esign and Analysis of

Stationkeeping Systems for

Floatin Structur

Upstream Segment

RECOMMENDED PRACTICE ZSK

THIRD EDITION, OCTOBER 2005

“r-1¢c:"».*-' :--- -. - - 5'; w. - .--“ .-.~¢<.;.-.-¢.1.-.-~

.. ..- »_... :-:_1;;;:-_9¢5:;.
. _
-. <§"\> ' . .-.<>_:>.._’ :>:_:-:.. './1 . .-f;>$s$é'. E
:61 . .- .-:=:¢:-:5:-‘/41;:-93¢» :'.i1=I-:I-IN-12
. - .. , ,
'-1&$"4-:-:3? 1:2;-$¢Z-11.47‘
:_§§3‘- 'T:I?f3_-1- '- 1:329:-""1:_-:?:<'ci:'<-.
¢\'-“-\' ‘| HG“ >r‘4I .
-. E:§:, '
' | :24-,1!"/1
3” " ‘ "“
'=i=:==; -‘H-I-\i -"1-\<-.*>:‘ . -. :> ’5:9'¥;|§3.
' ‘"’/4-\s.*
= E=:¢¢¢.-
n . . 1 " 2.23
n

V»: :d‘.-0%‘/;y
.-‘-:~:¥- "-; . .-> .- 14
' :{<$;{Z_';»;,¢;:;\==-:-z».=§<§§£
1 I

-; .<.W,.,<.
' r;fl‘.< ‘;:: . \ .\. , .
I xi’: :>:=:Z>:' .....-:9.-.->:~
555-q>>§. - '- ..:-:--:14»
¢?-I5!/é$,$:i:?;f;}13:j$.--.¢¢" 1'.\..q.».¢.<.
; ;:->'-
.- 2:‘? ‘=:=>
. ¢:‘
. :-:-z-:;.-:-
;:;:;:;:-.;:-.-:_->;;g4~.a:-'->. ' :;:c\?-*1;\-:<:.:l -- .-:-..'-- -- .-:-:<¢9<:- .-
-'¢'.:-;-:-:-;-:-v---W-2=-e.-mm: ' I -1:14:11‘);->:
:- 5:4 .1-‘ ;§.1'=:-
._e1: _ ..-:~. ;1.-.-.;.-I 1<-:.;--9;-1
-- -;'¢._._.'\_.>.->:;.;: -y .. ; ;.:.;.;e;_.;;;.;.;.
_-./r- ;;-.-.. -'
;1;Z§;!_:I£):*.:§ _:: ' '-‘#"¢?rl*
45:1:-' ' ' .-:‘<-;q53,;<§“~§f:§;‘:.-:?2.;: - ;=<:;g:;:(;;_¢5 -
-1» - '1':|:- : .- ._..->-1:->-:.
~~ . - ..-. :
-:->‘.¢'ei:-:=5:I:I:-:-:=:i I .=
-14-: .-;;,--
..1--{P
-1 <#.€ --
-' :15
. '-1'1‘:
- __,;;- Q.
...-;I=..;:_:_.‘_
.:>:-.1 . ..1;1;‘§§;§:;
. we>t<!€GQ<¥'.
 SPECIAL NOTES

AP] publications necessarily address problems of a general nature. With respect to partic-
ular circumstances, local, state, and federal laws and regulations should be reviewed.
Neither API nor any of API’s employees, subcontractors, consultants, committees, or
other assignees make any warranty or representation, either express or implied, with respect
to the accuracy, completeness, or usefulness of the information contained herein, or assume
any liability or responsibility for any use, or the results of such use, of any information or
process disclosed in this publication. Neither API nor any of API's employees, subcontrac-
tors, consultants, or other assignees represent that use of this publication would not infringe
upon privately owned rights.
API publications may be used by anyone desiring to do so. Every effort; has been made by
the Institute to assure the accuracy and reliability of the data contained in them; however, the
Institute makes no representation, warranty, or guarantee in connection with this publication
and hereby expressly disclaims any liability or responsibility for loss or damage resulting
from its use or for the violation of any authorities having jurisdiction with which this publi»
cation may conflict.
API publications are published to facilitate the broad availability of proven, sound engi-»
neering and operating practices. These publications are not intended to obviate the need for
applying sound engineering judgment regarding when and where these publications should
be utilized. The formulation and publication of API publications is not intended in any way
to inhibit anyone from using any other practices.
Any manufacturer marking equipment or materials in conformance with the marking
requirements of an API standard is solely responsible for complying with all the applicable
requirements of that standard. API does not represent, warrant, or guarantee that such prod-
ucts do in. fact conform to the applicable API standard.

All rights reserved. No part ofthis work may be reproduced, stored in a retrieval system, or
transmitted by any means, electronic, mechanical, photocopying, recording, or otherwise,
withontprior written permission fi"om the publisher Contact the Publisher;
/{PI Publishing Services, I220 L Street, N. W, Washington, D. C. Z0005.
Copyright © 2005 American Petroleum Institute
 FOREWORD

This recommended practice is under the jurisdiction of the API Subcommittee on Stan-
dardization of Offshore Structures.
Nothing contained in any API publication is to be construed as granting any right, by
implication or otherwise, for the manufacture, sale, or use of any method, apparatus, or prod-
uct covered by letters patent. Neither should anything contained in the publication be con-
strued as insuring anyone against liability for infringement of letters patent.
This document was produced under API standardization procedures that ensure appropri-
ate notification and participation in the developmental process and is designated as an API
standard. Questions concerning the interpretation of the content of this publication or com.-~
ments and questions concerning the procedures under which this publication was developed
should be directed in writing to the Director of Standards, American Petroleum Institute,
1220 L Street, N.W., Washington, D.C. 20005. Requests for permission to reproduce or
translate all or any part of the material published herein should also be addressed to the
director.
Generally, API standards are reviewed and revised, reaffirmed, or withdrawn at least
every five years. A one-time extension of up to two years may be added to this review cycle.
Status of the publication can be ascertained from the API Standards Department, telephone
(202) 682-8000. A catalog of API publications and materials is published annually and
updated quarterly by API, 1220 L Street, N.W., Washington, D.C. 20005.
Suggested revisions are invited and should be submitted to the Standards and Publications
Department, API, 1220 L Street, NW, Washington, DC 20005, standards@api.org.
 CONTENTS

Page
SCOPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

BASIC CONSIDERATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .....

2.1 Introduction to Stationkeeping Systems . . . . . . . . . . . . . . . . . . . . . ...........
2.2 Mooring Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3 Permanent and Mobile Mooring Systems . . . . . . . . . . . . . . . . . . . . ...........
2.4 Design Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5 Foreseeable Hazards for Mooring Systems. . . . . . . . . . . . . . . . . . . ... -O
co La;-
I
I.. [\.) JO\CJ\i-»+-~

ENVIRONMENTAL CRITERIA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
3.1 Environmental Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2 Environmental Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3 Wind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
3.4 Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
3.5 Current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
3.6 Water Depth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
3.7 Soil and Seafloor Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.8 Atmospheric Icing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . r-P1|- \ C>\CT'\C3'\U ‘\(.J1U‘l-3=-13>-
3.9 Marine Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1...... O\ -0 0 n -l 0

ENVIRONMENTAL FORCES AND VESSEL MOTIONS . . . . . . . . . . . . . . . . . . . .16

4.1 Basic Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16
4.2 Guidelines for the Evaluation of Environmental Forces and Vessel Motions . . .16
4.3 Simplified Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

MOORING STRENGTH ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.1 Basic Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I» .. I
Q .. I
I .. Q
Q I--»
I-~ Q--.._]
5.2 Mooring Analysis Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19
5.3 Vessel Offset. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19
5.4 Line Tension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19
5.5 Statistics of Peak Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20
5.6 Strength Analysis Based on Frequency Domain Vessel Dynamic. . . . . . . . . . . .20
5.7 Strength Analysis based on Time Domain Vessel Dynamics . . . . . . . . . . . . . . . .22
5.8 Strength Analysis Based on Combined Time and Frequency Domain
Vessel Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22
5.9 Thruster Assisted Mooring (TAM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22
5.10 Transient Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

FATIGUE ANALYSIS . . . . . . . . . . . ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . .24

6.1 Basic Considerations . . . . . . . . ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . .24
6.2 Fatigue Resistance of Mooring Components . . . . . . . . . . . . . . . . . . . . . . . . . . . .25
6.3 Fatigue Analysis Formulation . ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . .27
6.4 Fatigue Analysis Procedure . . . ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . .30

DESIGN CRITERIA . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30

7.1 Vessel Offset. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30
7.2 Line Tension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30
7.3 Line Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30
7.4 Anchoring Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30
7.5 Fatigue Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31
7.6 Corrosion and Wear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31
7.7 Clearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32
7.8 Supporting Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32

iv
 Page
8 MOORING HARDWARE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32
8.1 Basic Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32
8.2 Mooring Line Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32
8.3 Winching Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33
8.4 Monitoring Equipment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34

9 DYNAMIC POSITIONING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..34

9.1 Basic Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..34
9.2 Design and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34
9.3 Operating Personnel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36

10 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36

APPENDIX A INTRODUCTION TO MOORING COMPONENTS . . . . . . . . . . . . . . . .39

APPENDIX B RECOMMENDED WIND SPECTRUM . . . . . . . . . . . . . . . . . . . . . . . . .59
APPENDIX C SIMPLIFIED METHODS FOR THE EVALUATION OF
ENVIRONMENTAL FORCES AND VESSEL MOTIONS . . . . . . . . . .63
APPENDIX. D DRAG EMBEDMENT ANCHOR DESIGN . . . . . . . . . . . . . . . . . . . . . . 83
APPENDIX E PILE AND PLATE ANCHOR DESIGN AND INSTALLATION . . . . .93
APPENDIX F DETERMINATION OF AVAILABLE THRUST . . . . . . . . . . . . . . . . I19
APPENDIX G MOORING STRENGTH RELIABILITY . . . . . . . . . . . . . . . . . . . . . . I25
APPENDIX H MOOIUNG DESIGN FOR VORTEX INDUCED MOTIONS (VIM) 129
APPENDIX CH COMMENTARY ON APPENDIX H--~l\/IOORI.NG DESIGN FOR
VORTEX. INDUCED MOTION (VIM) . . . . . . . . . . . . . . . . . . . . . . . . . 135
APPENDIX I GLOBAL ANALYSIS GUIDELINES FOR DEEPWATER
FLOATING SYSTEMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
APPENDIX J MOORING STRENGTH AND FATIGUE ANALYSIS EXAMPLES I71

E? fU

Recommended Analysis Methods and Conditions. . . . . . . . . . . . . . . . . . . . . . . . . 19

Intact and Damaged TAM Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Mand.K Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Comparison of B-T and T-T Fatigue Life. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Tension Limits and Safety Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
c\ui.n.o>t\>- SE Drag Anchor Safety Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
7 Safety Factors for Pile, Plate, and Gravity Anchors (Dynamic Analysis) . . . . . . . 31
B.l Coefficients for NPD 2-Point Coherence Spectrum . . . . . . . . . . . . . . . . . . . . . . . 60
C.l Wind Force Shape Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
C.2 Wind Force Height Coefficients (for 1~Minute Wind) . . . . . . . . . . . . . . . . . . . . . 67
C.3 Wind Velocity Time Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
D.1 Estimated Maximum Fluke Tip Penetration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
E. 1 Recommended NC Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
E.2 Suction Pile Safety Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112
E.3 Design Factors for Finite Element Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112
F. 1 Correction Factor for Inflow Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
F.2 Thrust Losses in Reverse Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
I. l. Fairlead Motion Complex RAO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
J.2 Tension Spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I76
1.3 Environmental Condition, Mean Loads and Low-Frequency Motions
for the Analyzed Direction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
.I.4 Wire Rope Damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
J.5 Chain Damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
1.6 Annual Fatigue Damage as a Function of the Environment . . . . . . . . . . . . . . . . 181

V
 Page
Figures
1 Spread Mooring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2 TLP Lateral Mooring System . . . . . . . . . . . . . . . . . . . . . .
Differential Compliance Anchoring System (DICAS) . . . . . . . . . . . . . . . . . . . . . .
Typical External Turret Mooring Arrangement . . . . . . . . . . . . . . . . . . . . . .
Typical Internal. Turret Mooring Arrangement . . . . . . . . . . . . . . . . . . . . . . . .
Variations on Turret/Riser System . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Catenary Anchor Leg Mooring (CALM) with Hawsers . . . . . . . . . . . . . . . . . . . .
Catenary Anchor Leg Mooring (CALM) with Fixed Yoke . . . . . . . . . . . . . . . . . . .
Catenary Anchor Leg Mooring (CALM) with Soft Yoke . . . . . . . . . . . . . .
Single Anchor Leg Mooring (SALM) with Tubular Riser and Yoke . . . . . .
Single Anchor Leg Mooring (SALM) with Chain Riser and Hawser . . . . . . . . . .
4

Dynamic Positioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ;. . _-I .I I. I »-©\o 0\1o'u1-bc~al\J

Mooring Fatigue Design Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
-1

KO
O\
O0
U:
U3 -D-|>~£.»Jl\)" -‘CD
tiet=‘-I
<*P4 Major Elements of a Dynamic Positioning System. . . . . . . . . . . . . . . . . . . . . . . . . 35
I5 Dynamic Positioning Capacity Rosettes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
A.l Stud and Studless Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
A.2 Typical Wire Rope Constructions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
A.3 Wire Rope Socket for Permanent Mooring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
A.5 Submersible Buoy Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
A.6 Typical Mooring Connectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
A.7 Typical Subsea Mooring Connectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
A.8 Winching Equipment for Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
A.9 Chain Jack and Chain Bending Shoe Fairlead . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
A. 1 0 Drum Type Winch and Fairlead for Wire Rope . . . . . . . . . . . . . . . . . . . . . . . . . . 50
A.1.l Linear Winch and Bending Shoe Fairlead for Wire Rope . . . . . . . . . . . . . . . . . . 52
2 Traditional Drag Embedment Anchor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
l3 Suction Pile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 54
14 Suction Caisson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
15 Drag Embedded Plate Anchor (VLA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
PP???" 6 Suction Embedded Plate Anchor (SEPLA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
.B.1 Comparison of API and NPD Spectrum for 25-Knot Wind . . . . . . . . . . . . . . . . . 60
B.2 Comparison of API and NPD Spectrum for 50-Knot Wind . . . . . . . . . . . . . . . . . 62
B.3 Comparison of API and NPD Spectrum for 80-Knot Wind . . . . . . . . . . . . . . . . . 62
C.l. Semisubmersible Current Drag Coefficient for Members Having Flat Surfaces . 64
C.2 Wave Force Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
C.3 Wave Drift Force and Motion for Drillships Bow Sea . . . . . . . . . . . . . . . . . . . . . 68
C.4 Wave Drift Force and Motion for Drillships Bow Seas . . . . . . . . . . . . . . . . . . . . 69
C.5 Wave Drift Force and Motion for Drillships Bow Seas . . . . . . . . . . . . . . . . . . . . 70
C.6 Wave Drift Force and Motion for Drillships Quartering Seas, Surge . . . . . . . . . . 71
C.7 Wave Drift Force and Motion for Drillships Quartering Seas, Surge . . . . . . . . . . 72
C.8 Wave Drift Force and Motion for Drillships Quartering Sea, Surge . . . . . . . . . . 73
C.9 Wave Drift Force and Motion for Drillships Quartering Seas, Sway . . . . . . . . . . 74
C. 10 Wave Drift Force and Motion for Drillships Quartering Seas, Sway . . . . . . . . . . 75
C.1l Wave Drift Force and Motion for Drillships Quaitering Seas, Sway . . . . . . . . . . 76
C.12 Wave Drift Force and Motion for Drillships Beam Seas . . . . . . . . . . . . . . . . . . . 77
C.l3 Wave Drift Force and Motion for Drillships Beam Seas . . . . . . . . . . . . . . . . . . . 78
C. 14 Wave Drift Force and Motion for Drillships Beam Seas . . . . . . . . . . . . . . . . . . . 79
C. 1 5 Wave Drift Force and Motion for Semisubmersibles—Bow Seas . . . . . . . . . . . . 80
C. 16 Wave Drift Force and Motion for Semisubmersibles—Quartering Seas . . . . . . . 81
C.l7 Wave Drift Force and Motion for Semisubmersibles———Beam Seas . . . . . . . . . . . 82
D.l. Drag Embedment Anchors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
D.2 Anchor Holding Capacity in Soft Clay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
D.3 Anchor Holding Capacity in Sand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
D.4 Effect of Cyclic Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
D.5 Effect of Anchor Soaking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
vi
 Page
D.6 Percent Holding Capacity versus Drag Distance in Soft Clay . . . . . . . . . . . . . . . 90
E.l Three-Dimensional View of a Possible Failure Mechanism . . . . . . . . . . . . . . . . . 99
E.2 Example of Increase in Adhesion Factor with Time . . . . . . . . . . . . . . . . . . . . . . 101
E.3 Example of Failure Interaction Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
E4 Calculation of Required Safety Factor as a Function of Failure Mode . . . . . . . I04
E5 Schematic of Anchor Line Configuration During Embedment . . . . . . . . . . . . . 106
E.6 Anchor Trajectory a nod Fluke Orientation During Installation . . . . . . . . . . . . . . 106
E.7 Definition of Anchor Fluke Width, B, and Length, L . . . . . . . . . . . . . . . . . . . . . 108
E.8 Procedure to Measure Pile Out-of-Roundness . . . . . . . . . . . . . . . . . . . . . . . . . . 109
F.l Propeller Thrust, Open Propellers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I20
F.2 Propeller Thrust, Propeller with Nozzles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
F.3 Side Force, Tunnel Thrusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
H.l Eddies in the Downstream Wake of a Cylinder . . . . . . . . . . . . . . . . . . . . . . . . . . 129
H..2 Motion Trajectory of a Spar Experiencing VIM . . . . . . . . . . . . . . . . . . . . . . . . . .130
H.3 Example VIM Amplitude versus Reduced Velocity . . . . . . . . . . . . . . . . . . . . . . I32
1.1 Dynamic Offset in Terms of Total Offset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
11.2 Dynamic Tension for Most Loaded Mooring Line or Tendon . . . . . . . . . . . . . . 156
1.3 Surge Natural Period Increases with Water Depth-—Hurricane . . . . . . . . . . . . . 156
11.4 Surge Natural Period Increases with Water Depth~—Loop Current . . . . . . . . . . 157
I.5 FPSO Low Frequency Sway-Yaw Mode Natural Period . . . . . . . . . . . . . . . . . . I57
1.6 Low Frequency Surge Damping»--Hsurri.cane . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 8
1.7 Low Frequency Surge Damping Contributions in Hurricane . . . . . . . . . . . . . . . 159
1.8 Effect of Risers on Low Frequency Surge Damping . . . . . . . . . . . . . . . . . . . . . . 159
1.9 Effect of Neglecting Line Damping for SPAR . . . . . . . . . . . . . . . . . . . . . . . . . . 160
1.10 Effect of Risers on Current Load from Slender Bodies . . . . . . . . . . . . . . . . . . . I61
1.11 Effect of Current Drag Coefficient on Vessel Offset—Loop Current . . . . . . . . . 161
1.12 Effect of Current Drag Coefficient on Line Tension-—Loop Current . . . . . . . . .
1.13 Dynamic Offset Contributions in OM Hurricane . . . . . . . . . . . . . . . . . . . . . . . . 162
I1/|
Dynamic Offset Ccn.tributio.ns in GOM Loop Cunent . . . . . . . . . . . . . . . . . . . .
.l...1-I‘ 163
I.15 Dynamic Tension Contributions for the Most Loaded Line . . . . . . . . . . . . . . . . I63
I.l6 Mean Environmental Loads in GOM Hurricane . . . . . . . . . . . . . . . . . . . . . . . . . 164
I.l7 Mean Environmental. Loads in GOM Loop Current . . . . . . . . . . . . . . . . . . . . . . 164
I. 18 Difference in Wind Spectra Increases with Increasing Natural Period . . . . . . . . 165
I.l9 Sensitivity of Wind Spectrum on SD Offset from Wind-W»-Hurricane Condition 166
1.20 Sensitivity of Wind Spectrum on Total SD Otfset-~Hunicatne Condition . . . . . 166
I.2l Sensitivity of Wave Height for Hurricane Condition . . . . . . . . . . . . . . . . . . . . . 167
1.22 Sensitivity of Wave Peak Period for Hurricane Condition . . . . . . . . . . . . . . . . . 168
I.23 Comparison of Wave Frequency Tensions for Steel and Polyester Mooring . . . 168
I.24 Comparison of Line Damping for Steel and Polyester Mooring . . . . . . . . . . . . 169
J.l Mooring Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
J.2 Spectral Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
J.3 Selection of Input Motion . . . . . . . . . . . - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
J.4 Mooring System and Environmental Directions . . . . . . . . . . . . . . . . . . . . . . . . . 178

vii
 Design and Analysis of Stationkeeping Systems for Floating Structures

1 Scope semisubmersibles and spars. A third type of stationkeeping

The purpose of this document is to present a rational system is dynamic positioning (DP). Dynamic positioning
method for analyzing, designing or evaluating station-keep- can be used as the sole source of statioinkeeping or used to
ing systems ased for floating units. This method provides a assist a catenary mooring. Dynamic positioning can be used
uniform an.alysis tool which, when combined with. an under- with either ship shaped or semisubmersible vessels.
standing of the environment at a particular location, the char-
acteristics of the unit being moored, and other factors, can be 2.1.1 Spread Mooring
used to determine the adequacy and safety of the mooring
In a typical spread mooring system, groups of mooring
system. This document addresses station-keeping system
lines are terminated at the corners of the vessel, holding a
(mooring, dynamic positioning, or thrustenassisted mooring)
stable vessel heading. Figure l is an illustration of a catenary
design, analysis, and operation. Different design require-
ments for mobile and permanent moorings are provided. spread moored semisubmersible. Since the environmental
force on a semisubmersible or a spar is relatively insensitive
The design procedure specified in this document is based
to direction, a spread mooring system can be designed to
on a deterministic approach where the mooring system
hold. the vessel on location regardless of the direction of the
responses such as line tensions, vessel offsets, and anchor
loads are evaluated for a design enviromnent defined by a environment. However, this system can also be applied to
return period. The mooring system responses are then ship-shaped vessels, which are more sensitive to environ-
checked against the mooring strength, offset limit, and anchor mental directions. The mooring line can be chain, wire rope,
capacity to ensure a factor of safety against mooring breakage fiber rope, or a combination of the three. Either conventional
or excessive vessel excursion. It should be noted that mooring drag anchors or anchor piles can be used to terminate the
designs based on this approach may not have the same level mooring lines.
of reliability, as discussed in Appendix G. The combination of a spread mooring with vertical moor-
The technology of mooring floating units is growing rap- ing tendons to restrain a Tension Leg Platform (TLP) on loca~
idly. In those areas where data considered adequate were tics, as shown in Figure 2, has been ased to enhance both the
available, specific and detailed recommendations are given. operability and reliability of the basic TLP concept. The con-
.In other areas general statements are used to indicate that figuration and design of this spread mooring are similar to a
consideration should be given to those particular points. spread mooring system for semisubmersible based floating
Designers are encouraged to utilize all research advances production systems.
available to them. As offshore knowledge continues to grow, The Differentiated Compliance Anchoring System
this document will be revised. It is hoped that the general (DICAS) is a new mooring system that was initially devel-
statements contained herein will gradually be replaced by oped for FPSO operations off Brazil. The system is a spread
detailed recommendations. mooring system with different lateral mooring stiffness at the
This document does not address mooring inspection/mai.n- bow and stern of the FPSO, allowing the FPSO to partially
tenance requirements and synthetic fiber rope mooring. These weathervane. Because of this feature, production swivels and
issues are addressed. in the following API documents: turrets are not needed, leading to a reduction in the mooring
* API RP 21, Recommended Practice for fI’t-SEE"?/iC€' capital cost. However, the larger vessel offsets, particularly
Inspection oftl/[coring Hordwarefor Floating Drilling the yaw motion of the vessel, will require a more complex
Units (Reference l.). and costly riser system.
~ API RP 2SM., Recommended Practice for Design, An early example of DICAS is a mooring system consist-
Manufacturing, and Maintenance of Synthetic Fiber ing of l5 mooring lines, which are arranged in three groups of
Ropesfor Oflshore Mooring (Reference 2). five lines, with two groups located at the bow of the vessel
and one group at the stern (Figure 3). Another example of
2 Basic Considerations DICAS is a mooring design consisting of 18 mooring lines,
2.1 INTRODUCTION TO STATIONKEEPING which are arranged in four groups, with two groups of 6 lines
SYSTEMS located at the bow of the vessel and two groups of 3 lines at
the stern. The bow mooring groups provide the majority of
The stationkeeping system for a floating structure can be the restoring force. However, system stiffness also depends
either a single point mooring or a spread mooring. Single upon the lateral stiffness of the stern mooring groups. As the
point moorings tend to be used more frequently for ship stiffness of the stern mooring groups increase, the ability of
shaped vessels, while spread moorings are used mostly for the vessel. to weathervane decreases.
 AP! RECOMMENQEQ Pa/xcrrca ZSK

...i¢§u-:~.-' ' "'3

Winch or Windlass

__ ':_ ' ' —._

. ..:: .. '-'-;-‘-_._: " ' u , p::-
_. . . H
- .;: .;: ' '. ' -“*5,-:I"" ‘*1’ __; -
.2: : . _ .; - . . '1-.-,-.-.---.-.-.0-:1
at E‘ \' ' 9 .‘_- _ . - l?:1‘.<:I:i'=:3"7}4:=:=:= .
_ » .-4" ; , '-.~ I f \'- -:':E-. 1;-:-.1:-:<:;-<1:-:-:-: _ -
- 25‘ ‘ =':3:T§=§55:!=5II<?7.-$*£E=:Y:3'< \.- >- 5*:-:-.1-\:.=-5:3?-.
-me»-'' /.-»=-2-;-.-.»; ,:-.-.. /\'--.'
s-/-..;..--».-=..-.-
1542' -
.-{.01.’.
l
L
755551’/1$$55§‘\.'<.-:“»~»\»..
.T'{\\\-(2-I\v?. =EE$.». '=’-t§§==}<1§$E$§»‘i-*§'¢§:¥' -
.~»“\\.\\-.-.. :»:w:».»-.w~.-;:‘;-.-
~,~.~.\'.\'.w\.'4\.6-2.1.-.\\-.~\-.-.<.u»u.»w.-.\\>:\|a.-.(r,-._-
~»“w.-.\.\.,,._- I.-
.-.-.-.-»/-‘».-'.-.-\'.-/.n-¢.1 l w.w.-.-.w
, .~'-'*-r/M-re-r:---\*
-.>.-».- ?\'A-.‘|*.‘_1-VI . 1».-.4
‘-:-:-:+:-:-.- :-:-'<- \-.
\:-> l :>.-.-\.-.\.-.-.1.9».»»-.~»~=-.\...<.u»w.»v.-.w.~¢,-,w.
»/-“a
-».- w»
/4»-.4,-¢-.'¢».:.-.-..¢._ » . .». ...
~.-» .\- H l» ->'-‘-\~'-'-'-'->:v ->‘-\*-'-“V\'-\'*-§"'§*5>"5>-‘-A-FM-L!‘
.>/.“\.»“./4..~»w.»\-1-14.“.».»..»~.\-.a-/ 7
-e.\¢<\-».~.--.-A-,4.’ J-I»).-v ..
I gag-\'~;¢‘-yAr-'5.52¢¢-a.:.:,z.g~;.:.:¢.;.\:,;.;.1 -.-.11 ,.;.;.;.-.;.-.; 11\' l !>.\\w»»\»-..v.-.\\.-.\vA-
-.m»-.~“w.1.-.<.\w.-~u!w.w».- w.u.v.»-.'.r.'.u.'-\.=-n-.17-'3.
5I4-?\:Q2+‘-\-51/-I§?>>§>"'3-\!-I->"*-“ .~.- '-=.-.-.<--.*¢-
4*/<1-<\-'-w’/./',= -
ggg%gggefigg5=;;5=5;=;=,==;;;;;==;q.,,;=;;;1:;g;>-
k, .v _. _ .. .. 1 .1 _. . 1‘l :'5a:‘-:-:(c&-'-?<-2::-qt?-Kw?
1_.1,.1,.- -<1‘-:k‘\:"‘-5"‘-""
--.~,5;.c==~,<¢:_.\~¢:~>\# '-'-».~¢<‘—Q5:?/".<<',§

- - 415$?/'
'- $.'='<‘e<=.<:1'¢*,?:;=E‘.
'.‘I' ...’:€<;2z;<s.<;=:‘¢-:-.
"4'=.-s5;:,:E3:52=:=:2=:1;=22k=:=::2=:s:=:=:=:£=m-2am:
1 ‘
‘ I2‘! - --
-v
1 ">. .-
“““'-'-““"
\\nnl\'i\I\!H5?\ ‘ .-1.-41."",§'“"*:”
' -1-.2é1:%ws>=v<:»2.a::=Z%?4-‘
'~‘¢'-::;==1:£<.¢'
"
-
....,....,-.-._.,,.,,..,,......., - ",r:.<:=:-9'
. ,.,,_... ,.,..........,
"-.~'¢-,-.1»--wrw-,-.-.-M.-.-,-,-,-.-.1.val..»I.\ -.-1..-.v.<.1.1.'.1.-.- -
.-.1:-;-
.-:-\A\I
- .-.-:-4 \. ,/'41“ ~; v
‘Q/3:’
-:':.., -/-:-:1:1:1:-:1 aw
-:-:-:->¢.-:-:-:-:---as
. -:-:- \ -2*-:
‘I -:1:<:=I :"
1/.\:-:..--1%:-:-:-. 1- .'
'-=-I-24$:2=:=$s;1s:===:éé=:=;£:a;$$§-
"--:-:-:-:1:;m\..-:-- : £4»,--:-:-:-: .-1+.
;'<=;;z>€2===9 .-:1:1.<¢-..
';=.<;=,~> -_ -. - -- . - - ..;
=:=:1:1:1:421:%:t=:t
. :_:_;:W:_: ..?.¥' E€1:'<i:1:1 ll :1;=-25:’
:5¢....... ., -_._._/_
1'1‘! __ 2 - '
1.- - .».;.;.;.;.;. ;.;.;.;,¢.;.—.;.- ——~ .
' ' ' ' =:i:=?5ZFE.
- . .;¢, 1 :3;-:-11>:-: 5»:l. -:-:-.¥.~';.4:-4
, . . -"))’1-1-1\I\1
. .-».-.-<-.-.-. r .-:.-.- — -
I

Pendant bugy _ "e§§~'i 5Ii‘-:;‘i:. - ' ..: I’: - l

H
'1-9! °-ERR“ :55: .;_ H
" E552 _Ig£§E 1,4~-01-_-:I:I.
. ' --:75‘ 3:-:In-3:41>‘"I II..
_ . _. .. 1*:‘-l_*:*I
» I =5" ."f"j=‘1-""li§§-

Pendant line _..__,_ Mooring “fie

2

I Anchor

Figure 1--Spread Mooring

ll
F2
. . and
ornmg E7;
production an -
equipment _

Buoyant
superstructure

' :_': ;_ 2 "2

_ _
/ l
= M“ l l = r
\ lnllre .
submersrble
Ware rope and oharn l l l » l l buoys
lateral mooring I M l H ‘ A
line l M l
/ L l A Mooring \
3 A 2/ tendons
1: r l N ‘ 3
ll 1

‘:1 ‘ 1 “ ‘

r Package ofdrilring
* I A 1 tools being run
1 A; in A to weil
Production riser 9 \ 3 1‘ I A
from completed l L ‘
well 1 1l l kl PM ‘ l1 Seafloorfoundetion
. , r Al I A1 Drag
“ /Anchor l M M l amhor
__ pale l A _
______ . ._ “___ - -2 """"' “ . ____,;

Figure 2--TLP Lateral Mooring System

 Dessert AND ANALYSIS on STATIONKEEPING SYSTEMS FOR FLOATING STRUCTURES 3

.___..__.__ pI"'*"II\

wQ‘_-;-///'
-an Bin ‘W y
”,¢,'¢gw:--22:.‘/ll
‘id
\
I, \\

r §~
I 4 I \

4/
If I’, I p’ if /'
I T.
id ll £
\\ I
4! \
\.\ \
I
4

I \\
\
\
1"
r*'
\\.
‘
. s
r’
1
\
!‘\
I
\\
Ja-
Q

Q
¢-plI—

Figure 3—Differential Compliance Anchoring System (DICAS)

2.1.2 Single Point Mooring 2.1.2.2 CALM (Catenary Anchor Leg Mooring)
Single point moorings are used primarily for ship shaped The CALM system consists of a large buoy, which sup-
vessels. They allow the vessel to weather vane. This is neces- ports a number of catenary chain legs anchored to the sea
sary to minimize environmental loads on the ship shaped ves- floor (Figure 7). Riser systems or flow lines that emerge from
sel by heading into the prevailing weather. There is wide the sea floor are attached to the underside of the CALM buoy.
variety in the design of single poi.nt moorings, as discussed Some of the systems use a hawser, typically a synthetic rope,
below. between the vessel and the buoy. Since the response of the
CALM buoy is totally different than that of the vessel under
2.1.2.1 Turret Mooring the influence of waves, this system is limited in its ability to
withstand environmental conditions. When seastates attain a
A turret mooring system is defined as any mooring system
certain magnitude it is necessary to cast the vessel off.
where a number of catenary mooring legs are attached to a
In order to overcome this limitation, rigid structural yokes
turret, which includes bearings to allow the vessel to rotate
with articulations are used in some designs to tie the vessel to
around the anchor legs.
the top of the buoy. An example is shown in Figure 8. This
The turret can be mounted externally from the vessel bow
rigid articulation virtually eliminates horizontal motions
or stem with appropriate reinforcements (Figure 4) or inter-
between the buoy and the vessel. Another development,
nally within the vessel (Figure 5). The chain table can be
shown in Figure 9, is a buoyant yoke with a “soft” mooring
above or below the waterline. The turret also could be into»
connection using chains attached to the yoke.
grated into a vertical riser system which is attached to the
bow or stern of the vessel (or internally) through some kind
2.1.2.3 SALM (Single Anchor Leg Mooring)
of mechanism that allows articulation (gimbaled table, “U”
joint or chain connections). The base of the riser is often This system employs a vertical riser system that has a large
weighted through additional weight within the riser or sus» amount of buoyancy near the surface, and sometimes on the
pended beneath the riser (counterweight). These items affect surface, which is held by a pre-tensioned riser. The system
the performance of the mooring system. The configuration of typically employs a tubular, articulated. riser with a fixed yoke
the riser could include steel tubular, chain or wire rope com- (Figure 10). It is possible also to use a chain riser with soft
ponents and can vary considerably in diameter and length. mooring connections (Figure ll). The vertical buoyancy
The position of the chain table relative to the riser also can force acting on the top of the riser functions as an inverted
vary according to the design. Figure 6 shows some variations pendulum. When the system is displaced to the side, the pen-
in the turret design offered by the industry. dulum action tends to restore the riser to the vertical position.
 API Racomnarmaa Pnptcncs ZSK

Upper connection structure

=r
+‘~ __,,¥ _ __ __,,t— mg """1nn¢___ "W" fw"
pq~r—4wunuL WWW """"""“"*""':7’ ' "1"" ""‘*'* ' "" W
.- \0.»-\. . _.

maaamr _ '::""H'I ' \

\r§"%-'-‘-'=-" ' I
3,,‘

-tr»
-"E =‘*‘ S
. A .I .
-'= " !E."‘.'-
pnqmun4l I-‘Stat-t!JIr.|

A55; \
ir 1--‘M; 1
\‘ - L.*.t. r F7

Floating storage unit

twmt

“
(FSU)

/
/
,» /
¢
'\. '
@/;_0OQ9/6-
I
‘
Ft" 3
‘

r
n

r
\ . \
Vertrcal
Z } turret shaft
F t

13

r ‘_ i

l Lower connection structure V t V V

- —~~ - -- A - if - M A - ~ - —******* W, - -, »- M -7*; - Chain tab|e

r _ ‘._._:.mw
WF‘ .‘..__V_
‘Iran. 4; .$._*

ii II
-~@;'w r
<~ J)‘

:4 5| tr
Mooring chain (typ.) tr
I —r»- ~» ‘Q
Figure 4-»»~Typical External Turret Mooring Arrangement
 DESiC-SN AND Ai\tALYSiS 0? STATiONi<EEP¥NG~ Svseeres FOR Fzeemze Smocreees

ln~Iine swivel

Upper connection structure Toroiolal swivel

i ~ i*~~~~* a
. .- M or of r gr
‘IH.
;%--
or t ”
1-|_ i
3-,. -"
3

1'1"" *T W
....E?..
it M
_

/
___.;__'~_~—11~' —'“" 77
_
7
.1‘

/2’
f1-nil-F

= /Upper
connection
structure
oooooooooo L or
l l
_,___:ac o '\H=E.==!'<!'E=:
-5--t
_ _ _"'i_l1‘ _ _ n.
_______________ __
_ ___________‘________________1____________ _ ___________”_j__fi___ _E i | lkfi __________________ __

r
FF“.nanny' I 1 1 lllllllltollllllti _r
l
aa
"'
:,.u aw_ l I!
I

to /f
i i i wWmJ_WW o¢ rm“!-QMQAW l

E£Efl%§
l
l

il
i -
V".l
ll
ll

Floating storage unit

_ ;1|u lun¢w_+.nI.W {trot rt
U cu;

i Vertical
turret shaft

Tu fret well wall We

l<-- l
Z’ l
I
l

i r l
i I
i
g l I‘ i
i

Lower connection structure I

W W __ _

=:=fl|| //~o/— Chain table

'.

llrrlri
.. --- --~ -» 4» l

i»------- a a a a a a W r
Z»
l \
Mooring chain (typ.) i
t i /f
fii~@*
Z
__=’L-I

it

Figure 5---Typical Internal Turret Mooring Arrangement

6 API Rscorvrmenoeo PRACTICE ZSK

tit
= SWlV€| jOilT[
Swivel joint

I’
sorter bearing
_. l_ I"~11 , ....... - 1° Ftoller beans
5 ' *
""
.................. ,.
Mechanical 9 it =3-
; Riser I -1
..-v

' ' .. 5

W Cylindrical
IX shaft

@..»1 '\.-.
‘\-
1
2
Z a.
"M.
'\..
N5‘?-\..

=-
'
I II..1' ' /1”
. IIII _. ;e_~_-:'-1- _n@
.
ri
\
\
\'\_
\_\
Riser
" _
" 1
_.ji.-.¢-._ COUfii6'i'W9lgi'ti
\

I ‘
I
' fl
tII
K

""‘..\i;i
1 l _-.1-_-_.

r Riser -
1* unu-
Chain -' _
nun

i
unnu-

. II .- g. . . . .
Counterweight I. T
1 ;-er-—— Chain Ballast
_.__;._~_- in-.-.
._. .;

/ '\

/.
2,._/ | T"-_
.‘ Chain —// ‘
\

Chain Buoyant Riser

Counterweight Turret Ftiser System Counterweight
System System

Figure 6—\/arlations on Turret/Riser System

The tanker can be secured to the top of this SALM buoy 2.2 MOORING COMPONENTS
with either a flexible hawser or a rigid yoke as discussed in
the CALM description. The base of the riser is usually A mooring system consists of a number of components
attached through a U-joint to a piled or deadweight concrete such as chain, wire rope, synthetic rope, connecting hard-
or steel structure on the sea floor. In deep water, the riser sys- ware, clump weight, buoy, winch, fairlead, and anchor. A
tem usually has mid-span articulation. description of mooring components commonly used by the
offshore industry can be found in Appendix A.
2.1.3 Dynamic Positioning
2.3 PERMANENT AND MOBILE MOORING
Dynamic positioning (Figure 12) is a technique of auto~ SYSTEMS
matically maintaining the position of a floating vessel within
a specified tolerance by controlling onboard. thrusters which Permanent moorings are normally used for production
generate thmst vectors to counter the wind, wave and current operations with longer design lives. The mooring for a float-
forces. Dynamic positioning is particularly well suited for a ing production system (FPS), for example, is normally a per-
vessel designed to arrive and leave location frequently, such manent mooring since FPSs typically have design lives of
as an extended well test system. over 10 years. Mobile moorings often stay on one location for
a short period. Examples of mobile moorings include those
2.1.4 Thruster Assisted Mooring for mobile offshore drilling units (l\/lODUs) and for tenders
moored next to another platform such as floatels, drilling ten-
Many floating vessels designed to operate with moorings der, and service vessels. The division between mobile and
are also equipped with thlusters and thruster control systems. permanent moorings may not be clear for operations with
The thntsters can be used to control the vessel heading, design lives of a few years. In this case, the user should make
reduce mooring load under severe environment, or increase a judgment based on the risk of exposure to severe environ-
the workability of the floating vessel. ments and the consequence of a mooring failure. Differences
 Dasrom mo Amrvsrs o? S'?A"¥"%€>NKEEPiNG Svsrws FOR FLOATNG Smooroass

:~/-r f
F

,viv‘

<9,
f' an-

% '
0 fr? ‘
g ”
I. ... ' _ ’
'I'f:;5I'lI:§:I-Iiffiilli.-. ' '3"-'1-l-=-1:551:-§l;'E?ElI-.-23§l:i:I:T¥t5I-1'26?‘ 1
_._._.___._._”;\'.._.....55§:j;;:;:;.;;;:1I-l§I:l:?:3;3F3 -:l:§.;_\-I-I:-'-Q.-.._-.-:=:-I:<>-.;;f1_.__
.-.-.-'-"'-1--.-.-.\:-,,,,_‘. ___ __ _ _
I.-,;._._. ;.E;;:;.¢;;;;.;.-:-1-.-.3"-:{. _.;.7.7
._ -

I‘ . .. ..

- ' "
Hawser I
H
II
’ I . .

II

_.;:f; - :-'0.--;.:.:.;._ '“'*w’-'-""'1- 4 ,

1*‘. ._ ._ <7/.:-:-;-;.;\/_,;.;., .., " ""/»¢<
""-'-,:I:3F.i:1 3:' .‘.-;c-:.;:-;;:;:;'-'/-'- " '
'- - ‘$3515-: . T'-1.>:;-§3:':"i'?*><-:-re;c- .
-. .=;_._._._.;.-:-.;-2;:

_:;lI\n| Iun mu

J;
r\

pa“
\
I r '-..
1

i ; \ Mooring line

I \

I
-\

I
1

_I
\

..- “' . J,
-.1:l:- .. .-:I:I-:.-: ',:- , _ 1::-' -‘.»." ,1; I, ""':1
EP'1E'*'§= _.;E§>':-,,--:;. :i._'7.j2""j:§igi;:,;
1.
_I- _:{:';"-'-::;;;;;1:
r-.1.-=-5
-,,3':*»_1-:--:;
" ’1:~-

/
/ /
A.-
A/'

Figure 7—Catenary Anchor Leg Mooring (CALM) with Hawsers

 AF-‘i RECOMMENQE8 PRACTECE ZSK

r
in W
~Q3.>/ §\
iv
~
- H‘:
*-lr—~ ,,,_
h@ I <6“
‘I?
-_-7.5.
_ _,,_-| *-1"“
ff

' ---.
/
:3:3
‘Q
Y fr-:35;:-:¢!;:;i;:-:+;;I;i_: _p¢"’
'31-1-:-at :11? ' , .
5;-Ii i :§=::===‘=i=?=%555?ii§E ;:;;;:;E§1§§5; E_ "‘
:IE235;§§i§§§:§1:§:§;:;:;?;I;I
_
m¢;'.- - ;-

Q I
- "7'7"'3'{-'-ark-125* "/'-'.-.-;-:-:-:-> .. -- . .. ....
" J" " " ' " __ __ _- :-.-'-I-I-:2E::.;¢.;-.;.[.;'.1tI*l-:-:-:-
' '" ‘-,';‘;:-:-r-I-::-.-;--1:-;_;.;$;_:;;:_-I;;;;I4;;;;-:;:;.;;;-5;;
':'.3.?"'”-I 2/ -'_-.-;l§:'<l:-:-;-:-:¢:I:'I 3;:1 :-:-:-1;:l:I-l-'r-_-:- ' ""1"?-"2'-->~f'?-‘I77-’"'5'7""<"-'-I-I3?-'1 '3'-r-1-: :-._-11':-;<;:;:I:I.
. ‘ - .

-I-1-E:;;;::=ri:=:-55' .' . ':'-'§;';"':':l'=:i:=:l:-:1:-:1:5?: I}:

;:f=;.=.- -'-1,:.;._,:._, .:;=; . _._.__.__ E2;:;a;2;2;55:;2533;‘:;.;:;;;;;;€E§¢r252335553I=I5;}-i[§;f§.;§i§§ai- = - ~
j_fI.,_' ' :'/.{¢-1'.-:-:»=,,_._ -..;:;:;I;:;.. - ..
‘-'3.'. -"I:-:-' /Lr:-:2 ' : 1I;-:I:§;-:-.;:' ' "'f"'f7I:"' '
.:;__._,'.;< - . l_.5;:::_._._._. _
_.;,\_'<:\‘,.\I*.-I . ._.-:;.‘37,;;;:;',_:-'-'-'"-'5:g.
;;_-§;:;-..

I
-:-:1’-:-1;.-'1 . -_ .. - 5--"-'
-.!.;{q¢-;.§;, ,__._. _ -.-.;.;.;.;.;.
. \..,. Hr+4.;.;._._.;.;;;.;.;.;:;:
-:-:-:-:-:+:-.--:-:-;-:-:-'
=;;-;- ,. - . &£=E=;<;>:5:&=;- ':‘:::':'m:'LM": ' -=-=-=11->-=- §-§,§_:_§IEI3‘:""

\\ ,
Buoy / '1 .=7 ':§";?F§€?§I-.-_-

‘i
;:;:-:-1->:1.1;;;i-? i
1-

\
i

.
ii
ii
i

i \
i

’1 \

/ , \\X\g_//

/// ' \.
Mooring line

i K
1’

\.
A:

I
\

/ , I

<1 -€'-'»I:';._.-:'-§.\-I-

I'-§l>'lr
=:=r,»i‘ -.'5'-'F""
i'?j'I._£i2'i2l:-.- ‘N
- -5,-3.
;.;--- - -.;._¢ .491.-
_. .~;.__., . \‘
. --:;,
-:-:-4 '/'-' _I;!
I-,__. .v-"1;
.__,_>_: -by-:;::.'-;...
;.-;\- /=11,"-"-:;r.-"§:I"'f1. -. :-131357:-;
;'._ -.:.-;: . .:_:;-/ "Z;_.;;. .55.: " '21: J. "l:':3'?:3"
if 7l-:-’;-:-
' ' " ;I:!'!"_._.
"
_:':"5.
. -' .-'1:
/-,._-.

