAIN AND ROPES (FIBRE + WIRE) 197
15.1 Tension Bending at a Wildcat and its Effect on Fatigue 197
15.2 Tension Bending at Chainhawse 205
15.3 Tension Bending In Wire Rope 215
15.4 General Implications of Tension Bending Fatigue for the FPS Industry 219
15.5 Recommendations 222
16 FRACTURE MECHANICS AND CRITICAL CRACK SIZE 223
16.1 Required Data 223
16.2 Fracture Mechanics and Chains State of the Art Summary 224
16.3 Fracture Mechanics Critical Crack Size Implications 225
17 LINE STATUS MONITORING AND FAILURE DETECTION 226
17.1 Instrumentation Status - Survey Results 226
17.2 Existing Failure Detection Systems 227
17.3 Future Failure Detection Systems 232
18 INSPECTION, REPAIR & MAINTENANCE (IRM) 238
18.1 In Air-Inspection 238
18.2 Where to Inspect on a Mooring Line 239
18.3 In-Water Inspection 242
18.4 Marine Growth Removal 246
18.5 Manufacturing Tolerances and the Inspection Implications 247
18.6 Wildcat/Gypsywheel Inspection 247
18.7 Inspection Frequency Code Requirements 255
18.8 Outline Method To Break Test Worn Mooring Components 257
19 SPARING OPTIONS 261
19.1 Contingency Planning - Spares and Procedures 261
20 THE IMPORTANCE OF A COMPREHENSIVE MOORING DESIGN
SPECIFICATION 264
20.1 Installation Parameters 265
21 KEY CONCLUSIONS & FUTURE WORK RECOMMENDATIONS 267
21.1 Overview 267
21.2 Key Conclusions 268
21.3 Recommendations for Further Study 270
22 REFERENCES AND BIBLIOGRAPHY 272
23 APPENDIX A - SUMMARY OF PAST RELEVANT JIPS 278
24 APPENDIX B MOORING INTEGRITY QUESTIONNAIRE (EXCEL) 279
25 APPENDIX C 2005 OTC JIP PAPER 280
26 APPENDIX D HSE SAFETY NOTICE 3.2005 FLOATING PRODUCTION
AND OFFLOADING (FPSO) MOORING INSPECTION 281
3
A4163-01
LIST OF TABLES
Table 1-1 - North Sea Mooring Line Failure Data, 1980 to 2001 [Ref. 1] ................................... 12
Table 1-2 - Indicative Single Mooring Line Failure Costs ........................................................... 13
Table 3-1 Summary of Chain Design Parameters (modified from Vicinay Chain Catalogue).. 41
Table 3-2 Comparison of Manufacturing Parameters ................................................................ 41
Table 3-3 Chain Geometry Implications for Inspection and Maintenance ................................ 42
Table 3-4 - Summary of Ship or Marine Grade Chains [Ref. 13]................................................. 45
Table 3-5 Example of Indicative Surface Hardness Values for Various Chain Grades (courtesy
of Vicinay)............................................................................................................................. 46
Table 3-6 Illustration of Indicative Wire Rope Material Properties [Ref. 2] ............................. 49
Table 3-7 - Comparison of the Advantages of Spiral and Six Strand Wire (courtesy of Bridon) 50
Table 3-8 - Comparison of the Cons of Spiral and Six Strand Rope ............................................ 50
Table 3-9 - Wire Rope Recommendations for Varying Field Lives (courtesy of Bridon) ........... 50
Table 3-10 - Stipulated MBL and Proof Load Values for Various Sizes and Grades of Chain
(courtesy of Vicinay)............................................................................................................. 62
Table 5-1 - Line Failure Cost Estimate, 50,00bpd North Sea FPSO ............................................ 83
Table 5-2 - Line failure Cost Estimate, 250,000bpd West African FPSO .................................... 84
Table 7-1 - Example of Specified Corrosion and Wear Allowances from One Classification
Society ................................................................................................................................. 102
Table 12-1 - Summary of the Pros and Cons of Sliding and Roller Bearings [Ref. 48] ............. 167
............................................................................................................................................. 169
Table 13-1 - Example of the First Page of the Questionnaire see appendix B for a Full Listing
Table 13-2 - UK Sector of the North Sea Data [Ref. 49].......................................................... 174
Table 13-3 - UK Sector of the North Sea Data [Ref. 49]........................................................... 174
Table 13-4 Number of Anchor Incidents in the Period of 1990-2003 in the Norwegian Sector
[Ref. 50] .............................................................................................................................. 174
Table 15-1 Comparison between Chain Tension-Bending Fatigue Parameters Note that values
in italics are derived from BOMEL measured stress factor. ............................................... 203
Table 15-2 : Wire Rope Fatigue Reduction Due to Tension Bending [Ref. 31]......................... 216
Table 15-3 - S-N Parameters for Mooring Chain Fatigue.......................................................... 220
LIST OF FIGURES
Figure 1-1 - Red Arrows Indicate Key Areas subject to Degradation on a .................................. 14
Figure 2-1 - JIP Organisation ........................................................................................................ 19
Figure 2-2 - CTR Breakdown........................................................................................................ 19
Figure 2-3 Participants at the Steering Committee meeting in Paris ......................................... 21
Figure 3-1 Typical Turret Moored FPSO................................................................................... 22
Figure 3-2 Shallow and Steep Mooring Line Angle Illustration................................................ 23
Figure 3-3 Line Heading Illustration.......................................................................................... 23
Figure 3-4 Definition of Windward and Leeward Lines + Environmental Offset..................... 24
Figure 3-5 Offset Position and Tension Effect........................................................................... 25
Figure 3-6 Illustration of Load Excursion Curve [Ref. 2].......................................................... 25
Figure 3-7 - Typical Spread Moored Unit, Girassol FPSO offshore West Africa (courtesy of Stolt
Offshore) ............................................................................................................................... 26
Figure 3-8 Illustration of Catenary System................................................................................ 28
Figure 3-9 - Typical Spread Moored Catenary System (Courtesy of Vryhof) ............................. 28
Figure 3-10 Illustration of Taut-Leg system .............................................................................. 29
Figure 3-11 - Typical Spread Moored Taut-Leg System (Vryhof) ............................................... 29
4
A4163-01
Figure 3-12 Illustration of Surge, Sway, Heave, Roll, Pitch and Yaw ...................................... 30
Figure 3-17 - Illustration of Deepwater Breaking Wave Types (Plunging Break on the Left and
Figure 3-18 - Illustration of the Damage Caused to Schiehallions Bow by an Unusually Steep
Figure 3-19 - Model Illustration of the Effect of a Breaking Wave on a FPSO (Courtesy of APL
Figure 3-20 - Isambard Kingdom Brunel in front of Studded Chain for the Great Eastern steam
Figure 3-24 Indication of the Manufacturing Tolerances of Studless Links (courtesy of
Figure 3-28 - Accurate Drag Anchor Placement by Crane in Good Weather Conditions (courtesy
Figure 3-30 Anchor Pile + Chain Tail Deployed by a Twin Crane Construction Vessel (courtesy
Figure 3-34 - Example of Terminology during a Tensile Test (courtesy of Ashby & Jones, [Ref.
Figure 4-7 - Helicopter Rescue from the Free Drifting North Carr Lightship after Mooring
Figure 4-8 - Illustration of the North Carr Link Failure Relative to a 1999 North Sea FPSO Link
Figure 4-9 - Dutch Lightship Number 11 whose Mooring Failed in a Force 10 Gale in October
Figure 4-10 - Fifth Generation Deepwater Nautilus Broke free of all her Moorings during
Figure 4-11 - Petrojarl 1 which experienced two broken lines at the same time when hit by a
Figure 3-13 Example of Optimising the Stiffness of the Load offset Curve ............................ 32
Figure 3-14 - Illustration of Long Crested (Unidirectional) Seas ................................................. 35
Figure 3-15 - Illustration of Short Crested (Confused) Seas....................................................... 35
Figure 3-16 - Example of a Wave Breaking on a Column of a Semi-Submersible ...................... 36
Spilling Breaking on the Right) ............................................................................................ 37
Wave (courtesy of BP) .......................................................................................................... 37
website) ................................................................................................................................. 38
ship, circa 1858 ..................................................................................................................... 39
Figure 3-21 Comparison of the Geometry of Modern Studded and Studless Chain.................. 40
Figure 3-22 - SPATE Contour Map of a 76mm Loose Stud Chain Link [Ref. 8]..................... 43
Figure 3-23 - Example of the Arrangement of an Asymmetric Stud ............................................ 44
Vicinay) ................................................................................................................................. 47
Figure 3-25 Studlink Manufacturing Tolerances (courtesy of Vicinay) .................................... 48
Figure 3-26 - Illustration of the Make Up of Different Wire Rope Types (courtesy of Bridon) .. 49
Figure 3-27 Chronology of Deep Star Funded Synthetic Mooring Studies ............................ 51
of Stolt Offshore) .................................................................................................................. 52
Figure 3-29 Installation and Normal (Vertical) Load Position (courtesy of Vryhof ) ................. 53
of Stolt Offshore) .................................................................................................................. 53
Figure 3-31 Suction Anchor Deployment (courtesy of Stolt Offshore)....................................... 54
Figure 3-32 - Example of a Tensile Test on a Steel Sample cut out from a Chain Link ............. 58
Figure 3-33 - Example of a Chain Sectioned for Material Testing .............................................. 59
17]) ........................................................................................................................................ 59
Figure 3-35 - Stress Strain Curves for R3, R4 and R5 Chain Steel (Data courtesy of Vicinay).. 61
Figure 3-36 Approximation of the Stress Distribution in a Typical Chain Link ....................... 63
Figure 3-37 - Illustration of a Finite Element Representation of a Chain Link ............................ 64
Figure 3-38 Finite Element Representation of a Shackle Body ................................................. 64
Figure 4-1 Illustration of some of the Main Factors which Influence Mooring Integrity.......... 65
Figure 4-2 - North Sea Pioneer on the Argyll Field ................................................................. 66
Figure 4-3 Fulmar SALM after Breakaway (courtesy of BBC film clip) .................................. 68
Figure 4-4 Schematic of the layout of the Fulmar SALM ......................................................... 69
Figure 4-5 - Extract from On this Day BBC Website ............................................................... 70
Figure 4-6 Illustration of a Typical Lightship Weathervaning Mooring ................................... 71
Failure [Ref. 20] .................................................................................................................... 72
Failure (fatigue cracking followed by ductile rip out) .......................................................... 73
1991 which also broke a number of semi-sub moorings see Section 4.2........................... 73
Hurricane Ivan....................................................................................................................... 75
steep wave ............................................................................................................................. 76
Figure 5-1 - Summary of a Single Line Failure Scenario ............................................................. 78
5
A4163-01
Figure 5-2 Illustration of Riser Stretch After Loss of Position Following Mooring Line
Failure.................................................................................................................................... 79
Figure 6-2 - Illustration of the Weight and Handling Issues Associated with Mooring
Figure 6-5 Illustration of a Hockle in Spiral Strand Wire during Recovery of a FPSO
Figure 6-9 Pre-Stretching Polyester lines During Installation to Minimise the Requirement for
Figure 6-10 - Illustration of the Potential Difficulty in offshore alignment of pins on large
Figure 6-11 - Sledge used to Protect H Connector during Deployment over the Stern Roller
Figure 7-3 Illustration of the Extent of General Corrosion on a Recovered Floating Production
Figure 7-6 Arrow shows the Apparent Grinding Action on the Inner Face of One of the Links
Figure 7-7 Example of the Damage Caused to a Hanging Shackle Pin on a FPSO Mooring Line
Figure 7-12 -Test Rig Set Up for Break Testing of Mooring Components (Studless Chain in the
Figure 7-19 - Measured Wear Rates of U3 and U4 Chain at 8,170lbs (300 tonnes equivalent)
Figure 9-1 - Turret Design in which Chain Lengths can be Adjusted (courtesy of Chevron-
Figure 9-2 Generic Turret Design in which the Chains are Stoppered off at the Turret Base
Figure 5-3 - Potential Multiple Line Failure Scenario .................................................................. 80
Figure 5-4 - Example of how Mooring Integrity Philosophy can affect Production .................... 81
Figure 6-1 - Spooling Fibre rope onto a Powered Reel from Standard Containers ...................... 85
Components (Courtesy of Stolt Offshore) ............................................................................ 86
Figure 6-3 - Red Arrows Show Examples of Mooring Dog-Legs............................................. 87
Figure 6-4 - Illustration of Twist on a FPSO Mooring Line during Recovery ............................ 90
Mooring System .................................................................................................................... 90
Figure 6-6 - Example of Damage to the Bend Stiffener on an Open Socket ................................ 91
Figure 6-7 Illustration of Spiral Strand Wire Kinking during Installation................................. 91
Figure 6-8 - Mid Line Buoy Swivel Connection Link (courtesy of MoorLink AB). ................... 92
Future Line Length Adjustments [Ref. 27] ........................................................................... 96
Diameter Rope [Ref. 26] ....................................................................................................... 97
(Courtesy I. Williams)........................................................................................................... 97
Figure 7-1 The Balmoral Benchmark FPV which has been continuously on station since 1986
(Courtesy of CNR) ................................................................................................................ 99
Figure 7-2 Plan View of Mooring Incidents at Balmoral........................................................ 100
Unit Mooring Line after 16 years service ........................................................................... 103
Figure 7-4 Illustration of the Extent of Corrosion Pitting .......................................................... 104
Figure 7-5 Example of the Damage Caused to the Crown of the Links .................................. 105
............................................................................................................................................. 105
............................................................................................................................................. 106
Figure 7-8 Finite Element Stress Contour Plot (compare red areas with Figure 7-6) [Ref. 8] .. 106
Figure 7-9 - Example of Thrash Zone Wear .............................................................................. 107
Figure 7-10 - Illustration of the Extent of Pitting Corrosion...................................................... 108
Figure 7-11 - Example of Wear and Pitting Corrosion on the Shackle Pin ............................... 109
instance) .............................................................................................................................. 111
Figure 7-13 Illustration of Biologically Induced Pitting Corrosion in a Ballast Tank............. 112
Figure 7-14 - Crack Growth per Cycle versus Stress Intensity Range [Ref. 2] .......................... 113
Figure 7-15 Illustration of Excessive Chain Wear on a CALM Buoy [Ref. 34]...................... 115
Figure 7-16 Typical Temperature and Salinity Profile in the Tropical Oceans ....................... 116
Figure 7-17 Indicative Oxygen Concentration versus Water Depth (courtesy of BP)............. 116
Figure 7-18 Gulf of Mexico Snap Shot of Bottom Oxygen Concentration ............................. 117
[Ref. 34] .............................................................................................................................. 119
Figure 8-1 Illustration of Line Tension Variations during a Payout/Pull-In Test .................... 123
Texaco)................................................................................................................................ 127
(courtesy of Bluewater) ....................................................................................................... 127
Figure 9-3 - Spread Moored FPSO Single Axis Chain Stopper (courtesy of SBM)................... 128
6
A4163-01
Figure 9-4 - External Cantilever Turret which experienced Chain wear at the Trumpet Welds
which was halted by use of UMPHE (courtesy of Shell).................................................... 130
Figure 9-5 - Example of the Level of Inspection Detail which can be achieved using a Typical
Figure 9-6- Test Tank Mock-Up of Micro-ROV inspection of Chain Emerging from Turret
Figure 9-7 - Micro-ROV Photograph of Chain Wear Notches where Chain Emerges at the
Figure 9-9 - Artificially Introduced Notch on to Spare Chain Links, note also Red Circular
Figure 9-10 - Example of Stretched Chain during Break Testing, the Blue Mark Shows the
Figure 9-11 - Example of a Special Cobalt Chromium Anti-Wear Coating (courtesy of I.
Figure 9-12 - Photograph of a Recovered Link Showing a Wear Notch (courtesy of I. Williams)
Figure 9-13 - An Example of the Chain Damage noted after the Notched Chains had been
Figure 9-14 - Turret Arrangement where the Chain Stopper (in red) is Behind the Rotation Point
Figure 9-15 Illustration of Potential Wear at Metal to Metal Contact (courtesy of I. Williams)
Figure 9-16 - Fairlead Chain Stopper where the Chain Stopper is in Front of the Rotation Point
Figure 9-17 - As Installed Photo Graph of the Design Shown in Figure 9-16 (courtesy of
Figure 9-18 Typical CALM Buoy Chain Stopper (courtesy of The Professional Divers
Figure 10-1 Examples of the Subjectivity Associated with Assessing IWRC Rope Conditions
Figure 10-3 - Photograph of Disconnected Socket on the Sea-Bed (courtesy of BP/Stolt
Figure 10-4 - Note End Plate also seems to be Falling Off on the Right Hand Side (courtesy of
Figure 10-14 - Weighted Chain Option Utilising Parallel Chain Sections (courtesy of
Workclass ROV (courtesy of I.Williams) ........................................................................... 131
Trumpet (courtesy of I. Williams) ................................................................................... 132
Trumpet Bell Mouth (courtesy of I. Williams) ................................................................... 132
Figure 9-8 - Indication of the Extent of the Wear ...................................................................... 133
Infrared Target (courtesy of I. Williams) ............................................................................ 134
Location of a Typical Notch (courtesy of I. Williams) ....................................................... 134
Williams) ............................................................................................................................. 135
............................................................................................................................................. 136
recovered back to Shore (courtesy of I. Williams).............................................................. 136
(2 black concentric circles) ................................................................................................. 137
............................................................................................................................................. 138
(used on some Spread Moored FPSOs) (courtesy of Maritime Pusnes) ............................. 138
Maritime Pusnes)................................................................................................................. 139
Handbook [Ref. 38]).......................................................................................................... 140
Figure 9-19 - Amoco CALM Buoy- Note Inclusion of Rubber Casting (courtesy of [Ref. 38]) 140
Figure 9-20 - Comparison of Alternative Fairlead Arrangements (courtesy of Bardex) ........... 142
Figure 9-21 Example of a Wire Rope Bending Shoe (courtesy of API RP25K) ..................... 143
Figure 9-22 - Example of a Chain Bending Shoe Design [Ref. 39]............................................ 143
Figure 9-23 - Bending Shoe Design which includes an Angle Sensor [Ref. 40] ........................ 144
[Ref. 43] .............................................................................................................................. 145
Figure 10-2 - Illustration of the Mooring Layout and Connections ............................................ 146
Offshore) ............................................................................................................................. 147
BP/Stolt Offshore)............................................................................................................... 147
Figure 10-5 - End Connection Detail ......................................................................................... 148
Figure 10-6 - Illustration of Socket Minus End Plate ................................................................. 148
Figure 10-7 - Repair Utilised Bigger Bolts and Allowed the Socket Pin to Rotate .................... 149
Figure 10-8 - Example of Retrofitted Anodes to Control Corrosion Rate .................................. 150
Figure 10-9 - Example of Disconnected Anodes after approximately 12 months of Service..... 150
Figure 10-10 - Example of Detached Clump Weight on the Sea-Bed ........................................ 151
Figure 10-11 - Example of Recovered Clump Weights .............................................................. 151
Figure 10-12 Illustration of Where the Damage Occurred on the Mooring Catenary ............. 152
Figure 10-13 - Example of a Parallel Chain Excursion Limiter (courtesy of I. Williams) ......... 152
N.Groves) ............................................................................................................................ 153
7
A4163-01
Figure 10-15 - Red Arrow Illustrates the Local Wear can take place when utilising Parallel
Chain (courtesy of N. Groves) ............................................................................................ 153
Figure 10-16 - Example of Mid-Line Buoy Failures on a European FPSO................................ 154
Figure 10-17 - Gripper chock showing chain damage ................................................................ 156
Figure 10-18 - Upper Gypsy Wheel Arrangement before Failure .............................................. 156
Figure 10-19 - Gypsy wheel structure after failure, i.e. Gypsy Wheel No Longer Present ........ 156
Figure 10-20 - Illustration of a New Design of Kenter Shackle intended to have improved
Fatigue Performance ........................................................................................................... 157
Figure 10-21 - Example of Windlass Crack (Red Arrow) due to Stress Raiser caused by Sharp
Corner (courtesy of BP) ...................................................................................................... 158
Figure 11-1 - General Arrangement of the Brent Spar Mooring System (courtesy of Shell) ..... 160
Figure 11-2 - Brent Spar Fairlead Chain Stopper in the Hull (courtesy of Shell)....................... 161
Figure 11-3 - Close Up of the Stopper (courtesy of Shell) ......................................................... 161
Figure 11-4 - Indentation from where the chain bore down on the Stopper (courtesy of Shell) 162
Figure 11-5 Red Arrow Illustrates wear on the chain, where it sat on the stopper (courtesy of
Shell) ................................................................................................................................... 162
Figure 11-6 - Brent Spar Wire Sample Y1 prior to cleaning [Ref. 41]....................................... 163
Figure 11-7 FLP Mooring General Arrangement (courtesy of Shell)...................................... 163
Figure 11-8 - Example of Short Trumpets on a Long Term Moored Floating Loading Platform
(courtesy of Shell) ............................................................................................................... 164
Figure 13-1 - Comparison of Mooring Line Inspection Periods for Different FPS Categories. 173
Figure 13-2 Historical Failure Rates for Different Types of Units ......................................... 176
Figure 14-1 - Special Joining Shackle (courtesy of Vicinay Catalogue) .................................... 179
Figure 14-2 - H Shackle Pin Configuration (courtesy of I. Williams) .................................... 180
Figure 14-3 Illustration of Subsea Connectors which have been used on Pre-Installed Mooring
Lines .................................................................................................................................... 181
Figure 14-4 - Example of a Special Joining Plate - Note Electrical Isolating Bush ................... 181
Figure 14-5 Example of the Make Up of a Typical Closed Spelter Socket (courtesy of Bridon)
............................................................................................................................................. 182
Figure 14-6 - Example of an Open Socket .................................................................................. 183
Figure 14-7 - Example of a Closed Socket ................................................................................. 183
Figure 14-8 - Connector or Termination Design Flow Diagram - Initial Phase ........................ 187
Figure 14-9 - Connector (Termination) Detailed Design Flow Chart....................................... 188
Figure 14-10 Illustration of a Purpose Designed connector allowing limited compliance in Two
Planes .................................................................................................................................. 190
Figure 14-11 - Example of a Dynamic Analysis to Estimate the Angle for the V Slot Size on
the H Shackle ................................................................................................................... 191
Figure 14-12 - Example of Material with a Non Clearly Defined Yield Point .......................... 194
Figure 15-1 Broken Link from Fairlead .....................................................................................197
Figure 15-2 Mechanical Damage on Fairlead Link................................................................... 197
Figure 15-3 - Support of a Link in a Wheel Fairlead ................................................................. 198
Figure 15-4 - Photograph of Test Link Showing Bearing Plates [Ref. 10]................................. 199
Figure 15-5 - General View of Tension Bending Test Rig (protective screens removed for
clarity) [Ref. 10] ................................................................................................................. 199
Figure 15-6 - Broken Hardened Plates at the end of the First Test [Ref. 10] ............................. 200
Figure 15-7 - Twisted Link Due to Mis-aligned Butt Weld [Ref. 10] ........................................ 201
Figure 15-8 - Simple Out of Flatness Twist Measurement Jig [Ref. 10] .................................... 201
Figure 15-9 - Illustration of Failed Link Due to Tension Bending [Ref. 10].............................. 204
Figure 15-10 - Girassol Offloading Buoy [Ref. 55]................................................................... 205
Figure 15-11 - Girassol Offloading Buoy Failure in Chain Link 5 [Ref. 55] ......................... 206
Figure 15-12 - Girassol Offloading Buoy Failure in Polyester Rope [Ref. 55] ...................... 206
Figure 15-13 - Chainhawser Arrangement and Location of Critical Link [Ref. 55] ................. 207
8
A4163-01
Figure 15-14 - Out of Plane Bending Mechanism (See Section 25 [Ref. 56]......................... 208
Figure 15-20 - Tension Bending at Wheel Fairlead (Bearing Load Eccentricity) and Tension
Figure 17-3 - Sonar Display Screen Showing 12 Mooring Lines and 2 Risers Close to the Centre
Figure 17-7 - Indication of the Data Available from Instrumented Mooring Lines (courtesy of
Figure 17-8 - Illustration of a New Sonar System due to be Installed in the North Seas (courtesy
Figure 18-1 - Red Arrows and Black Line Indicate Key Areas subject to Degradation on a
Figure 18-3 - Typical Turret Cross Section Illustrating that the key Mooring Components are
Figure 18-4 - Chain Stopper View Prior to Chain Installation with Pull in Rigging Present
Figure 18-5 - Illustration of ROV Deployed Optical Calliper Measurement System (courtesy
Figure 18-10 - Red Zones Highlight the Importance of Checking all Relevant Structural
Figure 18-11 - Example of a Parted Lubrication Line Feeding a Submerged Wildcat or Gypsy
Figure 18-16 - Partially Buried Shackle Illustrates the Difficulties in checking locking pins
Figure 18-17 - Example of the Wheel Tappers Approach Used for Detecting Cracks on Railway
Figure 18-18 - Example of Anchor Handling and Heading Control Tugs during a Mooring Line
Figure 18-19 - Use of Divers from a RIB to open up the Chain Stopper during a FPSO Mooring
Figure 15-15 - Schematic of SBM Test Rig [Ref. 55] ............................................................... 209
Figure 15-16 - Photograph of SBMs Test Rig [Ref. 55]............................................................ 210
Figure 15-17 - Typical FPSO Chain Stopper Arrangement ........................................................ 211
Figure 15-18 Illustration of Wire Rope Failure Modes (courtesy of Bridon) .......................... 217
Figure 15-19 - The 1.0MN Wire Rope Bending-Tension Fatigue Test Machine ....................... 218
Bending from Interlink Friction (Torque at Contact).......................................................... 219
Figure 15-21 - Comparison between Various Mooring Chain S-N Curves ............................... 221
Figure 17-1 - Sonar Fish for Deployment through Turret (courtesy Chevron Texaco) ............. 227
Figure 17-2 Sonar Fish Deployment Method (courtesy Chevron Texaco) .............................. 227
(courtesty Chevron Texaco) ................................................................................................ 228
Figure 17-4 - Simple Pre-Installed Inclinometer with + or 1 Degree Accuracy ...................... 229
Figure 17-5 - Illustration of a Football Sized ROV (Courtesy of I. Williams) ....................... 229
Figure 17-6 - Instrumented Load Pin Shackle Link (courtesy of BMT/SMS)......................... 230
BMT/SMS).......................................................................................................................... 231
of I. Williams) ..................................................................................................................... 233
Figure 17-9 - Close Up of the Proposed Sonar Head (courtesy of Ian Williams)...................... 233
Figure 17-10 - Response Learning Without Line Tension Input ................................................ 234
Figure 17-11 - Illustration of Riser Monitoring Instrumentation (courtesy of 2H) .................... 236
Mooring System (leeward likely to have worst wear) ........................................................ 239
Figure 18-2 - Example of a Weight Discontinuity which may Result In Enhanced Wear ......... 240
Submerged........................................................................................................................... 241
(compare to Figure 18-3)..................................................................................................... 242
of Welaptega Marine Ltd) ................................................................................................... 244
Figure 18-6 Illustration of Heavy Marine Growth on Long Term Deployed Chain............... 246
Figure 18-7 - In-Situ Inspection of a Wildcat Pocket by Abseillers .......................................... 248
Figure 18-8 - Close Up Of Fairlead Pocket Note Slight Lip on the Right ............................... 248
Figure 18-9 - Example of Chain Wear From Sitting in a Wildcat Pocket .................................. 249
Connections (Courtesy of CNR) ......................................................................................... 249
Wheel (Courtesy of CNR)................................................................................................... 250
Figure 18-12 - Example of a Non Flat Link................................................................................ 251
Figure 18-13 - Buchan FPS Wire Rope NDT Inspection Head ................................................. 252
Figure 18-14 - Proposed Wire Rope Inspection Toll Delpoyed from a ROV............................ 253
Figure 18-15 - Example of a Difficult Area to Inspect .............................................................. 256
(courtesy of ENI)................................................................................................................. 256
Carriages and Locomotives ................................................................................................. 258
Repair Operation (courtesy of I. Williams)........................................................................ 259
Line Repair (coutesy of I. Williams)................................................................................... 260
9
A4163-01
Figure 19-1 - Example of a Plate Shackle which may be useful for a Temporary Repair (courtesy
of Balmoral Marine)............................................................................................................ 262
Figure 19-2 - Temporary Mooring Line Winch Deck on a Gulf of Mexico Spar....................... 263
10
A4163-01
1 EXECUTIVE SUMMARY
1.1 Overview of the Report
The main objective of this report is to improve the integrity of the mooring systems on
Floating Production Systems (FPSs). It is intended to be read and understood by non
mooring specialists such as FPS Operational staff - so that the people who live and
work on FPSs will be better able to become more involved in the vital task of looking
after their own mooring systems. Meanwhile the included feedback on the actual
performance of mooring systems in the field should assist designers and manufacturers
to improve future mooring designs. Hence, the report attempts to identify gaps in the
existing knowledge of mooring behaviour and components to provide a road map for
future work. Appendix C includes a paper presented at the 2005 Offshore Technology
Conference (OTC) which represents a stand alone summary of the key points of the
JIP.
1.2 Introduction
Unlike trading ships, FPSs 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 comes their way. Hence, depending on location, at times
their mooring systems need to withstand high storm loadings. Typically, during design
for harsh environments, mooring systems 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 as has been seen in the past see Section 4.
Failure of two adjacent mooring lines mooring lines at the same time due to wave
shock loading has been seen and this could have serious implications if the risers are
pressurised at the time.
This JIP is concerned with assessing how mooring systems have performed in the field
to identify the level of degradation which has taken place. Hence, the JIP has looked at
FPSOs, Semi-submersible production units and Spars through out the world. From the
survey it has become apparent that certain, potentially significant, 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.
11
A4163-01
1.3 Indicative Mooring Statistics
At the beginning of the project it was hoped that it would be possible to gain data on
the mooring performance on most of the FPSs (turret and spread moored FPSOs,
production semis and Spars) in the world. In practice the best data which has been
obtained is for North Sea Units, partly due to local contacts and also the rigorous
reporting regime in this area. In the absence of more comprehensive information, it
thus seems prudent to consider official statistics for this region to be the best available
indicator of the likelihood of mooring failure. Exactly how these statistics can be
related to milder environments is difficult to quantify based on the presently available
data set.
Table 1-1 summarises failure statistics for North Sea operations for different floating
units covering the period 1980 to 2001. It is clear from these statistics that the
probability of line failure per operating year is relatively high.
Type of Unit Number of Operating
Years per Failure
Drilling Semi-submersible 4.7
Production Semi-submersible 9.0
FPSO 8.8
Table 1-1 - North Sea Mooring Line Failure Data, 1980 to 2001 [Ref. 1]
Given the safety critical nature of mooring lines and the likelihood of failure one might
imagine that they would be heavily instrumented with automatic alarms which would
go off in case of line failure. 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 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,
- 50% of units cannot adjust line lengths.
12
A4163-01
1.4 The Cost of Mooring Line Failure
If a multiple mooring line failure should occur in storm conditions the potential cost
and the implications for the whole FPS industry could be extremely high, depending on
circumstances. Even a single mooring line failure would be costly as is illustrated in
Table 1-2 for two different case studies, further details can be found in Section 5.4.
Description Approx. Cost of
Single Line Failure
50,000bpd N. Sea FPSO 2M
250,000bpd W. African FPSO 10.5M
Table 1-2 - Indicative Single Mooring Line Failure Costs
1.5 Key Findings from the Survey
During the course of the project a few common themes emerged which are outlined
below:
Wear where the Lines Connect to the Surface Platform
Achieving material compatibility at the key turret interface is vital see Section 0.
Whether the trumpet pivot point should be in board or outboard of the chain stopper
needs further consideration for new designs. In addition, whether rotation should be
permitted in two planes, rather the one which is typically the case at present also
requires addressing based on further in field experience. This may have particular
implications for spread moored FPSOs. Access for inspection of these areas also needs
to be improved and this should be specified in the mooring design criteria see Section
20.
Wear/Corrosion Allowance for Long-Term Moored Units
On two North Sea FPSs chain wear and corrosion has been found to be significantly
higher than what is specified by most mooring design codes. More in field inspection
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.
Excursion Limiting Weighted Chain Designs
A number of excursion limiting weighted chain systems have experienced problems
see Section 10.3. Great care is needed in the design of such systems; particularly if
they are due to operate in adverse environmental conditions. Parallel chains seem to
have worked well, as opposed to clump weights (lump masses) or hung off chain tails.
Clean catenaries, i.e. without buoys or clump weights seem to work best, although
water depth, riser offset limits and environmental conditions may mean that this is not
always possible.
13
A4163-01
Local Design of Connectors
Connectors are vital components on any mooring system and they need to be very
carefully designed if they are to prove reliable over a long field life. Certain mooring
problems have been due to the local design of the connectors. Section 14.4 includes a
summary of key items which should be considered during detailed connector design.
There is an emerging need for the development of a fatigue resistant connector suitable
for use with mooring chains during repair/overhaul operations.
Friction Induced Bending
Friction induced bending fatigue appears to be a significant issue which has been
somewhat neglected and warrants further investigation. This was less of an issue for
catenary systems in moderate water depths. Deep water taut moored units seem to be
potentially particularly susceptible see Section 15.2.
1.6 Key Areas to Check on a Mooring System
Based on the survey results,
Figure 1-1 illustrates the key areas which should be inspected on a mooring line. The
FPS has been displaced by environmental forces, thus illustrating both windward and
leeward mooring lines.
Figure 1-1 - Red Arrows Indicate Key Areas subject to Degradation on a
Mooring System (leeward likely to have worst wear)
From a number of units it has become clear that the less loaded leeward lines can be
subject to greater degradation than the windward lines. This seems to be due to greater
relative rotation on leeward lines since the line is typically under lower tension.
14
A4163-01
1.7 What are the Best Ways to Detect Line Failures?
It is vital to detect line failures promptly or else there is a danger of a unit entering a
storm while still producing and thus being at an increased danger of loosing another
mooring line.
Detecting a line failure in the mud is difficult since the catenary shape, depending on
sea-bed conditions, may not change that much. Section 17 summarises the key
detection techniques available at present. It is clear that in-field trials are required to
identify what systems prove to be reliable over the long-term. Hence, this is an on
going issue which requires monitoring, assessment and publicising of the key findings.
1.8 Inspection Technologies
Inspecting moorings lines in situ is desirable due to the danger of damage during line
recovery to the surface and also during re-installation. There is also a significant cost
involved in mobilising intervention vessels to recover lines to the surface and then re-
install them.
In-water line inspection is difficult, particularly with respect to identifying possible
cracks. Despite this it has become clear that many possible problems can be identified
early on, using tweaks to existing technology. This has been successful as long as
suitably experienced people are involved in planning the inspection process and
examining the results.
Section 18 summarises the present available inspection technologies and includes a
prioritised list of possible future improvements.
1.9 Key Conclusions and Recommendations
The survey of past and presently operating FPS units has shown that serious incidents
have occurred in the past including loss of station. The survey has also shown that
even for more up to date designs, deterioration of certain areas of the moorings system
may be more rapid than expected. As well as the detail issues there is a more general
issue that requires addressing, namely the manner in which mooring integrity is
managed and audited on an on-going basis.
Since moorings are category 1 safety critical systems, if the deterioration is not detected
early and monitored/rectified the consequences could be severe. Hence, it is vital that
whoever offshore is responsible for the day to day operation and inspection of FPS
moorings should have a strong marine background, such as a Deck Officer or Marine
Engineer. Such personnel have a suitable mindset in that they really understand the
importance of moorings and their likelihood to deteriorate significantly over time. It is
important that these personnel should be provided with sufficient resources so that they
can be pro-active with regard to inspection and any possible repairs which may be
required.
15
A4163-01
Semi-submersible units have accumulated hundreds of years of mooring experience for
varied world wide locations. A key point to learn from such units is that chains, wire
ropes, gypsy wheels, stoppers and connectors have finite lives and do wear out.
Although drilling rigs deploy and recover lines fairly regularly, which can cause
damage, the wear seen on production semis is still significant see Section 7.1.
However, most large scale FPS with 20+ year design lives seem to have been built on
the expectation that the mooring lines will last for the life of the field and that safety
will not be compromised towards the tail end of the field life, when production rates
have dropped. If production rates have dropped there is less money available for
mooring line repairs. Hence, assessments should be undertaken during the field life to
assess whether line change outs may be required in the future and if so contingency
money should be allowed for to cover this later expenditure.
In general, moorings should be thought of as relatively vulnerable primary structural
members subject to constant dynamic motion. Expecting such systems to last for 20+
years without overhaul may prove to be optimistic. The commercial risks associated
with a line failure during the field life justify the selection of top quality equipment
from the outset. This equipment then needs to be regularly inspected and repaired as
required to ensure that it is still fit for purpose.
Availability of mooring line spares including connectors is extremely variable. Given
the several month lead-time associated with procuring new components, it is
recommended that each operator should identify short term remedial measures to repair
a line if it fails. This would involve identifying commonly available components which
can be obtained at short notice from marine equipment rental companies. Outline
procedures including the type of intervention vessel required should also be developed.
Mooring systems are not as simple as they first appear and they need careful
management through out their design lives. Thus a life cycle approach to mooring
design and operation is recommended. In this way designers can feedback their
inspection requirements to Operators and then learn from whatever is found during
inspection. Manufacturers should also be included in this feedback loop, since they
may be best placed to implement improvements to their products. Hence over time
mooring design and manufacturing should improve. At present designers and
manufacturers are not always involved with the in field behaviour mooring systems.
Therefore, they may not be aware of operational or inspection type issues. In general
there seems to be a need for periodic Mooring Audits to re-assess original design
parameters and review inspection records to assess whether the system is still fit for
purpose.
It is clear from this state of the art review that to continue to improve mooring integrity
a number of topics still require further investigation. A bullet point list is included in
Section 21.3.
16
A4163-01
2 INTRODUCTION AND SCOPE
2.1 The Need for a JIP
The number of Floating Production Systems (FPSs) operating in the world increased
substantially during the 1990s and there is now an ever-increasing body of FPS
operational experience. In 2001 Noble Denton was commissioned by the UK Offshore
Operators Association (UKOOA) to review available operational data from the British
sector. The key results to emerge from this study were as follows:
- There has been one FPSO line failure for every 5.4 operating years (this
figure has been updated during this study);
- Several cases occurred in which there was systematic damage to more
than one line;
- Particular problems have been experienced at connectors and interfaces;
- In no cases was the damage recognised immediately;
- Long-term failure rates remain uncertain.
The study concluded that the potential for multiple line failure is greater than is
commonly perceived, and this should be a major cause for concern. The main reasons
for this situation are:
- Available inspection and maintenance provisions can allow long periods
in which single or multiple defects can remain undetected;
- Most UK sector FPSOs can not detect if they have lost a mooring line;
- The risk of mooring line failure is often underestimated and the majority
of operators do not carry spares or have systems in place for dealing with
a line failure;
- Design codes and standards give little guidance on terminations,
connections, fair leads and stoppers which is where the majority of
failures has been seen;
- Similarly there is limited guidance on inspection, repair and maintenance.
17
A4163-01
2.2 Scope Development
The JIP scope was developed to extend the previous UKOOA study to include
international experience, and reassess the conclusions of the UK sector study in more
global terms. In addition, a follow up has been carried on the recommendations of the
UKOOA study to investigate the levels of exposure to duty holders, and developing
measures to reduce the associated risks.
Specifically the work has covered the following:
- Disseminate data gathered from international experience,
- Develop guidance for designing mooring connectors and interface
elements,
- Provide guidance on mooring line inspection,
- Summarise the pros and cons of line failure detection methods,
- Take a look to future deepwater and taut leg applications
- Investigate and report illustrative case studies
The JIP scope has been adjusted during the project to take into account results found to
date and also the difficulties experienced in obtaining international data.
2.3 JIP Objectives
The basic objectives of the JIP are to:
- Improve safety
- Help to safeguard reputation of FPSO/FPS industry
- Feedback operational and inspection to mooring designers
- Publicise the importance and potential vulnerability of mooring systems
This report is intended to be read and understood by non mooring specialists such as
FPS Operational staff. In this way the people who live and work on FPSs will be better
able to become involved in the vital task of looking after their own mooring systems.
18
A4163-01
DESIGN AND CONSTRUCTION ISSUES
CTR 2 : Transportation, Handling & Installation
Challenges
CTR 3 : Design of Connectors & Interfaces
INTERNATIONAL SURVEY OF
MOORING PROBLEMS
CTR 1 : Survey of International FPSO/ FPS Ex
CTR 4 : Consequences of Line Failure
DISSEMINATION
OF RESULTS
(CTR 10)
Lessons Learned
Detailed Report
Integrity Check List
2.4 Project Organization
The project organisation is illustrated in Figure 2-1.
Martin Brown
COMMITTEE
Project Manager
STEERING
Nigel Robinson
NDE Project Director
Consultants: I.D. Williams, R Stonor, R
Nataraja, D. Orr, R.V. Ahilan
ND Group Resources & Subcontractors
Figure 2-1 - JIP Organisation
The scope of work was broken down into Cost, Time, and Resource Modules [CTRs],
which were organized as follows:
:
p
:
:
Water Survey
:
DISSEMINATION
OF RESULTS
(CTR 10)
Bulletins/Steering
OTC paper
Detailed Report
DESIGN AND CONSTRUCTION ISSUES
CTR 2 : Transportation, Handling & Installation
Challenges
CTR 3
Design of Connectors & Interfaces
INTERNATIONAL SURVEY OF
MOORING PROBLEMS
CTR 1 : Survey of International FPSO/ FPS Experience
CTR 4 : Consequences of Line Failure
INTEGRITY MANAGEMENT
CTR 5
Status Monitoring and Failure Detection
CTR 6
Inspection, Repair & Maintenance, inc In
CTR 7
Sparing Options
Lessons Learned
Committee briefings
Figure 2-2 - CTR Breakdown
19
A4163-01
2.5 Project Participants/Sponsors
The following list details the organisations which have sponsored the JIP plus the
personnel nominated to the Steering Committee. It is worth noting that the Steering
Committee meetings provided an excellent mechanism to obtain and distribute data.
Thanks are given to all members of the committee and the Chairman for their
participation.
1. ABS, Rod Yam and Ernesto Valenzuela
2. Ansell Jones
3. Balmoral Group, Doug Marr
4. Bluewater, Simon Stauttener
5. BP, Richard Snell, Peter Gorf and Steve Barron
6. Bureau Veritas, Frank Legerstee and Michel Franois
7. Chevron Texaco, Matthew Brierley, Paul Devlin, and Jim Hughes
(corresponding member)
8. ENI (Agip), Les Harley and Bill Nicol
9. Hamanaka Chains, Yoshiyuki Kawabe
10. HSE, Martin Muncer and Max English
11. IMS/Craig Group, Alan Duncan and Mark Prentice
12. Lloyds Register, Douglas Kemp, Richard Bamford and Alwyn McLeary
13. MARIN, Henk van den Boom and Johan Wichers
14. Maersk Marine Contractors, Graham Kennedy and Vere MacKenzie
15. National Oilwell/Hydralift-BLM, Philippe Gadreau
16. Norsk Hydro, Tom Marthinsen
17. Offspring International, Nigel Grainger and Russell Glen
18. Petro Canada, Sherry Power and Scott OBrien
19. SBM, Philippe Jean (Chairman)
20. Statoil, Kjell Larsen
21. Vicinay Cadenas, Dave Nicol and Eduardo Lopez
22. Welaptega Marine, Tony Hall
Many people from various organisations helped out through out the JIP by providing
information. It is impossible to list them all, but their combined support has been
crucial in enabling a comprehensive picture to be pulled together. Particular thanks are,
however, given to Amerada Hess/Wood Group and Mr Ian Williams for making highly
relevant data readily available to the JIP. Thanks also to Diane for all her assistance
with the layout and editing of this document.
20
A4163-01
2.6 Steering Committee Meetings
The Steering Committee met four times during the course of the JIP in Monaco,
Aberdeen, Paris and Houston, all being well attended. The meetings in Monaco and
Paris were part of the FPSO Forum/JIP Week. The Aberdeen meeting was a standalone
meeting.
The final meeting in Houston was at the end of the 2005 Offshore Technology
Conference (OTC).
Figure 2-3 Participants at the Steering Committee meeting in Paris
21
A4163-01
3 MOORINGS OVERVIEW
3.1 Mooring Basics
3.1.1 Restoring Forces
To appreciate how to preserve the integrity of a mooring system it is helpful to have a
basic understanding of the different types of mooring systems and how they work. This
subject is covered in this chapter, which also includes a simple introduction to how
such systems can be analysed.
The primary purpose of a mooring system is to maintain a floating structure on station
within a specified tolerance, typically based on an offset limit determined from the
configuration of the risers. The mooring system provides a restoring force that acts
against the environmental forces which want to push the unit off station. In the
following diagrams the main components of mooring system restoring force are
explained.
The connection between the mooring system and the body of the vessel is where the
restoring force of the mooring system acts, see Figure 3-1. At this connection point
there are two force components present; horizontal and vertical. The horizontal
component of the mooring lines tension acts as a restoring force. The vertical
component acts as a vertical weight on the vessel. In deep water the vertical force can
be quite considerable. For some designs of FPS, with limited payload capacity, the
vertical mooring force can have significant design implications.
Figure 3-1 Typical Turret Moored FPSO
22
A4163-01
It is informative to understand the significance of the mooring line angle as it departs
the point of connection to the vessel. A low angle to the vertical will generate a low
restoring force, with significant vertical load on the vessel. If the angle here is large,
then the restoring force will be increased while the vertical load on the vessel will be
reduced. This relationship can be seen in Figure 3-2. The vessel needs to be able to
support the applied vertical loading.
Figure 3-2 Shallow and Steep Mooring Line Angle Illustration
The relationship outlined in Figure 3-2 is adequate for considering a 2 dimensional
scenario. The mooring of a vessel, however, is a 3 dimensional problem and to this end
it is necessary to consider the angle of the mooring line in the plane of the sea-surface.
With reference to Figure 3-3 it can be seen that the tensions in a mooring line are split
into two components; the restoring force that opposes the environmental loading, and
the lateral force, which may balanced by another mooring line.
Figure 3-3 Line Heading Illustration
23
A4163-01
3.1.2 Environmental Loading
When there is no external loading on the system the vessel will not move from its static
equilibrium position. When environmental loading does occur an imbalance in the
system will occur. To restore equilibrium the mooring system restoring force must
become equal to that of the environmental load. This is achieved through the vessel
offsetting from its original position. As this occurs the windward lines will pick up
tension and the leeward lines will shed tension. This is shown in Figure 3-4.
Figure 3-4 Definition of Windward and Leeward Lines + Environmental Offset
The vessel will offset until the windward lines have generated a restoring force that
balances the environmental loading. This means that the distance between the anchor
and fairlead will increase, and thus the tension at the fairlead will also increase. This is
shown in Figure 3-5.
24
A4163-01
The relationship between environmental load and vessel offset is often represented in a
Load Excursion curve, as shown in Figure 3-6. This figure illustrates the load
excursion characteristics of a 1,200m long, 76mm nominal diameter chain in 100m
water depth with a working or pretension tension of 100te. The plot emphasizes the
need to model the axial elasticity, even for chains, in order to get realistic results. Axial
elasticity depends on geometry and material. Since there are new materials and
geometries available in the market, it is important that designers should confirm with
manufacturers that the values they are using agree with full scale testing values.
Figure 3-5 Offset Position and Tension Effect
Figure 3-6 Illustration of Load Excursion Curve [Ref. 2]
25
A4163-01
3.1.3 Mooring Configuration
The most common mooring configurations are Spread Moored and Single Point
Mooring systems, which are taken to include turret systems. The key attributes of each
are discussed in this section.
Spread Mooring
This conventional mooring approach is widely adopted for semi-submersible
drilling/flotel/production units. For floating production applications, spread moorings
are used primarily with semi-submersibles and non-weathervaning FPSOs (i.e. no
turret) see Figure 3-7. Since the wave loading on a semi-submersible is relatively
insensitive to direction, a spread mooring system can be designed to hold a semi on
location regardless of the direction of the environment, although there is probably an
optimum heading. However, a spread system can also be applied to ship-shaped
vessels, which are more sensitive to environmental directions, as long as the
environmental conditions are relatively benign and the weather direction is fairly
uniform without strong cross currents. In a location such as the North Sea, the forces
which can be generated on the beam of a spread moored FPSO, plus the motions in
such conditions, effectively prohibit such a mooring arrangement.
The mooring lines can be chain, wire rope, fibre rope or a combination of the three.
Either conventional drag anchors or anchor piles can be used to terminate the mooring
lines.
Figure 3-7 - Typical Spread Moored Unit, Girassol FPSO offshore West Africa
(courtesy of Stolt Offshore)
26
A4163-01
Spread moorings are typically cheaper than turret moorings since they are mechanically
far less complicated. However, they are limited to where they can be used and they can
make offloading operations by a shuttle tanker somewhat more involved.
Single Point Moorings (SPMs)
Single point moorings (SPMs), such as internal or external turrets, are used primarily
for ship shaped units see Figure 3-1. They allow the vessel to weathervane, which is
necessary to minimise environmental loads on the vessel by heading into the prevailing
weather. There is a wide variety in the design of SPMs, but they all perform essentially
the same function.
3.1.4 Catenary and Taut Leg Moorings
Two main types of mooring system can be used for either the Spread or Single Point
system; Taut-Leg and Catenary. Both methods allow the system to withstand the
applied forces, but through different mechanisms.
A catenary system generates restoring force through the lifting and lowering of the
line onto the seabed, plus a limited amount of line stretch. This is shown in Figure 3-8
with a typical arrangement shown in Figure 3-9.
A taut-leg system makes use of the material properties of the mooring line, namely its
elasticity, as shown in Figure 3-10. A typical taut-leg arrangement is shown in Figure
3-11. Taut-leg moorings are relatively new and are typically used in deep water to limit
FPS offsets.
27
A4163-01
Figure 3-8 Illustration of Catenary System
Figure 3-9 - Typical Spread Moored Catenary System (Courtesy of Vryhof)
28
A4163-01
Figure 3-10 Illustration of Taut-Leg system
Figure 3-11 - Typical Spread Moored Taut-Leg System (Vryhof)
29
A4163-01
3.1.5 Vessel Dynamics
Waves will cause a vessel to move in all six degrees of freedom; surge, sway, heave,
roll, pitch and yaw. These degrees of freedom are illustrated in Figure 3-6.
The motion of the vessel to individual waves is called its wave frequency or first-order
response. As a mooring line moves through the water it will be subject to dynamic line
drag and inertia loading and sometimes a whipping effect. It is possible to take this
into account by undertaking a dynamic mooring analysis, but this does increase
computing time significantly.
Figure 3-12 Illustration of Surge, Sway, Heave, Roll, Pitch and Yaw
30
A4163-01
The compliance of a mooring system is such that conventionally the presence of the
mooring system is not considered to affect the wave frequency response. The overall
mooring system stiffness and associated natural frequency will influence its second
order or low frequency slow drift response.
In deep water for certain floating objects, such as deep draft Spars, the wave frequency
motion is attenuated to a certain extent by the mooring system due to the higher system
stiffness. Hence, a coupled analysis is sometime undertaken. The general conclusion
from this type of analysis appears to be that the mooring quasi-static tension has an
impact on a floater's wave frequency response, which in turn will affect the mooring
dynamic tension. On the other hand, the effect of dynamic tension is less important to a
floater's wave frequency response. For deep water the effect of risers on the vessel
response becomes increasingly important and this should be taken into account.
The coupled wave frequency motion of a floater can be calculated in the time domain
using the wave force, wave frequency added mass and damping, and mooring force at
each time step. Usually a convolution method needs to be adopted in the radiation
force calculation. Although the coupled wave frequency motion calculation in the time
domain is slower than the Response Amplitude Operator (RAO) based wave frequency
motion calculation, it is still acceptable. Typically a 3 hour simulation will take a few
minutes. However if there is very high mooring stiffness or if a mooring dynamic
analysis is performed, then the computing time will be high.
3.1.6 Mooring Design
The tensions experienced by a mooring system at any time are driven by the following:
- Static component from Wind, Mean Wave Drift and Current,
- Wave frequency component, caused by 1st order wave frequency motions and
drag/inertia effects on the line,
- Low frequency component, due to 2
nd
order low frequency waves and wind
dynamics.
The essence of mooring design is to optimise the behaviour of the mooring system such
that the excursions of the surface vessel do not exceed the allowable flexible riser
offsets, while at the same time ensuring that the line tensions are within their allowable
values. Thus the mooring system load offset curve should not be too hard or too soft
see Figure 3-13. Hence, considerable iteration work may be required to optimise a
system for a particular location.
31
A4163-01
It is worth noting that spring buoys (mid water buoys) and clump weights can also be
used to obtain an optimised mooring system stiffness by extending the resistive forces
over greater distances, hence allowing clearance over subsea features. However, their
use should be treated with caution, particularly in areas subject to harsh environmental
conditions, where they have been known to come adrift see Section 9.3. Buoys and
clump weights are also likely to introduce bending effects which may have an
undesirable impact on the fatigue life see Section 15.
Figure 3-13 Example of Optimising the Stiffness of the Load offset Curve
3.1.7 Mooring Analysis Calibration with Full Scale Behaviour
The determination of maximum tensions for a multiple line system requires application
of specialist computer programmes, which in many cases have been under continuous
development for a number of years. Despite this, there are still uncertainties in
estimating mooring loads using analysis software and model tests. Hence, it would be
desirable to compare the behaviour of a full scale FPS in known weather conditions
versus predictions. Surprisingly little work has been done on this topic, although this is
partly due to the difficulties associated with obtaining reliable weather and
instrumentation readings.
32
A4163-01
3.1.8 Active Winching and Thruster Assistance
Since there are now hundreds of years of accumulated mooring experience from semi-
submersible rigs, it is informative to understand the basis of their mooring operations.
This is reviewed in this section, which considers active winching and thruster
assistance.
Active Winching
Active winching can be undertaken on semi-submersible production, drilling and
accommodation units. There are two basic options, namely:
1. Leeward line slackening,
2. All round length adjustment, including windward lines, so that the tensions are as
well balanced as possible at the limit of vessel surge.
If the leeward lines are slackened down too much this can result in greater
yawing/surging and reduced direction control which can lead to higher line tensions. In
other words, if there is too much slack in the system, there is an increased danger of
high line snatch loadings.
Windward line tension optimisation can also be problematic. To quote from Robert
Ingliss informative 1992 paper [Ref. 3]:
in practise rig operators are reluctant to adjust windward line tensions in
severe weather conditions and usually restrict adjustments, if any, to
slackening leeward lines. This is partly to do with limitations in winch stall
capacity and the risk of a winch or brake failure, but most importantly the
majority of rigs are not provided with suitable tension monitoring devices and
computerised winch control systems which would make extensive line tension
optimisation a realistic possibility. The general situation is that analysts
frequently utilise line optimisation to reduce tensions to meet acceptance
criteria but these line tension optimisation procedures are almost never
implemented in practice on a rig.
Based on this type of feedback the latest mooring design codes (e.g. ISO [Ref. 4] + OS
E301 [Ref. 5]) do not permit either windward or leeward active wincing to minimise
mooring line tensions apart from going from one operational state to another.
Thruster Assistance
A number of semi-submersibles and a relatively small number of FPSOs are equipped
with thruster assistance. The thruster assistance can be categorised as either Thruster
Assistance (TA) or Automatic Thruster Assistance (ATA). TA is based on manual
joystick thruster control. ATA makes use of automatic remote control algorithm
system to control the behaviour of the thrusters.
33
A4163-01
It has been found that operation of the thrusters can be very effective in reducing peak
line tensions; even though the thrust delivered can be modest. Typically in a mooring
analysis the thrusters are considered to reduce the mean load applied to the mooring
system. However, thrusters also seem to damp down the magnitude of the slow drift
second order offsets. They can also be helpful with respect to heading control. This
can be particularly useful on a production vessel, if a small change in heading can result
in reduced vessel motions, thus improving the efficiency of the oil/water separation
process.
In practical terms, when operating in manual thruster mode, high line snatch loads can
be avoided by applying thrust as the wave train approaches. This will tend to push the
vessel in the direction of the advancing sea. As the wave passes it is necessary to ease
down on the thrust to avoid over slackening the windward lines. If these become too
slack there is an increased danger of snatch loading when the next wave train passes
through.
3.1.9 Metocean Parameters and their Impact on Mooring Integrity
For relatively benign environments, such as off West Africa, there is a much smaller
difference between operational and survival sea states compared to say the North Sea.
This means that if the metocean parameters, or the response of the vessel due to these
parameters, is underestimated, there is significantly less of an in built safety margin
compared to harsher climates, particularly with regard to fatigue.
The degree of spreading of the waves (see Figure 3-14 and Figure 3-15) can also affect
mooring analysis results. The geographic area and fetch distance will influence the
type of waves likely to be encountered in practice. Conventionally, short crested seas
are considered to result in reduced wave frequency response and hence reduced
mooring line tensions - see section 3.3.2 of [Ref. 6]. However, recent model test results
at DHI in Denmark has shown that for certain vessel sizes the mooring loads in short
crested waves can be higher than in long crested waves [Ref. 7]. Thus the key point is
to ensure that the response of the system is thoroughly evaluated for the worst expected
conditions (ie short or long-crested) both from a fatigue and a strength point of view.
34
A4163-01
Figure 3-14 - Illustration of Long Crested (Unidirectional) Seas
Figure 3-15 - Illustration of Short Crested (Confused) Seas
35
A4163-01
3.1.10 Rogue/Steep Breaking Waves and Shock Loading
Mariners have used phrases such as Freak Waves, Rogue Waves, Walls of Water or
even Holes in the Sea, to describe some of the conditions they have experienced at
sea. Trading vessels are typically weather routed to avoid the worst of predicted
weather conditions. However, permanently moored FPSs have to ride out whatever
weather is thrown at them.
From a statistical sense the longer a FPS is on station the more likely it is to experience
100 year + conditions. If an elderly FPS with a mooring system which has seen wear,
corrosion and has accumulated some hair line cracks is subject to such conditions, the
likelihood of single or even multiple line failure is increased.
Very occasionally an unusually steep wave slam load could occur at the same time that
a floating structure is around its maximum slow drift offset. The resulting shock or
spike load on the mooring might be quite considerable. How much this shock loading
is transferred to the mooring lines will depend to a significant extent on the degree of
structural damping in the hull structure, the vessel inertia, how long the load acts and
where the moorings are relative to where the wave impacts. For a semi, where you
might get wave slam/slap right into one of the corners (see Figure 3-16), the amount of
structural damping might well be less than compared say to a FPSO with an internal
turret (see Figure 3-19). Hence the loading could be higher.
Figure 3-16 - Example of a Wave Breaking on a Column of a Semi-Submersible
In deep water steep elevated wave fronts with breaking or near breaking crests can
occur see Figure 3-17. In addition, a "Three Sisters" wave group can occur in which
the second wave is generally the highest and is often preceded by a long trough.
Hence, a moored object may ride the first wave, but then plunge submerged into the
base of the second steep fronted wave that then inflicts the greatest shock loading.
36
A4163-01
Figure 3-17 - Illustration of Deepwater Breaking Wave Types (Plunging Break on the
Left and Spilling Breaking on the Right)
In November 1998 the Schiehallion FPSO was struck by a wave which was felt
throughout the vessel. The wave caused tears in the forward shell plating of the
forecastle superstructure, buckling of supporting stiffeners and permanent deformation
of the forecastle tween deck see Figure 3-18. Production was shut down and non
essential personnel were evacuated to a nearby drilling rig. In this instance no damage
was reported to the mooring system, but it illustrates the danger presented by infrequent
steep breaking waves.
Figure 3-18 - Illustration of the Damage Caused to Schiehallions Bow by an Unusually
Steep Wave (courtesy of BP)
37
A4163-01
Present day standard mooring analysis tools do not evaluate this potential shock load
effect on the mooring systems. Hence it is difficult to quantify. But there is a
possibility, based on the wave description, that it could have been a factor which led to
the virtually instantaneous multiple line failures experienced by Petrograd 1 in the
early 1990s (see Section 4.3). This might also be a factor in the relatively frequent
mooring line failures experienced by semi-subs. It is recommend that this topic should
be investigated further and that appropriate cross checks should be made with the real
life recorded response of FPSs in severe/steep sea weather conditions. However, it also
should be noted that such weather conditions do not occur very often.
Figure 3-19 - Model Illustration of the Effect of a Breaking Wave on a FPSO (Courtesy
of APL website)
The right hand side photograph of Figure 3-19 is perhaps an example of the type of
wave conditions which could impart a shock loading to the moorings, depending on the
FPSO offset at the time. If a mooring line had already broken and its failure had not
been detected (due to a lack of failure of instrumentation) the chance of additional line
failures would be high in these conditions.
38
A4163-01
3.2 Mooring Line Constituents
3.2.1 Introduction
Various different materials can be used to assemble a mooring line. This section
provides a brief description of the main components that typically constitute a mooring
line. The pros and cons of the various types of line components are explained. This
helps to aid understanding when considering how actual systems have performed in
situ. Connectors and terminations are considered separately in Section 14.
3.2.2 History of Studded and Studless Chain
Early mooring lines tended to make use of simple links without studs. Development of
this design led to usage of studded links, see for example Figure 3-20. Ease of
handling and avoidance of kinking were the primary reasons for the introduction of
studs. The resulting link geometry (see Figure 3-21) took advantage of the ability of
the stud to resist some of the bending loads in the links. The studded link standard
geometry of length of 6 x Bar Diameter (D) and breadth of 3.6 x D was approved by
the British Admiralty in the 1860s.
Historically anchor chain used on ships was, in general, only required to meet
intermittent short term loading and therefore, even over a long ship service life, fatigue
was unlikely to be a problem.
Figure 3-20 - Isambard Kingdom Brunel in front of Studded Chain for the Great
Eastern steam ship, circa 1858
39
A4163-01
Studless Link Studded Link
Figure 3-21 Comparison of the Geometry of Modern Studded and Studless Chain
[Note: DNV Cert Note 2.6, states 3.3D to 3.4D for the of studless link width]
Fairly recent long-term applications of chains in the moorings of floating production
systems have brought about the development of studless chain. The studless chain link
has been redesigned with a smaller breadth to reduce the bending loads. These designs
are increasingly used for long-term moorings because loose and missing stud problems
are eliminated. Unfortunately, however, the fatigue life of studless chain has been
shown to be half that of comparable studded chain, based on the results of fatigue
testing [Ref. 8]. In other words the fatigue endurance of studded chains is twice that of
studless if the studs remain tight. Of the 70 fatigue failures reported in the Houston JIP,
52% occurred at an inner Half-Crown position, 34% at an inner Crown position and
14% at a mid leg position. The Crown refers the area of maximum bend and Half-
Crown essentially refers to the area of the link where bending commences.
The studless link standard geometry of length of 6 x D and breadth of 3.35 x D came to
the market after 1989 as consequence of collaboration between DNV and Vicinay for
the Veslefrikk B project. For this chain the first tentative specification went out in
1995 with the DNVs Certification Note 2.6. More recent developments include
customised chain geometries also known as Variable Geometry and Weight (VGW) as
discussed in OTC paper 8148, 1996 [Ref. 9]. VGW provides flexibility to modify link
geometry and weight to suit a particular application see for example Section 18.8.2.
Table 3-2 and Table 3-3 summarise the relative merits of studless and studded chain in
terms of design, manufacturing, inspection and maintenance.
40
A4163-01
Requirement
Recommended
Chain
Reason
Lower static or dead weight in the catenary Studless Lower weight per metre
Access for shackle and accessory
connection
Studless More interior link space
Versatility, similar to end links Studless Completely open links
Greater safety factor with same weight per
metre in the catenary (strength to weight
ratio)
Studless Larger possible diameter with less
weight per metre
Greater stiffness in the mooring line Studlink Higher elasticity modulus
Higher Break Load Both Same break load, but different
proof loads
Transition through windlasses and
fairleads
Both But studless more likely to knot or
twist
Long fatigue life Open to discussion See previous page
Table 3-1 Summary of Chain Design Parameters (modified from Vicinay Chain
Catalogue)
Requirement
Recommended
Chain
Reason
Better inspection of weld and crown area Studless Greater access due to lack of stud
Elimination of stud locating problems Studless Lack of stud
Oversizing of the link in the weld zone Studless Elimination of the flattening and
material expansion in the weld
zone
No links with stud looseness Studless Lack of stud
Table 3-2 Comparison of Manufacturing Parameters
41
A4163-01
Requirement
Recommended
Chain
Reason
Eliminate premature fatigue due to loose
studs
Studless No stud, therefore no notch effect
Reduce inspection/repair costs Studless Easier access + no loose studs to
repair
Eliminate galvanic reaction between the
stud and the link
Studless No stud, therefore no possibility of
reaction
Increase reliability of the chain over time To be determined Although there are no loose studs
issues with studless, the fatigue
performance of studless is less
good than that of studded
Handling and connectability with D
shackles and hooks
Studless Better access for the through pin.
Minimal requirements and
restrictions
Early indication of system degradation Studlink Condition of studs likely to be
representative of system as a whole
Handling and Manoeuvrability with
respect to bending shoes and chain
stoppers
Both Certain restrictions, in general
studless chain is more likely to
knot than studded
Table 3-3 Chain Geometry Implications for Inspection and Maintenance
3.2.3 Effect of Loose Studs
Offshore oil industry experience with studded chains has shown that during use, the
studs start to get loose and the seat of the studs is often the initiation point for fatigue
cracking [Ref. 2 and Ref. 11].
The BOMEL JIP [Ref. 10, p.54] showed that the Stress Concentration Factor (SCF) in
a studless link (a link in which the stud has been removed as opposed to a link designed
to be studless which has a different geometry) is much higher than in a studded link, i.e.
5.3 compared to 3.8. It follows therefore that a link with a loose stud is likely to have a
much shorter fatigue life than one where the stud is properly fitted.
42
A4163-01
Sandbergs report [Ref. 12] discusses the effect of loose studs and reports the
following: (Page 11) This ovalization occurred at the point where the edge of the
footprint forms a notch effect, which in some instances appeared quite severe. It was at
this point that the fatigue cracks were initiated and then propagated through the side of
the link. However, on page 16 Sandberg advised that Not all loose studs had
developed cracks even where the end of stud play was up to 4.0mm indicating that
many factors play a part in the initiation and propagation of these cracks including the
ultimate tensile stress (UTS) of the studs. The mechanical properties for the stud are
important, particularly with regard to yield strength, for the setting of the chain under
the proof load. If the properties are not correctly balanced then fixing of the stud may
be significantly impaired which can affect the future serviceability of the chain.
Figure 3-22 shows a SPATE contour map which gives a crude indication of how a
loose stud can affect the stress distribution in a studded link where the stud has become
loose. The basic theory behind SPATE (Stress Pattern Analysis by Thermal Emission)
is the detection of minute changes in surface temperature due to the pseudo adiabatic
response of a material under stress. Through an infrared detector, scanning the surface
of a given material, relative changes in temperature are fed to a computer system for
correlation and finally presented as a pictorial colour image of the stress pattern over
the scanned area. These pictures can be interrogated further to obtain stress values at
any given point. As the stresses in the three principal planes contribute to the overall
temperature change, stress values obtained are a summation of the principal stresses
generated by the dynamic loading in each plane.
Figure 3-22 - SPATE Contour Map of a 76mm Loose Stud Chain Link [Ref. 8]
43
A4163-01
Asymmetric Studs
Asymmetric studs were developed to reduce the amount of loose studs experienced in
the field. The Asymmetric stud is designed in such a way that gives equal foot prints
on either side of the link see Figure 3-23. In studlink chain, the asymmetric stud
design is claimed to provide more equal stud indentation and contributes to a more
symmetric stress distribution in the link.
Figure 3-23 - Example of the Arrangement of an Asymmetric Stud
During manufacturing of the formed link, the flash butt weld (FBW) side of the link is
still very hot by the time the process brings the link to the stud press. Hence there is a
very hot side and the other side of the link, the parent material, which has already
started to cool, is therefore at a lower temperature. With a normal stud the cold stud,
when pressed into the link will, on the hot side, sink deeper into the FBW and to a
much lesser extent on the parent material side. Thus, if you inspect stud link chain with
a stud missing you can quite often see there is almost no foot print on the parent
material side.
The asymmetric stud provides equal foot prints on both sides. The asymmetric stud is
then pressed to allow it to expand thus locking the stud in place. The actual
configuration of the stud faces are different compared to the standard studs, because the
edges of the asymmetric stud are more rounded to reduce the chance of notches or
crack initiation under the studs, the studs are also flatter on one side so the stud cannot
sink into the hot FBW and more rounded on the other side to fit the form of the link on
the parent material side. In addition, when the studs are expanded it puts a spring effect
into the link thus assisting to keep the studs in place.
44
A4163-01
3.2.4 Chain Grade and Ultimate Tensile Stress
There are a variety of chain grades available, each of which are distinguished by the
differing yield strengths of the steel that are used in their manufacture. The Grade 1
chain (designated Q1, U1 or K1 depending on whether Continental, UK or Norwegian
specification) was developed using mild steel. This grade of chain is not now used in
offshore mooring systems. The most important chain grades for the offshore industry
are as follows:
- Oil Rig Quality (ORQ), dating from the beginning of the 1970s with 641MPa
issued by API
- R3, dating from the mid 1980s with 690 MPa to meet ORQ + 10%
- R3S, with 770 MPa to meet ORQ + 20%
- R4, with 860 MPa
- R4S with 950 MPa or R4 + 10%
- R5 with more than 1,000 MPa
R is the standard International Association of Classification Society (IACS)
terminology for offshore mooring chains. Interestingly, none of the Certification Notes
for any of the Certifying Authorities actually lay down detailed minimum alloy content
for specific grades.
It is important not to confuse standard Ship or Marine Grade chain with offshore R
grade chain. In other words Grade 3 is different from R3. Table 3-4 below summarises
the characteristics of Marine Grade chain:
Description Ultimate Tensile
Strength (MPa or
N/mm2)
Marine
Grade U1
Wrought iron or mild steel 310
Marine
Grade U2
Special Quality Steel 490
Marine
Grade U3
Extra Special Quality Steel 690
Table 3-4 - Summary of Ship or Marine Grade Chains [Ref. 13]
45
A4163-01
3.2.5 Chain Grade and Hardness
Surface hardness values based on, for example, Brinell or Vickers Hardness Testing are
not normally reported for different chain grades, although API 2F [Ref. 65] (section
5.5) and DNV Certification Note 2.6 [Ref. 18] (section 6.6.4) mention that such values
should be obtained during manufacture.
In Table 3-5 Vicinay has reported surface hardness values for various grades of chain.
It should be noted that the tests were undertaken on non polished services and thus the
values should be considered to be indicative only.
Chain Grade Surface Hardness (Brinnell (HB)
Grade 3 220 250
R3 235 -260
R3S 250 275
R4 275 305
R5 305 - 325
Table 3-5 Example of Indicative Surface Hardness Values for Various Chain Grades
(courtesy of Vicinay)
The applied tension in a link and hence the resulting stresses could conceivably result
in a change in surface hardness. Vicinay has investigated this point and concluded that
the increase in superficial hardness of chains, due to tension, is insignificant.
One oil company chain specification states that the maximum hardness should not
exceed 350 HV10 (HV = Vickers hardness, similar to Brinnell in this range of
numbers) mainly due to a sensitivity to hydrogen induced crack growth. In general the
higher the hardness the more the sensitivity increases, but this depends on steel
composition. It is the minimum yield stress that determines the minimum hardness.
3.2.6 Charpy Impact and CTOD Tests
Charpy Impact test energy measurements are available for different chain grades. A
Charpy Impact test will give an idea of the ductility of the material and its susceptibility
to brittle fracture depending on the temperature. Therefore a Charpy test does not give
a full picture of surface hardness. Charpy is concerned with mechanical fracture of the
material. For the same steel the mechanical of fracture is more related to structural
metallurgical.
46
A4163-01
During chain manufacture Crack Tip Opening Displacement (CTOD) values should be
evaluated both for the main body of the link and for the weld. For example Table 10-5
of DNV Certification Note 2.6 [Ref. 18] provides critical defect sizes which the metal
should achieve. CTOD gives a better indication of Fracture Toughness, compared to
Charpy-V impact test which only gives an indication of the toughness of a metal.
It is worth noting that the temperature of the test is important because almost all steels
have a zone of high stable values over a "plateau" region. But suddently, within very
few divisions of temperature variation, they fall to a zone of lower values.
3.2.7 Manufacturing Tolerances and Implications for Wear
Figure 3-24 and Figure 3-25 show typical chain manufacturing tolerances. The
implications of these types of tolerances are discussed in more detail in Section 18.6.
Figure 3-24 Indication of the Manufacturing Tolerances of Studless Links (courtesy
of Vicinay)
47
A4163-01
Figure 3-25 Studlink Manufacturing Tolerances (courtesy of Vicinay)
48
A4163-01
3.2.8 Wire Rope
Figure 3-26 shows the normal wire rope construction types used for offshore mooring
lines. Six strand independent wire rope core (IWRC) is typically used for mobile
drilling units due to its lateral flexibility and relative cheapness. Spiral strand wire is
generally torque balanced, the implications of which are discussed in more detail in
Section 6.2.2. Sheathing has been introduced to protect the wire from corrosion. It
will, however, be interesting to see if, over time, the sea water ingress causes corrosion
underneath the sheathing. At present there are no real techniques available to monitor
such corrosion, see also Section 18.8.
Figure 3-26 - Illustration of the Make Up of Different Wire Rope Types (courtesy of
Bridon)
The yield strengths of steel used in the construction of wire mooring ropes vary but are
very high, for example see Table 3-6.
Construction Ultimate Tensile Stress (N/mm
2
)
Six strand (IWRC) 1860
Spiral strand 1570
Table 3-6 Illustration of Indicative Wire Rope Material Properties [Ref. 2]
49
A4163-01
The two main wire rope types utilised have differing properties. The pros and cons of
the two rope constructions are summarised in Tables 3-6 and Table 3-8:
Spiral Strand Advantages Six Strand - Advantages
Higher Strength to Weight Ratio Higher Elasticity
Higher Strength to Diameter Ratio Greater Flexibility
Torsionally Balanced Lower Axial Stiffness
Higher Resistance to Corrosion
Higher Fatigue Resistance
Table 3-7 - Comparison of the Advantages of Spiral and Six Strand Wire (courtesy of
Bridon)
Spiral Strand Cons Six Strand Cons
Easy to kink during installation see Introduces torque into a mooring line
Figure 6-7
More expensive than six strand Typical design life of 5 to 8 years
Table 3-8 - Comparison of the Cons of Spiral and Six Strand Rope
Mooring wire is zinc galvanised to provide defence against corrosion; the major factor
along with fatigue determining mooring line service life. Heavier zinc coatings are used
on the larger wires of the spiral strand product enhancing corrosion protection
properties. The larger outer wires of the six strand product may also use heavier zinc
coatings to increase the attainable design life. An anti-corrosion blocking compound
may be applied during manufacture to add a further corrosion prevention measure.
Typical service life expectancy is shown in Table 3-9.
Design Life Recommended Product Type
up to 6 years Six Strand
up to 8 years Six Strand c/w zinc anodes
up to 10 years Six Strand c/w 'A' galvanised outer wires & zinc
anodes
10 years plus Spiral Strand
15 years plus Spiral Strand c/w Galfan coated outer wire
Table 3-9 - Wire Rope Recommendations for Varying Field Lives (courtesy of Bridon)
50
A4163-01
Wire rope suppliers can provide sheathed products in yellow polyethylene with a black
longitudinal stripe. The yellow colour aids in service inspection as damage shows
black against the yellow background. The black stripe can highlight any turn
introduced into the wire during installation.
3.2.9 Fibre Rope
High strength and high modulus fibre materials offer certain advantages for offshore
mooring systems. The use of fibre ropes has increased substantially with the move into
deep water and as test results become available. Figure 3-27 illustrates the chronology
of the Fibre Rope test programme undertaken as part of the US Deep Star programme.
Figure 3-27 Chronology of Deep Star Funded Synthetic Mooring Studies
OTC 12178 [Ref. 14]
Synthetic ropes are made of visco-elastic materials, so their stiffness characteristics are
not constant and vary with the duration of load application, the load magnitude, the
number of load cycles and the frequency of load cycles [Ref. 15]. In general, synthetic
mooring lines become stiffer after a long time in service. Synthetic ropes also creep
over time and this needs to be taken account of during design and installation see for
example Figure 6-9.
51
A4163-01
3.2.10 Anchoring Options (Drag Anchors, Vertical Uplift Anchors, Piles, Suction Piles)
Drag embedment anchors (see Figure 3-28) are the most popular type of anchoring
point available today. This type of anchor has been designed to penetrate into the
seabed, either partly or fully. The holding capacity of the drag embedment anchor is
generated by the resistance of the soil in front of the anchor. The traditional drag
embedment anchor is very well suited for resisting large horizontal loads, but not for
large vertical loads.
Drag embedment anchors are generally installed by applying a load somewhere close to
the maximum anticipated intact load. At this time the anchor will have penetrated to a
certain depth, but will still be capable of further penetration as the ultimate holding
capacity of the anchor has not been reached. By this stage the anchor will have
travelled a certain horizontal distance, called the drag length. Following installation the
anchor is capable of resisting loads equal to the installation load without further
penetration and drag. When the installation load is exceeded, the anchor should
continue to penetrate and drag until the soil is capable of providing sufficient resistance
to match the applied load or drag failure takes place.
Figure 3-28 - Accurate Drag Anchor Placement by Crane in Good Weather Conditions
(courtesy of Stolt Offshore)
Vertical load anchors (VLAs) are installed in a similar manner to a conventional drag
embedment anchor. During installation the load arrives at an angle of approximately
45 to 50 to the fluke. After triggering the anchor to the normal load position, the load
always arrives perpendicular to the fluke. As a VLA is deeply embedded and always
loaded in a direction normal to the fluke, the load can be applied in any direction.
Consequently the anchor is ideal for taut-leg mooring systems as long as complete
embedment is achievable.
52
A4163-01
Figure 3-29 Installation and Normal (Vertical) Load Position (courtesy of Vryhof )
Piled anchors (see Figure 3-30) are hollow steel pipes that are installed into the seabed
by means of a piling hammer or vibrator. The holding capacity of the pile is generated
by the friction of the soil along the pile and lateral soil resistance. It is usually
necessary for the pile to be installed at considerable depth below the seabed to obtain
the required holding capacity. Piles are capable of resisting both horizontal and vertical
loads.
Figure 3-30 Anchor Pile + Chain Tail Deployed by a Twin Crane Construction Vessel
(courtesy of Stolt Offshore)
Suction anchors (see Figure 3-31), like piles, tend to be hollow steel pipes, although the
diameter of the pipe is much larger than for the pile. The suction anchor is forced into
the seabed by means of a pump connected to the top of the pipe, creating a pressure
difference. When pressure inside the pipe is lower than outside, the pipe is sucked into
the seabed. The pump is then removed following installation. The holding capacity of
the suction anchor is generated by the friction of the soil along the length of the pipe
and the lateral soil resistance. The anchor is capable of withstanding both horizontal
and vertical loads.
53
A4163-01
Figure 3-31 Suction Anchor Deployment (courtesy of Stolt Offshore)
In most FPS applications the anchors are semi-permanent fixtures, unlike mobile
drilling units where they would be routinely recovered. The forces involved in anchor
recovery are high and can lead to damage.
54
A4163-01
3.2.11 Line Pull In (Winching) Options
The mechanical design of the line pull in system will influence the design of the
mooring system. There are different options which can be adopted. This section
briefly reviews the various types and outlines their pros and cons.
Pros:
Powerful
mean of
Chain
Jack
tensioning.
Cons:
Slow
manipulation.
Most common
Powered
Windlass
method for
handling and
tensioning
chain.
55
A4163-01
Linear
Most
when high
tension and
required.
Cons:
passes through
winch.
Most
for handling
wire rope.
Cons:
rope is under
wire rope.
Winch
applicable in
permanent
applications
large-diameter
wire rope are
Requires a
large diameter
take-up reel to
coil the wire
rope after it
the linear
Drum-
Type
Winch
conventional
method used
Pros:
Operation of a
drum-type
winch is fast
and smooth.
1-As the
requirement
for line sizes
and lengths
increases, the
size of the
winch can
become
impractical.
2-When wire
tension at an
outer layer on
the drum,
spreading of
preceding
layers can
occur causing
damage to the
56
A4163-01
Traction
Winch
The Traction winch has been developed for high tension mooring
Pros:
applications as well as for handling combination mooring systems. It
1-Compact
consists of two closely spaced parallel mounted powered drums, which
size.
are typically grooved. The wire rope makes several wraps (typically 6 to
2-Capability
8) around the parallel drum assembly. The friction between the wire
to provide
rope and the drums provides the gripping force for the wire rope. The
constant
wire rope is coiled on a take up reel which is required to maintain a
torque.
nominal level of tension in the wire rope (typically 3%-5% of working
-Ability to
tension) to ensure the proper level of friction is maintained between the
handle very
wire rope and the traction winch. This system has been favoured for use
long wire rope
in high tension applications due to the compact size, capability to provide
without
constant torque, and ability to handle very long wire rope without
reduced pull
reduced pull capacity [Ref. 16].
capacity.
57
A4163-01
3.3 Determination of Minimum Break Load (MBL) & Maximum Stresses
3.3.1 What is MBL and How is it Determined ?
A fundamental characteristic of any mooring component is its minimum break load
(MBL). This section discusses the load extension characteristics of typical steel
mooring components and reviews how this relates to MBL.
Initially it is helpful to review what happens when a material is loaded up as is seen, for
example, in a tensile test machine (see Figure 3-32). An indicative resulting stress (o)
strain (e) behaviour and associated terminology are shown in Figure 3-34. Figure 3-33
shows the remains of the chains which were sectioned for material testing in the
machine shown in Figure 3-32.
Figure 3-32 - Example of a Tensile Test on a Steel Sample cut out from a Chain Link
58
A4163-01
Figure 3-33 - Example of a Chain Sectioned for Material Testing
Figure 3-34 - Example of Terminology during a Tensile Test (courtesy of Ashby &
Jones, [Ref. 17])
59
A4163-01
Tensile testing will produce a load versus displacement curve which is then converted
to a nominal stress versus nominal strain or o
n
versus e
n
curve where:
o
n
= Force/Area prior to testing (A
0
)
e
n
= Displacement/Length prior to testing (l
0
)
The advantage of plotting o
n
versus e
n
is that it allows comparison of specimens with
different A
0
and l
0
on the same graph, thus examing the properties of the material
unaffected by speciment size.
The following terminology is illustrated on Figure 3-34:
o
Y
= Yield strength (F/A
0
at onset of plactic flow)
o
0.1%
= 0.1% Proof Stress (F/A
0
at a permanent strain of 0.1%. 0.2 % proof stress
is also sometimes quoted see also section 14.5. Proof stress is useful for
characterising yield of a material which yields gradually and thus does not show a
distinct yield point)
o
TS
= Tensile Strength (F/A
0
at onset of necking)
e
f
= (Plastic Strain after Fracture or Tensile Ductility. The broken pieces arr put
together and measured and e
f
calculated from (l l
0
)/l
0
, where l is the length of
the assembled pieces and l
0
is the legth at time 0.
The behaviour past the onset of necking (see Figure 3-34) depends on material
characteristics as well as the rate at which the load is applied. Thus the continuation of
the curve beyond the maximum load is difficult to obtain, both during tests and in a
theoretical analysis of a component. Hence it is of no practical value from the
viewpoint of the structural integrity of the mooring system.
Figure 3-35 illustrates how the shape of the stress strain curve will vary depending on
material charaterisitics/grade of chain. It can be seen that as strength has increased the
strain before fracture has decreased. Brittle fracture was a problem in the early days of
developing grade 4 mooring chain in the first half of the 1980s. However, by carefully
controlling the alloy constituents and the quenching and tempering process it is now
possible to manufacture a high strength steels, such as R4 or R4S+ (R5), with a grain
structure which is more ductile and hence is resistant to brittle fracture. High
tempering temperatures may be utilised to optimise Ductility/Toughness Properties.
60
A4163-01
Stress Versus Strain Plots for various Chain Grades
0
R4
200
400
600
800
1000
1200
S
t
r
e
s
s