551'?‘-"-2
-.';-____;_\_- ».~_-f_\"§:
» .~_., (“Q _.1r_.l_ --1-.--.=z-<-,:=;:. "-3
.§i§:§:' .5;_._.:$5-‘IEj"1_‘j.1E;:;:' --5; -,= I-. +21 -;-5
¢ '-'- '=r2=a5§I -Z?"fl-:;.':i?*§
' "$1-;_, _: *1"
/I _
-I __\';'-fw_._,_
""*"1""" ; -: _-_;.;:;:_ J
I --I: . - - ,1 -:-,;.,-. -:-,-:; _'/‘y' ' '
"===._.: -2 .3 -,,- . ’.|_.{_ .-:-: 5;: |_7.|_-:_
._._¢- :;

. "D .. ::-:--\a'-§:-|{;l’:£‘i<-’:::--
- _
C . .. " -“<1'¢~ -4"-,9» "'-":1-.1: ;-.j;- ;I;2;-"-::=:-

//
-' .:;" ‘»;:.:"' +>_;.--- :~',;._.-7 :-:;:-;-
' ...-5'15"
. _:i >513? -.-1.:-.I"' 1-5555"’-:iQ.~
-..-‘,,;,-.-1- . .i::F.i3.-. ""5. '_.;:f:-
f/ '-FF"'-I;I-:-I-:_-:-'-:-' -;-: .
~.=---fie‘E-I555"Iiizi i;I3=§§:-.'§§=-z--Ii'?E:I 5"=Y:_-i5“???-1:225;"=iii:H‘=‘1=2=5§55$i§i5‘; I551"-:"""§:;E;r*=--=:==-':I2
-:1. .-1-==§:=--.':=:--:. ;:"-=-1::;=
:_I '-:§;;:;_;;_____.<;:_§:::i;-:__:fl
'-- "
:-' A ';:-L--I-_ -:-;;;z::I'-':;:;;_:_;:[;3E;.;:§:I_ ".;§i"._.;,:I 555:: -._,_::;.;. I--3:}-
"
>'='== -‘ 1:1
=;;j=i%"';:;:;a;§§.=§;.j I \\
.-1--:=' -;=-2--l‘1%;"-33;:;==i:-.-. ,_==;:=.-3-2-1 -=-11>"
" .|
""'-- "'--_.____
-1931-. ' .. . -.. _-..
.;-.'
.I '-'..-“+1 1:-'=' .';?- .--
.; -V-.-.E_ £___.;;;:;.,.,_\_ '-

/»'

Figure 8—-Catenary Anchor Leg Mooring (CALM) with Fixed Yoke

 t

DES¥GN AND Ai\iAi_‘~r’SiS or-" $TATiONKEEPi¥\IG SYSTEMS FOR Fi.OAi"i¥\iG Smuororzes 9

Chain
Buoy '. I

1
P

»*.or5%
W‘Q)1-R‘ ¢1 1-
Q. 19$£3‘i;1'¢;§:¢:‘.!. I“

Mooring line Heavy Weight

Yoke

-:'-‘r 1'-3*‘./.-" .- "r

- -,. '-; -: .;:;:;'.---'-;;._._<-'.-:---4 _'. ." ;:;-: .. ..;.;!_.;:;I-.
:--; .. " '
:-;1r-'
.'»_ '-:;I;?~'-' -lg.-.;:;I-1;:-1;:-' ;.;1 +:' _ _._.;;;r_ -';:;. ' :;,;;: _:;;;:_.;___ ':§:;; __ ' _.;: _I-'-:-.;.;;-I;';i-3.» _.;:-_‘-,;. _ __,.. -. '55 __Eg§-§;._:;;; _:-I:;_.;.Ia’-__'_;_.,:;§-L’ ;:;
6 9 . - ;:.:i:5:=- .;;r'_.;;;.- 1-" .| =--'¢- ,',-,_.+.._._.,;';E 35115-g-‘=:; .Ii;EIE:§.:=E. ::-: =5;-'-"f,.;;;;§. '
:5":1ii ' ”" -i=1. ‘ii?-‘. =-;=.-.=?z-"=’3-1555}?-1&5;-.-r":5E-"-‘i ’--L-=1;-="i-.-..-&?ii -.l5§' ""1l5T.."‘ ‘IiI'I. "" ""5"'. ::="r&:=‘E 1;-'5"-'3---;
I. "»':':1’5--"=’='.-.""1=-:=;-‘Er-“ =5=5-‘-"=15;-. -1-.--.E§E:;;-r-:l:- ":51". "'?";“"~"""-' ' _ '-,:!:i :Et?'- ':= I:-:-.-:;'lr'.'-:1 »:f:fiF--Ii-.: .i:1: "';?=: ..-:1‘ "ii-';
,i;E;_’;?_."' -:§:=1=-":-=:.:.:=2;;-: -;z:=r- _.:=:*.;=;.; ~' :=:Y-._-.=.-- _._:;-= -;.-'-=.=:.=-..'=...‘:; ;=§=; Q:-;?I'. 1===“’ "1-. ====:; z:'-==z=z~' .51.. -;.; -.;.;.'=-'-'-r=' -£;=. =-1-‘=====.-=-.<':="' --IE:-L..\'
- sA.
-'E>'-_." -.-.::.<ii:-:-.-. .i:):' »,'i§-:- .“’-
;.;;.A.-I/.5 _.g.;. ._ \_ -
"'1;.;._
1 "7"-'37I":13"
_
1:1‘:-:1:="
-:-:- -:-
l:>.
"=*.=="5;'==:i==-* _<="-:,~.=-I -=2-' .-.==- .- -.---2:;-1 -=:-.=;=;'-.-- .-;:z: :l-az=.-.--==1:-="-1:.'-.-;:.=.==":=. ._.__=;=:=. ,=;=. -'g=__';-;:-- .;.;z: __:_:_ '-=:=;-':'=r:-=-' _ =:-; ,1. ;..'-1;:=--,.,=;

Figure 9—Catenary Anchor Leg Mooring (CALM) with Soft Yoke

Storage Tanker

\ 1"
41"‘

iA i
~ """'=1i:ia?2Iz1:I:'s'z I. . . E -=1‘
Swivel -., .
;-:>.~:-:-: _H_|+5_-l-I-I-L3;;5:555];=:;:5:g:;:;:-15:5.-5--gggyvffll_
,1.I.l.'.fi';'Il=i!IiI235I-Ii-3:-:::=:--.~<.-.-2?”. . .
._._-c¢;:;:_;;Q.,:_:;';';';' '_ ;.
_;--I;-1-;1;;;g;;;_;5§5;;;
-: :e:zE2E1i;
__=-:-:-:-:-:-'- -:-:-.-:-;-;-
;&;5_:Z_<Z;i;€§?;;§;*i=5;E;§f§5;"g5§§;Z=5;E:;: ''''' ' - -:- =- -=-=->»..
-4:1-;-.=ll;Z?E5%?&?E5E5Zr&"E"E - -,;:§'.7'7'7'T‘_5'3'

:1:-;-. .:-:.-...__._:...-_.L:.:::::.;-;-,-5"""'“>?;;F£9J:'::>;;r'
'-'-'-:;:;:;:;:;.__-.4:;;;;-:¢;-:--:-:-:- ' ' ""’--

-
-;-;'.-;:=;< rlr?-<""'"-'1'1'1'-*1‘ "'Y""'E;E§EgE;E;E;5f 57:-r-:-:-r-@T:';-;-:-!E3i- . . . . .9I:iI-:
. .'.§II'IfI."':I-;> --:;=-:----'-=-===:'-’.I
E5i;\3$'?3;“*">z4;.;.,;.;.;._.;- -I:-1:.-I5?gE;;;;,:;:;;§ --:-:-: - ;:.:-

"i'r
f-1:1:
. -.-..» /
-,1'.’/.:i--- . c.:\:-.3.3.lJ...-;:-...
.-.;.Z¢¢.;.-.--.;-*;:5;.;.-.-.-.;.;:;.-:-_-:-_ ', L_ .;. --:=:-::.'->1;
'-'-=-=:=:=E=E!I+ "<1-,. / ".;;;.;._ _,-/
/
::;:1:-:§£IEi:!%i;2f=f5=:2},, -.;-A-3.1371-'
| . w | . . _:-:{-:-:5;i;-:-;-:-;-;-- -:l--

"P355"-2-Y5‘ "5':3::'E3:-1-:-:-:-:-:-:-:=:i§-:"""
/ -:-;5%E1115155:7/'13‘-':5"3I-I-/:/R::'?Z:?A§11$j'I-FIEFI.-_-.‘-'-;!;IIl'.-I-I-:55;
':=§§'.i!-if:-:-.<>$E?d‘ . . . "/I:?3i§""""""::' i"
- .;.;.;.5;:;:;:;;;.;.;.;.;:;:2-'-. .;;:I:=:-;1.-.-;-.-. -:-;-:I:=.-.-.--;-;I5EI:I;.;-,-,- ""='==-:-=r.-,'"'9'
- '5'/5:;.;.H215;1;;3»----;53_:4:;:;.;;;§:;§§--
/‘

==="- =-=+-- ';:' . =-=-;-:-;. I. <=.a-=-.:.:,:."' ., """':.:..'-;-:.: M5” _:_ 1:-:;:;:i-11>:-:;;;.;:I;

-' / I:I;I.1:I:"='=-'=-=====-=.=...w ~, / />;_;:;:3::¢::j-':::="- 15:-g;§.§._.I.Ijfj'I'f j;._._,.:; §.1' >».r»»'>>¢==:=-==-=a:a; <:;.= ' "-'-=-14'::i=i:..:.;.;.::i§;- "
-1'1-I-$37‘-I-1'11-L-""I " -'-'-'-:-1-'-'-1-!;I;i .-g:-
. . ,5 "
II1-I:::::;;;:jL:Iji:I:I;E;;, ;}-;:;:;:;:;:;;--. ----.-.----1-:;:;:;:;:;;;:;>q:§;§j;:-'-- - -- _ .-.1;.;¢¢;;'.;:;:_:i_:_;_:_:::;§;':;:_§;:_:_:_:_:_:;5:':______;_:;..
. :1-i:'rI:{:E:5'=" *'i‘I:IE=:=:=r=:r-I
/;I-£1:-!--I-I;§;:;:;)-..

'§1§E21F?i-25'5"
:1-1-.-2:‘ 1:".3:-2¢.;.;.;.;._._
-=-:1:-1I:?r;:5:§.E-.:.; ' ""'§':I3:§.-“'_ "I
...i_ _I ';_-.5'3'3'5*'5'3'5:'r§§1}:-:\:;I-;;.;.-.\;
1:2. . _\ .;.... .. -I'-'-ii?!I5135$aw;>232‘"Em-1::-»i:§:E;.m*?7???IY...Il' -.-:E‘E'T:E:E:Ezi1
'- 1-:-5'--’-H91»:-iiH._:_,__,3@w';"-"-F:;;_;_I_1_'1';‘-';{:_'_'5'5';'_:51.‘.h'-"

:"f{¢I;§E§
.;.g;;.;-.s;:;:;§.__:
§§§§EE§E:
_:_
i{E§:§:§ 51:.
. - 2
:'H':':':':5"’
.;.; _;,,__
' _ =:-:;.-.;§
'-:-:-:-:-:-:-1-:-:-:;;;._\._._-;.;.-:-1&-
'{fE§w;q;;.g,’;“*'7-Qfiu
- .-,,;.,.,..-W-1-'1 '
E I
..

i i? --"' .:
"Ii.'§_§'}'7'5'7' ' ""l'.l:->:-:-13'*"c"""""" " . _
=-1+: "'1 "7.-.3"I".-one-1==E=%2:z:z:z:1?"="-""""=-é
' .'3:- __ 713:-:;.-_.;.;.-_
-.:.'.‘.;'I'

Buoy section

Riser section . _ l _
Ui'i|V€‘l'SE:l|_|Oll”!t
\'..
31:?‘ .;;§:';.c;:- .- !-:;- ._

i-‘E
"'-I-I:;.j.
.1-1-.-.'=ei'-=-."tI;.-.r;- "-‘£5552 ’5‘;I’- >'I~'i&- -'--=1‘-='='" ’=w<--.-?§‘'- i?="_ 1.
I I. _ .;; '-111;: _’ -'»I:_c;:!-.!:'
._.L_\ .-_.: ;l'i:=:'» ...;;:-' V. -. . 1 .;...:::.. j»/..,=‘
-- .

A Base

> -
<;. .-
-- -' ;-= -: '-1%.-;-. '-:= 1- .:-L-:-;.--gi;-. -.1
-_.1I;I;;.;.

Figure 10—Sing|e Anchor Leg Mooring (SALM) with Tubular Riser and Yoke
10 AFN Recorumemoeo §3RACTiCE 28K

--._____

ii?
, 1-""§'
i iI L. A
,,n ‘

\\.
I‘
is V. \4"’
96%; "
' ' H
HBWSQF 7 - '
. . .@. @. . -" @ - - .ZJII§. i"' "fiiic.
1:-=1" ..;.;.e>: 1;3;255:;-=-=-=="'-:..,,_.,._.,._.,. -:-=...,..-;=;.5=;...,.,.;;;§€fj:;i;:.;=:;Ij'f"""" ;....;§:;:;i1I
_ -:-:-:;:-:-;-. -- f:§"':::"'
- H__IWI:I:lI-I H _ :';3‘3:1;1§:§:§;§
=-.-.;.{_:._ .-.1-5-1%-9;-H.-355.;__:.E.:;_.;.;.;.i§;g._:__,:-;;.;_7_._. _______J:;.¢,_...;..,_ . .-.-,-;.E=;:_.;:._.>_-_-;,-;;:.

-
; .F -' . . .1 -:-:-:-:-r-r-'1*F1------;---1:r'.-av??? . . .-.-:-:':':1;211:1:'l':1>:-:-3:533;
".1. ""%3E§?'I1.'I 5""'¥:‘IE;i§l§5:§i=§w1-=-r-:5-1‘ -'-'-'-'-1.""""'"““""""’,-,3‘3,‘§i.~""“..i.lf..-..?if""'§E§‘5?5i§E1*”'--'I5=3=‘='-'"'"'
-;- - :- *5; . . '-~:--:;-:-:-:-:::-,.;-.-.-.i:1:'--:-:-.1.-:1'1'IrI:I:1:l;-:7-I-
31‘ ' ' --1:15--: 11'1"» - Y‘"I'5'51-:-:1:-:€1:1'1r§:3!E1E:§:jf:§:j:§i>}:§;-- :§?_‘f'f‘_"' '
= »

\
/

a 1
P" . ;:§:_<

_ _ _ _ _ _ __ '
..;.:-:-:"- ':-:-;-:-;-;--,5-__-,- . _ . . . -- -- .-'11‘ .’,;:;I9I;2 i;:;:-1-:; ._
;§:;;;:;§ .-...|.|.,._.j:f:;.=';?"I155"; _ __ ._.,.-;v.I;I¥"' ‘ -=::z;-..-._§.-..WY-';§if:jTij;f&;z;z;{ !z§:§:'"§.§;I:""' I
...:: ......::::,¢:':';:':
.. " ;:,:':,/j/'.
1.-' ]'- .. -. ;.

.,;.;.;:;:< _;.;.;;;.;:;.;:§::.- .-:-:-:-:-:-,,-1If-§Ej'1-" :1:1:-:1::-;-:-E1 '-'1r1:3:?5J;fi;§:;1§$j§E:1"‘1‘1:1-1-

-2-: '-:-.¢.;.;.;.;.;-:-';'-' .;:;"' " .;,,JZ__»_.;._:_:_:

Buoy ,
1

Chain riser
"M-

' I

' 1

l ‘M . __ H

;_-. o._ _ . ._ \

.>*'.'~'-= -.».-1---==5.::.1::= - . .- . -.

IQ -.
+'. .5 -.. . ., _,,- .. -'-,...,. ..,_.; _ . . - .1“ ._ ..';.1
7" '11- .:.; 3:-3 *3‘ . ;i:-I"'-:-.-. ‘T-1 '1- ‘:‘- ' .--"' ' .. " :3:>. -25.1-
===..=-??'==;ii="_.. --1:--.,.§.§".§;‘§ 1-;;;;1?.._E___;;.-=:...:.§- ._j-,_¢:<.~"31=;._._;,;-
- ‘j_]f-.1:.-;=-::":=-5==--_
.;:,...--§;;.1--§_ _ ' "-1 '-::w3'<1' - 1%: .;.1._ -\.- .. -_:;.~:;. _ _ ____ 11:55,.-.;:_.__.
'- .-:1 "QT ' ' . - ' --.' 11%..'II~?' .15‘ “Iv, l'~"' .5515.-5; “';E.i
E15‘. -;-.-.-.‘I .. ' ' ""-'1:=.‘1' -:1?15-"“'<+ --:F’:""»,-- - .. - - ...-:i"' . -.\.,
"'.’-:1: -351'-" '-‘F1:-.' '.
'-,;._-.-j-';:-:_C\
' ' '..-13.“~':
..';;f '-:1, - _
'1' ::.--
:2--1; ' - ' .__ ";:;:
.- . ;.'1.;-'
T:1 ;.-2.
1- .-:-:~i.I:I"-
3;
dz? ..
-I;.;.;¥¢-;;>:-
:::::_l:lI .3 :_ :5 _ > . _
.. .-.;‘§i'-;'§1§1\ __:::..;:;. _ .-:-: _.:1!;;. ... (1.
:;"",.;:§:_c1'<§.;.. :1: ' .-.£->'=:i1'1:..-..-EIE5E-:-- " '. ‘ "- . -' Y1-';',,
1' --1'-‘F1-" .-;:i:E=-'-‘Z- ""=
5151- '§.E: .\{.' 1' .~5:..;.. I'['E"-i 12:. . -- E.<:-‘F’?- " -;1:IT"I"I'II;'-.:I?'* '-51:1
' "-‘F, 1-. *+. ..3"11' ':1;1:1.-;-: .-:-:-.-.1-1 '1 -'1 .-.-:?:- . ' - '§:1::!£:!:‘€Z"
-1- - ' ' 5?]? “"5:-. --
"5-:"'"f:,'-=..1;-'-1-...-Q‘
1@==”==»,Ji§/i. z=..-.-:1:'
-='§.z§.>.=.1I‘_. .'--'<.\:.-
I:'"=-".1; j;2;:;:...'"‘ .
-1:-‘IE1I15-'.'.I.-;1E.-;=-1
- ;-;I;.. "1E1E1..’j-I
1 =,‘=ij»"'-#1:"
-:;.~;,; _:_"*1.-".-'
., ,f1_, 1' " - '
=-1:='=1,-..=ig£=§.==§
"
:'5.~ .'>' "1.-:5: '.3.3.-.. -1'1 ':1:' 1:;:':' :-.
' L .-?"':1:I:1...: ,:- '15?" '::':5.-..§ "1-- -
- - -‘A-' .;;._-,I:}'§.\:?.~':-
. .
k ‘P3?-...-.;.;.'<-.':;: ‘:15.-_-.~: .,.-:;.;.I:§ -;-'.j,_.;;,._"-'-1;;> :r:::;-'-1‘ _:5 1~-:1:'-'-"1/A i:I:». ‘ ' " .(\-:1; . '-
-:;:;-
.-.-_;.-,;.-. --, '1-‘r_:_- --:;;-.-t.{;<;»
_j._.1.____L._-i .¢- ..;:-..;;;;--:_.-,,_;:-:;;.;.
_ ._-:;:;'-..\,c,*' -' -.-.\ ,.r_I __ :.';1- .35;-,:1.
.,1\;_.,-4‘;-;-:;_._.1. _,_ .-.;;;-- ..
" ' ‘W ' H H .-!I;;€.~:"1. ' 1"}\.' '-1;. '-:1:-I‘:--. -r.-:-:-: “-1 -, '
( -.
'.,-1.. _._.-,'- ;:. :-:15?1-.:.::§_-_.__::E. -:-:\...1.-._q
\/ :;_ ‘-‘-:;';. _:;;- :_-;-:1-.__.;__ - 1.15; ._
>3 1~:I1':3;-'>. : .:-..3I=?1I'"=".€i;..i-E.22-;‘::.'" 11 ~

Figure ’|‘|---Single Anchor Leg Mooring (SALM) with Chain Riser and Hawser
 DESiGN arse Anarvsis or Srariorzxssezse Svsrsivis roe Ft.OATi¥\.i{3 Srneeroaas 11

/ X

/" \

//'/ ..;.. An -
U ' \ as
lnII|nn|I‘__-h‘

\ _~ ii?) "'"' W
,, . . . ......
1'3) ‘Hr
"' Hr" _ qu‘

JO
If‘

\
IL
i

/
/./
if/'
""'IIn|_,,

nl -0 - —-»~ - — —

a» ..:3)»
-I '-I24 ‘i

!I!.I.I._-_:-.i‘-'-i-
ri

Figure 12--Dynamic Positioning

between permanent and mobile moorings are significant, as the arrival of a storm. By contrast, an FPS is unlikely to be
discussed below. The discussion can be used as guidelines to removable from location, and may not even have quickly
determine to what category (permanent or mobile) the float- retrievable risers.
ing structure belongs.
2.3.3 Method of Analysis
2.3.1 Type of Moorings
A quasi-static analysis method is often used for evaluat-
A mobile vessel is normally equipped with a spread moor» ing the performance of a mobile mooring system, and the
ing, internal turret mooring, or dynamic positioning system. effects of line dynamics are accommodated through the use
However, a permanent vessel has more choices of mooring of a relatively conservative safety factor. A more rigorous
design because mobility is normally not required. dynamic analysis is required for the final design of a perma-
nent mooring system, and the factor of safety is relaxed to
2.3.2 Environmental Criteria reflect that some uncertainty in line tension prediction is
The design environments for mobile moorings are lower removed. Dynamic analysis can also be performed for
than those for permanent moorings. The lower design. envi- mobile moorings.
ronment for mobile moorings is based on the consideration A fatigue analysis is not required for mobile mooring sys-
that the consequence of a mooring failure would generally be tems. Because of abuse from frequent deployment and
less severe. This can be illustrated by comparing a Ix/IODU retrieval, many mooring components of a mobile mooring
with an FPS. In many instances, a MODU can at least discon- system are replaced before they reach their fatigue limits.
nect and may even lay down its drilling riser. In the case of However, for permanent installation, fatigue is an important
tropical storms, it may be possible to move the vessel before design factor, and a fatigue analysis should be performed.
12 API Rscorviusnoso PRACTiCE ZSK

2.3.4 Mooring Hardware Design. guidelines for riser can be found in API RP l7A/
ISO 13628-1, .Recon/mended Practtcefor Design and Opera-
Mobile moorings use the mooring hardware that can be
tion of Subsea Production Systems (Reference 3), API RP
rapidly deployed and retrieved. This limitation does not apply
17B, Recommended Practiccfor Flexible Pipe (Reference 4),
to permanent moorings. Many mooring components such as
API RP l 6Q, Recommended .Practice for Design, Selection,
anchor pile, linear winch, buoy, and chain jack that may not
Operation, and Maintenance of Marine Drilling Riser Sys-
be suitable for mobile moorings can be used in a permanent
tems (Reference 5), and API RP 2RD, Design ofRz'sersf0r
mooring. Also permanent moorings often require heavier
Floating Production (FPSs) and Tension Leg Platforms
mooring hardware because of the more stringent design
(TLPs) (Reference 6).
requirements.
2.4.3 Subsea Equipment Considerations
2.3.5 Installation
Mooring lines shoul.d be clear of subsea equipment such
The deployment of a mobile mooring is normally carried
as templates, riser base, satellite wells, and flowlines. Any
out with the assistance of work boats. This operation is sim- contact between mooring lines and subsea equipment dur-
ple and u.sually takes no more than a few days. The deploy- ing installation, operation or maintenance presents a high
ment of an FPS mooring often requires the assistance of potential of damage to both the equipment and the mooring
much heavier vessels such as a derrick barge or a purposely lines. If interference, or the potential for interference
built work boat. A portion of the mooring is often preset. appears unavoidable, it may be possible to alter the layout
Sometimes special design features are incorporated. in the and design of the mooring system through the use of an
mooring design to facilitate deployment. asymmetric arrangement of mooring lines, or the use of
clump weights or spring buoys. Coordination of the moor-
2.3.6 Inspection and Maintenance ing system design with the subsea equipment layout is
A mobile mooring can often be visually inspected during essential.
retrieval or deployment. Retrieving a permanent mooring for Guidelines for the design of subsea equipment are given in
inspection can be very expensive. To inspect a permanent API. RP 17A/ISO 13628-1, Recommended Practice for
mooring, divers or ROVs (Remotely Operated Vehicles) are Design and Operation ofSubsea Production Systems (Refer-
often used. Also, replacing faulty mooring components is ence J).
easier for mobile moorings than for permanent moorings.
2.5 FORESEEABLE HAZARDS FOR MOORING
2.4 DESIGN CONSIDERATIONS SYSTEMS.
The main objective of this section is to outline the causal
2.4.1 Primary Design Considerations factors and circumstances surrounding a number of individ-
The primary design considerations associated with a moor- ual and specific hazards relating to potential mooring system
ing system are design criteria, design loads, design life, oper- failures. It is important for mooring system designers and
ation and maintenance considerations, etc. These operators to recognize and minimize these hazards.
considerations are addressed in detail in the following sec-
tions. ln addition, a designer must also pay attention to the 2.5.1 Failure of Emergency Mooring Release
riser and subsea equipment considerations. Mechanism, or Its Premature or Delayed
Release
2.4.2 Riser Considerations In cases where an emergency m.ooring release system is
adopted, hazards arise if the systems do not activate on
Risers transfer fluids between the seabed and the produc-
demand, or can be activated when not required to do so. Rig-
tion or drilling vessel, and constitute one of the primary
orous procedures are therefore essential to ensure correct
design constraints of the mooring system. The riser system
operation. Planned maintenance and testing of such systems
often places limitations on the allowable vessel offset. In the
is also essential providing mooring integrity is not compro-
event of excessive vessel offsets, mooring line adjustments
mised in so doing.
such as slackening the leeward lines are sometimes per-
formed to avoid damage to the riser. An equally important
2.5.2 Failure of Structures supporting Anchoring
consideration is interference between mooring lines and ris-
Equipment, Fairleads and Winches
ers, during both operational and extreme weather conditions.
The mooring system and riser system must therefore be Hazards to the structure are caused by poor design and
designed to accommodate each other, and coordination of construction, and in service demand resulting in equipment
these two design efforts is essential. failure from load, vibration, corrosion and wear. In extreme
 Dssiou nun aaarvsis or Srariouxseeiuo Svsrsrvis eon Ftoaravo Sraucruess 13

cases, watertight integrity can be affected. Thus, a rigorous 2.5.6 In-service Degradation of Mooring
Planned Maintenance Schedule should include regular Components due to Corrosion, Metal
inspections of all mooring component foundations and inter- Fatigue, Abrasion, Deployment and Retrieval
nal supporting structure. l\/laintcnance should also cover the
Corrosion, wear, and damage can degrade mooring line
correct operation of all moveable parts and protection from
integrity. Corrosion is of major concern especially in the
corrosion.
splash zone for chain and wire systems. Individual wire
breaks are frequently attributed to corrosion. If a high corro-
2.5.3 Manufacturing Defects and Processes sion potential exists between components of a mooring sys-
tem then hydrogen levels can be sufficient to cause
Wire rope and chain, common links, connectors, chasing
embrittlement in any high-strength materials present.
equipment, pennants etc. are all subject to imperfections in
the manufacturing process. Such equipment must be accepted Abrasion or wear in wire ropes generally occurs at the
winch, fairlead and touchdown point. Normal operations will
on the basis of quality control procedures, testing and
ensure shifting of the contact areas. Abrasion in chain will be
approval, and constructed to recognized industry standards.
concentrated at lin.k contact points and any unforeseen con-
tact with vessel structure.
2.5.4 Mechanical, Electrical, and Hydraulic Damage during deployment and retrieval is common
Failures relating to Mooring Systems unless careful procedures are followed. For example, wire
Hazards from these supply systems can lead to the failure ropes are particularly subject to crushing damage on winches
of control, reference or sensors, load response, and loss of from high tension spooling. The spiral structure of wire rope
manual over»-ride. Mechanical. failures can occur to Windlass can cause a torque build-up if dragged along the seabed
brakes, clutches and sleeves, pawls, gears, splines, cracked resulting in a hock or loop should the rope tension reduce.
Also, the improper use of chasers during retrieval of wire
discs or drums, cooling, striker bar, wire spooling and ten-
ropes may lead to damage to the rope and connectors.
sioning devices, and wear to chain windlass/fairlead pockets.
A regular inspection program is essential to monitor the
Hydraulic failures can occur to pipework, seals, joints,
integrity of the moorings.
pumps, brake valves, greasing systems, oil contamination,
leaks and levels, or emergency release valves. Electrical. fail-
ures generally affect control and monitoring of position, ten-
2.5.7 Inadequate Operating, Maintenance and
Handling Procedures
sion and power systems. Regular inspection and planned
maintenance procedures should be set in place to minimize Operating procedures are produced to minimize the risk of
the likelihood of failure of these systems. mooring failure and injury to personnel and environment.
However, they do not always adequately address all potential
2.5.5 Mooring System Overload, Fatigue, and hazards which may exist during handling, deployment and
insufficient Anchor Holding Capacity retrieval. Also, inspection/discard criteria and maintenance
requirements are critical elements necessary to assure reli-
Overload is defined as any tension. which exceeds a pre- ability of any mooring system. Permanent moorings are less
defined lirnit or which exceeds the capacity of the anchor. likely to suffer from handling damage as they generally
Either event may cause a loss of position. Hazards include remain in place for longer periods. Handling procedures
inadequate use of propulsion, exceptional environmental con- should address normal handling operations with the focus of
ditions, inappropriate anchor penetration, excessive tension- priorities ensuring the safety of personnel, minimizing ser-
ing, equipment failure, poor installation and retrieval vice wear, scuffing, abrasion, chaser damage, running opera»-
operations. tions, winching wire onto drums or chain into a locker.
The adequacy of the mooring system to resist cyclic load- The condition of all mooring lines will deteriorate with
ing (fatigue) will be addressed in the mooring analysis. Prac- time in service and adequate inspection and maintenance pro-
tical measures can be adopted to move the sections of grams will be necessary to ensure continuous integrity.
mooring subject to concentrated fatigue e.g., at the fairleads
or at the touch-down point. 2.5.7.1 Stored or Trapped Torque Energy in Wire!
Anchor holding failure can be attributed to overloading Chain Catenary Systems During Retrieval
the soil foundation, inappropriate equipment or anchor Stored or trapped torque in mooring chain during retrieval
design for the soil conditions (e.g., fluke angle), equipment can pose a significant safety hazard to personnel. The chain
failure (including shackles), inadequate line lengths for torque energy is introduced into the system as a result of 6 or
extreme conditions. The consequence of these is likely to be 8 strand wire rope dynamic untwist/twist characteristics
anchor drag. under increased and decreased tensions. If a marine swivel is
14 API Rscoiaiaauoao Paacrica 28K

not present either at the top or bottom of the chain, torque 3 Environmental Criteria
may become trapped during recovery since the chain chaser
position at the anchor shaft prevents the release of torque. 3.1 ENVIRONMENTAL CONDITION
This condition could occur whenever a mooring line does not
incorporate an operable marine swivel. The industry recognizes two classifications £)f61'.i‘v'i1‘O}1ZZ}€i”i-
tal condition when evaluating mooring system strength: max-
Detailed procedures, which thoroughly address best prac-
tices for the safe and cautious handling of this potential haz- imum design condition and maximum operating condition.
ard, should be developed prior to retrieval of any catenary
mooring system. Several documented near misses have 3.1.1 Maximum Design Condition
occurred in the US GOM MODU operations as the direct
The maximum design condition is defined as that combi-
result of uncontrolled release of trapped chain torque. This
nation of vvind, waves, and current for which the mooring
potential situation presents a high risk of severe injury or loss
of life. system is designed. Mooring systems should be designed
for the combination of wind, wave, and current conditions
2.5.8 Exceedance of Mooring System Capacity to causing the extreme load in the design environment. In
Perform its Function practice, this is often approximated by the use of multiple
sets of design criteria. For example, in the case of a
Before construction, the design capacity of the mooring 100-year design environment, three sets of criteria are often
lines will have been determined from a mooring analysis. An investigated:
anchor holding test and winch capacity test will confirm
design conditions. It is unlikely however that the winch sys- a. the I00-year waves with associated winds and currents,
tem will be able to operate in the worst environment when the
load is borne by the chain stoppers. Operating out of the b. the 100-year wind with associated waves and currents, and
proven range will lead to the windlass being unavailable and
could lead to overload on the adjacent lines. Similarly, poor c. the lOO—~year current with associated wave and wind.
anchor holding capacity will lead to anchor drag which can
result in a less than optimum. mooring spread. The most severe directional combination of wind, wave,
Written operating procedures for managing mooring line and current should be specified for the permanent installation
tensions should be clearly defined and readily available to being considered, consistent with the site's enviromnental
those on board. This will include the use of thrusters if avail- conditions. Special attention should be given to certain float»
able, the redisttributiont of line tensions to prevent design ing structures such as large ship-shaped vessels, which are
exceedance of component, and including the course of dominated by low frequency motions. Since low frequency
action should anytiiitng unforeseen occur. motions increase with decreasing wave periods, the I00-year
In some areas of the world the practice on receipt of severe waves may not yield most severe mooring loads. Lower
weather forecasts is to reduce tensions on all mooring lines waves with shorter periods could yield larger low frequency
and to evacuate personnel. In other areas, personnel stay on motions and thus higher mooring loads.
board and initiate measures to protect the installation. Clear
operating procedures must exist for both scenarios. 3.1.1.1 Maximum Design Conditions for
All components should be maintained to a satisfactory Permanent Moorings
level achieved by the use of a robust Planned Maintenance
System. with unambiguous criteria for discard and replace- The l0O-year return period design criterion should be used
ment of key components. for permanent moorings. If the design life of the mooring is
substantially lower than 20 years, a shorter recurrence inter-
2.5.9 Operator Error val may be justified. In this case, the recurrence interval
should be determined by a risk analysis taking into account
Hazards which arise from operator error include the inabil-
the consequence of mooring failure.
ity to implement all items described in this Section and as
defined in the operating procedures. Operator errors can be For a permanent operation with a mooring system that per-
minimized with adequate training of responsible persons cou- mits rapid disconnection of the production vessel from the
pled with a company-wide culture to eliminate errors of all mooring, the maximum design condition is the maximum
types. This may require mandatory attendance at appropriate environment in which the production vessel remains moored.
training courses, the provision of clear procedural guides and However, the mooring alone (without the production vessel)
operations manuals, setting out the chain of command, drills should be able to withstand the maximum design environ-
and briefings before undertaking work. ment for permanent moorings.
 Dacron AND Auatvsrs or STATIONKEEPING SYSTEiviS FOR FLOATING Srsucruaes 15

3.1.1.2 Maximum Design Conditions for Mobile The operating environmental criteria should be known to
Moorings the people responsible for the drilling, offloading, or produc-
tion operations in order that timely plans to suspend opera-
3.1.1.2.1 Operations Away From Other Structures tions can be performed.
Moorings for mobile floating units such as MODUs oper-
ating away from other structures should use a maximum 3.2 ENVIRONMENTAL DATA
design environment with a return period of at least 5 years.
Special attention should be given to operations in an area of Models leading to the design responses of interest should
tropical cyclone such as Gulf of Mexico (hurricane) and consider the jointly distributed environmental phenomena.
South China Sea offshore China (typhoon). These areas are Environmental data, such as Wind, Wave, current and tide,
characterized by generally mild environment combined with have site-specific relationships governing their interaction.
severe storms during the cyclone season. For operations out When collecting data, various relationships should be
of the cyclone season, the 5-year environment can be deter- included. Of particular importance are the wind/wave,
mined using the environmental data excluding tropical wave height/period, and wave/current relationships and
cyclones. For operations during the cyclone season, the tropi- their relative directions. The directions of various environ-
cal cyclone data should be included. mental phenornena are especially important for single point
0

In some areas of tropical cyclone, the return period can be moorings.

reduced for operations during the season of tropical cyclone if The maximum design environments for mobile moorings
both of the following conditions are met: discussed in Section 3.1 should generally be determined by
annual statistics. However, if the operating season is well
a. A risk analysis is conducted to evaluate the consequence defined and seasonal environmental data are sufficient to
of a mooring failure. Such an analysis examines various provide meaningful statistics, these environments can be
scenarios of mooring failure, probability of occurrence of determined by seasonal statistics. There are areas governed
each scenario, and their effect on safety and environment. by special weather events, which may not be well repre-
The risk analysis may be used to determine the appropriate sented by typical return period statistics. For example some
return period, but it should not be less than one year. areas of mild climate may be subject to “sudden storms”
b. Operational personnel evacuation is planned and executed such as squaiis, and other areas may be subject to very high
before arrival of a tropical cyclone. currents. In these cases, the special weather events should
also be taken into consideration in determining the environ-
3.1.1.2.2 Operations in the Vicinity of Other mental criteria.
Structures The mooring system should be assessed. under the most
unfavorable wind/wave/current directions that can be reason-
The maximum design environment for mobile units oper- ably assumed to occur. The ability of the vessel to change
ating in the vicinity of other structures should have a return heading in response to changing environmental conditions
period of at least 10 years. In an area of tropical cyclone, for can be considered.
operations out of the cyclone season, the I0-year environ-
ment can be determined using the environmental data exclud-
3.3 WIND
ing tropical cyclones. For operations during the cyclone
season, the tropical cyclone data should be included. Opera- Two methods are generally used to assess effects of wind
tions that belong to this category are characterized by higher for design:
risk of collision or contact of structures or equipment. An
example of such an operation is a MODU with mooring lines 1. Wind is treated as constant in direction and speed,
deployed over a pipeline. Damage to the pipeline may occur which is taken as the 1-minute average.
if the anchors are dragged into the pipeline. Other examples 2. Fluctuating wind is modeled by a steady component,
include a drilling tender, a floater, or a service vessel moored based on the 1-hour average velocity, plus atime-va1y-
next to a platform. ing component calculated from a suitable empirical
wind gust spectrum. The recommended wind gust
3.1.2 Maximum Operating Condition spectrum is presented in Appendix B.
The maximum operating condition is defined as the combi- For the final design of permanent moorings, Method 2
nation of maximum. wind, waves, and current in which the should be used. However, Method 1 may be used if it can be
unit can continue to work, for example, to drill, produce, off- shown to be more conservative. For mobile moorings, either
load or maintain gangvvay connection. This condition shall method may be used. The design wind speed should refer to
not exceed the maximum design condition. an elevation of 10 rn above still-water level.
16 AP! RECOMMENDED PRACTICE 28K

3.4 WAVES 4 Environmental Forces and Vessel

The wave height versus wave period relationships for the Motions
d.esign seastate should be accurately determined from ocean- 4.1 BASIC CONSIDERATIONS
ographic data for the area of operation. The period can signif-
icantly affect vvave drift forces and vessel motions, and Environmental loads can be categorized according to the
therefore a range of wave periods should be examined. For following three distinct frequency bands:
fatigue analysis, the long-term joint distribution of wave
a. Steady loads such as wind, current, and wave drift forces
heights and periods (scatter diagram) is required, and a single
are constant in magnitude and direction for the duration of
best estimate of the associated wave period can be used for
interest.
each seastate.
b. Low frequency cyclic loads that excite the platform at its
3.5 CURRENT natural periods in surge, sway, and yaw. Typical natural
The most common categories of currents are: periods range from l to l0 minutes.

a. tidal currents (associated with astronomical tides), c. Wave frequency cyclic loads with typical periods ranging
from 5 to 30 seconds. Wave frequency cyclic loads result
b. circulation currents (loop and eddy currents), in wave frequency motions, which are typically indepen-
c. storm-generated currents, dent of mooring stiffness. The approach of neglecting
mooring stiffness in wave frequency motion calculation is
d. soliton. currents. always conservative. For small floating structures such as
buoys where the first order wave loads are not large,
The vector sum of the currents applicable to the site is the
accounting for mooring stiffness will yield more realistic
total current for each associated seastate. The speed and
wave frequency motions. Also if the natural period of the
direction of the current at different elevations should be spec»-
moored vessel is close to the wave periods, the wave fre-
ified. In certain geographic areas, current force can be the
quency rnotions can be dependent on the mooring
governing design load. Consequently, a selection. of appropri-
stiffness. In this case the effect of stiffness should be prop-
ate current profile requires careful consideration.
eriy accounted for.
3.6 WATER DEPTH
4.2 GUIDELINES FOR THE EVALUATION OF
The design water depth for the mooring system should ENVIRONMENTAL FORCES AND VESSEL
account for sea level variations due to tides, storm surges, and MOTIONS
seafloor subsidence, if applicable.
Environmental forces and vessel motions may be deter»
3.7 SOIL AND SEAFLOOR CONDITIONS mined either by model testing or calculation. The water depth
effects should be included. Guidelines for evaluating these
For permanent moorings, bottom soil conditions should be items are provided in API RP 2T (Reference 7). The wind
determined for the intended site to provide data for the spectrum recommended for the evaluation of low frequency
anchoring system design. Shape of the seafloor should be wind forces can be found in Appendix B. For mobile moon
properly accounted for in the mooring analysis. A bottom ings either API RP 2T or the simplified methods presented in
hazard survey should be performed. These measures should Appendix C can be used.
also be considered for mobile moorings to ensure adequ.ate Cylindrical. deep draft vessels such as spar under current
anchoring.
condition may be subjected to significant VIM (Vortex
Induced Motion). Guidelines for mooring analysis under
3.8 ATMOSPHERIC ICING VIM conditions can be found in Appendix H.
Increased wind area due to superstructure icing should
be considered in platform wind force calculations, as 4.3 SIMPLIFIED METHODS
appropriate.
Design equations and curves for a quick evaluation of
3.9 MARINE GROWTH environmental forces and vessel motions are provided in
Appendix C. These simplified analytical tools were devela
The type and accumulation rate of marine growth at the oped primarily for the analysis of mobile moorings. They
design site may affect weight, hydrodynamic diameters, and may be used for preliminary designs of permanent moorings
drag coefficients of vessel members and mooring lines. This if more accurate information is not available at the early stage
should be taken into consideration in the design. of the design process and if the limits for these tools are not
 Dacron AND Arou..\rs:s or STATiON§<EEF‘!NG Svsrei»/is son Froarruo Sraocrosas '27

exceeded. Simplified methods are available for the following between the environmental load and the mooring system’s
force components: restoring force. Wave frequency and low frequency vessel
a. Current forces for ship shaped and semisubmersible hulls. motions are calculated from frequency domain approach that
b. Mean wave drift forces and low frequency motions for yields standard deviation motion responses. Statistical peak
ship shaped and semisubmersible drilling vessels. values, such as significant or maximum responses, are then
evaluated based on certain peak response distributions.
c. Steady wind force.
Finally, the wave frequency and low frequency responses are
d. Wind and current forces for large tankers. properly combined to yield the maximum combined response
e. Forces due to oblique environment. for specific storm duration.
To perform analysis for weather vane vessels, using the
5 Mooring Strength Analysis approach of frequency domain vessel dynamics, the vessel
heading must be fixed. The fixed design heading, at which
5.1 BASIC CONSIDERATIONS
the mooring system responses are calculated, should be deter-
mined taking into consideration the mean equilibrium head-
5.1.1 Introduction ing and low frequency yaw motions.
Mooring analysis shall be performed to predict extreme
responses such as line tensions, anchor loads, and vessel off- 5.1.2.2 Time Domain Approach
sets under the design environment and other external loads
(e.g., riser loads, tandem mooring loads, etc.) The responses In this approach, the general equations of motion describ-
are then checked against allowable values to ensure adequate ing the combined mean, low, and wave frequency response of
strength of the system against overloading and sufficient the vessel are solved in the time domain. The forcing func-
clearance to avoid interference with other structures. tions include the mean, low frequency, and wave frequency
Active control of mooring system by mooring line adjust- forces due to wave, wind, current, and thrusters. The dynamic
ment may be performed for certain operations. However, equations describing the vessel, mooring lines, risers, and
active mooring line adjustment should not be considered in thruster forces are all included in a single time domain simu-
the mooring analysis for maximum design conditions. lation. Time histories of all system parameters (vessel. dis-
Guidelines for mooring strength. analysis are provided in placements, mooring line tensions, and anchor loads, etc.) are
the following sections. A strength analysis example can be available from the simulation, and the resulting time histories
found in Appendix J. Strength analysis under VIM conditions are then processed statistically to yield expected extreme val-
is presented in Appendix H. Guidelines for global analysis, ues. The time domain simulation should be long enough to
which is required for strength and fatigue design, can be establish stable statistical peak values.
found in Appendix I.
5.1.2.3 Combined Time and Frequency Domain
5.1.2 Simulation of Vessel Dynamics Approach
The following three approaches can be used for simulating To reduce the complexity and computational effort associ-
vessel dynamics. All three approaches include certain tech- ated with the full time domain simulation, a combined time
niques of approximation and therefore may not yield consis- and frequency domain approach is often employed. Time and
tent results. Also all three approaches have limitations. The frequency domain solutions for mean loads, wave and low
approach selected for the mooring design should be verified frequency motions can be combined in different ways. In a
by model test data, full-scale test data, or a different analytical typical approach, the mean and low frequency responses
approach. (vessel displacements, mooring line tensions, and anchor
Because of the weather vane nature of the vessel with a loads, etc.) are simulated in time domain while the wave fre-
single point mooring, the vessel may experience large low quency responses are solved separately in frequency domain.
frequency yaw motions. These yaw motions may signifi- The frequency domain solution for wave frequency responses
cantly affect vessel and mooring system responses, and there- is processed to yield either statistical peak values or time his-
fore should be accounted for in time or frequency domain tories, which are then superimposed on the mean and low fre-
simulations as well as in model testing. quency responses.

5.1.2.1 Frequency Domain Approach 5.1.3 Simulation of Mooring Line Response

In this approach, the general equations of motion describ- The responses of a mooring system to mean forces can be
ing the response of the vessel are de-coupled and analyzed predicted by static catenary equations. Generally speaking,
separately for mean, low, and wave frequency responses. the responses to low frequency motions can also be predicted
Mean responses are calculated from static equilibrium by the same method because of the long periods of these
18 AP! RECOMMENDED Pnacrrcs .2-SK

motions. The responses to wave frequency vessel motions changes, causing an interaction between this nonlinearity
can be predicted by either quasi»-static or dynamic analysis. and the geometric nonlinearity.

5.1.3.1 Quasi-Static Analysis 5.1.4 Riser Considerations

In this approach, the dynamic wave loads are taken into The riser system interacts with the vessel and the rnooring
account by statically offsetting the vessel by wave induced in several aspects. Wave and current loads on the risers
motions. Vertical motions and dynamic effects associated increase the environmental loads resisted by the mooring,
with mass, damping and fluid acceleration on the mooring while the riser system stiffness provides assistance to the
line are neglected. Research in mooring line dynamics has mooring. Furthermore damping from the riser system
shown that the accuracy of tension predictions based on this decreases the low frequency motions and in turn reduces the
method can vary widel.y depending on the vessel type, water mooring load. The net result of these effects depends on a
depth, and line configuration. number of factors such as type and number of risers and
water depth, etc. Mooring design should take into consider-
5.1.3.2 Dynamic Analysis ation the riser loads, stiffness, inertia, and damping unless it
Dynamic analysis acconnts for the time varying effects doe can be demonstrated that neglecting some or all riser effects
to mass, damping, and fluid acceleration. In this approach, the will result in same or more conservative mooring design.
time-varying fairlead motions are calculated from the vessel's Some of the floating production units are equipped with
surge, sway, heave, pitch, roll, and yaw motions. Dynamic steel catenary risers or midwater flowlines arranged in asym-
models are used to predict mooring line responses to the fair» metric patterns, which may impose large riser or flowline
lead motions. loads on the mooring system. In this case the riser or flowline
Two methods, frequency domain and time domain analy~ loads should be carefully evaluated. and properly accounted
ses, can be used for predicting dynamic mooring loads. In the for.
time domain method, all nonlinear effects including line
stretch, line geometry, fluid loading, and sea bottom effects 5.1.5 Damping of low frequency motions
can be modeled. The frequency domain method, on the other Low frequency motion of a moored vessel is narrow
hand, is always linear as the linear principle of superposition Ia A A ' i‘ _ .' .4‘ ‘ A ‘: + A L‘? Y % T M “““+
u€'tfiuCu 111 1I'CC'it1CfiC-3" S1fiC@ it i3 uGm1I'iE‘u.€u u_y tut: J.!E&DU11Cl11L
is used. Methods to approximate non—linear effects in the fre» response at the natural frequency of the moored vessel. The
quency domain and their limitations should be investigated to motion amplitude is highly dependent on the stiffness of the
ensure acceptable solutions for the intended operation. mooring system and damping. There is a substantial degree
There are four primary nonlinear effects that can have an of uncertainty in the estimation of low frequency motions,
important influence on mooring line behavior: particularly in the area of damping. There are four sources of
a. Nonlinear Stretching Behavior of the Line“-The strain or d.amping:
tangential stretch of the line is a function of the tension a. ‘Viscous damping of the vessel, including wind, wave, and
magnitude. Nonlinear behavior of this type typically current drag
occurs only in synthetic materials such as polyester. Chain
and wire rope can be regarded as linear. In many cases the b. Wave drift damping of the vessel
nonlinearity can be ignored and a linearized behavior
c. Mooring system damping
assumed, using a representative tangent or secant
modulus. d. Riser system damping
b. Changes in Geometry»»-—»The geometric nonlinearity is The technology to estimate viscous damping has well been
associated with large changes in shape of the mooring line. established, and viscous damping is normally included in the
low frequency motion calculations. Wave drift damping,
c. Fluid Loadi.ngm-The Morrison equation is most frequently
mooring system damping, and riser system damping, how-
used to represent fluid loading effects on mooring lines.
The drag force on the line is proportional to the square of ever, are more complex and are sometimes neglected because
of a lack of understanding in these damping components.
the relative velocity (between the fluid and the line), hence
Recent research indicates that these damping components can
is nonlinear.
be significant. They can even be higher than viscous damping
d. Bottom Effects--In many mooring designs, a consider- under certain conditions, and neglecting them may lead to
able portion of the line is in contact with the seafloor. The significant over—estimation of low frequency motions. In
interaction between the line and the seafloor is usually applications where low frequency motions is an important
considered to be a frictional process and is hence nonlin- design factor, such as large ship—shaped vessels, it may be
ear. In addition, the length of grounded line constantly warranted to evaluate damping from all these sources either
 DES§GN Ann Ateriwsis or STA'i'iOi\il<EE?°ii\iG Svsrsrvis son Ftorrrino Sraocroaes 19

by analytical approach or model testing. A more detailed dis- Table 1—Rec0mmended Analysis Methods
cussion on damping can be found in Appendix I. and Conditions
Conditions to be I
5.2 MOORING ANALYSIS CONDITIONS Type of Mooring j Analysis Method Analyzed l
. D ,_. , ............- - lI
Mooring Analysis can be carried out for intact, damaged or
Permanent Mooring
transient conditions according to Table 1. Definitions of those
conditions are given hereunder. Descriptions of analysis con- 3 Strength design Dynamic Intact/damaged
ditions for thruster assisted moorings are given in Section I Fatigue Design Dynamic Intact
5.9.2.
Mobile Mooring I
5.2.1 Intact Condition
Strength design Quasi-static or dynamic lntactfdamagedf
This is the condition in which all mooring lines are intact. transient 3
Fatigue Design Not required Not required
5.2.2 Damaged Condition
*1 Transient analysis should be performed to check vessel offset
This is the condition in which the vessel oscillates around a (check for tension is not required) for mobile moorings under any
new mean position after a mooring line breakage or a thruster one of the following conditions:
I * A floating vessel is moored in the vicinity of another structure
system failure.
~ A MODU conducts a deep-water drilling operation where
excessive transient motions may cause damage to the drilling
5.2.3 Transient Condition riser such as stroke-out of the slip joint or exceeding the flex
joint limit.
This is the condition in which the vessel is subjected to
transient motions (overshooting) after a mooring line break»
age or a thruster system failure. Swg,-g e Significant wave frequency motion
Sprmax R Maximum low frequency motion
5.2.4 Recommended Analysis Methods and
I"! ._ __ 45.1.: __ ___ ___
DUTIUIIIIQTIS
Q1.» . m $53. on 3. :1 ('5{D3 1'
1.1.11.1 LLB
Inn: -P
l.\J YT 1.1. ti,1}$2(Drs ('5 '-:: - i-3ofinn
.I. J L \v.I..\J',l..l.

The analysis methods to be used and conditions to be ana-

The maximum offset is the larger of the values determined
lyzed for various designs are defined as follows:
by the following equations:
5.3 VESSEL OFFSET
Smax 3 Smear: T SQ’;/nax T Swfsig (51)
The following definitions of vessel offset apply to any par»-
ticular points of interest in the vessel.
Smax T Smear: J“ Swfmax if Slfsig (5 -2)
5.3.1 Mean Offset
The combined offset from different degrees of freedom
The mean offset is defined as the vessel displacement due (for example surge and sway) shonld be defined as the vector
to the combination of the steady components of wind, wave, sum of individual degrees of freedom.
current, and other external forces. Alternatives to this approach are the time domain approach
(Section 5.1.2.2) and the combined time and frequency
5.3.2 Maximum Offset domain approach (Section 5.1.2.3), which involve statistical
When the frequency domain approach is used for the simu- processing of simulated time history to yield expected
lation of vessel dynamics (Section 5.1.2.1), the maximum off- extreme offsets.
set is defined as the mean offset plus maximum displacement The above discussion applies to the intact and damaged
due to combined wave frequency and low frequency vessel conditions. For the transient condition, maximum offset is
motions. Maximum offset can be determined by the follow- defined in Section 5.10.
ing procedure.
5.4 LINE TENSION
Let:
Smean e Mean vessel offset 5.4.1 Mean Tension
Smax 1 Maximum vessel offset
The mean tension is defined as the line tension correspond-
Swfmax = Maximum wave frequency motion ing to the mean offset of the vessel.
29 AP! Rscounsnoso Paacrics 28K

5.4.2 Maximum Tension natural period of the vessel/riser/mooring system Tn,

When the frequency domain approach is used for the simu- which can be estimated by the following equation:
lation of vessel dynamics (Section 5.1.2.1), the maximum
tension is the mean tension plus appropriately combined TF3 x 2a It/F71/,1. (5.8)
wave frequency and low frequency tensions. Maximum ten-
where
sion can be determined by the following procedure.
rn == system mass including added mass,
Let:
Tmax R" Maximum tension
k = system stiffness taken at the vessel's mean
' position.
Tmem, K Mean tension
Equation 5.6, vvhich is based on a narrow band Gaussian
Twfmax ="~"~ Maximum wave frequency tension
process with Rayleigh distributed peaks, may not always
Twfin-g I Significant wave frequency tension yield conservative predictions of maximum value. For non-
Tff,”ax H Maximum low frequency tension Rayleigh peak distributions, alternative approach such as
model test or time domain. simulation for the specified storm
Tim-g K Significant low frequency tension
duration is often used. If this approach is used, the simulation
or model testing should be of sufficient length to establish
The maximum tension is the larger of the values deter»-
reasonable confidence bounds for the expected maximum
mined by the following equations:
response in the storm duration. Typically responses in the
storm duration should be simulated several times, and statisti»
Tmax T Tmean J" Tlfinax T Tvvfisig (5-3)
cal fitting techniques should be used to establish the expected
maximum response (Refer to Appendix I for more guidance).
Tmax T Tmean T Twfmax T Tbbig (5-4) Of particular concern is the passive turret moored vessels
that will not maintain a constant heading because of low fre~
The above discussion applies to the intact and damaged quency yaw motions. The peak response distribution can be
conditions. For the transient condition, the maximum tension.
significantly affected by the variation in vessel heading, and
is defined in Section 5.10
the assumption of Rayleigh distribution can substantially
Alternatives to this approach are the time domain approach
under-estimate the maximum value.
(Section 5.1.2.2) and the combined time and frequency
domain approach (Section 5.1.2.3), which involve statistical
processing of simulated time history to yield expected 5.6 STRENGTH ANALYSIS BASED ON
extreme tensions. FREQUENCY DOMAIN VESSEL DYNAMIC

5.5 STATISTICS OF PEAK VALUES 5.6.1 Analysis for Spread Mooring Systems
For wave frequency or low frequency responses (motions, In a mooring strength analysis based on the approach of
tensions, etc.) that can be represented by a narrow band Gaus- frequency domain vessel dynamics, the mean position of the
sian process with Rayleigh distributed peaks, statistical peak vessel is first determined by the force equilibrium in the
values used in the frequency domain approach can be calcu- surge, sway, and yaw directions. For vessels equipped with a
lated from o (standard deviation) by the following equations: spread mooring system where the yaw moment will not have
a significant impact on the vessel heading and line tension,
Sig. Value I 2o (5.5) the yaw moment can be neglected. The responses to wave
and low frequency excitations are then calculated and added
Max. Value K A/2(ln N) o (5.6) to the mean position. The procedure outlined below is recom-~
mended for the strength analysis using a quasi—static or
N= T/Ta (5.7) dynamic approach:
l. Determine the environmental criteria such as wind and
where current velocities, significant wave heights and peri-
T e the specified storm period (sec), ods, their relative directions, storm duration, and wind
and wave spectrum for both the maximum design, and
Ta K the average zero up crossing period (sec) of the
operating conditions.
response.
2. Determine the mooring pattern, the characteristics of
A minimum of 3 hours should be specified for the storm chain, wire, and synthetic rope to be deployed, and the
period. For low frequency motions, Ta can be taken as the initial tension.
 Danton nun anatvszs or Srarzouaaaaiuo Srsraius eon Fi.O£iT§i\iG Sraucroaas 2t

3. Determine the mean environmental loads acting on the and wave spectrum for both the maximum design, and
hull using either model test data or the procedures operating conditions.
described in Section 4. Determine the mooring pattern, the characteristi.cs of
4. Determine the vessefs mean offset due to the mean chain, wire, and synthetic rope to be deployed, and the
environmental loads using static mooring analysis initial tension.
approach which should account for line stretching and Calculate the combined. mean environmental. yaw
friction. moment about the turret due to wave, wind, and cur-
5. Determine the low frequency motions using the data rent as a function of vessel heading. These yaw
moments may be based on model tests or calculated
and procedures described in Section 4 or through a
wind, current, and wave drift force and moment
h.ydrodynamic motion analysis. Since calculation of
coefficients.
low frequency motions requires the knowledge of
mooring stiffness, the mooring stiffness at the mean From the mean environmental yaw moment, determine
offset should be determined first using a static mooring equilibrium headings and their stability. Stable vessel
analysis approach. headings occur where the total environmental yaw
moment is zero and a perturbation of the vessel head-
6. Determine the significant and maximum wave fre- ing results in a yaw moment opposed to the direction
quency vessel motions from a hydrodynamic motion of the perturbation.
analysis or model test data.
Determine the yaw moment stiffness at the equilibrium
7. Determine the vessel's maximum offset, suspended heading. For an unlocked turret the yaw moment stiff-
line length, maximum tension, and anchor load using ness is the rate of change of the mean environmental
Equations 5.1 to 5.4. For quasi-static analysis, the yaw moment with respect to heading.
wave frequency line tensions are calculated by static Determine the standard deviation of the vessel’s low
catenary equations. For dynamic analysis, the wave frequency yaw response about the stable equilibrium
frequency line tensions are calculated by frequency headings using a hydrodynamic motion analysis pro-
domain or time domain line dynamics. gram. This requires knowledge of the low frequency
8. Compare the maximum vessel offset, suspended line yaw moment spectrum, the vessel yaw inertia and
length, maximum line tension and anchor load from added inertia about the turret, the yaw moment stiff-
ness, and the vessel and mooring system yaw damping.
step 7 with the design criteria stated in Section 7. If the
All of the above should be determined for the stable
criteria are not met, modify the mooring design and
mean vessel heading under consideration.
repeat the analysis.
In the absence of better information, the linearized
5.6.2 Analysis for Single Point Mooring Systems yaw damping coefficient about the turret can be
estimated from the sway d.amping as follows.
Because of the weather vane nature of the vessel with a
single point mooring, the vessel may experience large low 0,, 3 3:-c,(a3 +123)/(a+ a) (5.9)
frequency yaw motions, and time domain simulation or
model testing may be most appropriate for the mooring where
design. To perform analysis based on the approach of fre-
quency domain vessel dynamics, certain assumption on the CR2 K linear yaw damping coefficient, Nm/(Rad/
vessel heading must be made. The technology is still in a state sec),
of development, and a number of approaches have been Cy X linear sway damping, N/(m/sec),
investigated and used by the industry. An example approach, a 1 length of vessel forward of turret,
which is considered to be conservative, is presented below.
b I length of vessel aft of turret.
In this example approach, the design heading at which the
mooring system responses are calculated, is defined as the Calculate the design heading as the vessel’s stable
stable equilibrium heading of the vessel under mean environ- equilibrium heading plus or minus (which ever is more
critical) the significant (two standard deviation) low
mental loads plus or minus the significant low frequency yaw
frequency yaw response.
motion. The recommended analysis procedure is as follows:
Fix the vessel heading at the design heading deter-
1. Determine the environmental criteria such as wind and mined in step 7 above. Then follow the procedure
current velocities, significant wave heights and peri- described in steps 3 through 8 of Section 5.6.1 for cal-
ods, their relative directions, storm duration, and wind culating the mooring system responses.
22 AP! Racouusuoao Pnacrzca 2SK

5.7 STRENGTH ANALYSIS BASED ON TIME individual storm simulations can be expected to increase as
DOMAIN VESSEL DYNAMICS the number of low frequency cycles in the storm duration
decreases (low frequency natural periods increase).
As described in Section 5.1.2.2, time domain methods
For turret moored vessels the low frequency natural period
may be used to perform coupled simulations of mean, low,
of the vessel’s yaw motion will generally be significantly
and wave frequency vessel and mooring system responses.
longer than the surge and sway natural periods. When the
This approach requires a time domain mooring analysis com-
yaw natural periods are long, a large scatter in the realizations
puter program, vvhich solves the general equations of motion
of extreme values from individual storm simulations can be
for the combined mean, low, and wave frequency responses
expected. Consequently, a large number of repetitions of the
of the vessel, mooring lines, and risers. A significant advan-
storm simulation may be required to achieve confidence in
tage of this approach is that low frequency damping from the
the prediction of the expected extreme response values. More
vessel, mooring lines, and risers are internally generated in
detailed time domain analysis guidelines can be found in
the simulation. Also the coupling between the vessel and the
Appendix I.
mooring/riser system can be fully accounted for. This
approach requires, however, much higher computer
resources and engineering effort. It also requires that the 5.8 STRENGTH ANALYSIS BASED ON COMBINED
computer software be validated against either model test TIME AND FREQUENCY DOMAIN VESSEL
results or other analytical solutions. The recommended pro- DYNAMICS
cedure is as follows: In this approach, the mean and low frequency responses
are typically simulated in time domain which allows for non-
l. Determine the environmental criteria such as wind and
linearities in stiffness of mooring lines and risers, and in ves-
current velocities, significant wave heights and peri-
sel forces d.ue to quadratic terms and changes in yaw angle.
ods, their relative directions, storm duration, and wind
Constant or variable thruster forces may also be modeled.
and wave spectrum for both the maximum design, and
Transient motions resulting from line breaking or thruster
operating conditions.
failure may be evaluated by specifying the time of failure in
2. Determine the mooring pattern, the characteristics of the time domain analysis. Unlike the full time domain
chain, wire, and synthetic rope to be deployed, and the approach as described in Section 5.7, evaluation of low fre-
initial tension. quency damping cannot be included in this simulation
3. Determine the vesselis wind and current force coeffi- because of the absence of wave frequency components.
cients, and hydrodynamic model of the system Damping must be evaluated separately and treated as an input
including vessel, riser, and mooring. parameter.
4. Perform a time domain simulation for the storm dura- ‘Wave frequency vessel motions are calculated separately
tion using a time domain vessel/mooring analysis in the frequency domain from the vessel’s motion RAOs and
program. Repeat the simulation several times using the wave spectra. These motions can be combined with the
different seed values for generating the wave and wind low frequency motions in two ways. In the first method, the
time histories. frequency domain solution of wave frequency vessel motions
is transformed to a time history, which is added to the mean
5. Use statistical analysis techniques to establish the and low frequency vessel displacement to arrive at the com-
expected extreme values of vessel offset, line tension, bined vessel displacement. In this case the seed values for
anchor loads, and grounded line length. generating wave frequency and low frequency time histories
6. Compare the extreme vessel offset, line tension, anchor should be the same to yield consistent results. In the second
loads, and grounded line length from step 5 with the method, the mean and low frequency motions time histories
design criteria in. Section 7. are statistically analyzed to determine the peak values, which
are then combined with the peak values of wave frequency
The extreme value of a particular response parameter (ves-
motions to arrive maximum vessel offset, as described in Sec»
sel oifset, line tension, anchor loads, grounded line length,
tion 5.3.
etc.) realized in a single time domain simulation will vary
about its expected value. Consequently, statistical fitting tech-
5.9 THRUSTER ASSISTED MOORING (TAM)
niques and repetition of the simulation are required to estab-
lish reasonable confidence in the predicted extreme response.
The number of repetitions of the simulation that are required
5.9.1 Introduction
will depend upon the extreme value characteristics of the sys- In evaluating the capability of the thruster assist systems, it
tem response parameter and the sophistication of the statisti- is necessary to consider the equipment that support and con-
cal methods used to predict the expected extreme value. In trol the thmsters, their modes of failure and repair times, and
particular, the scatter of realizations of extreme values from the training and experience of the personnel operating the
 Dssion AND eiiatrsis or Srarionasseino Svsreras son Ftoarsve Srnucruass 23

systems. The determination of thruster forces is a complex restrictions. Guidance for evaluating available thrust is
mul.ti-disciplinary undertaking, requiring knowledge of provided in Appendix F.
thruster design, thruster performance in normal and extreme
2. Determine the worst thruster system failure. Failure
seastates, thrust allocation logic and optimi:z.ation, control and modes and effects analyses (FMEA) should be per-
monitoring systems, reliability of mechanical and electrical formed to identify the worst single failure. The
equipment, computer hardware and software, operational definition of the worst single failure should allow for
practices, operator trainin.g, and human factors, etc. This doc- thruster system availability (mean time to failure and
ument does not provide detailed recommendations and meth- mean time to repair) over the design life of the
ods for the design of thruster assisted moorings. Rather, the installation.
guidance and background information contained in this sec-
tion is intended to be used by the mooring analyst, when the 3. Determine the allowable thrust:
thruster systems have been designed and proved to acceptable For automatic thruster control systems, the allowable
industry standards. thrust can be determined as follows:
Thrusters may be used to assist the mooring system by
- For the intact thruster condition, the allowable
reducing the mean environmental forces, controlling the ves- thrust is equal to the available thrust or effective
sel’s heading, damping low frequency motions, or a combina- bollard pull when the thruster system is operating
tion of these functions. n.ormal.ly.

5.9.2 Analysis Conditions for Thruster-assisted ~ For the damaged thruster condition, the allowable
Moorings thrust is equal to the available thrust after
accounting for the worst failure as determined by
Intact and damaged. TAM system definiti.ons are given FMEA.
below.
For manual thruster control systems, the allowable
Table 2-Intact and Damaged TAM Definitions thrust as determined above should be multiplied by a
reduction factor of 0.7.
nnfini.f§nn onfl I
fai>
ML I I I I I U191! Irllll.-I l i
The allowable thrust used in the mooring analysis
ZE°5.oring Factor of Mooring System I Thruster System should be verified during thruster system sea trials. A
Safety Condition Condition
. _ .____ — ____________ ____;';_i~'***;'____i ‘ — — — —
detailed discussion on thruster assisted turret mooring
l
can be found in IMCA Report No. GM”.-2096-0600»
Intact i Intact Intact
2561 “Guidance on Thruster Assisted Station Keeping
Damaged ‘ Intact Damaged
by FPSO and Similar Turret Moored Vessels” (Refer-»
ence 8).
Damaged Damaged Intact
____... ______. "77 _____..t_ _ _ . __ ____ __ 5.9.4 Analysis for Thruster Assisted Mooring

Table 2 implies that if both the mooring system and the In a thruster-assisted mooring system, the load sharing
thruster system condition are “intact”, then the TAM defini- between the thruster and the niooring systems is complex and
tion and mooring factor of safety is “intact”. If either the can only be fully accounted for with a time domain system
mooring system or the thruster system condition is “dam- dynamic analysis. However, a simple mean load reduction
aged”, then the TAM definition and mooring factor of safety method would yield reasonable results.
is “damaged”.
5.9.4.1 Mean Load Reduction Method
5.9.3 Determination of Allowable Thrust In this simplified approach the thmsters are assumed to
When thrusters are used to assist the mooring system, it is counter only the mean environmental loads in the surge,
necessary to quantify the allowable thrust that may be used in sway, and yaw directions. Allowable thrusts from thrusters
the mooring analysis. The recommended procedure is out- are first evaluated and then subtracted from the mean load.
lined below: The remainder of the mean load, and the wave and low fre-
quency motions would be taken by the mooring system.
1. Deterniine the available thrust taking into consider- For vessels with a spread mooring where the vessel head»
ation the eiiiciency of the thrusters and losses due to ing is held stable by the mooring lines, the surge and sway
vessel motions, current, thruster/hull and thruster/ components of the allowable thrust can be subtracted from
thruster interference effects, and any directional the mean surge and sway mean environmental forces. The
24 API RECOMMENDED PRACTICE ZSK

mooring response can then be evaluated using analysis proce- l. Compute the equilibrium position under mean load for
dure for mooring systems without thruster assist. an intact mooring.
For vessels with a single point mooring, the main function 2. Break a line and compute the maximum transient
of the thrusters is heading control. Reduction. of surge and motion (overshoot) in the time domain with mean load
sway mean loads by thrusters should generally not be consid- only and with the mooring system stiffness, K, updated
ered. For vessels with high thruster capacity that significantly at each time step. Generally a model with three degrees
exceeds the heading control requirements, mean load reduc- of freedom (surge, sway, and yaw) is required.
tion can be considered for extreme environments. In this case,
a conservative portion of the thruster capacity should first be 3. Determine maximum vessel offset by the following
allocated to heading control. The remaining portion can then equation.
be used to reduce the surge and sway mean load.
Snzoar T Smear: T St T Sn/fisig if Slfsig (510)
5.9.4.2 System Dynamic Analysis where
A system dynamic analysis is normally performed using a Smear, I mean offset as calculated in step 1,
three axis (surge, sway, and yaw) time domain simulator. This
simulator generates the mean offset, low frequency vessel S, I maximum transient motion (overshoot)
motions, and thruster responses corresponding to specific with respect to the equilibrium position
environmental force time records. In this analysis, constant from step 1 as determined in step 2,
wind, current, steady wave drift forces, and the low frequency
Swjfw-g I significant wave frequency motion, cal-
wind and wave drift forces are typically included. Wave fre-
quency wave forces, which are not countered by the thruster culated in the frequency domain,
system, can be excluded in the simulation. The wave fre- S313‘,-g e significant low frequency motion, calcu-
quency motions can be computed separately using a vessel lated in the frequency domain using the
motion program and added to the output from the time damaged. mooring system stiffness.
domain simulator. To obtain a proper maximum value from
the time domain simulation, it may be necessary to generate a
5.10.2 Time Domain Analysis
number of force and response records for the storm duration
and calculate the expected maximum value using a statistical Time domain transient analysis is similar to the time
approach. domain analysis for vessel dynamics described in Section 5.7.
The only difference is that a mooring line is removed during
5.10 TRANSIENT ANALYSIS the simulation to model a line break. The simulation should
be repeated for a number of wave force records, and for the
A moored vessel will experience transient motions after a
break to occur at a number of times during each record. The
mooring line breakage or thruster system failure before it set-
maximum offset observed during these simulations, or the
tles at a new equilibrium position. Transient analysis check-
most probable maximum offset estimated from the results of
ing maximum offset is required for certain mobile mooring
these simulations, should be used for the design.
operations, as specified in Section 5.2.4. Transient analysis of
a moored vessel under wind, wave, current, and thruster load-
ing is complex and may require a time domain solution. To 6 Fatigue Analysis
simplify the analysis, a combination of time domain (tran-
sient motions) and frequency domain (vessel motions)
6.1 BASIC CONSIDERATIONS
approaches can be used. Fatigue life estimates are made by comparing the
long-term cyclic loading in a mooring line component with
5.10.1 Combination of Time and Frequency the resistance of that component to fatigue damage. For
Domain Analysis mooring systems, a T-N approach is normally used. This
In this approach, the maximum transient motion is first approach uses a T-N curve, which gives the number of cycles
determined using a time domain approach. Then vessel to failure for a specific mooring component as a function of
motions obtained from frequency domain approach are super- constant normalized tension range, based on the results of
imposed on the transient motion. The recommended. proce- experiments.
dure is as follows: The l\/Iiner’s Rule is used to calculate the annual cumula-
tive fatigue damage ratio D:

D -_ ENE
:2 (6.1)
 Dasiou nun Auaivsis on Srarrouaaanirie Svsraus son Froariue Sraocrunas 25

wliere curves from a simple regression analysis. The special consid-

11,- K number of cycles per year within the tension erations include:
range interval 1'. ' Fatigue failure probability analysis.
N5 Ia number of cycles to failure at normalized ten- ~ Experience check such as perforrning a fatigue analysis
using a T-N curve under investigation to see whether
sion range i as given by the appropriated T-N
the fatigue life prediction severely violates industry
curve.
experience.
The design fatigue life, which is 1/D, should be greater
- Other analytical approaches such as finite element for
than the field service life multiplied by a factor of safety
stress concentration or fracture mechanics.
defined in Section 7.5. For used mooring components, fatigue
damage from previous operations should be taken into * Comparison with other published T-N curves for the
account. component.
Quasi-static approach should not be used for calculating When selecting a T-N curve for fatigue analysis of a com-
tension ranges due to its severe deficiency in estimating wave ponent, such as studless chain, the designer should consider
frequency tensions. Both time and frequency domain the safety margin in all three integral parts of a fatigue life
dynamic approaches may be used for tension range predic- assessment (T-N curves, factor of safety, and analysis
tions. Alternatively, tension ranges can be obtained from method) against experience and available test data.
model testing.
Guidelines for mooring fatigue analysis are provided in the 6.2.1 Tension-Tension (T-T) Fatigue
following sections. A fatigue analysis example can be found
in Appendix J. Guidelines for mooring fatigue analysis under T-N curves presented in Equation 6.2 can be used for cal-
VIM conditions are presented in Appendix H. Fatigue dam- culating nominal tension fatigue lives of mooring compo-
age to piles from installation loading is addressed in Appen- nents.
dix E, Section E.7.
Na”-K (6.2)
QB IO
%
is‘6;0 m F?‘Z531co—S2|
O'11 OJ9
is1>:2 ("J I-11 O -rz E5 (DO E2 :2 ("D 2?.I-rii(9
N I Number ofcycles,
T-N curves for various mooring components should be
based on fatigue test data for these components and a regres- R I Ratio of tension range (double amplitude) to
sion analysis. The component T-N curve typically corre- reference breaking strength (RBS). For chain,
sponds to a lower bound defined as the lower bound of a two- RBS is taken as MBS (minimum breaking
sided, 95 percent prediction interval (2.5% probability of strength) of ORQ common chain link of the
fatigue resistance exceedance). However, this practice is not same size for ORQ, R3, R4, and R4S common
followed. precisely for some recommended T-N curves or connecting links. Guidance on increase of
because of uncertainties in the test data such as: chain diameter for corrosion and wear and its
effect on fatigue life calculation are given in
* Insufficient test data. 7.6. For wire rope, RBS is the same as MBS.
~ Lack of test data in the low tension regime where Let Ln, = Ratio of mean load to reference breaking
fatigue damage is most severe. strength for wire rope, M and K values are pro-
- Data acquired in unrealistic test environments such as vided in Table 3.
high frequency test in seawater environment.
Table 3--M and K Values
~ Lack of a broad representation of test samples such as
testing samples of l or 2 sizes from 1 or 2 manufac- Component M K
turers. - - -- -- ---- ---T7
_____ ___ __ _
------------------ --
____ _____
,
_7___
- -----— -—
_ ___________ _ _ mi
- -- - -- --—7—r

Common studlink 3.0 i 1,000 7

Most of the uncertainties are due to difficulties to meet
high funding requirement for fatigue testing of mooring com- Common studless link 3.0 3 l6
ponents. To address these uncertainties, special consider»
Baldt and Kenter connecting link 3.0 178
ations were sometimes taken in addition to regression
analysis for test data. As a result, some of the recommended Six/rnulti strand rope 4.09 10.‘3"2°“m*m’
T-N curves may represent the “best estimates” of the fatigue
resistance of the component instead of the lower bound
t S,,,,, ,,,,,,, .,,, 5,05 A 1Ow l
2.6 API Recorvaaanoao Fnacrice 28K

The mooring component fatigue design curves are plotted include galvanizing, jacketing, blocl<ing compound, and zinc
in Figure 13. Note the wire rope cu.rves are mean load depen- filler wires. Careful investigation considering the design life,
dent, and a mean load of 30% MBL is assumed in the plot. inspection, and change-out strategy should be carried out to
The equation for studlink does not applied to links with determine the combination of these elements needed for a
loose stud. Figure 13 indicates that studless chain has lower specific project.
fatigue life than studlink chain. However, Studless chain does Mean load has a significant influence on wire rope fatigue
not have the fatigue issues associated with stud, such. as loose life and therefore should be included in the design curve
stud, stud weld crack, sharp corners at stud foot print, corro- equations. A mean load of 0.3 RBS is considered to be repre-
sion between stud and link, and defects hidden behind the
sentative for conventional mooring systems. For wire rope
stud that cannot be detected by inspection. It is important to
fatigue analysis, the following methods can be considered to
consider all factors affecting fatigue life in the selection of
account for the mean load. effect:
chain type.
The above T-N curves should be used in conjunction with a 1. For each seastate, determine the mean load and the cor-
factor of safety 3, as specified in Section 7.5. A reliability responding design. curve, which is then used to
study indicates that the T-N curves combined with a factor of calculate the fatigue damage for that seastate. This
safety of 3 and Simple Summation method (Section 6.3) requires using different design curves for different
would yield fatigue designs with acceptable probability of seastates.
failure (see Reference 18 for details). Since the slope of the
above T-N curve for studless chains lies outside the 95% con- 2. Determine the average mean load for seastates causing
fidence range from a regression analysis on the available test significant fatigue damage and use the design curve for
data, the T-N curves based on regression. analysis of the test the average mean load for all seastates.
data presented in 2 chain fatigue test JIP reports (References 3. Use the design curve for a mean l.oad of 0.3 RBS for
19 and 20) may also be considered (Reference 21). conventional mooring systems. For a taut leg mooring
The T-N curves for wire rope are only good for wire ropes or TLP tether system, method (1) or (2) should be
protected from corrosion. Elements for corrosion protection used.

0.0 .=---Z-»--_--__-.-.~ -- . . --- - --- I .000

------ Chain-Studlink
_ 0.2 _g _____________________ ~. _ _ _ _ ——O— Chain-Studless _ M __ 0.63,
/
\ x
+. Connector, Kenter Link I

.. $3‘ e- ...t_._____ /
__ _ I ........l Si></MuIti-Strand, Mean Load = 30% Mas
- -- - -I 0.398
j —— --W Spiral Strand, Mean Load = 30% IVIBS I
Streng
9\
-0.6 _- ---------- -~ II/,
Zif5 . __........___._R__ 1iiIIIIiiI I. II i i I I I I i i i I l -~-I 0.257
k
th

O / _;_ _1_

/BrRefereakStnce rength ‘
._~._ II .__ // \.
/
Breance
-93 Jr sssssssssss __ ;>/ ..I.l I lirIIII II a
<5; /I 53>
i

Range _apm /I /i
-1.0 -1- ------------------------------ -- - - - - - - - --‘>-<>;----- _£__/
I“ Range/Refers
ens'0n
I
/
LJ.. ._
---<_-------W 0.100 I’!

I . p / \
‘*1 -1.2 ----------- -- i II 1iIII / eTenso
TR,og0R=
I

-1 4;-. r/I I I I 1 s
ii

/
9
I 4 ones R

—~—- I /i
as iiiiIIIIiiII I
ii i
I I I III I 1iiiI I
e -——I—— - ------- ——0.040
-,.-‘.Q,-t.._.-i¢-»‘~w‘.,»~m“,_-_._-
I -.

“T'”I
1
-1.6 _ _I . .! _ _ ‘alumni _ wn1____ Timur
0.025
UJ -lb U‘! 0‘
_ ._ Q. O0

Iog,O(N),N 1» Number of Cycles

Figure 13-—I\/Iooring Fatigue Design Curves
 QESKEN %i\iD §%i‘i§"\i.YSiS OF STAWONKEE?-‘ii‘~iG SYSTEMS i‘OR l:i.OATiNG $"i'Ri_}CTURES an -~i

Among the three methods, Method (l) is most accurate centration u:nder these conditions should be carefully evalu-
but requires more computational effort. If method (2) and ated by finite element analysis, especially for permanent
(3) are used, a sensitivity study should be performed. to moorings. Fatigue analysis should account for the additional
ensure these simplified approaches produce conservative stress concentration in these areas.
predictions.
Data for other types of connecting links are insufficient for Table 4—Comparison of B-T and T-T Fatigue Life
generating design curves. Limited test data indicate that the
i B-T Fatigue Life in
fatigue life of D-shackles is comparable to that of common Terms of Percentage of
links of the same size and grade, provided the shackle is , Rope Type D/d Ratio T-T Fatigue Life
W, ---------------W - ,,,,,,,,, ----------~ - l- kiwi-:,,,__,
machined fit with close tolerance, no locking pin is used pen-
Six strand I 20 3
etrating through shackle body. ._ —_———_—_—_ #1 — —————————— ——____—_ ———— ___7 — __
____

Six strand 70 8
6.2.2 Bending - Tension (B-T) and Free Bending
Fafigue l\/lulti-strand 20 L}!

Data for bending~tension fatigue of chain and wire rope are l\/Iulti-~strand 70 15
insufficient for generating design curves. In the absence of a Spiral strand 2l@i GL5
fatigue design, precautionary measures should be taken to l

avoid mooring failure due to bending-tension fatigue. For Spiral strand 2:1(2; 1.5
example, the fairlead to line diameter ratio (D/d) should be
large enough to avoid excessive bending. The portion of
6.3 FATIGUE ANALYSIS FORMULATION
mooring line in direct contact with a fairlead should be regu-
larly inspected. Also this portion should be periodically 6.3.1 Accumulated Fatigue Damage
shifted to avoid constant bending in one area. A study com-
paring the bending-tension and tension-tension fatigue lives The annual fatigue damage, accumulated in a mooring line
of mooring lines on a semisubmersible under the North Sea component as a result of cyclic loading, is summed up from
environment provided the following data which can be used the fatigue damage arising in a set of environmental states
as reference values for establishing operation policy to avoid chosen to discretise the long-term environment that the moor-
excessive bending tension fatigue for wire ropes. ing system is subjected to:
Note that the analysis for bending»~tension fatigue is very
complex and the values in Table 4 are rough estimates only. z'=n.

Additional more margin of safety is recommended when 0 a 21), (6.3)

these data are used.
Free bending at the wire rope ter1ninati.ons can also signifi-
cantly reduce the wire rope fatigue life. To avoid premature where
fatigue failure in permanent moorings, a bend limiting device
should be incorporated at these locations. Such a device is D, = the annual fatigue damage to the component
designed to smoothly transfer the loads from the termination due to environmental state i.
to the rope.
The di.scretisation into 1' =1, ...,n environmental states
As for tension bending of chain, the portion of mooring
should be sufficiently detailed to avoid any significant error
line in direct contact with a fairlead should also be regularly
in the total. Each environmental state is defined in terms of
inspected and shifted to avoid constant bending in one area.
the wind, wave, and current parameters and directions
In general, the worst load case is to tension-bend a horizontal
required to compute mooring system responses. The proba-
link over a shallow groove, which results in very high stress
bility of occurrence, Pi, is required for each environmental
in the stud weld region. Therefore, fairleads must be shaped
state. The calculated fatigue life of the mooring system is:
and sized to avoid this type of unfavorable bending of chain
links. Limited fatigue T-N tests of chains over a five--pocket
fairlead indicate 5% to 20% of B-T fatigue life in terms of T- L 1 1 / D (years) (6.4)
T fatigue life. A seven~pocket fairlead design generally gives
much improved B-T fatigue life. The annual fatigue damage accumulated in an individual
Industry experience indicates that chain links in direct con- state may be computed as:
tact with fairleads, bending shoes, chain stoppers, or hawser
pipes can be subjected to additional stress concentration,
D, K 3’-€"E[1i§?“‘] (6.5)
which in turn can cause premature fatigue failure. Stress con-
28 API RECGMMENDED F’RAC"f'iCE 28K

where the combined. low and wave frequency tension

spectrum.
M and Ki are defined in Table 3 and:
rt; I the number of tension cycles encountered in The time domain cycle counting (Method 4) is generally
state i per year, considered to be the most accurate method of calculating
fatigue damage, however the analysis is relatively time con-
E[R,,- SM] I the expected value of the normalized tension suming. Simple summation (Method l) will generally give an
range R; raised to the power M, in state i, acceptable estimate of fatigue life if the ratio of standard
The number of tension cycles per year in each state can be deviation tensions between wave frequency and low fre-
determined as quency response satisfies the following condition:

rt; X v,» 1?} K vi - P,-3.15576 >< 107 (6.6) Grif/%..2l..5 or “if/Q-,,§o.o5

where
vi K the zero up-crossing frequency (hertz) of the Where crwf and 0'1; are wave frequency and low frequency
standard deviation tension, respectively. However, this
tension spectrum in environmental state i,
method (Method 1) may underestimate fatigue damage com-
T, Z the time spent in environmental state i per year, pared to Method 4 if both low and wave frequency tensions
P; I the probability of occurrence of environmental contribute significantly to the total fatigue damage. The com-
state 1'. bined spectrum method (Method 2) is always conservative
and may significantly overestimate the actual fatigue damage.
The normalized tension ranges should be computed taking
The combined spectrum with dual. narrow-banded correction
into account the effects of pretension and the effects of the
factor method (Method 3) is an improvement, which yields
environmental loads due to wind, waves and current.
less conservative predictions than Method 2. It is suitable for
Although the cumulati.ve effect of the tension cycles is
the cases where both low frequency and wave frequency ten-
required in the fatigue analysis, rather than the extreme ten-
sions cause significant fatigue damage. However, when the
sion, it is still necessary to take care to compute the dynamic
fatigue damage is dominated by low frequency tensions, this
response of the mooring line to wave-frequency loads at a
method (Method 3) will overestimate the fatigue damage.
representative offset for each environmental state.
Analysis procedures for Methods 1, 2, and 3 are presented
below.
6.3.2 Methods for Combining Low and Wave
Frequency Fatigue Damage 6.3.2.1 Simple Summation
Four methods can be considered, as follows, for combining Wave frequency and low frequency fatigue damages for
fatigue damage due to low frequency and wave frequency
environmental state 1', are estimated by the following equa-
tensions.
tion, which is based on a Rayleigh distribution of tension
l. Simple Summ.ation~»--In this approach, low frequency peaks.
and wave frequency fatigue damages are calculated
independently. The total damage is assumed to be the
sum of the two. Di I %’i<~/iris.->M ~ r<1+M/2) i %<~/Eats)” ro+M/2) (67)
2. Combined Spectrum--In this approach, the combined
low frequency and wave frequency spectrum is first where
calculated. The total damage is then calculated using D; m annual fatigue damage from wave frequency and
the standard deviation of the combined spectrum. low frequency tensions in environmental state 1',
3. Combined Spectrum with Dual Narrow-Banded Cor- V1 W; e number of wave frequency tension cycles per year
rection Factor--In this approach, a correction factor is for environmental state z‘, from Equation 6.6,
applied to the result of the combined spectrum method
(method 2). RWG-5 =-"-"- ratio of standard deviation of wave frequency ten»
sion range to RBS. The standard deviation of the
4. Time Domain Cycle Counting--~In this approach,
tension range should be taken as twice the stan-
fatigue damage is calculated using a cycle counting
dard deviation of tension,
method, such as the Rainflow method, to estimate the
number of tension cycles and the expected value of the T = Gamma function,
tension range from a time history of tensions. The ten- nu I number of low frequency tension cycles per year
sion time history may be determined directly by a time for environmental state i, from Equation 6.6. The
domain mooring analysis or it may be generated from average zero up-crossing frequency may be esti-
 Dacron ruin Ari/u.\rs2s or Srarioretaasaio Svsrsivis son Froarino Srnucrunas 29

mated by .1/Tn, where Tn is the natural period. of cess. The fatigue damage for environmental state i is esti-
the vessel computed at the vessel‘s mean position, mated from Equation 6.12.
.RLU,- Z ratio of standard deviation of low frequency ten-
sion range to RBS. The standard deviation of the 5'25 M q
tension range should be taken as twice the stan-
B, e Q, En/ism.) -F(1+M/2) (6.12)
dard deviation of tension.
The correction factor is given by:
6.3.2.2 Combined Spectrum
In the combined spectrum method fatigue damage for
3, q l a+2 2‘, _ Mr ( 3--T-M)
2 V _ MW
environmental state i is estimated. from the fol.lowing equa-
eisifil (7‘~r.s)2 ll +"~im“.ti'7~irr‘ Tm fl“ 31' ifrrrriz
tion, which is based on a Rayleigh distribution of tension sq x ’~.r.r pf for
peaks.
(6.13)
D, -= Z-git./2i,,.>M -1“(l+M/2) (es) Where the subscript e refers to the envelope of the com-
bined tension process, and the mean up-crossing frequency
(hertz) of the envelope of the normalized tension process, ve,-,
In Equation 6.8 the standard deviation of the combined low is given by,
and wave frequency tension range, R6,-, is computed from the
standard deviations of the low, Rm,-, and wave, R WU,-, fre-
quency tension ranges by, v... - 6/;\*f.i'Vf.t + }\*Li7\'Wz"ViVz'6€Vi <6-14>
where
Roi 3 '\!Rii/ar+Ri.az (69) SW, I the bandwidth parameter for the wave fre-
quency part of the normalized tension process,
The number of cycles, n,-, in the combined spectrum is cal- which may be taken as equal to 0.1.
culated from Equation 6.6 with the zero up-crossing fre-
quency (hertz) of the combined spectrum, vg,-, given by,
6.4 FATIGUE ANALYSIS PROCEDURE
The recommended procedure for a detailed fatigue analysis
VG; T‘ '\1}'~.r.ri’.i.r+ )l‘*Z'i"z'ViVz' (6-10)
is described belovv.
where l. The long term environmental events can be represented
by a number of discrete environmental states. Each envi-
vm K the zero up-crossing frequency (hertz) of the
ronmental state consists of a reference direction and a
wave frequency tension spectrum in environ- reference seastate characterized by significant wave
mental state i, height, peak spectral period (or equivalent), spectral
vL,- ="- the zero up-crossing frequency (hertz) of the shape, current velocity, and wind velocity. The probabil-
low frequency tension spectrum in environ- ity of occurrence of each environmental state must be
mental state i. specified. In general, 8 to 12 reference directions provide
a good representation of the directional distribution of a
and 7tL,- and?tW,- are given by, long term environ_ment. The required number of refer-
ence seastates normally falls in a range of 10 to 50.
Fatigue damage prediction can be fairly sensitive to this
Rt.
it-:1“ T".'{""5'" = it-Wt ""‘ "r'“"'“'“'"if‘7
R2 (6-11) number for certain mooring systems, and therefore it is
Ru T R Wt -Rt; T R W1‘ best determined by a sensitivity study.
2. Each environmental state can be analyzed analogously
6.3.2.3 Combined Spectrum with Dual Narrow- to the procedure used for mooring strength analysis as
Banded Correction Factor described in Section 5. The wave and low frequency ten-
The combined spectrum with dual narrow-banded correc- sions can be computed about the position of the mooring
tion factor method uses the result of the combined spectrum system under mean loading only.
method and multiplies it by a correction factor, pi, based on 3. Determine the M and K values for Equation 6.2.
the two frequency bands that are present in the tension pro- (Table 3).
30 AP! RECOMMENDED PRACTICE ZSK.

4. Compute the annual fatigue damage from one environ» cient vertical load resistance for the soil condition under con-
ment (one seastate in one direction) due to both the low sideration. Guidelines for the use of drag anchor to resist
frequency and the wave frequency tension according to vertical loads are provided in Appendix D.
methods presented in Section 6.3. Shorter line lengths can be used for moorings with other
5. Repeat Step 4 for all environmental states and compute anchoring systems that can resist substantial vertical pulls
the total annual fatigue darnage D and fatigue life L such as pile anchors, suction caissons, or vertically loaded
according to Equations 6.3 and 6.4. anchors.

7 Design Criteria 7.4 ANCHORING SYSTEMS

7.1 VESSEL OFFSET 7.4.1 Drag Anchor

Vessel offset limits should be established by clearance The holding capacity of a drag anchor in a particular soil
requirements and limitation of equipment such as risers and condition represents the maximum horizontal steady pull that
gangways. can be resisted by the anchor at continuous drag. Evaluation
of anchor holding capacity is addressed in Appendix D. Fac-
7.2 LINE TENSION tor of safety for drag embedraent anchors, which is defined as
anchor holding capacity divided by maximum anchor load, is
A tension limit can be expressed as a percentage of the provided in Table 6.
minimum breaking strength (MBS) of the mooring compo- Table 64»-Drag Anchor Safety Factors
nent. MBS is defined as the breaking strength guaranteed by
the mooring component manufacturer. The minimum break» l A A A if Quasiisaac A
ing strength of chain may be taken. as the break test load Analysis Analysis
(BTL). Guidance on increase of chain diameter for corrosion Pennanent Mooring A
and wear and its effect on strength calculation are given in IITEEICT if T H W

Section 7.6. Damaged condition 1.0

Tension limits and equivalent factors of safety for various
conditions and analysis methods are provided in Table
Mebi1eMeeriee....
i,..i._..,..i .,_......4:.¢.;....l.z WT
l l l l l lif A A 7
irrtcuoi. bUilUli.1Ul.t . CD Q) 00
The criteria in Table 5 are intended for moorings which are Damaged condition 1 Not required l Not required
properly maintained and inspected, and have connecting
hardware with breaking strengths equivalent to the mooring
7.4.2 Pile Anchor, Vertically Loaded Plate Anchor,
lines.
and Gravity Anchor
The same mooring line tension safety factors are applica-
ble for thruster assisted moorings (TAM) assuming the Factors of safety, defined as anchor capacity divided by
thruster system is reliable and has significant contribution to maximum anchor load from dynamic analysis, for anchor
the station-keeping capability of the floating structure. piles, plate anchors, and gravity anchors are provided. in
Table 7.
Table 5--Tension Limits and Safety Factors A discussion on pile and plate anchor design and installa—
tion can be found in Appendix E. Further discussions on fac-
A Ti Analysisl Tension Limit i Equivalent tor of safety for anchors can be found in Section E.3.l (suction
5 Method (Percent of MBS) {Factor of Safety
anchor) and Section E.4.4 (drag embedded plate anchor).
Intact Quasi-static 50 2.0
l Intact Dynamic , 60 1.67
7.4.3 Mooring Test Load

Damaged Quasi»-static 70 1.43 After installation, the mooring should be test loaded. to
ensure adequate holding capacity of the anchoring system,
Damaged Dynamic so 1.25 eliminate slack in the grounded portion of the mooring lines,
detect damage to the mooring components during installation,
7.3 LINE LENGTH and ensure that the mooring line’s inverse catenary is suffi-
ciently formed to prevent unacceptable mooring line slacking
If drag anchors are used, the outboard mooring line length due to additional inverse catenary cut-in during storm condi-—
should in general be sufficient to prevent anchor uplift under tions.
conditions as specified in Section 5.2.4. This requirement is For permanent moorings with drag anchors, the mooring
especially important for anchors in sand and hard soil where lines should be test loaded to at least 80% of the maximum
anchor penetration is shallow. Uplift of drag anchor may be storm load determined by a dynamic mooring analysis for the
permitted if it can be demonstrated that the anchor has surfi- intact condition. This requirement is mainly based on industry
 Dsstos Ann ANAi..‘r’S¥S or $'¥“A'?¥ONKEEPif~tG €rrsrar.rs son Ftosrsivo Sraocroaas 31

experience with drag anchors in soft clay where deep anchor depends on type of steel and sea water environment, and is
penetration can be achieved. For drag anchors on hard or sand often significantly acclerated in the first few years of service.
seafloors where anchor penetration is typically limited to no If the corrosion rate is uncertain, a conservative approach
more than one fluke length, higher anchor test load may be using the chain diameter excluding the increase for corrosion
more appropriate. For permanent moorings with pile anchors and wear should be considered for the fatigue analysis.
and suction caissons, a test load should be determined based Corrosion of wire rope at connections to sockets can be
on the consideration of eliminating the slack in the grounded excessive due to the galvanized wire acting as an anode for
mooring lines, forming reverse catenary in the mooring line adjacent components. For permanent systems it is recom-
below the seafloor, and detecting damage to the mooring mended that either the wire be electrically isolated from the
components during installation. socket or that the socket be isolated from the adjacent compo»-
For mobile moorings with. drag anchors, test load should nent. Additional. corrosion protection can be achieved by add-
be determined by a number of factors such as type of anchor,
ing sacrificial anodes to this area.
soil condition, winch pull limit, and anchor retrieval. How-
ever, they should meet the following minimum requirements:
7.7 CLEARANCE
The test load at the anchor shan.k should n.ot be less than 3
times the anchor Weight.
The mooring test load at winch should not be less than the 7.7.1 Basic Considerations
mean line tension for an intact mooring under the maximum Contact between a floating vessel, its mooring cornpo~
design condition. This requirement is for close proximity nents, and other marine installations should be avoided with a
moorings only. comfortable margin, especially under severe environments.
Duration of the test load should be at least 15 minutes for For floating vessels moored in close proximity, it is common
both mobile and permanent moorings. Refer to Appendix E, practice to move one of the vessels away before a threshold
Section E.6.3 for test loading of plate anchors. environment is reached. Under normal operating environ»
mentor in areas of mild environment, close proximity moor-
7.5 FATIGUE LIFE ing can be acceptable, assuming some clearance criteria are
The predicted mooring component fatigue life shall be at met.
least 3 times the design service life of the mooring system. The clearances between a floating vessel, its mooring com-
ponents, and. other marine installations should be determined
7.6 CORROSION AND WEAR for the conditions specified in Table 1. To determine clean-
ance criteria, many factors should be considered, such as
Protection against chain corrosion and wear is normally environment, vvater depth, and risk of injury, asset and envi—
provided. by increase of chain diameter. Current industry romnental damage, etc. Conservative criteria should be estab-
practice is to increase the chain diameter by 0.2 mm to 0.4 lished based on these considerations. As minimum, the
mm per service year in the splash zone and in the dip or clearance requirements provided below should be met.
thrash zone on hard bottom. The diameter increase is reduced
to 0.1 mm to 0.2 mm per service year in the remaining length.
7.7.2 Mooring Line Crossing Pipeline
For strength analysis, the diameter of the chain should not
include the increase for corrosion and wear. For fatigue anal- Where a mooring line crosses a pipeline within the ele-
ysis, the diameters of the chain for different periods of service vated part of its catenary, a minimum vertical clearance of 10
life can be established if the corrosion rate can be predicted. m under the intact condition should be maintained. A moor-
In this case the chain diameter for a certain period is the norm» ing line may contact a protected pipeline provided this contact
inal diameter minus the expected corrosion and wear for the remains throughout the full range of predicted intact line ten-
time up to that period. It should be noted that corrosion rate sions thus the contact point must not occur in the thrash zone.