(
N
/
m
m
^
2
)

R3 - R3S
R4S+ = R5
0 2 4 6 8 10 12 14 16 18 20 22
Strain (%)
Figure 3-35 - Stress Strain Curves for R3, R4 and R5 Chain Steel (Data courtesy of
Vicinay)
For a strong, fairly ductile (non brittle) material one potentially has a choice as to what
to value to select for MBL. It could either be the load recorded in the test bed when the
material fails or it could be the maximum load before the loads drops away prior to
breaking. The load recorded at which the material breaks is related to the rate at which
the load is applied, so it is somewhat variable. Internation Association of Classification
Societies (IACS)W22 [Ref. 66] states: each sample shall be capable of withstanding
the specified break load without fracture and shall not crack in the flash weld. It shall
be considered acceptable if the sample is loaded to the specified value and maintained
at that load for 30 seconds. At the end of the break test, in general the tested
component should be scrapped.
It can thus be appreciated that manufacturers have to confirm by physical testing that
they have achieved the specified MBL and proof load values. In general it would be
desirable that sufficient testing should be undertaken to obtain a spread of results.
However, since chain is only as strong as its weakest link the minimum value achieved
should be considered to be representative of the MBL, not for example the mean or the
mean minus a number of standard deviations.
Therefore, the catalogue specified minimum break load (MBL) is actually an agreed
specified strength which has an associated testing requirement to ensure that this is
achieved. Table 3-10 below indicates the MBL and proof loads for various sizes and
grades of chain. The d in the expressions refers to the chain diameter in mm.
61
A4163-01
Stud Link Studless
Proof Load ORQ KN 0,0140 *d^2*(44-0,08*d)
Minimum Break Load ORQ KN 0,0211 *d^2*(44-0,08*d)
Proof Load ORQ + 10 KN 0,0154 *d^2*(44-0,08*d)
Minimum Break Load ORQ + 10 KN 0,0232 *d^2*(44-0,08*d)
Proof Load ORQ + 20 KN 0,0168 *d^2*(44-0,08*d)
Minimum Break Load ORQ +20 KN 0,0253 *d^2*(44-0,08*d)
Proof Load R3 KN 0,0156 *d^2*(44-0,08*d)
Minimum Break Load R3 KN 0,0223 *d^2*(44-0,08*d)
Proof Load R3 S KN 0,0018 *d^2*(44-0,08*d) 0,0174 *d^2*(44-0,08*d)
Minimum Break Load R3S KN 0,0249 *d^2*(44-0,08*d)
Proof Load R4 KN 0,0216 *d^2*(44-0,08*d) 0,0191 *d^2*(44-0,08*d)
Minimum Break Load R4 KN 0,0274 *d^2*(44-0,08*d)
Proof Load R4S KN 0,02376*d^2*(44-0,08*d) 0,0201 *d^2*(44-0,08*d)
Minimum Break Load R4S KN 0,03014 *d^2*(44-0,08*d)
Proof Load R5 KN 0,0251*d^2*(44-0,08*d) 0,0222 *d^2*(44-0,08*d)
Minimum Break Load R5 KN 0,03186 *d^2*(44-0,08*d)
Table 3-10 - Stipulated MBL and Proof Load Values for Various Sizes and Grades of
Chain (courtesy of Vicinay)
3.3.2 Determination of MBL for Worn Mooring Line Components
To be confident that worn components can withstand the maximum estimated mooring
loads it is recommended that break testing should be undertaken. Classification
Societies define MBL to be the maximum load which the unit can withstand for 30
seconds without failing. However, for worn components it is not possible to know
what strength the component should have to begin with. Section 18.8 discusses this
subject in some detail.
62
A4163-01
3.3.3 Areas of Maximum Stress in a Chain Link
Since chains are fundamental to the majority of long term mooring systems (normally
present in the thrash zone and at the fairleads) it is helpful to understand on a loaded
chain where the maximum stresses are likely to be experienced. Chain links are
complex, statically indeterminate structures subjected to a combination of bending,
shear and tension when loaded. Figure 3-36 gives a fairly crude approximation of the
stress distribution in a loaded link. Minimum tensile stress occurs in the outside fibres
at the intersection of the curved end with the long axis. Maximum shear stress is about
45 away from this axis and on a radial line through the centre of curvature of the end
of the link. For chain of low to medium hardness, failure is typically by shear. As
hardness increases, the typical failure mode shifts to tension because of bending. The
failure location shifts from the maximum shear plane to a plane in the long axis of the
link. Combined stresses reduce breaking strength to about two thirds of that computed
by assuming that the load is uniformly distributed in simple tension across the two
circular cross sections of the straight side of the link. If a link is turned sideways, the
long sides of the link are subject to high bending stresses and the load carrying capacity
of the link is greatly reduced.
Figure 3-36 Approximation of the Stress Distribution in a Typical Chain Link
[Ref. 19]
A more accurate determination of stresses in a link or connector can be achieved by
undertaking a finite element analysis. Figure 3-37 and Figure 3-38 illustrate typical FE
plots for a chain link and a shackle body. Such an FE analysis can consider geometric
and material nonlinearities.
From a FE analysis or measurements the value of 1.8P/A in the Crown of the link is
found to be 2.1P/A and the value in the weld section changes from 1.8P/A to 1.65P/A.
The outer Stresses change from -0.77P/A to 0.02P/A.
63
A4163-01
During forging and the flash butt welding residual stresses will be introduced into a
link. Such stresses will tend to be minimised during the heat treatment process.
However, Vicinay have evaluated the influence of residual stresses created during the
proof loading process. During two separate analyses it was found that the effect of the
residual stresses were negligible. Still it would be useful to have more data available
on the effect of residual stresses see Section 14.5.
Figure 3-37 - Illustration of a Finite Element Representation of a Chain Link
(Courtesy of Vicinay)
Figure 3-38 Finite Element Representation of a Shackle Body
(Courtesy of Vicinay)
Linear elastic analyses are performed for fatigue life assessments and non-linear elastic
plastic analysis for ultimate load predictions. The behaviour of the components is
typically analysed when working under a particular loading boundary condition.
64
A4163-01
W ea r a nd f a t i g u e
4 CONTEXT SETTING - HISTORICAL INCIDENTS
AND THEIR SIGNIFICANCE
So far the modern (say post 1995) Floating Production Industry has not made headline
news after a spectacular mooring failure resulting in a unit breaking completely free of
its moorings and risers. However, a review of some of the more significant mooring
failures seen over the last few decades shows that there have been serious incidents in
the relatively recent past. Given the substantial increase in the number of FPSs and
their increasing age, the probability of a future incident does increase. Hence this
section is included to set the context of mooring problems and to try to help to guard
against complacency that technology has advanced to the extent that major failures will
not happen again.
4.1 Long-Term Degradation Mechanisms
Today we have accumulated hundreds of rig years of semi-submersible operating
history based on drilling, accommodation and production units. Semi-submersibles
have relatively good motion characteristics compared to mono hulls (ships). It will be
interesting to see if mono-hull mooring systems suffer from greater degradation
compared to semi-sub mooring systems. Overall, it is important that the lessons learnt
from semi-subs are transferred to the design of long-term mooring systems. This
section attempts to aid in this process.
Figure 4-1 illustrates some of the main loading or degradation mechanisms a long-term
moored system must be able to withstand for possibly in excess of 20 years. At the end
of this period the mooring system still needs to be able to withstand a 100 year return
period survival storm. In general, from an engineering perspective, it is worth
considering that there are not many mechanical systems, which after 20 years of hard
use can still be expected to meet their original design specification.
B en d in g & T en si o n
H i g h es t T en si o n s
C o rro si o n
I m pa ct & A b ra s i o n
Ben d in g & T en si o n
Hi g h es t Ten si o n s
Co rro si o n
I mpa ct & Ab ra s i o n
W ea r a nd f a t i g u e
Figure 4-1 Illustration of some of the Main Factors which Influence Mooring
Integrity
65
A4163-01
Some of the most severe conditions occur in the so called Thrash Zone, when the
chain comes in contact with the sea-bed at the end of the catenary. It is at this point
that most cyclic movement occurs and increased stresses may be encountered due to
momentum of the moving material. This movement may also cause damage if the sea-
bed is hard where the chain contacts it. Another problem resulting from this
movement, if the chain is studded, will occur if the studs are able to move. When the
studs are loose enough to move freely in the link, then corrosion in the footprint, of
both the stud and side of the link will be enhanced due to the fretting action removing
the corrosion products and exposing a clean surface. As the process progresses, the rate
of corrosion will become higher. This may also be enhanced by crevice corrosion in
the early stages [Ref. 12].
4.1.1 Argyll Transworld 58 Breakaway
Hamiltons Argyll development in the UK sector of the North Sea in 1975 was the
earliest use of a semi submersible as a production platform see Figure 4-2. The
converted drilling rig Transworld 58 was renamed North Sea Pioneer and was
moored in 79 metres of water. Multiple steels risers were used, since at that time
flexibles were not considered to be sufficiently developed. The risers had to be pulled
during rough weather, which happened 28 times over the first 6 years.
Figure 4-2 - North Sea Pioneer on the Argyll Field
66
A4163-01
After 6 years on station the TW58 broke completely away from its moorings. The
following description describes the sequence of events that led to the breakaway. The
information is taken from a HSE database, in which the identity of the units involved
has been deliberately removed.
In 1981 the converted semi-sub producing from a subsea manifold abandoned
production due to the fierceful weather. At 01:19 hrs a 9-ft heave and an 82-knot
wind was recorded and the conditions continued to deteriorate and at 02:36 hrs,
anchor chain no. 4 parted (tension: 200,000 lbs). Some 20 mins later two other
anchor chain parted. Now the rig was 65 to 80 feet off location. At 05:13 hrs
anchor chains 6 and 7 parted after the rig had been hit by an unusually large wave.
Two helicopters were mobilized. The weather continued to build and 20 mins
later the breakaway of the rig appeared imminent and anchor chains 10, 11 and 12
were cut. This action was taken to prevent the overrun of these anchors and
possible capsizing of the rig as a result. Only anchor chain no. 1 was left
dragging. This prevented the rig to drift directly towards the sbm. The rig's mat
tanks scraped over but cleared the <...>'s mooring lines. Once clear of the <...>
The evacuation could start. 48 persons were evacuated while 22 remained
onboard. The rig continued to drift another 1.5 days before being secured by a
towline. The rig had drifted some 27 miles from original position. The rig was
towed to <...> for inspections and replacement of anchor chains.
The breakaway of a production semi submersible with a wire mooring system may not
seem too relevant to most modern custom designed FPSOs. However, the fact that
conditions developed which caused a multiple line failure is reasonably significant in
itself. For the TW58 it was extremely beneficial that the steel risers were designed to
be disconnected in severe weather. This is not the normal situation for as FPS with
flexible risers. Hence, if all the moorings fail on a FPSO which has no dynamic
positioning assistance, the likelihood is that the risers would all be ripped out at either
at the sea-bed or the vessel, or a mixture of the two - see also Section 5.3.
4.1.2 Points of Significance
In summary, the key issues, from an up date perspective, to note from this incident are
as follows:
1. Single line failure followed by multiple line failure led to a complete loss of
station.
2. There was a potential danger of capsize which was prevented by cutting the
remaining mooring lines, which fortunately in this instance, were accessible
from the deck.
3. Subsea damage was likely from a dragged chain and anchor.
4. High potential loss of life.
5. Extended period of deferred production.
6. High collision danger with neighbouring facilities.
67
A4163-01
4.1.3 Fulmar SALM Breakaway
The first Floating Storage Unit (FSU) in the North Sea was installed at Shells Fulmar
field in 1981. The converted 210,658 dwt Medora was yoked by the bows to an
articulated Single Anchor Leg Mooring (SALM) see Figure 4-3 and Figure 4-4.
Due to fatigue cracking the Fulmar SALM broke free at wedges at the base of the
mooring column in 1988, after 7 years on station. Weather conditions at the time,
although severe, were below the mooring survival design criteria. The Fulmar SALM
breakaway made headlines news as can be seen from the extract from the BBC in
Figure 4-5.
At one stage the Fulmar SALM was freely drifting in the North Sea with the anchor
column nodding up and down as the ship tracked across the sea-bed. It is understood
that the 3,700 tonne mooring column made tracks in the sea-bed every time it came in
contact with it. Apparently the column made a seabed depressions each side of the
Forties pipeline, but fortunately did not hit it!
Figure 4-3 Fulmar SALM after Breakaway (courtesy of BBC film clip)
Although the Fulmar SALM is an unusual design, it does have basic similarities to
present day turret moored FPSO. The damage it could have caused if it had collided
with other oil and gas installations while it was freely drifting can be imagined.
The time on station for both the TW58 (6 years) and the Fulmar SALM (7 years) before
problems occurred roughly ties in with the statistics reported in Noble Dentons
UKOOA report. Overall both incidents illustrate the importance of not being
complacent about mooring integrity as systems age.
68
A4163-01
Figure 4-4 Schematic of the layout of the Fulmar SALM
4.1.4 Points of Significance
The key issues, from an up date perspective, to note from the Fulmar breakaway are as
follows:
1. A fatigue failure in the detailed design of a connecting element.
2. A long-term degradation mechanism was involved.
3. The BBC reporting probably led to some reputation damage.
4. There was potential huge subsea damage and pollution damage.
5. There was a real collision risk with neighbouring facilities.
6. An extended period of deferred production resulted in which the complete
FSU was replaced by another vessel.
69
A4163-01
Figure 4-5 - Extract from On this Day BBC Website
70
A4163-01
4.1.5 Weathervaning Navigational Lightships
Weathervaning lightships have been riding out the worst of winter storms since 1732
[Ref. 20]. Lightships have to stay on station whatever the weather and only rarely
come into port for repairs. Figure 4-6 illustrates a typical lightship mooring. Although
lightships are much smaller than FPSOs, it can still be informative to learn from the
considerable SPM mooring experience they have accumulated over many, many years.
Figure 4-6 Illustration of a Typical Lightship Weathervaning Mooring
(courtesy of No Port in a Storm [Ref. 20]
On Lightships it was well known that the most vulnerable point for a mooring line was
where it left the ship, otherwise known as the nip. Here a length of canvas was
wrapped around the chain to reduce wear upon it (see also the CALM buoy rubber
bung on Figure 9-19). Thus the line length was regularly adjusted to minimise wear at
the nip. Also swivels were introduced in the mooring line to reduce the chance of the
cable knotting or kinking.
Figure 4-7 shows the North Carr Lightship, which broke free from her moorings in
1959 and was free drifting while fully manned. This failure led to an ultimately tragic
rescue operation. The left hand side of Figure 4-7 shows the links that failed in the
grounded line section. The right hand side shows a link that failed on a North Sea
FPSO in 1999. As can be seen, there is a certain similarity between the two failure
mechanisms.
71
A4163-01
Interestingly a certain number lightship mooring failures did not occur at the height of a
storm, but a short period after the worst weather had passed. A similar behaviour has
been noted for mooring line failures on semi-submersible rigs.
It is significant that most British lightships were not equipped with any form of
propulsion and hence were at the mercy of the waves if they broke free. Many FPSOs
also do not have any form of propulsion and thus would be unable to steer out of
trouble if they should break free. Although the likelihood of a FPSO totally breaking
free is low, if it should happen the consequences of a collision could be extremely
severe. Since thrusters can help stabilize offloading operations and can help process
equipment motions by adjusting FPSO heading, it can be argued that there are three
good safety reasons for specifying their inclusion on FPSO projects from the outset.
Even in recent times lightship moorings fail with depressing regularity, see Figure
4-9). Hence, all mariners on Light Ships live with the perpetual fear that the moorings
might fail and thus take appropriate safety precautions. It would be highly desirable if
this mariner watchfulness about the integrity of moorings could be more widely
distributed.
Figure 4-7 - Helicopter Rescue from the Free Drifting North Carr Lightship after
Mooring Failure [Ref. 20]
72
A4163-01
Figure 4-8 - Illustration of the North Carr Link Failure Relative to a 1999 North Sea
FPSO Link Failure (fatigue cracking followed by ductile rip out)
Figure 4-9 - Dutch Lightship Number 11 whose Mooring Failed in a Force 10 Gale in
October 1991 which also broke a number of semi-sub moorings see Section 4.2
73
A4163-01
4.2 Multiple Line Failure Incidents
Mooring systems are typically analyzed to check that station keeping can be maintained
in case of one line failure. Thus, it is tempting to conclude that multiple line failure
scarcely ever occurs. However, studying the literature reveals that this is not
necessarily the case. For example, during storms in October 1991 and January 1992
there were 8 semi submersible mooring incidents, of which 3 were multiple failures,
two double and one quadruple. The quadruple failure resulted in the vessel having to
abandon its location. There were so many incidents in these two storms that the HSE
commissioned a special investigation report, OTO 92 013 [Ref. 21].
In both storms adequate warning was provided in weather forecasts available to all rigs.
Both storms were less severe than the design storm, although in the case of the January
storm which affected northerly areas most strongly, only marginally so. It is interesting
to see that semi subs were not alone in suffering during these storms, Figure 4-9 shows
a weathervaning lightship whose mooring also failed and the vessel ran aground.
Given the moorings lessons learnt over the years it might be expected that a semi
submersible which was upgraded in 1997 and was fitted with new mooring lines would
not be subject to multiple line failures. However, this was not the case for the Bideford
Dolphin which in 2000 lost three moorings lines during a storm in the Snorre field.
Coming even more up to date in December 2004 the Ocean Vanguard semi-
submersible lost two of its eight anchor chains and had its riser torn off when drilling
for Eni Norge on Haltenbanken in Norway. The wind was reported to be 43 to 60 knots
and the waves were approximately 17 to 18m in the area where the incident took place.
In this instance it is believed that it was not a mooring line failure but instead the pawls
which lock off the winch failed. This illustrates the key point that all arrangements for
stopping off the chain including connections should be strong enough and able to
tolerate fatigue loading over the life of the field and beyond, if re-deployed, which is
likely with a FPS.
4.2.1 Gulf of Mexico Experience
Incidents of multiple mooring line failures are relatively common in the Gulf of Mexico
during hurricanes [see Ref. 22]. This paper reports that during hurricane Lili the semi
Ocean Lexington broke her moorings and drifted to the beach. In the same storm the
semi Glomar Celtic Sea lost 6 out of 8 of its mooring lines.
In 2003 Typhoon Fay in SE Asia is believed to have caused damage to some of the
Venture series FPSOs, but details have been hard to obtain.
74
A4163-01
Hurricane Ivan, which passed through the Gulf of Mexico in September 2004, is
reported to have left the following in its wake:
- 5 Mobile Offshore Drilling Units (MODU) were adrift.
- One MODU was reported to be leaning about 3 degrees
- The 5
th
generation Deepwater Nautilus semi (see Figure 4-10) broke all her
moorings and drifted 120 kilometres north-east from its pre-storm location. It is
understood that the Rig had a composite fibre rope and steel mooring system.
Full details of the failure are, unfortunately, not in the public domain, but it is
understood that all the failures occurred in the steel wires not the fibre rope. The
steel wires were of 3 (95mm) diameter and each line was designed to
withstand a force of 693t (6,800kN or 1,500kips)
Figure 4-10 - Fifth Generation Deepwater Nautilus Broke free of all her Moorings
during Hurricane Ivan
Results to date indicate that hurricane Ivan did not cause any damage to the mooring
systems of the permanently moored Spars and production semi submersibles in the Gulf
of Mexico. It would, however, be interesting to follow the track of Hurricane Ivan and
to see how close it came to any of the permanent mooring systems. It is also worth
noting that the majority of long-term mooring systems in the Gulf of Mexico are still
relatively new and thus fatigue, corrosion and wear effects will not yet have fully
developed.
75
A4163-01
4.3 Petrojarl 1 Multiple Lines Failure (1994)
The Petrojarl 1 FPSO (see Figure 4-11) experienced multiple line failures while
working on the Hudson Field approximately 60 miles north-east of the Lerwick in the
Shetland Islands in January 1994. Petrojarl 1 was subject to 50 to 55 knot NW winds
and lost two lines at the same time after being hit by a 20 to 25m high wave. Overall 4
out of 8 lines broke over an 8 hour period (lines no. 2, 3, 4 and 7 parted). After the
initial failure production was shut down and the vessel kept on station using her
thrusters and the remaining mooring lines. Petrojarl 1 was never off station and
started reconnecting mooring lines the next day, personnel were not evacuated. It is
worth noting that Petrojarl 1 had the option of quick disconnection of the remaining
mooring lines and risers [Ref. 23]. Planned riser disconnection is not possible for the
majority of FPSOs, apart from those designed to operate in areas subject to ice bergs or
typhoons.
Figure 4-11 - Petrojarl 1 which experienced two broken lines at the same time when
hit by a steep wave
4.3.1 Points of Significance
The key issues, from an up date perspective, to note from this incident are as follows:
1. There was double line failure when hit by a smaller than the design wave.
2. The presence of thrusters prevented an uncontrolled break away.
3. The multiple failures were due to fatigue damage which developed at around
about the same time to a number of mooring lines.
4. Unusually this particular type of turret design allowed rapid identification that a
line failure had occurred.
4.3.2 Overall Message from the Case Studies
Even during the course of this JIP mooring failures have continued to occur both in the
North Sea and also in the Gulf of Mexico. Other case studies are covered through out
the remainder of this report. In general it is clear that a great deal can be learnt from
the case studies which it is believed is highly likely to be relevant to many of the FPS
units operational in the world at the present time.
76
A4163-01
5 CONSEQUENCES OF MOORING LINE FAILURE
5.1 Single Line Failure
As discussed in chapter four, there have been several examples of single line failures on
moored floating offshore structures in the history of the industry. Experience has
shown that line breakages are not restricted to extreme sea-states. Failures have
occurred in both moderate and severe conditions, although the failures in moderate
conditions tend to follow on fairly quickly from storm loading.
Various sources of potential failure have been identified, namely:
- Overload/overstress
- Fatigue in a catenary, at a sheave or connection
- Brittle fracture
- Corrosion
- Wear and abrasion
- Mechanical failure of the mooring line handling/lock off system
In many cases several of the above will interact, for example fatigue damage at a
sheave followed by overload of the remaining ligament under storm conditions. A
potential failure scenario is illustrated overleaf in Figure 5-1.
Although, at present, there are no fixed regulatory requirements for the timescale under
which damaged lines should be replaced, the continued operation of a unit with one
damaged mooring line should lead to re-appraisal of the reduced mooring system using
the higher intact condition safety factors. Where the full design condition cannot be
achieved, consideration should be given to reduced operating conditions. In fact, in the
era of uncertainty following a line failure, it can be argued that the safety factors on the
remaining lines should be raised until it has been determined with reasonable
confidence why the original line failed.
Revised operating conditions could be identified prior to loss of a line, permitting a
more rapid assessment of the damaged condition. It is recommended that a summary of
reduced environmental parameters for production should be available on board all
operating FPSs in case of line failure.
This single line failure condition falls within the normal design criteria for the mooring
system and should not threaten the integrity of the unit directly. Under certain
conditions the falling chain could damage sub sea infrastructure. For example, a
flexible could be swept below mooring lines under high current conditions, exposing it
to impact. This could be of most concern to a gas riser and should be considered as
part of the detailed design process.
For specific units in deep water hydrocarbon conditions could cause riser blockage due
hydrate or waxing in case of an extended shutdown. If a unit from a process point of
view cannot tolerate shut down, it will be cost effective in the long-run to add in
additional contingency in the mooring system. This should be specified in the design
brief, see Section 20.
77
A4163-01
The progressive deterioration of a
Deterioration
component of the system under fatigue,
corrosion or wear.
Failure Followed by failure of the component
under moderate storm conditions.
Line loss might be detected through tension
Detection
monitoring equipment where that is
installed. It is possible that the line failure
could be undetected until a routine subsea
check of the mooring system.
The system is likely to be shutdown until
Shutdown
the continued integrity of the mooring
system has been verified and new operating
limits defined.
The mooring and production systems
Inspection
would be inspected to identify any related
damage.
Reduced operations
Resumption of operations under
reduced weather criteria.
Repair Reinstatement of the full mooring
system.
Figure 5-1 - Summary of a Single Line Failure Scenario
78
A4163-01 79
5.2 Multiple Line Failure
The failure of several mooring lines could overload remaining lines in sequence,
resulting in loss of position of the unit. Multiple failures in the mooring system are
ranked as safety critical risk category 1 which is the highest category [Ref. 23].
Y
Z
20 m
g_ g _ _ _ _ _ q _ _ j _ ( ) ( ; ) p
Figure 5-2 Illustration of Riser Stretch After Loss of Position Following Mooring
Line Failure
There have been several instances of multiple line failure of a fixed floating unit.
Factors which contribute to the likelihood of multiple (as opposed to single) line failure
include the following:
Design: the presence of a systematic weakness in the mooring system will
apply to all lines, increasing the likelihood of multiple line failure.
Age: fatigue, corrosion and wear will tend to deteriorate all mooring lines,
particularly in the same quadrant, to roughly the same extent over
time.
Detection: where no line tension or equivalent monitoring system is available,
failure of a single line may go undetected (see Figure 5-4). This may
expose the remaining lines to higher loads for an extended period.
Mooring system technology continues to change as operations move into deeper water
and new techniques are developed for more marginal fields. The extension beyond
proven technology can introduce unexpected problems. As the existing fleet ages,
fatigue, wear and corrosion will become more significant, again exposing any design
weaknesses in the mooring systems.
A4163-01
A potential multiple line failure scenario is illustrated in Figure 5-3. Figure 5-4
illustrates how prompt detection of line failure and good quality inspections can reduce
the danger of multiple line failure.
Deterioration
The progressive deterioration of a type of
component under fatigue, corrosion or
wear.
1
st
Failure
Followed by failure of the component
under moderate storm conditions. This
could go undetected
Unzipping
Overload of adjacent lines, perhaps after
further deterioration if the initial failure
was undetected.
Excursion Loss of integrity of the mooring system
could be identified from the loss of station
keeping after failure of several lines.
Shutdown The risers should be de-pressurized and
isolated prior to damaging distortions in the
system.
Figure 5-3 - Potential Multiple Line Failure Scenario
80
A4163-01
In the event of some combination of circumstances leading to rupture of a riser without
depressurisation, for instance where initial mooring line loss was undetected; the rate of
gas / oil escape would be much larger. Rupture at the base of the turret could result in a
significant release of gas or oil close to the FPSO. If the unit is still producing and
sources of ignition such as the flare are lit, a gas explosion could result. This illustrates
the importance of early detection of mooring line failure. It is also conceivable that
wave grouping and rogue waves might cause multiple line failures to an intact system if
it is under designed or if the integrity of lines from a particular sector has been
compromised by wear, corrosion and fatigue. Petrojarl 1 appears to have lost two
lines at the same time when she was struck by a large wave in 1994, see Section 4.3.
The formation of a large gas bubble underneath the FPSO itself could potentially affect
the stability of the whole unit.
5.4 Business Interruption Consequences - Two Case Studies
Clearly the commercial impact of shutting down production and repairing the mooring
system would depend upon both production rates and the extent of damage.
Scenario 1 Medium Sized North Sea FPSO producing 50,000bpd
This scenario assumes the following after 7 years on station:
- Loss of one mooring line and as a result shutting down for 2 days to
identify the extent of the damage
- Dive Support Vessel (DSV) mobilisation to survey subsea infrastructure
(risers, condition of remaining lines, etc.) and to support the line repair
- Mobilisation of 2 x anchor handler tugs (AHTs) to reinstate the damaged
line plus a FPSO heading control tug.
An indicative costing for the above incident has been worked out inTable 5-1. Note it
is assumed that suitable spares, including connectors, are available. Hence the capital
cost of spares and replacements has not been included.
82
A4163-01
Item Calculation Value Notes
2 x 50,000 Assuming
Deferred Prdn x $25 1,470,000 1.7$ to 1
(2+2) x
DSV 70,000 280,000
Heading (2+2) x 2 days mob /
control tug 15,000 60,000 demob
2 x (2+2) x plus 2 days
AHTs 15,000 120,000 diving
1,930,000
Total 2M
Table 5-1 - Line Failure Cost Estimate, 50,00bpd North Sea FPSO
The $25 per barrel rate for oil is based on deferred production cost and will be field
specific depending on operating costs, etc. It has been estimated using the following
typical formula:
Value of deferred production = Deferred volume x Margin x Discount factor
The following terminology applies:
Margin = the prevailing oil price less the production facilities cost of delivery
including all appropriate costs (depreciation, variable lifting and transportation
costs, etc.)
Discount factor = 1/(1+discount rate)
n
The Discount rate has been taken to be the fairly industry standard level of 10%
and n is the period in years.
If one assumes that the present oil price is about $45/barrel and that the lifting or
recovery cost is approximately $10/barrel, the price per barrel of the deferred
production in n years time is as follows:
Deferred production cost ($/barrel) = (45 10) x 1/(1.1
n
)
Hence, if a line failure occurs in year 7 and the anticipated field life is 20 years the
calculation becomes:
Deferred production cost in 20 years = 35 x 1/(1.1
20-7
) $10/barrel
Value of product today $45 - $10 = $35/barrel
Lost value in deferring production $35 - $10 = $25/barrel
83
A4163-01
Scenario 2 Large West African FPSO Producing 250,000bpd
This scenario assumes the following:
- Loss of one mooring line and as a result shutting down for 2 days to
identify the extent of damage
- DSV mobilisation to survey subsea infrastructure (risers, condition of
remaining lines, etc.) and support line repair.
- Mobilisation of 2 x anchor handling tugs to reinstate the damaged line. It
is assumed that the unit is spread rather than turret moored and hence a
heading control tug is not required.
An indicative costing for the above incident has been worked out Table 5-2. Again it is
assumed that suitable spares including connectors are available. Hence, the capital cost
of spares and replacements has not been included. It is worth noting that the
mobilisation time and costs are significantly higher for a West African FPSO than for a
North Sea FPSO. In practice it may be the case that suitable vessels are available
locally, but this cannot be relied upon.
Item Calculation Value Notes
Deferred Prdn 2 x 250,000 x $25 7,353,000 Assuming 1.7$ to
DSV (30+2) x 70,000 2,240,000 30 days mob / demob
plus 2 days diving/repair
operation
AHTs 2 x (30+2) x
15,000
960,000
Total 10.5M
Table 5-2 - Line failure Cost Estimate, 250,000bpd West African FPSO
Two simple conclusions can be drawn from the above calculations.
- Financial costs associated with mooring line failure are large, particularly
relative to the capital cost of the failed component.
- Both lost production and vessel costs are significant.
Where the platform is remote from the main offshore operating centres, deployment of
suitable vessels may take several weeks. If the integrity of FPSO or subsea
infrastructure after the incident cannot be demonstrated using local resources, a lengthy
shutdown may be required. A long shut down would increase the cost of deferred
production dramatically.
Repair of the system may require the procurement of special connectors or replacement
line segments. These may have lead times of 4 to 6 months, requiring medium term
operation either with one line down or a short term connector solution see Section 19
on contingency procedures and spares. The cost of the repair will increase if two repair
mobilisations are required, in other words an initial short term fix followed by the long
term repair.
84
A4163-01
6 HANDLING, TRANSPORTATION/TRANSFER AND
INSTALLATION
6.1 Transportation/Transfer
Mooring lines are not simple items to transport due to their length and weight. Any
damage to lines during transportation of transfer can have serious implications for long-
term mooring integrity. Manufacturers typically have detailed instructions for
transportation and transfer of their products and these instructions should be followed
to the letter. Poor practice during transportation and handling potentially can destroy
project schedules.
Figure 6-1 gives a good indication of the great care needed while handling fibre ropes
with careful level winding and proper back tension when the ropes are installed on the
reels.
During transportation, transfer to installation spools and installation of sheathed wire
rope, particular care must be taken to ensure that sheathing remains undamaged.
Figure 6-1 - Spooling Fibre rope onto a Powered Reel from Standard Containers
[Ref. 24]
85
A4163-01
6.2 INSTALLATION OF MOORING LINES AND CONNECTORS
The installation procedures for FPS moorings are highly dependent upon the design of
the system, the water depth and the available installation vessels. There are several
issues in the deployment of FPS mooring lines additional to those encountered with
mobile offshore units.
x A FPS mooring system must operate continuously with restricted inspection
and maintenance over a long period. Installation damage to protective
sheathing, connectors and mooring lines can cause fatigue and corrosion
problems over time.
x Many FPS units operate in deep water locations. As a result a complex line
make-up including wire rope, man made fibre segments and occasionally mid
line buoys may result.
x The FPS requirement for the mooring system to support extreme conditions
without moving off station can lead to large heavy mooring system
components (Figure 6-2 gives an indication of the manual handling issues). It
is clear that items such as bend stiffeners can be relatively easily damaged.
This has been seen on recovery of a North Sea FPSO mooring system at the
end of a relatively short deployment see Figure 6-6.
Figure 6-2 - Illustration of the Weight and Handling Issues Associated with Mooring
Components (Courtesy of Stolt Offshore)
86
A4163-01
6.2.1 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.
To date non-straight mooring lines have been noted on two North Sea FPSOs see for
example Figure 6-3. 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. However, so far, these FPSOs have not yet
experienced storm line loadings as severe as the maximum loadings evaluated during
the mooring design process. Analysis results indicate that with a tension of 500 tonnes
it will not be possible to drag a dog legged mooring line through the mud. This seems
to tie in with what happened offshore during pre-tensioning. However, it should be
noted that this is an area of uncertainty, since there is a lack of data to assist with, for
example, selection of axial and normal drag coefficients.
Figure 6-3 - Red Arrows Show Examples of Mooring Dog-Legs
87
A4163-01
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 occurs the implications for mooring line tensions and fatigue loading
should be re-evaluated. If Dog Legs are still present at the end of the field life, this
could indicate some conservatism in the design process, particularly if during this time
the FPSs have experienced survival conditions.
6.2.2 Torque Implications for Mooring Line Installation
The design of a mooring system requires consideration of the potential for torsion in the
lines. This includes the behaviour of each component with respect to imposed tension
and torsion. Hence, during the installation of heavy components in a chain or spiral
strand system, consideration must be given to the introduction and control of torque.
Application of tension to a wire rope will tend to straighten the individual rope fibres,
resulting in either rotation or a corresponding restraining torque. The torque developed
is approximately proportional to applied tension and wire rope diameter. For an
ordinary lay six strand steel independent wire rope core (IWRC), Bridon Ropes quote
the following expression for torque developed under tension.
Torque 07 . 0 Diameter Tension
More complete expressions, taking into account twisting of the rope, the increased
torsional stiffness of rope under tension and even cross terms between these various
components are given by Chaplin [Ref. 25]. The simplest of these expressions is given
below.
A Rotation
Torque 085 . 0 Diameter Tension +
Length
A where 000531 . 0 G Diameter
4
+ 187 . 0 Tension Diameter
2
Newton
G 000 , 75
2
mm
The first term is equivalent to the coefficient presented by Bridon. The coefficient on
rotation represents the geometric torsional stiffness of the wire rope, plus an additional
term reflecting interaction between tension and torsional stiffness.
It should be noted that the numerical coefficients listed above vary even within the
reference quoted above, and as such predictions of the numbers of turns for a given
condition should be treated with care.
88
A4163-01
The four different types of line segment used in the deployment and operation of
mooring lines exhibit very different torsional characteristics. Both the torsional
stiffness and the tension induced torque vary.
- Six (or eight) strand wire rope is not torque balanced. This means than
when an axial tension is applied, torsion is developed in the line. The
greater the tension, the larger the torsion.
- Chain demonstrates little torsional stiffness for low levels of rotation, very
high stiffness for greater rotation (in excess of 3 degrees per link). A chain
with no twist will not develop torsional moments under tension.
- Spiral strand wire rope presents a relatively high torsional stiffness. It is
essentially torque balanced, developing a much reduced torsional moment
than the corresponding six strand rope. If a torque is applied to spiral strand
wire it can easily become damaged see Figure 6-7.
- Polyester rope has a low torsional stiffness, due to the small diameter of
individual fibres. There is little tendency to develop torsional moments
under tension.
In the design of mooring systems, consideration must be given both to the interaction of
the individual components in the operating condition, and to the implications of this
during installation. It is very important when deploying chain that no twists should be
included, but in practical terms for a long length of chain this is not simple to achieve in
practice.
Clearly, where one component has a tendency to rotate and develop line torsion, this
may result in the twisting of adjacent components. Each line type has different issues
associated with the imposition / release of torsional loading.
- Where six or eight strand wire rope is subjected to dynamic axial loads with
no torsional restraint it will rotate. The combination of tension and rotation
is much more subject to fatigue than tension cycling with ends restrained.
There may also be issues associated with the whirling of heavy fittings or
adjacent chain segments increasing damage rates.
- The performance of chain when subjected to torsion plus tension is not well
understood. Where line tension drops below a limiting value there is some
possibility of knotting of the chain, which will reduce strength and fatigue
resistance. Under significant tensions chain is able to accept small levels of
rotation without apparent damage.
- Spiral strand wire rope is both relatively stiff in torsion and sensitive to
damage when twisted. This damage occurs due to slippage between layers
of (torque balanced) wire. In extreme cases this can develop into hockles,
where the lay of the wire is so distorted that some wires twisrt right away
from the body of the rope see Figure 6-5 and Figure 6-7).
Fibre ropes appear to be able to accept quite large levels of rotation without a
significant impact on their performance.
89
A4163-01
Figure 6-4 - Illustration of Twist on a FPSO Mooring Line during Recovery
Figure 6-5 Illustration of a Hockle in Spiral Strand Wire during Recovery of a
FPSO Mooring System
90
A4163-01
Figure 6-6 - Example of Damage to the Bend Stiffener on an Open Socket
Figure 6-7 Illustration of Spiral Strand Wire Kinking during Installation
91
A4163-01
6.2.3 The Use of Swivels
Where it is desirable to prevent the imposition of torque on mooring line segments, in
line swivels may be used. There are two main types of offshore mooring swivel. Slide
bearings provide a robust, low maintenance torque release, but will only operate under
quite high torque levels. Roller bearing swivels provide a low friction torque release,
but require maintenance.
Swivels are sometimes used during the installation of deepwater moorings to avoid the
introduction of twist in a heavy chain or spiral strand line segment.
If a mid-line buoy is to be used in a mooring line, a connection link such as that
illustrated in Figure 6-8 may be used. This connector allows the central section which
is attached to the buoy to rotate but the padeyes on either side for the main mooring
legs are fixed relative to each other. This type of swivel is intended more for use in a
permanent mooring system as opposed to as a temporary installation measure.
Figure 6-8 - Mid Line Buoy Swivel Connection Link (courtesy of MoorLink AB).
Mid line buoys can result in greater relative rotation at the connections which in certain
instances has been known to lead to premature failure see Section 10.3.2. Therefore,
the use of mid line buoys should be treated with caution.
6.2.4 Pre-Installing Mooring Lines
It is often desirable to pre-lay mooring lines. This permits location and securing of
anchor points prior to the arrival of the FPS. Separation of the installation programme
into discrete segments reduces the vulnerability of the programme to weather windows
and removes these operations from the critical path.
92
A4163-01
Deploying an anchor pile, chain, spiral strand wire or polyester rope as one operation
can be problematic, unless a high specification construction vessel is employed and
great care is taken. Potential difficulties include:
- Risk of rotation, possible interference and damage
- Difficult to control simultaneous lowering of multiple handling systems
- Difficult to reverse the process.
However, a reliable subsea connector is required if lines are going to be pre-installed.
If the subsea connector is not reliable, over time a weakness may be introduced into the
system. Assuming a reliable subsea connector is available its use may help with
respect to possible mooring line repair operations which may be needed at some stage
during the field life. It is important, to minimise relative rotation and wear, that the
weight per metre of the connector should not be too much higher than that of the
mooring line to which it attached.
6.3 Installation Watch Points from a Mooring Integrity Standpoint
Over the last few years there have been a number of notable deepwater projects
completed in the Gulf of Mexico, which have used either spiral strand wire or fibre
ropes. Very useful experience has been gained form these projects. This section
attempts to summarise some of the key lessons learnt.
Suction Pile Rotations
It is important that the orientation of the padeye lines up with the mooring line direction
when it is tightened up. Cases have been reported of suction piles rotating as they are
sucked into the sea-bed, which can be problematic.
Spiral Strand Wire
The key watch points are:
- Requires handling within tight tolerances for twist, friction and compression.
- Sheathing can be easily damaged.
There have been several cases where spiral strand has been irretrievably damaged
(usually through kinking) during installation.
Polyester / Synthetic
The key watch points are:
- Large diameter ropes, have a large storage volume requirement.
- Multiple spooling operations from storage reels to installation winch are
normally required.
- The outer braiding layers are susceptible to damage.
93
A4163-01
Although polyester is a durable material, the braided jacket and even the core, can
be subject to damage during installation if not properly handled, much like sheathed
spiral strand wire [Ref. 26].
At present it is customary not to allow polyester rope to come in contact with the sea
floor due to concern that particle ingression will cause harmful abrasion of the fibres.
With the introduction of soil particle filter clothes just under the jacket this may be no
longer necessary [Ref. 26] but at present it is customary to adhere to and this will
impact the installation procedures. Balmoral Group Norways experience with
MODUs and fibre ropes indicates that this may not be required. However, it is still
difficult to know what would happen during a true long-term deployment.
Installation Ground Rules
It can be helpful to provide Installation Contractor with succinct ground rules for
installation including any special considerations, e.g. handling of polyester ropes, such
as:
- Limits on twist
- No sea-bed contact
- Acceptable means to handle and stopper
- Temporary storage and transport requirements
- Contingency measures
Past projects have successfully utilised a management of twist procedure to identify
how twist will be monitored, assessed, recorded and summed up over a mooring line. In
particular, it is important to specify low torque or torque balanced wires for messenger
line or slings during installation. Twist can be monitored by a ROV viewing a pre-
painted stripe onto the mooring chain and a colour stripe marker built into the polyester
ropes jacket during the manufacturing process.
Petruska reports [Ref. 26] Installing a polyester mooring system is similar in many
ways to installing a sheathed, spiral strand wire system when using similar/identical
installation vessels, but a few differences do exist. For example sheathed, spiral strand
has special requirements on minimum bending radius and the associated tension in
order to prevent damage to the sheathing and also to prevent kinking wire strands.
Both have limitations on twist, although different, since spiral strand is not perfectly
torque balanced while polyester ropes can be made to be torque neutral. On the Mad
Dog project two complete twists (i.e. 720) per mooring line were permitted for the
fibre rope.
Although polyester weighs much less both in air and in water, it does take up more
volume, which needs to be taken account of during installation.
94
A4163-01
6.3.1 Polyester Rope Line Length Implications
Polyester rope lengths can vary and it is important to understand the different
categorisations namely:
- Manufactured length,
- Installed length at the specified pre-tension,
- Lengths expected at various phases during the installation.
The total variation through out this process may vary as much as 50 to 100m. Short
Term Creep and Long Term Construction Stretch may lead to a need for the rope being
manufactured somewhat shorter than its final required length.
Common practice calls for polyester to never come into contact with sharp edges, high
heat or steel work wires. It is vital to ensure all equipment free of sharp edges and
where necessary to use special padding material such as burlap or lamiflex to
further aid in protecting the rope jacket from snags and tears.
On fibre rope moorings the majority of fibre rope creep should occur in the first year of
service. This creep is likely to result in a requirement to re-tension the mooring system.
On the Red Hawk Spar there is no requirement for spar offsetting for well drilling or
maintenance operations. Hence a single chain windlass located at one position on the
Spar deck with fairleading access to the six mooring stations was assessed to be
sufficient for pre-tensioning and mooring line adjustment purposes if required. This
single chain windlass was integrated into the topsides rather than at a dedicated
winching deck as on previous spars. To reduce the necessity of future line length
adjustments, some of the fabrication stretch was removed as illustrated below. This
required application of a tension level of 40% of the MBL for 1 hour, namely
approximately 500 t. The geometric amplification provided by this means seems to be
capable of achieving such a tension. It appears that this method of tensioning up the
lines is not very precise and there must be a danger of increased dynamic loading of the
tensioning tow line due to tug motion/changes in tow line angle. Hence it will be
interesting to see how such lines perform in situ.
95
A4163-01
Figure 6-9 Pre-Stretching Polyester lines During Installation to Minimise the
Requirement for Future Line Length Adjustments [Ref. 27]
Fibre Rope Protection
It is important to limit fibre rope exposure to ultraviolet light by the use of lamiflex
sheeting and tarpaulins. Also there should be no welding or flame cutting in the
vicinity of fibre rope. Hence there may be a need for bolted clips for sea fastening the
rope reels to transportation cradles.
Fibre Rope Connectors and Thimbles
The design of connectors for use with fibre ropes is still evolving. Figure 6-10
illustrates one design that has been used in the Gulf of Mexico.
Such connectors need to be designed to simplify offshore lining up of pins. For
example in Figure 6-10 the H-link is not a true H-link in the sense that the two face
plates are not rigidly connected. This allows differential movement of the two plates
which can cause problems with alignment and getting the pin back through especially
at the hang-off platform under load see Figure 6-10.
The illustrated design includes a thimble which should take most of the wear. But
tight fits may still be encountered offshore, since the polyurethane protective coating
around the eye of the polyester splice is manually applied. Also when spreading the
eye of the polyester splice to insert the thimble, tearing of the polyurethane would often
occur in the crotch region, thus either special care/an improved procedure is required or
the polyurethane needs to be reinforced.
96
A4163-01
Not all fibre rope connectors have made use of thimbles so it will be interesting to see
if over time wear/abrasions becomes an issue. Unfortunately inspection access to this
area is difficult see Section 18.6.1/Figure 18-15.
An important point for the use of fibre ropes is careful labelling for
identifying/avoiding confusion on fibre rope segments.
Figure 6-10 - Illustration of the Potential Difficulty in offshore alignment of pins on
large Diameter Rope [Ref. 26]
6.3.2 Overboarding Operations
Overboarding of heavy items (anchors, sockets, etc.) may need special protection as
may be provided by a sledge arrangement see Figure 6-11 for example which shows
an H shackle launch. The sledge can be recovered using a work wire on to a capstan
winch after the heavy item has been deployed over the stern roller.
Figure 6-11 - Sledge used to Protect H Connector during Deployment over the Stern
Roller (Courtesy I. Williams)
97
A4163-01
6.3.3 Anchors
Once anchors have been installed and successfully pre-tensioned on FPSs they seem to
have proved reliable in situ. The difficulties which have been experienced in the field
are typically when soil conditions turn about to be different than predicted. Hence, it is
desirable to collect sufficient soils information prior to the FPS deployment.
If project schedule and vessel availability allow, it is recommended that the following
site survey work should be undertaken prior to installation:
- Carry out bore hole soil sampling at two locations on each mooring line.
- The first location should be the anticipated anchor landing point.
- The second location should be the predicted final anchor position.
In certain instances only limited borehole data may be available. In such cases it makes
sense to be on the conservative side when selecting the size and weight of the proposed
anchors. Anchor steel is relatively cheap compared to the day rate of installation
vessels!
When a drag anchor is installed it is very difficult to determine the depth of the sea-bed
penetration. This can make accurate determination of line pretension difficult if, during
installation, the length of all the mooring line sections was carefully noted on the basis
that this can be used to back calculate the pre-tensions.
Drag anchors normally have minimal corrosion protection, just a basic paint coating.
Despite this corrosion has not been a problem even for anchors on drilling rigs, which
have a much harder life than an anchor which sits deep into the sea-bed. Still given that
field lives can be extended and that high quality coatings are available, it would seem
logical to make greater us of such coatings.
Drag anchor fatigue life is typically far superior to that of the chain, which they are
attached to. Hence, anchor fatigue life is normally only checked if specified by the
anchor manufacturers client.
98
A4163-01
7 CORROSION, FATIGUE AND WEAR (CASE
STUDIES)
7.1 The Balmoral FPV An Industry Benchmark
The Balmoral Floating Production Vessel (FPV) represents an early North Sea semi-
submersible production unit (see Figure 7-1). Unusually for the time, it was a purpose
built production unit utilising a new GVA design and was built in Gothenberg in 1986.
Hence, today (2005), it has been in continuous operation without dry docking for some
19 years.
It is also worth noting that the Buchan, Amerada Hess 001 semi-submersibles and
the Brent Spar have also seen long deployment periods. Some of the experience
which has been gained from these units is discussed in Sections 8 and 11.
The Balmoral FPV was provided with a Rolls Royce mooring system consisting of
driven anchor piles and 92mm R4 studded chain made in accordance with the new
DNV standard to avoid brittle failures. The chain, when new, had a minimum break
load (MBL) of 853t MBL. In addition, the FPV has 4 x 39 tonne maximum nominal
thrust azimuthing thrusters, which are used in storm conditions to reduce mooring line
tensions.
Figure 7-1 The Balmoral Benchmark FPV which has been continuously on station
since 1986 (Courtesy of CNR)
Despite some built in redundancy the FPV has experienced a number of line failures
which are summarized in the plan view in Figure 7-2.
99
A4163-01
Chart of the FPV Mooring System
N
o
r
t
h
i
n
g