Table 7—Safety Factors for Pile, Plate, and Gravity Anchors (Dynamic Analysis)
K iiii if if ifWWguctign/Driven PileandGravityAnchor i i i i i i if an t.i..a...t.1ii i i i i i i i i i i i if
t tPi.i.i..ni i i i i i i i i i iii iiiiiiif W iiiii iiiif A iii if
Condition Lateral 1 ii.leaa.1 if Laiéiéi .tia1 Permanent Mobile 1
Intact 1.6 2.0 1.2 1.5 2.0 1.5
2 1 . .......... ... . . . 2 iiiii2 . _ - , 2
Damaged 1.2 1.5 10 1.2 . 1.5 1.2 1
__ _ _. ,,,,,,___ l
32 API Recorvmnenoan PRACTICE 28K

7.7.3 Mooring Line Crossing Mooring Line 8.2.1 Mooring Wire Rope
Where a mooring line crosses another mooring line, a min- Mooring wire rope should have no fiber core. Blocking
imum vertical clearance of 10 m under the intact condition compound of good. quality should be used to fill the spaces
should be maintained if one of the mooring line at the cross»- between the wires. The ends of each rope section should be
ing is grounded. The minimum clearance should be increased terminated with resin or zinc poured sockets. Mooring wire
to 20 m if both lines are suspended at the crossing. ropes and end sockets should meet the material, design, man-
ufacture, and testing requirements specified in certain classi-
7.7.4 Horizontal Distance Between Installations fication rules such as those provided in Reference 9.
For non.-torque-balanced mooring wire ropes, such as six.-
A minimum horizontal clearance of l0 m should be main- strand and eight-strand wire ropes, the torque or twisting
tained between the moored unit (or its mooring lines) and any characteristics of a wire rope should be considered in the
other installation. This clearance is required for all conditions design of mooring line configuration and in mooring line
as defined in Section 5.2. handling procedure to ensure proper mooring application and
safety of handling crews. Wire rope manufacturers should
7.7.5 Clearance Between a Drag Anchor and Other provide users torque/twist data for the allowable tension
Installations range as part of the wire rope basic properties.
Contact of wire rope in the dip or thrash zone may cause
If a marine installation lies in the dragging path between excessive wear in the rope/j acket or excessive free bending at
the anchor and the floating unit, the final anchor position the socket. This condition should. be avoided for permanent
should allow at least 300 m drag before contacting the marine moorings under normal operating environments. This condi-
installation. Otherwise the anchor should be at least 100 in tion may be acceptable under extreme environments.
from the marine installation.
8.2.2 Mooring Chain
7.7.6 Clearance Between Mooring Lines and Other Mooring chain should be manufactured according to one
Vessel Structures. of the following specifications:
r"-1 '1 1' 1 111 ' . .1 1.‘ .1 rv».
LOHSIGCTEIIIOH snau be given to the oetrimentai errects or n - API Spec 2F, .S‘pecif.icati0n ficwr .Mc0ring (Refer-
contact between mooring lines and other vessel structures ence 10)
such as anchor bolsters. ~ RCS (Recognized Classification Society) Rules for
Offshore Mooring Chain (Reference ll)
7.8 SUPPORTING STRUCTURES
8.2.3 Connecting Link
Supporting structures such as chain stopper, fairlead and Connecting links such as shackles and detachable links
their foundations should have equal or higher design strength. should be made of forged or cast material. They should be
than the mooring line. Special attention should be given to the fully inspected by non-destructive testing (magnetic particle,
design of supporting structures such that failure of the sup- die penetrant, eddy current, etc.) according to recognized
porting structure will not result in multiple line failures. standards. Cast connecting links should also be examined by
x-ray or ultrasonic testing to detect internal casting defects. In
8 Mooring hardware addition, forgings and castings should satisfy a Charpy V-
notch energy requirement of 40 J at -20° C. Kenter connect-
8.1 BASICCONSIDERATIONS ing links should meet the requirements specified in API Spec-
ification 2F (Reference 10). Inspection, mechanical, proof
'\

Specifications and conditions of the mooring components

and break testing of other types of connecting l.inks should
should be in accordance with those assumed or required by
meet similar requirements or other recognized standards.
the mooring analysis.
8.2.4 Mooring Buoy
8.2 MOORING LINE COMPONENTS
Mid~line buoys are typically constructed from steel or syn»
Requirements for mooring line components such as wire thetic materials. They should be rated for the maximum buoy
rope, chain, connecting link, and buoy are given in this sec- submergence depth derived for the mooring system intact and
tion. Requirements for synthetic fiber rope can be found in 1—line damage analysis. The maximum safe working depth
API RP ZSM, Recommended Practice for Design, Manufac- for mid--line buoys should be based on analyses and/or testing
raring, and Maintenance of Synthetic Fiber Ropes for Off- using applicable and recognized design and manufacturing
shore Mooring (Reference 2). codes.
 Dasion AND Anmvsis OF STATIONKBEPING Svsr n1 E on ‘T1 C) R Ftoanno Srnucrunss co co

Surface mid-line buoys should be designed. to meet the fol» devices du.e to buoy motions during the intended service life.
lowing conditions: Mid-line buoys subject to high motions should be in.spected
on a regular basis to ensure the integrity of the buoy connec-
~ Buoys should be designed to remain at the surface in all
intact and one-line damaged conditions, unless the tions to the mooring lines. If there is concern about the failure
buoy is rated for the maximum submergence that could of the buoy’s connection to the mooring line during the
occur in any of these conditions. intended service life or period between inspections, the con-
sequences of a loss of the buoy should he investigated in the
~ Buoys should be designed to remain at the surface in mooring system design as a damaged condition. In this case
case of one compartment flooding in all intact mooring the criteria for the damaged condition should be met.
conditions, if applicable such as for hollow steel buoys,
unless the buoy and the inter.nal compartment bound- 8.3 WINCHING EQUIPMENT
aries are rated to withstand the submergence resulting
from the flooding of any compartment. Winches should meet the requirements specified. in ISO
9089 “Marine Structures---Mobile Offshore Units--Anchor
* Filling of floodable compartments with foam prevent»
Wiuches,” First Edition, 1989-l2-Ol (Reference l2) and Sec-
ing the ingress of water in case of compartment damage
tion 4.ll. of Code for the Construction and Equipment of
may be considered to eliminate the I compartment
Mobile Qflshore Drilling Units, I989, International Maritime
flooded design, provided that the foam is capable of
Organization (Reference 13).
withstanding the hydrostatic pressure at the maximum
design submergence of the buoy.
8.4 MONITORING EQUIPMENT
' Certain low-density foams used to fill buoy compart-
ments, may take on water slowly over time. In this 8.4.1 Line Tension
case, some percentage flooding of the compartment
should be considered in the design of the buoy. Moored floating units should be equipped with a calibrated
system for measuring mooring line tensions if the operation
~ Buoys should be rated. to withstand overtopping by requires mooring line adjustment, and line tensions should be
storm waves in the maximum design environment. continuously displayed at each winch. For units that do not
~ Buoys should be fitted with draft marks and/or other require a tension measurement device, a device for detecting
suitable means for monitoring flooding of the buoy. mooring failure should be considered.
For units with thrusters that are intended for mooring line
Any mid-line buoy used to maintain a safe separation tension reduction, a means of indicating line tension and/or
between the mooring line and other critical systems, such as vessel offset should be provided. This means should be suit-
pipelines, other mooring legs, etc. shall be designed to main- ably redundant to cover the single failure requirement.
tain adequate clearance in all applicable intact, l-line damage
and 1 buoy compartment flooded conditions, and shall be
8.4.2 Line Payout
rated for the maximum buoy submergence in any of these
conditions. Moored floating units should be equipped with a system
For mid-line hollow buoys that are not close to other struc- for measuring mooring line payout if the operation requires
tures, the consequences of buoy collapse should be investi- mooring line adjustment.
gated, if applicable. For example, if the flooding of a
compartment in such a hollow mid-line buoy may result in 8.4.3 Vessel Position
the buoy submerging below its rated depth, the consequences
of a complete collapse of the buoy should be considered and Moored floating units should be equipped with a system
analyzed in the mooring system design. The consequences to for monitoring the position of the vessel if the operation
be analyzed should include, but may not be limited to: requires restraining the vessel offset. If available, a semi-rigid
dynamic mooring line and mooring equipment loads due to link to a fixed object (for example, link bridge from tender to
buoy implosion and rapid descent, impact of rapid changes in platform) may be used to monitor the floating unit’s position.
vessel position resulting from buoy collapse, recovery of col- For MODU operations, a position system should be available
lapsed buoy and mooring line, etc. to provide the unit’s bearing and distance from the wellhead
The connecting hardware of mid line buoy to the mooring or point of riser attachment. For units with automatic thruster
line should be designed for the maximum hydrostatic and assist system, the measurement of position should be by at
dynamic loads using applicable and recognized design and least two different means. For units with manual thruster
manufacturing codes. Also the design should consider the assist system, the measurement of position should be by at
wear or loosening of the connecting hardware and its locking least one means.
34 API RECOMMENDED PRACTICE 2-SK

8.4.4 Vessel Heading surements are required. If the heading control is automatic,
the accuracy and frequency of both measurements should be
Floating units with a single point mooring should be
adequate to meet automatic control requirements.
equipped with a device for measuring heading. If the head-
ings are to be controlled, at least two different heading mea-

I’x
ran wanna :___ uuunuun 11:" '""" """' "W :______ T7 7"

5 DP Control Room
Power system ' l Wind sensor | ‘ Computers

/
I >(}4.K4l
r
---—----1» | s Gyrocompasses
Wifid Venical Reference Units
Environmental _ _ _
forces < ' We
I' i ,_-... _______ _ I __ '

_ ‘P _ —flw_____9— 7 -.-wile’ - ‘*__,7‘-q,\?—’ W17 77 Y ‘—————gJ~' fin-can-¢_*;_,;,~:__ ___r_.§;_""'fii:~4I"""‘;‘i-_ — _ #::__r1~___, T __ i 4.f"“~ 1 ' 1,gqpi_" In-_________,¢u—~—__=r rrrr —— T 7777*-'-w—

i . .
—;___ 7 —_— *7 7 _nl_fi '

Wave S ® I if I I llllllllllllllllllllll”® if

l MM... Current
!*\
\
I. | \\
‘X Tnrusters

I \;\\ I
/"-'*""""-\____\
Hydrophones

/fl l ’ \

/\'—l
+-l-}~
thy)
:4’ ——-;»-,_ __
_

____ W
4,
.

Figure 14—l\/lajor Elements of a Dynamic Positioning System

:
W _
e-*;:;:
_ ____~__
_ _—_
Wt" ""
—~ _ *___
_
ii
1'?
;~- _

7 ~___
________ __

.__”w
Subsea beacon
-q. _-__
ll?
W
————————— e ' ""'

9 Dynamic Positioning a Control system: The DP computers and associated soft-

ware, including position references, vessel sensors,
9.1 BASIC CONSIDERATIONS
operator interface, power management, etc.
Dynamic Positioning (DP) is a technique of automatically
A more detailed description of DP system can be found in
maintaining the position of a floating vessel within a speci-
Reference 14.
fied tolerance by controlling onboard thrusters which gener»
ate thrust vectors to counter the wind, wave and current
forces. As shown in Figure 14, the major elements of a DP 9.2 DESIGN AND ANALYSIS
system include:
9.2.1 Basic Design Philosophy
~ Power system: Prime movers and auxiliary equipment,
generators, switchboards, cabling, etc. A DP system should be able to keep a vessel in position
within certain excursion limits under the design environment.
= Thruster system: Thrusters and auxiliary equipment, Since the consequences of losing station can be serious, DP
including main propellers and rudders, associated systems should be designed to have high reliability and a oer»
cabling and thruster control equipment, etc. tain amount of built-in redundancy.
 DESIGN AND Anntvsis OF STATIONKEEFWNG Svsretus FOR FLOATING Sraucruntzs 35

9.2.2 Failure Mode and Effect Analysis. floating production, shuttle tanker, etc. The particular risk
analysis is likely to take into account some of the following:
Failure modes and effects analysis (FMEA) should be con-
ducted for floating vessels with a DP system. Failure modes - The time to reach a safe situation, or recover from the
to be considered in the FMEA should include the following: immediate danger,
* The sudden loss of major items of equipment r Speed of loss of position (drift-off, drive-off, or a large
excursion),
* The sudden or sequential loss of several items of equip-
ment with. a common link. ~ Environmental limitations,
' Control and monitoring instabilities and failures, and ~ Operational procedures,
methods of detection and isolation
~ Human factors etc.
- Faults that can be hidden until another fault occurs
The risk analysis can be general and cover different work-
DP systems should be designed so that, as far as is reason- ing situations and types of work. However generic assump-
ably possible, there are no common single-point failures. The tions and principles should be considered for each project,
DP system FMEA should be proved in sea trials, as far as is location, and procedure to ensure the analysis is valid and/or
reasonably practicable, to demonstrate the effects of the vari- changes are made to maintain its validity and applicability.
ous failure modes and to ensure that both equipment and pro-
cedures are in place to safely cope with failures. 9.2.4 Determination of Station-Keeping Capability

9.2.3 Guidelines for Design, Test, and 9.2.4.1 Basic Considerations

Maintenance A holding capability analysis should be performed. to
Detailed guidelines for design, test, and maintenance of DP determine whether a DP system can maintain the position of a
systems can be found in the IMO document MSC Circular floating vessel within an acceptable watch circle under the
645, Guz'delz'aes for vessels with Dynamic Positioning Sys- operating environment. This analysis should be performed for
tems (Reference 15}. it defines vessei redundancy into three new designs as well as for individual operations. Two meth-
equipment classes, where Equipment class l vessels have the ods can be used to analyze the holding capability of a DP sys~
least redundancy, and vessels complying with Equipment tem. A time domain system dynamic analysis is normally
class 3 have the most redundancy. In this context ‘equip- perfonned for new system designs and critical operations,
ment’ refers to all the equipment (power, control, and refer- especially those in shallow water. For routine operations in
ences), together with its location/layout on the vessel, that deepwater, a simplified method addressing only the mean
goes to define the degree of redundancy. Equipment Class environmental forces can be used.
definitions are:
9.2.4.2 Environmental Loads and Vessel Motions
~ Class 1: Loss of position may occur in the event of a
single fault. A DP system counters steady loads and darnps out low fire-
quency motions. Wave frequency motions are not affected by
* Class 2: Loss of position should not occur from a sin-
the DP system and therefore can be excluded from the hold-
gle fault of an active component or system such as gen-
ing capacity analysis. However, they should be included in
erators, thrusters, switchboards, remote controlled
the determination of maximum offset.
valves etc. Static components such as cables, pipes,
manual valves etc. should be adequately protected
against accidental damage.
9.2.4.3 Available Thrust

' Class 3: Loss of position should not occur from any Guidelines for determining available thrust are provided in
single failure including a completely burnt fire sub- Appendix F, which deals with typical propulsion devices and
division or flooded watertight compartment. installation scenarios for DP vessels supporting offshore
operations.
Using these classifications and the results obtained from
the FMEA, it is possible to allocate the vessel with an equip» 9.2.4.4 System Dynamic Analysis
ment class notation. Selection of a DP vessel with its inherent
redundancy or class should be based on a risk analysis for the System dynamic analysis for a DP vessel is similar to that
particular type of DP operation. The risk analysis should take for a vessel with a thruster assisted mooring (Section 5.9.4.2).
into account the risks involved with specific operations such The major difference is that the mooring stiffness is not
as drilling, diving, flotel services, heavy lifting, pipe laying, included in the analysis.
36 AP] RECOMMENDED PRACTICE. 23K

9.2.4.5 Simplified Method API RP 16Q, Recommended Practicejor Design, Selec-

tion, Operation, and Maintenance of Marine Drilling
In this approach, we assumed the DP holding capability is
Riser Systems, Isl Edition, August 2001.
satisfactory if the DP capability is greater than the mean envi-
ronmental load. The procedure is as follows: API RP 2RD, Design of Risers for Floating production
(FPSs) and Tension Leg Platjbrms (TLPs), Isl Edition,
1. Establish an operating environment and a vessel head- June 1998.
ing relative to the operating environment.
API RP 2T, Planning, Designing, and Constructing Ten-
2. Calculate the mean surge and sway forces and the yaw sion Leg Platforms, 2nd Edition, August 1997.
moment due to wind, waves, and current.
IMCA Report No. GM-2096-0600-2561 “Guidance on
3. Determine the required output of each individual Thruster Assisted Station Keeping by FPSO and Similar
thruster based on the DP system algorithm for thrust Turret Moored Vessels.”
allocation.
DNV Certification Notes No. 2.5 “Certification of Off-
4. Determine the available thrust for each thruster.
shore Wire Ropes,” Det norske Veritas, 1995.
5. Calculate the DP capability according to IMCA M 140
API Spec 2F, Specification for Mooring Chain, 611‘ Edi-
Specification for DP Capability Plots, August 1997
tion, June 1997.
(Reference 16).
RCS (Recognized Classification Society) Rules for Ofl-
6. Repeat the previous steps for different headings, oper-
shore Mooring Chain
ating environments, and thruster failure cases.
7. Produce plots of calculated DP holding capability ver- ISO 9089, Marine Structures»-<Mobile Offshore Units -
sus vessel heading for different environments and Anchor Winches, First Edition, 1989-12-01.
thruster failure scenarios. Examples of DP holding IMO, Code for the Construction and Equipment of
capability rosettes for a semisubmersible and a drill Mobile Ojfihore Drilling Units, 1989, International Mar-
ship are presented in Figure 15. itime Organization.

5::ca C) '13 "ii :1

D
Z G) 1-J m23 U) Q 4!Z I11 |.-
OPL Oilfield Seamanship Series Volume 9, Dynamic
Positioning, ISBN 1 902157 02 8.
Specially trained personnel are required to operate the DP IMO document MSC Circular 645 “Guidelines for ves»~
system with its sophisticated electronic equipment. Guide»- sels with Dynamic Positioning Systems.”
lines for training DP operators can be found in the IMO docu-
ment MSC Cir. 738 “The Training and Experience of Key DP IMCA M 140, Specification for DP Capability Plots,
Personnel” (Reference 17). August 19.97.
IMO document MSC Cir. 738 “The Training and Experi-
ence of Key DP Personnel.”
10 References
Memo to API RP 2SK Work Group, “Studlink and Stud-
1. API RP 2I, Recommended Practicefor In-Service Inspec- less Fatigue Curves for Mooring Lines,” Exxonl\/Iobil
tion of Mooring Hardware for Floating Drilling Units, Upstream Research Company, March 2003.
zed Edition, Nov. 1996.
Joint Industry Study Report, “Fatigue Testing of Large
2. API RP ZSM, Recommended Practicefor Design, Manu- Diameter Anchor Chains,” Noble Denton & Associates,
facturing, and Maintenance ofSynthetic Fiber Ropes for Inc., August 19, 1994.
Ofiishore Mooring, ls‘ Edition, March 2001.
Joint Industry Study Report, “Corrosion Fatigue Testing
3. API RP 17A/ISO 13628-1, Recommended Practice for of 76mm Grade R3 & R4 Studless Mooring Chain,”
Design and Operation ofSubsea Production Systems, 3rd Noble Denton & Associates, Inc., May 15, 2002.
Edition, September 2002.
Memo to API RP 2SK Work Group, “T-N curve for Stud-
4. API RP 17B, Recommended Practice for Flexible Pipe, less Chain Proposed in API RP ZSK draft,” Shell, April
3rd Edition, March 2002. 2004.
 DESKBN AND /3.NAL‘tS¥S or 5"{A?iONKEEF=‘l¥‘\iG SYSTEMS roe Frosting Steucrut-ass ea --1

SEMISUBMERSIBLE

coo
340 35-100 .
O
2 040

2:: é\ ; Number of tnrusters available: 8

vessel drilling draft: "Z00 ft
Wind 54.0 knots from 000 deg

.’\/tA\\§§i;
-=0: .
Current 0.90 knots from O00 deg
2
Waves 21.0 ft from 000 deg
Riser restoring force 0 kips
250

240

230

2 ‘Qfiil
W 13$
unto
200 190 180 179
4 it 160
Percent of avallabie thruster loading versus vessel heading

DRILL SHIP

see 35° 009

we mo - 020
s o 3 3
so R UP 40

280

270
{\\\~//Q’ Number oi tnrusters available: F-6 A-6 M-6
Vessel drilling draft:
Wind
Current
Waves
54.0 knots from
0.90 knots from
21.0 ft from
22.0 ft
000 deg
000 deg
000 deg
2
Riser restoring force 0 klps
250 -

/1
advI :.\\~
-4 20

220

!9¢%.\
TQQ 186 76
1 40

Percent of available thruster loading versus vessel heading

Figure 15»--Dynamic Positioning Capacity Rosettes

 APPENDIX A--INTRODUCTION TO MOORING COMPONENTS

A.1 Mooring Line which is costly and difficult to deploy at deepwater sites. A
chain/wire rope/chain combination system is sometimes
A.1.1 CLASSIFICATION
used. For example, a wire rope is often inserted in an all chain
Mooring lines for moored vessels may be made up of mooring line on a MUDU to increase the water depth capabil-
chain, wire rope, synthetic rope, or a combination of them. ity of the drilling vessel.
There are many possible combinations of line type, size, and When fiber rope is used, it is normaI.ly placed in the cate-
location and size of clump weights or buoys that can be used nary portion of the mooring line to avoid contact with the sea-
to achieve given mooring performance requirements. Follow- floor and the winching equipment. Chain or wire rope
ing are typical systems used by the industry: segments can be used to connect the fiber rope to both the
vessel and the anchor.
1. All Wire Rope System
Because wire rope is much lighter than chain, wire A.1.2 CHAIN
rope provides a greater restoring force for a given pre-
tension. This becomes increasingly important as water The choice of material and fabrication of large diameter
depth increases. However, to prevent anchor uplift chain for a moored vessel requires careful evaluation. It is
with an all wire system, much longer line length is desirable to have chain used for this application manufac-
required. A disadvantage of an all wire rope mooring tured in continuous lengths for each mooring leg. This elimi-
system is wear due to long term abrasion where it con- nates the need for chain connection links and the associated
tacts the seabed. For these reasons, all wire rope moor- problems with fatigue. Otherwise, connecting links with suf-
ing systems are seldom used for mobile or permanent ficient fatigue life should he used.
moorings. Chain can be obtained in several grades with Grade 4 or R4
being the highest strength. Oil Rig Quality (ORQ) or R3
2. All Chain System
chain has been sold in large quantities over the years and has
Chain has shown durability in offshore operations. It generally performed well. Grade 2 chain is not recommended
has better resistance to bottom abrasion and contrib- for maj or mooririig operations.
utes significantly to anchor holding capacity. How- A grade of chain somewhere between R3 and R4, for
ever, in deep water an all chain system imposes an example R3S, is preferred by some designers since it is easier
increasing weight penalty on the vessel's load carrying to manufacture than R4 chain. In any case it is recommended
capacity by its own self weight and high initial tension that considerable care is taken in establishing correct chemi-
requirements. cal composition of the bar stock, manufacturing techniques
3. Combination System which incorporate precise quality control and finally, compre-
hensive testing of samples of the final manufactured product.
In this system, a mooring line may be a combination of
Recently a new grade, R4S, chain was introduced by some
chain, wire rope, and fiber rope. In a chain/wire rope
chain manufacturers. The R4S studlink chain is stronger than
mooring, a length of chain is typically connected to the
the R4 chain, with the increase in strength achieved by using
anchor. This provides good abrasion resistance where
a larger diameter bar stock within the overall dimensions of a
the mooring line contacts the seabed and its weight
specific chain link size and by improving the ultimate tensile
contributes to anchor holding capacity. The choice of
strength of the base metal. The outside length and width of
chain or wire rope at the vessel end and the type of ter-
the link are virtually unchanged, making it possible to use
mination also depends on the requirements for adjust-
existing chain windlasses and fairleads, thus enabling a sig-
ment of line tensions during operations. By proper
nificant increase in mooring chain breaking strength with
selection of the lengths of wire rope and chain, a com-
minimal changes in rig mooring equipment. The disadvan-
bination system offers the advantages of reduced pre-
tage of this chain is its relatively short service experience.
tension requirements with higher restoring force,
Stud chain (Figure A. l.) has been used by the offshore
improved anchor holding capacity, and good resistance
to bottom abrasion. These advantages make combina- industry for more than 30 years. For R3 grade, studs are often
tion system attractive for deep water mooring. welded on the side opposite to the flash weld. Studs are nor-
mally not welded for higher grades. The industry has experi-
An alternative to the above system is the wire ropelchainl enced significant problems associated with studs, including
wire rope combination system where wire rope segments are loose stud, fatigue crack and fracture at the stud weld or stud
connected to both the vessel and the anchor. A length of chain footprint. In the 1990s a new chain, studless chain (Figure
is used in the dip zone where the mooring line is in dynamic A. 1), gained wide acceptance in the application of permanent
contact with the seafloor. This minimizes the amount of chain moorings. Studless chain is about 10% lighter than the stud
40 API RECOMMENDED PRACTICE 28K

. ‘\\\ I
Stud footprint
Link

i it
; i
- F . il 1
.5
/ s ¥ Stud we-Id G _ i
I h"'§-
A
é . Stud lg ‘Ki'“" ‘jmKT_‘—flIT_ A
t, t NF{ { { { ,. _ ~ - M ????? /see//ié
Il uup-un .

; =1 i 1t " r tr 1 " 1

.' i
l
l l i
l\ i l ‘
4
| = ' i
l I

Flash weld

X, ' 1 “‘ Link

i
I
I

Studless Chain Stud Chain

Figure A.1—-Stud and Studless Chain

chain, but has the same breaking strength. Limited fatigue incorporated to provide additional corrosion protection. A
testing indicates that studless chain may have lower fatigue filler material is used to block the inside spaces between the
life, but the industry’s experience with studless chain has wires to minimize the spread of corrosion with ingress of salt
been favorable so far. In fact most of the permanent moorings water.
recently installed or designed. use studless chain instead of Life expectancy of different types of wire rope is provided
stud chain. below:
= Galvanized 6- strand 6-8 years
A.1.3 WIRE MOORING LINE
~ Galvanized unjacketed spiral strand 10-12 years
The wire rope sections of the moorings can be of various ~ Galvanized unjacketed spiral strand
constructions as shown in Figure A.2. The wire rope con- with zinc filler wires l5-l'7 years
struction type includes a number of strands wound. in the
same rotational direction around a center core to form the ~ Galvanized j acketed spiral strand 20-25 years
rope. The number of strands and wires in each strand (i.e., ~ Galvanized jacketed spiral strand with
6x36, 6x42, 6x54), core design and lay of strands are gov- zinc tiller wires 30-35 years
erned by required strength and bending fatigue considerations The ends of each mooring line section should be termi-
for the rope. This construction generates torque as tension nated with sockets. A resin material is preferred over zinc for
111CI'€€1S€S. pouring the sockets. For permanent moorings, the sockets are
The spin~resistant strand type constructions (spiral strand typically provided with flex relieving boots (bend stiffener)
and multi-strand) are attractive for use with permanent moor- joined to the socket in a manner to seal out the ingress of
ings since they do not generate significant torque with tension water and limit free benching fatigue. Zinc anodes are
changes. Both constructions use layers of wires (or bundles of attached to protect the socket from corrosion, and isolation
wires) wound in opposing directions to obtain the spin resis- washers are used to electrically separate the two connected
tance characteristics. segments (Figure A.3).
For corrosion resistance in permanent moorings, typically Careful quality control and testing should be exercised
a polyethylene or polyurethane jacketing is employed. The prior to and during the fabrication of the rope to ensure that
jacketing material should be a high density type. Also all the rope meets design specifications and the final product
wires should be galvanized. Zinc tiller wires are sometimes produces guaranteed minimum break strength as specified.
 i:FESiGi\f ANS F\i\iALY'S1iS OF s”i“A'?iOi\iKEEF'iN{3 SYSTEMS FOR FLt'}A'i'ii\.iG QTRUCTURES A

:-$2
og-Z99
0 its O0‘ K.
.‘Q0
00090
O.
OOOO
9*?
9!

~ .O9OQ
9r$0$:i§:§:,Q,
'e3~:*3’~°
<%%xd¥%$%: as-s-as ‘9b: si2k,:0‘Oe?*g:}QO;':o}Q5.O{;?{¢g°9YQo?,o9-@.~°o'3§o!§.'\a0: a~.'%t~%~: t~a:*i?;'
OOQ
0'0?-0 8 90 -%-<3*’-° Q00
t
wt .Q.O9QOQ:8OO
.:.{:"I*.tQxg?;‘,.
Non-sheathed spiral strand with Zinc Fillers Nornsheathed spiral strand rope

1! 5!;o,:.,oq_Q.g‘Q! -9}.‘
E I‘ O U Q “O ‘I O -

:*0F’oP*0?'oQi'o*Qq{fio90 9; r 5.‘. ‘='.*'" i.. 01.“;-%Q”' 1--ll"? §%f1i':-0.1!-9*‘”,0;%”’g’;1Q:;jfi

Z i.}.I..I..lt. Q ~lg-cl.xi Q" 1 -Q; or--+‘=o 1*:'-;'%!!ei'¢‘!£5#;<;!o-~!'> -
'."6'-'%*‘S95-9 ii" Q . '0‘ =0-9.. 0.1‘
“Ii I .
°"f;35"1i
__
1-‘=2-{@990
.. I

.. Q».-Q Q. -Q‘!

Sheath spiral strand rope Multistrand ropes

6-‘Strand IWRC 8~Strand EWRC

Figure A.2--Typical Wire Rope Constructions

 API RECOMMENDED PRACTICE ZSK

isolation System

Strand sheathing
\\>~
:3.-.'-.

§?§’f;;=-
*

Bend stiffener

Open socket 0

. Hg,’

Isolation bushing ‘C.

ocket pin J’ C C /\

V/'

Handlmg Padeye ‘ I Anode

9 1 ,6‘
_-'Q Q k
-' ,-

\ isolation washer

$
Anode

Ciosed socket
Strand or rope

Closed Socket Open Socket

l© <¥l\_'J__F'>
|I|% Illa.“ |n|; £1 V
Figure A.3--Wire Rope Socket for Permanent Mooring
 Dssioiv ans Aniugrsis or-' Srarionksssiuo Svsrsus roe Ftoarazo Sraucroass 43

A.1.4 SYNTHETIC FIBER ROPE A.1.4.3 Rope Properties

A short introduction to synthetic fiber rope is given in this Urilike steel components, fiber rope stiffness increases
section. Detailed design guidance can be found in API RP with mean load, and decreases with cyclic load range and
ZSM, Recommended .Pmctice for Design, Manufacturing, with load relaxation over tirnc. After the rope has been ten-
and Maintenance of Synthetic Fiber Ropes for Ofifshore sioned to allow bedding-in, and cyclic load and relaxation has
.Mo0rz'ng. occurred for som.e time, the stiffness of fiber ropes tends
Fiber ropes may be used. as segments in steel catenary toward a linear function of mean. load and load range.
systems, or in taut leg systems. The subtle differences Little data is currently available on creep in ropes. A gen-
from that of the steel wire rope/chain mooring systems eral indication of the effects of creep is given by yarn data.
include the non-linear stiffness, minimum tension. require»- Continuing elongation under load may lead to a need to re»
ments, location of fiber rope segment to be away from tension of mooring lines. HMPE fibers show significant
fairlead and seafloor, creep phenomenon, and different creep, which can lead to creep rupture failure. Creep of poly-
handling procedures. ester and aramid fibers is much lower than HMPE.
Factors which may limit the life of synthetic fiber ropes for
A.1.4.1 Fiber Type deepwater mooring and which should be checked include
hydrolysis, heating and internal abrasion, tension-tension
The fibers currently being considered for use in permanent fatigue, axial compressive fatigue, and creep rupture.
or temporary deepwater moorings are polyester (polyethylene
terephthalate), aramid (aromatic polyamide), HMPE (high A.1.5 CLUMP WEIGHT
modulus polyethylene), and nylon (polyamide). Currently,
Clump weights are sometimes incorporated in mooring
polyester is considered to be a good candidate for the offshore
legs to improve performance or reduce cost. By providing a
mooring application due to its low cost, low stiffness which
concentrated weight to the mooring leg at a point close to the
induces less dynamic tension, good resistance to axial com-
seabed, a clump weight can be used to replace a portion of
pression fatigue, good fatigue properties, good strength to
chain, and increase the restoring force of a mooring leg.
weight ratio, and good creep resistance. Polyester rope is the
Using clump weights in a mooring design requires careful.
only fiber rope that has been installed in permanent moorings.
consideration of potentially adverse effects, such as increased
Other fibers such as I-(lb/{PE and ararnid may be more suitable use of connecting hardware and installation complexity,
for applications where a smaller rope diameter is required undesirable dynamic response of the mooring line, and
(e.g., for frequent handling) or for ultra—deep water mooring embedrnent of the clump weight in the seabed. In some moor»
applications. Nylon rope has been used for hawsers in CALM ing designs, heavy chain segments are used in place of clump
systems where high elasticity is an important property. These weight for easy installation and lower cost.
hawsers can be inspected frequently and replaced. Also in
shallow water locations, a length of nylon rope is often A.1.6 SPRING BUOY
inserted in the mooring line to absorb the energy from vessel
dynamics. Spring buoys are surface or subsurface buoys that are con»
nected to a mooring line. Benefits of spring buoys are:
A.1.4.2 Rope Construction - Reduced weight of mooring lines that must be sup-
ported by the vessel hull; this is particularly advanta-
Note: Figure A.4 was intentionally ornittted.
geous to floating vessels moored in deep water.
There are many different rope construction types. Two
' Reduced vessel offset for a given line size and preten-
types of rope construction, “wire-rope construction” (WRC,
S1011.
as used in steel wires) and “parallel strand”, are most com-
monly used. Jacketing should be used on fiber ropes where ~ Increased vertical clearance between the mooring line
external abrasion is expected to occur while in service and and the equipment below.
during installation and recovery. Certain rope constructions Adverse effects of spring buoys are:
must have a protective jacket to keep the strength core strands
~ Increased use of connecting hardware and installation
together, such as parallel strands. A jacket may also provide
complexity.
some protection to soil ingress, marine growth, and fish bite.
A sand filter is often placed between the jacket and load bear~ ~ Potential increased design loads on the mooring lines
ing fibers to give additional protection against soil ingress. due to dynamic response of the buoy in heavy seas.
Fiber ropes are typically constructed to be torque balanced. Spring buoys used with permanent moorings are typically
A torque matched rope is sometimes used when it is con» constructed from steel or a combination of synthetic material
nected with a mooring component that is not torque free, such surrounding a steel structure. A high density foam material
as a six-strand wire rope. (glass spheres encased in a high density foam) has been suc-
44 API Recomruewoso Paacncs 28K

cessfully used to provide buoyancy for deepwater drilling and A..2.1 WINDLASS
production riser and mooring operations. Steel buoys have
been found to provide a cost competitive solution. The buoys The most common method of handling and tensioning
can be built either spherical in shape, using unstiffened dished chain is through the use of a windlass. The win.dlass consists
ends welded together, or with ring stiffened cylindrical bodies of a slotted “wildcat” which is driven by a power source
and ends. Buoys can be placed in line with the mooring (with through a gear-reduction system. As the wildcat rotates, the
a strength member through the buoy) or attached separately chain meshes with the wildcat, is drawn over the top of the
to the mooring through a tri-plate as shown in Figure A.5. wildcat, and lowered into the chain locker. Once the chain is
When using the in-li.ne buoy approach, care must be taken to hauled in and tensioned, a chain stopper or brake is engaged
allow for rotation in the end connections. to hold the chain. Windlass has proven to be a fast and reli-
The buoys should be so designed to have adequate strength able method for handling and tensioning chain (Figure A.8).
for maximum operating depth. During fabrication of the
buoys all welding should be tested with appropriate A.2.2 CHAIN JACK
non-destructive testing. Also, corrosion protection should be Chain jack is a device which reciprocates linearly to
adequately provided. haul-in and tension chain. Usually powered by one or more
hydraulic cylinders, chain jack engages the chain, pulls in a
A.1.7 CONNECTING HARDWARE short amount of the chain, engages a stop, retracts, and
Connecting hardware such as shackles, swivels, fishplates repeats the process. Although chain jack can be a powerful
and detachable links (Figure A.6) are used. to connect means for tensioning chain, it is very slow and is recom-~
together the main mooring line components. Inspection and mended for applications not requiring frequent line manipula-
replacement of connecting hardware in a permanent mooring tion (Figure A.9).
are difficult, therefore fatigue life and corrosion protection
become important considerations. The design of all connect»- A.2.3 DRUM-TYPE WINCH
ing hardware to be used in permanent mooring lines should Conventional drum-type winch is the most common
be thoroughly evaluated to ensure that stress concentration method used for handling wire rope. Operation of drum-type
factors are correctly identified, and that fatigue life and corro~ winch is fast and smooth. Drum~type winch consists of a
sion protection is adequate. lvlanufacturi;ng of connecting large drum on which the wire rope is wrapped. The base of
hardware should be subject to an appropriate level of quality the drum is often fitted with. special grooves sized specifically
assurance. to the size of wire rope being handled. The groves control the
Connecting links such as Kenter and Baldt links are often positioning of the bottom layer of wire rope on the drum. For
used in mobile moorings. They can pass through chain fair- subsequent layers of wire rope, an external guidance mecha»
leads and windlasses and can be periodically inspected and nism such as a level-wind is often used to control positioning
replaced. of the wire rope on the drum. The tensioning capacity of the
Recently subsea connectors were developed to allow con- winch is a function of number of wraps on the drum (Figure
nect and disconnect of two mooring line segments und.er A.l0).
water. These connectors typically have a male part and a Drum-ty_pe winch can be a cumbersome method of han-»~
female receptacle, which are attached to two different line dling wire rope for deepwater or high strength mooring sys-
segments to be connected. The under water operation of con- tems. As the requirement for line sizes and lengths increases,
necting or disconnecting the male and the female parts are the size of the winch can become impractical. In addition,
performed by an ROV (Figure A.7). when wire rope is under tension at an outer layer on the drum,
spreading of preceding layers can occur causing damage to
A.2 Winching Equipment the wire rope.
The type and design of winching equipment required in a
A.2.4 LINEAR WINCH
particular mooring system depends on the type of mooring
line to be handled, and whether or not the floating vessel Linear winch is similar in principal to chain jack. Two sets
itself must initially tension the mooring lines or test load of grippers, one stationary and one translating, are used to
anchors. A floating vessel often has the means of adjusting haul-in and tension the wire rope. Linear winch is available in
mooring line tension, retensioning after anchor drag, and dis- a single-acting form in which case the wire rope moves inter»
connecting individual mooring lines. Besides, a floating ves~ mittently as the gripper is retracted to begin another stroke,
sel is often used for combined drilling and production. This and in a continuous double~acting form in which case two
will require the capability for finite surface positioning for translating grippers are used alternately for continuous
maneuvering the risers. This positioning can be achieved by smooth motion of the wire rope. Linear winch is most appli»
paying-out and heaving»-in mooring lines. cable in a permanent application when high tension and
DESKBN AND Anzuxsss or STA'%'iONKEEF'iNG SYSTEMS FQR Fioarzno Sraucroses

__ _ _ _;_ ,..»——————---

iii l|/
*1?“

as
Steel buoy pt‘-.l_ ........................"

H Buoy Attached to Mooring

._ _ .. .. .. - /’'
__ ¢~* - p - - \\
- X - —-- - —--
- --—

Buoy In Line With Mooring

.11..

1
l

tr J ‘H
Buoy

W clevrs

Spelter
socket A§a »~@
BUOY
tnplate i

Spefier “
socket A
Synthetic Foam Buoy

Figure A.5---Submersible Buoy Configurations

 AP! Racorvrmauoso PF-2ACTlCE ZSK

_ pppppppg A p M p p p g rt

A. Common link B. Enlarged link C. End link

..i ta a aA
“pgggggggif i_ t_l jjjjjjj7* iH' __ I HM“"lii Lm**Wfl_E
t e_ ......W

D. Joining shackle type D E. Joining shackle type Kenter F. Anchor shackle type D

. _ i_______________ ._

" (Iee e e F I“ “ ----L

_.l_L
L "il c‘

G. Anchor shackle type Kenter H. Swivel l. Swivel Shackle

N :' 2----~' ~ --A | 7' ' i-r---s.--».........».....

- all l I '

fl i
K. l -""“"'. i
~ i l l _ ' |_ i I l

l l .. Jvnniuur_______
LI
ha ll_
L.‘
manua---.. . . . .:":\TI"2
'31:
':-=' .1--'
T‘.
l. I
X
l.
l- ~==:::.-P‘ i
‘ . t
I
M..
I
\
'1
Y “Iii,
-

as;A-~as .

J. Swivel Shackle ASW K. Baldt detachable anchor connecting links

Figure A.6--Typical Mooring Connectors

 Desren AND ANALYS-BS or Srarionksssare Svsraias son Froarsva Sraocroaas

<9

.
-‘I: l,
i
ti

Delmar Connector Ballgrab Connector

Figure A.7“-Typical Subsea Mooring Connectors

 AP! Rscoaaaeaoeo Paacrica ZSK

Windlass
/
/P

/
""“-

. -
I|........
W
l"'|%i!!;€ . up

ii!
l-I J
4.

L . rt

'*--"'---_._'?"'._"!?:_u-1'-.q_,._, ,_

chain
locker
___‘o.'_A .:-‘_1"M4_-u.-—u-_.v-._:-._-|i_:—“.‘"_'.-“"I;_~,?\-_.._'--_.-M.I*"-.,
n*‘r-;_~fuoi"__;:_
5 '--T,,,;,
_7--I--.
-..-,-
_ ._"1"_x-'i;-.'-"_'“-_."';:._-'\?l"—I":- ‘_.1».-.‘_.'i _-“"'—'._; .-"I .1; -
1'
Fl?
._l;

\ ‘

Falrlead

r n\\ \h\\

as I

Figure A.8—Winching Equipment for Chain

 D E S w N A N D A N anL HY HmS OF 8TM Q N K E E _P W G S Y S T E M S F O R F L GA_i W G S T R U C T U R _t S 14GU

\\“W

_ ”_ _A _
U__ J§U>fig
Igsg IlIi‘l :1“
Mb
Hu
_
W\_p_ qL§“p Ii
_p:
_I ‘_-
W
_ _‘

hp __z‘Im§-I

@
_1
W_g___ ____ l Fr m__1_WW
i_ L w___ L
W
N
1 I Mt JW_
myp
__
I Maw _ M r_ E._
3_
H__I
L
i‘‘1 1‘V‘
EgoE6
_ Uv_Ug
gcwm%_
@_ ®_g:$OH|m_ <9_)_m_®“m_
Ego
®gg__wCEOUEE EEOV_%_ Q‘
\k\\
\\¢\
C_@£U
gcwmC_ UC®m_
Ugo QEgo
kg‘
E£_W>w
V_Ofl_
x\
\\
X‘
__\“\\
_\

' l l

E_ mE_@>E:
_ w WO_ O
EcmgaaUQXE

gnaw
fig
dmmi
QigI‘iIg___nq

M5‘MI
__/2
‘ “wig
ii
l*1
i
‘HMwpM
_m___W'
_ Q
Elrm\
_p_\g RM/_
7 5 %
/I
I_
_____'
Mm M M M %
/‘FIT

D\ M M M M M Q‘
ll
0 __m

Ml_@_ _
9 “N

- _¢—_V7
Q
Q
Om
hm;Q

_I _J /
p__
__O
u_

M?
Mx UM
Q
‘EH _L__
__w
H
UC

%f
Q
F_U
C
_>>
A\\_
g Q
Q

i
_ ‘
>
“Q? _'
E
3
LO
£0
__/X

Aw“
-* H ®
l__
m
_H
‘X
my
N
 Dacron ano ANALYS¥S or Srarronxessrno Svsraus FOR Froaravo Sraocroaes 51

large-diameter wire rope is required. A take-up reel is neces- A.3 Anchoring System
sary in this case to coil the wire rope after it passes through
the linear winch. A winching system using linear winches is The options that are available for anchoring floating ves-
sels include:
illustrated in Figure A. 1 1.
' Drag Embedment Anchors
A.2.5 TRACTION WINCH - Pile Anchors (driven, jetted, drilled and grouted)
Traction winch has been developed for high tension moor- ~ Suction pile and Suction Caisson
ing applications as well as for handling combination mooring ~ Gravity Anchor
systems. It consists of two closely spaced parallel mounted - Plate Anchor (drag embedded and direct embedded)
powered drums, which are typically grooved. The wire rope
makes several wraps (typically 6 to 8) around the parallel in selecting anchor options, consideration must be given to
required system performance, soil conditions, reliability,
drum assembly. The friction between the wire rope and the
installation, and proof loading.
drums provides the gripping force for the wire rope. The wire
rope is coiled on a take-up reel which is required to maintain
A.3.1 DRAG EMBEDMENT ANCHOR
a nominal level of tension in the wire rope (typically 3% to
5% of working tension) to ensure the proper level of friction Traditional drag cmbedment anchors (Figure A.l2) were
is maintained between the wire rope and the traction winch. initially used for mobile mooring operations. Drag embed-
This system has been favored for use in high tension. applica- ment anchor technology has advanced considerably in recent
tions due to the compact size, capability to provide constant years. Engineering and testing indicate that the new genera-
torque, and ability to handle very long wire rope without tion of fixed fluke drag cmbedment anchors develop high
reduced pull capacity. holding power even in the soft soil conditions. High effi-
ciency drag embedment anchor is generally considered to be
A..2.6 FAIRLEAD AND STOPPER an attractive option for mooring applications because of its
easy installation and proven performance. In fact, many exist-
Mooring lines are subjected to high wear and stress at the ing permanent and mobile moorings use drag cmbedment
fairlead and stopper arrangements. The long tenn service of a anchors. The anchor section of a mooring line can be
mooring system requires that fairlead. and stopper arrange- pre-;installed and test loaded prior to platform installation.
ments be carefully designed to minimize wear an.d fatigue.
Mooring chain and wire rope are often stopped off at the A.3.2 PILE ANCHOR
vessel in order to take direct mooring loads off the winch.
A pile anchor’s resistance to uplift and lateral loading is
Chain stoppers and wire rope grips used for permanent moor-
primarily a function of pile dimensions, the manner in which
ing systems must be designed so that the stress concentrations
the pile is installed and loaded, and the type, stiffness, and
and wear within the chain or wire rope are kept at acceptable
strength of the soil adjacent to the pile. Horizontal capacity
levels.
can be increased considerably by adding special elements
Fairleads sh.ould provide sufficient sheave to rope diameter such as skirts or wings to the pile top. Pile anchors can be
ratio to minimize tension-bending fatigue. Typically 7 to 9 designed. to develop high lateral and vertical resistance, and
pocket wildcat sheaves are used for chain. Sheaves for wire be very stable over time. Piles are generally installed using
rope have diameter (D/d) ratios of 16-25 for mobile moor- driving hammers although other methods such as jetting and
ings, and 40-60 for permanent moorings. There are other drilling and grouting techniques have been used. Installation
devices which provide attractive alternatives for fairleading ofjetted or drilled and grouted piles can be handled by a con-
large diameter mooring lines. An example is the underwater ventional drilling rig without major modifications. However,
swivelling bending shoe shown in Figure A.ll. This device, disturbance of soil daring jetting and drilling operations
which is used initially with wire ropes, incorporates a shoe to should be carefully evaluated.
rope diameter ratio of more than 70 and a special high density
nylon bearing material secured to the bearing surface on the A.3.3 SUCTION PILE AND SUCTION CAISSON
shoe. Replacement of the material is possible by slacking
down the mooring line and removing the bearing material A.3.3.1 Suction Pile
which is bolted to the bearing surface in sections. Industry Suction piles can be used for large deepwater mooring sys-
experience indicated that this device can also be used for tems and can be designed for very high mooring line loads.
chain (Figure A.9). However, operating with high chain ten- They are typically tall steel cylindrical structures with or
sion and frequent vessel move should be avoided because it without internal stiffener systems. The cylinder unit is open at
can cause serious damage to the bearing material. the bottom and normally closed at the top (Figure A.13). A
F0WW
2__ A WW R E C O M M E N D E D P R A W E _E 2 S K

QO$@ufi30%
_Ufi@_ M fiH_

~04 1AIég2&31mmn magma@525

i ‘ _€_2mE
__ ___
W_
W___;
um
_:7*
_RWTi:
_ W_
_‘U4_h_
W __igs ‘___i5_W___km\
__‘i____ ___”fix W\%\W%_1____\__:___\W\_\_\
__%\__% _\_ _ :N
___\\
_ _7”_ __H_“HT
___mh_ Wi%EfA|HM_%I DE
mcmhmgMwgg
l X _é
_ mCEEOO
£m_ mEw

_f _'_h_h_W__ U_Hh_ _%
W
_/f
_L_\_‘__V
“_ ” W___[__f__‘_%_“___
$\J_
M%
k
%If
\
_ r____“\
_ ?_h_Fr_’

WW_flfi

mg
U©@5@E
r‘gswmE_
9;}6%
®30¢
_8_OE>>E|m_H¥_: < H_C®m

\
\ if \\_\_

MW_w$E
_
‘_
__
\\_ \L _ l[_liT\_
‘ w
_E_Q_
_
_ _ _mu_N___
___
I_ ___n__
“__

rm$% ”¥Hyfin% fi¥&@%£fi€ig$m $m%fim%fim%

sum“$®E|? _
$1
H
_ W_“__*In_$____
§F
w
%”$@?wW%H
{mWm¢% WE _
W_ M% H
X_g__
Emmmhofim
mgugm
mosm_
WHlhi:yFw,‘W
%
w>mmcwi ,6Q %
_\_m%_$m_
%%%%N
%%
_HH_‘__
 j\ i_
A1 % %%_

X005
tnwfigfi

\\
X ‘ xogw
5*“/\\4/Om
(J
%\\
_% _x
_E

\\j
__\ __

Dom _M_
_x :8
Ea0%E2§_Eg0%
Egg

_ __
@
\
Q8
gm
EEO
A 1 _H

\
/
®_%gm_ O£U_ & _
_g‘ OQEQE
ago
_'E QO5©‘E_'UiN_g‘UE<mC_<

J‘ Ll3 !‘

‘ ‘ ‘ _‘_ _ :‘ _I ‘

%%%
_% E_gE%n_%Q_ ; H%_ _%_um

\\
\

\_
 APIRECOMMENDHJPRACHCE 2*.) Q2Fa

c::il“i%§IL
c::3=i~'(‘EI’

i
K

\ \

¥ 1

l |-un u Uln,u ‘"vl

\\\Q
P

_guru!"Awlunlunlr \
\
M
1 @117

Padeye 1

i I l‘ .