6456500
6456000
6455500
6455000
6454500
6454000
6453500
Kenter links
Touchdown
Missing or
loose studs
D-shackles
Pile 1
+2.7%
Pile 2
No measurements
Pile 3
-1.9%
Pile 4
+11.6%
Pile 5
+7.5%
Pile 6
No measurements
Pile 7
+33.5%
Pile 8
-25.7%
563500 564000 564500 565000 565500 566000 566500
Easting
Figure 7-2 Plan View of Mooring Incidents at Balmoral
Historically, Balmorals mooring lines were inspected and the studs pressed every 5
years on the back of an AHT. In 2001 one of the most heavily loaded windward lines
was recovered and taken to Haugersund for detailed inspection by Chainco. Every
other link was examined. Just one crack was discovered on the outer shoulder of a
single link which was thought to be a random manufacturing problem
One section of the line had very loose studs and this was cut out and transferred to the
chain locker on board the FPV. On this basis DNV accepted Welaptega Marines (see
Section 18.4) in water ROV inspection for the other lines, rather than inspection of each
line on the back of an anchor handler.
However, in November 2002 a leeward line broke. Despite a drop in the reported
tension it took time to confirm that the line had definitely broken. This was because the
break was in the mud and the line still had some catenary profile. Hence, one option
was a partial line run out. The break was only confirmed when the line was pulled in
on the chain windlass and a ROV saw the chain end emerge from the mud. This has
definite implications for possible line failure detection methods see Section 17.
100
A4163-01
7.2 Corrosion and Wear Allowance Discussion of Code Requirements
For long term integrity it is vital that wear and corrosion are correctly accounted for in
the design process. This section reviews the existing guidance which is available in
mooring design codes/recommended practices and then compares the specified values
with what has been recorded in the field. It is worth noting that some earlier mooring
systems were designed with no corrosion allowance. This ties in with what was
specified in the design codes available at the time, e.g. POSMOOR code of July 1989
[Ref. 29]. Hence, if these early systems did not include much design margin (i.e. they
just met their allowable loads) then wear and corrosion may fairly quickly cause a
reduction in their capacities, such that they no longer meet their allowable loads. In
some instances the safety factors in some of the earlier codes may have been higher,
which thus by default effectively included some in built allowance for wear/corrosion.
In a similar vein fatigue life calculation were not required by POSMOOR 1989.
In the absence of alternatives API RP 2I [Ref. 30] is sometimes applied to long term
FPS moorings. API RP 2I has universal allowable reductions in chain diameter (see
Section 7.5.3). These may not be appropriate for a long term FPS which does not have
an inbuilt allowance for corrosion and wear. In such cases a new evaluation of the
worn chain break strength should be undertaken. However, as is discussed in Section
7.3.3 an accurate assessment of the strength of worn chain is difficult to determine.
Draft ISO standard (19901-7) Part 7, Section 10.6 [Ref. 31] states for chain in the
splash zone or in contact with a hard bottom sea-bed the diameter should be increased
by 0.2mm to 0.8 mm per year of the design service life. The 0.8 mm per year is a
significant increase compared to other codes.
API RP 2SK Section 3.1.2 states the following Protection against chain corrosion and
wear is normally provided by increasing chain diameter. The allowance in chain
diameter for corrosion and wear is a complicated issue that still requires significant
research and service experience to address. Currently industry practice is to increase
the chain diameter by 0.2mm to 0.4mm per service year in the splash zone where
oxygenated water tends to accelerate corrosion and in the dip or thrust zone on hard
bottom where heavy corrosion takes place.
101
A4163-01
OS-E301, Table F1 Corrosion & Wear Allowance for Chain
Corrosion allowance referred to the chain diameter
Part of Mooring
Line
No Inspection
(mm/year)
Regular inspection
1)
(mm/year)
Requirements for the
Norwegian
continental shelf
(mm/year)
Splash zone
3)
0.4 0.2 0.8
2)
Catenary
4)
0.3 0.2 0.3
Bottom
5)
0.4 0.3 0.4
1) Regular inspection e.g. in accordance with the Classification Societies or
according to operators own inspection programme approved by national
Authorities if necessary. The mooring lines have to be replaced when the
diameter of the chain with the breaking strength used in the design of the
mooring system is reduced by 2%.
2) The increased corrosion allowance in the splash zone is required by
NORSOK M-001 and is required for compliance with NPD regulations.
3) Splash zone is defined as 5m above the still water level and 4m below the
still water level.
4) Suspended length of the mooring line below the splash zone and always
above the touch down point.
The corrosion allowance in the Table is given as guidance; lower values
may be accepted provided it is documented.
Table 7-1 - Example of Specified Corrosion and Wear Allowances from One
Classification Society
Section 59.2.2 (Concentrated Corrosion) of BS6349-1 2000 [Ref. 28] defines
accelerated or localized corrosion as concentrated corrosion. Relevant factors
include:
1) Repeated removal of the protective corrosion product layer.
2) Bi-metallic corrosion, where steel is electrically connected to metals having
nobler potentials or where weld metals are significantly less noble than the
parent metal.
3) Accelerated corrosion associated with microbiological activity.
In such circumstances typical corrosion rates of 0.5mm/side/year and as high as
0.8mm/side/year have been observed in the relatively cold UK coastal waters.
102
A4163-01
7.2.1 Wear/Corrosion Rates Experienced in the North Sea
Figure 7-3 Illustration of the Extent of General Corrosion on a Recovered Floating
Production Unit Mooring Line after 16 years service
The data considered in this section is based on a North Sea semi submersible based
floating production facility that has been continuously operating for almost 20 years.
On this unit a line failed in the thrash zone and a number of links close to the break
were recovered back to shore. Dimensional checks of the most worn areas of this chain
revealed 10mm of apparent wear/corrosion over 16.25 years based on the nominal
chain diameter. This gives a wear rate of 0.615 mm/year in the thrash zone. As can be
seen this is 50% higher than the value specified in OS E301 [Ref. 5]. If a higher wear
rate is experienced than has been allowed for it is possible that after a set number of
years the mooring system will no longer be capable of with standing the maximum
anticipated storm loading. Since all the lines will be subject to wear, although not
necessarily at the same rate, this could mean that if one line fails the remaining lines
may no longer be strong enough to withstand the one line failed design case. In such a
case there is a real danger of the mooring system starting to un zip itself and the unit
loosing its station keeping capability.
103
A4163-01
Figure 7-4 Illustration of the Extent of Corrosion Pitting
Figure 7-3 is a picture of some of the links which were recovered following the line
failure. As can be seen the chain has experienced fairly heavy corrosion. In addition,
Figure 7-4 shows the extent of corrosion pitting. The materials testing laboratory
which examined the chain reported the following:
The metal loss observed on all the links took the form of large areas of
pitting where the metal loss was at least 2-3mm, with isolated areas of deeper
pits with more severe metal loss. The entire outer bend region of some links
was affected in this way, as well as large areas of the straight sections.
A somewhat unexpected result from the examination of the links recovered from the
thrash zone was damage to the crown of the links see Figure 7-5. It is believed that,
as tension is cyclically reduced, some type of impact or grinding action on the on inner
edge of an adjacent link seems to be occurring (see Figure 7-6). It is also worth noting
that the chain had very loose studs, hence it is possible that contact between the crown
and the stud is occurring. However, there was no particular evidence on the stud itself
of such a contact happening.
104
A4163-01
Figure 7-5 Example of the Damage Caused to the Crown of the Links
Figure 7-6 Arrow shows the Apparent Grinding Action on the Inner Face of One of
the Links
Another example of how the dynamic action of a moving link may cause damage to an
adjacent item is shown in Figure 7-7. This photograph shows the beginning of a failure
of a small hanging shackle which attaches an excursion limiting weighted chain section
to the main links of a FPSO mooring line. The failure of the hanging shackle pin is
likely to have been caused by the dynamic pinching action of the adjacent link plus the
general rotation of the hanging shackle pin see also Section 10.3.
105
A4163-01
Figure 7-7 Example of the Damage Caused to a Hanging Shackle Pin on a FPSO
Mooring Line
When a chain is subjected to an applied load it is subject to a complex combination of
tension, bending and shear loads. A finite element derived indicative stress pattern for
a loaded link is shown in Figure 7-8. In this plot the highest stresses areas are coloured
red. Comparing Figure 7-8 with Figure 7-6 shows that the area of apparent grinding
damage approximately corresponds with one of the areas of maximum stress (see also
3.3.3). Hence damage in this area could result in a relative rapid reduction in break test
capability. Another related factor here is the effect of corrosion pitting which in certain
cases can be in excess of 3mm (see earlier).
Figure 7-8 Finite Element Stress Contour Plot (compare red areas with Figure 7-6)
[Ref. 8]
106
A4163-01
In Figure 7-8 it should be noted that the stress contours show a weak asymmetry
about the X-Y plane due to simply supported constraints applied to the static end of the
chain model and a static load applied to the dynamic end.
7.2.2 Wear in the Thrash Zone
The data in Figure 7-9 is based on a detailed measurement programme on a line which
was bought back to shore after many years of use on a North Sea semi-sub FPS.
Although there is quite a lot of scatter, the black poly line on the graph indicates
maximum wear at the touchdown point indicated by the red dashed vertical line.
80
82
84
86
88
90
92
94
650 850 1250
946m
Chain Thickness vs Link Number
450 1050 1450 1650
Link Number (for line 7 section)
C
h
a
i
n

T
h
i
c
k
n
e
s
s

(
S
p
e
c
i
a
l

M
e
a
s
u
r
e
m
e
n
t
/
2
)

(
m
m
)

Line 7 Chain Thickness API Minimum Line 4 Thickness Poly. (Line 7 Chain Thickness)
BREAK POINT TOUCH DOWN @ 128Te
FPV PILE
835m 761m 687m 909m 983m 1057m
Distance from Pile (m)
Figure 7-9 - Example of Thrash Zone Wear
107
A4163-01
7.3 North Sea FPSO Apparent Corrosion and Wear Data
This section is based on recent measurements on components which were recovered
from a North Sea FPSO mooring system. Figure 7-10 and Figure 7-11 show the
condition of a special shackle. The shackle was positioned approximately 40m out
from the wildcat at the base of the turret at the transition from studless to studded chain.
This shackle is much more pitted than expected considering that it was in the water for
less than 7 years. It appears that some degree of galvanic type or perhaps sulphate
reducing bacteria (SRB) induced corrosion has taken place. Galvanic type corrosion is
perhaps more likely given the localised deterioration of the shackle pin (see Figure
7-11) where it has been in contact with studded chain.
On the studded chain heavy pitting of 2 to 3mm depth was noted where the chain
connected to the shackle. This localised effect again supports the hypothesis of galvanic
type corrosion between the shackle and the chain, which has affected some 26 links
before the effect is dissipated. Pits can act as stress raisers see Section 7.5.
Given that there is a length of studded chain without pitting between where the pitting
has been observed and the fairlead, there is some evidence that this pitting is not due to
any cathodic action from the FPSO itself. Since this phenomenon has developed either
side of the shackle, it is logical to assume that the material characteristics of the shackle
may be a contributing factor.
Figure 7-10 - Illustration of the Extent of Pitting Corrosion
108
A4163-01
Figure 7-11 - Example of Wear and Pitting Corrosion on the Shackle Pin
The as forged dimensions on this shackle are not known with certainty, but typical
dimensions are known. The pin of the shackle goes through the end of a common link
of studded chain and the bow of the shackle goes through the studless chain. Based on
nominal or typical dimensions significant wear appears to have occurred at the bow of
the shackle with the bar diameter down from 170mm to 158mm (12mm) a major
reduction in less than 7 years.
7.3.1 Chain Wear/Corrosion Assessment (Studded and Studless)
Since chains and shackles are typically forged the final dimensions after manufacturing
are not known with any certainty, unless as built data is measured, recorded and the
item can be identified. If this is not done the final as manufactured bar diameter at the
inter-grip area may well not be known. As chain is manufactured it is bent around an
anvil when red hot and this tends to reduce the bar diameter particularly where it is
bent.
Based on the nominal chain diameter of the studless 142mm chain this shows an
apparent maximum in field combined wear/corrosion of (142 134.5) 7.5mm over less
than 7 years which at 1.07mm/year is high. So far the apparent wear and pitting
corrosion seen on this chain has been 3.6 times (1.07/0.3) higher than was allowed for
during the design process.
Based on the nominal diameter of the studded 137mm chain this gives a maximum
combined wear/corrosion of (137 132) 5mm over less than 7 years which at
0.71mm/yr is also in excess of what was allowed for in the design process.
109
A4163-01
In this case it would be useful to compare the relative shackle surface hardness with the
existing chain to see if this or galvanic type corrosion is the cause of the high apparent
wear/corrosion rate.
7.3.2 Discussion of the Consequences of the Apparent Wear/Corrosion Rate
It is appreciated that the quoted apparent North Sea wear/corrosion rate may well not be
applicable to all geographical areas and unit types. However, if the rate is even roughly
correct this may well have potentially serious consequences for units intended for long-
term field lives. Thus it is important that this area is investigated further as a matter of
priority - see Section 21.1.
7.3.3 Break Testing of all Chains
At the end of perhaps a 20 year deployment period the minimum break load of all
mooring components on a FPS should still be able to meet the calculated maximum
design load multiplied a suitable safety factor. However, we do not know how
grinding, wear or pitting corrosion will affect the chains break load. An approximate
estimate of the break test load could be obtained by using a finite element model
representation. With such a model it would be difficult to have confidence that the
finite element model is representative, particularly when hairline cracks may be
present. Hence, it is recommended that as used mooring lines and components become
available, either due to line failure or the completion of a FPS assignment, that
representative lines should be break tested to see what their actual break load is after
X years service. Figure 7-12 illustrates a test rig set up from a mooring chain break
test.
Break testing such lines may also reveal the presence or otherwise of any fatigue
cracks. Such cracks may not always be detectable using conventional inspection
techniques. For example, Magnetic Particle Inspection (MPI) on recovered semi-
submersible chain has found crack like indications at the inner bend region of all links.
The materials testing laboratory doing the inspection judged these to be Laps. This
feature occurs as the result of two mating links rubbing together, which causes a fold on
the material surface. Due to the rough nature of the surface in this area it is not
generally possible to do ultrasonic testing to assess the depth of these cracks like
indentations. Hence, the desirability to obtain confirmation of the presence or
otherwise of any obscured fatigue cracks.
110
A4163-01
Figure 7-12 -Test Rig Set Up for Break Testing of Mooring Components (Studless
Chain in the instance)
111
A4163-01
7.4 Sulphate Reducing Bacteria (SRB) Induced Pitting Corrosion
Sulphate reducing bacteria (SRB) have been known to cause pitting corrosion in areas
such as ballast tanks see Figure 7-13. Unconfirmed rumours have indicated that SRB
may have also caused rapid corrosion damage to mooring systems in the North Sea,
south-east Asia and off Brazil. In certain areas, such as the Black Sea, it is believed
that the concentration of SRB is higher and this is believed to have caused some
difficulties for drilling contractors.
Figure 7-13 Illustration of Biologically Induced Pitting Corrosion in a Ballast Tank
It is understood that biologically induced pitting corrosion tends to be more prevalent in
warm oceans. Deep isolated pitting is a text book classic example of SRB attack. Thus
microbial induced corrosion has potential implications for floating production units in
the tropical oceans. SRB are anaerobic and can develop in a < 1mm thick layer of
slime. These bacteria can cause severe corrosion by accelerating the reduction of
sulphate compounds to corrosive hydrogen sulphide. Concern has also been expressed
about the use of high strength mooring line steel in high H
2
S environments as it may
lead to hydrogen embrittlement. This can also be affected by the amount of cathodic
protection being applied (see also Par Ohlsson paper, 3
rd
Int. Offshore Mooring
Seminar [Ref. 5]).
Standard bacteria cultivation tests exist to check for the presence of SRB. It is believed
that it would be possible to collect a slime sample from a mooring line by means of a
ROV or if necessary by diver. If a likely candidate FPS can be identified it would be
interesting to undertake such a test to assess the concentration of such bacteria. In
general 1 SRB per litre of sea water is fairly normal. Higher concentrations can be
found in the sea bed top soil.
112
A4163-01
7.5 Stress Corrosion Fatigue
Investigations undertaken by Vicinay Cadenas, Labein R&D, Bilbao University and
others have shown that corrosion clearly affects the fatigue behaviour of steel mooring
components.
Hostile environments, such as seawater, can accelerate the initiation and growth of
fatigue cracks, particularly in the presence of mean tensile stresses. One mechanism is
the development of corrosion pits, which then act as stress raisers. In other cases the
environment causes cracks to grow faster by chemical reactions and dissolution of
material at the crack tip.
To make fatigue life estimates (see also Section 16 Fracture Mechanics and Critical
Crack Size) it is possible to apply the fracture mechanics characterization as
represented by the Paris equation as shown below in terms of a curve of da/dN versus
'K:
da

m
C 'K
dN
In this expression a is the crack length, N is the number of cycles, 'K is the range of
stress intensity factor. C and m are material and environment dependent constants
which are typically determined in the laboratory. For chain, Vicinay has measured
values for the exponent m in different environments. The figures for dry air are lower
(around m = 2.7), compared with the values of free corrosion in seawater that are m =
2.88. Below a threshold value of 'K cracks do not grow at all. Above a high level of
'K crack growth is much more rapid as is illustrated in Figure 7-14.
Figure 7-14 - Crack Growth per Cycle versus Stress Intensity Range [Ref. 2]
113
A4163-01
Vicinay, in conjunction with fracture mechanics, metallographic and materials science
experts has further investigations in process to study the environment-assisted cracking
(EAC) and particularly the hydrogen-assisted cracking (HAC) behaviour of chain.
7.5.1 Latest Work on Chain Corrosion
For certain oil and gas projects the required design life for production facilities may
reach 30 years. An example of such a project is the Belanak offshore liquefied
petroleum gas (LPG) FPSO facility offshore Indonesia. This hull has been designed
and built to last 30 years without the need for dry docking and all mechanical
equipment has been specified to last for this period. Such a long design life presents
real challenges for a system which is exposed to continuous wear and corrosion, yet at
the end of the field life must still be able to withstand a 100 year return period storm.
The recent OMAE Speciality Symposium on FPSO Integrity in Houston August 30 -
September 2, 2004 included a paper looking at Mooring Chain Corrosion Design
Considerations for an FPSO in Tropical Water [Ref. 33]. This paper reviewed US
Naval Research Laboratory (NRL) data on corrosion rates from its 16 year test
programme in a tropical area and from corrosion data for other geographical areas from
other sources. In summary the US NRLs test results indicate that a corrosion
allowance of 0.2mm per year on one side should be sufficient for compensating the
actual corrosion damage. This gives 2 x 0.2 = 0.4mm/year on diameter which ties in
quite well with the existing codes. However, this is lower than the North Sea reference
number reported in Section 7.2.1, i.e. 0.6mm/year.
What is perhaps significant here is that the 0.4mm/year rate discussed in the OMAE
paper seems to only correspond to corrosion, the effect of wear seems to have been
neglected. North Sea experience seems to indicate that wear can be quite considerable.
In locations such as West Africa, where less extreme but regular FPSO motion can be
expected year after year, the effect of wear is expected to be significant. Hence it is felt
that a 0.4 mm/year rate to cover corrosion and wear is not conservative, at least for the
North Sea. But still more data is needed from other types of units, which have seen
long-term deployments in different geographical locations.
The design of the surface floating facility, the type of mooring and metocean conditions
will affect wear rate. For example in 1982 4.5 inch diameter U4 grade chain on a
CALM buoy failed due to excessive wear after two months, see Figure 7-15 [Ref. 34].
In this case the buoy anchor pattern was asymmetric with distinct strong and weak roll
stiffness axes and surge stiffness axes. However, this incident shows that accelerated
wear can be a real issue.
114
A4163-01
Figure 7-15 Illustration of Excessive Chain Wear on a CALM Buoy [Ref. 34]
7.5.2 Temperature, Salinity and Dissolved Oxygen Content
The following factors will influence corrosion rate all of which will vry to some degree
depending on geographical location. :
- Dissolved oxygen
- Temperature
- Salinity
- Velocity of water particles
All other factors being equal corrosion rates are approximately proportional to the level
of dissolved oxygen in the water. Oxygen content is influenced mainly by water
particle velocity and temperature. As can be seen in Figure 7-16, (also see Figure 7-17)
temperature drops with increasing water depth and hence oxygen content increases.
Thus this is a potentially undesirable effect from a corrosion perspective for deepwater
FPSs.
115
A4163-01
Figure 7-16 Typical Temperature and Salinity Profile in the Tropical Oceans
Figure 7-17 Indicative Oxygen Concentration versus Water Depth (courtesy of BP)
116
A4163-01
Figure 7-18 Gulf of Mexico Snap Shot of Bottom Oxygen Concentration
(courtesy of BP)
Salinity in terms of the chloride content of sea-water increases corrosion rate by
increasing electrical conductivity and adversely affecting development of the protective
films on the chain surface. It can be seen from Figure 7-16 that there is an approximate
change in salinity of approximately 7.7% from the surface down to a depth of about
3,000m.
It can be seen from the values reported in this section that the corrosion behaviour of
deep water mooring systems is presently uncertain.
7.5.3 Inconsistency in API RP 2SK and RP 2I
Wang and DSouzas OMAE paper identified an inconsistency between the mooring
chain inspection requirements defined in API RP 2I (In-service Inspection of Mooring
Hardware for Floating Drilling Units, and RP 2SK (Recommended Practice for the
Design and Analysis and Stationkeeping Systems for Floating Structures.) Although
RP 2I is not necessarily appropriate for a FPSO it would be logical to apply it to a semi
submersible production unit.
Section 3.4 of RP 2I states Links having any of the following problems should be
removed: an average diameter of two measured diameters less than 95% of the nominal
diameter (about 10% reduction of cross sectional area) or a diameter in any direction
less than 90% of the nominal diameter. This is a different approach to the corrosion
rate specified in RP 2SK.
117
A4163-01
An example helps to illustrate the difference between the two approaches. Take a chain
designed to have a net diameter of 105mm for a 30 year service life. Applying a
0.4mm per year corrosion allowance results in a final diameter of (30 x 0.4) + 105 =
117mm. After 25 years assuming the 0.4mm per year corrosion allowance the chain
will still have a sound diameter of 117 (25 x 0.4) 107mm which would still meet
the original design requirement. However, this remaining diameter of 107 fails the RP
2I inspection criteria of 0.95 x 117 111mm.
Obviously it is undesirable to have an inconsistency between two API reports. It is
believed that API 2I is due to be revised and it would be desirable for this inconsistency
to be resolved at this time.
7.6 Wear Analysis (Shoup and Mueller Work)
As was mentioned in Section 7.5 an interesting example of how wear can lead to
mooring line failure is provided by the failure of a CALM buoy just two months after
installation. This was investigated by Shoup and Mueller in their OTC paper 4764
from 1984. Although this is a now a fairly old paper, it is still a particularly useful
work in the respect of surface hardness and wear prediction.
Wear is a complex process involving material properties, forces, sliding distances and
environmental factors, such as sea-water immersion. Hence, rather than relying solely
on theoretical analysis, Shoup and Mueller undertook an experimental wear study.
Because of the cost of full scale component testing, it was decided to perform wear
tests on smaller size specimens simulating as closely as practical the actual service
conditions. Figure 7-19 shows the wear results obtained from the crossed cylinder wear
tests. Both U3 and U4 marine/ship grade chain had high initial wear rates, followed by
a distinct knee and a nearly linear lower rate after approximately 150 cycles. The knee
and the plateau were probably caused by the decreased contact pressure and reduced
sliding distance resulting from wear. The presence of sea water which provided
lubrication caused a distinct reduction in wear. This has implications for external
turret moored FPSOs in benign climates.
118
A4163-01
Figure 7-19 - Measured Wear Rates of U3 and U4 Chain at 8,170lbs (300 tonnes
equivalent) [Ref. 34]
These experimental tests identified wear rate coefficients which are dependent on
applied tension and whether the chains were in air or sea-water.
Using this data a modified form of Archards wear equation was developed of the
following form:
t=
180
|
|
.
K ( )( = r

_=
N 1
1 Fi += += Fi |
=
\
|
=
\=
|i
|
|
.
| ( i 1= += i) | | 3 ........ ) TMV
2
i 1
where:
F = chain tension
= roll angle (degrees)
r = radius of the chain barstock
K = wear coefficient (dependent of F)
N = number of records
TWV = total wear volume for the duration of the test
119
A4163-01
7.6.1 Shoup and Muellers Key Conclusions
The conclusions from Shoup and Muellers paper are interesting and have potential
implications for the reliability of the mooring systems on deep water floating
production facilities. Hence, they are reproduced in full below:
The most important result of the study is the realization that wear is an important
criteria for anchor leg design, especially for deepwater systems. Deepwater catenary
systems are prone to anchor chain wear because:
1. Overall system elasticity and surge motion increases with water depth. As
surge motion increases, interlink motions also increase.
2. Catenary chain moorings have large pretension interlink forces in deep
water. The wear study shows wear rate increases dramatically with
increasing load (particularly at the floating structure interface).
Catastrophic wear failure of catenary anchor leg lines (at the floating structure
interface) can be prevented by:
1. Placing large links below the chainstoppers to keep the gross contact
pressure below the high wear rate regime.
2. Using a stopper casting support which is free to rotate about two
perpendicular axis. This will eliminate most of the wear generating
interlink motions.
3. Studying the behaviour of links in the wear zone to determine if a
particular mooring arrangement generates large relative sliding distance
between links.
With respect to point 2 it is worth noting that that most FPSOs only allow stopper
rotation about one axis rather than two (see Figure 9-3). For spread moored FPSOs it
will be interesting to see if wear experienced in the field may make adopting a twin
axis approach worthwhile.
120
A4163-01
7.6.2 Calibration of Up to Date Wear Analysis Model with Offshore Recorded Values
With modern dynamic analysis mooring packages, it is possible to predict the relative
rotation between individual links for different line tensions/sea states. There are now a
number of FPSs which have been operational for a number of years and from which
indicative wear rates are available. Hence, there is benchmark data which can be
potentially used to validate a wear analysis assessment and assist with the selection of
wear rate coefficients. Thus, it is recommended that such an exercise should be
undertaken (see Section 21). Once a good validation has been achieved it should be
possible to apply the developed methodology to a planned new deep water long-term
FPS. It should be possible to take into account the system specified pretension and
expected environmental conditions and vessel response. In this way the calculated
wear rate can be compared with the code specified wear and corrosion rates (see
Section 0). If the calculated value is higher than the code specified value a cost benefit
analysis may be required to assess whether increasing the line diameter is more cost
effective than carrying out a replacement operation some time during the field life.
7.6.3 Enhanced Wear and the Possible Development of Loose Studs in the Chain
Lockers
On all the lines in the chain locker on the FPSO with adjustable lines discussed in
section 9.1, there will be two slack sections where the chain hangs off from the bitter
end shackle and down from the ceiling mounted gypsy wheel. As the FPSO responds
to the environment these slack chain sections will move around and may be subject to
wear within the locker, which might not normally be expected for un-tensioned chain in
the chain locker. The motion of the chain added to the possibility of corrosion inside
the chain locker could lead to the development of loose studs. On this unit loose studs
were found in a 17 link, chain section, which was pulled out from the chain locker.
Previous experience with chain storage in lockers on semi-subs indicates a potential for
corrosion pitting damage. To quote, chain which normally remained in the locker
exhibited severe localized corrosion in the form of deep pitting. This was unexpected.
However, the severe pitting probably resulted from the formation of oxygen
concentration cells at points of contact within the stored pile of chain. The moist salty
environment provided electrolyte and the varying local concentrations of oxygen
provided the anode/cathode galvanic potential [Ref. 35].
Pitting corrosion and the wear are both potentially significant points, since you do not
want to adjust line lengths to reduce wear and possibly by doing so introduce weak
links into the system, which were not previously under high tension see Section 9.1.
121
A4163-01
8 UNBALANCED LINE PRE-TENSIONS (CASE
STUDIES)
8.1 North Sea Semi-Submersible FPS
When a mooring analysis is undertaken the pre- or working tensions are set at specific
values, which are often identical. This is a reasonable approach as long as the unit in
the field can set their line tensions to comparable values. If the set up line pre-tensions
on a FPS are unbalanced, this can lead to increased maximum line tensions and reduced
fatigue lives. In addition, in case of a single line failure this can lead to an increased
transient excursion, which might exceed the allowable watch circle.
On a North Sea semi-sub FPS the offshore personnel doubted the tension readouts were
accurate because:
- Sometimes the wire became partially bedded into, and/or damaged the lower
wrap on the winch drums
- When grappling for certain components on the mooring line they were not
found at the expected depth.
Therefore, an underwater ROV survey was taken of the flounder or tri-plate connectors
on the mooring lines to obtain their x, y and z co-ordinates. From these positions and
knowing the submerged weight of the line it was possible to undertake a catenary line
calculation to determine the actual line tension. These tensions can then be compared
to the tension readouts on the rig at the time that the ROV position check was made. It
was found from this process that the calculated tensions and the measured tensions
could be out by up to 160% in the worst instance!
There are a number of potential reasons why the tension meters were so far out. These
include:
- The meters have not been calibrated or the calibration has drifted over time
- The gypsy wheels may be seized
- The instrumentation is not sufficiently sensitive
- The tensions are measured at the base of the winches in board of the fairleads
122
A4163-01
8.2 Line Payout/Pull-In Test
To confirm whether or not the gypsy wheels were seized and to assess the sensitivity of
the tension meters, a carefully controlled Line Payout/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 by the same amount and the line tensions recorded. If this
test is undertaken relatively quickly in good weather, it would be expected that the
same tension would be obtained for the same line payouts. However, this was revealed
not to be the case in all instances, see for example the plot below for Line Number 11.
Line No11
W
i
r
e

p
a
y
o
u
t

(
m
)

195.0
194.0
193.0
192.0
191.0
190.0
189.0
188.0
187.0
186.0
185.0
0.0 20.0 40.0 60.0 80.0 100.0 120.0
Tension (te)
Figure 8-1 Illustration of Line Tension Variations during a Payout/Pull-In Test
The wiggles on this graph are believed to be due to due to the sheaves binding and
then becoming free and then binding again. It is understood that a similar wiggle
pattern has been recorded during a Payout/Pull-In test on a Gulf of Mexico Spar.
123
A4163-01
8.3 North Sea FPSO
This FPSO is fortunate, compared to FPSOs where the mooring lines are stoppered off
at the turret base, in that loadcells are available which can provide a rapid means of
verifying mooring line tensions. In principle loadcells should be able to detect both
extreme event line tensions and normal working tensions. However, on this particular
unit not all the loadcells are working properly and the offshore personnel have little
confidence in the reported values. In addition, the wildcats could be partially seized,
which would influence the line tension readings.
Obviously if loadcells are uncalibrated the results cannot be treated with confidence.
But also on a single loadcell unit, with a very large reporting range, the sensitivity to
accurately establish the lower pretension value is uncertain, as seems to be
demonstrated by this FPSO. Hence consideration should be given to using two
loadcells on each mooring line. The first would be accurate at the pre-tensioning load
( 100t) level. The second loadcell would have sufficient range to monitor loads
during storm conditions ( 1,000t).
8.3.1 Chapter Conclusions and Recommendations
Historically semi submersible drilling units have been subject to relatively frequent
mooring line failures which equate to approximately one failure per three operating
years. Sometimes these failures cannot be attributed to obvious causes. 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 may well not be detected by the Operator.
Such a Rig would thus be in greater danger of mooring line failure.
If the tension meters are well positioned, working properly and their calibration is in
date, a likely cause of unbalanced line tensions is partial seizure of the gypsy wheels.
This can be confirmed by a simple line Payout/Pull-In test. If this reveals that some of
the gypsy wheels are partially seized an attempt should be made to free them up.
However, if the unit is on station it may not be feasible to undertake such work in situ.
In such a case the line tensions out with the fairlead should be determined by other
measures such as:
- ROV or possibly diver monitoring of the chain angles where they emerge
from the fairleads
- Acoustic monitoring of the x, y and z positions of specific connectors on the
mooring lines
From these measurements it is possible to back calculate the actual line tensions as long
as this is done in calm conditions with minimal tidal variations.
124
A4163-01
At present it is not known how common a problem this could be for other operating
units. Seized wheels may be more likely on a wire sheave than on a chain
gypsywheel/wildcat. Hence, it is recommended that similar Payout/Pull-In tests are
repeated for a number of different ages and designs of Semi-Submersibles. This is
recommended in the HSEs recent research report 219, Design and Integrity
Monitoring of Mobile Installation Moorings [Ref. 36].
Azimuth Checks and Marine Growth
If a gypsy wheel is partially seized with respect to rotation it may also be seized relative
to azimuth rotations. Hence, as well as checks gypsy wheel checks on free running, the
ability of the fairlead assembly to freely slew or azimuth should also be confirmed. If
the fairleads cannot azimuth freely increased chain wear is likely to occur. In practice
the best way to achieve this in the field may be to examine the marine growth at the
fairlead to see if it has been displaced as the gypsy wheels azimuth. If there is no
evidence of removal of marine growth it is likely that the fairleads may be seized in the
azimuth direction and may also have problems rotating!
125
A4163-01
9 MOORING BEHAVIOUR AT THE VESSEL
INTERFACE (CASE STUDIES)
The design of the vessel interface needs to minimize the potential for wear, corrosion or
other forms of degradation. However, in field experience is demonstrating that this is
not always being achieved. This is discussed in this chapter. The key points are
relevant to mooring systems in general, not just to one particular design or even type of
floating platform. Although turrets are discussed in detail the key points are relevant to
Spars, spread moored FPSOs and semi-subs.
9.1 Permanently Stoppered Off Versus Adjustable Lines
There are a number of different turret designs available on the market. On many turret
designs the chains are stoppered off at the base of the turret see Figure 9-2. There are
also a fewer number in which the line lengths can be adjusted during the life of the unit
see Figure 9-1. In addition, there is at least one unit which uses wire into the turret as
opposed to chain. Although there are many different 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 normally 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 semi-submersible drilling unit 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.
Being able to chain the line lengths has the following beneficial effects:
1) Distributes the high wear point on the chain over several links thus prolonging
chain life.
2) Distributes the wear over several gypsy wheel pockets, thus prolonging gypsy
wheel life.
126
A4163-01
Figure 9-1 - Turret Design in which Chain Lengths can be Adjusted (courtesy of
Chevron-Texaco)
Figure 9-2 Generic Turret Design in which the Chains are Stoppered off at the Turret
Base (courtesy of Bluewater)
127
A4163-01
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 (see Section 18).
Being able to adjust line lengths can introduce its own perils, although these should be
controllable. During a regular line tension adjustment operation on one North Sea
FPSO there was a failure of the lifting and locking mechanism resulting in a complete
line run out (see Section 0).
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. This may be
particularly relevant to spread moored FPSOs which cannot weather vane. Hence,
there may be more wear at the chain/hull interface when the weather is not directly on
the bow. Depending on location the weather coming in on the vessels quarters may
occur for a significant proportion of the time.
Figure 9-3 - Spread Moored FPSO Single Axis Chain Stopper (courtesy of SBM)
128
A4163-01
Trumpets or guides are normally 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 see Section 11. However, in this case the chains in the stoppers were
probably pre-rigged before the Spar was towed out to location. A kenter joining
shackle was then used to connect up the chain in the field before the line was tensioned
up. If you have a reliable method of connecting up in the field this approach does have
some advantages. For example, the trumpets can be dispensed with and it is also easier
to undertake a change out of the top chain section at some stage during the field life if
required without cutting the chain. This illustrates the importance of having long-term
reliable connectors, which is an area which still requires further work.
129
A4163-01
9.2 Wear at Trumpet Welds Internal and External Turrets
9.2.1 External Turret
On a number of 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 trumpet section (see Figure 9-6). This was first experienced on an early S.E.
Asian external turret moored FPSO. For this 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 (UMPHE)
sheeting). This whitish material can just be seen on Figure 9-4 poking out of the
trumpets. In this case the weld beads were left as they were with no attempt to grind
them down smooth. On this project UMPHE has been successful in stopping the chain
wear, however, the sheeting needs regular inspection and replacement when it becomes
worn or damaged. Hence, this is a solution which is only suitable where access is
good, not for a submerged turret, in a harsh environment.
Figure 9-4 - External Cantilever Turret which experienced Chain wear at the Trumpet
Welds which was halted by use of UMPHE (courtesy of Shell)
Considerable wear has also been noted on the chains which are normally in air on a
benign climate external turret unit. Water lubrication may be a possibility to minimize
the wear rate on such units see section 7.6.
130
A4163-01
9.2.2 Wear at Trumpets - Internal Turret
In the North Sea mooring lines are typically inspected utilising a work class ROV
which performs a fly by of the lines. During one of these surveys a slight shadow was
seen on one of the chains at the trumpet interface during the annual workclass ROV
chain survey. Unfortunately, the large work class ROV was unable to get close in
enough to inspect this shadow to determine whether it was simply removal of marine
growth and mill scale, or if a notch was being ground into the chain (see Figure 9-5).
To investigate this apparent anomaly further, a test tank mock up of the chain and
trumpet assembly was built so that the capability of using a football sized micro-ROV
(see Figure 17-5) to get in close to the bell mouth could be evaluated. This test tank
test is illustrated in Figure 9-6. Micro-ROVs are particularly attractive for inspecting
around the base of the turret since they can be deployed from the FPS itself rather than
employing the services of a ROV support vessel. A micro-ROV can typically be
deployed over the side of the FPSO either by hand or using a simple lowering frame.
In addition, on some FPSOs there may be a spare I tube which is wide enough for the
micro-ROV to be lowered down through. The test of the micro-ROV was successful
ands it was subsequently deployed in the field. Figure 9-7 illustrates one of the
photographs taken by the micro-ROV in the field. Marks can be clearly seen on both
left hand and right hand faces of the chain where it has been in contact with the
trumpet.
Figure 9-5 - Example of the Level of Inspection Detail which can be achieved using a
Typical Workclass ROV (courtesy of I.Williams)
131
A4163-01
Figure 9-6- Test Tank Mock-Up of Micro-ROV inspection of Chain Emerging from
Turret Trumpet (courtesy of I. Williams)
Figure 9-7 - Micro-ROV Photograph of Chain Wear Notches where Chain Emerges at
the Trumpet Bell Mouth (courtesy of I. Williams)
132
A4163-01
6
1
Even with the better image resolution provided by the micro-ROV, quantifying the
exact extent of the chain wear was difficult. However, it can be seen from
Figure 9-8 that it was potentially significant if the illustrated reduction in bar diameter
is correct. In addition, it can be seen from Figure 9-10 that the location of the notch is
in an area which is subject to significant reduction in bar diameter when a chain is
loaded up to its MBL. Unfortunately, there is little data available on how reduction in
bar diameter can affect chain strength. To try and determine as reliably as possible how
a notch in the chain would affect strength, a notch was ground into some spare chain
links left over from the original installation see Figure 9-9. This link was then break
tested to assess how much the chain MBL had been reduced by the presence of the
notch.
6
1
Figure 9-8 - Indication of the Extent of the Wear
133
A4163-01
Figure 9-9 - Artificially Introduced Notch on to Spare Chain Links, note also Red
Circular Infrared Target (courtesy of I. Williams)
Figure 9-10 - Example of Stretched Chain during Break Testing, the Blue Mark Shows
the Location of a Typical Notch (courtesy of I. Williams)
134
A4163-01
The results from this break test of the notched chain indicated that it was likely that
some of the as installed mooring lines would no longer meet the required mooring line
safety factors. Even if the lines were still within the required safety factors, it would
just be a matter of time until they became out of specification and when this might
happen could not be reliably quantified. In addition, there was some possibility that
fatigue cracks could have developed due to the regular knocking action which, if
present, would reduce the break strength considerably. At present no technology exists
which can check for fatigue cracks underwater, particularly in such an inaccessible
area. Therefore, the decision was made to undertake a repair operation to change out
the links going through the trumpets by custom built chain links of the same length as
the existing chain, but made from a larger bar size. In addition, the new links were
given a special hard cobalt chromium anti wear coating see Figure 9-11. Further
details of the repair operation can be found in Section 18.8.2.
Figure 9-11 - Example of a Special Cobalt Chromium Anti-Wear Coating (courtesy of
I. Williams)
9.2.3 Actual Chain Condition after Recovery
It is interesting to compare the actual condition of the recovered compared to its
expected condition. Figure 9-12 shows one of the recovered links. This figure clearly
demonstrates that the wear was gradually eating into the side of the chain, thus
progressively weakening the link. Although the extent of the wear was not as bad as
some of the earlier predictions, it was clear that the wear would get worse over time.
Fortunately, Magnetic Particle Inspection (MPI) of the key links did not reveal any
hairline cracks, but this was not known beforehand. As well as the wear due to contact
with the weld beads, additional damage was noted along the chain which was lying
along the trumpet and sitting in the stopper see Figure 9-13.
135
A4163-01 136
Figure 9-12 - Photograph of a Recovered Link Showing a Wear Notch (courtesy of I.
Williams)
Figure 9-13 - An Example of the Chain Damage noted after the Notched Chains had
been recovered back to Shore (courtesy of I. Williams)
A4163-01
9.2.4 Reasons for the Wear
It is significant to note that the chain stopper on type a) designs is typically inboard of
the pivot point - see Figure 9-14. 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 see Figure 9-21. It
is this contact, plus an associated sliding/sawing action, which seems to have led to the
chain notches.
Figure 9-14 - Turret Arrangement where the Chain Stopper (in red) is Behind the
Rotation Point (2 black concentric circles)
Should the Stopper be behind or in front of the Pivot Point?
It is helpful to consider the pros and cons of having the pivot point behind the chain
stopper (i.e. the rotation point is closest to the hull). Some spread moored units have
gone the other way (see Figure 9-16). This approach seems to ensure that the
compliance introduced by the bearing takes out as much of the motion as possible and
the metal to metal contact as illustrated in Figure 9-15 is avoided. It will be interesting
to see how much wear is experienced in the field by the designs with the stopper
outboard of the pivot point. It will also be interesting to see what happens to the chain
which is under low tension from the stopper up to the deck of the FPS see Section
9.2.5.
Implications of Long Trumpets
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 FPSOs which fit this category
[Ref. 37].
137
A4163-01
Figure 9-15 Illustration of Potential Wear at Metal to Metal Contact (courtesy of I.
Williams)
Figure 9-16 - Fairlead Chain Stopper where the Chain Stopper is in Front of the
Rotation Point (used on some Spread Moored FPSOs) (courtesy of Maritime Pusnes)
138
A4163-01
Figure 9-17 - As Installed Photo Graph of the Design Shown in Figure 9-16 (courtesy
of Maritime Pusnes)
Compatible Surface Hardness
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.
Having the pivot point behind the chain stopper may date back to the original design of
CALM buoys (see Figure 9-18). However, as far as can be determined the early Shell
buoys did not have long trumpets and thus wear at the end of the trumpets may not
have been an issue. Given that a tried and tested working design from CALM buoys
was already available it is not surprising that this detail was incorporated into early
FPSO turret designs which were not initially deployed in harsh environments.
139
A4163-01
Figure 9-18 Typical CALM Buoy Chain Stopper (courtesy of The Professional
Divers Handbook [Ref. 38])
On the subject of CALMs Figure 9-19 shows an Imodco buoy with a rubber casting
used to minimise wear at the lip of the trumpet. Thus, it is clear that potential wear in
this area has been an issue for a number of years. Significantly during installation it is
apparently difficult to get the rubber castings in exactly the right place.
Figure 9-19 - Amoco CALM Buoy- Note Inclusion of Rubber Casting (courtesy of
[Ref. 38])
140
A4163-01
9.2.5 Alternative Fairlead Designs
The traditional fairlead arrangement on a semi-sub is illustrated on the upper two
sketches of Figure 9-20. With this design the chain where it runs round the lower
wildcat is under high tension, particularly in storm conditions. With this type of design
the first free link hinges on the last link in the fairlead pocket as the chain catenary
angle changes. Thus the chain scrubs the surfaces of the pocket and whelp under high
contact pressures. Hence, over time, both the chain and the wildcat will suffer from
wear and damage see Figure 18-8 and Figure 18-9. An alternative design is presented
in the lower two sketches of Figure 9-20. In this design the in the motion between the
chain and the surface platform is mainly taken out at a horizontal pin which attaches the
stopper to the floating vessel and also by a freely azimuthing assembly.
With this alternative design it is important that the chain should not be actually slack
from where it runs from the chain stopper to the windlass or chain jack. If the line is
too slack there may well be excessive movement between links as the surface platform
responds to wave excitation. Excessive movement can lead to accelerated wear. If
wear happens above the stopper and then line is let out a weak point may be introduced
in the system. However, not too much back tension should be included since it is
important that the whole chain stopper assembly should still be able to azimuth freely.
It is important that the in field performance of these new designs of fairleads should be
studied after a few years of operational experience to check whether they are
performing as well in situ as hoped. This information then needs to be fed back to the
wider mooring community.
141
A4163-01
Figure 9-20 - Comparison of Alternative Fairlead Arrangements (courtesy of Bardex)
142
A4163-01
9.3 Use of Bending Shoes
Before concluding this section on mooring lines at the vessel interface it is appropriate
to also include mention of bending shoes. As can be seen from Figure 9-21 and Figure
9-22 bending shoes can be used both for wire rope and chain. At present there is little
data available in the public domain comparing the performance of bending shoes to
either wildcats/gypsy wheels or permanently stoppered off designs. It would be
extremely helpful to track down such data, since it could be that a well designed
bending shoe could help to preserve the life of the mooring line at the vital vessel
interface.
Figure 9-21 Example of a Wire Rope Bending Shoe (courtesy of API RP25K)
Figure 9-22 - Example of a Chain Bending Shoe Design [Ref. 39]
143
A4163-01
For wire rope cyclic stresses from bending will shorten its service life because of
fatigue. Fatigue resistance (service life) increases as a ratio of bend shoe diameter to
wire diameter (D/d) increases. Individual wires move relative to one another and to the
bearing surface as the rope bends causing abrasion. Abrasive wear increases as D/d
decreases. Under heavy loads, the rope flattens against the bearing surface, increasing
relative motion between strands and wires. Lubrication and large D/d ratios mitigate
the adverse affects of bending. Minimum D/d ratios are available for different rope
constructions [Ref. 19 7-2.12].
Chain works most efficiently when loaded in pure tension. Tensioning chain that is
bent over a surface introduces bending stress that reduces load carrying capability. It is
thus recommended that Chain should not be tensioned over surfaces with diameters less
than seven times the chain diameter. Thus sharp bends and corners should be avoided
[Ref. 19 7-29].
The bending shoe design illustrated in Figure 9-23 includes an angle sensor which can
be used to back calculate the static line tension. However, given the problems outlined
in Section 0 it will be interesting to see if the dynamic behaviour of the chain at this
point over time may cause wear problems. The particular application illustrated is in
deep water and hence line dynamics (whipping/fluid drag) will affect tension. In other
words the recorded angle may not give an accurate idea of the tension in the line.
Figure 9-23 - Bending Shoe Design which includes an Angle Sensor [Ref. 40]
Another point to note on this project is that the chain is locked off and it is not planned
to be moved regularly. In fact the chain jacks were removed after installation and will
be re-installed as required during chain inspection. Not being able to work the chain
will affect its fatigue life, so it will be interesting to see how well this mooring system
performs over time.
144
A4163-01
10 FURTHER MOORING CASE STUDIES
10.1 Wire Rope Systems
A number of floating production/storage units ropes within their mooring systems
have seen periods of extended operation in the north sea including :
- AH001
- Buchan FPS
- Emerald Producer FPSO
A number of wires from these units were removed and examined as part of two
previous JIPs [Ref. 41 and Ref. 42]. The inspection of these lines confirmed that wire
will be subject to degradation at the fairlead region and in the thrash zone. Hence, if
IWRC wire is used in these locations it will typically need to be replaced after about 8
years service see Table 3-9. Further information on when to discard IWRC mooring
lines can be found in Chaplin 1992. Figure 10-1 from Chaplin 1992 [Ref. 43] gives an
idea of the type of degradation which IWRC rope can be subject to :
Figure 10-1 Examples of the Subjectivity Associated with Assessing IWRC Rope
Conditions [Ref. 43]
145
A4163-01
10.2 Unintended Line Disconnection
On a North Sea dis-connectable FSU (see Figure 10-2) the mooring wire and socket
was found to have parted from the triplate assembly. An in-water survey showed the
line to be in normal alignment, but separated by 36m from the triplate assembly, which
was still securely attached by the mooring chain to the suction anchor. Inspection of
the mooring line socket showed the socket retaining pin to be displaced, as one of the
circular retaining plates which keep the pin in place had parted from the socket body
(see Figure 10-3). It should be noted that the initial design prevented pin from rotating,
also Section 14.1.2.
Figure 10-2 - Illustration of the Mooring Layout and Connections
146
A4163-01
Figure 10-3 - Photograph of Disconnected Socket on the Sea-Bed (courtesy of BP/Stolt
Offshore)
Figure 10-4 - Note End Plate also seems to be Falling Off on the Right Hand Side
(courtesy of BP/Stolt Offshore)
147
A4163-01
Figure 10-5 - End Connection Detail
Figure 10-6 - Illustration of Socket Minus End Plate
148
A4163-01
-
10.2.1 Probable Causes of the Failure
There was a relatively steep change in mass per unit length at the triplate. This meant
that the line at the forged tri-plate was subject to:
repeated pick up and set down contacts with the seabed, and
- quite large relative rotations between chains, the wire and the tri-plate
elements.
This resulted in rotational torque being transmitted from the wire socket through the
retaining pin into the tri-plate and finally out to the chain cable via the LTM shackle.
The pin retaining plate is bolted both to the pin and to the socket body the pin. The pin
cannot rotate and the torque must be resisted by these bolts. These bolts either became
loose and fell out, or failed in shear/fatigue.
How was it Rectified?
The problem was rectified by using more and bigger bolts on the end plate and
allowing it to rotate see Figure 10-7. The issue of whether or not to allow the pin to
rotate is discussed in greater detail in Section 14.1.2.
Figure 10-7 - Repair Utilised Bigger Bolts and Allowed the Socket Pin to Rotate
149
A4163-01
10.2.2 Anode Failures
Excessive corrosion was noted on the moorings wires discussed in Section 10.2.
Therefore a series of anodes were retrofitted on the lines to control the corrosion level
see Figure 10-8. The anodes were inspected after approximately 12 months service
see Figure 10-9, where it can be seen that a number had become disconnected. There
are a number of possible reasons for the anodes becoming disconnected and, due to
commercial reasons it is not possible to discuss these in detail. However, from a
mooring integrity point of view the key message seems to be keep your catenaries
clean see also Section 10.3. In other words avoid adding anything on to the
catenary, particularly in the thrash zone.
Figure 10-8 - Example of Retrofitted Anodes to Control Corrosion Rate
Figure 10-9 - Example of Disconnected Anodes after approximately 12 months of
Service
150
A4163-01
10.3 Excursion Limiting Weighted Chain Problems
Excursion limiting weighted chain designs (see also Section 3.1.6) have been adopted
for certain FPSO projects. However, as can be seen from the two projects discussed in
this section, their use can cause problems. Figure 10-10 and Figure 10-11 illustrate
clump weights used on a North Sea FPSO. Unfortunately, as can be seen from Figure
10-10 a number of clump weights have come off in service. It is not feasible to re-
attach the clumps without recovering the lines to the surface, which is a major costly
exercise.
Figure 10-10 - Example of Detached Clump Weight on the Sea-Bed
Figure 10-11 - Example of Recovered Clump Weights
151
A4163-01
General Location of
Damaged Shackles
General Location
of Excessive
Wear
Figure 10-12 illustrates an alternative weighted chain design utilising hung off chain
tails. However, this design has also seen problems as is discussed in Section 7.2.1 and
illustrated in Figure 7-7.
Wear
General Location of
Damaged Shackles
General Location
of Excessive
Figure 10-12 Illustration of Where the Damage Occurred on the Mooring Catenary
10.3.1 Use of Parallel chains to Increase Weight
Figure 10-10 illustrates an excursion limiting weighted chain design which has
operated successfully in the North Sea since the later half of the 1980s. As can be seen
it utilises a parallel chain design.
Figure 10-13 - Example of a Parallel Chain Excursion Limiter (courtesy of I. Williams)
152
A4163-01
If adopting such a solution with more than 2 parallel chains it is necessary to appreciate
that the chains will not be of identical length and thus will experience different tension
ranges. Hence the chains need to be suitably sized. In addition, it is important to still
inspect the chains regularly. Figure 10-15 shows the extent of wear that can still occur
due to the dynamic motion in the thrash zone. Thus whatever connectors are selected
need to be robust.
Figure 10-14 - Weighted Chain Option Utilising Parallel Chain Sections
(courtesy of N.Groves)
Figure 10-15 - Red Arrow Illustrates the Local Wear can take place when utilising
Parallel Chain (courtesy of N. Groves)
153
A4163-01
Line 10,
shackle
buoy
failed,
2001
Line 1,
shackle
buoy
failed,
2004, sea
bed wire
corroded
Line 2,
buoy chain
cracked &
replaced
Line 3,
buoy chain
failed
10.3.2 Buoys on Mooring Lines
Problems with submerged mooring line buoys have been encountered on a European
based FPSO (see Figure 10-16). Figure 10-16 and a semi-submersible production unit.
It is interesting to note that in Figure 10-16 the failures have occurred on leeside or
breasting lines, not windward lines.
Line 10,
shackle
buoy
failed,
2001
Line 1,
shackle
buoy
failed,
2004, sea
bed wire
corroded
Line 2,
buoy chain
cracked &
replaced
Line 3,
buoy chain
failed
Figure 10-16 - Example of Mid-Line Buoy Failures on a European FPSO
154
A4163-01
10.4 Line Run Outs and Quick Releases
A complete mooring line run out from the turret of a North Sea FPSO occurred during
the course of this JIP. The line back to the bitter end shackle was whipped out of the
turret and fell down to the sea-bed. It was a serious uncontrolled incident which could
easily have lead to a serious injury or a fatality. There was an added danger of damage
to sub sea assets. Again semi sub drilling units have suffered from chain run outs and
thus this was not a known failure mode. Figure 10-17 illustrates the damage done to
one of the chain gripper chocks.
The following list of operations chronicles the events leading up to the failure:
- Raised hydraulic oil pressure to approximately 250 barg.
- Chain load was taken on the gripper and the chain rose up to remove load
from stopper.
- Opened the stopper and the chain was pulled up by fully extending the lift
cylinders.
- Attempted to close the stopper but was not possible because the stopper
was contacting chain link. The operator considered that either tensioner
was not fully extended or that chain links were too long.
- The chain was lowered and the exercise repeated, but it was still not
possible to engage the lower stopper.
- Whilst lowering the tensioner a loud noise occurred and operator thought
that chain was slipping through the tensioner and ran for cover.
- Due to dust from the chain being detected by smoke detectors, a platform
general platform alert (GPA) occurred and all personnel were mustered.
The FPSO was shut down until an ROV could be mobilized through fear
of damage to risers and other mooring lines.
Late Design Changes and Subsequent Modifications
The design of this particular mooring system was revised during the latter stages of
fabrication. This was a result of further load cases which required a stronger mooring
system. This was identified when the turret fabrication was well advanced. Thus, with
the positions of equipment fixed, compromises in the design were made. Critical to
these were the relative position of the tensioner to the chain locker spurling pipe which
had the effect of fixing the size of the gypsy wheel and therefore the number of pockets
in the wheel - see Figure 10-18. Also the tolerances of forged chain links had not been
properly taken account of. 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 [Ref. 45] and OTO 98086 [Ref.
46].
155
A4163-01
Figure 10-17 - Gripper chock showing chain damage
Figure 10-18 - Upper Gypsy Wheel Arrangement before Failure
Figure 10-19 - Gypsy wheel structure after failure, i.e. Gypsy Wheel No Longer Present
156
A4163-01
Chain Run out Implications for the Industry
The Manufacturer of the Linear Tensioner assembly which failed has confirmed that
this FPSO is unique in its use of a combined upper gripper and lower stopper assembly
in one installed unit. BUT this does not necessarily mean that there is no danger of
possible chain run out on other units which are able to adjust mooring line tensions.
Hence the lessons learnt need to be distributed through out the industry. 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.
10.4.1 North Sea FPSO Repair of Loose Studs
On an early North Sea FPSO it was discovered that a number of mooring lines had
loose studs. The lines were repaired with a new design of kenters (see Figure 10-20).
However, at present the classification society is stating that, despite the expected
superior fatigue performance of these kenters, they will still need to be examined in the
dry after 5 years service. Recovering kenters on to the back of an anchor handler, so
that they can be dissembled and examined, is a major cost. Since kenters themselves
are not that expensive relative to boat time, it makes sense to replace any kenters which
have to be recovered for inspection. The replaced kenters can then be examined in
detail back on land to evaluate whether there is deterioration or cracking. If this shows
that the new improved fatigue life kenters have behaved well in the field, there would
be more of an argument for leaving them in situ for longer between inspections.
Figure 10-20 - Illustration of a New Design of Kenter Shackle intended to have
improved Fatigue Performance
157
A4163-01
As FPS units get older, the desirability of repairing lines using components, such as
kenters, which are as similar as possible to common link chain, will increase. Hence, it
will be important to record the performance of all new designs of kenters both on FPS
and semi submersible applications. Since flotels and drilling semis recover their
mooring lines regularly this should give increased scope for inspection compared to a
permanent FPS mooring. It is worth noting, however, that at present no type of kenter
is permitted to be part of a permanent mooring system in the Gulf of Mexico.
10.5 Windlass Failures
This incident relates to a South China Seas production semi. Eleven mooring lines
made of 5 inch spiral strand wires and 4.75 inch chains moor the semi. The windlasses,
by paying in and out of the upper chains, are used to adjust the position of the semi
over the subsea trees. In the early phase of the project the wells were batched drilled
and completed. This led to a significant chain mileage. A year after first oil, one of the
windlass wheels started to wobble. A closer inspection discovered that the windlass
wheel was split into two pieces by a circumferential crack, see
Figure 10-21. Inspection of the other wheels indicated similar damage. Since the
windlasses could not be used, all the chains were locked on the chain stoppers. The
wheels were not of a standard design. The wheels designed for a 5 inch chain had
been modified to accommodate two chain sizes: 5 inch and 3 inch. The 3 inch chain
was used to pull in the 5 inch mooring chain at installation. An additional
circumferential groove had been machined in the wheel to accommodate clamps
between the 3 inch to 5 inch chain. The roots of the main wheels were squared up to
accept these clamps. The circumferential groove had no fillet.
Figure 10-21 - Example of Windlass Crack (Red Arrow) due to Stress Raiser caused by
Sharp Corner (courtesy of BP)
158
A4163-01
An important principle of practical design is the scrupulous avoidance of sharp corners
or other "stress raisers" if there is any suspicion of alternating stress. So the cause of
the failure was the stress raiser. Two independent fatigue analyses were made that
showed the wheel could not have failed even with the bad fatigue detail. Further
investigations revealed that the chains were not being locked off by the chain stoppers
when the chains were not being adjusted, thus increasing the wave cycling loading on
the wheels. Including the wave and wind tension cycle damage continuously was still
far from sufficient to explain the failure. Knowing the answer the investigators dug
deeper and added to the fatigue estimate the damage caused by the wheel rotation under
the chain load. The rotation fatigue damage greatly exceeded the environment fatigue
damage and easily explained the failures. New wheels without the groove were air
shipped to Hong Kong and installed. The windlasses have operated without problem
since then.
It is perhaps significant to note that fatigue damage caused by wheel rotation under
chain load is not typically evaluated.
10.5.1 Operators Conclusions from this Incident
The following summarises the Operators conclusions from this incident which are
informative from a mooring integrity point of view:
1. One of a kind designs or modifications of old designs sometimes fail
prematurely.
2. When one designs a first-of-its-kind system that is critical to the operation
of a one billion dollar facility, one should make the design robust. There
are many ways to increase robustness. One very effective way, and
practically free in comparison to the consequences, is to remove all stress
raisers and all bad fatigue details.
3. Experienced specialists should perform detailed reviews of the design and
finished product.
4. In a one of a kind design there are many unknown unknowns that may
load the system in unanticipated ways.
5. Failure investigations should be well publicised to help educate others - part
of the purpose of this JIP!
159
A4163-01
11 SPARS AND OFFLOADING BUOYS (CASE
STUDIES)
11.1 Brent Spar Buoy
The Brent Spar Buoy, although now famous for the nature of its abandonment, was a
successful design from a mooring integrity point of view. It had a 19 year operational
life and minimum wear was found on the chains at the stoppers when they were
examined when the Spar was cut up in Norway, see Figure 11-4 and Figure 11-5 . The
MBL of the IWRC wire rope was found to have had no loss of strength when break
tested after the line had been recovered see Figure 11-6. Indeed if the strength had
changed at all it had marginally increased.
Figure 11-1 - General Arrangement of the Brent Spar Mooring System (courtesy of
Shell)
Brent Spars motion characteristics are probably significantly better than either a semi-
sub or a FPSO. Loop currents do not occur in the North Sea and hence vortex induced
hull vibration on Brent Spar does not seem to have occurred, unlike some Gulf of
Mexico Spars. Brent Spar is interesting in that, relative to most FPSO designs of today,
there were no trumpets or hawse pipes to guide the chain into the stopper. Figure 11-3
and Figure 11-2 show the fairlead arrangement used on Brent Spar.
160
A4163-01
The chain is likely to have been pre-rigged which would have made things easier in the
field, but this meant that a kenter was introduced at the connection. Thus this does not
seem to be a particularly desirable solution for a long-term moored FPSO. Still it
would be good to find a way of lining up chain in a stopper without using trumpets and
angle iron as a guide, since these items can cause problems over time, see Section 9.2.
It is believed there was one mooring failure on Brent Spar, 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 see Section 10.4.1. In addition, one of the Brent Spar
mooring lines got damaged by an anchor line from a drifting vessel.
Figure 11-2 - Brent Spar Fairlead Chain Stopper in the Hull (courtesy of Shell)
Figure 11-3 - Close Up of the Stopper (courtesy of Shell)
161
A4163-01
Figure 11-4 - Indentation from where the chain bore down on the Stopper (courtesy of
Shell)
Figure 11-5 Red Arrow Illustrates wear on the chain, where it sat on the stopper
(courtesy of Shell)
162
A4163-01
Figure 11-6 - Brent Spar Wire Sample Y1 prior to cleaning [Ref. 41]
11.2 Floating Loading Platform (FLP)
The FLP illustrated in Figure 11-7 is continuing to enjoy a 12 year deployment in the
Northern North Sea. During this time no problems have been experienced with the
mooring system. What is perhaps significant about the FLP is that the trumpets when
the chains come into the platform are short see Figure 11-8. In addition, this unit was
fitted with simple, but reasonably accurate inclinometers (see Figure 17-4). Micro
ROV inspection of such inclinometers, in calm weather conditions, can identify if the
lines are intact and whether or not a breakage could have occurred in the mud.
Figure 11-7 FLP Mooring General Arrangement (courtesy of Shell)
163
A4163-01
Figure 11-8 - Example of Short Trumpets on a Long Term Moored Floating Loading
Platform (courtesy of Shell)
164
A4163-01
12 TURRET MECHANICAL IMPLICATIONS FOR
MOORING INTEGRITY
12.1 Introduction to Turrets and Failure Modes
Turrets are reasonably complicated mechanical constructions which are subject to the
following :
- A long service life with limited opportunity for in depth inspections while
deployed.
- Regular fatigue loading.
- High storm loading.
- Gradual wear of bearings, gripper/locking units, etc.
There are numerous different types of turret, some of which are driven and others are
freely weather-vaning. Turrets can be located at different positions on a FPSO and this
tends to influence the turret type. Active turrets are supported on sliding bearings while
other suppliers tend to use wheels or rollers. Table 12-1 summarises the advantages
and disadvantages of the two bearing approaches [Ref. 48].
A key concern from a mooring integrity point of view is if the turret fails to rotate
which could result in the FPSO becoming partially or totally beam on to survival storm
conditions. This may well lead to twisting of the mooring system which could cause
damage.
Active or driven turrets are not normally at the bow or the stern of a FPSO. Thus such
systems tend not to naturally weathervane. Hence, the FPSOs thrusters combined with
the turrets turning and locking system are used to turn the FPSO so it stays head on the
weather. It can thus be appreciated that an active turret is probably more susceptible to
FPSO power loss than a naturally weather-vaning turret. In practice blackship or no
power conditions have occurred in the past on active turrets which have led to the
FPSO being exposed to beam sea conditions. What is significant from a mooring
design point of view is that FPSOs with turrets are not analysed for survival beam sea
conditions. The wave frequency motion of a FPSO exposed to survival beam sea type
conditions will be high and in certain cases this could lead to extremely high mooring
line tensions. As well as blackship conditions active turrets can also be susceptible to
thrusters coming out of the water in extreme storm conditions. In such cases if has
been known for the thrusters to race in air, overload and trip. Even temporarily losing a
thruster in the middle of a storm is undesirable.
Good references for turret behaviour in the field are HSE Offshore Technology Report
2001/073 [Ref. 47] and Turret Operations in the North Sea: Experience from Norne
and Asgard A [Ref. 48].
165
A4163-01
-
From a mooring integrity point of view the key points for the two different turret
systems seem to be as follows:
If the active turret turning system or power supply fails or is operated wrongly
there is a danger that the FPSO could end up broadside to the waves.
However, the turning system includes redundancy, for example two of the
four cylinders have sufficient capacity to turn the turret, even for maximum
friction. These need to be designed for good access for servicing and repair.
- For the passive turret if the bearings fail and the turret seizes up there is a
danger of ending up broadside to the weather. A serious failure may require
talking a FPSO off station to dry dock. Depending on location, even in an
emergency it will take several days or weeks to put in place arrangements to
take a FPSO off station. Hence the FPSO could have to ride out storms
broadside to the weather, condition which the mooring lines are typically not
designed to be able to withstand. The probability of complete bearing is
likely to increase with age. It would be interesting to know what level of
bearing deterioration has been noted when FPSOs have been removed from
station at the end of a particular assignment. A related point is how quickly
these systems can deteriorate if, for some reason, there is inadequate
lubrication.
Active turrets do not utilise the turret turning and locking systems all the time. Instead
the system is only activated when the turret has absorbed about 7 degrees of twist. This
is different to passive turrets and hence it will be interesting to see if this results in any
different wear mechanisms than active turrets. In actual fact this will be difficult to
differentiate since passive turrets tend to be stoppered off at the base of the turret while
active turrets have a gypsy wheel arrangement at the turret base. Wear at the gypsy
wheel may be more similar to that which is typically encountered on a semi
submersible submerged gypsy wheel fairlead.
Recommendation
A check should be made on a typical FPSO to see how great the increases in line
tension are if the vessel cannot weathervane and thus has to ride out a storm broadside
to the weather.
12.1.1 Line Tension Behaviour over Time
An interesting question is whether on active turrets the mooring line tensions have
decreased over time due to straightening of the chain on the sea bed? On two
Norwegian FPSO the line tension monitoring has not revealed any tension values close
to the maximum design values. The line tension has not been found to decrease (or
increase) significantly during the first years of operation.
Fatigue cracking was experienced on the grippers of a Norwegian FPSO at a stress hot
spot. All grippers on this unit have been upgraded to improve their fatigue
performance by removing the sharp notch.
166
A4163-01
Sliding Bearings Roller Bearings
Advantages - Extremely high - Passive system
vertical load capacity requiring no daily
- Redundant system
operations
allowing partial repair - Promotes passive
or substitution weathervaning, hence
- Stable turret position
suitable for vessel with
limited or no thrusters
- Minimum wear on
swivel
- Less risk of human
errors
- Wide fabrication
tolerances
- Adapts to vessel
deformations, hence
promotes a central
turret position with
minimum riser loads
Disadvantages - Active turret turning - Non redundant system
system needed (failure leads to
- Daily turning
Stuck Turret)
operations - Greater wear on swivel
- More frequent
maintenance
due to frequent
rotations
- Risk of excessive twist
in case of turning
- Small fabrication
tolerances
system failure or faulty - Vulnerable to vessel
operation possibility deformation
of uncontrolled twist
back. This may
happen if the torque
from the mooring
system overcomes the
frictional torque from
- The forward turret
position gives riser to
higher riser motions
and increased mooring
loads
the bearings.
Table 12-1 - Summary of the Pros and Cons of Sliding and Roller Bearings [Ref. 48]
167
A4163-01
12.1.2 How Marine Issues fit in with Normal Operations
It is important to be aware that turret operation/mooring behaviour is normally a
secondary task for control room operators. Their main task is to keep the topsides
process plant running and optimise the production of oil and gas. Given the importance
of moorings it is vital that regular emergency drills are undertaken to keep training up
to date.
On one Norwegian FPSO, to achieve optimum ventilation of the vessel topsides
facilities the vessel is normally orientated against the weather with the wind
approaching on the bow port side. This will tend to result in a slight corkscrew motion,
which over time could influence the wear/fatigue behaviour of the mooring system.
It is important to understand that a Floating Production Platform is the whole hull plus
process equipment plus risers and moorings. Thus the whole system needs to be
considered as one inter-related unit. This is somewhat different to a fixed structure
where the supporting structure is highly unlikely to suffer from progressive failure.
12.2 Implications of Mechanical Repairs
Turrets and thrusters are mechanical systems which are typically operating for an
unusually long period of time in harsh environments. The option for dry dock
inspection is not normally available and on going production operations may limit
when and how inspection can be undertaken. Turret or thruster mechanical repairs can
have safety case implications and any repair operation can be difficult and potentially
dangerous. The following extract gives an idea of what can be involved in repairing a
thruster in situ.
Faced with removing a 13 ton thruster motor from deep within the
bowels of the vessel and transporting it to the beach for overhaul onshore,
the team instead chose to dismantle thruster 2 in-situ onboard the FPSO.
There were significant safety concerns regarding the removal of the motor
in its entirety. The lifting route involved a 15-foot vertical lift, cross-
hauling through the bulkhead hatch with only 90mm clearance in winter
sea conditions, lifting over thruster 1 and a lift through an engineering
space of 90-foot to reach the main deck. Due to these safety concerns, a
safer and better method was needed to achieve the overhaul. The eventual
solution was to strip down the motor into manageable pieces, onboard the
FPSO, and to remove the rotor and bell housing for repair onshore. On
completion of the repair, the rotor and bell housing were successfully
reinstated using state-of-the-art technology from specialist vendors to
realign the motor onboard. The entire process was achieved without any
disruption to production.
Active turrets also make use of computer controlled systems. Issues can arise of
computer/software obsolescence. This needs to be taken account of in the
planned maintenance system.
168
A4163-01
13 GENERAL TRENDS AND STATISTICS
13.1 Questionnaire Process
A custom designed questionnaire was developed to undertake an international survey of
worldwide FPS operations including FPSOs, production semis and Spars. A typical
custom designed questionnaire is included in appendix B. The questionnaire is based
on an Excel spreadsheet based with drop down boxes in an attempt to make completion
as quick and easy as possible. The questionnaire was partially filled in making use of
information in the public domain before emailing to Operators and Contractors. If the
answer to certain issues is yes a standard series of questions are generated in a new
worksheet. A filled out example is also included as part of the spreadsheet.
m
Sep
No
No
No
A. GENERAL DETAILS
A1. Unit Name A2. Field Name
A3. Unit Type
A4. Water Depth A5. Geographical Area
A6. Date Installed
A7. Is the FPS classed?
A8. Has the unit ever been used elsewhere?
A9. Was the unit ever removed from site and then re-installed?
A10. Can the mooring system be disconnected in case of typhoons or ice bergs ?
A11. If the moorings can be disconnected, how often has this happened to date ?
JOINT INDUSTRY PROJECT: FPS MOORING INTEGRITY
QUESTIONNAIRE
Kuito FPSO Kuito, offshore Angola
Spread FPSO
383 West Africa
1999
Classification Society:
??????
Yes
N.A.
Table 13-1 - Example of the First Page of the Questionnaire see appendix B for a Full
Listing
The questionnaire is now available in the public domain and it is hoped that the format
for reporting incidents will continue to be of use to the offshore community after the
completion of the JIP.
169
A4163-01
13.1.1 Difficulties Obtaining Data
Initially it was hoped that offshore personnel would be able to complete the majority of
the questionnaires. Despite considerable time spent preparing and chasing up
questionnaires and the ease of email communication, it became clear that the
questionnaire was not given a high priority by hard pressed offshore and office based
personnel. A degree of past knowledge is necessary to complete the questionnaire
properly and with personnel change outs on units this knowledge can easily get lost.
This is itself is a somewhat worrying result for such complicated production facilities.
Getting detailed information on units operating outside the North Sea was particularly
difficult. The problem was not necessarily lack of interest, just a lack of time with
operational issues taking precedence. It is noted that good data could be obtained by
visiting FPSs and auditing the condition of the set up of their mooring systems and
reviewing inspection records. Quite often there is reasonable data available and the key
problem is gaining access to this data which may not be centrally stored.
Although it was difficult to get data on as many units as initially hoped, the data which
was obtained was in general of high quality (see Case Studies) and is believed to be pof
relevance to the vast majority of FPSs in the world today.
13.1.2 International Survey of FPS Experience
Types of Units and Geographical Location
The international survey has been an important part of the JIP. The approach has been
to collate data in a similar fashion to the UKOOA FPSO study but extending it to the
worldwide fleet of FPSOs, Production Semi submersibles and Spars. Mooring
performance will depend on the type of unit considered and the environment which it is
exposed to. The following generic types have been considered:
- North Sea turret moored FPSO
- North Sea Semi-sub FPS
- West African spread moored FPSO
- Brazilian deepwater FPSOs (turret and spread moored)
- Brazilian Semi-sub FPS (very limited data received)
- Gulf of Mexico Spar
- South East Asia FPSO
- Special FPSOs e.g. dis-connectable and ice resistant
- Worldwide FSU (limited investigation)
170
A4163-01
13.2 Summary Statistics for Unit Type and Geographical Area
The following graphs summarise some of the more reliable and significant results
obtained from post processing the returned questionnaires. Only statistics where a
reasonable sample size was acheievd have been reported.
North Sea Turret Moored FPSOs
Lines
Adj
Adjustable Line Lengths
Locked Off
50%
Can Be
usted
50%
67%
33%
Real Time Offset Monitoring
Units with
Unit s
witho ut
50%
Real Time Line Tension Monitoring
Unit s with
50%
Unit s
witho ut
171
A4163-01
78%
Line Failure Alarms
Units with
22%
Units
witho ut
Mooring Line Spares
Unit s with
33%
Units
wit ho ut
67%
87%
Existing Repair Procedures
Units with
13%
Unit s
witho ut
172
A4163-01
It is interesting to compare the average mooring line inspection periods for various
different types of FPS units in different locations. This is illustrated in Figure 13-1.
5
0
1
2
3
4
5
1.2
1.6
0.5
1.5
2.5
3.5
4.5
Years
Average Inspection Periods
North Sea Turret Rest of World Turret Rest of World Spread
Figure 13-1 - Comparison of Mooring Line Inspection Periods for Different FPS
Categories
173
A4163-01
13.3 HSE UK Sector and Norwegian Statistics
Interrogation of existing health and safety databases provided useful validated data
see Table 13-2, Table 13-3 and Table 13-4.
96/97 97/98 98/99 99/00 00/01 01/02 Total
Moorings/DP 0 0 1 2 1 3 7
FPSO/FSU
Total
Incidents
11 10 15 22 10 11 79
Mooring/DP
percentage of
FPSO/FSU
incidents
0% 0% 6.7% 9.1% 10% 27.3% 8.9%
Table 13-2 - UK Sector of the North Sea Data [Ref. 49]
Period 1980 to 2001 (ORION database)
Drilling Semis Production Semis Accommodation
Semis
FPSOs
N F N F N F N F
Anchor
Failure
170 0.211 8 0.111 23 8 0.113
Table 13-3 - UK Sector of the North Sea Data [Ref. 49]
Where N = number of events and F = occurrences per unit year. Anchor failure defined
as Problems with anchor/anchor lines, mooring devices, winching equipment or
fairleads (e.g. anchor dragging, breaking of mooring lines, loss of anchor(s), winch
failures.
Incident Description Mobile Drilling Units Production
Single Failure 9 3
Multiple Failures 3 -
Table 13-4 Number of Anchor Incidents in the Period of 1990-2003 in the Norwegian
Sector [Ref. 50]
174
A4163-01
Even though considerable resources have been devoted to the international survey, it is
quite possible that that only a fraction of the total number of mooring incidents which
have occurred outside of 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 the 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.
These statistics indicate that it would not be totally unexpected for the crew on a FPS to
expect a mooring line failure sometime during a field life which exceeds 9 years.
Exactly how these statistics can be related to milder environments is difficult to
estimate without access to more data. But it is worth noting that but fatigue may be
more of an issue for milder locations - see Figure 15-11.
This concern that offshore staff on a FPS should always be prepared and ready to react
appropriately in case of mooring line failure is reflected in the following quote from
from Ref. 3. Although the background of this paper is mooring operations of semi-
submersible units, the basic sentiment is felt also to be appropriate for FPS units in
general.
The high failure rate of individual mooring lines means that a unit must
always be prepared to deal with a mooring line failure as an almost routine
operational matter. It is unrealistic to operate a unit on the assumption that a
mooring line failure is unlikely to occur.
It is not perhaps surprising that when FPSs first became more prevalent that their
mooring systems suffered some initial teething troubles as new designs were
introduced. These early problems seem to have been largely resolved and thus in
general terms one might expect a plateau period of mooring failures. What will be
interesting to see is how the number of failures increase as the mooring lines age and
are subject to corrosion, wear and fatigue.
As a FPS operator, it is probably helpful to think in terms of average historical line
failure rate per FPS unit operating year.
Figure 13-2 illustrates historical failure rates for different types of North Sea units. The
failure rate for FPSOs is closer to the UKOOA study reported value of once every 5.4
operating years. Given that North Sea designed FPSOs are carefully regulated it would
be expected that the reliability of these units should be good. Hence it is disappointing
that the failure rates for FPSOs is only slightly better than production semis and only
about twice as good as a drilling semis.
175
A4163-01
NDEs UKOOA report referred to DNVs data for 1980 to 1998, which reported 7
FPSO/FSU anchor system failures. This gave 0.186 failures per unit operating year or
one failure every 5.4 operating years. Hence, it can be seen that the failure rate seems
to have improved somewhat from 1998 to 2001.
0
1
2
3
4
5
6
7
8
9
10
Comparison of North Sea Failure Rates for Different Unit Types (1980 - 2001)
Drilling Semi Production Semi FPSO
Type of Unit
N
u
m
b
e
r