F4fi%4‘1lIti-lIU5#ITU"I I
I
1‘
1

Figure A.13-“Suction Pile

 Gaston Ann Anarvszs or Srnrionnseelno SYSTEMS eon Ftoarsvo Sraucruaes 55

su.ction pile is installed by first lowering it into the soil to self» A.3.3.2 Suction Caisson
penetration depth (i.e., penetration due to submerged pile
weight). The remainder of the required penetration. is Suction caisson is a suction embedded anchor that is rela-
tively shallow in height and is designed for relatively small
achieved by pumping the trapped water from the inside of the
penetration. The suction caisson’s submerged weight makes
suction pile. The pressure differential thus created will result
up a large part of the anchor’s vertical holding capacity. A
in an additional driving force on the anchor top, which will
multi-cell concrete structure with a large footprint and a shal-
drive the pile into the soil. As the penetration increases, the
low skirt penetration would be an example of a suction cais-
driving force needed normally increases, requiring a grad.u~
son (Figure A.l4). The vertical load capacity is mainly from
ally increasing differential pressure.
its own weight plus possibly some skin friction and internal
After penetration, the water outlet is normally closed, and suction. Horizontal load capacity is generated by skirt pone»
a suction pile may achieve substantial capacity to resist ver- tration and friction between the soil layers being sheared.
tical downward loads, horizontal loads, vertical uplift loads,
moments, and combinations of these loads. For suction piles A.3.4 GRAVITY ANCHOR
embedded in clay and with a closed outlet, the capacity to
resist environmental loads is governed by an undrained Gravity anchors are deadweight anchors which commonly
shear failure in the soil around and beneath the pile. The consist of concrete or steel blocks, scrap metal or other mate-
capacity depends on depth of skirt penetration, cylinder rials of high density. Skirt penetration is obtained through
d.iarneter, shear strength in the clay, shear strength at the self-weight penetration, and the design uplift capacity is
clay/wall interface, the load inclination, and the location of dependent on the submerged weight of the anchor. Horizontal
the load attachment point. In the case where the top part is capacity is a function of the friction between the anchor and
left open or retrieved, or for long-term uplift load compo- the soil and shear strength of the soil beneath the anchor.
nents, pull~»out of the skirts may also be a possible failure Gravity anchors can be used for small mooring systems, but
mechanism. typically are not used for large deepwater mooring systems.
The holding capacity is generally greater if the pile is pre-
vented from tilting. A translational failure mode without tilt- A.3..5 PLATE ANCHOR
ing can be achieved by lowering the load attachment point Plate anchors were initially used by the US Navy for
from the top of the anchor to the anchor wall at an optimal anchoring of fleet mooring buoys. They are installed at deep
depth below the seabed. The location of the optimal load penetration beneath the seafloor where the generally higher
attachment point depends on the shear strength profile, the soil strength allows the use of relatively small plate anchors
shear strength at the clay/wall interface, the load inclination, for high mooring loads. Plate anchors typically have signifi-
the submerged anchor weight, and the depth/diameter ratio of cant vertical holding capacity. This allows the use of taut leg
the pile. The optimal location is typically two-thirds of the mooring systems where the anchor line can intersect the seaf-
length of the pile downwards from the pile top. loor at significant inclinations. Plate anchors can be placed in
As suction piles are shallow structures compared. to driven two broad categories: drag embedded and direct embedded.
piles, deep soil borings are not required, but more detailed
soil data are needed at shallow depths than for driven piles. A.3.5.1 Drag Embedded Plate Anchor
Suction piles have mainly been applied in cohesive clay type Drag embedded plate anchors are embedded to deep pene-
soils. Suction cmbedment penetration through thin sand or tration in a manner similar to drag anchors. During installa-
granular layers may be feasible, provided the suction pile tion, the anchor is first placed on the seafloor, and as the
design takes this into account. Penetration in non-cohesive anchor is pulled along the bottom, it penetrates the soil. Ini-
granular type soils requires special considerations, which are tially, the anchor dives more or less parallel to the flu.ke,
not covered in this document. eventually rotating such that the target penetration depth is
Suction pile length to diameter ratios may range from 2:1 achieved. Following the cmbedment, the anchor is “set or
for stiff clay soils to as much as 7:1 for very soft clay soil. keyed”, i.e., the anchor fluke is oriented such that it becomes
Suction piles are often designed with large depth/diameter nearly perpendicular to the anchor line, thus providing high
ratios in soft clays, since the upper part of soft clay deposits horizontal and vertical holding capacity. These drag embed.-
provide limited bearing capacity and skin friction. A suction d.ed anchors are often referred to as VLA, which stands for
pile can consist of two sections, an upper driving section and }__/ertically _L_oaded Anchor. Two VLAs are commonly used by
a lower pile section, which are connected during installation. the offshore industry: Stevmanta and Denla. The Stevmanta
Once full penetration is reached, the two sections are disen» anchor uses a bridle system to convert from its installation
gaged from each other and the upper section is recovered, configuration to its plate anchor operational orientation,
leaving the lower section in the soil. The upper section is then whereas the Denla anchor uses an articulated shank (Figure
reused to drive other anchor sections. A.l5).
56 API RECOMMENDED PRACTICE ZSK

Tendon

Tsndon
receptacle

Bailast
(2 res": are) _
-I-I= -"-'.;r={; .'{ ,-53 I"_’.1;'F' ' 315;. :‘-zfi '7,"-:5; r~,f<?' ‘;-3; 1'=j'...
_
:I;I_-';§;I-. -1:;-.-_;;;"--_.,;; >.»s-.;,g-' ’ - + ..@;;;-- _¢‘-.-
-I. - :{=..-:-;-
-_ ‘M
__,,_q. ‘ii’__' :-:-.
._-V» ,-’ -F?- »{-_
-- .- .'11;)
.-\ ix
- _ _ 1
H, . Q5 \-r. -.-.;; ;. n -- - -\- _.:;:;.
,5’I :-:;:;. -
\. '\ -\ .7 ’
. Q55 .393 ‘%E:E:'.;:f:>';;§f:-" Q1252: .;:§: <9;-n1
%-;-:-: "5" -_,_ +,5-:h- ..;-955', '99 .
.4‘ ‘V
'9'$5.i\q-- --\
’
1"" _
_ ‘\. _.}b J
. .-W. .. . - -- -
. vs:
.‘-€-:=:=- T"-'._:'-:1-",=?;. 5* -15:; ~l;;5_j -' 3.5;; _>:-1);; 2&3. -91¢-';z=. i=*;. - .
:;;.:: .-»_1;' -:=_-.-,5 ' .-.-_:7;T- '-221' -:[:3" '_§;:§ 4§i:;f__¢':l- ¥3,;:§:5'

.-:§:7: ..' 31‘?

4

F53.-55:: - {vi .1111 .:I:I- ‘ J:::- ' '1

‘Q~:'%;;_LI.§.;_€; 2:-:5‘
--¢§g:-"Ig;3E=- ‘€:I<§¢ -.I:="§;;;- +.i:iE;= . .¢:;;“¢§Z'
. \ VA/J
‘M1? X
_:-S .p. -:;.-. 25.41.?"-;;: 5;. $3; 1-:;:-J? .»..;.: . 4; ;.1_. .;;-_-kg 1,._i..- ;.;-..
:- - ,:-.;-;- _-V. . 9'.-;;:_ ,3 /--.
' -’:-- '.’--
_;§<
' 25>‘: .1_,;:5- ' 3.15.. I-:-:__;6;
<. . Y.>_7_<_ <62?
,;:' .'1'11 ;:;* ' 55’
If ,.;:;
V5],
,:v
,6-.\ '-:5. . -'
~. %‘~1113:3-
"~.*__¢;;-
as
’;_.
*2
I ' 1‘:1 *,s1:=""'
I-.;;-I;
’-.-. -. "n 5-:-.551.
-:-,-. .;.;.: . .
.._... , -_ "_:=E;;- I-'.1.=i‘;E5:;
¢/
-:-:-
-.’5-_.;.-
. W . .. :§"'$ _. -3‘ "'-§‘-- '4‘. -.-1
pa; .-13>.-r; -'__:&:=. £‘=*-.¢:<=- ' A-_ &"==-.e;<= sis’ "‘=".--:5==- [:55 ;‘=‘1-.a‘~Ii= 1 c
.V Pafapat
_- -‘P "2 -. ....'\'.‘.
f-:- ,_;. ‘c.-_ 9'1"» .
.->5 .-. J: > ;i"’P‘..:-=‘:'..¢z3 "
“°-1"‘=’§‘ l%5"‘?’I'.<-55.'.I¢£ *’"-‘5‘?';- . _-.;.;..- -- .- z;.;.-_ -::-,- 5} 1:?-.;-.;;-' _'.;_;,’ ._ g 2"» _1\ _. .1 , -!i:i'-,g+j:-
-'.::-."’:I' .--: ' .11-1-.2;-:-. I‘
,1. I¢*£:- '2: 5"'J,:, "-:-:- . »/----= '~->»¢:~ éfr»"',; 1,;.;.~_;-.;=; L; "¢?€.5;513'-.5 -'5/.:'-:1:-1: 2-:v_:'I:I' -: 43:-'_;-1:1:-._;;-. ¥fi:'1§I:I-'f_5" ‘It’-“>5
.- /~:,"""';'i:..
¢§_¥:‘3-:-_:- -,;:::- "9.1-1;: 1275.. 'I:I:-'{""'..-..
31:: !:<:;:?-"-:;:- '
';_-:1: "$§~‘:' ‘PE-'
..;'- - %-' "-.:.l.
.-Y-in ..
.. .’_‘
>1:/,»_====:.:-*¢¢ ’ » ’ é i = ' § as*:;;;;- =>'::=‘--= £4255; - . ; , : - _ <; ; = ; = =' r=,_.2/1.11.; <5a':»23-+'=;i&="1:;.
.»;'__.:_.
» ""'f".=;iEE=_j./er."
_:~ o __/1; '"=-'-.-1;‘!-'_.,.==;= :-";'
1 " ' 1. .- /-- ' #2 "'1' -"i: -' 'rI'. '-I-1;-1.; > 5-:-.
J’ /A
{:2 ’ J:=£ 5'.-.¢ ' " -‘-:I- -* -' .~:-.' ii ;f:‘ ‘ -:-:-‘ ‘I55 ‘:1 ' J-;-I iii "..<i;5= £1»..:'-P2?’2;=.f1*,-:=z:" _'-‘mi-'~'.=:=:-.'=‘=".§.i=sé ;"=‘=",=2=-'<ar=I.=¥='§=i2
. ,»- . -. -
. - _.;'>
" ' v.5‘.- I .;I;I'-._/4
If \___
-_: _é_
-' 1:". __|_:_I
.-_-:i.3- _.§:_: .. \-,_-.:_ X9 “1.2:
fa .. r-.1.:- __j;; -.:\.-_ ‘.55: :3-,3. 1 . 3% .. .-\,\.E._: 3:5: -4...
__;|,_:; . IQ, .mI¢.;.:r'_q;_ -.:\ A19 -:_|.;.;.E.:_: ‘,3, r\'|:|"'|:|:| \1;.;.- ._
-' ".'-:;I==,; §i'.+I,'-'-I-I-.;- $54‘--:-_.-; $3-1.:I-1:1-.~ \>
$1.-__~I-';,= 1:5!’-_-I-1:1 ,;- 4
>51-1:!-,>; 2-5: ‘A5 *.;::t,'»5;.=.,.; . ' - - 5_<-:*->:=:_4- §;:;:_-=-:;-- ".. "-’--
-5.. .1W > . ',-§?»_.-:-..::1.-' .\'-..
. .91‘
'r .- .-:-.-,-.-' 1/.- -§,.-"-I " --av. .- :1:-:1: " :-:--. 1115-..~:-.'-
‘:1-Ln 4-.-. -
-+:..
’»2+'r1u|lnul::_1£ui1_____)u1nl1____J&.'\Lz;'_
===-====;'f;s*,- i-§;:=# -;:;="I1i=i'= 1'-§>' 1:1:-‘=22: -‘»1==i%+ .-1-5;-,<.~
.. I 3"} =i&§=?=-.-;=:-is .;:::"=J;za
+ 5‘ J-:55’; i E5:=i=iw-:-"E5¢"+ - ' - ' .~ \<~:-.;‘~"i=-=.=r-§;=;< E;15=5:.
"f;:;:- '53.’-:-g/_"”
<5?: -£1:;:c_1;:;.{;;:;:- -.351-.._-:;'._“',+,>; -:;:;:;:-';;;:-gfi? ‘ii.’-‘:1?-.;:;:-4:-‘ -:-:*. .-:- ':-fr. 31;:->_:-.11 .-,

S 6 ab ed i _l}§Y7{§__E{{{—}}}]____E—I—I{{{{§§{{;] _[ZIZZZZIZIZIZZZZS___ F_'_'_'_'_'__ "'1'! ILIZL5ZZZZ{{{55{{{{{{§Li_

I .7-:3-------3
i i ________E---------I-I
jiiff; . J,». . _’ ;r__:s-j
. ..
.
*§;3»i?:-' '1-Zea; *;j=lI:’ F? ’-".“5fi-'".- ‘
._ i=5-:-.~_:-I
5:-2-*9. J >;:1=¢>< " -.%>:'===-
. /_:¢;;._;I_::‘ -;;:;.‘a_1;.:¢._?
I F 5:
:;».:.¢.':p__,_:A- 12-.1§»'< gié. .+'*--
-

1-!
&:-
'3-'lI§s~ Q31
.5:-13;}; ._ 11'-
:2; -
“L1
I
I
i

__ __run___________ nnu1____________ ___ 7 ’ ’

Installed Anchor
Front View

fi\

@5000

Anchor skirt

' \/ l _ Anchor skirt

Stifiensrs Y V V /
(beams) - I i V

/O
Towing padeye

Section 1 -1

Figure A.14---Suction Caisson

 DESKBN AND i5\NAi_“fS§ S OF ST.‘3tTlC}i\iKEEP§¥‘~iG SYSTEMS FOR F‘:_ o/moo Srnocrones £37

Normal loading mode ,

‘X
u
‘-
Installation mode
M

—-—-»-. - . -,.
'\<\-\“‘J-I

man“-_ J
_-.)i_ _-___-
_ %M w,.¢\_;'

Llr_fi*j *“—~—~_ ma»wW”*i;-“ 0

Bruce Denla Mk4 -u

-. . _ 0
H-T.»
' ., .m; _-;_I¢_Z':(;"'
- »».._

"‘“ - - - . ._

-._ ._, _ _

e5

g_,,,_W ,

/
”\
rs/F’
, .154? .K:
Normal IOQGIGQ mode \
,.;.<;,<f\\2 gtW
’i;w E
"I § /1 iii}
—-_ F Kg! I
"‘\ i~!M
ynfl
xA--. ’//

'\ ‘-\\ {l‘\

.,_, K .. #‘~ _:__-t_.
_. qx_
I '"}/ >,-3-kinX=~..
/ -‘___-.

/ /41/
*’/5'/y .~ H. -.___-.,_

“'\-.
//////' .4’
f.3?
////F //I
.1 / _-~\\\}\~?'~'-1
\_':‘___,/

. ///I’ // ) §\\\\,\
_ . / ;/ ///'>,Ib
. l // “I:-),:,i';-._.i;j

.. ff ‘xix;
w 1

p , W, /i p Wviii-Z“-'~ ~ t
I 6)<k\;~iflflrrt
_ -- X1),U '2 i

/I 1 .-
-
; . /Ll
Stevmanta . /I
3' . l' -. ifI; nstallation mode
-
.g. 2
,j b
' -. . I
/K ’fl-I
"'r
,p Xv I _
/ l

. '
§\

Figure A.15--Drag Embed ded Plate Anchor (X/LA)

A.3.5.2 Direct Embedde d Plate Anchor wt'tht its tip slotted
t for insertion of a plate anchor. The suction
follower is immediately retracted by reversing the pumping
Direct cmbedment of plate anchors can be achieved by sue» action once the plate anchor is brought to the design depth, and
tion, impact or vibratory hannncr, propellant, or hydraulic can be used to install additional plate anchors (Figure A. 16). In
ram. The section embedded plate anchor has been used for the SEPLA concept, the plate anchors fluke is embedded in
major offshore mooring operations. As an example, the vertical position, and adequate fluke rotation ' is achieved dur-
SEPLA (§uction Embedded, _E_I_,_ate gnchor) uses a so-called ing a keying process by pulling on the mooring line.
' follower, Wh1Ch
suct1on ' 1S' cssentia
' ll y a reusable suction anchor
5, 8; A P R E C O M M E N D E D P R A C T E E 2 QU K
$1
K‘fi_

2_M_5>_
__ _ M”My W W _p_H_W0W8 “XW0 ‘J fwd UVfi__ _"
“U _ __ +_ ii?_
71!I _ _*_ ~_| _g®Eg
ENE
B
\@§O_§Egujm
pEmwcga_@M%%%___
\
Q qawmp
Em_U%n_ pW392E_§_%w_
\W\\\\
v UmgO§
N5p _y _p?_@__§®>OU@__
y‘
_ ‘__ \___ __wmm®>
mE_H_
mm:
CODm__BSw:WB_y‘_U_\_®
m _ \\\\\\\\\_
_ gfff.‘ M
/f/0*
Iif
JW_ \S?
\ _\T \/0*__ \ __\ \_R__N
“‘\IxWIf_I<\W‘x0“,\\\_“\NW fig
E
p\\I§\pp_@__;_®Q__m§‘ /1/_'IT’ $\‘pg _\R
_ W _\I
/t_U /I\_\\\
X “_\\\\
fig\ H‘”/
m/IWxyj
/z_ _
\H_\ _\\\_"\___
2;
\\“\“:\\
m%8_>_p \\
M___
\ _\
\ __
h\M/
\r
__
h_
ff
\\
\\\\v\\\

?_®W>OU_®W_ X
/_hg uWmpw\otg\umi
R
_ If}! “W
X
57
Wxx;
{fig;
or/I _
\\_I,
\\\
/___,
_ /‘_m_5E®%QC_EEOEE
WEgmwp‘ OE“Max
a_'sfi@W_EOQpgWwm wmg O_mg
H“
/I
p_'__/Ab?
ff /_f/

Q
’//
xcsn
I g€_ SmC_

I;
_\
I’

“X
I’1’
I

_i8n_CQmwv§OC<%_&
g_O:gmE¥w$_<9_ 9H_
X/
_\___
/

L____\_

gmggw
h®‘_50_ ou_
\mO_‘_UC(®H©_n_
 APPENDIX B--RECOMMENDED WIND SPECTRUM

B..1 Basic Considerations fig Z 3600 [S],

As discussed in Section 3.3, fluctuating wind can be mod- .2: I elevation above sea level [tn],
eled by a steady component, based on the l-hour average
velocity, plus a time-varying component calculated from a t e wind speed. averaging time period, r< 3600s [s].
suitable empirical wind gust spectrum. A number of wind
spectra have been developed from various resources, such as and the l-hour mean wind speed U(z) at elevation 2 is
given. by,
the Ochi, Davenport, Harris, API, and NPD spectrum. Cur-
rently only the API and NPD (Reference Bl) spectrurn are
commonly used by the offshore industry. The API spectrum, on/) -= U° [1A + c1n(-ii]
10) rs ' 2)
which was published in earlier editions of API RP 2A, has
much smaller empirical database than the NPD spectrum.
with
The uncertainty of the API spectrum is addressed. through
specifying a range instead of a single value for the dimen-
sionless peak frequency. This results in a spectrum defined by
c I 0.0573./1+0.15U, (B.3)
upper and lower bound values. In the latest edition of API RP and
2A, the API spectrum was replaced by the NPD spectrum,
which was also specified by the draft ISO standard. -0.22

The recommended wind spectrum for this document is the I,,(2:) e O.06[1+O.043U0](%) (a4)
NPD spectrum. However, this spectrum may have significant
uncertainty in the region of long periods, say over 500 sec- where
onds. Por responses with a natural period longer than 500 sec»-
onds, the API upper bound spectrum can be considered. A U(2:) = l-hour mean wind speed at elevation z above
sensitivity study on the effects of API and NPD wind spec- sea level [m/s],
trum to global responses for spar, TLP, semisubmersible, and
U3 ="-= l-hour mean wind speed at elevation of l0m
FPSO in various water depths can be found in Appendix I,
Section 1.7. above sea level [m/s].
It should be noted that the NPD spectrum formulas pub-
lished in the 213* Edition of API RP 2A (Reference 8.2) and B.2.2 WIND SPECTRUM
its Supplement l published in 2002 contain errors, which The NPD wind spectrum describes the energy density of
have been corrected in this document. the longitudinal wind speed fluctuations at a point. The 1-
point energy density is given by,
B.2 Equations for NPD Wind Spectrum
The Norwegian Petroleum Directorate, NPD, [B.l] wind
profiles, gust factors, and spectra are all defined by a single
parameter. The defining or characteristic parameter is the l- SNPDU)
M 3 Zr-1 (-3
U6 2

10
Z

l0
0.45

(1 +}Op1468)3_5,,-I (B5)
hour mean wind speed at 10m above sea level, U0 (m/s). Note
that the equations provided below for the NPD wind profiles,
gust factors, and spectra assume metric units. where

B.2.1 WIND PROFILES AND WIND GUST SPEEDS SNpD(fl Z is the spectral energy density at frequency
f [(rn/s)2/Hz],
The maximum wind speed, in 1 hour, averaged over 2‘ sec-
onds (t < 3600s) at a height of z meters above sea level is f I frequency [Hz].
given by,
and
u(z, 1) = U (Z)[l W 0.411,(z)m(z/i0)] (13.1)
~ 172f(fi)2/3

where
f K """""""'"""""—
U0 3/4 (B-6)
u(z,z) is maximum t-second averaged wind speed in l
hour at elevation z above sea level [m/s], lei
 60 APi RECOMMENDED PRACTICE ZSK

B.2.3 COHERENCE SPECTRUM B.3.1 WIND PROFILES AND WIND GUST SPEEDS
The 2-point coherence spectrum describes the squared cor» The average l-hour wind speed at a height z above sea
relation between the spectral energy densities SQ‘) of the lon- level (the l--hour mean wind profile) is given by,
gitudinal. wind speed fluctuations of frequency fbetween two
points in space.
The coherence spectrum between two points at levels .12,1 org) 2 o,[ZiA)m (13.10)
and 2; above sea level, with across-wind positions y1 and j/'2,
and along-wind positions xp; and x3 is given by,
where
U(z) K l.»-hour mean wind speed at elevation z above sea
c0hNP,,(/) e exp{-5; lid (13.7) level [m/s, ft/s],
U0 K l-hour mean wind speed at elevation of 10m
(33ft) above sea level {m/s, ft/s},
where
2: 3 elevation above sea level {rn, ft],
C0hNPD(f) 2 the squared correlation between spectral
ZR X 10m (3311) K reference elevation above sea level.
energy densities S0) for two points
with The wind gust speed averaged overt seconds (1? < 60s) at a
height of z meters above sea level is given by,

A, 3 Oi,/i"ii?iZ;“' (B.8) u(z,z‘) e U (z)[l + g(t)I(z) (B. 1 l)

where
and
u(z,t) e the I-second averaged wind gust speed at eleva-
tion :2 above sea level. [m/s, fifs],
Z, ...77 -7176“ 1
(B.9) Z e wind speed averaging time period, r < 60s [s],
with
and with the coefficient Ot, p, q, and F and the distances 5/ are
given in Table B. 1 below.
g(t) 2 3 71-1n[®0'6] for iseos
1
(13.12)
Table B.1—Coe1'ticients for NPD
2-Point Coherence Spectrum and

1 5 J5: qr Pi P2 5 A Z\\ ~—€}.i 25

.r 777 7 7 7 O.l5(;_-J forz.§2,,
1 Ix;-x_;| 1.00 0.40 0.92 2.90 1(2) == $275 (B13)
2 1y;-y,1 1.00 0.40 0.92 45.00 015(5) forzézs

3 12%| 1.25 0.50 0.ss 13.00

where
The two points are 2:3 e 20m (66ft) K thickness of the “surface layer”.
= at levels 2; and 23 above sea level {m],
1- with across-wind positions y,1 and y3 {in}, B.3.2 WIND SPECTRUM
1' and with along-wind positions x; and x; [in].
The API wind spectrum describes the energy density of the
B.3 Equations for API Wind Spectrum longitudinal wind speed fluctuations at a point. The l-point
energy density is given by,
The American Petroleum Institute, API, [B2] wind pro-
files and gust factors are defined by a single parameter, the l-
hour mean Wind speed U0 at 10m (33ft) above sea level. on 0) e "W (B14)
However two parameters are required to define the API wind 5,(1+ 1.51%)
spectrum, U0 and 0..
 Dasion amp A1\1a1.vs1s or SrAr1o1v1<ese11~1e SYSTE¥‘v¥S eon Froartno Srnucrones

where B.4 Comparison Plots of NPD and API

1514P100 =1 is the spectral energy density at frequencyf Wind S p ectra
/S)2/HZ}
(ft/S)2/HZ], See Figures B. l., 13.2, and B3.

I frequency [Hz],
B.5 References
with B.1. NORSOK Standard: Actions and Eflecrs 117003
6(2) 2 K2) U(Z) (B 115) Rev.l, Norwegian Technology Standards Institution,
Oslo, 1999.
and
B.2. API RP 2A-WSD, Planning Deszgnmg and C01/z—
stmcting Fixed Oflshore PZcnjfora2s- We; irzrzg Stress
1;, e 90(2) with 0.01 <'" 01. <r 0.1 (B16) Design, 215* Edition, December 2000
Z

For measured wind spectra the average value off}, is given

by or 2 0.025.

Uo(i hr, 10m) = 12.8‘? mis (25 knots)

Period, T = 1/f, seconds
‘1 E+5 iE+4 1E-1-3 1E+2 1E+1
500 1 ‘MM .. .. . _. .. .
1 1 1 . 1 1
1 1 ii

ii . _ ..

400 1 ‘ q W”, ‘
4“‘“l-_€T{_i____ _____ .1"--fi1._. _1‘7 .7. _ . ».»».»~.-11».»...............

1 —-C-1 API Upper Bound (alpha: 0.01)

1 1

k \ 1 1 —é% API Lower Bound (alpha = 0.1)

i 1 p p . .
/-"\\
1
1
1
1
1 1
1
1 1- 11 1
1 1 1 1 1: 1
‘
1
1 1 1 1 1 1 1
1 1 1 1 1 1 1
1
1
11111. 1 1 1111111111 1i11111 1 i
1 1 1 1 1

2/rn/s
Hz 300 >77 L. . 1 . .11 1
K12 1
fl\
/""'~ 1
1 1
1
1
1 1 1 1
1 1 1 1
1 1 1 1 1
--w

1 11 1 11 .::_. 1 1 . . . . . ..

200 ‘ ‘R 1 1 1
________11 u 1
1
Q 1
_ 11 1
..1

1 1 111 1 i
l
1 111 1
1 11 111 1 i1.1 1
1 1.1“ 1 ‘ 11
1

WndSpectrum,Sf 1 1 1 1 111
1
11%
11 l 1 1
.1
1 11 1 1 1
111
1 1
3
1 1 1 i 1 1 1 1 1 1 1 1 1 1 ‘1 1 1 ‘ 3
1 1

100 _1 1 1 11 1
1_1______1___,, ____________, 11 11 1 L __,_ ,__ _, ,, .1111,
1 1 11 _ .._,._. 1. 1. .

/ _ 1,._ ‘A-Q1“ .1... .1: . 4.“ 1

1 1 i T l I. u ‘V. ‘I '.-"‘ V i H H i 1

1 1
1 1
1 11 11 ' ' ,2;
1 . 1

iE~5 ‘1E-4 lE—-3 1E-2 1E-‘l "1E 0

Frequency, 1‘ = 1/T, Hz
Figure B.1—Comparison of API and NPD Spectrum for 25-Knot Wind
 API Recemmeweee PRACTICE ZSK

Uo(_1hr, 1Dm)= 25.74 mfs (50 knots)

Period, T = *1/1‘, seconds
1E+5 1E+4 1E+3 “TE4-2 “1E-H 1E+0
5000 aponuunnm
1 1
_______1“____________1-unnn|1|""' 1 jrnqynumnu
1 11 11 11 11 11
_u1_________
11
_________ r’;’"'|uuuuI\y~M' 7 —: __ 1 7 - - 1 *—* 7
11 11 11 1 11 1111 11
1 1 1 1111
1 1M1
1%.
1 E 1

1 1 m W

121 --11-11---1-111--1\1Pe
4000 1* _ W,” _____ __ ____ __—__._.. --1-1-..... __ __ ____________________ _j _—_—1>,—,1~—

11
__

A
~— —

-1-C— API UpperB0und (a|p1~1a=o.o-1)

----1---1 1
‘
--

______? 1 1 ____________11
1A 1 --->-é--- API Lower Bound (alpha = 0.1)
111
1 1. 111111 1 1 11111 1
/—"°~
1 111111111 111111 1 111‘, 1 - 1-
‘1 1 ‘ 1 ‘ 1 11 ‘ 1 1 1 ‘ 1' 1; 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1-

2/Hz
m/51> 3000
1‘ ‘ 1

11> 1W1 1 ._.__ ---. 1 ;T1LT_.__: W ‘1 1 111 - 0 1 Q 1.1_._1_11_.1

~1-.,,,,,/
»\ 11
1
11111
1111
1 11
1111
._ 1 . 0.
1
/-‘M

2000 11 1 777 1

_ 1_
_ _ _,1_1 =71... __, 1 _. __ _____1__ __ _. _ 1 7 1 "7 77.,,...1...J[-1-_. _____ 4 1 AW 1

_.1 1
111111 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11
11 1111
WndSpectrum,S(f 11
1 1»
1
1 1

1 1 1
.. 1 1 1 1
1
1 111
1
1 1 11
1
1 1 1 1 1
1 1
1
1

1111111 11 1 1» 1 - 1 1111
11 1‘ 11 1
1
11
11
1- 11 1 11.
11 11 ‘1‘ 1
111111
1 1
1 1 1 1‘ 1 1 1 1 1
1
1
1 "I'.
1
1 000 ‘1
1
1 1
____________ _____1.~
, 1 1___ _____.110- 1 ______1 ____W1 1 1 __1.____ _ .11
1
._1..1...1\ 1
1

I
‘E-.:
.1;

£0
‘ . _ . .. . ... . 1 K4. ‘ . . .. . _ . /. . .1-.41. - §*. 1..-e .1 .+ . _:;=:_ .E:':':':1.- .1:-I-'-1:; .. . . ..
O KZZflIfl_flI 7______.__ flIfl" "ER291’-QBVJ$101!E"-i!'8.'Il'tlb!!.\lRI2'hYIL*;1nWiiii'ji*Ki'i|i¢i|iiEi'§-:..-Eiii:§j:..._.1I_I1EIE._.____._E§§;_| _____ .. ‘ -------

1E-5 1E~4 1E~3 1E-2 1E-11 3 E-O

Frequency, fm HT, Hz
Figure B.2—-Comparison of API and NPD Spectrum for 50-Knot Wind

Uo(1 hr, 10m) :-. 41.184 mls (80 knots)

Period, T = 111‘, seconds
'1 E+5 15+-4 *2 E~1~3 *1 E-12 35+"? ‘¥E+0
5000 1 1 1 1 **** 11111111 1 1 11 11 1 1 1 11 ii. 1

‘ 1 1 1
4000 1 7 77————___ ——— 1-.-1-1...1

A 11 1 11 -0-— AP1 Upper Bound (alpha: 0.01) 1 1

1

1 1 ‘ -—->-<-- AP11_ewerBeund(a11:>11a=O.1)
‘ 1111
H—"\ 1151
1: 1111111 1 11.111
1 1 1 1 11111
1 1 1 1 11 1; I 1111111
I1 1
2/Hz
rn/s
--1_,1,./
3000 11111111111”_ .............. __1_--
1111111111
............ .........._.1_e._1.__W.11“..1....1 1.. L
1
W_1.........................._1fi».e
1111111
1“
1 1 111- 1

/"""\-.
Q3;
'1-I'-"I
-1.,,./

-\

2000 W :1» 1 We i T e 11_W_.. ~*.11 .... 1 W

11 11 11 1111 11 11 1 1 1 1 1 11 Aral, 1 1 11 1. 11 1 1 1 11 1 1 ‘ 111‘ 11 1. 11 1 1
1 1
1
1
11
1
1
1
1
1
1
1
1
11 1 1 1
1111 1 ‘ 1 11 1 1 1 11
111‘ ‘ 1 11
1‘
11
11
1111
111
11
11
1:111
11111
.1
1
1 1 1 1 1 1
1
1 1 ‘ 1111
11111.1
111111‘
“I
111

1‘1‘r 11‘ 1
1 ‘F 1 1 I 1 1 1 1 1 1 1 1 1 1
1 1 1 1 51 1 1 1 1
Spect
W
Sndrum ‘ 111111111
1 1 1 1 1 1 1
1111
1 1 1 l
1 1 1 1 1 1 1 11 1 1 1 1 1:. 1 1 1 1 1. 1 1 11 1 1 1 1 1 51 -
1 1
1

1 000 i ‘ ‘ ______________________.____..¢<_;+ 1:._, .~1......1».~_ ___________________ __________1‘._______71——77%-..-r........_______ ______.______~:;=11_.._.......,....1......

I4,.

1 1 1 1 1 1 1 1 1 11 -1 1:. 111

. . \ ‘ 1 ..1l ‘ ‘ . ‘ . . 1 . ‘ 1‘ :b..Q~-uni-».§~. _'- A .-1.0!». A. . _ 1

1 . 1 . LA“. A ,;; , ‘_ _' * . . . . . . . . . . . . .. 1 1 1
Q litiihmQtinInQIiimum!thanmmIuamzamzqtnavzataaura:wearslausmma>;s:&v.@w.»-I51;-Q1wgih...-T1-1.»a@...§=.; .-. .-1. . ‘ .

*1 E~5 ‘1 E-4 "iE~3 1 E1-2 1 E4 YE1-0

Frequency, 1‘: 1/T, Hz

Figure B.3-“Comparison of API and NPD Spectrum for 80-Knot Wind

 APPENDIX C---SIMPLIFIED METHODS FOR THE EVALUATION OF ENVIRONMENTAL
FORCES AND VESSEL MOTIONS

C.1 Basic Considerations b. Current force due to beam current on ship-shaped hulls:
Design equations and curves for a quick evaluation of
environmental forces and vessel motions are provided in this F6), H CWSVC2 (C.2)
appendix. These simplified analytical tools were developed
primarily for the analysis of mobile moorings. They may be where
used for prelirnina1"y designs of permanent moorings if more
.FC,, I current force on the beam, lb (N),
accurate information is not available at the early stage of the
design process and if the limits for these tools are not CC}, K current force coefficient on the beam,
exceeded. For the final design of permanent moorings, how-
ever, the more rigorous approaches as outlined in 4.2 are rec- I 0.4011»/(re - 1<r2) (72.37 Nsecz/m4).
ommended.
Note: Equations C.l and C.2 were developed for estimating current
forces on drillships. They are applicable only to production vessels
C.2 Current with similar hull form and size.

Current forces are normally treated as steady state forces in c. Current and wind forces for large tankers: Current and
a mooring analysis. They can be estimated by model tests or wind forces for large tankers can be estimated using the
calculations. report Prediction of Wind and Current Loads on VLCCs
published by Oil Company International Marine Forum
C.2.1 MODEL TESTS [C.l]. This report presents coefficients and procedures for
computing wind and current loads on very large crude car»
Model test data from towing tank or wind tunnel tests may riers (VLCCs), namely, tankers in the 150,000 to 500,000
be used to predict current loads for mooring system design dwt class. Wind/current force and moment coefficients are
provided that a representative underwater model for the unit presented in nondim.en.s;i.on.a.l. for moored vessel. in
is tested and that the contribution to current load. made by various draft and under keel clearance conditions. While
tllrusters, anchor bolsters, bilge keels, and other appendages the analysis of mooring restraint has not been addressed,
is accounted for. Care should be taken to assure that the char- coefficients are provided for use with either computer ori-
acter of the flow in the model test is the same as the character ented or hand calculation techniques for design of tanker/’
of the flow for the full-scale unit. terminal mooring equipment.
d. Current force on semisubmersible hulls:
C.2.2 CURRENT FORCE CALCULATIONS
If current forces are to be calculated, the following equa- F6, = Cg5(CdAC + c,,.-4;) V3 (0.3)
tions should be used:
where
a. Current force due to bow or stern current on ship shaped
hulls: FCS e current force, lb (N),

CS, I current force coefficient for semisubmersible

Fcx I CcxSVc2 (C-1)
hulls,

where = 2.85 lb/(ftz ~ ktz) (515.62 Nsecz/mi’),

Fex e current force on the bow, lb (N), Cd I drag ¢°@fll°i@1'1ll (dime1'15i°n1@$5),

C6,, e current force coefficient on the bow, e 0.50 for circular members (see Figure C.l for
members having flat surfaces),
2 0.016 lb/(ftg - ktz) (2.89 Nsecz/m4)
’ AC K summation of total projected areas of all cylindri-
S K wetted surface area of the hull including cal members below the waterline, ftz (mg),
appendages, ft2(m2)
’ Af I summation of projected. areas of all members hav—
V6 ="= design current speed, kts (m/sec). ing flat surfaces below the waterline, ftz (mg).
64 API Recoumsuoso Paacnct-2 28K

2.0
E IIIIII if M mmimmmmmmii i i iiiiiiiiiiiiiiiiiiiE ~IIIIif T TIIIIII W t
V E . E. "
1 \ —------1» h r HIE

1.5 ............ ..
i ___ _ ' TE“?
............

.1 4.l, .
"""\"!""""""""’* ‘if r E
jfliiiflwfliii i Tiiiiiiiiiiiiiiii h 1 1' R Z u-CITIFH '

Co.. l I eeeeeeeeeeeeeeeeeeeeee~""'"""'***********************1 *4--‘I '_

nt 1.0? 3
51
a
ll
T 1

‘*"" 1 1I l

tf'c'e
lull! i

l
‘ li ‘:

l . _ l{ W” _______ ._
‘F’ l *************************"Tl*********"Z*****" 7'7 l
l l
DragCoo
0.5 .. . . . . . . . . . . . . _

_ _l _______________________________

0.0 d ________________________________________________________________ __ ———————————— —— . . .

0.0 0.1 0.2 0.3 0.4 0.5

Figure C.1-~Semisubmersib|e Current Drag Coefficient for Members Having Flat Surfaces

e. Current force on mooring lines and risers: The effect of assure that the character of the flow in the model test is the
current loads on mooring lines and risers on the overall same as the character of the flow for the full-»scale unit.
mooring design should be evaluated. This is particularly
important for deepwater locations with high currents. Cur- C.3.2 WAVE FREQUENCY VESSEL MOTIONS
rent loads on mooring lines and risers can be calculated
using appropriate current profiles and drag coefficients. In The motions of the vessel at the frequency of the waves is
high currents, drag coefficients should be adjusted for the an important contribution to the total mooring system loads,
presence of vortex~induced vibrations. particularly in shallow water. These wave frequency motions
can be obtained from regular or random wave model test data
C.3 Waves or computer analysis using either time or frequency domain
techniques.
Interactions between ocean waves and a floating vessel
Wave frequency motions have six degrees of freedom:
results in forces acting on the vessel that can be conveniently
split into three categories (Figure C.2): (a) first order forces surge, sway, heave, pitch, roll, and yaw. They are normally
that oscillate at the wave frequencies inducing first order considered to be independent of mooring stiffness except for
motions known as high frequency or wave frequency floating systems with natural periods less than 30 seconds.
motions; (b) second. order forces with frequencies below
wave frequencies inducing second order motions known as C.3.3 MEAN WAVE DRIFT FORCE
low frequency motions; and (c) steady component of the sec-
The mean wave drift force is induced by the steady compo-
ond order forces known as mean wave drift forces which can
nent of the second order wave forces. The determination of
be estimated by model test or calculation.
mean drift force requires motions analysis computer pro-
grams or model tests. Design curves for estimating mean
C.3.1 MODEL TESTS
wave drift forces for drillships and semisubmersibles are pro-
Model test data may be used to predict wave forces for vided in Figures C.3 through C.l7. The curves are applicable
mooring system design provided that a representative under- to typical MODU type vessels. However, for large drilling
water model of the unit is tested. Care should be taken to and production semisubmersibles (with displacements over
 Dusters awe Anatvsts or Srarrouxssetno Svsrarvss son Froarzno Srnucrunas 65

Low frequency component .1.

‘~11

“',"',J? l l pl’ ‘
Force

( ( (t it { { { { { { { { { { to

7
i High frequency component
K

7 7 "W" ""' |unv|v|L___;* """""" "nnnnunpnu nmmt _ ____nuua|v1___: ***** »~~~...................___J.....aa* .”m~~~. ........._.. .... _
Time

Figure C.2-~Wave Force Components

30,000 short tons) and large tankers, the use of these curves is low frequency surge, sway, and yaw motions are included in
n.ot recommended. a mooring analysis.
The curves for sernisuhmersibles represent the upper Low frequency motion of a moored vessel is narrow
bound of the mean wave drift forces generated by a motions banded in frequency since it is dominated by the resonant
analysis computer program for four semisubmersible designs response at the natural frequency of the moored vessel. The
including typical 4, 6, and 8 circular column twin hull designs motion amplitude is highly dependent on the stiffness of the
and the pentagon design. The curves for drillships were gen- mooring system. The motion amplitude is also highly depen-
erated for ship lengths of 400 ft to 540 ft. For drillships which dent on the system damping so that a good estimate of damp-
are outside this length range, the mean drift force can be esti- ing is critical in computing low frequency motions. Methods
mated by extrapolation. However, extrapolation for ship for predicting the low frequency motions are still in a state of
lengths below 350 ft or above 600 ft is not recommended. development. There is a substantial degree of uncertainty in
The data presented. are appropriate for ship-shape vessels the estimation, particularly in damping.
with normal hull form. Care should be used in applying this There are three sources of damping:
data to vessels with blunt bows or stems or other unusual hull a. Viscous damping of the vessel.
features.
b. Wave drift damping of the vessel.
C.3.4 LOW FREQUENCY VESSEL MOTIONS c. Mooring system damping.
Low frequency motions are induced by the low frequency The technology to estimate viscous damping has been well
component of the second. order wave forces, which in general established, and viscous damping is normally included in the
are quite small compared to the first order forces. Because of low frequency motion calculations. Wave drift damping and
this, the low frequency forces do not play a significant role in mooring system damping, however, are more complex and
the motions in the vertical plane (i.e., roll, pitch, and heave are often neglected because of a lack of understanding in
motions) where large hydrostatic restoring forces are present. these damping components. Recent research indicates that
However, in the horizontal plane (i.e., surge, sway, and yaw wave drift damping and mooring system damping can be sig-
motions), where the only restoring forces present are due to nificant. They can even be higher than viscous damping
mooring or dynamic positioning systems and production ris~ under certain conditions, and neglecting them may lead to
ers, the motions produced by the low frequency forces can be significant over-estimation of low frequency motions. In
substantial. This is particularly true at frequencies near the applications where low frequency motions are an important
natural frequency of the mooring. Therefore, in general, only design factor, such as for large tankers, it may be warranted to
66 AP! Rscoivwzsuoao Pnacrica ZSK

evaluate damping from all these sources either by analytical C.4.2 WIND FORCE CALCULATION
approach or model testing.
C.4.2.1 Constant Wind Force
The determination of low frequency motions requires
motions analysis computer programs or model tests. Design The steady state force due to wind acting on a moored
curves for estimating low frequency motions for drill ships floating unit can be determined using Equation (1.6.
and semisubmersibles are also provided in Figures C.3
through C.l7. These curves are applicable to typical MODU Fw E Cl/VZ(CsChA) V1412 (C-6)
type vessels. However, for large production and drilling
semisubmersibles (with displacements over 30,000 short where
tons) and large tankers, the use of these curves is not recom-
mended. F,,,. K wind force, lbs @\l),
The curves presented are appropriate for mooring spring
CW K 0.0034 lb/(ftz ~ kt3) (0.615 Nsecz/m4),
stiffness of l8 kips per ft of vessel offset. For other mooring
stiffnesses, the results from Figures C.3 through C.l7 should CS K shape coefficient,
be adjusted by Equations C.4 and C.5:
Ch K height coefficient,
A -"= vertical projected area of each surface exposed
X. e (<14) to the wind, ft2(m2),

VW K design wind speed, knots (m/sec).

11.: (<15) The projected area exposed to the wind should include all
columns, deck members, deck houses, trusses, crane booms,
derrick substructure, and drilling derrick as well as that per»
where
tion of the hull above the waterline. Wind shielding in accor-
(XS)REF = rms single amplitude low frequency surge from dance with acceptable methods may be considered.
Figures C.3 to C17, In calculating wind areas, the following procedures can be
followed:
k 1 mooring system spring stiffness in kips/ft taken
at the vessel’s mean position, a. The projected area of all columns should be included.