o
f

o
p
e
r
a
t
i
n
g

y
e
a
r
s

p
e
r

f
a
i
l
u
r
e

Figure 13-2 Historical Failure Rates for Different Types of Units
176
A4163-01
14 CONNECTORS AND TERMINATIONS
14.1 Background
Connectors and line terminations (e.g. spelter sockets or fibre rope splices) are vital
components in a mooring system, since they are typically necessary to join up different
types of line e.g. chain to wire, or to suit manufacturing/transportation limitations on
line lengths. However, the need for some type of opening and closing mechanism
means that, to achieve the required strength, the connectors tend to be heavier than the
lines to which they are attached. Thus connectors and line terminations tend to be areas
of discontinuity on a mooring system with respect to weight per metre and also bending
and torsional stiffness. This is because they are unlikely to flex in the precisely the
same way as the chain, wire or fibre rope to which they are connected.
In general, where there is a weight discontinuity on a mooring line there is an increase
in relative rotation. This rotation can result in wear plus possibly some fatigue loading.
Yet it is typically fairly difficult to inspect connectors in situ for wear see for example
Figure 18-15. Hence, due to the long-term effect of these degradation mechanisms,
failures have occurred see for example Section 10.2 and the problems associated with
traditional Kenters, Baldt, Pear and C Links on drilling rigs.
The design of chains and wire rope does not tend to change dramatically from one
project to another. This is not necessarily true of connectors which may need changes
for new applications. Hence this section attempts to summarise what connectors and
line terminations are available at present. It then goes on to consider what should be
taken account of when designing new connectors see the flow diagram in Section
14.3. The section concludes with any gaps in the existing knowledge base and
identifies topics which merit further investigation.
14.2 What Type of Connectors Can be Considered for Long Term Mooring
(LTM)
Class Societies typically have special requirements for Long term Mooring (LTM)
systems. For example DNV applies this categorisation for mooring systems which will
be at the same location for more than 5 years. LTM D shackles typically have a
double locking mechanism, such as a nut and locking pin restraining the primary
restraint system, see for example Figure 14-10.
The following section reviews the type of connectors which are fairly readily available
and might be considered for a long term mooring.
177
A4163-01
14.2.1 Traditional Kenter
The traditional design of Kenter link dates back to a 1905 patent by Max Kenter. The
patent description is as follows :
Improvements in Chain Coupling-links - A coupling-link for chains
consists of two similar parts a, b, which are adapted to engage laterally and
are locked in their engaged position by the grooved piece f. Movement of
the piece f is prevented by the inclined taper-pin h secured in place by the
lead plug i. [Ref. 52].
The great merits of a kenter are that it can pass through a gypsy wheel in the same way
as normal chain and its dynamic behaviour is very similar to chain, since it is of
comparable weight and geometric arrangement. If kenters are tight fitting and are
properly assembled with a lead plug added after assembly, they can perform quite well.
For example some drilling rigs end up with kenters in the thrash zone and these can last
quite well for a temporary application. Still assembly tolerances vary in practice, and
thus having a kenter in the thrash zone should only be considered as a temporary repair.
Kenters are discussed in more detail in Section 10.4.1. It should benoted that if the lead
plug comes out in service there is a danger that the kenter could open up.
14.2.2 Special Joining Shackle (SJS) and Shackle Pin Rotation
A SJS can be used to connect studless common link chain to studless common link
chain without the need for an enlarged end links, which would typically be required if a
normal shackle were to be used. Enlarged end links can only be added to a chain at a
Forge so their inclusion reduces flexibility with regard, for example, to trimming chain
during line hook up operations.
The bow of a SJS needs to be trimmed compared to a standard D shackle to allow
studless link connection. To qualify for LTM designation typically a double locking
mechanism for the shackle pin is required, as well as a demonstration of fatigue life. In
addition, the same quality material is typically used for the shackle body, pin and
locking nut.
In the design illustrated in Figure 14-1 the pin of the shackle is oval which means that it
normally cannot rotate in the shackle body. Normally when two chain links rotate
against each other the surface profile and the harness for the links is very similar.
However, when a chain is connected to a SJS the surface hardnesses of the chain link
and the shackle, as well as the geometry may be somewhat different. If the
combination is such that this leads to accelerated wear of the chain link, which is of
thinner section than the shackle, then this could lead to early loss of integrity. In
addition, due to the weight discontinuity it is likely that there will be more relative
rotation between the link and the joining shackle.
178
A4163-01
Whether or not to allow pin rotation on a connector is a difficult question to answer. In
the case described in Section 10.2, the initial approach adopted to preventing pin
rotation was not strong enough and ultimately failed. The solution was to allow the
system to rotate and using a much stronger mechanism to keep the end plates in
position. For the H shackle illustrated in Figure 14-2 the pins are oval where they pass
through the shackle body. This is needed to make them small enough to pass through
the chain links. However, the holes in the H shackle body are round. This allows the
oval pin to rotate in the H shackle body. Since the H shackle body and pin are
oversized any wear in these components should not be significant. But it is important
that wear in the attached chain link should be minimised, since when it moves it should
not be grinding against a fixed, not rotating surface (pin).
Some precautions are possible; for example the shackle shown in Figure 14-2 had its
pins specifically fitted to the chain links and a set of baseline pin angle measurements
were taken using photographs so that they can be compared in the future to ROV
photographs. Whether this logic proves to be demonstrated in practice will only
become clear over time!
Figure 14-1 - Special Joining Shackle (courtesy of Vicinay Catalogue)
179
A4163-01
Figure 14-2 - H Shackle Pin Configuration (courtesy of I. Williams)
Other Types of Connectors
Other types of subsea connectors are illustrated in Figure 14-3. Requirements for
subsea connectors are discussed in more depth in section 6.2.4. More temporary types
of connectors include Baldt or C connectors and pear links. Sometimes these are a
rattling good fit and the general perception is that they fail more often than kenters.
They should not be considered for long term mooring systems, even for temporary
fixes, unless they are all that is available. Standard kenters are better machined, fit
tighter and are more suitable for a short term repair.
Subsea
Designed to
facilitate
Connector
connection &
(Delmar)
disconnection
Female Male
Female and Male being assembled
180
A4163-01
Subsea
Connector
(Ballgrab)
Designed to
facilitate
connection &
disconnection
Figure 14-3 Illustration of Subsea Connectors which have been used on Pre-Installed
Mooring Lines
Figure 14-4 - Example of a Special Joining Plate - Note Electrical Isolating Bush
181
A4163-01
14.3 Terminations General
To be of practical use a rope must be provided with means for connecting it at its ends
into the mooring system. The task of the termination is to transfer the predominantly
axial load in the rope into an engineering component which can be attached to standard
mechanical/structural components which form part of the platform being moored.
Nearly all rope terminations depend to some degree on developing radial forces - and
through them friction - to allow the axial load in the rope to be transferred to another
element. The splice is the basic example of this in which, when the rope is placed in
tension, the geometry of the splice generates radial loads between the rope strands and
these allow sufficient friction forces to be generated to transfer the load from one strand
to another. In other terminations the radial forces are generated by means of a
mechanical device.
It is well documented that during break or fatigue testing many rope specimens are seen
to fail at or very close to the terminations. This is due to the additional stresses
introduced into the rope at or close to the termination.
The termination components may have to support many other additional loads, such as
bending and shear, other than the axial load in the rope. Finally the termination may
have to survive abrasion, fatigue and corrosion.
14.3.1 Spelter Sockets
Splicing of wire ropes is complex and difficult on account of the weight and stiffness of
the large size of typical mooring ropes. An alternative which has developed is the use
of the spelter socket (see Figure 14-7) in which the rope is inserted into a metal collar
with an internal conical hole. The wire rope is cleaned and teased out to form a brush
which adopts the internal conical space of the termination. Into the cone is poured a
molten spelter alloy which solidifies and acts to grip the wire when it is pulled.
Figure 14-5 Example of the Make Up of a Typical Closed Spelter Socket (courtesy of
Bridon)
182
A4163-01
Wire rope terminations or sockets can be either open (Figure 14-6) or closed (Figure
14-7).
Figure 14-6 - Example of an Open Socket
Figure 14-7 - Example of a Closed Socket
An alternative to using spelter alloy for terminating wire ropes is to use an epoxy resin.
This has become a popular and efficient method of terminating wire rope. Potted
sockets have the advantage that they can be applied in the field if necessary without the
complications of providing a means of melting the spelter alloy.
The potted resin socket has also been used for terminating fibre ropes. As in the case of
wire rope the end of the rope is inserted into a socket and the yarns/strands splayed out.
A compound such as epoxy is then used to fill the socket which, when it sets, forms a
strong bond to the fibre material and socket. Conventional wire rope sockets have been
used successfully on small aramid ropes.
183
A4163-01
Although epoxy potting has been used for some small size fibre ropes it has not proved
effective for larger ropes due to the difficulty in providing sufficient circumferential
area over which to distribute the shear loads needed to carry the axial load out of the
rope. Solutions have been proposed in which the rope is divided into component
strands each with its own potted termination but so far the complexity of this approach
has been a major obstacle to its use.
It is important that, when being prepared for potting or speltering, a rope must be
accurately set so that the termination is not at an angle to the axis if the rope. If this
happens the rope will be subjected to a degree of bending when the load is applied and
this can seriously weaken the capacity of the rope. This effect is particularly apparent
when only a short specimen of rope is being terminated as in such cases the limited
length of the rope means that the uneven loading over the rope cross section has less
chance of being absorbed in the stretch of the rope. This leads to a concentration of
bending effect and earlier failure. This can be particularly important when preparing
short specimens for prototype testing.
14.3.2 Spelter Socket Fatigue Assessment (S-N curves) + Bend Limiters
At present during detailed design S-N curves for common link chain seem to be often
applied. This is not really appropriate and requires further consideration on live
projects.
The Bend stiffener and attachment mechanism also needs to be suitably designed, see
for example the damage shown on Figure 6-6.
Other areas to consider for sockets include:-
- the potential for pin rotation,
- the need for anti rotation keys,
- whether or not the spelter sockets are in a vertical orientation.
If the spelter sockets are not vertical they will be subject to cyclical bending stresses,
which over time might cause a fatigue problem, depending on their design. It is also
important to be able to check whether the insulating PTFE bushes are present or not.
14.3.3 Fibre Rope Splices
This is the prevalent form of termination for fibre ropes throughout the industry. There
is a large degree of experience available when considering terminating large diameter
fibre mooring ropes with eye splices. Other techniques (Grip and Potted) are much less
well documented. A splice in a polyester rope has been shown to have an efficiency
approaching 1.00, depending on the quality of the splice. It should be noted that the
certified minimum break load (MBL) of a fibre rope is that of the spliced rope ([Ref.
53] or OCIMF hawser guideline).
184
A4163-01
Splicing has been described as an Art; there is a movement within the rope industry
pushing for this to be changed to a Technology. There are papers which describe
modelling of splices (see for instance Ref. 67) and recent offshore projects in the Gulf
of Mexico have described various means of controlling and documenting splice
production in ropes with a breaking strength in the region of 2,000 tonnes.
In terms of fatigue testing, most assessments have been made with splice terminations.
Here the fatigue lives for polyester and aramid ropes are described as being well above
that of steel wire rope at normal working loads.
For large diameter synthetic fibre ropes under long term cyclic loading the only verified
technology for their termination is by use of a splice.
Two types of spliced eye hardware are described. The first is a construction involving
a metal thimble and shackle arrangement which has been used successfully for years at
single point mooring terminals. This type of connector is, however, described as
making rope handling particularly cumbersome and awkward [Ref. 68].
To-date most offshore experience of large fibre mooring ropes has relied on this type in
which the thimble can be slipped into a prepared soft-eye splice when the connection is
being made-up on the deck of the mooring installation vessel. The thimble is supported
by a shackle and provides a suitably large diameter over which the fibre rope can be
bent. Advice on the choice of spool diameter is available from guidance documents
such as Ref. 69. The thimble diameter should be large enough to develop the best axial
tension and fatigue strength while minimising abrasion and wear as the stretching rope
slides against the metal of the spool. In order to minimise this problem the splice eye is
often wrapped locally with a binding tape in order to minimise wear on the fibre rope.
When the line goes into tension the spliced eye pulls tight and prevents the thimble
from falling out. However, during over-boarding when there may low line tensions and
eccentric loads on the connection care must be taken to stop the line slipping-off the
thimble and being caught on the shackle instead.
It is important that the fitting between the eye and the spool is tight enough so that the
two do not become separated. This is described as being most likely to occur during
rope handling when the rope is slack.
This is often achieved through encapsulating the thimble in polyurethane. However if
the splice is provided with a permanently fitted hard eye in this way it means that there
are additional problems when handling the rope on its transportation and deployment
spools as the thimble must be prevented from abrading and damaging the rope on the
spool.
Various novel terminations are currently being proposed and evaluated. These seek to
improve the efficiency of terminations for very large ropes by splicing the sub-ropes
individually to themselves to create a number of eyes. These are then supported on
multiple pins. High performance materials (e.g. titanium or super duplex) are used in
order to minimise the weight of the metal components.
185
A4163-01
14.4 Connector/Termination Design Flow Chart
Figure 14-8 and Figure 14-9 together illustrate a connector design flow chart based on
approaching the subject with a blank piece of paper. The flow chart is also generally
applicable for the design of terminations. Comparing this flow chart with present day
design practice shows that the following:
- A wear analysis is typically not undertaken
- The dynamic motion of the connector is typically not evaluated
- Calculations are not typically undertaken to size locking pins based on high
line tensions and frictional forces
- Electrical isolation needs to be considered early on in the design process
- Inspection is not given a high priority during connector design.
186
A4163-01
NO
Could any existing
NO YES
UNSATISFACTORY
YES
designs be used?
Asses the pros &
cons of the existing
design & decide
whether to proceed
with modifications?
Develop a Design Brief specifying strength,
fatigue life, installation weight, material
properties etc
Obtain Client written approval of Design Brief
& Class agreement in principle
Requirement to join two lines
Assess the ease of deployment, including
connection/disconnection & the likelihood of
unintentional disconnection
Identify what is special about the new
application & propose a new design
Quality Plan for design & manufacture to
describe activities to be performed, frequency
and type of inspection/tests, criteria to be met as
well as give reference to applicable controlling
documents
Submit Quality Plan for Class Approval
Figure 14-8 - Connector or Termination Design Flow Diagram - Initial Phase
187
A4163-01
NO
Calculations
- Codes
Factors Identified?
- Past Observation
- Calculations
- Hand/Simple
Approval
& Long Term Insp
NO
Wear Assessment
- Proposed Methodology
- Justification
- Past Observation
- Guidelines
Fatigue Life Assessment
- Suitable S-N Curves
- Stress Concentration
- Physical Testing
- Hand/Simple
Calculations
- Tension/Tension
Assessment
- Tension/Bending
Assessment
- Cumulative Damage
Evaluation
Corrosion Assessment
- Avoidance e.g.Anodes
- Allowance
e.g.Corrosion Margin
- Physical Testing
Strength Assessment
- Finite Element Analysis
- Material Testing
- Physical Testing
Commence Detailed
Design Process
Manufacture in accordance
with Quality Plan & Class
Society inspections
Test Connector as per
Quality Plan
Documentation to be
stamped with Class
Issue Recommendations
for Connector
Transportation, Installation
ection
Finalise Design Reports &
Drawings; issue to Class
Society & Client
Design
Iterations
Successful?
YES
Figure 14-9 - Connector (Termination) Detailed Design Flow Chart
188
A4163-01
14.5 Detailed Design Guidance
There are numerous different types of mooring line connectors depending on whether
chain, wire or fibre ropes are being connected. Operational experience from a number
of floating production and drilling units has shown that failures quite often occur at
connectors. In certain instances the connectors have failed by coming undone when
their locking mechanisms have failed, see Section 10.2.
To ensure the long term reliability of a FPS mooring system it is important that the
design of ALL the proposed connectors in a mooring system should be carefully
reviewed. This also needs to take into account where on the catenary and on the sea-
bed the connectors are situated. The following general guidelines are provided for the
design and selection of connectors:
- Try to minimise connector weight to avoid increased rotation at each end due to
an abrupt change in weight per metre of the mooring line.
- Avoid placing connectors in the thrash zone.
- Allow shackle or spelter socket pins the ability to rotate.
- Add marks on connectors so that any wear can be measured.
- Ensure connectors have compliance in all required planes of motion see below.
- Match material characteristics to minimise wear.
- Pin locking devices need to take in to account applied loads and friction developed
lever arms.
- Secondary locking devices should be provided which are capable of withstanding
the full anticipated loading in case the primary device fails.
- Sharp edges where any pin is stepped down to the locking nuts should be rounded
to minimise the danger of cracking occurring.
- In general rounded rather than chamfered edges around the edge of the shackle
should minimise damage if contact occurs with chain links.
- It may be beneficial to put a groove in shackle/spelter socket pin to reduce point to
point contact. In other words contact between the link and the pin would be
similar to link to link contact.
- Thought should be given on how to handle the connector given its weight.
Welding lifting eyes on to it should be avoided.
- The weight distributions of the connector, including the locking mechanism,
should be balanced to ensure that it does not lie to one particular side.
189
A4163-01
Depending on configuration/pre-tension as a catenary leg is laid down on the sea-bed
and the sea-bed takes up the weight, the tension may drop down to zero. Hence if you
have a heavy connector on the sea bed you can see a lot of rotation at this point and thus
potential wear. Modern day dynamic mooring analysis programmes can assess if the
lines are likely to go slack at any time.
Chain, wire and fibre rope can all rotate in numerous planes. However, most connectors
have limited compliance and it is thus important to confirm that, where they are placed
on a mooring system, they can allow the predicted motion without resulting in lock up
and thus relatively sudden increases in bending stress. Figure 14-10 illustrates a
purpose designed X shackle for connecting between studless common links at the
base of a FPSO turret. In this case a dynamic analysis was undertaken to determine the
required compliance in two planes and then the V in the X shackle was designed to
suit, including a contingency factor, see Figure 14-2 and Figure 14-10.
When designing a connector the most appropriate proof stress should be assessed which
will give a good fatigue life see Section 14.5.
Figure 14-10 Illustration of a Purpose Designed connector allowing limited
compliance in Two Planes
190
A4163-01
Figure 14-11 - Example of a Dynamic Analysis to Estimate the Angle for the V Slot
Size on the H Shackle
Certain connectors, such as Long Term Mooring (LTM) D shackles, have been
available on the market for a number of years and thus have established track records.
Still it is prudent when selecting connectors to determine where they have been used
before and if any problems have been identified. Even with an established track record,
connectors must still be evaluated on a case by case basis, since different pretensions,
vessel motions and position on the mooring line can affect their long term behaviour.
14.5.1 Forging versus Casting
From a mooring perspective castings are normally considered only suitable for mass
produced, intricate items. For example, Wire rope sockets are cast because of their
inherent elaborate design. Forging is the preferred method of manufacturing mooring
accessories, since it helps to ensure that the items are strong but ductile. Impact
strength is a key element in any mooring accessory design.
The following list details the advantages of forgings versus castings [Ref. 51] :
- Higher strength
- Unmatched toughness
- Longer service life
- Higher structural integrity (absence of internal defects)
- Use of higher design stresses
191
A4163-01
- Greater transverse properties
- Increased safety margin (due to high ductility)
- Greater strength-to-weight ratio (lighter parts, reduced sections)
- 100% density (no porosity)
- Higher overall quality / reliability
- Reduced product liability concerns
- More acceptable parts / fewer concerns
- More uniform heat properties (lot to lot, and part to part)
- Directional or isotropic property profile
- More consistent machining (uniform microstructure and chemical
composition)
- Better hardness control for abrasion / wear resistance
- Extended warranties more probable on critical part / assemblies
- More versatile processing options and combinations
- Option to optimise the grain flow directions of the component
14.5.2 Discussion of Connector/Terminations Type Approval
At present it appears that each Classification Society has their own rules and
regulations for assessing whether a connector is suitable for a long term mooring
application. It would be helpful if, perhaps under the auspices of IACS (International
Association of Classification Societies), that a standard protocol could be developed for
designing and testing these connectors/terminations.
14.5.3 Fatigue and Wear Assessment of Connectors
API RP2SK provides the following Data for other types of connecting links (i.e. apart
from Kenter or Baldt links) are insufficient for generating design curves. Limited data
indicates that the fatigue life of D-shackles is comparable to that of common links of
the same size and grade, provided that the shackle is machined fit with close tolerance,
no cotter pin is used through the shackle body and the shackle is the narrow throat
type. Since spiral strand sockets are not D shackles or Baldt or Kenter links it can be
argued that it is not valid to assess their fatigue performance using a S-N curve for
common link chain based on API RP 2SK. However, this technique appears to have
been used on a number of North Sea FPSOs.
192
A4163-01
Apart for a few general types of connectors, such as kenters or Baldt links, specific S-N
curves are not available. In such cases it is common practice to assume that the
behaviour of a large, heavy connector will be superior to that of the chain or wire to
which it is connected. Considering the cost associated with mooring failure or
intervention, this is not a re-assuring situation in terms of long term system reliability.
Hence it is important that a valid fatigue assessment is undertaken for each individual
part of a mooring system. This should be reflected in the mooring design specification.
As is discussed further in Section 0, wear can be a significant issue over the life time of
a mooring system. Where there is an abrupt change in weight per metre of a mooring
line at a connector, greater relative rotation can be expected. This may not be too much
of an issue for the connector itself which may be oversized. However, over the long
term it may become an issue for whatever is connected up to the connector, e.g.
common link chain. This should be assessed on a case by case basis and if considered
to be a possible cause of concern, suitable analysis or testing should be undertaken.
14.6 Proof Load Testing and Its Impact on Fatigue
Chain strength is established by break and proof load tests. In a break test, a sample
length of chain is loaded in tension to an estimated break load value which it is held at
for 30 seconds without failure. The minimum breaking load specified is typically 75%
of the stress level for the minimum ultimate tensile strength. The proof load test is
nominally the highest load carried by the chain without deformation. The ratio of proof
load to breaking strength depends on chain grade see IACS W22 [Ref. 66]. In
studded mooring chain proof load is about two thirds of its break strength. For example
for setting the chain to its final shape and locking the studs into position, a proof load
based on 90% of the load at minimum yield stress, or 78% of the minimum breaking
load is typical.
The proof load concept needs some further refinement for materials, such as certain
higher strength steels, which do not exhibit a particularly defined yield point. These
materials initially respond under load with a linear elastic response but this gradually
softens until the ultimate tensile strength (UTS) is reached. As no yield point can be
used as a reference point, the concept of proof stress has been advanced, essentially as a
material property, where the 0.1% proof stress corresponds to the stress from which a
0.1% strain permanent set would result following elastic unloading. Depending on the
material, the proof stress (PS) may be defined variously as a 0.1% or 0.2% proof stress
see Figure 14-12 and Figure 3-34. From a materials perspective a proof test is to
confirm that the degree of elongation under application of the proof load (stress) is not
exceeded.
193
A4163-01
Figure 14-12 - Example of Material with a Non Clearly Defined Yield Point
Proof loading of chain is carried out for a number of purposes including to check the
stiffness (elongation) of chain and to ensure the studs are fixed following heat
treatment which otherwise relaxes the initial clamping forces applied. Proof loads are
generally defined in codes and standards as a proportion of the minimum 0.2% proof
stress or minimum UTS combined with the nominal section area.
Proof loading of the chain into the plastic range leaves a small permanent set when the
load is removed. The component geometries means this induces locked in residual
stresses in the chain and these are compressive at the inner shoulders of the links. This
means that applied tensions have to reverse the residual compressive stresses before
tension is induced in these fatigue prone areas and the proof loading may therefore be
considered to be beneficial to fatigue endurance.
Evidence for this was obtained, inadvertently in the BOMEL JIP. The first two tension
fatigue tests delivered extraordinary results and were halted, without evidence of
cracking, when the predicted lives were exceeded by a factor of around three. In
consultation with the chain supplier, it was concluded that the chain had been subjected
to an excessive proof load during manufacture. It was noted that such treatment was
allowed under the specification and the practice is not uncommon to stretch the chain
when it is under-length.
194
A4163-01
Amoco (one of the BOMEL JIP sponsors) proceeded to undertake further testing to
assess proof loading effects on the fatigue life of chains, in conjunction with the
University of Tulsa [Ref. 54]. A range of proof load levels was investigated (up to
82% of break strength) and it was concluded that proof loading substantially increases
the fatigue life and this was attributed to the residual stresses generated. Importantly
they noted that in addition to the level of proof loading, the ability of the material to
sustain the residual stresses without redistribution under cyclic loading was another
factor affecting the consequences for fatigue life.
A difficulty for high strength (e.g. R4) chain is that the proof stress is cited with
reference to the minimum UTS/breaking load. If a batch has a significantly higher
UTS (which in many senses is desirable), the degree of plasticity brought about at the
proof load level may be significantly less than assumed with the minimum specification
material. This means that the degree of residual compressive stress within the links
will vary depending on the actual material properties. Furthermore with the high
strength steels used for chain, the proof stress to UTS ratio can exceed 0.95, something
that is generally precluded with a limit around 0.85 in steel for structural purposes.
High actual PS/UTS ratios would further limit the degree of plastic deformation /
residual compressive stresses achieved through the standard proof loading procedure.
A more consistent approach for specifying the proof load would be in relation to the
batch UTS, something that is invariably tested.
Although the above discussion relates the effects of proof loading to the consequences
in terms of fatigue performance, in the case of studlink chain appropriate levels of
proof loading are equally important. If the degree of plasticity achieved under proof
loading is less than anticipated, studs will be more likely to become loose in service.
The above discussion highlights the importance of:
- Developing a more meaningful specification for effective proof loading during
chain manufacture
- Undertaking research (using finite element analysis and physical testing) to
define any beneficial effect of proof loading of chain for fatigue performance
and translating this into manufacturing specifications, as appropriate.
14.6.1 Break Testing Rate of Load Application
OTC paper 10798 1999 [Ref. 70] states on page 268 that a connector failed the
minimum break load test because the rate of load application was faster than had been
specified by the manufacturers test procedure. Another connector plate was tested
using a slower rate of load application and passed the minimum break load test.
195
A4163-01
This is of considerable interest since, depending on the slow drift offset of the FPS, a
mooring system can experience very rapid increases in line tensions due to first order
wave frequency response. As far as can be determined classification societies
(including IACS) typically do not specify the rate of application of the break test load.
Hence it is believed that this is a matter which requires further consideration, since
intrinsically it is not desirable to have a means of testing which:
a) poorly replicates the offshore loading environment,
b) can lead to either a pass or a fail depending how the test is done.
Whilst performing a "real" scale test would be a rigorous/ideal way forward, it is
unknown whether all the test houses / manufacturers would be able to undertake this for
the larger sized items and more research is needed. In general the ramping period does
not appear to be recorded unless specifically requested by reference to a manufacturer's
or client specification. This seems a simple and informative piece of information which
should, by default, be included in the as built documentation.
It is appreciated that failure through shock loading may be caused by brittle failure
instead of the traditional tensile failures (necking). Therefore, one option may be to
have more onerous requirements for impact tests or at lower temperatures. This may be
of more particular interest with high tensile steels which may be more prone to the
brittle failure due to defects / flaws. But still there is a need to correlate this back to the
ramping speed break test performance of real connectors.
196
A4163-01
15 OUT OF PLANE BENDING CHAIN AND ROPES
(FIBRE + WIRE)
15.1 Tension Bending at a Wildcat and its Effect on Fatigue
As chain is fed over a fairlead it is subjected to tension plus an out of plane bending
moment resulting from the local geometry of the contact between chain link and gypsy
wheel bearing surfaces. Fatigue issues associated with chain links in gypsy wheel
fairleads have been reasonably well documented and incorporated in design practice
(API 2SK RP2SK [Ref. 31] and DNV OS-E301 [Ref. 5] for example).
Despite this, failure of mooring systems continues to occur. Fatigue calculations are
often restricted to links in the catenary, neglecting the reduced fatigue life local to the
end termination. One North Sea unit suffered a link breakage at the fairlead, after only
one winter.
Figure 15-1 Broken Link from Fairlead Figure 15-2 Mechanical Damage on
Fairlead Link
The photographs above demonstrate both fatigue cracking and the mechanical damage
with can result from the high stresses experienced by a chain link at a fairlead.
197
A4163-01
15.1.1 The Load Mechanism
The load mechanism is related to the eccentricity of thrust lines with respect to the bar
neutral axis (centroid). This effect is enhanced where the change of angle is greater
(for a five pocket as opposed to a seven or a nine pocket fairlead). It may also be
increased where wear of the fairlead moves the contact point away from the end of the
link. The imposed hogging moment is balanced by a counter effect at the other end of
the link.
The chain links within the fairlead are thus subjected to an out of plane bending
moment which is proportional to the tension in the mooring line. Tension variations in
the mooring line result in a stress range due to both the axial and out of plane loading in
the link.
Figure 15-3 - Support of a Link in a Wheel Fairlead
The fluctuating bending stress in the link is referred to as tension bending in API and
bending of the chain links in the fairleads in DNV. In both cases the fatigue
mechanism appears to be identical to that examined by BOMEL in their anchor chain
JIP see Section 15.1.2. Fatigue damage (SN) curves and implied stress concentration
factors are broadly consistent between the three sources.
15.1.2 Tension Bending Fatigue Testing Undertaken by BOMEL
During the early 1990s BOMEL conducted a Joint Industry Project into the design of
anchor chains [Ref. 10]. As part of this study work they conducted a series of tests to
examine the influence of the tension bending effect on mooring line fatigue.
198
A4163-01
Figure 15-4 - Photograph of Test Link Showing Bearing Plates [Ref. 10]
The programme included the use of a test rig representing a five pocket fairlead (see
Figure 15-4 and Figure 15-5). The horizontal link was supported at four points on a
mounting point which was cycled vertically in order to develop varying tension in the
two fixed links. The tests were conducted on 54 mm diameter K3 chain with welded
studs.
Figure 15-5 - General View of Tension Bending Test Rig (protective screens removed
for clarity) [Ref. 10]
199
A4163-01
BOMEL monitored the mean and range of the imposed loads in the horizontal link and
used strain gauge readings to assess the response of the chain. The result set was
limited to 6 tests. The test conditions represented a different number of stress ranges.
The figure below demonstrates damage to the mounting plates during the first test. It
can be seen that two of the four hardened bed plates are cracked diagonally opposite
each other. The wear marks on the plates and on the links indicate that the two which
cracked were more heavily loaded than the other pair. The crack in the link occurred
over one of the fractured bed plates, also indicating heavier loading at this location.
These wear and crack locations demonstrate the significance of twist or out of
flatness in the unstressed link. The initial out of flatness was measured for all
subsequent tests see Figure 15-8. As can be seen in Figure 15-7 in certain cases out
of flatness can be quite significant.
Figure 15-6 - Broken Hardened Plates at the end of the First Test [Ref. 10]
200
A4163-01
Figure 15-7 - Twisted Link Due to Mis-aligned Butt Weld [Ref. 10]
When set in the fairlead the link position support restraints include bearing of the
shoulders of the link against the fairlead plus of the bend at each end against the
corresponding sections of adjacent links. As a result imperfection in any of the three
links, or in the fairlead itself, can prevent the link from initially bearing on four
shoulders.
As the link is relatively stiff BOMEL found that the load required to deform it
sufficiently that load is transferred through all four bearing points may approach or even
exceed the 0.2% permanent strain value.
Figure 15-8 - Simple Out of Flatness Twist Measurement Jig [Ref. 10]
201
A4163-01
15.1.3 Conclusions Regarding Fatigue Life Calculation
Despite the limited data set, BOMEL presented a fairly consistent relationship between
tension range and number of cycles to failure. Scatter appeared to be associated with
initial out of flatness in certain links.
A factor on the nominal stress in the link was defined as the local (measured) stress
divided by the nominal link stress.
Fac
nom
= Factor on nominal stress = Local (Measured) stress/ Nominal link stress
Nominal link stress = LineTension/Area = LineTension /(2 x x (Bar Dia/2)
2
)
Nominal link stress = (2 x LineTension) / x (Bar Dia)
2

The measured Fac
nom
value associated with two point bending was 5.2. A lower Fac
nom
of 3.6 was measured under 4 point bending. The two point bending Fac
nom
compares
well with a value of 5.9 derived from the difference between the intercept for the
fatigue performance curves for tension bending and for pure tension (all conducted in
the same JIP). The stress factors for a given material can be compared using an
expression of the form given below.
log( A )
sion TensionTen
)log( A
ding TensionBen
Fac
m
ding TensionBen nom )
10 Fac
sion TensionTen (
This expression can be re-arranged as follows for a pedagogic understanding :
1 1
m Fac
ding TensionBen nom )
(A
ding TensionBen
) Fac
sion TensionTen nom )
(A
sion TensionTen
)m
( (
Where m = the slope of the TN curve.
The stress factors of 2.5 and 1.5 quoted in DNV for 7 and 9 pocket fairleads are applied
to the in-plane bending stress for a link in the catenary. BOMEL record a factor of 3.7
on the nominal stress for this condition. If we use this to convert the DNV values to
factors on the nominal stress we produce comparative values of 9.3 and 5.6.
202
A4163-01
Stress factors for 7 and 9 pocket wildcats are summarised below:
DNV 7 pocket
Comparative factor of 9.3 <= 2.5 x 3.7
Factor on fatigue life of 0.06 <= 2.5^(-3.00).
DNV 9 pocket
Comparative factor of 5.6 <= 1.5 x 3.7
Factor on fatigue life of 0.30 <= 1.5^(-3.00).
API
Effective stress factor of 1.61 <= 0.2^(-1/3.36).
Comparative factor of 6.0 <= 1.61 x 3.7
API quote an upper bound fatigue life reduction factor of 0.2 which given the curve
gradient of 3.36 amounts to an effective stress factor of 1.61, in this case producing a
comparative factor of 6.0 on the nominal stress in the link.
Source SN Curve Factor on Nominal Stress Factor on T-T Fatigue Life
Gradient 5 Pkt 7 Pkt 9 Pkt 5 Pkt 7 Pkt 9 Pkt
BOMEL JIP 3.173 5.9 0.23
DNV OSE301 3.00 9.3 5.6 0.06 0.30
API RP2SK 3.36 9.0-6.0 0.05-0.20
Table 15-1 Comparison between Chain Tension-Bending Fatigue Parameters
Note that values in italics are derived from BOMEL measured stress factor.
Clearly both API and DNV take account of the twist / two point bearing effect and the
impact that this has on fatigue at wheel fairleads. The upper bound factor from API is
generally consistent with the BOMEL results for 2 point bending, with the DNV
guidance appearing somewhat more onerous.
Use of the two point bending factor can be further justified by consideration of stiffness
of a chain link. Even for a relatively perfect bearing geometry (0.8 mm link out of
flatness on a machined support bed) approximately half of a load cycle occurs under
two point bending. For a link with 3 mm out of flatness in the test rig the entire load
cycle occurs under two point bending. Clearly this indicates that the higher SCF is
applicable for fatigue calculations, as reflected in both DNV and API guidance.
203
A4163-01
Figure 15-9 - Illustration of Failed Link Due to Tension Bending [Ref. 10]
Reference should also be made to Vargas et al Stress concentration factors for studless
mooring chain links in fairleads, OMAE 2004-51376 which tends to demonstrate that
the usual SCFs recommended by some Rules are overestimated. In addition, DNVs
requirement for stress factors of 2.5 and 1.5 for 7 and 9 pocket fairleads is no longer
present in the latest edition of OS-E301 (Oct 2004) [Ref. 5].
15.1.4 Issues Affecting Operations and Inspection
The progress of fatigue damage involves several separated phases. In the early stages
the changes in the material microstructure will not be detectable. It is not until the
development of actual cracking that detection becomes possible. Even then it may not
be possible to detect relatively large cracks without the use of NDE techniques. Upon
development of the first crack detectable under laboratory conditions BOMEL quote an
average proportion of life remaining of 34 %, though the value is low as 17% in some
tests. Other investigators e.g. Vicinay have reported higher values. Detecting critical
crack size is obviously important for different material types. Further information can
be found in Section 16 on Fracture Mechanics.
In some test cases propagation of the fatigue crack through the bar thickness did not
lead to unloading of the system. It was noted that in a real fairlead, the containment
offered could result in a cracked link supporting line tension. It is suggested that this
could explain some reported failures occurring shortly after anchor repositioning. On
this basis it could be recommended that subsequent use of fairlead chain sections in the
catenary should be avoided unless detailed inspection of the chain has been undertaken.
However, from a practical point of view this is unlikely to feasible.
204
A4163-01
API makes recommendations regarding the design of the fairlead and management of
the line to ensure that a link is only exposed to tension bending for a limited period of
time.
15.2 Tension Bending at Chainhawse
15.2.1 Girassol Offloading Buoy Experience
This mooring system suffered a series of line failures well within the design life of the
system. Subsequent engineering review of the system indicated that the principal
failures were driven by fatigue damage.
The Girassol offloading buoy [Ref. 55] is a 19 m outside diameter circular buoy
moored in approximately 1,320 m water depth. The original mooring system
comprised 3 x 3 mooring lines. Each line was made up principally of polyester line
with chain sections at upper and lower terminations. It was designed to act as a taut
system with working tensions of approximately 95 Te.
Figure 15-10 - Girassol Offloading Buoy [Ref. 55]
Approximately 7 months after installation of the system one of the mooring lines failed,
rapidly followed by the two remaining lines in that group. The first two failures were
caused by a breakage of the fifth link from the chain stopper. The third line failed due
to overload of a section of the polyester line which appears to have been damaged
during installation (see Figure 15-12). The corresponding chain link on the last line
was also found to have suffered fatigue damage.
Subsequent to the first three breakages, failures have occurred in both of the remaining
groups of mooring lines. Clearly the system was subject to much more onerous fatigue
conditions than had been considered during design.
205
A4163-01
Figure 15-11 - Girassol Offloading Buoy Failure in Chain Link 5 [Ref. 55]
Figure 15-12 - Girassol Offloading Buoy Failure in Polyester Rope [Ref. 55]
206
A4163-01
A review of the original design calculations indicated that the design had been
generally consistent with the load mechanisms within the body of the chain. Repeated
failures at the fifth link however indicated that an additional effect must be inducing
greater fatigue loading at this location.
15.2.2 Explanation of the Failure Mechanism
It was determined that fatigue cracking of the fifth link in each mooring line was the
result of a cyclic out of plane bending moment applied to that link. It should be noted
that the geometry of the chain trumpet essentially restrains links 1 to 4 to move with the
trumpet. Rotation between the buoy and mooring line is concentrated on link 5 and to a
lesser extent on subsequent links.
Figure 15-13 - Chainhawser Arrangement and Location of Critical Link [Ref. 55]
Under low tension this rotation would be provided by slip between the links. Under
higher tensions a significant inter link friction has to be overcome before this slip can
occur as well as local flattening at the point of contact. These effects result in torsion in
the bar at the contact point, which is represented by out of plane bending in the body of
the link.
207
A4163-01
Figure 15-14 - Out of Plane Bending Mechanism (See Section 25 [Ref. 56]
The critical bending stress required to overcome interlink friction is proportional to the
line tension and friction coefficient. For a friction coefficient of 0.1 the nominal critical
bending stress amounts to approximately 40% of the nominal axial stress (total value,
not range) in the chain, which is surprisingly high value. The local flattening
/embedment pushes this figure higher.
n Calculatio
3 2
2tD 2tD TD
Z , A , M , 1 . 0
nom nom nom
32 4 2
2
o
bnom
M A TD 32 2tD
nom nom
40 . 0
3
o Z T 2 2tD T 4
tnom nom
Notation:
D chain bar diameter (m)
T Tension force (kN)
A Area (m
2
)
M Bending Moment (kN.m)
Z bar section modulus (m
3
)
o Stress due to bending or tension (kPa)
Friction coefficient
nom
Suffix nominal
b
Suffix bending
t
Suffix tension
208
A4163-01
Notes :
i) The bending component reverses, doubling the range of extreme fibre bending
stress.
ii) Stress will tend to cycle tension to tension.
iii) The above comparison is based on nominal stresses. It is anticipated that the SCF
associated with interlink friction will be less than that for a link in the catenary
under a tension range.
Individual links are extremely stiff. As a result very high out of plane bending moments
and thus stresses can be developed, even at relatively small angular deflections. Under
high tensions the interlink behaviour thus prevents links from slipping, providing a
beam like behaviour in the chain sections.
15.2.3 Physical Testing Following the Girassol Failure
SBM carried out a series of tests using a rig designed to reproduce the load regime at
link 5. Strain gauges were mounted on the link in order to quantify the out of plane
bending stress developed. The test was carried out under both dry and in-water
conditions.
Figure 15-15 - Schematic of SBM Test Rig [Ref. 55]
From their laboratory work SBM derived a relationship between interlink angle and the
out of plane bending stress in the link. The measured value amounted to 288MPa
change in extreme stress for an angle of 1 degree between consecutive links.
209
A4163-01
Using this relationship with the DNV B1 curve [Ref. 57] and a limiting range in
interlink angle of 3.8 degrees (imposed by geometry of chainhawse) SBM computed a
life to failure of 107 days. This calculation assumed that the full pitch rotation was
imposed on a single link, and that the friction coefficient was sufficient that no slip
occurred up to the limiting angle of 3.8 degrees (amplitude) where the 7
th
link touches
the chainhawse.
Figure 15-16 - Photograph of SBMs Test Rig [Ref. 55]
210
A4163-01
15.2.4 Relevance of the Girassol Failures to Other Systems
It can be seen from the Girassol failures that under certain conditions, the interlink
friction driven fatigue can result in very rapid fatigue damage. This mechanism
appears not to govern for many mooring systems, though the basic mechanics will be
present for all chains.
holding a
horizontal link
Pivot point
Stopper plates,
of trumpet
Figure 15-17 - Typical FPSO Chain Stopper Arrangement
The requirement for rotation between the end links of the chain was not a result of the
curved chainhawser. As illustrated above, where the stopper is located above the
assembly pivot point, rotation of chainhawser is unlikely until the chain contacts the end
of the trumpet. Small angles of rotation will thus be taken out in the first free link,
either as slip or in bending.
Sensitivity to this failure mechanism appears to result from three factors in the Girassol
system.
i) The nature of and natural frequencies of the moored structure resulted in angular
rotations having a high frequency of occurrence at a significant magnitude.
ii) The design of the chainhawse imposed the restraint that the majority of the rotation
(principally due to buoy pitch) was taken out in a single link.
iii) The relatively high working tension for the chain size permitted the development of
significant interlinks friction forces and local yielding.
211
A4163-01
The relative significance of these is not yet totally clear at present. Due to the
individual stiffness of chain links, large bending stresses can be developed by relatively
small end rotations. These will tend to be relieved, either by slip or contact with the
trumpet, for larger motions. The proportion of the rotation imposed at a single link may
depend upon the chainhawse design, but this has not been studied. Friction coefficients
for interlink friction combined with the line tension will determine the stress developed
prior to slip, but little data is available in this area.
The simplified calculations below illustrate the accumulation of fatigue in a mooring
chain. The first calculation relates to Girassol (high pre-tension and an assumed annual
damage of 1.0 from the failure history). Example 2 relates to real example for a FPSO
in moderate water depth and a relatively low pre-tension. Example 3 relates to an FPSO
in deep water with high pretension. These calculations are intended to be indicative
only, it is appreciated that the wave climate, directionality and chain hawse geometry
are not represented.
In the following examples the notation listed below is applicable:
D chain bar diameter (mm)
T Tension force (tonnes)
A Area (m
2
)
M Bending Moment (kN.m)
Z bar section modulus (m
3
)
SCF Stress Concentration Factor
m Slope of the S-N curve
N Number of cycles to failure
N
1 year
Number of cycles in one year
o Stress due to bending or tension (kPa)
Friction coefficient
nom
Suffix nominal
b
Suffix bending
t
Suffix tension
212
A4163-01
3
Example 1 is a summary of the behaviour of the Girassol offloading buoy.
Example _1
, D 81 T mm 5 . 123 Te, 1 . 0
2tD TD
Z , M
nom nom
32 2
M TD 32 8T
nom
o 47MPa
bnom
3 2
Z 2 2tD tD
nom
log(N ) log A m 2 log( SCF o
bnom
)
Take log( : A) , 46 . 11 m , 173 . 3 ge AnnualDama , 1 N
1year
6 10
6
) 46 . 11 173 . 3 2 log( 47 SCF log(N
1 year ratio
)
46 . 11 778 . 6
( )) 94 log(
10
173 . 3
SCF 318 . 0
ratio
NB : SCF 7 . 3 SCF 18 . 1
ratio
However, the interlink frictrion will impose a bending force on both sides of the link.
Above example developed from BOMEL JIP ([Ref. 10] Section 3.11.1, page 29).
Log
10
N = 11.32 3.173 log
10
(Ao
nom
)
The report expression is:
Log(N) = logA mlog(2 x SCF x o
bnom
)
M = 3.173 is the SN curve gradient. Ao
nom
is the nominal stress range in a link during
fatigue testing. This has been set to be equal to 2 x SCF x o
bnom
. The 2 takes into
account that the link can move both clockwise and anticlockwise about an axis through
the line of the chain.
Within the catenary there are many chain links, increasing the likelihood of failure.
The design value of logA is typically 2 standard deviations below the test mean.
BOMEL took a value of 4.4 to reflect the number of chain links with a standard
deviation of 0.184. BOMEL results are from tests in air. Submerged chain will fatigue
more rapidly (logA increases by 0.301 based on standard fatigue texts for in air and
cathodically protected in water).
In this instance Log A has been taken to be 11.46 rather than 11.32 as shown below.
Considering a single link seems appropriate for the link being bent at the hawsepipe.
logA = 11.762 relates to a single link in air.
logA = 11.320 relates to a 4000 link chain in air.
logA = 11.461 relates to a single link in water.
logA = 11.019 relates to a 4000 link chain in water.
213
A4163-01
N
1 year
= 6 x 10
6
cycles is based on a wave frequency indicative period of 5.3 seconds,
i.e. (365 days x 24 hrs x 60mins x 60 sec)/5.3 seconds = 6 x 10
6
cycles. This is a crude
approximation.
The annual damage amounts to (6 x 10
6
cycles)/Inv. Log(11.46 3.173 x log(2 x 47 x
0.318)) = 1.0002, which means the fatigue life is consumed in less than a year. This is
not too different to what happened in practice on the Girassol buoy
Example _ 2
, D 120 T mm 60Te, 1 . 0
3
2tD TD
Z , M
nom nom
32 2
M TD 32 8T
nom
o 10MPa
bnom
Z 2 2tD
3
tD
2
nom
log(N ) log A m 2 log( SCF o
bnom
)
Take log( : A) , 46 . 11 m , 173 . 3 SCF , 18 . 1 N
1year
6 10
6
18 . 1
log(N ) 46 . 11 173 . 3 log( 2 ) 10
7 . 3
N 10
91 . 8
N
ge AnnualDama
1year
007 . 0
N
Example 2 results in a fatigue life of approx 143 years
Example _ 3
, D 114 T mm 143Te, 1 . 0
3
2tD TD
Z , M
nom nom
32 2
M TD 32 8T
nom
o 27MPa
bnom
Z 2 2tD
3
tD
2
nom
log(N ) log A m 2 log( SCF o
bnom
)
Take log( : A) , 46 . 11 m , 173 . 3 SCF , 18 . 1 N
1year
6 10
6
18 . 1
log(N ) 46 . 11 173 . 3 log( 2 ) 27
7 . 3
N 10
54 . 7
N
ge AnnualDama
1year
17 . 0
N
Example 3 results in a low fatigue life of approx. 6 years. This is potentially a
significant and worrying result which merits further investigation.
214
A4163-01
15.3 Tension Bending In Wire Rope
15.3.1 Significance of Tension Bending in Fatigue of Wire Ropes
Both DNV OS-E301 and API RP2SK [Ref. 31] make reference to bending tension (B-
T) fatigue of wire ropes at sheaths, pulleys and fairleads in addition to providing
guidance on the calculation of tension-tension (T-T) fatigue.
API offers some indication of fatigue life reduction factors taken from experience of
operations of Semi-submersibles in the North Sea. It is interesting that DNV does not
provide any specific guidance on the increased rate of fatigue under tension-bending.
The lack of clear guidance for calculation of this mechanism is partly due to its
complexity. The objective of this section is to increase understanding of this area,
rather than to provide guidance.
15.3.2 Testing at the National Engineering Laboratory (NEL)
A series of tests were carried out at the National Engineering Laboratory to examine
fatigue of wire ropes under bending tension conditions in 1988 [Ref. 58]. These tests
were carried out on a 40 mm diameter six strand rope with an independent wire rope
core. A test rig was developed for the purpose - see Figure 15-19. The set up permitted
the rope to be cycled both with respect to tension (load amplitude) and movement over
the sheave (bending length) under a given mean tension.
The conclusions were as follows:
i) The rate of fatigue damage is highly dependent upon the bending length up to a
travel of 25% of the lay length.
ii) The mean tension is the next most significant parameter in determining the rate
of fatigue damage.
iii) The load amplitude (i.e. the range in tension) on the wire rope has relatively
little influence on tension-bending fatigue.
The last item is perhaps surprising; indicating that load cycling on the rope is not the
dominant source of fatigue for tension bending.
The requirement to consider a loading source other than the tension range is reinforced
in OTH 91 341 [Ref. 59] where it is concluded that although the tension-tension fatigue
life is driven by the load range, the bending-tension fatigue life is governed by the
mean load and bending over sheave (BOS) behaviour.
215
A4163-01
15.3.3 Guidance on Damage Calculation
API give some specific guidance on the reduction in fatigue life associated with this,
tabulating 3 types of wire rope at 2 bend to wire diameter ratios. Similar information
can be found in table A6 of ISO 19901-7 [Ref. 4].
Wire Rope
Type
Gradient
m
D/d
Ratio
Factor on T-
T Life
Six strand 4.09 20 0.03
Six strand 4.09 70 0.08
Multi-strand 4.09 20 0.05
Multi-strand 4.09 70 0.15
Spiral strand 5.05 20 0.005
Spiral strand 5.05 70 0.015
Table 15-2 : Wire Rope Fatigue Reduction Due to Tension Bending [Ref. 31]
From the tabulated values it can be seen that the fatigue lives associated with tension-
bending (T-B) are very small in comparison those for tension-tension (T-T).
Unfortunately the above data makes no reference to either mean tension or bending
length, the two parameters identified as critical to bending tension fatigue. The
implication is that B-T fatigue can be estimated by directly factoring the calculated T-T
fatigue life, which lacks consideration or either parameter. This approach could only
be useful for a particular class of vessel under a specific climate.
The serious deterioration in fatigue life when bending is present has implications for
Flex Boots/Bend Limiters at wire rope sockets. Evidence has been seen in the field of
Flex Boots becoming detached over time. If the Flex boots become detached the wire
at the socket is more likely to fail due to tension combined with bending. Figure 15-18
shows how wires which have failed due to fatigue can be recognised. Failure of Flex
Boots may also result in additional torque induced fatigue at the spelter socket. This
will tend to be more a factor for non torque balance IWRC rope.
216
A4163-01
Figure 15-18 Illustration of Wire Rope Failure Modes (courtesy of Bridon)
217
A4163-01
OTH 91 341 [Ref. 59] introduces a useful distinction within B-T fatigue between an
upper bound T-T life and a lower bound Bending over Sheave (BOS) value. T-T
fatigue damage rates would be derived from the load range history of the mooring line
and would apply to the full length of the wire rope. BOS fatigue damage rates would
be developed from a combination of the mean line tension and mooring line to sheave
rotations.
It is possible that the two mechanisms can be considered independently. This would
depend upon the potential for bending length due to wire stretch under the tension load
range. It is understood, however from OTH 91 that the BOS damage rate would be
dominated by mooring line to vessel rotations, justifying independent treatment of BOS
and T-T fatigue.
15.3.4 Conclusions
A large body of research has been conducted toward understanding bending tension
fatigue of wire ropes. But no generally applicable quantitative guidance is offered by
DNV or API on this subject.
The fatigue process and loadings associated with bending tension appears to be well
understood. Substantial test work has been carried out permitting the definition of
load-cycle curves for various configurations.
Therefore, the development of specific design guidance for the estimation of bending
tension fatigue for offshore mooring systems appears to be feasible. It is recommended
that this be incorporated into calculations for integrity of wire rope mooring systems.
This is likely to become of increasing importance as the age of the wires in
combination wire/chain systems increase.
Figure 15-19 - The 1.0MN Wire Rope Bending-Tension Fatigue Test Machine
[Ref. 42]
218
A4163-01
15.4 General Implications of Tension Bending Fatigue for the FPS Industry
15.4.1 Implications for Design
Both API and DNV guidance includes recommendations on the design for tension
bending fatigue of chain at wheel fairleads. Both sets of recommendations represent an
extension to the calculations already performed for fatigue of chain within the catenary.
Depending upon the design approach used, tension bending in a wheel fairlead may be
calculated directly from the catenary line tension by applying a factor on the nominal
stress, or from the catenary fatigue life by applying a different factor.
Neither design code provide any guidance on the interlink friction effect associated
with the Girassol buoy. It is recommended that within design of the mooring system
this fatigue mechanism be considered in addition to wheel fairlead tension bending, as
applicable. Note that the two effects will coexist, their relative importance governed by
the relationship between line working tension and dynamic tension range.
Figure 15-20 - Tension Bending at Wheel Fairlead (Bearing Load Eccentricity) and
Tension Bending from Interlink Friction (Torque at Contact)
The calculation of fatigue damage due to interlink friction at requires consideration of
the chainhawse geometry and the friction coefficient. Having developed a relationship
between link bending stress and motions of the unit, taking account of limiting values
due to the chainhawse geometry and slip between links, a conventional fatigue analysis
can be carried out. Various S-N curves have been proposed for the damage calculation.
As the cracking appears to occur away from the flash weld, in a similar location to that
for tension fatigue, a modified chain fatigue curve is thought to be appropriate.
219
A4163-01
15.4.2 Selection of Design Fatigue Parameters for Chain
Fatigue curve parameters proposed by DNV and API are outlined in the Table below:
Source Gradient
m
Intercept
A
Notes
BOMEL JIP 3.173 11.46
For use with single link in catenary
Adjusted for in water use
BOMEL JIP 3.173 11.02
For use with 4000 link catenary.
Adjusted for in water use
DNV OSE301 3.00 11.079 For use with catenary
API RP2SK 3.36 11.653 For use with catenary
Table 15-3 - S-N Parameters for Mooring Chain Fatigue
Note that the three guidance notes also differ in terms of the recommended safety factor
on fatigue life. This factor is typically dependent on the inspection regime and may
also be affected by the criticality of the adjacent mooring lines.
The BOMEL JIP contrasts mooring chain failure, where loss of any one of many
similarly loaded components results in total loss of the system, to failure within a
system where a degree of redundancy is present (e.g. a jacket). In order to account for
the increased probability that one out of the large number of links in the mooring chain
will fail a slightly more onerous curve is developed. Where a specific link (such as a
link in the wheel fairlead) is considered, the single link SN curve parameters would be
appropriate.
Traces of the three catenary fatigue SN curves, BOMEL catenary (BMLn), DNV
studlink OSE301 (DNVs) and API RP2SK (APIc) are illustrated in
Figure 15-21.
220
A4163-01
Figure 15-21 - Comparison between Various Mooring Chain S-N Curves
Factors on stress or fatigue life proposed in the various guidance notes may be
represented by factoring stress or life values, or alternatively by adjusting the intercept
values to produce a case specific S-N curve.
15.4.3 In Service Inspection
In service inspection of chain sections at wheel fairleads and chainhawsers is made
extremely difficult by location and poor access.
It should be noted that a visual inspection may not be sufficient to identify fatigue
cracking in the chain links. Even under laboratory conditions [Ref. 10] quite large
cracks could not be readily identified without the use of dye penetrant or some
alternative NDE technique.
221
A4163-01
15.5 Recommendations
The constraint offered to a chain link in a wheel fairlead is such that severance of one
limb of a chain link can occur without release of the mooring line. On this basis it
could be argued that repositioning of fairlead chain links into the mooring catenary be
avoided unless detailed inspection of the affected links has been possible. However,
for normal field operations this is not a feasible process.
No guidance on the applicable friction coefficient for sliding between chain links is
available. As this value is critical to the interlink friction tension bending fatigue
problem, it is recommended that further work be done to identify suitable values for
this.
Interlink friction tension bending fatigue and where applicable wheel fairlead tension
bending fatigue should be addressed in the design of permanent mooring systems.
Bending at sheave calculations should be performed for wire rope mooring systems to
identify rates of fatigue damage associated with bending-tension at the fairlead.
222
A4163-01
16 FRACTURE MECHANICS AND CRITICAL
CRACK SIZE
Experience has shown that chains and connectors are susceptible to fatigue cracking.
Thus this section looks at the potential of Fracture Mechanics to help reduce the
likelihood of cracks leading to complete failure.
The basis of Fracture Mechanics is that it provides quantitative answers to structural
integrity questions, such as the following:
- What is the critical crack size at service loads?
- How safe is the system if it contains a crack?
- How long might it take for a crack to grow from initial to critical size?
- How often should a particular structure be non-destructively inspected?
Fracture Mechanics provides a quantitative relationship, between material, design and
fabrication, or more simply between stress, flaw size and toughness. Fracture
mechanics is not only a powerful tool for analytical evaluation of Non Destructive
Testing (NDT) flaw indications, but is also helpful for the initial design, materials
selection and any subsequent failure analysis.
In theory, a Fracture Mechanics analysis, coupled with appropriate inspection
procedures, can provide a rational and quantitative method for enabling a component to
be kept in service safely, at least until a scheduled inspection or maintenance outage.
At this time it may be possible to undertake a repair with minimal loss of production.
Overall, therefore, it can be appreciated that Fracture Mechanics is potentially a very
useful tool to assist the mooring design and integrity monitoring process.
16.1 Required Data
A fracture mechanics evaluation of a particular flaw requires accurate knowledge of the
following:
1. The size and shape of the flaw,
2. The loading conditions/stress levels in the region of the flaw (Finite Element
Analysis or Direct Measurement)
3. The operating environment, e.g. sea-water, splash zone, etc.
4. The fatigue/fracture mechanics properties of the material.
223
A4163-01
There are real practical difficulties in obtaining, particularly in situ, the above data for
mooring lines (see Section 18). If it is not possible to detect and assess crack sizes
accurately the remaining life of the component under consideration will be highly
uncertain. For mooring components there can be very large variations in applied loads
depending on the severity of any storms and the steepness of any waves which are
experienced.
Although it is difficult at present to detect cracks in situ, inspection technology will
continue to improve. Thus it is important to continue to develop our Fracture
Mechanics understanding with relation to mooring components. Although good work
has been done in this area see below, it is clear that more work is still required. For
example, it is important to identify the maximum defect size that can be permitted
during manufacturing, while still allowing a satisfactory mooring component field life.
16.2 Fracture Mechanics and Chains State of the Art Summary
A number of groups have been using Fracture Mechanics as a tool to understand and to
analyze fatigue induced crack growth of mooring chains in dry air and in hostile
environments. These include Vicinay Cadenas, Labein, Agder College of Engineering,
Grimstad, and the Department of Metallurgical Engineering and Materials Science of
the Bilbao Engineering Faculty (UPV-EHU).
To reveal fatigue crack behaviour in mooring chains, measurements were carried out
for high strength steel (Grade R4) in the Labein Laboratory, Spain, using Compact
Tension (CT) specimens with 12.7 mm thickness. These tests were performed in 1999
and 2000 in accordance with standard ASTM E 647 procedures. The specimens were
tested in air and in sea water with a frequency of 1 Hz. The fatigue crack growth was
monitored by an Alternating Current Potential Drop (ACPD) method and the crack
growth rates were plotted as a function of the Stress Intensity Factor Range (SIFR).
In 2003 a second series of tests using specimens of steel grade R4 were undertaken at
Agder University College, Norway. In this case Compact Tension (CT) specimens
with 25 mm thickness were tested in various environments under constant amplitude
loading. The fatigue crack growth was monitored by an Alternating Current Potential
Drop (ACPD) method and the crack growth rates were plotted as a function of the
Stress Intensity Factor Range (SIFR). The tests were carried out in dry air, in sea-water
without protection against corrosion and in sea-water with cathodic protection. The
cathodic potential was set to 890 mV and 1100 mV relative to an Ag/AgCl reference
cell.
224
A4163-01
The measured growth rates were compared with the rates for medium strength carbon
manganese steels as found in standard codes e.g. BS7910. The measured growth rates
were well within the scatter band given for these steels freely corroding in air. The
crack growth rates found in seawater with cathodic protection were, however,
substantially lower than the rates given in BS7910. When a cathodic potential of -1100
mV was applied, crack closure was observed at medium levels of AK. The explanation
is the formation of calcareous deposit in the wake of the crack front that gives
significantly reduced growth rates and finally leads to crack closure. This finding is a
surprise for high strength steel. The results are promising and should be investigated
further including the implications for the offshore operation of cathodic protection
systems.
A linear elastic fracture mechanics model was established to study the fatigue
behaviour in a studless link. The recorded growth parameters were used in conjunction
with a crack-like initial flaw with depth in the range from 0.12 to 0.25 mm. The
difference found between the growth rates in dry air and in free corrosion were in
accordance with tested fatigue lives for these two environments.
In general, after chains have been broken during fatigue testing microscopic
observations have been undertaken of the fatigue fracture surfaces to confirm the crack
growth process and any initial defect. In this way it has been possible to adjust the
validity of the fatigue model parameters.
16.3 Fracture Mechanics Critical Crack Size Implications
Knowing the critical crack size which could lead to rapid chain failure is important for
chain inspection. Additional research is need in this area to identify what are the
critical crack sizes. This will be an important input in helping to develop new
technology capable of detecting these sizes of cracks before they propagate through the
material thickness.
225
A4163-01
17 LINE STATUS MONITORING AND FAILURE
DETECTION
17.1 Instrumentation Status - Survey Results
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 the
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.
Anchor handling experience and calculations has shown that very 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 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.
- 50% of units cannot adjust line lengths.
If the level of instrumentation/alarms in the North Sea is patchy, it seems likely that
units operating in less heavily regulated regions will have even less instrumentation.
226
A4163-01
17.2 Existing Failure Detection Systems
17.2.1 Simple Sonar Probe
This system is illustrated in Figure 17-1 and Figure 17-2 and is employed on a North
Sea FSU. The Sonar head is deployed through the centre of the chain table to
approximately 15-20 metres below the hull. The head is deployed every 2 weeks in
calm weather or after a storm to confirm that all the mooring lines are present.
Figure 17-1 - Sonar Fish for Deployment through Turret (courtesy Chevron Texaco)
Figure 17-2 Sonar Fish Deployment Method (courtesy Chevron Texaco)
227
A4163-01
The illustrated system is fairly simple and is easy to repair if something does go wrong
with it. Figure 17-3 illustrates the image which can be seen on the sonar display screen,
namely 12 mooring lines and two risers close in to the centre. However, the system
does have two important limitations, namely:
1. If a line breaks in the mud (not unknown) it will still have some
tension/catenary and thus the change in the screen appearance may not be
sufficient to indicate that a line has failed.
2. A line could fail and not be detected for 2 weeks, during which time a
severe storm could develop.
Figure 17-3 - Sonar Display Screen Showing 12 Mooring Lines and 2 Risers Close to
the Centre (courtesty Chevron Texaco)
17.2.2 Use of Micro ROVs
The possibility of a line failing in the mud and not being detected is a realistic concern.
For example one FPSO has experienced a break in the mud line approximately 8 years
into its field life. If, however, FPSs were equipped on installation with the type of
simple inclinometer shown on Figure 17-4 it would be possible to determine, in calm
weather if any of the mooring line angles have changed to a significant extent. The
inclinometers could be checked using a Football sized ROVs which can be deployed
directly from the deck of the FPS itself. This removes the need for expensive ROV
intervention vessels see Section 0. These small ROVs can be stored on the FPS itself
or can be sent out by a helicopter as the need arises. Simple inclinometers overcome
the difficulties sometimes encountered with damage to power and signal distribution
cables on more complex systems. Being able to do a fly by in good weather to read
all the inclinometers would show whether the mooring line tensions are still in balance,
or if for slack or dog legs have been pulled out of the system see Section 6.2.1.
228
A4163-01
Figure 17-4 - Simple Pre-Installed Inclinometer with + or 1 Degree Accuracy
(Courtesy of Shell/SBM)
Figure 17-5 - Illustration of a Football Sized ROV (Courtesy of I. Williams)
229
A4163-01
17.2.3 Instrumented Mooring Lines
Intrinsically the simplest way to finds out if a mooring line has failed is to include a
load cell in the line, ideally close to the fairlead, where the tensions are normally
highest. Such a system is illustrated in Figure 17-6. Figure 17-7 shows the very useful
data which can be obtained from such a system as long as it is working properly.
However, particularly for submerged turrets, because the mooring lines are not readily
accessible, if the sensor in or the wiring/connections fail, you are in the difficult
situation of not knowing whether the line has failed or the sensor has failed. If you
record all line tensions and a line fails you should see tension pulses on the adjacent
lines. These should be detectable if the recording interval is frequent enough and the
load cells are sufficiently sensitive, although this does mean that you end up
accumulating a lot of data. On one instrumented North Sea FPSO a mooring line
failed, but it took two weeks of data processing from the other lines to reveal the
tension spike that confirmed it was a real failure rather than an instrumentation fault.
Power & data
transmission
cable
Figure 17-6 - Instrumented Load Pin Shackle Link (courtesy of BMT/SMS)
The power and signal transmission cables are areas of particular weakness for systems
exposed to long term offshore loading conditions. They may not even survive the
installation operation. To quote from one project,
The load monitoring systems are not commissioned yet. The load cell cables did
not last the tow from the conversion yard as they all came off their cable trays
running up the side shell and were damaged - site did not use the specified
number and size of cable ties. All cables still to be replaced.
230
A4163-01
This quote shows the low priority which is typically assigned to mooring line
instrucmentation.
Figure 17-7 - Indication of the Data Available from Instrumented Mooring Lines
(courtesy of BMT/SMS)
Theoretically you could go for a simpler system, for example using limit switches
placed on the trumpet assemblies. However, without moving the trumpet assemblies it
is not clear how these could be tested in situ to confirm whether or not they are still
functioning. Moving the trumpet assemblies with the chains in situ is not feasible
without the use of powerful anchor handling tugs.
External Turret or Spread Moored Moorings Lines Visible
For FPSOs with internal turrets or spread moored units, where the chains come up onto
the deck, it is relatively easy to confirm that the mooring lines are still present by
simple visual observation. Again, however, there can be a difficulty if a mooring line
fails in the mud. For lines which cannot be seen clearly from the FPSO deck regular
checks should be made, for example by supply or standby vessels that all the lines are
present. This should be written into the standard operating procedures.
231
A4163-01
17.2.4 FPS Offset Monitoring and Line Failure Detection
If a mooring line fails the resulting equilibrium position would change and theoretically
it should be possible to detect this based on offset monitoring. However, apart from in
deep water, if a mooring line fails in moderate weather conditions it is difficult to
distinguish the change in offset from the normal offset changes due to wind, wave and
current effects. Perhaps surprisingly mooring lines do fail quite often in moderate
conditions, sometimes following on from storm loading.
In addition, the direction from which the weather comes from may influence the
effectiveness of offset monitoring for line failure detection. For example if a line fails
and the weather pushes the unit in the direction of the failed line, the offset from the
equilibrium position will be small compared to the weather pushing the unit in the
opposite direction to the failed line.
Satellite drift and possible gyro malfunction can affect the accuracy of offset
monitoring. For example a system is installed on one North Sea unit and this has
indicated out of position alarms when there were no line failures. This can be due to a
poor Global Positioning System (GPS) fix depending on the number of satellites
available at a particular time. In another incident the FPSOs gyro became unstable and
this resulted in high apparent offsets. However, as long as false alarms do not happen
so often that they are automatically discounted, the odd false alarm helps to keep
people thinking about moorings. In addition, it was helpful that the gyro problem was
noticed early on before it could have had an impact during say an offloading operation.
Overall offset monitoring and recording is cheap and worth having since it is surprising
what mariners can deduce from experience and relatively little data. For example, if
you are used to the FPSO taking up a certain offset in moderate south westerly
conditions and this, plus the overall response of the vessel seems to change, would be a
good trigger to deploy a micro-ROV to check out the condition of the lines see
Section 17.2.2.
It should also be noted that offset monitoring is a potential input to other possible line
failure detection techniques, see for example Section 17.3.1.
17.3 Future Failure Detection Systems
New methodologies to detect a mooring line failure typically feature scanning acoustic
transponders deployed through the turret, attached to the hull of the FPSO (see Figure
17-8 and Figure 17-9), or installed on the seabed to provide an indication of the
catenarys profile. The advantage of these compared to the simple dipping sonar
described in Section 17.2.1 is that, in calm conditions, it is hoped that they should be
able to detect the change in catenary profile which is likely to be associated with a line
break in the mud. Two different systems based on this basic approach should be tested
in the near future in the North Sea.
232
A4163-01
Figure 17-8 - Illustration of a New Sonar System due to be Installed in the North Seas
(courtesy of I. Williams)
It will be interesting to see if this system proves to have sufficient resolution in practice
to pick up a line failure in the mud see Section 17.2.1.
Figure 17-9 - Close Up of the Proposed Sonar Head (courtesy of Ian Williams)
233
A4163-01
Vessel/ Mooring
mathematical
model
Predicted
Vessel
Position &
heading
Difference
between
predicted &
actual
Vessel response
to environment
Measured
Position &
Heading
Revise
mathematical
model
coefficients
Predicted
Tidal Current
& direction
Draft Sensor
Measured
Wind Direction
& Magnitude
Vessel/ Mooring
mathematical
model
Predicted
Vessel
Position &
heading
Difference
between
predicted &
actual
Vessel response
to environment
Measured
Position &
Heading
Revise
mathematical
model
coefficients
Predicted
Tidal Current
& direction
Draft Sensor
Measured
Wind Direction
& Magnitude
17.3.1 Response Learning System (RLS) for Automatic Line Failure Detection
Another line failure detection option may be a Response Learning System (RLS) 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. If a unit is equipped with an environmental detection and recording system
and a DGPS (Differential Global Positioning System) location system it should be
possible to utilise learning algorithms, similar to those used by Dynamic Positioning
(DP) systems, to evaluate, for a given applied environment, what the excursions should
be for an intact and one line failed condition. Figure 17-10 includes a flow chart which
illustrates this process. Hence, if the excursions do not match the predictions then an
automatic alarm should be sounded, alerting the crew that a line may have failed.
Overall this is a fairly complicated procedure and will require investment to develop
further. However, it has the real benefit that it would be a relatively simple retrofit to
existing installations, avoiding the need for expensive intervention work such as
installing load cells and wiring. Also if the system breaks down it should be possible to
fix it without any wet intervention.
Vessel/ Mooring
mathematical
model
Predicted
Vessel
Position &
heading
Difference
between
predicted &
actual
Vessel response
to environment
Measured
Position &
Heading
Revise
mathematical
model
coefficients
Predicted
Tidal Current
& direction
Draft Sensor
Measured
Wind Direction
& Magnitude
Figure 17-10 - Response Learning Without Line Tension Input
234
A4163-01
17.3.2 Utilization of Riser Monitoring Technology
Converting Motion Data into Line Tensions
The subsea motion sensors can monitor six degrees of freedom motion. However, what
is really wanted from a mooring perspective is real time mooring line tensions. This is
more complicated, but is believed to be achievable.
It is suggested that the following approach could be utilized to evaluate real time line
tensions. The advantage of this approach, relative to strain gauges, is that no major
intervention is required to install load cells as part of the mooring line.
1) Install, at a suitable in air locations on the FPS a motion reference unit (MRU)
and data logger.
2) Based on the data from this MRU the real time motion at each of the fairleads
can be determined and recorded.
3) Input the motion time trace into a high quality line dynamics mooring
analyses programme to evaluate line tensions.
Although the previous approach will give an estimate of the line tensions the accuracy
will depend on whether the drag and damping evaluated by the line dynamics
programme is reasonably correct. Hence a further refinement recommended to cross
check the line tensions results. This cross check would comprise the following:
1) Strap on a motion sensor to a mooring line at a known distance from the
fairlead, say 30m.
2) Collect motions data for this point on the mooring line while also recording at
the mooring line while also recording at the same time data on fairlead
motion.
3) Compare the predicted motion at 30m down the mooring line with the actual
behaviour.
4) If there is a significant difference modify the drag and damping parameters in
the line dynamics programme until convergence is achieved.
What is interesting about this approach is that it provides a means to identify the
maximum tension in a mooring throughout its length. This is because, depending on
dynamic behaviour, the maximum line tension may not be at the top of a mooring line.
Remote sensing technology has been utilized to monitor the behaviour of flexible risers
see Figure 17-11. These sensors tend to be battery powered and low power. The
signal from the sensor can be acoustically transmitted to a data logger on the platform
without the need for wires. On risers the sensors have been changed out by a ROV
when their in-built batteries become exhausted after several months.
235
A4163-01
Figure 17-11 - Illustration of Riser Monitoring Instrumentation (courtesy of 2H)
Real Time Monitoring of Line Tensions
Real time mooring monitoring provides an option to resolve the uncertainty which
always exists as to whether real behaviour is close to the initial predictions. This has
the following benefits:
- A check can be made on storm loadings.
- Fatigue life predictions can be updated.
- IMR strategies can be modified as required.
The required sensors for mooring monitoring are basically low power. Hence there may
be some option to power sensors from the fluctuations in mooring line tensions and
transmit the signal acoustically to a transponder mounted on the FPS hull.
236
A4163-01
17.3.3 UK HSE Position on Failure Detection
The present position of the UK Health and Safety Executive (HSE) on failure detection
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 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
see section 17.2.4. Satellite drift is also a potential factor affecting the reliability of
offset monitoring.
237
A4163-01
18 INSPECTION, REPAIR & MAINTENANCE (IRM)
18.1 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.
238
A4163-01
18.2 Where to Inspect on a Mooring Line
Figure 18-1 illustrates the areas on mooring line which are subject to the highest
degradation and should be most closely inspected. In particular, in field experience
suggests that the less loaded Leeside lines (see Figure 1-1 and Figure 3-4), which see
more relative rotation and motion, are subject to the greatest amount of wear.
The length of mooring line which seems maximum wear on the sea-bed is quite
localised. Hence it is important to ensure that the ROV measures the right links on the
sea-bed section. On Figure 7-9 the blue dotted vertical line shows the location of the
no applied load touch down point. It is interesting to see that the blue line is
approximately at the bottom of the black poly line which is a curve fit through all the
line 7 diameter measurements. In other words the maximum wear has occurred at this
point.
Orc aFlex at 08:30 on 05/04/02: Block 5_Multiple_Statics_Adj
50 m
X
Y
Z
m usted_Load_4_5_6_Uni_Sets_Steep_Sea_MPM_Exc ur_Laden.dat (az imuth=280; elevation=5) Static s Co
X
Y
Z
Figure 18-1 - Red Arrows and Black Line Indicate Key Areas subject to Degradation
on a Mooring System (leeward likely to have worst wear)
In Figure 18-2 where there is a red arrow there will be weight per metre discontinuity
as you change from wire to chain. Where there is a weight per metre discontinuity one
may experience increased relative rotation and thus wear. This seems to be particularly
pronounced on leeside lines.
239
A4163-01
Figure 18-2 - Example of a Weight Discontinuity which may Result In Enhanced Wear
18.2.1 The Problem of Inspection when you cannot Adjust Line Lengths
For the majority of FPSOs (i.e. non Tentech designs) which have the lines stoppered off
at the base of the turret, adjusting line lengths is only possible as part of a major
operation with a dive support vessel present (DSV). Line adjustment is not intended to
be undertaken during annual or bi-annual inspection operations.
Not being able to adjust line lengths presents a real issue with regard to access for line
inspection operations. As has been seen from the Section 9.2.2 case study it is very
difficult to see what is happening inside or even at the outer edge of the trumpet or
hawsepipe. If it is possible to drop the line tension, so that the links which have either
been held in a chain stopper or otherwise been working in the pocket of a gypsy wheel
are accessible, this makes inspection much more straight forward. To fully inspect the
touch down point and thrash zone it is desirable to lift the chain off the sea-bed. The
simplest and safest way to do this would be to pull in on a winch in the turret. If this is
not possible, due to the FPSO design, theoretically it might be possible to grapple for
the chain using a J hook from anchor handler. However, this might damage subsea
infrastructure and may not be permitted. What this means in practice is that the static
touch down point and thrash zone, which for the static equilibrium position extends
both up the catenary towards the FPSO and along the sea bed section towards the
anchor, is difficult to inspect. In other words on many FPSs a section of mooring line
subject to some of the worst degradation is difficult to inspect properly.
In certain cases it may be possible to operate thrusters to pick up some of the sea bed
chain or even to utilise a tug to pull a FPSO away from its normal equilibrium position.
This will need to be risk assessed on a case by case basis. If a tug is used to move an
FPSO it is vital that all components are strong enough to take the applied loads
including any closed chock fairleads on the FPSO. There has been one incident of a
closed chock fairlead failing while load was being applied by a tug.
240
A4163-01
18.2.2 Inspection Access to the Stoppers
In general, all the links in the trumpet area are difficult to inspect. Hence, if a link is
going to be exposed to 20 years of dynamic motion, does it make sense to place it
somewhere where it cannot be inspected, even if you have a high fatigue safety factor.
Figure 9-12 and Figure 9-13 show the level of wear noted on a mooring line which was
recovered back to shore after six years of deployment in the North Sea. At the turret
interface there are bending and twisting stress raisers, plus non perfect link geometry,
which make the situation worse compared to a pure tension-tension situation.
There are different designs of internal turrets and some may appear to give more ready
access to the chain stoppers than others. However, even for the type of turret illustrated
in Figure 18-3 and Figure 18-4, the room at the base of the structure is flooded. Thus
for example the picture in Figure 18-4 was taken by a camera mounted on a pole.
Figure 18-3 - Typical Turret Cross Section Illustrating that the key Mooring
Components are Submerged
241
A4163-01
Figure 18-4 - Chain Stopper View Prior to Chain Installation with Pull in Rigging
Present (compare to Figure 18-3)
It is important, for future designs how to improve accessibility for inspection. This has
implications for mooring design brief or specification see Section 20.
18.3 In-Water Inspection
To date chain mooring components have been the subject of the greatest effort to
develop in-water inspection methods. 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 3-4. 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.
242
A4163-01
18.3.1 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 callipers to a prototype
stand-alone robotic system and ROV deployed systems.
Diver inspections are, in general, not a favoured option. Mooring chains are highly
dynamic and therefore are potentially dangerous when divers are in close proximity.
Also diver inspection has been 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 to date this seems to have been
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.
Possible ROV-deployed systems include both mechanical calliper and optical calliper
systems. Mechanical callipers 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 calliper developed by Welaptega Marine Ltd. It comprises 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 18-5). 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 jewellery such as connectors, shackles and kenter links.
The optical calliper 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.
243
A4163-01
block
guide
Camera
Underwater
light
Deployment
Figure 18-5 - Illustration of ROV Deployed Optical Calliper Measurement System
(courtesy of Welaptega Marine Ltd)
18.3.2 Loose Chain Stud Detection
As discussed in section 3.2.3 in studded chain, loose studs have been implicated in
crack propagation and fatigue. Accordingly studded chain inspection and re-
certification 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 Welaptega Marine Ltd has developed an ROV-deployable loose stud
detection system. 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.
244
A4163-01
18.3.3 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).
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 artefact 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 systems [Ref. 60]. 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
footage.
As part of any in-water inspection it would be prudent to identify and inspect all
kenters and D shackles to confirm that they appear to be intact and that all split pins are
present.
245
A4163-01
18.4 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 season (see Figure 18-6). This growth needs to be
removed so that the underlying mooring components can be inspected.
Figure 18-6 Illustration of Heavy 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 (GUI), 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.
246
A4163-01
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.
18.5 Manufacturing Tolerances and the Inspection Implications
From an inspection point of view it is extremely important to have a good idea of what
the dimensions were of a mooring component when manufactured. Thus, the
significance of any changes in component dimensions over time can be assessed
correctly. With forged components, such as chains and shackles, there will tend to be
an inevitable variation in dimensions. Section 3.2.7 provides indicative chain
manufacturing tolerances. Therefore, for key areas, such as at the turret interface, it is
important to obtain key bench mark measurement data during the original installation
process. At the present time this does not normally happen.
18.6 Wildcat/Gypsywheel Inspection
In general, whenever in water mooring line inspection is undertaken, a check should be
made of the condition of the wildcat pockets. The chain must be pulled in or let out to
expose the wildcat pockets which are hidden at a given chain position / fairlead
orientation. It has been found on semi-submersibles that if the pockets are damaged
badly worn this typically leads to accelerated chain wear and damage see Figure 18-7,
Figure 18-8 and Figure 18-9.
At the gypsy wheel the key links are those that make regular contact with the pockets
on the gypsy wheel. These should be clear of marine growth and thus fairly easy to
identify once the line is slackened off. If this is not the case it would be good to mark
one of the links before the line is slackened off. If this is difficult to achieve the line
should be slackened off a precise number of links so that one knows which links were
on the gypsy wheel. Taking some still photographs before the chain is slackened off is
a wise precaution. It is desirable to take sufficient measurements at the top of the
catenary such that one can compare the wear on links on and close to the gypsy wheel
with those further down the catenary.
247
A4163-01
Figure 18-7 - In-Situ Inspection of a Wildcat Pocket by Abseillers
(Courtesy of CNR)
Figure 18-8 - Close Up Of Fairlead Pocket Note Slight Lip on the Right
(Courtesy of CNR)
248
A4163-01
Figure 18-9 - Example of Chain Wear From Sitting in a Wildcat Pocket
(Courtesy of CNR)
Figure 18-10 - Red Zones Highlight the Importance of Checking all Relevant Structural
Connections (Courtesy of CNR)
The structural connections between the wildcat fairlead assembly and the hull structure
(see Figure 18-10) should also be regularly checked. Problems have been known to
develop in this area [Ref. 61].
249
A4163-01
Figure 18-11 shows damage to a submerged wildcat or fairlead lubrication line as noted
during an abseiller based inspection operation. Although not all wildcats require
lubrication, if a system is designed to have lubrication it can be seen that it is fairly easy
for, even steel, lines which go through the splash zone to become damaged. This
illustrates the difficulty experienced running power and signal transmission lines
through the splash zone see Section 17.2.3. If the wildcats are designed for
lubrication and are without it for an extended period the chances of seizure are
inevitably increased see Section 8.
Figure 18-11 - Example of a Parted Lubrication Line Feeding a Submerged Wildcat or
Gypsy Wheel (Courtesy of CNR)
18.6.1 Flatness of Chain Links and Torsion Implications
If a link which sits in a wildcat pocket or chain stopper is not flat (see for example
Figure 15-6, Figure 15-7 and Figure 18-12) it will be subject to regular bending
stresses. Over time this will have an impact on the fatigue life of the supported link
(see section 15.1.3).
Unfortunately, which a FPS goes on station, it is impossible to know in advance which
link will be sitting in a chain stopper or wildcat. Therefore, it is desirable to check the
flatness of all the links at the end of the chain which may be held/constrained. Figure