(Y5)R;y:~ = rms single amplitude low frequency sway from b. The blocked-in projected area of several deck houses may
Figures C.3 to C. 17. be used instead of calculating the area of each individual
unit. However, when this is done, a shape factor, C3, of
The drillship curves in these figures are for drillships of l.l0 should be used.
400 ft to 540 ft in length. For drillships that are outside this
c. Isolated structures such as derricks and cranes should be
length range, the low frequency motions can be estimated by
calculated individually.
extrapolation. However, extrapolation for ship lengths below
350 ft or above 600 ft is not trecommended. d. Open truss work commonly used for derrick mast and
booms may be approximated by taking 60 percent of the
C.4 Wind projected block area of one face.
The force due to wind may be determined by using wind e. Areas should. be calculated for the appropriate hull draft
tunnel or towing tank model test data or Equation C.6. The for the given operating condition.
wind speed used is defined in 3.3.
f The shape coefficients, Cg, of Table C.l can be used.
C.4.1 MODEL TESTS g. Wind velocity increases with height above the water. In
order to account for this change, a wind force height coef-
Model test data may be used to predict wind loads for ficient, Ch, is included. The height coefficients, Ch, of
mooring system design provided that a representative model Table C.2 can be used. This table applies to the approach
of the u.nit is tested, and that the condition of the model in the using l-minute constant wind (see 3.3).
tests, such as draft and deck cargo arrangement, closely
matches the expected conditions that the unit will see in ser- h. Equation C.7 may be used to adjust the wind velocities of
vice. Care should also be taken to assure that the character of various average time intervals.
the flow in the model test is the same as the character of flow
for the full scale unit. m-am, (on
 Dearest ans Annrvsis or-“ Srsrionxaseisio Svsretvis FOR Fronriivo Sweetness 67

where C.4.2.2 Low Frequency Wind Force

V, e wind velocity for the average time internal 2‘, As stated in 3.3, wind force can be treated as constant or a
combination of a steady component and a time varying com-
ca e time factor from Table C.3, ponent. The time varying component is also known as low-
Vh, K 1 hour average wind velocity. freqnency wind force. Similar to the low frequency second
order wave forces, low-frequency win.d force also induces
Table 0.1--Wind Force Shape Coefficients low frequency resonant surge, sway, and yaw motions. Low-
frequency wind forces are normally computed from an empir-
Exposed Area CS ical wind energy spectrum. Low-frequ.ency wind and wave
Cylindrical shapes 0.50 forces are normally combined to yield low frequency vessel
Hull (surface above waterline) 1.00
motions due to both effects.
The recommended wind spectrum is presented in Appen-
Deck house 1.00
dix. B.
Isolated structural shapes 1.50
(cranes, channels, beams, angles) C.4.2.3 Wind Force for Large Tankers
Under deck areas (smooth surfaces) 1.00
Steady wind forces for large tankers can be estimated using
Under deck areas (exposed beams and girders) 1.30 the report, Prediction of Wind and Current Loads on VLCCS,
Rig derrick 1.25 published by Oil Company International Marine Forum
[C. 1], as discussed in C.2.2.
Table C.2—-»-Wind Force Height Coefficients
(for 1-Minute Wind) C.5 Oblique Environment
Height of Area Centroid Above Water Level The equations presented are convenient for calculating
n Tm Ffiifii W M€1€fS 777777777 W
wind and current forces for bow and beam environments. For
Over Not Exceeding Over Not Exceeding C;, environments approaching from an oblique direction, Equa-
tion C.8 can be nsed to evaluate wind and current forces if
0 LII CD 0 15.3 0 more accurate predictions are not available.
50 100 15.3 30.5 1.18
100 150 30.5 46.0 1.31 F, m %~Fy[ ] (cs)
150 200 46.0 61.1 1.40 1 + cos 6 1 + sin qt
200 250 61.0 76.0 1.-47 where

Note: This table applies to the approach using l-minute constant Ff 1 force due to oblique environment, lbs (N),
wind (3.3). It is based on the following equation for wind velocity:
V2 Z, 1/10 Fx I force on the bow due to a bow environment, lbs
».>-1, ~ tal Fy:
(N)
force on the beam due to a beam environment,
where
Z e Height of area centroid above water level (m), lbs (N),
VZ 3 Wind velocity at .2, J‘: direction of approaching environment (degrees
V10 = Wind velocity at 10 m height. off bow).

Table C.3--Wind Velocity Time Factor C.6 References

Average Time Period Z Time Factor a C.l Prediction of Wind and Current Loads on VLCCs, Oil
Companies International Marine Forum, Second Edition,
1 hour 1.000
1994.
10 min. 1.060
1 min. 1.180
15 sec. 1.260
5 sec. 1.310
3 sec. 1.330
68 A M R E COM M E N D E S D‘ R A C T WM E 2 Q“ HR

pa H
W - mp w eL KWm WW
HS} _ upi t§iLi § il!§!i_®o_~O":t_Mm: IpO
A_F141‘giIp‘1g‘p5Ii_p p_i1

M G8H Ve D fi FO 1

Om
_®W@
ON@
anmm
ENmm

‘p‘“ ‘W

pi_L[“‘
’il“§““““MWHW“
W_t
_“ p[T“1‘bpEitH‘pl1k;_H_:i“kM.kH‘p.”“‘MMi
lHi~“W‘‘
U‘_H‘np
HU_"“““““““‘p pp;

Z
W
WWp p
Wi

“ “ If

Wm__WWW_W_
W
_m_7
WW_ j
T__
i
7 H_Z
f _f
Wi
%
” 7"H
J
___
_ T_ _ _WMg WWWW _ A___ ___Z
fl_

E5
3%
®>C8>>Om_wQE2__5.
gm Og_6“_>HiOm_>O_@__éH_ _O9_

5 __ MZ k _ M _ L__ _

p p p pp
_ _p_ _ _ ___‘L1i_‘p:i§“‘_L“
“ W‘‘§“HM‘%:
EVEQQI
§Om_%@_>§_ w
__ “Q
_ H _H_

wmwacx
NW p Hp y

_ _

_ ___i
ml“‘£W1“‘H H M‘H“ l 1lp‘ _
an mm 8 go“: wg _ _ p_
_ Efir“ %
Y
_HHHH_“
i EW
i _t
___3

i i i

EVEgg__ m§O Cl:

E®E®O®_Q£m_
Np1

___mEw
@“gm
W@$> EVfiugm ‘iii
MW pI
gEa %_
V_mug?3
2I3
QBF

WpMp

w U O HO W __AO U 6 H _D 6 M M O _i 8 p n M w V 8 hmU WM 8 W “Q
 M e aH W8 V 8 D rm FO C6 r K_ mQ
Wfi

O2p
_pp
H]
O
WD p_Ipp‘itpp pw“‘npp£p1kp!‘pi_ _p.ilp_MUflpi fip!p p pHpHHUpip flwp p p p

mm3
OWm_pQVm
2“
mw
D

p _ _‘__p
gpW _pp
___ _ p_ E5
$82
_pm
2_ _p_ _

ppp}
p

p““p L
WW
p_p

goW®w2gmgmg_%__56%C
bg
>§ pO_H_p w_O®__§H_ O_ _O_2
mp p p p ‘p p p p p ‘p p p ‘ p p p p p pép p p p p p p p p pi“‘l_fip ‘p p p p p p p p p p p p p p p

p pwW
p p__ pp W_
_K
Ww
_
W _ _H €W
ooqg m_Qm5 W
:CwEHgpp pM
Evpg__pgp3pp EWI
Sp
mmuM
p W w
Wp
p___W_____ W
i”_
i _p
W_ M_pp pp p 7
p
q7p
Wf
fl_
N
_p‘Wp
p ip pI p
_
__pp_p
H_H_HHmHH_H_H_H__N\
___ _O W
_ _ HU%
H H _ __“__ H _? MppflWp pgWpp__
_1‘
m_
p p_pp_
p p_ WW_
__ i !i %pi__‘_i‘i W
:
ppW
E
Wm
ppW
_m%_Hgmw
_w_m®>UmS_

amp“Hp Hpp a p__pp ip‘_ _i iWfi

"W
Hi
i _
W
M_ __pppp_
p_ fi
pW
_p W_
W
_ _W
p__7 ”
_

_ _pi
pg
p‘ 7
p rpifip p p_H“ ‘p W
pp _
__ _
p W i_
_ WpMppMp
W _ WWW
_ppp_W 7__ _

g
3Emgm;%:%p_m_w
7
HW

Wp_WW7

A5
£_%m___g>O
ctEnmgm
EV
Ea

W U O HO W pflS U 8 Hb 9 NJ M O _p_ 6 p mp MwV 6 mmU E _S W H

78 A M R E C O M M E N D E B Dr Q R C T E E 23K

M 8 aU MVE D m FO C Qr r wK___j_m_1 S \_|_ _ _ _

_gmA W?Om_A \_éA1\ _ __’H‘’L®

“M}%‘§_““‘1Qk“M‘1:“l!I“‘_‘Li
§_“
‘
Q
E
8
9“mfl
gma
2@
hl“H[_‘%"\§3v\LHi_ i
“ ‘“iWV“
: ‘_!‘i “‘:35 ‘“ ¥tkfUHHH‘N“ i‘"“

___ nW__AUU_ M 1j E§l§¥‘_‘N__‘Z fl_uflMQ”

%_
i\ jA1 f
Ni __W W
__” _ M"__
_ _
W
W A
_
WW: W_HH H _C _w__ _W
W_
_jw

W”
_A _ __

iv‘;
‘L1“‘:‘1_‘‘

E_m_
DEM
gm§OH_
CO:gomgEw__t
wgw
§>|Om_>O_w__§H_OL8
_ __
:
_T i _
_ __ m_ __ i
MW
i
ii
7_W
_____
_

i _ W_____%€__5
V__H

__ i i _ %__
_

“ 1‘ _“‘ “ 1‘ “1‘

L1
Q309%{
StEmwwl
Emu_ m_ag§>w
WW
J‘,
d
_ _

ab
yr
o
_W
_I _H Hm_nH _ _H_ _it

Cg
AN“

__
W__
_ _ Wi

Q5“ 9“ a _5
EO32mgHUm_%_O
GW 1}“ “ “M“ ‘ ‘1‘I:“‘1‘ ‘ ‘

R EV
%g_Hm _mEV
at£9_m__ Ea§/OH7
_N,

’ *‘3“‘Il ‘ ‘ ‘
i W §W
_§i“i‘_
TE ‘_

_@_“gm
@w§_Q@®_> 2
%_g _HX
Mgf
©_mUg£

§ _ W _ W W __
W_ __ _ W_ W _ i# _

(
1+ J ) U 0 "___ O W FA3 U 8 H _D_ 9 J 4__ M O “I 8 p n U_ ‘Ww V B it U S S W H
 Q ES E N AN D AN _fiLY S mm mT nu N K E E P W G
my Fr 8 cu Y S Y E Mm S F AU R F‘ L nuAT W G CO Wm U C ? U R E S 7;Z €,M

M 9 Hfl W8 V G m t FO rC9 M my ‘S;

2Q
@P
Q
ON
QMmN
QWR@@

WO
i_A;_‘iQINJW:
_\A;W\ \ §W_ _ _ _'_

__J
_ W ‘QNNQ Ii“_Ii “‘ _I_'! .-l‘.“i!‘ _'. I‘Ji_‘l _T‘.l_“ l:I_l!

_ _ " ‘j“_“1‘ _ E5
0%
®®O_6%C
Qpswd_w®mmgmwt
>_®>_EO©“,_O®O‘_#O'_9>H_ _ gggE2__5

i““:‘W
_W
W_ __
_W_ j_ NJ1H”‘1‘“@LN‘
I u‘HHu1hu‘u‘u_“ _Qm WM
W
H“__ _W_ _‘A_
‘r\éi __
_
W _
W
j$22
Ea

A‘‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ 1

W
L
WFCwh_”
W_Z _

Y
_ _

PG
39$:
Hc_m%um>gm
wWON 7
_

50
flog
NEG

5_
T‘‘1 fi““1 _
fi
W_W fi
i
W"
Ego_ __
“gape i i fimww ‘§_i ‘I|_Il_L_l‘- r“‘.“.“.‘I-l‘_

gm:
2aUP
W_

O9“ mm O60”: ‘i
‘1‘‘_i
‘ i“
_ “ 1!‘“ L“““ “ ‘M

_W
_
__
i

EVEgg__m§O Cg
EwEgm_m_m5
W

__mEw“gm
Egg EV
r_%m9m
tE_
Y”
QVQE
H
g
ED
_ "_ j_ _

Q
A
m% flu_
_

__

{ 1+ }_
II U m_ w W InO U 6 H b 6 NJ W O _I 6 p m_ M w V 6 Fm‘U B 8 W H
7; _?_ F
A R E C G NW E N D E D “P Q A! C T w E 2 S K
_‘

O2‘
_ _ Oi _cwmm
ammrOPm
Ommm

M 6aD MV 9 W FO m 6 M ml Q

_ _ ____ _ mQ\_Et_

‘“N
_U“ M‘H“_hH‘!“H‘WE“ ‘ 1_ A W
_ :T _ _ ___H _ _ _ _ _ _ _ WW
_
_f WN W
_
__ V W
ji Hm_, ____ _W
_

Nk !§i_N _iQAN

_7 _fi in______
W
2 __ WM__
*__

‘_‘_Q\]‘ \‘ _‘ ‘ ‘ ‘

__ __ _ _ fig
“®>®p_O“
he_
Q95dgm@E_
O§>ED¥N__Omgmm
§HO_>_ _®%_6g_Ew_=5

_
_ _ A_ _ f‘ _ __

_ _ __

W_ifi % m______
1‘1iwh‘_“7%i ‘:“¢kHWu_h_“ ‘_:“\h$‘_: W_H E!‘§i“‘dQl§

__ w_ AI
W
_W_
_H H H H _ _ _ _ H H _ H H_ _ §§¥“%;\W“¥Wi
1‘“i_‘M‘3i“n_M‘“H_‘l“‘“‘i ‘“ ‘ ‘
l‘Huh‘l“1
1‘ _““1
U“H “‘H‘H1‘|Tl ‘ 1
>¢®WW_L
i‘N S:
H“cg®m_u=ym:m c_m >m
_ W
_W_M _
_ _H H _H H H

‘W
_W C
_

\L

EW“ I

OE“ E 25h©_ \Q;

Ev
ggg__g_ /O cg
E®E8_m_%_O
1‘ “ _‘“ 1

_@E“$6
nWw®>%_@_>_ 2_W_
"H_
_W fi
_ W
M
_

Ct
éum hm
z
CO3:
M_u__5_H
EVEa
gm:
S’
mmH

W‘ 1___ L
L _

Q
© @ W m NW F G

w U O __HO W A O U G H b 6 3 M O _| 6 D_ HH___ M W UV 8 flwU B S W H

‘J
 ___

M 8H H MV9 D rm FO r C E M my Q
O Om?_ C2_

"1‘:t
“1_‘“1“‘Wi“‘M_1:“‘T_H3H‘5¥iN“‘HyfiHh‘i 3‘“iM1 HMU“
Om
2“mm
‘1]‘
Ii‘‘ W‘HH‘1 H‘1M_WT“H H“

T_
W W_’ ii__ ‘LV‘ ‘ I ‘ ‘

_V 1_E5{I1__I{_‘ii_ __M_2_m
r
W_W

W‘‘“ 1 “ ‘1“ “‘

E6
DEW
g§H_
he
®8_OH_
Cmgmvt
mvmawgm
>O@_>H_O¥w_>*O_gagEw___5

_
_
_

‘iLU‘

Q
““
_ll‘‘;“1‘_I1‘
:1‘“ l,“i‘ ‘ 1‘£:“ ‘ 1‘ii““

_ 7
_ _7 _Q’
__ W_L
W____
O2ESNmmW

émé g
Ew$55QQIg_C9m
aw
Jo
_ yM
o

L““\i‘“!\
“VA/O
_
_
ROG’?
_

iii!

\O

yo
aT _ _ 71
_
£0
9% aw GOO“K
‘_

_ i_
77
__

II
ii_“
‘M“““M “ “ “_““‘

Ev%C®___§5 CI:
E$E@OM_Qm_mQ
1‘ i\ *_
_@$93
“gm
M_wg> Qt
£_u$_m
atgm E8
gqgH
gm:
S
2H

v_
W

W WW

Nd

w U O HO W M O Mi
__h3 U 8 H _U 6 J Id 6 p m M w V 6 fimU E "S W H
7*4H AnM R E C O M M E N O E D D‘ R A C T Q E 2 HQ “H
_j__ ____f
W7 W
___Wk
___WE‘
_ __ 1JIii;A
Q
I_A1_lA % A

M 9HH W8 V G nu r m FO C 9
r M m Aw
GE OE
Gm
\ \W
__
N
W
FO?
ONm
mm
mfiOW
Q‘

Il_I __
.‘U
_i'-i .I -Il_.‘_Imo
‘_‘.“‘. ‘l ‘. -‘1.l
b_ _ 3 M_ M __
V ___
W_

_ _ _ _ j_ ”YW
U2
H
gm
_ ___‘_ _ __ _ __ _W
W
__\

E5
Ug
m36%C
bwgw
figmm
vL>_§>O¥®H__O@O$5m_“_O, _m>_5t_m_6wQ_fi_ _5

_
_ _
7
LélNéM;_A“Té _N_NN

5
i_A
“zfi

_ A __

yawn?“
_ M
? ‘J
W
“_%‘t‘i
W
N_

W
\WE,»
1 A fix
_ A_
_‘“fl¥“_L“A“_“:1’]‘“N:I
EQQI
A5
Hgm;C§ _@_w

E950
_"WW

i\\};
max
NR0___ %H__ _ _ _ aopo€$

A1 *AQ _
_ %
fifiv
_&i x

___mEw
m“ww®>mum W
gEgg_
cgH EV
SfimmnHgm _mEVED
ow§O3 €_O8:
vHCmEHOm_QwWQ M %_
XgagWa W
_ __ W_
_ L“
i“WW ;“
WZW““‘l
Wa

__

gm:
SI
gH

w U m_ m W FAO U G H __D 9 J ___: Mmy fil G p m I1:

w WV 9 fi U E 8WH
 Q ES w N R N D AFM anL S g V; O F ST Wmw N K E E mm N Gd S Y ST E M S F O R F LG A _TW G ST R U C T U R E S FED
0%
®8heCO:
bgwdgwmgat
$LO“> £O_"_Om7:jvgma
O5_>_ gowQEw__ 5

M98H MV 8 D m FO C 9 M ml Mm;
r r

QGN CW3 Jog:

O2AI_ii
3
{A WV\ Om_ _OAl Om0%mm
fi
J‘NIiA1‘ A ‘ \
\ “W_“L‘_Fh\\
iii‘; Wé‘1 AN“
%N%? _%O/flu
Wf_W i
WW _ éAAAAi2j
WAA”JWW Q Wv_ [LL\
i\WL
1W y\
“CO/yo
7W
WH_ HW
_ H_ _ H W‘_i _____‘‘_W__Y
___‘V
$2‘___ _ _W__ _ _(___ _ _

W__v_N W
_Ww8Ww%;j Mfiwtéo
_ _ _______ H
_fi
yO%y_N /_VIQW”W?Li W_____ _ _ _ H,_1_ _ , jW
fij
(\\__
W _ W
_W1Wm
_
_
J1‘!‘iii“

__Km
‘__
31‘“ii‘

__
_
““‘“ _“ “‘1_W“‘
“‘“3W;‘ ‘111‘ ‘ ‘5‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘_““ 1“

“it‘
W“‘HUWH“ H““_1,‘“ ‘l“i\W1“‘:¥n“‘H_“M_Hh1“UM“ “i‘__:ii

WWZ W(
_ W_

2D
OW@f
G
MN
E
Wfj _ E
_ WWW_YjiW(____‘WW
‘xx ___“

g2%:
Emgm;U_%_ §w
”
_ _w\j\\_
H

.___‘( _“

_
I “‘H“H‘_“_‘l‘‘,‘¥
VH_1:!‘ :“‘
l ‘i‘““ 1‘ i‘‘1I: ‘ ‘“;I‘!i'‘_,‘T‘‘ ‘>‘,>‘“,h‘vquvhé

Ev£€®_
_ _§>O Ca
Ewgwufiga
_®CEUQE
@Hwfi@@>mw _ __,

g
gumgm
EVEa §_
K
gqg3
gwz
£1
®_®H

M U O fi__O W A O U 6 H b BH MO 1 8 D_ fifi__ IN
w UV 6 ii U Ab S W H
76 A M RE C O M M ENDED P RA C T EE 2S K

M 98 H MV9 D rm FO m 9 M m Q)

‘iIL
Q
3
£
% %
Q"I
M!ON\i\lN»;
\‘W i\_iH‘
I__\‘i i
__ W
H_ _ “cm?
T
W_iWE__g_‘_
£
vA_’
g ad
_ MMW
__H“_‘W_WW
_ H_ ___ W‘___
____HHWHW
___U__¥o
g _gW
__ “ ‘!_}i“rm W‘_ W
i_1H‘vl3“hi1 _W__ W
H‘g M
___ “ 1‘ w __T_H___
It‘"_1 _WH_WW_
__"__W____"
I i
i_ M
__Z
O
ONM:O__m
mmommm
Ow

ac/woe
{ ___‘ ___‘w\___\W
H __

m H____
l“ L:“EL“ _:‘H_“!‘

_*,_
W__ _ _’_’_

H\_\1\€‘ \1___\ \ \ w”
‘___” _HM__ 3%
WED
®3$26figwmgmt
>\m_w>O“|_ ‘_OQEUE gowQEw_:_ O6*CO$O_>_

H_HH___H__
W
G _
__ _’,
___

___D7
_ W
_
W__ W$1_
_ \__
_ W _ __ I_\
__ m_

_A_\
A_H_ _\I MW

EV
2%:
Emmg;owtgm
‘ ‘L’‘
" l 1l“ “___““ 1W
_ __WV

__ _
K‘ _\\___‘
WW WQ_ T‘___

25 g LWNa
95“

__ _ _ _ W X‘_ é L
fl
”W‘ X_U MMW
_W_

at
gag__§>O pl:
E@E®U_m_ Qw_Q
_?\k_W\-6\%%
Ema?
mwm_“gmMm|_ gEngm
"
Qt:65 EQ_O_O_wPHx
gf
QM"; *7

_l|J_m AWm NA

w U O fluG W ffi0 U 8 Q H“ 9MM M O 3 9 p H B“ Q w V 9 hm”!U Q 8 W H

 S__
_ wO2% j QNmrCFm
G?mmOWmm
O_

M 9m W3 V e D Ym F_ O C 8
F
W p )

omYA f4iQ
_L\:li‘£“駥5A;IiWl3\A!4§_i|{l\A¥i%!i\

E5
E6
Egm
®COME:
wmaw
m§8Ew__m5‘
>_w_>O|NH__\Omw__Ou__ §H_
__
W_“““‘ Dci _ __\NJ;é1;\ O_(
q_omI“£_m
_R _ _
W
\T1A
A4WW_A} W\A\H1‘(O
000%’
ZA W i_WQ!YJVWAb?
AMWiji_jW
/vwvv W {W 1_
\;_my
W\
iii‘
W
Qovor L‘U _A___
_ “ ‘flHH§“‘1t“_1‘ ‘ ‘ “1‘ f_f\“‘_
M __ i

O
___ gW
W_

fiW?77

EV39%
HCm_mU_%§m>_w
k
fl_W ‘ ‘“ i‘“J_“

é?
:T $2
OS V_cgE/Ogj_ W
_W _ 5w _ _W
_x_ _\_ \__ v
\\
W{I

____
‘_““ “ “ “I‘‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘ 1‘i“ 1‘ w‘
__

ON
fi
L‘
W
K___‘
Ff

A5Em;
_ m§_O pl;
E®E®UM_Qw6

___mEw
$“gm
®W®> A5Engm
Evfig

{ mm U O O W" AW 6 H b6 ‘dM O*1 6 p RH “W w UV 8 AmU B 8 W H I. 1 ___}

?8 ArH R E C O % M E N S E S D‘ R A C T WM E 2 8 UK
mQ
E2
ON
W
8ma
Q“
Q“

M e aH MV6 D m r F O rC 8 M‘ m Mm,
OONA1I Gm?Aé O2A1‘W
omfi% O
A

‘1l ‘: {ii

iii
“1 ‘

iii‘

_
W__gw$>
CmwnHwwmwmfi 7
_ _H H _ _H H H_ H H _ _ WA
{ ‘ W

W_
___‘_
E5
UCW
®®O_COMHOE
Egma%__E“Q
wgw
>_m>_O|gH__O®S@H_
*__\
i i“1‘“‘“““““_‘

nH___H

H‘““i‘:“Hhf_F““

1Aif
WW_ _ 3 _flfl_ WWWW_
_ _W _ T AW
_
W W"W
___J
_ W

W_\K

“W
é“
‘“
Em?
at
Egm;m0%m_w
_
3 “1“‘‘ \~__\1
L“‘‘_H H“mH_ _ HmQ9
fig QM?
‘A W\_ Mi
i,

WJV
_
aw
W_Hmm__ éU_
_HH_m_
__ _

¢®_%
Jx

O
go“?
”
fi W L/Km
W_ \ WU HfiWW
T
_W__

\
W
any/V
%O__“m“H_
\ __ W_

1
OE“ ON x A566? iK

X
;
k
_JWW
I I

““‘

A5£Q_m___g>O C]:
EmE_8m_%WQ \_
_\
_ _\_
g
Egg
go:
V_
COOM:
H
atEm
gmr_
3mUmmP XW

© @
_ _ f __T W Qui W IN
W _ W W.
‘O
w U O HO W InO U 9 H b 6H M O “___. 6 p R n M w V 6 hmU E S W H
 Q ES mN A ht W AN ALY S $ G F ST M m N K E E P w G S Y ST E M S F O R F L O Ar 3 W G STR U C Y U R E S _{_
9
EV
gwW
b_mHw_km j Q2

M 98H MV8 D m F r O rC 8 M mQ

W
JE_ 1
__Q Q?
wL
IA £Qi
2A i AQif
‘A
T‘Aé&A\IN AAG
F
_
i H_\A‘HUI\iN_“‘N_‘1‘ 1‘

i_m_
WEV
EOSE
I
_NwEwHUm_Qm_XQW OKiE2I08$
___‘_WRg0_ _
__“6
H W
H___K
_W W
“\AA1“M
IMMO
\ ] J_A \mO¢
at
_ _ _ JA1A‘éWQ}Z\11A A _ iW
W,

AJQVG
WT QNW
J‘
_‘1
AléN I‘ Vé\‘]‘J ?:Aé‘ 1

J“ ‘ W_Z_ W6
j W_W__‘W l¥
uv, __\
_HHH_HH__
__Hn"____
_%
W
QWk
WWT T_ HrWH_\
W _0&0_
V
_W _j___W HL
W\\
___H_ _
W_(_\
_ ___M ___
7 f_*if WW
W
W _
WW
_Wj
iJ‘
A_‘‘
_‘i¥L‘_‘_1‘"“3[‘W‘“iHI1‘z“L’‘
hi‘_ji‘
‘HwM“5H:‘iVH_“J 1WEMH @__@O
M:
ON
QMma
QM

www$W>_ __ g£
__
$93$NWw OWHE @jOwC®__Ego WW
7
_H AH\
N__” _M‘U
_W\____iW\__ _ia_W
w
W
\ W
W\
W_ W_

tug
Em@_m®9_Emmmg_%__Ehe
Hwmmw
>_w_g>OoJE_H93>_HO _
‘Xf
WWW olviw/A
OW__
_j7 /_
V \7q{{Z W\
__
_W_ \
W W_W1_

EvEm?
Emg;§_Em_w
\
W

JIVQJVV

\A
\W_

{
1 W U m_ m W ehS U 6 fi b 9 H N 0 _i 6 p H “___ M w V 6 FmU E S W H
8_7 O A D__ RE C O M M ENDED D. R A C T m E Q“ SK
pWiS

M 6 a H WH V G D m F O m 8 MW W

A5owi‘ AI_l_e QN
_“ %“1HL‘i“L_1;:ul‘N“ i_“__
_“ _“mH_"M_“ f7W3 W\ow
H__1‘\1LAVAlufl‘:\“ “ L
om
_R \N \N3 \N \LN ®_w
“ ‘“ ‘“ ‘ ‘ ‘ ‘ Om
ji_AW
Q
A \l[‘L*1A ?\ oxF
L1‘“‘lIx ‘i_“:‘i“iNt‘ ‘i‘ ‘ ‘ ‘ I Ov
A;
Q
W
W4I““‘
“ ‘W

M
E5_ _W
O_E ___§_eggmHmfig_mmgoog_w_omI
_W
\fl
gmH _ _”HH_HH“
WT” M
N
_w
_W _i__
WW _W
"7___ W
___N11
__W_ __W
_WW
W _W i W‘|1f1‘ ‘‘ ‘:‘:‘ :‘ ‘ ‘ ‘ ‘ ‘ ‘ ‘W‘
_ W _W _Z__WHW 32
_ __ flW_"
__
maSN
flu0%
QQ
Om

_W
__7 Wm
_in
W
W
_

w_ ___ _
__
_ W_
W_
W_ _TW PO
UCG
E5
9:m§OH_
kO%COMHOE
_> Omwi$_£9_w$m
%>&®EgH_m>_M;Emw

W
W
_
_H_ _
”___
_

_
_ 7”
fi _
W,
_P '40]
i
W
W__

g
EEgm;mOglECgw

_
”W_

Q0

jO

©Q

QQW gm gm“ “EN OQN 3? @©__‘ Om__ Q QQQ

w U O fin O W __AO U 8 n b 9 In inM O "I 8 D_ H M Imw V G hm”!U WM S W H

 HM 8 & H MV8 D r mm FO C 6 r
M MW MW

%HH““ i aA1 _ 2L
tI_
W1 WM I‘“1 __ __ ______: L WW HH W m HM HH E
A:
i _l‘
\“ IE QQ
Ofi
@fi

W\}Iii

_
_AA1‘ %_goé32%@5602
H
\?
rw \ _ W
_ __ i i_ __ _ TW
_:‘_“l““1 “‘ “ “1

568E5 gf
mmHUP
E5
DEM
®®O___O“Qwéw9
>_§®>E_©O_PH__CgO_>_
aO_EHtmw_‘ U¥mMw_£w‘6ED_‘ 2E“w
T ‘F‘ ‘ ‘ ‘ ‘ ‘ “‘13 _”_ __ WW _ 7%

HM“’“ 1‘ H

_ _ _
AA A}A\ “1
W_
J
WcodA‘ 1Q‘N1A 2
m8QQN2
mfimG

“““\‘ _ _ W 7V f ‘l _ ____ f Wfi _fi _ _ W_

QtEUEI
Emugm;gcgw

f? ‘\|_

“ “ “1‘i ‘

“ :“‘ “mi
‘

2““‘Hi;

gm mWA 1 g? :V“ “ii‘:!\u fi“

i“1

Om;
A‘W\H“ wi‘L%‘}15% mgAQ

”_
_

A}\ ;\ \

GOM“
‘A%\;AA}3%
mgA1
V wk‘A“l%i€{¥
? fi7

(1 }
J U O “___ O W rAO U 6 H b 6 J d N O 7‘ 9 p m_ M w V 6 hm QM nu W ma U
82 AM R E C G M M E N D E B _P R A C T Q E 2 S HR

M 88 H MV 6 D m F_ O r C G M ml S J

WEiIRA
IWA W
W AG %__egg
_w§H_mOcO_gm>_mAl gm:
WgmHU___ H‘“:i‘“\“T\ \ I ___ "__H_ W
__.HH__"_
W
__H H__ ‘W:A@
_%H@U __\1 ‘T
W ii
i_ W_
M?A iW_
N_ W_
__ _W T‘ Z IW;_:: Z_
_'‘‘ “ ‘1 ‘ W‘ ‘ i
W _1_ Ak \A W
Wm_
_W _ N AQL1 * LT
WW W( _ _ LiAWWA‘WAi:“H‘!\Ahié“ ;A mmgm
Q?
Gmmi“

”
j
__ W
IWW_

£5
COMHOE jwj
E6
fig
95mm
heC0302
®8__OH_
mmwwE
>m> £t_Ow®m¥w®_£ W_®ED3_wME@w

A ‘A ‘ A \‘ W1 _ A __ __ W

__ H _
ctEQQI
wgwimorémgw

“:“ }‘i“1lgr;_‘“A
(h:‘i3L§“

_ NI:
; Q\ UHWiwi “A L _ago
*0L wT
IH_ W
J
W __H‘
A
J"
N
jfifi :I1‘ ‘1 if gE2@O
mm

;_

‘QWi_i ‘:i‘

gm 83”“ 2:“ “EM Oqfi OQN QQN 9:‘ 2:1 S6 cg

W 1' W
J U O aO W InO U B H b 9 Ia N O ml
In 9p H H__ ‘miW V G 6‘ U S S W H
 APPENDIX D---DRAG EMBEDMENT ANCHOR DESIGN

D.1 Basic Considerations issues are addressed in the following sections which are
based on centrifuge and fiill scale tests, analytical investiga-
The holding capacity of a drag cmbedment anchor in a par-
tions, and field experience. A significant portion of the infor-
ticular soil condition represents the maximum horizontal
steady pull which can be resisted by the anchor at continuous mation is from Reference D.2 which presents the results of a
drag. This load includes the resistance to the chain or wire centrifuge test program and an analytical investigation using
rope in the soil for an embedded. anchor, but excludes the fric- limit equilibrium analysis. The study was conducted for two
tion of the chain or wire rope on the seabed. Drag cmbedment high efficiency anchors--Bruce FFTS Mark lll and Stevpris
anchor holding capacity is a function of several factors, Mark Ill. It is uncertain whether the results are applicable to
including the following: conventional anchors such as Moorfast or LWT.

1. Anchor type--fluke area, fluke angle, fluke shape,

anchor weight, tripping palms, stabilizer bars, etc. Fig-
D.3 Effect of Soil Shear Strength Gradient
ure D.l shows drag cmbedment anchors commonly
in Soft Clay
used by the offshore industry. According to U.S. Navy’s publication [Reference D. l], the
generic design curves in Figure D.2 are valid for soft clays
2. Anchor behavior during deployinent—-Opening of the
with shear strength gradients ranging from 8 psf/ft. to 16 psi?’
flukes, penetration of the flukes, depth of burial of the
ft. A study [Reference D.2] indicates that soil shear strength
anchor, stability of the anchor during dragging, soil
behavior over the flukes, etc. gradient may influence anchor holding capacity. Adjustment
of anchor holding capacity may be justified if the soil under
Due to the wide variation of these factors, the prediction of consideration is significantly different from the soils in which
an anchor‘s holding power is difficult. Estimates of anchor the anchor tests were performed. The U.S. Navy’s data were
holding capacity are normally achieved through empirical obtained frorn anchor tests carried out in soft clays with shear
approaches, as discussed in the following section. Analytical strength gradients of about 9 to 13 psf/ft. If the soil under
tools based on limit equilibrium principles for anchor embed- consideration has a similar shear strength gradient, no adjust-
ment and capacity calculation in soft clay are now available. ment is necessary. Otherwise, adjustment of anchor holding
However, exact holding power can only be determined after capacity according to the shear strength gradient should be
the anchor is deployed and test loaded. considered.
Limited analytical and centrifuge test data indicate that
D.2 Anchor Holding Capacity in Typical anchor holding capacity tends to vary more or less linearly
Soft Clay and Sand with the shear strength gradient value. However, this has not
Anchor performance data for the specific anchor type and been fully verified by field tests and deviations from the lin-
soil condition should be obtained if possible. In the absence ear distribution have been observed in the centrifuge test data,
of credible anchor performance data, Figures D.2 and D.3 depending on anchor size and soil strength profile. Therefore
may be used to estimate the holding power of anchors com- the adjustment of anchor holding capability is not a simple
monly used to moor floating vessels. exercise. It requires judgment from the designer. Reference
Figures D.2 and D.3 are primarily based on Techdata sheet D.2 provides some data and a discussion on limit equilibrium
83-08R, “Drag Embedment Anchors for Navy Moorings,” analysis which may be useful for this exercise.
Naval Civil Engineering Laboratory, 1987 (Reference D.l),
combined with industry anchor test data and field experience. D.4 Effect of Lead Line Type in Soft Clay
The design curves presented in these two figures represent in
general the lower bounds of the test data. The tests used to The design curves in Figure D.2 are based on testing of
develop the curves were performed at a limited number of anchors connected to chain. Field tests and analytical studies
sites. As a result, the curves are for use in generic soil types indicate that when the lead line is wire rope, an anchor may
such as “soft clay” and “sand”. Industry studies indicate, yield significantly higher holding capacity in soft clay. The
however, that several parameters such as soil strength profile, anchor was found to penetrate deeper into stronger soil. For
lead line type (wire rope versus chain), cyclic loading, and the limited cases studied, anchor connected to wire rope pro-
anchor soaking may significantly influence anchor perfor- vides l5% to 40% higher holding capacity than the same
rnance in soft clay. Also some high efficiency anchors have anchor connected to chain. This is in good agreement with the
demonstrated substantial resistance to vertical load in soft results from a full scale test program. It should be noted that
clay. Furthermore, there are new versions of high efficiency the studies have been limited to high efficiency anchors in
anchors which are not covered by these two Figures. These soft clay with a fairly constant shear strength gradient.
84 A W RE C OM M E N D E B D2 R A C T E E 2 S K

AN _A
‘
9‘Q;0AG AA
\J
A‘ “
‘_ iin " @E@Q \_N3MA
3WE
_H

MyR
é
g
QQ 1%
f
_Wm
_WWWWjm"_W"_
’”__ _W
_
__w

_C_
W fi
_

W
W
_
_ W

_WW"
_
WW"
J

_E®mE9___gH_
_m§n_
_%_O@_>n_WEm_'irO_£O C<

E‘@
 Desien AND Anarvsrs on Srarronxseerrse SYSTEMS son Froarezo $rnocrunss

51000 iii 1 r w" 1‘ 1""*"""“""1"""*"""'"—-

1 11 1 _?????? 1 j j j j j j1 __
1 11 1 1 _ _ 1
1 1 1 1 1 1

W
‘ni iji‘I‘
1 1 111 1

_1
1 1 11
., .1 »., . . . . . . ._
1 1 1111 1
I 11

K 1 1
<1 1(( ( 1
Fluke angles set for mud 1 1 1
seafloor condition as per 1 1 1 1
2:000 manufacturers‘ 3peQifiCafiQn_ W 1*" 127:3 ”””””””" T ********7”“ T

1 1
1 1
1 1 1

1,000 — 1oo o~ _ r. .M. 1 1 1111 1

» ooooo ~ i T 1
1
1re 1
1 1
1
»- 1
1 K
1
‘ N

1
1 __ _ 1 1
111
1 00000000011 1 s
1
1 _
~ 1-- 110000000000000
1 1 1 1 1 1~1 1
1

Weooooooooooooooooooooooo as 'e sfas 1 11 1 1 1

®~\Q1\ 1
5()g 1 1 i :1oooooooooooooooooooooo 1* 1 co\ 1
11 ; 1 1 1 1 \1%‘ %1 1 1
1 __ _ W 1 1 $1‘ Ge ‘
_J % % ‘ + 7 777777777 7 7 ..... ._ . ._.14,“

1 1 1 1 <<<<'<~ @1>oC5@;\‘<‘ 1 1 1 I-\P!F\I_.P'\VsP.~I 1_ . .1. .1. .1

k'ps)
\-._,,./

mum»
1 1 1
1<ej@@<;s@e (_\c,~e ¢\ 5
1 1 ‘
1
1 1 1 .\s’<>da'<. gxgst 1
1 Q3’ e~ \
200 1 111 1 1 1 1 1 1 1 1 1~ 1 1 1 1 as as <23 <>"-111” gl*\\ 1
Capac
ty
ng 1 11 e‘ C1 11 11
1 1 1 cl; 1 1
1 §§o{\@' :~$&‘
1 1
1 1 1

horH0d 1
1

oqlc,“ 1
1

An: 1 1 1 -65*
1001 ~ as 11-: ~
1
1
11 1= 111
11111111
. 1119'?’
1 1 111111111
1 '7" is
<<\ _ , . \ < 1
~ 111 *
11 11 11111 1 1 1
111111 111111
11 1
1 111 <<"» _ 1 1 1 1 11 1 1 1 1 1 1 111
1 V C5 1 1 1 :1
1 I 1 1 1 ’"1~ 1 1
11 11 1 1
T 1'
Til

,, 7 7 ,,,,,,,,,”_1,, ,,,,,,,,,,,,, 7 7 1 1 \ ,,,,,,, W ., "1.-. W \ , , ,,,,,,,,,,,,,, , ,,,,,,,__l_,,_ ,_:,_"_, . 1 , 1144‘_ ;_i1n n r

s 1 1 1 1 1 1? 1 1'
snip
1 1 1 '5 1 ¢5\
_, ,_, ,, , , ",1 ,,,,,,,,, ,, 1 W,,,,,,, 1”,” ,_,_ ' \&‘ ,,,,,,,,,,,,,,,,,,,,,,,,,,_,,, 1,, 1
50 1 1111111111111'1 1 iiiiiJ 1 7 1 1 1 1 1 11 W 1 1 11111111111"1111111111111111111111111111W

Go ‘*0
_ W 1W"_~ ,, H,”

1
,,_ K11111111111111111W_ _ ___ ,,,,,,,,,,W 4 ,

0f
‘"9
%,. 1
,,,_,WMm”1W,

J _ ,, ,,,,,,,,H
1 1 r 1 1 Wo
1(.9\O
er? 1111111111W 1
1 1 1 1 111 I
3 1 1 1
1 ?

20 111111111111as 1 L11111111W11 1 ~ *Anchors require specie

handling to ensure fluke tripping
(possibly fixed open flukes).