15-8 illustrates a simple gauge which can be used to check whether links are flat.
250
A4163-01
Figure 18-12 - Example of a Non Flat Link
18.6.2 In Situ Inspection of Wire Rope
Wire rope is particularly difficult to inspect and, at the present time, tends to be
somewhat subjective, see Figure 10-1 and Figure 11-6. Sheathed spiral strand wire is
even more difficult to inspect, since it is obscured by the sheathing. However it is
difficult to assess if the sheathing becomes damaged during installation, for example
going over the stern roller. If the sheathing is damaged letting in sea-water, this could
result, over time, in accelerated undetected corrosion. Also abrasion could occur on the
seabed and this would be difficult to determine.
251
A4163-01
A previous JIP (Subsea Electro-magnetic Appraisal of Wire Mooring Lines (SEAL)
[Ref. 62] was intended to develop a ROV deployed in water wire rope inspection tool.
However, this JIP only went as far as Phase I Design. The proposed subsequent two
phases, which did not attract sufficient funding, were:
- Phase 2 construction of a prototype and completion of onshore trials.
- Phase 3 offshore proving trials.
An in air based wire rope inspection unit which was tested on ther Buchan FPS wire
ropes is illustrated in Figure 18-13. Figure 18-14 shows a sketch of the proposed SEAL
tool deployed on an inclined mooring line from an ROV. Given the difficulties
involved in inspecting wire rope and the age of some of the wires presently in use, it is
recommended that consideration should be given to moving forward with Phases 2 and
3 of SEAL.
Figure 18-13 - Buchan FPS Wire Rope NDT Inspection Head
252
A4163-01
Figure 18-14 - Proposed Wire Rope Inspection Toll Delpoyed from a ROV
253
A4163-01
18.6.3 Acoustic Emissions
Acoustic emissions are short bursts of elastic energy released as stress waves resulting
from irreversible deformations in the material under test. Very small changes in
conditions at any point in a material normally produce a large number of emissions. In
theory these emissions can thus be used to detect, locate and characterise defects. A
prototype acoustic emission system was developed at Cardiff University but never got
to the stage of being sold on a commercial basis. A more update review of acoustic
emissions can be found in HSE Research Report 328 (2005) [Ref. 63]. The system
relied heavily upon instrument software and a very powerful post-processing and
analysis package [Ref. 64].
18.6.4 Cathodic Potential Checks (Impressed Current & Sacrificial Anode Systems)
For mooring systems which are designed to have corrosion protection via impressed
current or sacrificial anode systems it is important to check whether the system is
operational. For example, on one FPSO the earthing cables to provide electrical
continuity between the FPSO hull and mooring chanins were never installed.
The status of the system can be checked by undertaking a Cathodic Potential (CP)
survey. NACE Standard [Ref. 71] states that using a silver/silver chloride reference
electrode readings between -800mV to -1,100mV suggests adequate protection.
Further information can also be found in European Standard EN 13173 Cathodic
Protection for Steel Offshore Floating Structures [Ref. 72]. Reference should also be
made to section 16.2 where the interesting observation is made that at a cathodic
potential of -1,000mV crack closure was observed due to the formation of calcareous
deposits in the wake of the crack front.
Lloyds Rules and Regulations for the Classification of a Floating Offshore Installation
at a Fixed Location (May 1999, part 8, ch 2.1.3, section 1.2.4) states a more negative
value may be used for those locations where sulphate reducing bacteria may be active.
Where higher cathodic protections are applied it is necessary to watch out for hydrogen
induced embrittlement see sections 7.3 and 7.4.
18.6.5 Checking Mooring Line Pre-Tensions
The importance of a balanced mooring sysem in terms of pre-tension values is
discussed in detail in section 8.1.
With a ROV in the field it is possible to check the absolute accuracy of the mooring
line tensions. This can be done in two ways:
1. Measure the x, y and z co-ordinates of the mooring line touch down point. This
will be difficult to do precisely if the FPV is moving around much.
2. Use a ROV to temporarily mount an inclinometer on the chain close to the
fairlead.
254
A4163-01
18.7 Inspection Frequency Code Requirements
Annual Surveys would typically be carried out by Classification Societies on mobile
offshore drilling units (MODUs) comprising visual inspection of the accessible chain
links on or adjacent to the windlass. Intermediate Surveys are then normally performed
at the second or third surveys following on from a 5 year special survey. The
intermediate surveys would normally be undertaken at a rig move and would include
100% visual inspection of all the chain, excluding that which remains in the chain
locker during normal operations.
Since FPSs are not subject to rig moves, intermediate survey chain inspections are
difficult/costly. Hence FPSs which are classed tend to concentrate on the 5 yearly
special periodic surveys. However, the fact that annual and intermediate surveys are
felt necessary for drilling semis shows that there is a need for regular inspection.
Lloyds Rules and Regulations for the Classification of a Floating Offshore Installation
at a Fixed Location, (May 1999) state the following:
2.2.10 For positional mooring systems a rota of component parts of the mooring
system is to be examined at each Annual Survey. A periodic inspection
programme is to be developed by the Owners/operators and submitted to LRs
Headquarters for approval. Annual Surveys should be capable of determining as
far as practicable the general condition of the mooring system including cables,
chains, fittings, fairleads, connections and equipment. The Surveyor is to be
satisfied that all components and equipment remain in an acceptable condition.
Particular attention is to be paid to the following:
- Cable or chain in contact with fairleads, etc.
- Cable or chain in way of winches and chain stoppers
- Cable or chain in way of the splash zone.
It is interesting that no mention is typically made of the initial requirements for
calibration of tension meters, nor how often they should be recalibrated once the unit
has been installed. As is discussed in Section 8 out of balance line pre-tensions could
be a key factor leading to mooring line failures.
255
A4163-01
18.7.1 Difficult To Inspect Areas
Inspecting fibre rope for potential wear in situ is potentially difficult see Figure
18-15. It is noted that certain projects have elected not to use thimbles and it will be
interesting to see if this leads to increased abrasion over time.
A c c e s s
d i f f i c u l t i e s
f o r i n w a t e r
i n s p e c t i o n
f o r j a c k e t
a b r a s i o n
Figure 18-15 - Example of a Difficult Area to Inspect
Chain in the trumpets and at the stopper is obviously difficult to inspect as previously
discussed see Section 9.
In theory it is important to confirm that all pin locking devices are in place and secure.
However, as can be seen from Figure 18-15, during a normal ROV survey, this can be
difficult to achieve.
Figure 18-16 - Partially Buried Shackle Illustrates the Difficulties in checking locking
pins (courtesy of ENI)
256
A4163-01
18.7.2 Pile, Padeye and Anchor Inspection
The padeyes on the piles are typically buried several metres down. From a simplistic
point of view jetting out soil from around the pile would loosen it and thus does not
seem desirable !
However, if one pile or padeye degenerates and fails the rest will probably be in a
similar condition. Thus the danger of the system unzipping with multiple line failures
cannot be discounted.
It is important that the fatigue capacity of the padeyes and piles should be checked and
be satisfactory based on a generous safety factor, probably in excess of 10. This is
because it is pretty well impossible to inspect these components in situ.
18.8 Outline Method To Break Test Worn Mooring Components
As mooring lines and connectors wear, corrode and fatigue it is likely that there will
become a stage when the true Minimum Break Load (MBL) of the line is no longer
known with any real confidence. With the desire to sometimes extend field lives
beyond the original design life there is a need to confirm that the as installed system is
still fit for purpose.
However, if you no longer know what the break load is likely to be this makes testing
somewhat more problematic. The following outlines a method that has been used
successfully at a chain test bed facility. It is worth noting that break testing can be
cheaper than Finite Element (FE) analysis and the results are likely to be more certain.
1) Undertake say 100te load steps initially, then say 50te steps nearer the expected
yield point - check at each step the load extension graph is a straight line;
2) When the first reading shows the line is just starting to tail off, call that the
yield/proof load - come back down and there should be a small amount of
plastic deformation;
3) The next check is to repeat the same line again, to the same load amount, and
that should be slightly displaced from the first run;
4) Repeat again to check that there is an accurate repeat of the elastic line;
5) Continue in whatever appropriate steps until one is chasing the load, i.e. it starts
to fall away.
It is worth noting that it is easier to set the load on some test machines as the tension
increases, rather than as it decreases; so it is better to do the steps only in one direction.
It is understood that chain elongating - hence knowing it is not holding (chasing the
load) - is normal, rather than a clean break. The chain extension can be automatically
logged into a data file monitored using infrared deflection monitoring equipment see
the chain infrared target shown on Figure 9-9.
257
A4163-01
18.8.1 Future Inspection Possibilities
As FPSs come off station at the end of their field lives this provides a good
opportunity to test worn mooring components.
There is also possible cross fertilization with flexible riser and particularly steel
catenary riser experience including touchdown zones and inspection techniques. For
example:
1. Sea bed troughs at riser touch down zones
2. Dynamic behaviour including snatch loading and compression which may be
detectable by installed instrumentation
Listening for the Sound of Cracks
Ref. 20 states Sounding the chain with a heavy hammer will reveal cracked or
internally corroded links or fittings. A sound link returns a clear, ringing tone; a bad
link has a dull flat tone.
This is a bit like a Wheel Tapper detecting cracks on railway locomotive/wagon wheels
see Figure 18-17. Sounding tests represent an interesting approach to inspection
which might avoid the need to remove marine growth see section 18.4. Hence, it is
suggested that further investigation of this topic should be undertaken.
Figure 18-17 - Example of the Wheel Tappers Approach Used for Detecting Cracks on
Railway Carriages and Locomotives
258
A4163-01
18.8.2 Repair Case Study Replacement of FPSO Trumpet Chain
For the chain damage reported in section 9.2.2 a long term repair was required. This
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 14-10) 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 (birdcaging). 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.
There were two potential options for changing out the links going into the turret,
namely:
1) Crop some links from the top section of chain, add a connector and re-install
2) Disconnect the chain at the sea bed
During the repair operation there was a strong desire not to disturb the spiral strand
wire since this can be relatively easily damaged see Figure 6-5. Hence option 2 was
selected.
Figure 18-18 and Figure 18-19 give an idea of the complexity and hence the cost of
such a repair operation including anchor handling plus heading control tugs, Dive
Support Vessel (DSV), divers and winch operations on the FPSO.
Figure 18-18 - Example of Anchor Handling and Heading Control Tugs during a
Mooring Line Repair Operation (courtesy of I. Williams)
259
A4163-01
Figure 18-19 - Use of Divers from a RIB to open up the Chain Stopper during a FPSO
Mooring Line Repair (coutesy of I. Williams)
260
A4163-01
19 SPARING OPTIONS
19.1 Contingency Planning - Spares and Procedures
Based on the indicative failure statistics reported in Section 13 it is quite conceivable
that a FPS may lose a line during its operational life. The failures are most likely to be
in the following areas:
1. At connectors or terminations
2. At wildcats/gypsywheels
3. Somewhere around the trumpet general area
4. In or just above the thrash zone area at the no load equilibrium position
5. At mid line buoys
6. At clump weights or hung off chain connections
Certain FPS Operators have spares, but they tend to be in the minority. In addition,
although spare sections of line, particularly for example polyester, may be available
suitable connectors are not always sitting in a warehouse or on board the FPS. In
certain cases although spares may have been ordered their present whereabouts or
condition is uncertain. On one project the spares disappeared over the horizon on an
anchor handling tug never to be seen again. On another project the spare chain was
stored out in the open and rusted to pieces.
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
14-10. 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. It is worth nothing that
one Operator has seen more degradation of spare polyester rope sitting onshore
compared to the same rope deployed under load offshore.
261
A4163-01
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. This may require the temporary use of second hand components such as
chain. However, the impact of introducing non standard elements (see Figure 19-1)
into a mooring system is best considered before a failure occurs. Long term mooring
(LTM) shackles should ideally be used as the connectors. 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 19-1 - Example of a Plate Shackle which may be useful for a Temporary Repair
(courtesy of Balmoral Marine)
In a post project Lessons learned exercise on one FPSO the following was reported:
262
A4163-01
Spares (all spares. commissioning and 2 year) should be purchased with
the initial order so they are available during construction, pre-
commissioning and commissioning. Sort out the mechanism for budget
allocation early so it does not impact spares purchasing. Consider spare
instrumentation/transmitters; since these are long lead and critical for
commissioning.
19.1.1 Operators Spares Club
Not surprisingly no two FPS mooring systems are identical, since they are in different
water depths and exposed to different environmental conditions. However, certain
components can be common between different units, for example, use of 120mm
studless chain. Other items such as LTM connectors (special shackles and H shackles
see Section 14.1) are likely to be needed for any repair work. Hence it would be logical
for Operators to form a Spares Club which could order key spares which would then be
available on a first come, first served basis. The established Marine Equipment
rental companies would probably be suitable organisations to store, look after and
promptly dispatch the spares when required.
19.1.2 Designing FPSs for Mooring Line Repairs
For a long field life the need to undertake mooring line repair is fairly high, even for
relatively benign climates, which will still suffer from wear and corrosion. Thus, it is
important that FPS facilities should be designed from the outset, such that mooring line
change out is relatively straight forward. However, this is not always the case which
may prove problematic in the long run. For example some Gulf of Mexico Spars have
utilised a temporary mooring line pull in winch deck, which is removed prior to setting
the main process equipment deck see Figure 19-2.
Figure 19-2 - Temporary Mooring Line Winch Deck on a Gulf of Mexico Spar
263
A4163-01
20 THE IMPORTANCE OF A COMPREHENSIVE
MOORING DESIGN SPECIFICATION
One of the best ways to solve problems is to prevent them starting in the first place.
Hence, if the mooring design brief at the beginning of a project is well thought out, it
can help to avoid difficulties which may develop during the course of the field life.
This section outlines what should be included in the original mooring design brief from
a mooring integrity point of view.
It is important that all interested parties should approve and support the mooring design
brief or specification. To ensure that a system proves to be reliable in operation the
design specification should consider and make reference to operations and long term
integrity.
To assist with long-term reliability it is necessary to be able to undertake inspection to a
level which gives real confidence in the condition of the as-installed system. This
ensures that intervention can be carried out early on, before detected anomalies get
worse and place the system at risk. Mooring line inspection should ideally be
undertaken in the water, since recovering lines is extremely expensive and may cause
damage such as wires birdcaging. However, mooring design briefs typically pay little
attention to the importance of inspection. Thus it can frequently be the case that key
components, such as chain stoppers, can be virtually inaccessible hidden away by long
hawse pipes or trumpets.
If the mooring design specification insists that key components of the mooring system
should be readily accessible for inspection, this will force designers to pay more
attention to this long term integrity/reliability issue. In addition, the specification
should state that the FPS design should allow for straight forward replacement in the
field of mooring lines, preferably using anchor handlers rather than specialist and
expensive construction vessels.
Mooring line instrumentation is another area which typically receives little attention at
the design brief stage. However, good quality instrumentation can potentially improve
mooring integrity to a significant extent. At the same time instrumentation can help to
detect problems early on which, without early intervention, can prove extremely
expensive to repair at a later date. This is particularly the case if the intervention work
results in deferred production.
The cost of instrumentation is relatively low if it is incorporated in the design from the
outset. However, it is vital that instrumentation needs to be reliable. Offshore
represents an exacting environment and thus best quality should be specified at the
beginning. The following parameters should typically be specified:
264
A4163-01
- Line tensions monitored and the data permanently recorded at a suitable sampling
interval
- 24 hours monitored over tension alarms
- FPS offsets and bearings monitored and permanently recorded
- 24 hours monitored lines intact alarms
- 24 hour monitored FPS excursion alarms
For moderate environments, such as off West Africa, there is a much smaller difference
between operational and survival sea states compared to say the North Sea. This means
that if the operational sea state, or the response of the vessel in the operational sea state,
is underestimated there is significantly less of an in built safety margin compared to
harsher climates, particularly with regard to fatigue. Therefore, depending on the
criticality of the fatigue assessment, it may be appropriate to undertake sensitivity
studies to assess the effect of an under prediction of actual vessel motions.
20.1 Installation Parameters
It is important that the mooring design process should take due consideration of the
capabilities of the likely installation vessels and their past performance on previous
projects. It is appreciated that during the early design phase that the particular
installation vessels may well have not been identified. Hence, a degree of conservatism
should be incorporated in the design process, such that the required installation
tolerances do not prohibit otherwise capable and perhaps cheaper vessels. This means
that the mooring design brief should include suitable loadcases to account for lines at
non uniform pre-tensions and anchors which may be tens of metres away from their
planned positions. This is particularly likely for the case of drag embedment anchors,
since it is extremely difficult if not impossible to predict where they will hold and
whether they will follow a straight line as they are dragged during pre-tensioning.
20.1.1 Mooring Design Team Participation during Installation
During the mooring installation process it is important that the installation crew should
be fully aware of the key design criteria, such as handling of fibre ropes, pretension
accuracy, anchor placement accuracy, avoidance of chain twists, etc. Therefore, it is
recommended that a suitably experienced member of the mooring design team should
go offshore during the mooring installation and FPS hook up operation. This person is
thus ideally placed to answer operational questions as they arise. Again this should be
specified in the mooring design specification so that contractors expect this and work
with the mooring designers during the development of the installation procedures.
265
A4163-01
In practice it is unlikely that a mooring system will be installed exactly to specification.
If the as installed condition does not correspond to any of the loadcases analysed
during the design process, it is important that key load cases should be re-run to ensure
that the system is still fit for purpose. Having sent a mooring designer offshore during
the installation helps to ensure that whatever is analysed back in the office corresponds
to the as installed configuration. In addition, feedback from the field helps to ensure
that the design of systems is continually improving and that sub optimum solutions are
not repeated. Again specifying a post installation check of the system performance
using as built parameters should be specified in the mooring design specification
20.1.2 Accurate As-Builts and Baseline Surveys
On the majority of FPSs the initial survey after mooring installation appears to have
been only Close Visual Inspection (CVI) and General Visual Inspection (GVI), no
measurements are typically undertaken. But there is a need for accurate as built
baseline dimensions so that the extent of any future wear can be assessed see section
18.5.
20.1.3 Mooring Design and Maintenance Based on a Life Cycle Approach
At present mooring systems are typically designed by specialists who may have little
further involvement after installation. It is only if serious problems occur that the
designers may learn more about how the moorings have performed in situ. This is
particularly the case if a FPS is provided by a contractor who then hands over operation
to an Oil Company.
However, mooring systems are not as simple as they first appear and they need careful
management through out their design lives. Thus a life cycle approach to mooring
design and operation is recommended. In this way designers can feedback their
inspection requirements to Operators and then learn from whatever is found during
inspection. Hence, over time, mooring design should improve. At present designers
are not always involved with the in field behaviour mooring systems. Hence they may
not be aware of operational or inspection type issues. Thus new projects may repeat
designs from the past, which in some instances have certain limitations. Obviously if
something has been demonstrated to work well over a long deployment this is a good
argument for not changing it.
266
A4163-01
21 KEY CONCLUSIONS & FUTURE WORK
RECOMMENDATIONS
21.1 Overview
In this project, an extensive investigation has been carried out of materials, design,
operations and management issues affecting the long-term integrity of mooring systems
for floating production systems. A broad survey has been conducted of units around
the world, especially those in harsh environments. The in-depth experience of the
participating equipment suppliers, designers, regulatory authorities, Operators and
Noble Denton has been collected and compiled into this state of the art report. The
resulting document is intended as a reference point for designers and operators alike,
with guidance on current and future practices and lessons learnt from the past.
There are many success stories in operation around the world, but there are also a
number of cases where the integrity of the moorings has been compromised to some
extent by unforeseen reasons. In part this is to be expected in any innovative
technology, but there also appear to be some critical omissions in design and integrity
management strategies. Significantly, during the course of this project failures have
continued to occur. Clearly there is still much to learn on this subject and key areas
requiring further work are identified later on in this section.
Overall, based on the evidence acquired during the course of this JIP, as systems age, it
seems quite probable there will be mooring failures in the future, unless more proactive
inspection and remedial work is undertaken. Areas to watch include:
1. Excessive wear and corrosion of chain in thrash zone,
2. Tension bending in deep water taut moored systems,
3. Problems due to the chain stopper being outboard of the pivot point
resulting in dynamic link wear + possible wear on the trumpet structure,
4. Weighted chain problems.
Given the difficulties associated with repair operations in deep water and the lack of
suitable spares, such failures would be expensive to repair and might attract publicity
which could be detrimental to floating production systems in general. Overall,
Operators need to get into the way of thinking that moorings are an integral part of their
production facility. This will encourage them to give them the attention that they merit
given the serious consequences associated with failure.
267
A4163-01
21.2 Key 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. One of the aims of this report is to
educate both offshore and onshore operational staff.
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 undertaken has shown the
importance of achieving compatible surface hardness, since it affects wear.
Unfortunately, at present chain hardness and wear do not normally seem to be
considered in any detail during the standard design process.
In situ in-water inspection techniques are continuing to improve, but further
developments are needed to provide dimensional data on links all around 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 which could include acoustic means see section 0.
Inspection access needs to be improved and this should be stipulated in the mooring
design brief or specification.
On two North Sea FPSs chain wear and corrosion 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 in field 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.
268
A4163-01
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, Magnetic Particle Inspection
(MPI), photographs and comprehensive dimensional 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 detecting quickly 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 have readily available on board the
maximum sea state in which they can continue to produce in case one line fails. This
assessment should be based on intact system mooring line safety factors. 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. During design the susceptibility of risers to be swept under moorings should be
assessed, since if a line fell on a pressurised riser the consequences are likely to be
serious.
Some large floating production projects have design lives of 20+ years. If a field is still
profitable there will always be a desire to continue production in excess of the design
life, but at this stage the moorings may no longer be fit for purpose. Hence, for long
field life projects a FPS Operator should review the budget for line repairs /
replacements part way through the field life based on up to date inspection findings
taking into account the experienced, rather than the anticipated, wear / corrosion.
A relatively simple wear model is reported in Shoup and Muellers OTC paper of 1984.
Given that there is now a limited amount of in field chain wear data from a few long
term deployed units, it would be desirable to undertake an up to date wear assessment
to see how the calculated values tie up. Once a validated methodology has been
developed it would be possible to use such an approach to estimate wear rates for 20
year plus required field lives.
A possible contributory mechanism for the relatively high line 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 in field Pay-In/Pay-Out tests should be undertaken to check whether
the line tension readings can be relied upon see section 0.
269
A4163-01
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.
In general there seems to be a need for periodic Mooring Audits to re-assess original
design parameters and review inspection records to assess whether the system is still fit
for purpose. When considering possible mooring line remedial works and when it
should be done it is logical to look at the anticipated future life of the chains based on
wear/corrosion rates experienced to date.
21.3 Recommendations for Further Study
Overall the JIP has helped to publicise the importance of mooring integrity to a larger
audience. However, it has also identified a number of areas which warrant further
investigation to improve safety and reduce life cycle cost. The following list identifies
the key topics:
1. Obtaining field data for different regions/FPS types on the combined
wear/corrosion rate particularly in the thrash zone/fairlead areas and the
implications for units which cannot adjust line lengths.
2. Calibrate an up to date wear/corrosion analysis model with long-term offshore
data (see section 7.6.2).
3. Engineering guidance for relative surface hardness for components expected
to be subject to long-term wear.
4. Assess how increases in proof stress may help the fatigue endurance of
mooring components see Section 14.5.
5. Development of improved in water inspection techniques for hard to access
areas with the goal of being able to detect cracks.
6. The potential for cost effective micro/mini ROV mooring line inspection from
the FPS itself.
7. Determining how chain strength is reduced by wear/corrosion is infancy and
more research and physical break testing of used lines and connectors is
required. This work should also consider the applied ramp rate during break
testing
8. Based on in field data, assess how removal of marine growth for inspection
affects corrosion rates.
9. Possible methods to check the integrity of connectors in the water, including
new designs of fibre rope connectors.
10. Collation and assessment of Pay-In/Pay-Out test data see Section 0.
270
A4163-01
11. Publicise the benefits of offshore visits/audits to brief personnel on mooring
integrity and to review existing instrumentation and procedures.
12. The dynamics leading to relative axial movement and wear between chain
links near the chainhawse needs to be understood so that it can be taken
account of either by adopting a new configuration (stoppers out board of the
pivot point plus possibly a twin axis design see Section 9.2) or by using a an
accurate wear/corrosion rate at the design stage.
13. Report on the in field performance of new mooring line instrumentation and
failure detection systems. Assess the feasibility of a Response Learning
System (see Section 17.3.1).
14. ROV collection of soil samples from the proximity of mooring lines to assess
the concentration of sulphate reducing bacteria (SRBs).
15. It is believed that there could be beneficial cross fertilization with flexible
riser and particularly steel catenary riser experience including touchdown
zones and inspection techniques.
16. Identify key common spares plus contingency connectors which could be held
by a shared Operators Spares Club see section 19.1.1.
17. Encourage the installation of simple inclinometers to aid in detecting line
failure if it occurs in the sea-bed mud see Section 17.2.2
With respect to steel components there is a need for additional reliable strength data to
assist with the following:
- to evaluate fatigue in connectors, terminations, etc,
- to evaluate bending and tension-bending fatigue in chains and also to
measure how chain surface finish can affect the friction between links,
- to better understand T-T fatigue for chains, currently given by T-N curves,
derived from full scale tests made in the late 1990's. A hot-spot S-N
approach, i.e. stresses by Finite Element analysis, strength derived from tests
on small scale specimen could be fruitfully used.
There are still uncertainties in estimating mooring loads using analysis software and
model tests. Hence, it would be desirable to compare the behaviour of a full scale FPS
in known weather conditions versus predictions. It is recommended that further work
should be done on this topic, although it is appreciated that there are difficulties
associated with obtaining reliable weather and instrumentation readings. The effect of
mooring shock loading when subject to breaking waves should also be assessed see
Section 3.1.10.
Overall there is a continuing need to monitor and report back to the mooring
community on issues which arise in future years as systems age, e.g. wear at FPSO
trumpets. This has been a particularly useful aspect of the Steering Committee
meetings to date and it would be desirable for this dialogue to continue. It is hoped
that this will be achieved through a Phase 2 Mooring Integrity JIP.
271
A4163-01
22 REFERENCES AND BIBLIOGRAPHY
Ref. 1 HSE Research Report 047, Analysis of Accident Statistics for Floating
Monohull and Fixed Installations 2003.
Ref. 2 The Centre for Marine and Petroleum Technology Floating Structures: a
Guide for Design and Analysis Volumes1 and 2.
Ref. 3 Experience with Mooring Integrity Assessment for Semi Submersibles
R.B. Inglis, Hydrodynamics: Computations, Model Tests and Reality
Proceedings of MARIN workshop on Advanced Vessels, Station Keeping,
Propulsor-Hull Interaction, and Nautical Simulators, Elsevier Science
Publishers B.V. , Amsterdam, 1992.
Ref. 4 International Organization for Standardization Draft International
Standard ISO/DIS 19901-7 Petroleum and natural gas industries Specific
requirements for offshore structures Part 7: stationkeeping systems for
floating offshore structures and mobile offshore units.
Ref. 5 DNV Offshore Standard Position Mooring, DNV-OS-E301, June 2001.
Ref. 6 Appendix A, Supplementary Requirements of the Norwegian Maritime
Directorate (NMD) and the Norwegian Petroleum Directorate (NPD)
POSMOOR 1996.
Ref. 7 Experimental Study of Load on an FPSO in Design Environmental
Conditions Skourup, J., Sterndorff, M.J., Smith, S.F., Cheng X., Ahilan,
R.V., Soares, C.G., and Pascoal, R., OMAE-FPSO04-0069, Houston).
Ref. 8 Noble Denton & Associates Inc Joint Industry Study Report Corrosion
Fatigue Testing of 76 mm Grade R3 & R4 Studless Mooring Chain dated
15 May 2002 (Report No: H5787/NDAI/MJW Rev 0).
Ref. 9 New Mooring Chain Designs by Luis Caada, Javier Vicinay, Alejandro
Sanz, Eduardo Lpez Vicinay Cadenas, SA., OTC 8149, 1996.
Ref. 10 Billington Osborne-Moss Engineering Limited (BOMEL) Design
Guidelines for Anchor Chains Final Report (Report No: C538R002.04
Rev A) dated June 1992.
Ref. 11 W.K. Lee and C.Z. Hua, "Theoretical and Experimental Stress Analysis to
Evaluate the Effect of Loose Studs in Anchor Chain," Conf. Proc.
Engineering Integrity Assessment, East Kilbride, Glasgow, 11-12 May
1994, pp. 171-191.
Ref. 12 Assessment of Mooring Chain from Mobile Drilling Unit, 19
th
Jan.
1994, Sandberg Consulting Engineers (Report No: M/5771/SCC/pb/03).
Ref. 13 Vicinay Chain Catalogue (Red).
Ref. 14 Development of API RP 2SM for Synthetic Fiber Rope Moorings by
Ming-Yoa Lee, American Bureau of Shipping, Paul Devlin, Texaco Inc
and Chi-Tat Thomas Kwan, Consultant.
272
A4163-01
Ref. 15 Guidance Notes on the Application of Synthetic Ropes for Offshore
Mooring by American Bureau of Shipping Incorporated by the Legislature
of and State of New York 1862 dated March 1999.
Ref. 16 Revised API RP2SK, Appendix A, under development.
Ref. 17 Michael F. Ashby and David R.H. Jones, Engineering Materials 1 An
Introduction to their Properties and Applications, Pergamon Pess Ltd.,
1980.
Ref. 18 DNV Certification of Offshore Mooring Chain, Note 2.6 dated August
1995.
Ref. 19 Marine Casualty Response: Salvage Engineering American Society of
Naval Engineers and JMS Naval Architects and Salvage Engineers.
Ref. 20 No Port in a Storm by Bob MacAlindin, published by Whittles (ISBN:
1870325370).
Ref. 21 Offshore Technology Report Review of Mooring Incidents in the Storms
of October 1991 and January 1992 Issued January 1992.
Ref. 22 Offshore Technology Conference 2004 Paper Post Mortem Failure
Assessment of MODUs during Hurricane Lili BP Malcolm Sharples,
Offshore Risk & Technology Consulting, Charles E Smith, Minerals
Management Service and Robert G Bea, University of California at
Berkeley.
Ref. 23 Offshore Technology Report 2000/086 Operational Safety of FPSOs:
Initial Summary Report prepared by Norwegian University of Science and
Technology (NTNU) for Health and Safety Executive.
Ref. 24 Jatar, S., Haslum, H., and Tule, J., The Design, Testing and Installation
of the Red Hawk Spar Polyester Taut Leg (TLM) System, 16
th
Annual
Deep Offshore Technology (DOT), New Orleans, Nov. 30
th
Dec. 2
nd
.
Ref. 25 Chaplin, Rebel & Ridge, Tension/Torsion Interactions in Multi-
component Mooring Lines, OTC012173.
Ref. 26 Mad Dog Polyester Mooring Installation, Petruska, d., Rijtema, S.,
Wylie, H., Geyer., J., 16
th
Annual Deep Offshore Technology (DOT),
New Orleans, Nov. 30
th
Dec. 2
nd
.
Ref. 27 Deep Offshore Technology (DOT 2004), New Orleans The Design,
Testing & Installation of the Red Hawk Spar Polyester Tank Leg Mooring
(TLM) System, Sanjai Jatar, Herbjorn Maslum, Jenifer Tule.
Ref. 28 British Standard BS 6349-1 2000 - Maritime Structures Part 1: Code of
Practice for General Criteria.
Ref. 29 DNV Rules for Classification of Mobile Offshore Units Special
Equipment and Systems Additional Class, Part 6 Chapter 2 July 1989
Position Mooring (POSMOOR).
Ref. 30 API Recommended Practice 2I, In-service Inspection of Mooring
Hardware for Floating Drilling Units, Second Edition, December 1996.
273
A4163-01
Ref. 31 API Recommended Practice for Design and Analysis of Stationkeeping
Systems for Floating Offshore Structures Second Edition December
1996 API RP2SK.
Ref. 32 Long term Mooring chains and components by Mr Pr Ohlsson,
Technical Manager, Scana Ramns AB, 3
rd
international Offshore
Moooring Seminar.
Ref. 33 Mooring Chain Corrosion Design Considerations for an FPSO in tropical
Water by Mark Wang and Richard DSouza, Deepwater Technology at
Kellogg Brown Proceedings of OMAE-FPSO I2004 OMAE Specialty
Symposium on FPSO Integrity, Houston USA 2004 - Paper No: 04-0046.
Ref. 34 Failure Analysis of a CALM Buoy Anchor Chain System by G. J. Shoup
and R. A. Mueller, Cities Service Oil & Gas Corp. - OTC 4764, 1984.
Ref. 35 Dowdy, M.J. and Graham, D.J., A Method for Evaluating and extending
the useful Life of In-Service Anchor Chain, OTC 5719, 1988.
Ref. 36 HSE Research Report 219 Design and integrity management of mobile
installation moorings P.J. Donaldson, M. Brown and M. Pithie (Noble
Denton).
Ref. 37 HSE Safety Notice 3.2005 Floating Production Storage and Offloading
(FPSO) Mooring Inspection issued April 2005 see Appendix D.
Ref. 38 The Professional Divers Handbook, published by Submex Limited
(ISBN: 09508242 0 8) 1982.
Ref. 39 Design and Analysis of West Seno Floating Structures Jafar Korloo
(Unocal), Jared Black (Unocal), Chunfa Wu (SEA Engineering), J. Hans
Treu (SEA Engineering) Presented at Offshore Technology Conference
held in Houston 3-6 May 2004, OTC 16523.
Ref. 40 Na Kika Deepwater Mooring and Host Installation A.K. Paton (Shell
International E&P Inc.), J.D. Smith (Shell), J.A. Newlin (Shell), L.S.
Wong (Shell), E.S. Piter (Edmar Engineering Inc); C. van Beek (Heerema
Marine Contractors BV). Presented at Offshore Technology Conference
held in Houston 3-6 May 2004, OTC 16702.
Ref. 41 Noble Denton Europe Limited Joint Industry Project The Evaluation of
Wire Mooring Line Strength and Endurance, Additional Testing of Steel
Wire Rope Final Report (Report No: L17294Rev1/NDE/RWPS) dated
16 February 1996.
Ref. 42 Noble Denton Europe Limited Joint Industry Project The Evaluation of
Wire Mooring Line Strength and Endurance (Old Ropes II) Final
Report (Report No: L18085/NDE/RWPS) dated 2 April 1997.
Ref. 43 Reading Rope Research The Inspection & Discard of Wire Mooring
Lines C. Richard Chaplin December 1992. Prepared as a supplement for
participants in a joint industry study on an Appraisal of Discarded
Mooring Ropes.
Ref. 44 Final Report of a Joint Industry Study on Prediction of Wire Rope
Endurance for Mooring Offshore Structures, working summary by C
Richard Chaplin, Department of Engineering at university of Reading
August 1991.
274
A4163-01
Ref. 45 Noble Denton Europe Limited Report for TotalFinaElf Exploration UK
Plc Investigation into the Run-out of Number 6 Mooring chain on
Transocean John Shaw (Report No: A4071/01/NDE/CLC/Ls) dated 20
February 2003.
Ref. 46 HSE Offshore Technology Report OTO 98 086 Quick Release Systems
for Moorings issued April 1998.
Ref. 47 HSE Offshore Technology Report 2001/073 Failure modes, reliability
and integrity of floating storage unit (FPSO, FSU) turret and swivel
systems.
Ref. 48 Turret Operations in the North Sea: Experience from Norne and Asgard
A by Borre Knudsen and Bard A. Leite - Procs of the Eleventh (2001)
International Offshore and Polar Engineering Conference, Stavanger,
Norway 17-22 June 2001.
Ref. 49 Health & Safety Executive Analysis of accident statistics for floating
monohull and fixed installations prepared by Martin Muncer , Research
Report 047, 2003
Ref. 50 Petroleum Safety Authority Norway Trends in risk Levels Norwegian
Continental Shelf summary Report Phase 4 2003.
Ref. 51 Forging Solutions #17 published by the Forging Industry Association.
Ref. 52 British Paper GB190607951, 1905, Improvements in Chain Coupling-
links.
Ref. 53 Bureau Veritas Guidance Note Certification of Synthetic Fibre Ropes for
Mooring Systems 1997 NI 432 DTO R00 E 1997.
Ref. 54 OTC 6905, 1992, The Influence of Proof Loading on the Fatigue Life of
Anchor Chain, Shoup, George J., Tipton, S.M., and Sorem, J.R.
Ref. 55 WADO Deepwater Mooring Conference, Paris, June 2003.
Ref. 56 Floating Production Mooring Integrity JIP Key Findings, OTC 17499,
2005, Martin G. Brown, Tony D. Hall, Douglas G. Marr, Max English,
and Richard O. Snell see Appendix C.
Ref. 57 DVN Fatigue Strength analysis of Offshore Steel Structures, DNV-RP-
C203, October 2001.
Ref. 58 N.F Casey, National Engineering Laboratory, Department of Trade and
Industry, Monitoring the Properties of Wire Ropes Subjected to Bending-
Tension Fatigue around Sheaves, DE/7/88, November 1988.
Ref. 59 Richard Chaplin & Andrew Potts, Wire Rope Offshore A Critical
Review of Wire Rope Endurance Research Affecting Offshore
Applications, HSE, OTH 91 341, 1991.
Ref. 60 Cost Effective Mooring Integrity Inspection Methods, Hall, A.D., OTC
2005, May 2-5, Houston, paper 17498.
Ref. 61 Performance and Testing of Components of the Ivanhoe/Rob Roy Floating
Production System Mooring, J.R. MacGregor and S.N. Smith, Amerada
Hess Ltd, and J.E Paton, JP Kenny (Caledonia), OTC 7492 1994.
275
A4163-01
Ref. 62 Joint Industry Project - Sub Sea Electrol Magnetic Appraisal of Wire
Mooring Lines (The SEAL Project) Report No: L17770/NDE/RWPS
dated 31 May 1996.
Ref. 63 HSE Research Report 328 Acoustic monitoring of the hulls of Floating
Production Storage and Offloading facilities (FPSOs) for corrosion and
damage prepared by Mecon Limited 2005.
Ref. 64 Section 4, Wire Rope Research at the NEL an Overview, N.F. Casey,
Nov. 1988.
Ref. 65 American Petroleum Institue Specification for Mooring Chain API
Specification 2F, Sixth Edition June 1997.
Ref. 66 International Association of Classification Societies (IACS)
Requirements concerning Materials and Welding W22 Offshore
Mooring Chain.
Ref. 67 The modelling and analysis of splices used in synthetic ropes; Leech, C
M Procedings of the Royal Society, published online,
doi:10.1098/rspa.2002.1105 (2003).
Ref. 68 The Analysis of Splices used in Large Synthetic Ropes, C M Leech,
Europmech 334 (Textile Materials and Textile Composites) Universit
Lyon, France. 1995.
Ref. 69 Engineers Design Guide for Deepwater Fiber Moorings, 1
st
Edition,
NDE/TTI Joint Industry Project, January 1999.
Ref. 70 OTC paper 10798, 1999 Genesis Spar Hull and Mooring System : Project
Execution, (W.F. Krieger, Chevron Petroleum Technology Co., J.C.
Heslop, Chevron U.S.A. Inc., B.E. Lundvall, Exxon Upstream
Development Co. and D.T. McDonald, Chevron U.S.A.
Ref. 71 NACE International, The Corrosion Society Corrosion Control of Steel
Fixed Offshore Structures Associated with Petroleum Producetion NACE
Standard RP0176-2003 Item No 21018.
Ref. 72 European Committee for Standardization EN 13173 ICS 47.020.01;77.060
Cathodic Protection for Steel Offshore Floating Structres January 2001.
Ref. 73 Health & Safety Executive Offshore Technology Report ITC 96 033 A
Review of Available Laboratory Test Data on Mooring chain
Applications dated April 1998.
Ref. 74 Health & Safety Exeuctive Offshore Technology Report ITO 96 018
Improved Reliability of Mooring Chain Systems dated May 1997.
276
A4163-01
23 APPENDIX A - SUMMARY OF PAST RELEVANT
JIPS
- Engineers Design Guide to the use of Deepwater Fibre Mooring Lines
- 1996, 31 participants, RWPS most knowledge
- Corrosion Fatigue of Studless Mooring Chain
- 2002 final report, Noble Denton Houston
- Subsea Electro-Magnetic Appraisal of Mooring Lines (SEAL)
- 1995 Noble Denton London
- Evaluation of Wire Mooring Line Strength and Endurance
- 1996/97 Old Ropes 2 and 2.5
- Mooring Code Joint Industry Study,
- Noble Denton & Associates, October 1995
- Fatigue Tests for Large Diameter Mooring Anchor Chains
- 1994 Final report H3241/NDAI/JIS, Houston work,
- High-Technology Fibres for Deep Water Tethers and Moorings
- 1995, otherwise known as Fibre Tethers 2000
The Appraisal of Discarded Mooring Lines
1992 Richard Chaplin Blue Book
The Prediction of Wire Rope Endurance for Mooring Offshore Structures
1990 final report, NEL involved
Mooring Integrity: A State-of-the-Art Review
1991-1993, Dr. Ahilan report for Shell
HSE Mooring Guidance
1989-1993
278
A4163-01
24 APPENDIX B MOORING INTEGRITY
QUESTIONNAIRE (EXCEL)
279
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
A. GENERAL DETAILS
A1. Unit Name A2. Field Name
A3. Unit Type
A4. Water Depth m A5. Geographical Area
A6. Date Installed
A7. Is the FPS classed?
A8. Has the unit ever been used elsewhere?
A9. Was the unit ever removed from site and then re-installed?
A10. Can the mooring system be disconnected in case of typhoons or ice bergs ?
A11. If the moorings can be disconnected, how often has this happened to date ?
B. MOORING SYSTEM MAKE-UP
B1. Line Make-up Non coated spiral strand wire
B2. Configuration Sub surface buoys used no swivels
B3. Approx. symmetrical Assymetrical system due to likely typhoon direction
system ?
B4. Can the lines tensions be adjusted during normal operations?
B5. How often is the line "worked" to prevent localised wear?
B6. Anchor Type?
B7. Length & make up of first line segment FROM ANCHOR (eg. 25m of 120mm ORQ studless chain)
m
B8. Length & make up of grounded length section
m
B9. Length & make up of catenary section
m
B10. Length & make up of final line section into FPS
m
B11. If applicable position, make up & length of any weighted line sections or buoyancy modules
JOINT INDUSTRY PROJECT: FPS MOORING INTEGRITY
QUESTIONNAIRE
310
Chain
ABS
No
No
?
?
? a
?
No
No
No
of
of
of
of
Nan-Hai-Sheng-Li Liuhua, South China Sea
Turret FPSO
South East Asia
Mar 1996
W ire Rope Polyester Rope
Catenary
Classification Society:
Not Possible
Drag Embedment
Studlink
4.5" Bridon Non-Coated spiral strand wire
4.5" Ramnas studded chain
4.5" Bridon Non-Coated Spiral Strand W ire + subsea buoy + spiral str
4.5" Ramnas Studded Chain
A draw ing of the 25te ne t buoya ncy subse a buoys
w ould be a ppre cia te d.
Yes
N.A.
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
C. HEADING CONTROL + THRUSTERS
C1. Thrusters Type
C2. Are thrusters used for normal heading control ?
C3. Are thruster used to reduce line tensions?
C4. Any thruster problems encountered which could affect mooring integrity?
C5. Any significant problems encountered during line "winching" operations?
C6. Any turret problem e.g. bearing or jacking/locking difficulties, but excluding swivel faults?
C7. Turret Designer?
C8. The turret is
D. MOORING DESIGN/CODES & STANDARDS
D1. Which Mooring Code was the system designed to ? Please Advise ?
D2. Presently anticipated field life ? years
D3. Calculated fatigue design life excluding safety factor ? years Please Advise ?
(if safety factor is included please advise)
None
No
No
?
20
No
Not Applicable
SOFEC
Free weathervaning
Don't know
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
E. LINE LOCK OFF & TENSIONING
E1. Means of Tensioning? Please Advise ?
E2. How is the line locked-off?
E3. Where is it locked-off?
E4. Freedom to Rotate?
W W ire Rope Drum-type
Chain Stopper
Submerged - Base of turret
Vertically
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
F. CONNECTORS & INTERFACES INCLUDING FAIRLEADS & SEA-BED TOUCH DOWN
F1. Connector type to anchor?
F2. Connector type between anchor line and grounded sections?
F3. Connector type between grounded and catenary sections?
F4. Connector type between catenary and final line section?
F5. Split pins/nylocs used on all shackle pins?
F6. Have you experienced connector failure or significant degradation ?
F7. Type of fairleads?
F8. Fairlead Fredom to Rotate?
F9. What is the approximate make up of the sea bed at the touch down points?
F10. Has signigicant wear been experienced at the sea bed touch down?
F11. Have trenches been excavated at sea-bed touch down?
??
mm
??? m
Other, please specify Anchor Shackle
Other, please specify
Socket - triplate - shackle
Other, please specify
Shackle - triplate - socket. Please advise
connections to subsea 25te net buoyancy
buoy?
Other, please specify
Socket - triplate - shackle
Don't know
Combined Stopper & Trumpet Assembly
Vertically
Other, please advise Please advise ?
Yes
How Much?
Yes How Deep?
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
G. TRANSPORTATION & INSTALLATION
G1. Who was the principal mooring installation & hook up contractor?
G2. Any line damage during transportation?
G3. Any line damage during installation?
G4. Do you know of any twists in the mooring lines?
G5. Are all the lines straight from the anchors to the fairleads ?
G6. What is the approximate maximum tension variation between lines in dead calm conditions ? Te
G7. Any lessons learned during transportation and installation ?
???
Clough Stena
Yes
Please, provide details
Don't know
Don't know
???????
Please advise ?
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
H. LINE TENSION MONITORING
H1. Are line tension read outs available in real time ?
H2. How are the tensions measured ?
H3. If applicable what is the approximate accuracy of the tension readout (e.g. +/- 20 tonnes) Te
During installation
H4. Are the line tensions recorded and the data permanently preserved ?
H5. Are the offsets measured from no load equilibrium position?
H6. Accuracy of measurements (e.g. +/- 0.5m) m
H7. Offsets recorded and data permanently preserved?
H8. Has the recorded data been validated against the original mooring design estimates?
No
??
No
No
Other, please advise The line tensions, in terms of chain angle
at the trumpets, are only measured during
initial installation and during inspection
surveys ???
Yes
????
Don't know
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
I. LINE FAILURE DETECTION
I1. Line failure detection devices on all lines?
I2. Basis of line failure detection devices?
I3. Line failure detection operational on all lines?
I4. 24 hours a day monitored alarms on failure detection devices?
I5. Please, provide details of any breakdowns of line failure detection devices, if known
I6. Is it possible to quickly distinguish between instrument failure and line breakage ?
I7. Please provide any details if there is any plan to install or add on existing system
No
No
No
No
Other, please specify
????
Not applicable ?
????
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
J. INSPECTION, REPAIR & MAINTENANCE (IRM)
J1. What is the frequency of mooring line inspection?
J2. How is the scope of line inspection determined?
J3. Is mooring inspection part of planned maintenance system?
J4. What is the principal mean of inspection?
J5. How is wear measured?
J6. In case the chain stoppers are submerged, how can they be inspected?
Other, please specify
Please advise ?
Classification Society Requirement
Yes
Other, please specify General visual inspection by ROV ?
Other, please specify
Have you used the W elaptega Marine ROV camera based
system for detecting wear? If not, have you used any vaguely
similar system ?
Is visual inspection almost impossible due to the chain stoppers
being at the end of a trumpet or hawse pipe surrounded by the
spider structure ?
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
J7. During inspection, is a check made on the security of all pins?
J8. Anchor connections
J9. Any significant problems encountered with anchors during or since installation?
J10. Are the risers normally inspected at the same time as the moorings ?
J11. Has significant wear been detected where the chain emerges from the
trumpets at the base of the turret or at the fairleads ?
J12. For systems with studded chain, have loose or missing studs been detected?
J13. Have you detected defects that are common to more than one line?
J14. Has wear been greater on windward (most heavily loaded) or leeward lines?
J15. How is corrosion prevented?
J16. Any particular corrosion problems?
J17. If the lines are electrically isolated, how is this checked?
J18. How often are the lines recovered for inspection?
J19. If weighted line/clump weights are used, have they stayed intact?
No
No
Only when easily visible
They can't be inspected, they are under mudline
Sometimes
Yes, please supply details
If wear please specify the level in mm if possible ?
Don't know
Other, please specify Corrosion allowance on the chain ?
Measuring cathodic potential levels via a ROV deployed
probe ?
Not Recovered
Not Applicable
Not Applicable
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
K. PRODUCTION OPERATIONS AFTER ONE LINE FAILED
K1. Does the exisiting Emergency Response Plan/Safety Case allow continued operation after one line failed?
K2. Are there pre-defined maximum environmental limits for continued operations after one line failed?
K3. Are there 24 hours a day monitored alarms if the offsets exceed a pre-defined level?
K4. What triggers the decision to suspend production? n
Yes
Yes, please supply details ????
Yes
Instrumentation readouts + Experie
Questionnaire
Nan-Hai-Sheng-Li-(CNOOC-SOFEC)-sketches.xls
L. HISTORY OF STATION KEEPING FAILURES
L1. How many stationkeeping failures have you experienced?
L2. Have you experienced multiple failures at the same time?
L3. Are these failures covered by completed Incident Summary Worksheets ?
M. SPARES
M1. What mooring line spares do you have which are immediately available?
M2. Do written contingency procedures exist for rapid deployment of a replacement mooring line?
M3. Are any mooring components routinely changed out?
N. PERSONNEL INFORMATION
N1. Name
N2. Position
N3. E-mail adress
N4. Email address of back to back cover
?
No
No
No
Connectors
W ire Rope
Chain
Fibre Rope
Instrumentation
Other, please specify
Do you hold any spares ?
No, I want to create Line Failure Incident
Reports
Questionnaire
1. Incident Description:
2. When did failure occurr?
3. Details of Failure:
JOINT INDUSTRY PROJECT: FPS MOORING INTEG
INCIDENT REPORT EXAMPLE
Failure of the retaining bolts on a wire open socket which
allowed the pin to come free and the mooring line to part.
Two to three years after installation.
On a turret moored moored floating storage unit the connection
detail between the chain and the wire consisted of a wire end
socket, a triplate and a D type shackle connecting to the
grounded chain. The wire open socket was connected to the
triplate by a round connecting pin that was held in place by an
end plate secured to the socket by 3 bolts aorund its
cicumference ands to the pin by three bolts in a line. During a
subsea survey it was found that the end plate had dropped off
and the pin dropped out. It was also noted that the end plate
bolts had failed or backed off on a number of the other socket
connections althouh the pins had not yet dropped out.
4. Probable Incident Cause (if known) including weather conditions at the time of failure
5. Incident Consequences:
6. How was the failure detected?
The connection was inb the sea bed working section of the
catenary (the thrash zone) and as the wire socket was repeatedly
picked up and set down there was a large relative motion
between the socket and thge heavier triplate and grounded chain
section that typically remains on the sea-bed. It is thought that
this introduced a large torsional/friction load between the pin
anbd the body of the socket that could not be accommodated by
the end plate retaining bolts and these failed allowing the end
plate to drop off and the pin to fall out. There had also been a
failure to insulate properly the wire section from the chain
section and the cathodic protection on the wire weas drained
down by the grounded chain section resaulting in a corrosive
environment that might have contributed to the failure.
Temporarily restricted offloading operations plus repair costs.
Subsea ROV survey
7. Remedial action?
8. Mean time to repair?
- Replacement of pins, modification to the retaining bolts and re
instatement of the cathodic protection.
Initial repair approximately 12 weeks, total refurbishment several
months,
A4163-01
25 APPENDIX C 2005 OTC JIP PAPER
280
OTC 17499
Floating Production Mooring Integrity JIP Key Findings
Martin G. Brown, Noble Denton Europe Limited
Tony D. Hall, Welaptega Marine Limited
Douglas G. Marr, Balmoral Marine Limited
Max English, U.K. Health and Safety Executive
Richard O. Snell, B.P. 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., 25 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 (FPSs),
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
2 [OTC 17499]
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.
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.
Corrosion
Wear & fatigue
Bending & Tension
Highest Tensions
Impact & Abrasion
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 10 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
3 [OTC 17499]
is affecting all lines in a particular quadrant to approximately
the same extent, see Figure 2.
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 100m
3
to
2,500m
3
. 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.
4 [OTC 17499]
Localised Wear
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 APIs RP2SK and DNVs 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.
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
5 [OTC 17499]
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 bending
9
.
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 90 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
7 [OTC 17499]
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]
reported to the UK Health and Safety Executive (HSE).
Although the North Sea is a hostile climate, units intended for
Line No11
use here are in general designed to a high standard. In
addition, a number of units in the North Sea have been around
W
i
r
e

p
a
y
o
u
t

(
m
)

195.0
194.0
193.0
192.0
191.0
190.0
189.0
188.0
187.0
186.0
185.0
0.0 20.0 40.0 60.0 80.0 100.0 120.0
Tension (te)
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
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.
9 [OTC 17499]
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.
Figure 13 - Illustration of ROV deployed optical caliper
7
measurement system
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 available
7
. 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 systems
7
. 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 footage
7
.
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 season
10
, (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
11 [OTC 17499]
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 catenarys 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,
Lloyds 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.
A4163-01
26 APPENDIX D HSE SAFETY NOTICE 3.2005
FLOATING PRODUCTION AND OFFLOADING
(FPSO) MOORING INSPECTION
281
Floating Production Storage and Offloading (FPSO) -
Mooring Inspection
- Safety notice: 3/2005
- Issue date: Apr 2005
Introduction
1. This notice is for operators of monohull weather vaning FPSOs and FSUs. It
explains why they need to ensure that the top sections of their mooring chains are
not subject to excessive wear that can affect the integrity of the mooring system.
Background
2. It has come to the attention of HSE that premature and unexpected mooring chain
wear has been experienced on one UKCS FPSO inside the trumpet connected to
the turret's chain table (spider). Essentially, the mooring chain is directed through
a carefully designed trumpet that has the ability to rotate about a horizontal axis
and thus accommodate the vertical motions of the FPSO without transferring
significant bending or twist into the mooring chain.
3. Damage to chain links at the trumpet bell mouth, within the trumpet body itself,
and around the chain stopper suggests that unexpected wear is occurring. The
exact cause of the wear has as yet not been ascertained. Causes might include
inaccurate offshore installation, defective design of the trumpet and/or unforeseen
load conditions at the trumpet.
4. The wear experienced is generally manifested as a loss of cross-sectional area. In
some chain links the loss of material has been such that retained strength would be
insufficient to achieve design factors of safety. Furthermore, the deterioration in
the chain links has occurred after just 4 years into a 20-year design life.
5. The principal concern to HSE is that similar wear mechanisms may be taking
place on other floating installations that have mooring chains passing through a
trumpet. Defects that affect more than one mooring chain can increase the risk of
multiple mooring line or system failure.
Action required
6. Operators of FPSO and FSU installation need to be aware of such occurrences,
and to ensure that they have suitable inspection routines in place.
-
-
-
-
-
7. Operators should inspect the mooring chains around and inside the mooring
trumpet during 2005, and take any necessary remedial action to ensure the
continuing integrity of the mooring system.
8. Periodic inspection of the chain around and inside the trumpet should be carried
out based upon the findings of the initial inspection.
Further information
Any queries relating to this notice should be addressed to:
Health and Safety Executive
Hazardous Installations Directorate
Offshore Division
Lord Cullen House
Fraser Place
Aberdeen
AB25 3UB
Tel: 01224 252500
Fax: 01224 252615
This guidance is issued by the Health and Safety Executive. Following the guidance is
not compulsory and you are free to take other action. But if you do follow the guidance
you will normally be doing enough to comply with the law. Health and safety inspectors
seek to secure compliance with the law and may refer to this guidance as illustrating good
practice.
- Updated 27.09.05
Copyright
Disclaimer
Freedom of information
Accessibility
Back to top
Printed and published by the Health and Safety Executive
C30 1/98
Published by the Health and Safety Executive
05/06
RR 444