1 1 1
1 1 1
‘l 2 3 5 1O 20 30 56 100
Anchor Weight (kips)

Note: This figure was reproduced from Techdata Sheet 83, "Drag Embedment Anchors for Navy Moorings," Naval Civil
Engineering Laboratory, 1987, except that the holding capacity curves for the Moorfast (or Offdrill II) and the Stevpris
anchor were upgraded. The upgrading of these two curves was based on model and field test data and field experience
acquired i11 recent years. The design curves in the figure represent in general the lower bounds of the test data. They reflect
data valid for anchor designs as of 1987. New anchor designs have since been developed. However, performance data. for
these new designs were insufficient and therefore their design curves were not included. Some guidelines for the performance
evaluation of late anchor models are provided in D9. The design. curves do not include a factor of safety.
Figure D.2—Anchor Holding Capacity in Soft Clay
86 API RECOMMENDED PRACTICE 28K

5,000 11- ( 1 ------- 1—--~1111-111111 — 1 1

1 {{{{{ ‘W {{{{{if A if if
| L???????????????? if 1 * ” j j j1* *
1

I Fluke angles setforsand as per ‘ pk 3 1

2,993 manufacturers‘ specification.
1‘ i
-.0 - 1 “" " " - _ _
1 1. _

k / //1
_0__1nnn-|_7 11
‘ 1 1

,
1,000 K K 1 11 1
1
~~~~~~11 ‘
1
l
1jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj11
iii gm
1 //
‘I './
// i
11
i
"’
A

1 ""0 K _____ P __

jj1 1 1 as O0
11111 1 1

0000000000 0 es 1 1
00 0 00000000000000000 00 0 £26
65‘ /1 i1 = i‘ 001 , ...

509 00000 00 0 00000000000000000 100W0E0. 0. 0_l 0000 000000001 0 /AW 00000000000000000 0 0 ..'
1 if j x g gi g gigi*iiggiigg g gii g g "1jjjj ‘ j j j j A jjjjjjjjjjjjjjj l iiiiiiiiiiiiiii H iiiiiiiiW 1

kps)
---
I. _/ / 1 1

1
.

************11 1 1 1 ~<\<>°*6 11 1
'\-.,_,¢

nun»

1 1 / *3 1 0‘? 1
1 1 \> . '\ 1 84*@7-
209 0 00000000000000000000 00010 0 Q) ,@'~\\<\ \\§$ QS§?\ 100 10 00000 1
Capac
ty
ng
aim“
1 , A <5 ‘=5 1 1
1 3 1 <~.‘<‘*© 1~;\\\/ 1 1
sags
\
nonunion
/'

1 1 eéflil 1 1 1 0&1 OQ\{‘}\I£~’ 1 ‘

AnchorciHo
100 ‘ 0000000000000000000000000000 1 11111 0 ... 15/ 1 1 ?????????????11
‘ Q a ‘ 0 0,0000. 0 _
”i":‘ \"’ " " i ””*1” "”;\OQ, v$\ \
000000000000000000 __ 1 Q1 _ J 0
eOQ&c9 Po O ?\\\’$
<g<->° 9» 0 0 0 0 0 0 0 or 1 or 1/i>?> 0 0 0 0 0 0 0 0 0 or 100 0 is 0 0 0 0 is
6‘\.'27
K9 l
0 / 0 001 0 1 ‘
0 1
<5 0 ‘ 1
‘
0 ‘
0000000
50 _ _ ___100_0_0000000000 1 0 0 0___Q_)’\O 00000000000000000000000000000 00 000 0000000000 00 000000000 00 00 } ‘

1 1 1 g
1 11
1
. 111111 11 ~ 11 — 11 1""""""1171 1 — —— 2 1 1 ———————————————— 1 1‘ 1 1
1 ***********" 1******************W 1 1 2 ; 1

11 1
V if ‘ ? ? ? ? ? ? ? ? if ? ? ? ? ? ? ?if l ? ? ? ? ? ? ? ? ? ?U l * Up to 40% reduction in A
1-MW
1 -»1W.
capacity with fluke angle set at
35°.
20 1 11111111 1 11 1 1 1 1 1 1 1111111111 1
1 ** Up to 25% reduction in dense
1 nearshore sands.
1 1

1 2 3 5 10 20 30 56 1 100
Anchor Weight (Kips)

Note: This figure was reproduced from Techdata Sheet 83, "Drag Embedment Anchors for Navy Moorings," Naval Civil
Engineering Laboratory, 1987. The design curves in the figure represent in general the lower bounds of the test data. They
reflect data valid for anchor designs as of 1987. New anchor designs have since been developed. However, performance data
for these new designs were insufficient and therefore their design curves were not included. The design curves do not include
a factor of safety.

Figure D.3-—Anchor Holding Capacity in Sand

 Dacron nun Answers os Srarionassnino Svsraias roe Ftoarzuo Sraocroaas 87

D.5 Effect of Cyclic Loading in Soft Clay ity. This occurs because the disturbed soil. that surrounds the
anchor consolidates with time and thereby gets stronger.
The effect of cyclic loading was studied by centrifuge test-
ing [Reference D.2]. Three tests were performed at 32.2 g’s Anchor soaking was also studied by centrifuge testing. The
simulating 1.1-kip prototype Stevpris anchors. The first test tests were performed for l.l-kip and l5»~kip prototype
was a steady pull test performed to establish a baseline static Stevpris anchors. These tests typically were performed by
anchor capacity. The subsequent cyclic tests began with a pulling the anchor to capacity permitting it to rest in the soil
steady pull that approached the anchors‘ capacity, followed up for about an hour, and then resuming the pull until the ulti-
with a series of loading cycles, and ended with a steady pull mate capacity was achieved. ln this case, the one-hour wait in
to establish the post-cyclic capacity. A sample load-displace the centrifuge is equivalent to a wait in the field of about 247
rnent curve is shown in Figure D.4. lt was found that cyclic days. A sample load-displacement curve for the 15-kip proto~
loading did not damage the anchors capacity. Cycling actu- type Stevpris anchor is shown in Figure D.5. As is seen, the
ally improved anchor capacity by about 25 to 50 percent, by anchor soaking produced a substantial increase in capacity
causing the anchors to dig deeper and become embedded in upon re-initiation of movement, but this benefit quickly dis-
stronger soil. appeared as movement conti.nued. The anchor soaking effect
Also there are studies indicating that anchors can resist therefore appears to be localized to poi.nts near where the
higher maximum cyclic loads than static loads. Although the anchor initially rested within the soil, and is not dependable if
studies were conducted for pile anchors, we expect sirnilar there is any possibility that the anchor could be loaded
conclusions for drag cmbedment anchors. beyond its capacity resulting in dragging during subsequent
Based on above discussion and other industry experience loading.
for anchors in soft clay, the mooring test load requirement for
perrnanent moorings (Section 7.4.3) should be test loaded to D.7 Capacity in Soft Clay Under Inclined
at least 80% of the maximum storm load determined by a Line Loading
dynamic analysis for the intact condition.
In the centrifuge test program, the Stevpris Mark III and
Bruce FFTS Mark III anchors were pulled such that, at the
D.6 Effect of Anchor Soaking in Soft Clay end of the tests, the pulling line penetrated the mud.line at an
A “soaked anchor” one which h.as been embedded for angle of about 30°. The maximunt holding capacity
some time. In normally consolidated clays, the effect of earlier was maintained at this angle. This leads to the conclu-
anchor soaking presumably is to increase the anchor's capac- sion that these anchors are able to sustain significant vertical

9-8 *r""r r"

-.--l 1 1 1 1 1 I

0,7 11 1 1 1 11 1 1 1 1 1 1 1 1 1 1 ’ 1111 1 1 1 1111 11

1
Cycling Test ‘A’; 32-g Test 1 1 1 1
Q_6 0 00 011 000000 0 0 00 000 0 0 0 0 0 0 0 0 00 00000 0 00000

,-:-.‘ ,

N 0.5 1 1111111 11 1
1
1 3 1 1

(k
nee 0.41 11111111111111111111 1 1111 1 1 1 11 11 1
1 1
sta
O wu-
1
1
1
1 1 1
1

Fies 0.3 1 1 1 11111111 Increased capacity

1 ‘ vs. initial static
Initial Static Capacity A
Mnchor 0.2 1 111 1 1 /1
1 1- 1 1 1 1 1 1 1 1 1 1 111
Aode cycling
M 0.1 1 111111 11 111 ll\j

0.0 0 0000000000000000000
1
0 1
0I l
0000000100 0 0 0 0 0 -0-
1 10 Cycles 10.5"
110.1 1 11 1111 1 1 1 11111111 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1111111
-9 2
n
‘ ..............l..........................................._... 1
10 50 1 O0 150 206 250 300 356
Ti me (sec)
Figure D.41-1Effect of Cyclic Loading
88 API Rscoivsviaivoao F-‘sacrtcs ZSK

1 1 1
77-g Test
1i z 11
E

111111111 1 111 11
E

‘nlnlulu4"ic
1:1

/_'\

‘-.._/
1 Brittle during
1 * subsequent
0000 0 0 0 0 0 1 0 00000700000000000000011 1 I 0000000000000;.000000000000000000000000010 1111111 movement 0 0 0 000
22 Percent increase 1
kN
esstance
"T 000 0 0 0 0 0 000 1
R
nchor 1 "WWI "TIT F 1 HWITTJW
100 0 0 0 1
"1 iiiiiiiiiiTTTF I

1 Hour rest period; equals

l\/lodeA T’ 247 days in DFOIOTYDB W" 17””

-er;rrrr I 1 ~~;*"*""" "1 1 1 rrrr*1

‘ 2
1 1
-0.1 1 0000000 h l 00, ,0 0_00001. 01

1
1 1

O 100 200 300 400 500 600 700 8100

Displacement (mm)

Figure D.5--Effect of Anchor Soaking

load in soft clay. Full scale tests and field experience also sup- 4. The maximum line angle at the mud line (including the
port this conclusion. There were reports that the operation effect of wave and low frequency vessel motions)
personnel had difficulties to retrieve a high efficiency anchor should be less than 20°‘ u.nder the maximum design
after the completion of a drilling operation. In some cases the environment for the intact and damaged condition.
pendant line or the chaser line for anchor retrieval broke. In This angle should be zero at the early stage of test load-
other cases the anchors were retrieved by vertical force from ing to ensure anchor penetration. Furthermore, the
the mooring line in excess of 25 times the anchor weight. In a holding capacity should be reduced by a factor R,
full scale anchor test program carried. out in the Gulf of Mex- which is a function of the angle at the mudline and
ico, the pull out forces in the pendant line for the test anchors takes into account the reduced friction due to shorter
were close to their holding capacities. embedded line length. For example the following R
Significant evidence is present to support the allowance of values may be applicable for Bruce FFTS Mark IV and
vertical loads on some drag cmbedment anchors. Following Stevpris Mark V anchors:
are the guidelines for drag embedment anchors subject to ver-
tical loads. Muaiifiéfiiibgiief 0 it 5 T 10 i i i i i iT5 20
l. Vertical loads are applied to anchors under extreme
(Des) 0000000 ggg 000000 ,,,,, g , , _,
R Z W1.0 . _0.95 0_.s9 0.s1W_;
environment only. Drag cmbedment anchors should
not be subjected to vertical loads under normal operat- The factor of safety should still be satisfied after reduction
ing environments.
of holding capacity.
2. It is applicable only to certain high efficiency anchors There are cases for permanent moorings where a taut-leg
for which sufficient research has been conducted and mooring spread may be utilized. In such a configuration, the
much field experience has been gained. mooring lines will have an initial angle to the seabed and will
3. The anchors are deployed in soft clay where deep pen- always have vertical and horizontal loads imposed at the
etration is expected. This may exclude certain anchor. It has been common practice to use pile or plate
operations with mobile moorings where the soil condi- anchors to resist these loading conditions, and design guide»
tions have not been thoroughly investigated or the lines for these anchors are provided in Appendix E.
anchor test load is insufficient to ensure deep
penetration.
 DESKSN ans Assatvsis or Srarionxaaasvo Svsraivis son Fi.{}P=Ci‘¥NG Sraocroass 89

D.8 Anchor Drag Distance and mation, the holding capacities of these new versions can be
Penetration Depths in Soft Clay estimated using the following equation:
Anchor drag distance and penetration depth estimates from
Reference D.l are presented in Figure D.6 and Table D.l. He I Ha -(A12/As) H (D.l)
Again, this information is valid for chain lead line and a shear
strength gradient of 9 to 1.3 psf/ft. Deviation from this may where
affect these values, especially the penetration depth estimates. H” I holding capacity ofnew version,
Table D.1-W-Estimated Maximum Fluke Tip Penetration HS I holding capacity of reference (for example
i\lorrna.lized Bruce FFTS Mark III or Stevpris Mark III in
Fluke Tip Penetration Figures D.2 and D.3) anchor of same weight,
(Fluke Lengths)
An K fluke area of new version,
Sands/Stiff Mud
Anchor Type Clays (e.g., Soft Silts and Clays)
AS e fluke area of reference anchor of same weight,
Stockless I 1 iiiiiiiiiiiiiii 3*

rt I 1.4 for commonly used high efficiency anchors.

Moorfast l 4
The fluke area ratio An/AS can be obtained from anchor
Offdrill II
manufacturers.

Boss D.10 Analytical Tools for Anchor

Danforth Performance Evaluation in Soft Clay.
Flipper Delta Analytical tools based on limit equilibrium principles for
GS (Type 2) 1 4%, anchor cmbedment and capacity calculation in soft clay are
LWT now available. These tools allow modeling of different
._.-..1-..-_ .1__:.__...a “...: ._....--:.1_ ..‘l..4...I1..-..] .._-_1--..- --3--r..-....-.........
Stato ictl1U.l1U1' UU5l.g,l1i'>' EJHIU PIUVIUU UV/L'£11.1€JU ?.1l1U1lU1' [)lo1‘1Ui'11'1'ct1.lUCJ

Stevfix information such as anchor movement traj ectory, anchor rota-

tion, mooring line profile below seafloor, and ultimate anchor
capacity, etc. However, there are certain requirements for
Stevpris MK III
these tools to yield reliable predictions:
Brace FFTS MK III
Bruce TS 1 5 1. The analytical tool needs to be calibrated by field or
Hook centrifuge test data for the anchor of interest.
Stevrnud 2. The analytical tools require that the soil properties are
well known, which may not be the case for many drag
*Fixed fluke stockless. cmbedment anchor applications. If there is uncertainty
in the soil properties, then suitable upper and lower
bound soil parameters should be established, and the
D..9 New Anchor Design anchor design should be based on the more conserva-
Driven by economics and frontier development, new tive predictions.
anchor designs and improvements to existing anchors con- 3. The user of these tools should be aware of the lirnita~
tinue to evolve. For example, Bruce FFTS Mark IV and tions of the tools and the mooling operation. For
Stevpris Mark V are advanced versions of the Bruce FFTS example some tools typically show that the anchor
Mark III and Stevpris Mark III anchors, for which design continues to penetrate gradually to a very deep penetra-
curves are presented in Figures D.2 and D.3. Also other new tion, picking up more and more anchor holding
anchor designs have become available. Anchor test data and capacity. In this case the user should consider limiting
service experience are insufficient for these new designs and the drag distance for calculating the anchor holding
therefore design curves cannot be presented at this point. capacity to a distance that will not result in unaccept-
Studies have shown, however, that the efficiency of the new able vessel excursion.
version is improved through improving the fluke area to 4. Ernpirical formulas or field experience, if available,
weight ratio. Based on this and in the absence of better infor— should be used to check the analytical predictions.
90 API Rscommsnoeo PRACTICE ZSK

Tenant

o
tn...
l @
@
l

@ I
Curve Anchor
1. Stockless (fixed)

2. Hook
SafgetyFactor
3. Bruce, F FTS MK Ill
Brace TS
Danforth
Max
Capac
ty
ceatof
mum GS (type 2)"”
LWT*
Per Moorfast Cor espondn
Navmoor
Offdrill ll*
Stato
Stevmud
. 1..1-.. Stevpris MK lll
4. Boss*
l
l
l Flipper Delta*
Stet/dig*
Stevtix
Stevirf‘
*Assumed based on geometric similarities. ---»

i
r
l

10 20 30 40
Drag Distance/Fluke Length

Figure D.6—Percent Holding Capacity versus Drag Distance in Soft Clay

 Dearest ano Anarvsis or-* 8TAT¥ONKEEPii\£G Svsrervis son Fi_£>AT¥NG Sraocruass 91

5. The analytical tools may not be suitable for layered soil F = coefficient of friction between chain or wire
profiles. rope and the ocean bottom, dimensionless,

D.11 Anchor Holding Capacity in Sand .Lcw 3 length of chain or wire rope in contact with the
ocean bottom, ft,
No significant study on the behavior of drag cmbedment
anchors in sand has been carried out since the U.S. Navy's WCW I submerged unit weight of chain or wire rope,
Study. Therefore, the design cu.rves presented in Figure D.3 lb/ft.
still represent the best available third party information on
anchor holding capacity in sand. A nnchors d.o not penetrate The coefficient of friction,f depends upon the actual ocean
deep in sand and therefore will not provide substantial uplift bottom at the anchoring location and type of mooring line.
resistance (Table D.1). Design for vertical loads, in this case, Generalized friction coefficients for chain and wire rope are
is not recommended. given in the following table. The static (starting) friction coef-
ficients are normally used to compute the holding power of
the line and the sliding coefficients are normally used to com-
D.12 Anchor Holding in Soils Other than pute the friction forces on the line during mooring deploy-
Soft Clay and Sand ment. These generalized coefficients can be used for various
The above discussions address anchor holding capacities in bottom conditions such as soft mud, sand, and clay if more
soft clay and sand. There are other soil conditions such as specific data are not available. Industry experience indicates
hard clay, calcareous sand, coral or rock seafloor and layered thatf can vary significantly for different soil conditions, and
soil profile. Predicting anchor holding capacity under these much higher slidingfvalues have been recorded.
conditions is more complex, and anchor design guidelines are
not available at this point. We hope that more research for Coefficient of Friction. (f)
these conditions will be carried out and some guidelines can Static Sliding
be developed in the future.
Chain l .0 0.7
Wire Rope 0.6 0.25
D.1-3 Holding Capacity from Friction of
Chain and Wire Rope
The holding capacity from friction of chain and wire rope D.14 References
on the seafloor may be estimated using Equation D.2. D.1 “Drag Embedment Anchors for Navy Moorings,” Naval
Civil Engineering Laboratory Tech Data Sheet 83-08R, June
Pcw :fLCWWCW (D.2)
1987.
D.2 Dunnavant, T. W. and Kwan, C. T., “Centrifuge Model-»
where ing and Parametric Analyses of Drag Anchor Behavior,”
Paper No. 7202, Offshore Technology Conference, Houston,
Paw e chain or wire rope holding capacity, lb} May 1993.
 APPENDIX E--PILE AND PLATE ANCHOR DESIGN AND INSTALLATION

E.1 Basic Considerations cal and l'1.0I‘iZOl].lEIl extent of the final geotechnical investiga-
tion (i.e., number, depth, and location of soils borings andfor
E.1.1 PURPOSE AND SCOPE
in-situ tests such as PCPTs) and to aid in the interpretation of
This appendix addresses a number of design and installa- the acquired geotechnical data. Previous site investigations
tion issues for driven piles, suction piles, and plate anchors, and experience may permit a less extensive site investigation.
all of which are capable of resisting vertical loads. The issues Some examples of these integrated geoscience studies are
include soil investigation requirements, anchor capacity eval» given in [E22] and [E23].
uation, structural design, fabrication, handling and transporta»
tion, installation, and pull testing. A general description of E.2.1.2 Soil Sampling and Laboratory Testing
these anchors can be found in Appendix A, Section A.3, and
Should the designer choose to rely on soil sampling and
the geotechnical factors of safety for these anchors are pre~
laboratory testing instead of in~situ testing during design, the
sented in Section 7.4 in the main body of this recommended
designer should be aware that the measured properties of soil
practice.
samples retrieved from deep waters may be different from in-
Some of the technological aspects of the design of suction
situ values. Without special precautions, the relief of hydro-
piles and plate anchors are still under developnient. Specific
static pore pressure and its resulting effect on any dissolved
and detailed recommendations are given in this appendix to
gases can yield soil properties significantly different from in-
the extent currently possible. General statements are also
situ conditions. Because of these effects, in-situ or special
used to indicate that considerations should be given to some
laboratory testing to determine soil properties is warranted.
particular points, and references are given for further guid—
Some of the geotechnical tools available when rotary drilling
ance. Designers are encouraged to utilize all research
techniques are employed for deepwater investigations are dis»
advances available to them. Designers can find an additional
cussed in [E24]. Coring with “jumbo” or “long” coring
discussion on plate anchor design in [E35].
devices has also been shown to provide shear strengths equiv»
alent to those obtained by rotary drilling methods and holds
E.1.2 FOUNDATION REQUIREMENTS
‘E3Q 22' OD (e as an alternative coring method [E25] and [E26].
The design of the foundation structure should ensure that,
in the anchor and the soil surrounding it, permissible limits of E.2.1.3 In-Situ Testing
stress, displacement, and fatigue are not exceeded during and
ln—sitn testing may allow a more reliable estimate of soil
after installation. The foundation systern above the rnudline
parameters and alleviate issues with sample disturbance. Typ~
should include provisions for inspection and maintenance.
ical tools used include: the remote vane (either seabed or
The extent of inspection, timing of the inspection, and main-
downhole units), the piezoprobe (to obtain estimates of in»
tenance should be commensurate with the redundancy rela~
situ pore pressure and permeability), PCPT (Cone Penetrome»-
tive to overall safety and performance.
ter Tests equipped with pore pressure transducers). Advan-»
tages of the PCPT include obtaining a continuous profile of
E.2 Site investigation soil resistance that allows for detailed stratigraphy. A detailed
E.2.1 PERMANENT MOORING discussion of PCPT data interpretation can be found in
[E32]. Other promising tools include the T-bar penetrometer
E.2.1.1 General Consideration [E48].
The areal extent of the foundation system for floaters
greatly exceeds that of fixed structures and TLPs. Require»- E.2.1.4 Recommended Sequence for Site
ments for site investigations should be guided primarily by Characterization
the type of platform to be installed, the availability and qual-
A site investigation program should be accomplished for
ity of data from prior site surveys, and the consequences that
each platform location. The program should, as a minimum
would result from a partial or complete foundation failure.
and preferably in the order listed, consist of the following:
It is recommended that a high-quality, high~resolution geo-
physical survey be performed over the entire areal extent of a. Background geophysical survey
the foundation. This survey should then have a realistic geo- Regional geological data should first be obtained to pro»-
logical interpretation and be integrated with the possibly vide information of a regional character winch may affect the
existing geotechnical data to assess restraints imposed on the analysis, design and siting of the foundation. Such data
design by geological features. Such an integrated study can should be used in planning the high-resolution surveys and
then serve as a guide to develop a scope of work for the verti— geotechnical site investigation, and to ensure that the findings
94 AP! Rscotiiisuoao Pnacrica 28K

of the subsurface investigation are consistent with known The minimum vertical extent of the site investigation
geological conditions. Site-specific background data should should be related to the expected zone of influence of the
include a re-examination of the 3-D, multichannel data loads imposed by the base of the foundation and must exceed
obtained for exploratory purposes and a review of the “geo- the anticipated design penetration by at least the anchor diam-
hazard” study used to site the exploratory wells. The 3-D data eter or anchor fluke width, B (see Figure E7). If Reverse End
set may be re-processed to enhance its high frequency con- Bearing (REB) at the suction anchor tip is to be taken advan-
tent. Suggested reading for further information is given in tage of in the vertical capacity analysis, soil characterization
[E.27]. up to three diameters for suction piles or three fluke widths
for plate anchors below the design penetration depth may be
b. Seafloor and sub-bottom survey
more appropriate. It is critical to ensure that no high-perme-
A site--specific, higheresolution geophysical information ability layers are present within the zone influenced by the
should be obtained relating to the conditions existing at and mobilization of the REB, particularly if the anchor is to resist
near the surface of the seafloor. The survey should include the long~duration loads such as those imposed by loop currents.
mapping and description of all seafloor and sub-bottom fea- If the soil investigation is performed primarily using PCPT,
tures that may affect the foundation system. Such features it is recommended that at least one boring and/or long core be
may include: seafloor contours, seabed slope angles, shallow taken to properly calibrate the PCPT results. This boring/core
stratigraphy, position of bottom shapes which might affect should be taken at one of the PCPT locations.
scour, boulders, obstructions, and small craters, fluid expul- The site investigation should also consider that during the
sion features, pockmarks, shallow faults, slump blocks, drill detailed platform and mooring design process, the seabed
cuttings plume, previous usage of seafloor, and gas hydrates. location of the anchors may change due to changes in moor-
The survey should use geophysical equipment and prac- ing lines lengths and/or headings, field layout, platform prop-
tices appropriate to the water depth of interest and provide erties, and mooring leg properties.
high»~»resolution imaging of the seafloor as well as detailed Some examples of the scope of deepwater investigations
stratigraphic information to a reasonable penetration below are given in [E.2] and [E33] and examples of data interpreta-
the zone of influence of the structure. tion are given in [E28] and [I-3.29].
c. Geotechnical. investigation d. Soil testing program

The sampling and in-situ testing intervals should ensure The main goal of the laboratory testing program should be
that each significant stratigraphic l.ayer is properly character- to properly evaluate all input parameters required for geo-
ized. The design soil parameters in various soil strata should technical and structural design, for all significant strata.
be determined from a field program that tests the soil in as When applicable, testing should be performed in accordance
nearly an undisturbed state as feasible. Because the quality of with recognized standards (i.e., ASTM or others).
soil samples can be expected to decrease with increasing Additional testing should be performed to define the creep
water depth, the use of in-situ testing techniques is encour- and cyclic behavior of the soil to allow prediction of soil
aged for deepwater sites. In addition, soil samples may be structure interaction due to sustained and cyclic loading. Con»
required to provide advanced engineering soil properties. sideration should be given to the performance of permeability
The scope of the geotechnical site investigation (i.e., num- and consolidation tests in order to understand set-up effects
ber, location, depth of bo-rings and/or long cores and/‘or for driven piled structures and capacity consideration for suc-
PCPT, etc...) will depend on the mooring system, and the tion piles and suction caissons.
quality and interpretation of the high-resolution geophysical In all»-clay profiles, the site investigation and laboratory
study (i.e. inferred lateral variations in soil properties). Typi- testing program should consider providing the following
cally, soil characterization (i.e., boring, long cores, or in-»situ information needed for the reliable design of pile and plate
tests) is performed at each anchor (or at each anchor group anchors, as applicable for the type of anchor, size, and anchor
for a group mooring pattern), or at least at two locations over loading:
the anchor pattern if the interpretation of the high resolution ~ General soil description, classification, and index
survey indicates little variation in soil properties across the testing.
pattern.
~ Soil stress history and over-consolidation ratio (OCR),
However, if high-quality geotechnical data already exist in
soil compressibility (i.e., unload and reload moduli), as
the general vicinity of the anchor pattern and little variation
measured in Constant Rate of Strain (CRS) tests or
of soil properties is inferred over the areal extent of the foun-
constant load tests.
dation, or if extensive experience with the chosen foundation
concept in the area can be drawn upon, the above recommen- ~ Soil permeability.
dations may be modified as appropriate [E.33], [E.3-4]. ~ Remolded shear strength and soil sensitivity.
 Dsszon AND Answers or Srarionxaasirio Svsraras son Fi_GATi¥\iG Srsocruaas 95

~ Monotonic and cyclic shear strength under appropriate data. The design and selection of pile and plate anchors
average and cyclic stresses for triaxial compression, should then be based on the best available soil data from
extension, and DSS stress paths; Samples should pref- nearby surrounding areas. Such inforrnation should be inter-
erably bc anisotropically consolidated and cyclic tests preted by recognized geotechnical experts whenever possible.
should preferably be performed at the expected load
period. E.3 Geotechnical Design of Suction Piles
' Creep data to define possible loss of shear strength
under sustained load. (in cases where large sustained.
E.3.1 BASIC CONSIDERATIONS
loads, e.g., loop currents, are important). Cyclic The design of suction anchors for floating systems includes
stresses should be superimposed on the sustained the following aspects: penetration and removal, capacity, and
stresses if relevant for the actual load conditions. soil reactions or soil structure interaction analyses for struc-
' Remolded soil consolidation characteristics (compress- tural design. In areas, such as the Gulf of Mexico, where trop-
ibility and permeability). ical cyclonic storms may exceed the capacity of the mobile
~ Reconsolidated remolded soil strength characteristics. mooring or mobile anchoring system, the design of suction
piles should consider an anchor failure mode that reduces the
- Soil thixotropy. chance of anchor pullout.
~ Parameters needed for generation of P-y curves (i.e., The calculation of the geotechnical holding capacity of the
50% strain factor, 85,9). anchor should be based on best-estimate soil properties.
Database for cyclic soil properties are available in [E2] Anchor adequacy with respect to installation should be
and [E30] for Gulf of Mexico clays. Such database should be checked against upper bound soil strength properties.
used to interpret tests results and reduce the number of site- If the uncertainty in the geotechnical data is greater than
specific cyclic tests. typically encountered (i.e., unusually large scatter in shear
strength measurements), consideration should be given to
e. Additional studies increasing the safety factors given in Table 7 of the main text.
As applicable, additional analytical studies or scaled tests The impact of the mooring line geometry in the soil on the
should be performed to assess the following aspects: anchor loads should be considered, since this geometry may
change the relationship between the horizontal and vertical
' Scouring potential anchor loads. The inverse catenary of the mooring line in the
~ Earthquake ground response studies or analysis soil may make the mooring line angle steeper at the anchor
padeye than at the mudline. This steeper angle could result in
' Seafloor instabilities in the area where the foundation a reduced horizontal load but an increased vertical load at the
system is to be placed anchor padeye. Both an upper and lower bound inverse cate-
~ Set-up effects nary should be checked to ensure the worst-case anchor load-
ing is established.
For drag embedded plate anchors, an alternate design phi-
losophy, currently used mainly in Brazil and South East Asia,
has consisted of performing a reduced site investigation, E.3.2 ANALYSIS METHOD
gaining experience with the site specific performance of plate E.3.2..1 Penetration Analyses
anchors through extensive load testing, and performing proof
load tests after the anchor installation to at least 80% of the A typical penetration analysis includes the calculation of
maximum storm load. determined by a dynamic mooring three quantities, for all penetration depths. These are:
analysis for the intact condition. ~ The penetration resistance exerted on the anchor by the
soil.
E.2.2 MOBILE MOORINGS
' The required underpressure to allow anchor embed-
If detailed site-specific soil data are available for a mobile merit.
mooring location, this information should be used for the
design and selection of the mooring anchors. However, this - The critical pressure that will cause the soil plug to fail.
information may not be available for some mobile mooring It is of paramount importance to properly estimate the
locations such as drilling locations in new exploration areas. underpressure required to penetrate the pile to its design
The only geophysical data available may be the 3-D multi- depth. Required underpressure is a critical input parameter to
channel data obtained for exploration purposes. Evaluation of the structural design of the anchor and it must also be verified
geohazards and constraints to the mooring system should still that the predicted under pre sure can actually be generated by
be performed, based on a re-examination of the above 3-D the pumps to be used during installation.
96 API RECOMMENDED PRACTICE 28K

E.3.2.1.1 Penetration Resistance presently the most common. The typical range of (rm, for
Gulf of Mexico deepwater clays is 0.2 to 0.5.
The penetration resistance can be calculated as the sum of
the side shear and end bearing on the side Wall and any other There may be uncertainty in the sensitivity since it is influ-
protuberances. Protuberances that might be present include enced. by the quality of the intact strength that it is related to.
mooring and lifting padeyes, longitudinal or ring stiffeners, Alternatively, the side friction, (1.,-CC - Supgg, can be equated to
changes in wall thickness, mooring chain, launching skids, the direct measurement of the remolded shear strength,
and others. through fall cone, UU triaxial, or minivane tests. The
For an anchor penetrated in clay without protuberance and remolded strength used in design should reflect both the
with a flat tip, the installation resistance, at a given penetra- directly measured value and the value derived from the intact
tion depth, Z, can be calculated by Eq. E.l strength divided by the sensitivity.
Some installation records have, however, shoyvn that the
interface shear strength mobilized during installation can, at a
Q20: x Qsz‘rie + Qti;J
given depth, be less than (ling - Snags. In cases where the full
Qizae E Await? ' (Oi»m.¢ S7/inssla VB 1) interface shear strength, o.,-,.,_,., may not be mobilized along the
anchor wall, such as when the anchor is painted or subjected
Qtip 2 (Nrr ' + “fl i Z) ' Arip
to unusual surface treatment, a correction factor may need to
be applied to the factor to properly predict the penetration
where
resistance [E.9] [E.52]. Ring shear tests, with the actual wall
QIC, I Total penetration resistance, surface modeled in the tests, may be used to measure the
actual interface shear strength.
QCMC K Resistance along the sides of the pile,
QC}, Z Resistance at the pile tip, E.3.2.1.1.2 Bearing Capacity Factor, NC
AW” e Sum of inside and outside Wall area embed- The values of the bearing capacity factor NC to be used to
ded into soil, calculate the penetration resistance of the anchor tip or of a
given protuberance depends on the shape of the protuberance
AC,-C s Pile tip cross sectional area, and the ratio of the width of the protuberance over the
o.,~,,C K Adhesion factor during installation (see embedment depth of the protuberance. Values of NC ranging
from 5.1 to 9.0 for round and strip footings are recom-
Section E.3.2.l.l.1),
mended in [E. 31].
Saggg e Direct simple shear strength, Because the anchor vi/all thickness is usually small corn-
pared to the anchor diameter and the cmbedment depth, the
Otfm SHDSS K fZ'iCiiOI1,
pile tip area is usually considered to be a deeply embedded
(oi,-,.,C - SuD5;g)M;e Average side friction from mudline to depth strip footing with an associated NC equal to 7.5. The values of
Z1
NC to be used in Eq. E.l are summarized in Table E.l.
A detailed example of NC calculation is given in [E.ll].
NC I Bearing capacity factor (see Section Values of NC different than those of Table E.l are acceptable
E.3.2. 1-1.2), provided that they can be documented by appropriate model-
ing and test results.
Suf,-EB H Average of triaxial compression, triaxial
extension, and DSS undrained shear
E.3.2.1.1.3 Changes In Penetration Resistance
strength at anchor tip level,
Due To Protuberances
y‘ X Effective unit weight of soil, Equation E.1 should be modified if protuberances are
present. The change in penetration resistance d.ue to the pres-
z ITip penetration depth.
ence of mooring and lifting padeyes, longitudinal or ring stiff-
eners, mooring chain, launching skids, pile tip other than flat
E.3.2.1.1.1 Adhesion Factor During Installation,
(i.e., beveled) or any other internal or external protuberance
gins should be considered carefully and require assessing the
The adhesion factor during installation, ot,,,C, is usually changes in friction and end bearing resistance caused by the
defined as the ratio of rernolded shear strength over undis- protuberances. Most protuberances will cause an increase in
turbed shear strength, that is as the inverse of the soil sensitiv- penetration resistance, except for internal ring stiffeners,
ity. The adhesion factor can be determined by various which may cause a decrease in internal side friction if they
methods but fall cone, UU triaxial, and miniature vanes are are closely spaced [E36], [E.54].
 Gaston AND Auatvsts or Srarionxeseiaio Svsreais son Fi.OA'i”¥¥\iG Sraucruaas 9?

Table E.1--Recommended NC Factor

I Purpose , Shape of area NC
Calculation of pile tip penetration resistance i Strip 7.5 A
a.la.laa;;?;;Za;a underpressure causing an plugii i i if if caaiiai if ST“ i6.2to9.0dependirrg on embedirieritiiiratio
I failure (Section E.3.2. 1.3) A i“ tE.3l l
Calculation of penetration resistance of protuberances Varies 5 to 13.5 [E.36]
(Section E.3.2.1 .1.3)

E.3.2.1.2 Required Underpressure etiy, mode of penetration (i.e., self-weight penetration vs.
penetration by underpressure) [E37]. It is commonly
The required underpressure, aU,.C,,, to embed the anchor
can be calculated as follows: assumed the 50% of the soil displaced by the cross sectional
area of the anchor will go inside the anchor during self-
weight penetration if the tip of the anchor is flat.
,__ Q-o W W! -
AUrer] W mmgmmwm The final elevation of the internal plug surface may also
depend. on the wall thickness variations, internal soil plug sta-
bility, spacing and type of internal stiifeners [E37].
where
Soil heave should be accounted. for in calculating the
Qm e Total penetration resistance, required pile stick-up and total length.

W’ e Submerged weight during installation,

E.3.2.1.5 Items of Special Consideration
AC, K Plan view inside area where underpressure is
Sand layers: If sand layers are present, they should be
applied.
given special attention. The penetration resistance in layered
profiles consisting of interbedded sands and clays may be sig-
E.3.2.1.3 Critical and Allowable Underpressures
nificantly higher than through clay, depending on the density,
The critical underpressure at a given depth, AUCC,-Z, defined degree of cementation, grain size distribution, thickness,
as the undeipressure that will cause a general reverse bearing spacing, and depth of the sand layers. The penetration rate
failure at the anchor tip and large soil heave within the through sand layers will need to be high enough to prevent
anchor, can be calculated at a given depth as follows: excessive flow of water through the sand layers ahead of the
anchor tip, as this may cause large plug heave.
AU(:rif x NC . A- . - . ~s
H'i.§‘Iff€ CCCCCCCC .
E.3.2.2 Removal Analyses
The geotechnical analysis should also consider anchor
where
retrieval for the following cases:
Am,-C=C re Inside lateral area of anchor wall. * Mobile moorings where anchor removal is needed for
reuse of the anchor or to clear the seabed. The suction
In shallow water one must check that the critical under-
pile retrieval procedures and analysis should account
pressure does not exceed. the water cavitation pressure.
for the estimated maximum setup time;
The recommended allowable underpressure, AUC”C,,,,
defined as the maximum underpressure that should be applied ~ Permanent moorings where it is required by authorities
to the anchor, can be calculated as the critical undeipressure that the anchors be removed after the system service
divided by an appropriate factor of safety. This safety factor is life. The suction pile retrieval procedures and analysis
typically a minimum of 1.5. Lower values may be acceptable should be based on full soil set-up.
provided that, during installation, the plug behavior is moni- - Mobile or permanent moorings: installation tolerances
tored and it is confirmed that no plug failure occurred. are exceeded, a mooring line is damaged during instal-
lation and for other contingencies.
E.3.2.1.4 Soil Plug Heave Inside Anchor The overpressure required to retrieve the anchor,
(AUrCC),.C,,, can be calculated by Eq. E.4.
The soil heave inside the anchor during installation may be
estimated by assuming that a percentage of the clay volume
displaced by the cross sectional area of the anchor goes inside 0 (5 F it-l “L W’
(A Ureq)retr in Qt I
the anchor. This percentage may depend on: anchor tip geom-
98 API RECOMMENDED PRACTICE ZSK

where described in API RP 2A are also considered adequate if suit-

ably modified. A method to modify P-y curves to account for
Q,;,._,,i(t 3 1;.) I Total soil resistan.cc at time of retrieval, 2%,.
the larger diameter of suction piles and to ensure lateral
Time re 0 is defined as the time at the end of
deflection is not overestimated can be found in [E5]. The
penetration,
merits and shortcomings of each method are discussed belovv.
W ’ ==Subrnerged weight during retrieval,
E..3.2.3.1 The Finite Element Method (FEM)
A5,, n Plan view inside area where overpressure is
applied. As discussed in [E36], the finite element method is the
most rigorous general method of analysis available for com-
plex structural systems (including soil continua and soil-
When calculating the total soil resistance during retrieval, structure interaction). The FEM will find the critical failure
Qm,(t I tr), Eq. E.l can be used with some modifications. It mechanism without prior user assumptions, provided an
should be noted that the interface shear strength might be appropriate constitutive model is used. The FEM also has
higher than its values during installation, due to soil set»up. many advantages including the ability to include complex
Paragraph E.3.3 gives guidance on assessing the increase in geometries, spatially varying soil properties, and non-linear
adhesion factor vvith time. The designer should also be constitutive behavior vvith failure criterion, to name a few.
mindful of possible differences between end bearing resis- Major disadvantages include the required specialty knowl-
tance in tension and compression for the protuberances that edge of advanced numerical analysis and the large time
may be present. In addition, the maximum extraction pres- investment to set up a model.
sure used should not be higher than the pressure causing soil In ductile plastic systems (foundations on soft clays are
plug failure. usually in this category) the ultimate capacity of the system is
The vessel removing the anchor is often capable of apply- independent of the sub-failure properties (e.g., Young’s mod-
ing a lifting force on the anchor with the recovery line. This ulus, Poisson’s ratio) [E38]. It has been shown that carefully
assistance can significantly reduce the required extraction formulated and executed analyses give system load carrying
pressure and should be included in the removal analysis. capacities that compare favorably with the few exact, analyti-
Therefore, if a load is taken by the lifting wire during cal solutions available [E39].
YCt.1i€'v’Ei.i, that 1O3'tu Cfifi uC Si.1utI'E'iCi.Cu 1YUi"i'i £116 fihfiiulatur 111
FEM programs are widely available and have been used to
Eq. E.4. advantage for assessing specific suction piles configurations,
The effect of the maximum extraction pressure on the steel matching the few experimental results available, and provid-
structure of the suction pile should be considered (see Section ing ground truth calibration for simpler models. As men-
5.2.3 and 5.2.5). tioned above, such analyses require special expertise and a
significant investment in time and are therefore not yet Well
E.3.2.3 Holding Capacity suited to parametric studies or conventional design iteration
Analysis and design tools to determine the capacity of suc- (such as are required for finding the optimum anchor line
tion anchors can be classified as one of three general methods attachment point).
[E36]. These are, in order of detail: FEM analyses may however be warranted for complex
load and/or soil conditions where little experience is avail-
' The Finite Element Method (FEM) advanced numeri- able, or to gain insight on specific behavioral aspects of the
cal analysis, foundation (i.e., assessment of pore pressure changes and
~ Limit equilibrium or plastic limit analysis methods effective stress path at any point within the soil mass).
(models involving soil failure mechanisms),
E.3.2.3.2 Limit Equilibrium or Plastic Limit
* Semi-empirical methods (highly simplified models of Analysis Methods
soil resistance including beam column models).
As discussed in [I-3.36], these models are more approximate
For the analysis and design of suction anchors for anchor- than FEM) models but are generally much easier to use than
ing deepwater floaters, the central focus is the ultimate capac- general FEM programs. The methods involve estimating the
ity of the suction anchor and not the load deflection behavior. ultimate capacity of plastic systems using assumed failure
It is recommended that suction pile design for permanent mechanisms. These mechanisms are typically based on a
moorings use FEM, limit equilibrium techniques or limit combination of experimental observation, more rigorous
analyses (see Sections E.3.2.3.l and E.3.2.3.2). For mobile numerical or analytical studies, and engineering judgment.
moorings with mainly horizontal loads, semi-empirical meth- These methods may also include the ability to incorporate
ods such as beam-column analyses (see Section E.3.2.3.3) complex geometry and soil strength variability and do not
using load transfer-displacement curves (i.e., P-y, T-Z, Q»z) require characterizing sub-failure behavior.
 Dusters nun Auarvsls or Srariouseeeiue Svsrsns roe Froarzuo Srnuctunes 99

Disadvantages of these methods are the approximate In general both limit equilibrium and limit analysis meth-
nature of the analysis and the difficulty of generalizing ods give upper bound estimates of ultimate loads such that
results, i.e., the need to calibrate the models to experiment or minimizing the ultimate load with respect to the geometric
more rigorous analysis for specific structural configurations parameters gives the “best” answer for the particular mecha-
and soil profiles. For example, changing either the soil nism. However, the “best” answer may or may not be close to
strength profile, the anchor geometry, the load inclination, the exact answer depending on the assumed mechanism. In
the load attachment point, the load type (i.e., duration, fre- the limit equilibrium method the result will not be a true
quency, rat.io of cyclic to mean load component, etc.) may upper bound if the mechanism. does not satisfy kinematic
require a change in the basic geometry of the assumed failure constraints. A discussion of these methods is provided in
mechanism. [E38].
In general there are two approaches that can be taken using
There are a number of existing computer programs and
assumed mechanisms; the limit equilibrium method and the
spreadsheets that implement these methods, but they are
plastic limit analysis method. In the limit equilibrium method,
generally of a one-off variety and are primarily used in-
a failure mechanism is assumed, usually described in terms of
house by various consultants, contractors, or research
one or more geometric parameters [E.3], [E.l4]. The body
groups. A few of these programs are publicly available or
force distribution, stress boundary conditions, and the stress
available for purchase. While some programs have been
or force distribution on failure surfaces are estimated, and a
rather widely used, there is no single general, industry
search is conducted to find the geometry that is closest to
equilibrium conditions. The plastic limit analysis method also accepted, program or procedure. Selected models have been
uses an assumed failure mechanism with the added require- shown to compare favorably with more rigorous FEM
ment that the mechan.ism satisfies kinematic constraints (i.e., results for soft clay profiles and various anchor geometry
incompressibility for a purely cohesive material, displace- and load attachment points [E.53].
ments continuity, etc.) [E.4], [E.4l]. This is, however, an active area of development. Auto-
A possible failure mechanism is shown on Figure E.l. mated solutions (programs, spreadsheets, etc) using these
Other proposed mechanisms can be found in [E4], [E50], approaches generally require much less input description and
and [E.5 l]. Depending on the failure mechanism, the anchor are much easier to use than general FEM programs. As a
is shown to resist vertical uplift loads by self-weight, skin result they are well suited for conducting parametric studies
friction, reverse end bearing (REB) and/or shear and/or rota- and design iterations. However, as mentioned above these
tional failure at the pile tip, passive and active earth pressure, solutions do not necessarily converge to correct capacity esti~
and soil flow around the pile. mates even with great care and analyst skill, and results from
In some limit equilibrium methods, the circular area is different formulations may give significantly different
transformed to a rectangle of the same area and the width answers. Thus, obtaining accurate results is very dependent
equal to the diameter, and 3D effects accounted for by side on the analysts understanding of the methodology and his/her
shear factors [E.3]. engineering judgment.

Boundary of soil ll//» Applied load

volume shown
;,_ _— Tp .”T ._—*_lIIIIu-|-|.._
__.,,,f' -—— —-w~fi_,,_

. :3:-.=z:=§'.‘-I
5' I '
Active wedge '
failure zone I . Passive wedge
-. 1 I . . failure zone

V
wj

i p _ . , F Flow around zone

Center of /’
rotation
ill
‘l~
Tip rotational resistance

Figure E.1-—Three-Dimensional View of a Possible Failure Mechanism

100 API RECOMMENDED PRACTICE 28K

E.3.2.3.3 Semi-Empirical Methods: Beam-Column ll. Ignore that the resistance elements depend on the
Analyses deformation mode and ignore the coupling between the
resistance elements. This can lead to large errors, par-
As discussed in [E36], these models are the most approxi- ticularly for relatively short piles.
mate, but may be the easiest to use if computer programs or
spreadsheets with FEM, lim.it equilibrium or plastic limit 2. Do not include independent side shear resistance com-
analysis methods are not available. They are labeled semi- ponents on active and passive sides to model different
empirical to suggest that they incorporate the basic mechan- relative shear displacements between the soil and the
ics of a suction pile loaded to failure, but depend on a set of pile on the two sides.
empirical rules to represent the soil resistance. These rules are 3. Do not include the coupling between the horizontal
typically less general than the methods discussed in Sections and vertical soil resistance components along the pile
E.3.2.3.l and E.3.2.3.2. For example, they do not explicitly sides and thus do not show the effect of inclined anchor
incorporate soil failure mechanisms, but instead represent the load. It is possible in principle to couple these elements
soil resistance as a load distribution varying along the bound- (P-y and T-2: curves), but this has only been done in
ary at the soil-pile interface. It is difficult to generalize such a special cases [E46].
load distribution for a wide range of soil profile types so a
4. Require input that is not essential to the capacity
particular solution may apply say, only to a normally consoli- assessment such as pile bending stiffness and sub-fail-
dated strength profile. Rules for constructing these distribu- ure soil response and produce output that is of little
tions are typically based on a combination of experimental interest for the analysis such as moment and shear pro-
and analytical results. In the so-called beam-column model, files and load deformation response that is probably
the soil is represented by uncoupled, non-linear, soil springs not very accurate. Because most piles are stiffened
along the pile boundary. The beam column m.ethod can pro- shells, the beam equations are of doubtful. validity and
vide estimates of the load displacement history up to and are largely irrelevant with regard to stresses in the pile.
including the full capacity of the soil-pile system. A better pile model for these purposes is actually a
In the beam-colu1n.n. model the soil resistance is repre- rigid body that can be approximated by setting the pile
sented by uncoupled, non-linear soil springs (P-y curves) EI to an arbitrarily large value (see Section E.5.3 for
which describe the sub-failure behavior of the local soil resis- recommendations on structural design).
tance as well as the peak capacity [E42] and [E43]. In the
5. Require user intervention to determine the pile capac-
API RPZA P-y formulations for piles, the curves exhibit soft-
ity. In most beam column programs the ultimate
ening behavior (reduced resistance with continued displace-
capacity is determined by trial and error, gradually
ment) to account for the effects of cyclic loading [EA-4]. It
increasing or decreasing applied loads until the mini-
has been argued in [B45], however, that ultimate capacity
mum load that produces numerical instability
estimates for piles, and thus presumably for suction piles as
(interpreted to be the failure load) is found.
well, should be based on non-softening (static) P-~y curves. In
this model the governing equations of a beam on an (non-lin- 6. Require special elements for rotational, vertical and
ear) elastic foundation are solved iteratively until an equilib- horizontal tip resistance.
rium solution is found for a given value of the applied load. 7. Do not explicitly include effects such as soil-pile inter-
The user can gradually increase the load in subsequent steps face roughness and loss of soil contact on the back side
until the solution no longer converges, a point which is inter- of the pile.
preted as failure.
It is possible to formulate and implement a beam-column
The beam-column model has been used by geotechnical program that overcomes most of the above limitations. There
engineers for almost 50 years for the analysis of laterally seems to be little incentive to do so however, as other meth-
loaded piles. Hence it has the decided advantage of being a ods are available that are simpler to implement and can be
familiar tool. There are many versions of beam»-column pro- especially tailored to suction pile analysis.
grams in use, including general purpose programs where
loads as well as non~linear springs can be prescribed at virtu-
E.3.3 INCREASE OF SIDE FRICTION WITH TIME
ally any point on the pile, as well as special versions where
non-linear spring construction is automated based on mini- As described in Section E.3.2.1.1, the side friction at a
mal soil property input. Thus, there might be an understand- given depth can be calculated as ct,-,.,S - Sapgg, with am rang-
able tendency for engineers to select these programs for ing in value from 0.2 to 0.5 for Gulf of Mexico deepwater
section pile analysis. However, the user should be aware that clays during installation. With the passage of time after instal-
these programs have significant limitations. As detailed in lation, the side friction increases through soil thixotropic
[E36], among the limitations, the conventional beam col»- effects and pore pressure redistribution at the pile interface.
umn models: This phenomenon is often referred to as “set-up”. Set-up
 Dearest AND Answers on STA'¥'iONi<EEPiNG Systems son Ftoaraie Sraocruass 101

effects are often addressed by estimating the change in the for different parts along the side of the anchor. The set-up
adhesion factor, oi, with time. Set-up mainly influences the along the part of the anchor penetrated by underpressure may
vertical capacity, and to a lesser extent, the horizontal capaci- occur much faster, but the permanent reduction may be larger.
ties of a suction pile [E7]. The method in [E20] was developed for driven piles witli
The set-up process may be different for the part of the ratio of diarneter over wall thickness less than 40. The
anchor penetrated by weight and the part penetrated by method in [E7] was proposed for penetration by underpres-
underpressure. For suction piles in highly plastic clays, the sure. Both methods should be applied with caution outside
set-up time can be long and. there can be a permanent loss of the range of data used in their development. Other methods
shear strength, whereby the ultimate side friction after full developed for driven piles include the one described in
set-up is less than the original undisturbed shear strength (i.e, [B49]. There is no single industry-wide accepted setmup
the adhesion factor, (Z, is less than 1.0 after full set~up), both curve.
for the portion of the anchor penetrated by weight and the As with other piled foundation systems, the calculated
portion penetrated by underpressure. anchor ultimate capacity should be reduced if soil set-up will
Some researchers [E.6] have reported that the portion of not be complete before significant loads are imposed on the
the anchor penetrated by underpressure is expected to typi- anchor pile.
cally have a shorter set-up time and a lower ultimate side fric- Set--up can be addressed in various Ways during design.
tion. after full set-up than the portion penetrated by weight. The designer can ensure adequate anchor capacity if:
Figure E.2 shows a typical soil set-up prediction graph for a
' The suction pile is designed with partial soil set-up;
large diameter suction anchor in typical Gulf of Mexico soils
and illustrates the current uncertainty in calculating soil set» ~ The suction pile is installed well in advance of platform
up. The methods in [E.7] and [E.20] are shown to illustrate hook-up and platform first-oil to ensure adequate soil
the potential differences in set~up time and ultimate friction set-up when the mooring may experience design loads.

r*""i"'"'* to "t "W" '" t" l —'"'"W

/'—"'H
0.9 ~ ~~~~~~~~~ l

rength as ~ ~ aaaaaaaaa~ l i_ _______. i i ___, *,i,,_..,_._... .. __

2 . .
. i ,

shearst 0.7 are —aaaaaaa~ as

‘Q

urbed
- 0.6 W .... H , l

/(undst
I i it l
9,5 J T aaaaaaa~
/-‘H.

1 i l
il ll il
l‘ l l

mtme
OT]
0.4 *

fr'dect'
mm
0.3 I ~ aaaaa~ -
"\-q—,/

an
s’
0.2 l
~ E
ils
Aph l_..

Bogard 2001 - Diameter = 6 ft; wall thickness = 1.8 in.; average cuwe; Reference E-20, E-21
0.1
-------?~’t---- Andersen and Jostad 2002 - alpha during installation = 0.3; alpha at 90% set-up = 0.65 -~ Reference E-7 I
0 --____I__-------_----_---- . .
0 20 40 60 80 100 120 140 160 180
"lime After Installation (Days)

Figure E.2--Examp|e of Increase in Adhesion Factor with Time

102 API Racouuaunso Pancncs .2SK

~ For a limited amount of time between installation and gram shows that if the load is primarily vertical, with padeye
first—oil, reduced extreme load criteria are assumed, angles from 40° or 45° to 90°, the failure mode is controlled
based on suitable risk analysis. by vertical pullout and 100% of the vertical capacity is avail-
able. In a similar manner, if the load. is primarily horizontal,
E.3.4 COUPLING BETWEEN HORIZONTAL AND with padeye angles from zero to 15°, the failure mode is con-
VERTICAL CAPACITY trolled by horizontal pullout and 100% of the horizontal pile
capacity is available. In this case, the maximum horizontal
When a suction anchor resists the design loads, the vertical
capacity is equal to 1.8 times the vertical capacity. If, how-
and horizontal components of the anchor capacity are not
ever, the load angle at the padeye is between 15° and 40°, less
mobilized independently. Coupling between vertical and hor-
than maximum vertical and horizontal capacities are avail»-
izontal capacities may be important in some cases. Studies
able. In the example shown, only 90% of the vertical capacity
have shown that for mooring line angles at the padeye
is available; and the available horizontal capacity is reduced
between 15° and 45° (as measured from the horizontal), it
to 150% of the vertical capacity, from the original 180%.
may be non»-conservative to neglect this coupling [E3] and
Examples of failure interaction diagrams can be found in
[E.8].
The following discussion is for illustration only and there-
[1-3.55] and (E561.
fore should not be used for design. The sample failure inter-
E.3.5 FACTORS OF SAFETY
action diagram shown in Figure E3 is typical for suction
piles with a length to diameter ratio of 5, in a linear increasing Factors of safety for holding capacity, defined as the calcu-
shear strength profile with low shear strength at the seabed. lated capacity divided by the maximum anchor load from
The mooring padeye is located on the pile shell, about 2/31‘(l.S dynamic analysis, are provided in Table 7 (ref. main body of
of the way down from the pile top. In this example, the dia- this recommended practice) for axial and lateral loads. Infor-

Interaction Diagram
(Illustration Only----Do Not Use for Design)

"2!_T_i"i"i"T°° ° "T°° ° TT"'i"T_l

0.8 - - ........ - _ ._ ~ . .. .... 1-*4 - l .......

3X
CI) '0: i 77: "___ i V __ __,__
H W_W_ _ 1
l
_

v/v.,
mmm.
—uni

4 40 to 45 deg
Y
Alt‘

0.4 n** ************ ............ ************** ******“J ** * . ******* ************* * **** * * F*"J*********************** r ' ' * t

- 15 deg
n_ - I
_4;uII"'
0.2 _ ___ ______?:'— — 7 l

0.0 ___, ... l I .........;A ‘ * .. * . l l

0.0 0.5 1.0 1.5 2.0
HNMGX
Mooring padeye angles measured from horizontal.
with:
V: Vertical load component.
H: Horizontal load component.
Vmaxz Vertical ultimate capacity of the anchor, for
purely vertical loads.
Figure E.3—Example of Failure Interaction Diagram
 DESEGN AND Anarvsis or Sratsonxseaino Svsratvis eon Fioartno Srsocrusss 1&3

mation on coupling between vertical and horizontal capaci- trolled by the axial capacity. The axial
ties can be found in Section E.3.4. Axial safety factors capacity is defined as the capacity under
consider that the pile is primarily loaded in tension, and are purely vertical loads.
therefore higher than for piles loaded in compression.
As the lateral failure mode for piles is considered to be less For a given failure interaction diagram, Figure EA illus-
catastrophic than the vertical one, lower factors of safety have trates how to calculate the required safety factor.
been recommended in Table 7 for lateral pile capacity. Use of
separate factors of safety for vertical and lateral pile capaci- E.3.6 OTHER SPECIAL CONSIDERATIONS
ties may be straightforward for simple beam-column analysis
l. Closed vs. open top: The top of the anchor should remain
of, for example, mobile moorings (ref. Section 13.3.2), but
sealed throughout the life of the field, if the RJEB at the
more complex methodologies do not differentiate between
vertical and lateral pile resistance. anchor tip is to be relied upon in design. Note that, with
increased soil set-up and side friction, the need for REB
The safety factor used in design should be based on the
decreases and, thus, the requirement to maintain a sealed
failure mechanism controlling the capacity and not only on
the load angle. Although load angle and failure mechanisms top cap. For anchors with essentially horizontal loading, a
are related, other parameters such as soil profile, anchor sealed top is not essential for the capacity, and the top part
geometry, and load attachment points are also important in can be removed. after installation (Reference E47).
determining failure mechanisms. For cases where axial pull- 2. Strength oriisotropy: Capacity calculations should be per-
out controls, the minimum safety factor should be as per formed with anisotropic shear strength, including effect of
Table 7, regardless of load angle. For cases where lateral pull- combined static and cyclic loading history.
out controls, the minimum safety factor should be as per
3. Internal ring sti'ffeners: For large long-term loads and for
Table 7, regardless ofload angle. Eq. 5 is proposed in order to
suction piles that are not sealed at the top, the skin friction
provide a combined factor of safety for situations where nei-
along the inside skirt wall is an important contribution to
ther the axial nor the lateral capacity control the design:
the capacity. The inside wall friction may be significantly
For a given geometry, load attachment, and soil profile, the
lower than the original shear strength due to the distur-
combined safety factor can therefore be calculated as follows:
bance during installation, especially if the anchor has
internal stifteners. In cases with series of ring stiffeners,
if 9 S Qlatemz , FOSc0mbined = FOSlateral clay from the upper part of the profile, and also water,
E8
may be trapped between the stiffeners and give low capac-
If 9 5 Qax,-dz , FOSc0mbined = FUSax.ial
ity at larger depth. In such cases, the compartment
H gmial 5 65 auxin! ’
between the ring stiffeners may also act as a drainage
channel [E36].

9.... 9 4. Gapping: A gap may form on the outside at the active side
FOScombined w FOSlateml + ~ ~ (FOSrixial " FOSleteral) (i.e., backside) of the anchor. There are uncertainties on
Qaxial W elateral
how to predict gap formation, unless the clay is soft and
with essentially zero strength intercept, in which case a
where gap is not expected to form. Therefore, one should make
conservative assumptions with respect to whether there
FOSCO,,,g,;,.,,,,,,; II Combined factor of safety (FOS), will be a gap or not. One should consider conservatively
.FOS;,,,,,,.,,; Y-= Lateral FOS from Table 7, placing the load attachment point far enough below the
optimal load attachment depth for the suction anchor top
FOSW-,,; I Axial FOS from Table 7, to move “backwards” (i.e., away from the direction of the
mooring line) during loading to prevent gap formation.
8 K Angle of mooring line from horizontal at
pile attachment point, 5. Instailation tolerances: The allowable installation toler-
ances (e.g., tilt and orientation) shall be included in the
t3;a,_.,m; =1 Load angle, measured from horizontal, capacity calculations, as tilt and out of plane loading may
below which the ultimate capacity is con- reduce the holding capacity of the pile.
trolled by the lateral capacity. The lateral
capacity is defined as the capacity under 6. Change in outer diameters: It is considered that variations
purely horizontal loads, in the outer dianieter with depth could reduce the outside
interface strength and it is recommended that, in general,
Gm-Q; K Load angle, measured from horizontal, designs with variations in outside diameters should be
above which the ultimate capacity is con- avoided.
104 AF" RECOMMENDED PRACTICE ZSK

. FOOmbined:FOaXia| l . . ‘ _ . .. _. .. . ., .. _ . _.
1.2

1 O i 1 __,_::;_;11 1 \ _1_1:Z;1_111 _ 1 |FOSgQmb§ngd 8.8 p@r .

0.8 I: \ e e i

$3 G)
V/\/max

0.4

1 i §
Gaxiai 1
0.2
I 7 0 7 "T;iiiii0 7 7 7 Fosgombined 3 Fosiateral ifiwwww
l \ Hlatereu ‘ . . . . .. _ .

0.0 W44 »__ _..__.J_____I Lm 1.

0.0 0.5 1.0 1.5 2.0
H/\/max

FOSflXi&| --iwlnujuiunnmnuninnnllnunnnlnallltiiwxigmllmiiinynuun

Foslateral we

SafFFactOof
S¢0emobtr-yfie,

Qlateral Qaxiai

Load Angle (from horizontal)

Figure E.4--Ca|culation of Required Safety Factor as a Function of Failure Mode

 Dacron sun At\tAt.YSiS or Srariousaseiue Svsrsns Pea Ftoataio Sraucruass 165

7. Stmd layers: sand layers, if present may have a significant tant that the anchor's penetration depth can be established
effect on the holding capacity. lt should be ensured. that during the installation process. Furthermore, a plate anchor
the sand layers will not cause excessive drainage and pore gets its high ultimate pul1-out capacity by having its fluke ori-
pressure redistribution. that could negatively affect the ented nearly perpendicular to the applied load. To ensure that
REE, particularly if the anchor is to resist long-duration the fluke will rotate to achieve a rnaxirnurn projected bearing
loads. area, the plate anchor design and installation procedure
8. Distance between irzstatlatton locations: In the event that should:
an anchor needs to be retrieved and re-installed, the deter- — Facilitate rotation of the fluke when loaded by environ-
mination of the minimum distance between the first mental load.s or during installation or both;
location and the subsequent location should ensure that
the volume of soil disturbed during the first installation - Ensure that no significant or unpredicted penetration is
will not be mobilized when the anchor resists the design lost during anchor rotation, which may move the fluke
load at the subsequent location. into weaker soil;
£2

9. Sustained Load: The duration of sustained loads (e.g., ~ Have the structural integrity to allow such fluke rota-
creep under loop current load) and the period of cyclic tion to take place during installation and keying opera-
loading should be considered and the anchor capacities tions or while subject to the ultimate pull-out capacity
should be adjusted to account for these effects. Examples load. Depending on the type of plate anchor and its
of capacity reduction as a function of load hold time can installation orientation, this item may also apply to
be found in [E.57] for vertically loaded anchors in Gulfof fluke rotation about both horizontal and vertical axis.
Mexico clays.
As appropriate, the anchor capacities should be reduced to
A combination of these considerations may be used to account for anchor creep under long-tenn static loading and
arrive at a suitable suction pile design. Due to the complexity cyclic degradation.
of analyzing the load capacities of large perrnanent suction Factors of safety for holding capacity, defined as the calcu-
piles, a recognized geotechnical expert should be consulted. lated soil resistance divided by the rnaxirnurn anchor load
from dynamic analysis, are provided in Table 7.
E.4 Geotechnica! Design of Plate
Anchors E.4.2 PREDICTION METHOD FOR DRAG
EMBEDDED PLATE ANCHOR
E.4.1 BASIC CONSIDERATIONS
Three aspects of drag embedded plate anchor performance
For plate anchors, the ultimate holding capacity is often
require prediction methods:
defined as the ultimate pull-out capacity (UPC), which is the
load for the soil around the anchor reaching failure mode. At ~ Anchor line mechanics;
UPC, the plate anchor starts moving through the soil in gener-
~ Installation performance; and
ally the direction of the applied anchor load with no further
increase in resistance or the resistance starts to decline. The ' Holding capacity performance.
ultimate pull-out capacity of a plate anchor is a function of
All three mechanisms are closely linked and influence one
the soil undrained shear strength at the anchor fluke, the pro»
another, as explained below.
jected area of the fluke, the fluke shape, the bearing capacity
factor, and the depth of penetration. When analyzing the plate
anchor ultimate pull--out capacity, the disturbance of the soil
E.4.2.1 Anchor Line Mechanics
due to the soil failure mode should be considered. This mode As discussed in [E58], [E59], [E60] and [E61], the anchor
is generally accounted for in the form of a disturbance factor line mechanics influence strongly the prediction of the drag
or capacity reduction factor. The bearing capacity factor and embedded. plate anchor’s final orientation and depth below
disturbance factor should be based on reliable test data, stud»- the seabed, which in return governs the holding capacity of
ies, and references for such type of anchors. Typically, the the anchor system. Figure E.5 is a schematic of an anchor line
plate anchor's penetration is in a range of 2 to 5 times the configuration showing the reverse curvature of the line as it
fluke width, B (see Figure E.7), depending on the undrained cuts through the soil. As the load in the anchor line increases,
shear strength of the soil, in order to generate a deep failure the inclination of the line with the horizontal at the anchor
mode [E.l0]. If the final depth does not generate a deep fail- attachment point decreases, giving rise to an interaction
ure mode, a suitable reduction in bearing capacity factor between the anchor line and the holding capacity of the
should be used. anchor.
Plate anchors get their high holding capacity from their In general, this problem is approached in the same manner
ernbedment into more competent soil. Therefore, it is impor- as that for predicting the displaced shape of a catenary, fixed
106 API Rscouusmcao PRACTICE 28K

~fl 7 "T ******* " rmmnvmn _______ __7 '**"1lu\\\uvuau|»r nwm\\1IvmrnrIL__ II " '" lI\\\\\\\IIw~L _ *—||Iu\\vlmnn—r '' __lII__ _ __ ____ "'7 _____ 7' ' ' |vnn\1___ 7*’ ''' " mwnuuwuu

I, Seafloor Mudline force

i nunnu|____ 7"-it nun _______ 7 flnnllnnuuuul nwnvnuuur ‘|nunr"' nuuwumu 1 _____ __ 4:1 7' ""1-uunmunuuunnt _7__ *7"1lIuI"* lnueuuum _ 7"»: Jvnr____Junnuu '7 " lulnnvinuvnuc "

lj

'1

Figure E.5—Schematic of Anchor Line Configuration During Embedment

at both ends, and defonned. only by its own weight plus bear»- l Empirical Methods are typically based on correla-
0

ing pressures exerted from the soil normal to the line and tions with observed anchor performance and
shear resistance tangent to the line. The governing differential dependent on anchor characteristics (weight) and an
equations for this system of forces are nonlinear and require approximated measure of soil resistance. However,
an iterative numerical solution. many of those field studies remain proprietary, and are
therefore not readily available.
E.4.2.2 Installation Performance 2. Limit Equilibrium Methods take into account a more
As discussed in [E58] and [E59], the capacity of a drag detailed description of soil and anchor geometry/
embedded plate anchor depends strongly on its final orienta- weight. The method is based on an estimated soil force
tion and depth below the seabed, hence prediction of the distribution on the anchor at failure condition; site spe-=
anchor trajectory during installation is a critical issue. Figure cific soil and anchor information can be incorporated
E.6 is a schematic diagram showing a typical anchor trajec- in more detail. This approach is most commonly used,
tory and sequence of anchor orientations as the anchor line is and commercial software based on this approach is
dragged along the seabed. available.
Methods for predicting this scenario generally fall into 3. Plastic Limit Analysis is in many ways very similar to
four groups: the Limit Equilibrium Methods. Virtual work princi»

Drag distance
’4

I
4 %“Iun n-Innun,1-was
fa‘

S." J‘

i Anchor depth
EI

Figure E.6-—Anchor Trajectory a nd Fluke Orientation During Installation

 Desrorv AND A%\iAi..YSiS or-" STA'i'iO¥\ii<EEPING Svsreivts eon Froarrrso Srsocroaes 107

ples are used to minimize the calculated failure load E.4.3 PREDICTION METHOD FOR DIRECT
with respect to the geometric parameters defining the EMBEDDED PLATE ANCHOR
failure mechanism at any anchor depth, anchor orienta-
tion, and anchor line conditions. Anchor capacity determination for direct embedded plate
anchors is identical to that shown for drag embedded anchors
4. Advanced Numerical Methods include usually the
with the following exceptions:
finite element (FE) method. It has the potential. of
obtaining a rigorous solution for all aspects of anchor ~ Final penetration depth is accurately known;
behavior. In practice, however, it has considerable lim-
- Nominal penetration loss during keying should be
itations. A complete solution would require a FE
incl.u.ded (usually taken as 0.25 to 1.0 times the fluke’s
model defining nonlinear material behavior, nonlinear
vertical dimension, or B in Figure E.7, depending on
boundary conditions, large strain and large deforma-
shank and keying flap configuration);
tion theory. Hence a simple anchor trajectory
prediction would require a prodigious effort to formu- ~ Calculation of effective fluke area should use appropri-
late, set up, and solve. However, FE models can be ate shape factor and projected area of fluke with keying
easily used to check calculations or enhance other pre— flap in its set position.
diction methods.
E.4.4 SPECIAL CONSIDERATION ON FACTOR OF
E.4.2.3 Holding Capacity Performance SAFETY FOR DRAG EMBEDDED PLATE
As discussed in [E58] and [E.59], anchor holding capac» ANCHORS
ity, as previously mentioned, is only a special case of the Factors of safety for drag embedded plate anchors are
installation sequence and, hence, the methods underlying higher than for drag anchors because overloading of anchor
installation prediction described above are directly applica» normally results in pullout of the anchor, while drag anchor
ble. This problem is considerably simpler than the installs» may drag horizontally or penetrate deeper, developing con-
tion problem since the ultimate holding capacity for a single stant or higher holding capacity under similar situation (see
location and orientation is of interest only. The ultimate Tables 6 and '7 in Section 7.4 of main text). For plate anchors
holding capacity can, therefore, be expressed on the basis of that exhibit similar overloading behavior as drag anchor, con-
conventional bearing capacity theory in conjunction with the sideration may be given to using drag anchor factors of
anchor line solution: safety, assuming the behavior can be verified by significant
field tests and experience.
Fmax m NcAeflTI Sir

E.5 Structural Design of Suction Piles

where
E.5.1 BASIC CONSIDERATIONS
Fmax K Ultimate holding capacity,
The purpose of this section is to provide guidance and cri»
A85 *1 Effective area of the anchor accounting for
teria for the structural design of suction piles. Some of the
shape and projected. area,
guidance and criteria are also applicable to driven piles.
NC K Bearing capacity factor determined for example Structural design for plate anchors is not addressed because it
from method of characteristics or finite element is typically performed by anchor manufacturers.
solutions,
E.5.1.1 Fabrication Considerations
n H Reduction for soil disturbance du.e to penetra»
tion and keying, The structural design criteria given in the following sec-
S,, -== Measure ofthe local undrained shear strength at tions assume the suction pile has been fabricated to be within
certain dimensions and tolerances. Although, API Specifica-
the design penetration depth.
tion 2B offers guidance in this area, it not considered suffi»~
Overall, considerable judgment and experience is required cient for the fabrication of suction piles. At a minimum, the
to evaluate the input parameters for any of the predictive following dimensions and tolerances should be specified in
methods. An example of anchor selection can be found in the suction pile fabrication specification in addition to the pile
[E.-40]. diameter and wall thickness schedule:
l. Pile length: Total length of the pile should be specified
with a suitable tolerance. The minimum acceptable
108 API Rscomnemoeo PRACTICE 2SK

i
Q

-3
*

§-9
!\
...
X
“Z.
, ll i

s
l1:
l

it

l
l
l
l ________ Wqmi-177 lnunmmsvu ___

Q
\
l

ft
ll
.-/"

Figure E.7—Definition of Anchor Fluke Width, B, and Length, L

length so specified should be acceptable with respect to outlines a procedure to measure OOR on cans with
the geotechnical design. their longitudinal axes horizontal. This technique, to a
large extent, will remove the effect of ovalization of the
Note: Specification 2B allows a 1.5 in. tolerance per 10 ft of length;
thus if the pile were 100 ft in length, it would be theoretically accept- can due to gravity in the OOR calculations. Alterna-
able if the pile were 15 in. short. This may not be acceptable if the tively, OOR measurements can be made with the axis
pile design were highly optimized. of the can vertical during diameter measurements.
2. Out-of-Roundness (OOR): Out»of-roundness, the 3 Circularity: Circularity is a measure of the pile wall’s
difference between the major and minor outside (or local deviation from the theoretical shape; in this case,
inside) diameters at any point along the length of the the theoretical shape is an arc with same radius as the
pile should not exceed 1% of the nominal outside (or pile. It is measured using a sweep gauge that has one
inside) diameter. The 1% roundness value is the maxi- edge cut to the theoretical inside or outside radius, as
mum OOR assumed in the buckling formulations appropriate. The recommended sweep gauge are
given API RP 2A, Bulletin 2U and other codes. length is 1/10th of the circumference of the pile. Mea-
suring circularity ensures that dents, flat spots or other
For each cross—section checked, a minimum of two sets geometrical imperfections do not adversely affect the
of two diameters each should be checked (i.e., eight buckling resistance of the cylindrical pile wall during
points along the circumference of the pile). In Figure suction cmbedment.
E.8 that would be A-E and G-C for the first set and F-B
and D-H for the second set. Note that Figure E.8 also
 Daszeu nae Auluvszs or STA.TtONt<EEP¥l\i€B Svsraus eon Froanne Sraucruaes 109

A G
H V' e F H

F o o . e
O O O s O
Step 1: Step 2:
- Measure AE & GC —- Rotate 90°
-- Measure GO & AE
- Average AE values
-~ Average GC values
-~ Calculate OOR using average values

F D
--1-#1140-It-_____g

E I ,l G 0 D i E

l
o ---- -- :-: B --~---- ---- F

c l A A o
s H
Step 3: Step 4:
-- Rotate 45° - Rotate 90°
-~ Measure FB & DH -- Measure FB & DH
-~ Average FB values
- Average DH vaiues
- Calculate OOR using average values

Figure E.8—Procedure to Measure Pile Out-of-Roundness

As the sweep is moved around the circumference of the can forming process. Individual cans should not pro-
pile in a plane perpendicular to the axis of the pile, the ceed to pile assembly until passing OOR and circular-
gap between the sweep gauge and pile wall is mea- ity requirements. The completed pile should also be
sured. The acceptance tolerance can be determined subject to final OOR and circularity checks in addition
from non-linear buckling analysis of the pile wall; a to the survey of straightness and length.
typical acceptable gap value would be 0.5 in. The num-
ber of circumferences checked along the length of the Straightness: Generally, the requirements of Specifi-
pile should be sufficient to capture all potential dents or cation ZB are sufficient with respect to straightness.
flat spots in the pile. A straightedge, held parallel to the This is due to relatively large diameter to buckling
longitudinal axis of the pile, should be used to survey length ratio of suction piles. Thus, suction piles are
the extent of any dents found using the sweep gauge. usually not subject to a column buckling mode of fail-
ure that lack-of-straightness would exacerbate, but
It is recommended that the dimensional control pro» only to a local buckling failure in which global
gram include OOR and circularity checks as part of the straightness is not factor.
116 API Rscouuauoeo Paacuca 28K

E.5.1.2 Handling and Transportation eye for a maximum load equal to a factor times the break
Considerations strength of the mooring line may lead to a significantly over-
designed padeye, which may not integrate well with the
In order to achieve economic and weight control goals, it is
anchor shell and bacloup structures. The mooring line padeye
not uncommon for suction piles to have large sections of thin
should be designed for the controlling load case, and sensitiv-
walled steel. These thin pile walls are especially vulnerable to
ity checks should be performed to ensure that a load. case with
damage by inadequate temporary support during handling
less than the maximum load but applied at a more onerous
operations in the fabrication yard. It is recommended that
angle does not control the design. The orientation of the
temporary supports be pre-eng:i.neered prior to handling and
applied load at the padeye will be affected by the inverse cat-
that rigging personnel be briefed on proper suction pile han~
enaiy of the mooring line, vertical misalignment due to
dling techniques.
anchor tilt, and rotational misalignment due to deviation from
Another potential source of pile damage is during loadout,
the target orientation. These factors should be properly
transportation and offloading operations. Care should be
accounted for.
taken when loading and unloading suction piles from their
cradles to minimize side or vertical impact. Cradle design and
fitment of the pile into its cradle should not invalidate the E.5.2.3 Embedment Loads
transportation design assumptions. For example, if nearly full For anchor cmbedment, the estimated upper bound suction
cradle contact is necessary to keep the pile stresses generated. pressure required to embed the anchor to its design penetra»
during the design environment below allowable values, the tion should be used for the design of anchor wall and anchor
actual pile fitment should match the assumptions or the cradle cap structure. However, the maximum suction pressure used
design modified accordingly. should not be higher than the suction at which internal plug
uplift occu.rs.
E.5.2 DESIGN CONDITIONS
The suction pile structure should be designed to withstand E..5.2.4 Extraction Loads
the maximum loads applied by the mooring line, the maxi- With respect to anchor extraction, there are two conditions
mum negative pressure required for anchor cmbedment, the that require evaluation:
maximum internal pressure required for anchor extraction,
and the maximum loads imposed on the anchor during lifting, l. Temporary condition: Extraction. of a suction pile
handling, launching, lowering and recovery. Fatigue lives of may be required for permanent moorings. For exam-
critical components and highly stressed areas of the anchor ple, after all suction piles have been preinstalled along
should be determined and checked against the required mini- with the mooring lines, one of the mooring lines is
mum fatigue life. accidentally dropped to the seafloor and damaged dur-»
ing the hookup operation with the vessel. At this time,
E.5.2.1 Mooring Loads on Global Anchor a decision to extract the suction pile and recover the
Structure mooring leg may be made. Typically, such situations
The load case that provides the maximum horizontal and may occur 30 to 60 days after the first suction pile has
been installed.
vertical loads at the mooring padeye should be used for the
global structural design of the anchor. The soil reactions gen»- For mobile moorings, the suction piles are often
erated by the geotechnical analysis will be used in these cal- extracted at the end of the current drilling or testing
culations. Sensitivity checks should be performed to ensure operation and reused in other locations.
that a load case with less than the maximum load, but applied
at a more onerous angle at the padeye, does not control the 2. Terminal condition: The suction piles for a permanent
design. mooring may be extracted at the end of its service life.
The estimated maximum internal pressure required to
E.5.2.2 Mooring Loads on Anchor Attachment extract the anchor for these two situations should be
used for the design of anchor wall and anchor cap
The mooring line attachment padeye or lug is a critical structure.
structural component. In order to meet fatigue resistance cri~
teria, the padeye is often an integral cast lug and base struc-
E.5.2.5 Transportation and Handling Loads
ture. This avoids the use of heavy weldments, which can
result in a lower fatigue life. The attachment padeye should The suction pile structure and its installation appurte-~
be designed to satisfy both strength and fatigue requirements. nances should be designed for the maximum loads gener-
The padeye should be designed for the controlling design ated during suction pile handling, transportation, lifting,
load with an appropriate factor of safety. Designing the pad» upending, lowering, and recovery. The suction pile designer
 DESiGi\i AND Anarrsrs or Srarionxasaiuo Svsraras FOR Ftoarino Srsuctuaas ‘£11

should interface closely with the installation contractor Shells, and API Bulletin 2V, Design ofFlat Plate Structures.
when determining these load cases. Design of appurte- As an alternative, buckling and. post-buckling analysis or
nances for these load. cases are typically performed using model tests of specific shell or plate structures may be per»
the installation. contractor’s in-house design guidelines or formed to determine buckling and ultimate strength.
other recognized codes. Nevertheless, all lifting appurte-
nances and their supporting structures should meet the min- E.5.3.6 Dynamic Response
imum requirements of API RP 2A.
Significant dynamic response is not expected for the
E.5.3 STRUCTURAL ANALYSIS METHOD anchor in its in»-place condition, therefore anchor structures
are often analyzed statically. Transportation analysis, how-
Pile analysis in accordance with Section 3 of API RP 2A is ever, will typically include dynamic loads generated. by
appropriate for piles with diameter to thickness ratios (D/t) of harmonic motions of a simple single—-degree-of-freedom
less than approximately 100 to 120. For cylindrical piles with model.
D/I ratios exceeding 100 to 120, it is recommended that a
detailed structural finite element model be developed for the E.5.4 STRUCTURAL DESIGN CRITERIA
global structural anchor analysis to ensure that the anchor
wall structure and appurtenances have adequate strength in E.5.4.1 Design Codes
highly loaded areas. Supplementary manual calculations may The design method adopted in this document is the work-
be appropriate for members or appurtenances subjected to ing stress design method, where stresses in all components of
local loading. the structure are kept within specified values. In general,
cylindrical shell elements should be designed in accordance
E.5.3.1 Space Frame Model with API RP 2A. for diameter to thickness ratios (D/1.‘) less
than 300 or API Bulletin 2U when D/t exceeds 300, flat plate
A space frame model generally consists of beam elements
elements in accordance with API Bulletin 2V, and all other
plus other elements needed to model specific structural char-
structural elements in accordance with API RP 2A or AISC,
acteristics. This is appropriate for piles with D/I ratios less
as applicable. In cases where the structure’s configurations or
than approximately 100 to 120 and for prelimi.nary design of
loading conditions are not specifically addressed by these
the top cap or padeye backup structures on large diameter
codes, other accepted codes of practice can be used. In this
piles (i.e., D/t> 120). case, the designer must ensure that the safety levels and
design philosophy implied in the API Recommended Practice
E.5.3.2 Finite Element Model ZSK are adequately met.
Finite element analysis is recommended for the global In API RP 2A and AISC, allowable stress values are
shell structure, top cap plate and supporting members and the expressed, in most cases, as a fraction of the yield or buckling
padeye backup structure for piles with D/t greater than stress. In API Bulletin 2U, allowable stress values are
approximately 100 to 120. Complex shapes such as the pad- expressed in terms of critical buckling stresses. In API Bulle-
eye casting or welding should also be analyzed by finite ele- tin ZV, the allowable stresses are classified in two basic limit
ment methods. states: ultimate limit states and serviceability limit states.
Ultimate limit states are associated with the failure of the
E.5.3.3 Manual Calculations structure whereas serviceability limit states are associated
with adequacy of the design to meet its functional require-
Manual calculations using empirical formulas and basic ments. For the purpose of suction anchor design, only the
engineering principles may be performed where detailed ultimate limit state is considered.
finite element analysis is not needed.
E.5.4.2 Safety Categories
E.5.3.4 Stress Concentration Factors
There are two safety categories: Category A safety criteria
Stress concentration factors can be determined by detailed are intended for normal design conditions, and Category B
finite element analysis, physical models, and other rational safety criteria are intended for rarely occurring design condi-
methods or published formulas. tions. The criteria in Table E.2 are recommended.

E.5.3.5 Stability Analysis E.5.4.3 Allowable Stresses

Formulas for the calculation of the buckling strength of For structural elements designed in accordance with API
structural elements are presented in API Recommended Prac- RP 2A or AISC, the allowable stresses recommended in these
tice 2A, API Bulletin 2U, Stability Design of Cylindrical codes should be used for normal design conditions associated
112 AP! Rscouuanoso F’RA{)TiCE ZSK

Table E.2--Suction Pile Safety Criteria Table E.3--Design Factors for Finite Element Analysis
Load Condition Safety Criteria Load Condition Design Factor "r|,;
Maximum intact A Maximum Intact 0.67
Maximum one-line damaged B Maximum Damaged 0.90
Anchor cmbedment A
Anchor embedment 0.67
Anchor extraction (temporary) A
Anchor extraction (temporary) 0.67
Anchor extraction (terminal) B
Anchor extraction (terminal) 0.90
Handling / lifting / lowering / recovery A
Transportation B Handling / lifting / lowering / recovery 0.67
Transportation 0.90
with safety criteria A. For extreme design conditions associ-
ated with safety criteria B, the allowable stresses may be applied loads. Note that primary stresses are not self-limiting;
increased by one-third if the working stress design method is i.e. primary stresses that exceed the yield strength of the
utilized (e.g., API RP 2A-WSD). material in question can result in failure. Secondary stresses,
For shell structures designed in accordance with API Bul- on the other hand, can be developed by local structural dis-
letin ZU, a factor of safety equal to 1.67 ‘P is recommended continuities or by constraint of adjacent parts; such stresses
for buckling modes for safety criteria A. For safety criteria B, are self-limiting. In some cases, it may be acceptable to
the corresponding factor of safety is equal to 1.25 W. The exceed material yield. for secondary stresses in elastic design.
parameter ‘P varies with buckling stress and is defined in API If the designer allows local yielding, he should ensure the
Bulletin 2U. It is equal to 1.2 for elastic buckling stresses at
material has sufficient ductility and that load will be able to
the proportional limit, and reduces linearly for inelastic buck-
redistribute to adjacent areas of the structure.
ling to 1.0 when the buclcling stress is equal to the yield stress
For flat plate structures designed in accordance with API
Bulletin 2V, the allowable stress is obtained by dividing the E.5.5 FATIGUE DESIGN
ultimate limit state stress by an appropriate factor of safety, E.5.5.1 Fatigue Analysis
which is 2.0 for safety criteria A and 1.5 for safety category B.
For cylindrical elements with D/t" ratios exceeding approxi- In-place fatigue of the anchor structure is caused by the
mately l00 to 120, it is recommended. that global strength be tension—tension cyclic loading of the mooring line attached to
analyzed using finite element techniques. Local buckling for- the anchor. The fatigue analysis for suction anchor is similar
mulations for axial compression, bending and hydrostatic to that for the mooring system, which is discussed in Section
pressure given in API RP 2A (for D/if <300) and API Bulletin 6 in the main body of this recommended practice. The major
2U (D/I S300) are considered valid if due consideration is differences are:
made for variable wall thicknesses (when it occurs) and buck-
ling length (which may extend below the mudline when per- l. The S-N (stress range versus number of cycles to fail-
forrning suction cmbedment analysis). ure) approach is recommended for suction anchor
The nominal Von Mises (equivalent) stress at the element’s fatigue analysis instead of the T-N (normalized tension
extreme fiber should not exceed the maximum permissible range versus number of cycles to failure) approach
stress as calculated below: used for the mooring system. Appropriate S-N curve
formulations that include the effect of member thick-
(SA 3 T150‘), ness should be utilized.
2. Normalized tension ranges from mooring fatigue anal-
where ysis are converted to stress ranges for suction anchor
GA === Allowable Von Mises stress, using highly refined finite element models.
Fatigue should. be checked not only at joints, but also at
11,- =1‘ Design factor for specified load condition,
any details with high stress concentrations; e.g., the padeye
cry K Specified minimum yield stress of anchor material. casting at the base of the lug and in the eye.

Design factors for the listed load conditions are given in E.5.5.2 Fatigue Life Requirement
the table E.3.
The design factors recommended in Table E.3 above are The predicted minimum fatigue life shall be at least 3 times
limits on the structure’s primary stresses generated by the the design service life of the suction anchor.
 Dacron AND Answers or STAT¥O¥\ii<EEPiN<3 Srsrasis son F¥.€)A'fiNG Srnucrosss "H3

E.6 Installation of Suction Piles, Suction - Allowable deviation from target heading of the moor»-
Caissons and Plate Anchors ing line attachment to limit padeye side loads and rota-
tional moments on the anchor padeye,
E.6.1 SUCTION PILES AND SUCTION CAISSONS
- Minimum penetration required before keying or test
In order to verify that the suction pile installation is suc- loading to achieve the required holding capacity,
cessful and in agreement with the assumptions in design, the
following data should be monitored and recorded during the - Allowable loss of anchor penetration during plate
installation of suction piles, for permanent and mobile anchor keying or test loading.
moorings: In order to verify that the plate anchor installation is suc-
~ Distance from intended seabed location, cessful and in agreement with the assumptions in design, the
parameters listed in Section 6.1 should also be recorded.
' Underpressure,
' Penetration depth, E.6.2.2 Drag Embedded Plate Anchors

- Penetration rate, For drag embedded plate anchors used in permanent moor-
ings, the installation process should provide adequate infor-
' Verticality, mation to ensure that the anchor reaches the target
* Orientation. penetration, and that the drag cmbedment loads are within the
expected load range for the design soil conditions. Typical
For permanent mooring systems, other parameters usually information to be monitored and verified is:
monitored include plug stability at all depths and plug heave
~ Line load in drag installation line;
at final penetration.
~ Catenary shape of installation line based on line tension
E.6.2 PLATE ANCHORS and line length to verify that uplift at the seabed during
cmbedment is within allowable ranges and to verify
E.6.2.1 Direct Embedded Plate Anchors anchor position;
"rm" . 1 1 . n 1. 1 1
uirect cmbedment or plate anchors can be achieved by
1' 11
r Direction of a C3('3CT‘CD >-1 (Da CT(D C2.W‘::(D Z3 5'!“
9'- l-

suction, impact or vibratory hammer, propellant, hydraulic ~ Anchor penetration.

ram, or gravity. The suction embedded plate anchor has been
used for major offshore mooring operations. As an example, E.6.3 TEST LOADING OF ANCHORS
the SEPLA (Suction Embedded _P_Late gnchor) uses a so-
called suction follower, which is essentially a reusable suc- P-or suction piles, suction caissons, and plate anchors, the
tion anchor with its tip slotted for insertion of a plate anchor. installation records should demonstrate that the anchor pene-
The suction follower is immediately retracted by reversing tration is within the range of upper and lower bound penetra-
the pumping action once the plate anchor is brought to the tion predictions developed during the anchor geotechnical
design depth, and can be used to install additional plate design. In addition, the installation records should confirm
anchors. In the SEPLA concept, the plate anchor’s fluke is the instal.lation behavior, i.e. self weight penetration, embed-
embedded in vertical position and necessary fluke rotation is rnent pressures, drag embedment loads, and that the anchor
orientation is consistent with the anchor design analysis.
achieved during a keying process by pulling on the mooring
Under these conditions, test loading of the anchor, as per Sec-
line.
tion 7.4.3 of main text, should not be required. However, the
Installation procedures should be developed and installa- mooring and anchor design should define a minimum accept-
tion analyses should be performed for direct embedded plate able level of test loading. This test loading should ensure that
anchors to verify that the anchors can be penetrated to the the mooring line’s inverse catenary is sufficiently formed to
design depth. The installation analysis should also consider prevent unacceptable mooring line slacking due to additional
plate anchor retrieval if applicable. inverse catenary cut-in during storm conditions.
For the cmbedment analysis, the risk of causing uplift of Plate anchors should be subjected to adequate keying loads
the soil plug inside the suction cmbedment tool should be to ensure that sufficient anchor fluke rotation will take place
considered. The allowable underpressure to avoid uplift without further loss of anchor penetration. The keying load
should exceed the required cmbedment pressure by a factor required and amount of estimated fluke rotation should be
of 1.5 (see Section E.3.2.l.3). Plate anchor installation toler- based on reliable geotechnical analysis and verified by proto-
ances should be established and should be considered in the type or scale model testing. The keying analysis used to
anchor’s geotechnical, structural, and installation design. establish the keying load should also include analysis of the
Typical tolerances to be considered are: anchor’s rotation when subjected to the maximum intact and
114 AP! Rricouuauoao Paacrica ZSK

l-line damage survival loads. If the calculated anchor rotation documented in API RP 2A. The recommended criteria in API
during keying differs from the anchor rotation in survival RP 2A should be applied for the design of driven anchor
conditions, then the anchor’s structure should be designed for piles, but with some modifications to reflect the differences
any resulting out-of-line loading to ensure that the anchor’s between mooring anchor piles and fixed platform piles. Some
structural integrity will not be compromised. of the guidance provided in Section E5 for structural capacity
In cases where the installation records show significant of suction piles may also be applicable for driven piles. The
deviation from the predicted values and these deviations design of a driven pile anchor should consider four potential
indicate that the anchor holding capacity may be compro- failure modes:
mised, test loading of anchor, as per Section 7.4.3 of main
l. Pull-out due to axial load.
text, may be required and may be an acceptable option to
prove holding capacity for temporary moorings. However, 2. Overstress of the pile and padeye due to lateral
testing anchors to the maximum intact load does not neces- bending.
sarily prove that required anchor holding safety factors are 3. Lateral rotation and/or translation.
met, which is of special concern for permanent mooring sys- 4. Fatigue due to environmental and installation loads.
tems. Consequently, if the installation records show that the
anchor holding capacity is significantly smaller than calcu- Factors ofsafety for holding capacity, defined as the calcu-
lated and factors of safety are not met, then other measures, lated soil resistance divided by the maximum anchor load
as listed below, to ensure adequate factors of safety should from dynamic analysis, are provided in Table 7. Information
be considered.
on coupling between vertical and horizontal. capacities can be
found in Section E.3.4. Axial safety factors consider that the
' Additional soil investigation at the anchor location to pile is primarily loaded in tension, and are therefore higher
establish and/or confirm soil properties at the anchor than for piles loaded in compression. As with other piled
site, foundation systems, the calculated ultimate axial soil resis-
- Retrieval of the anchor and re-installation at a new tance should be reduced if soil set-up, which is a function of
undisturbed location, time after pile installation, will not be complete before signif-
icant loads are imposed on the anchor pile.
~ Retrieval of the anchor, redesign construction of As the lateral failure mode for piles is considered to be less
the anchor to meet design require E;
g(E)Q:3 $0recd 1'€*l.HSt3li&-r
(“PG0
Q
catastrophic than the vertical, lower factors of safety have
tion at new undisturbed location, been recommended in Table 7 (see Section 7.4) for lateral
~ Delay of vessel hookup to provide additional soil con- pile capacity. Use of separate factors of safety for vertical and
solidation. lateral pile capacities may be straightforward for simple
beam-column analysis of, for example, mobile moorings (ref.
Drag embedded plate anchors should be test loaded, as per
Section 7.4.3 of main text, unless one of the following condi- Section E.3.2.3), but more complex methodologies do not
tions are complied with: differentiate between vertical and lateral pile resistance. The
safety factor should be in accordance with the guidelines of
~ The anchor installation load (drag-in load) is equal or Section E.3.5.
higher than the anchor required test load, as per as per
Section 7.4.3 of main text, and the anchor is not keyed E.7.2 GEOTECHNICAL AND STRUCTURAL
in the opposite direction, STRENGTH DESIGN
~ Soil properties at the anchor locations have been estab- In most anchor pile designs, the mooring line is attached
lished in accordance with E.2, the depth of the anchor to the pile below the seafloor, to enhance the lateral capacity.
after keying is known with reasonable accuracy and is As a result, the design should. consider the mooring line angle
not less than the minimum depth used for the design of at padeye connection resulting from the reverse catenary
the anchor. through the upper soil layers. Calculation of the soil resis-
tance above the padeye location should also consider remold-
E.7 Driven Pile Anchor ing effects due to this trenching of the mooring line through
E.7.1 BASIC CONSIDERATIONS the upper soil layers.
Driven pile anchors in soft clay typically have aspect ratios
Driven pile anchors provide a large vertical load capacity (penetration/diameter) of 25-30. Piles having such an aspect
for taut mooring systems. The design of driven pile anchors ratio would be fixed in position about the pile tip, and conse-
builds on a strong industry background in the evaluation of quently would deflect laterally and fail in bending before
geotechnical properties and the axial and lateral capacity pre- translating laterally as a unit. Driven pile anchors are typi-
diction for driven piles. The calculation of driven pile capaci- cally analyzed using a beam-column method with a lateral
ties, as developed for fixed offshore structures, is well load-deflection model (P-y curves) for the soil, with an
 DESiGN auo i’3%.NAi.‘r’SiS or STATiON¥<EEPiNG Svsreeis FOR Froarrno Sraucrunss 115

awareness of th.e limitations of such models as described. in tions are typically at the padeye, at the girth welds between
Section E.3.2.3.3. These computati.ons should include the the padeye and the pile, and between subsequent pile cans.
axial loading in the pile, as well as the mooring line attach» The evaluation of SCFS for girth welds needs to account
ment point, which will influence the deflection, shear, and for the local thickness misalignment at the weld. Equations
bending moment profiles along the pile. Pile stresses should for SCP-‘s are given in {E15, E16}. Note that the calculated
be limited to the basic allowable values in API RP 2A under SCF needs to be corrected by the ratio of the nominal thick-
intact condition. Basic allowable stresses may be increased ness used in the pile response analysis to the lesser of the pile
by one~third for rarely occurring design conditions such as wall thicknesses joining at the weld.. The SCF is to be applied
one-line damaged condition (see Sections E.5.4.2 and E.5.4.3 to the nominal pile stress range obtained at the weld location
for more detailed discussion). due to in-place loads, from which damage is to be calculated.
As argued in {B45}, “static” P-y curves may be considered
for calculation of lateral soil resistance. “Cyclic” P»-y curves E.7.3.3 Installation Loading
may be more appropriate for fatigue calculations. A rnodifi-
cation to the API RP 2A P-y curves has been proposed in Dynamic loads due to hammer impact during pile installa-
[E.5], to ensure that lateral deflections are not over-predicted. tion will induce fatigue damage on both pad.eye and pile girth
Consideration should be given to degrading the P-»y curves welds. The evaluation of the cyclic loads involves the
for deflections greater than 10% of the pile diameter. In addi- dynamic response of the pile-soil system due to the hammer
tion, when lateral deflections associated with cyclic loads at impact. This requires a wave equation analysis per blow for a
or near the mudline are relatively large (e.g., exceeding ye as given hammer type and efficiency, pile penetration, and soil
defined in API RP 2A for soft clay), consideration should be resistance. Various such analyses are to be conducted for judi-
given to reducing or neglecting the soil»-pile adhesion (skin ciously selected pile penetrations. For each analysis, traces of
friction) through this zone. stress versus time at the critical locations along the pile are to
The design of driven anchor piles should consider typical be developed, as well as the number of blows associated with
installation tolerances, which may affect the calculated soil the assumed penetration.
resistance and the pile structure. Pile verticality affects the For either welds or padeye, fatigue load calculations
angle of the mooring line at the padeye, which changes the should be carried out at various pile locations using local
components of horizontal and vertical mooring line load that stress range, derived from the wave equation analysis at the
the pile must resist. Underdrive will affect the axial pile selected pile penetrations. The location of the girth weld
capacity and may result in higher bending stresses in the pile. should be determined by the pile makeup schedule. The local
Padeye orientation (azimuth) may affect the local stresses in response should include the corresponding SCF effect. The
the padeye and connecting shackle. Horizontal positioning number of cycles of the stress history per blow is obtained
may affect the mooring scope and/or angle at the vessel fair- using a variable amplitude counting method, such as the res-
lead, and should be considered when balancing mooring line ervoir [E.l7] or rainflow methods.
pretensions.
E.7.3.4 Fatigue Resistance
E.7.3 FATIGUE DESIGN Applicable SN curves depend on manufacturing processes
E.7.3.1 Basic Considerations and defect acceptance criteria. Typically, pile sections are
welded by a two-sided SAW process and are left in the as»-
Anchor piles should be checked for fatigue caused by in- welded condition. For this case, the D curve, as defined in
place mooring line loads. Fatigue damage due to pile driving [E.l8], may be used. Use of a higher SN curve for this appli»
stresses should also be calculated and combined with in-place cation, without additional treatment of the weld, should be
fatigue damage. For typical mooring systems, fatigue damage demonstrated by relevant data. Use of weld treatment meth-—
due to pile driving is much higher than that caused by in- ods, such as grinding, may support the upgrading of the SN
place mooring line loads. curve, provided that (1) the grinding process is properly
implemented, (2.) weld inspection methods and defect accep-
E.7.3.2 In-Place Loading tance criteria are implemented, and (3) pertinent fatigue data
are generated to qualify the weld to a performance level
A global pile response analysis accounting for the pile-soil higher than that implied by the D curve.
interaction should be carried out for the mooring line reac-
tions due to the fatigue seastates acting on the system. The
E.7.3.5 Total Fatigue Damage and Factor of Safety
local stresses that accumulate fatigue damage in the pile
should be obtained by calculating a SCF (Stress Concentra- Once the fatigue loading and resistance are determined,
tion Factor), relative to the nominal stresses generated by the fatigue damage due to in-place and installation loads can be
global analysis, at the fatigue critical locations. These loca~ evaluated using procedures similar to those described in Sec-
116 API Recount-moan PRACTICE 2SK

tion 6 and Section E.5.5. The total fatigue damage should sat- E.5. J-B. Stevens and J-M. Audibert, “Re-examination of P-y
isfy the following equation for the critical structural elements Curve Formulations,” OTC #3402, May 1979.
E.6. NGI (1999): “Skirted Foundation in Clay. State Of The
D = F(D1) + F(D2) -< 1 (E8) Art. Set-Up Effects Outside Skirt Wall,” Joint Industry
Project. NGI Report 524071-2; 11 May 1999.
where
E.7. Andersen, K. H. and Jostad, H. P. (2002), “Shear
F I Factor of safety, equal to 3.0, Strength Along Outside Wall of Suction Anchors in Clay
D1 e Calculated fatigue damage for Phase 1, i.e., After Installation,” Proceedings, 12th ISOPE Conference,
Houston, ISOPE 10824.
installation (pile driving) phase and transporta-
tion phase, if significant. E.8. Randolph, M. and A. House (2002), “Analysis of Suc-
tion Caisson Capacity in Clay,” Proc. Offshore Technology
D; K Calculated fatigue damage for Phase 2, i.e., in-
Conference, Houston, OTC14236.
service phase, during the service life (e.g., 20
years). E.9. Dendani, H. And J-L. Colliat (2002), Girassol: “Design
Analysis and Installation of the Suction Anchors,” Proc. Off-
Further discussions on fatigue damage design for driven shore Technology Conference, Houston, OTC14209.
piles can be found in [E.18, E. 19].
E.10. Naval Civil Engineering Laboratory, Handbook for
E.7.4 TEST LOADING OF DRIVEN PILE ANCHORS Marine Geotechnical Engineering, Karl Rocker; March
1985.
The driven pile installation records should demonstrate
that the pile self weight penetration, pile orientation, d;riving E.11. Erbrich, C. And P. Hefer (2002), “Installation of the
records and final penetration are within the ranges established Laminaria Suction Piles---A Case History,” Proc. Offshore
during pile design and pile driving analysis. Under these cir- Technology Conference, Houston, OTC 14240.
cumstances, test loading of the anchor to full intact storm E.12. Clukey, E.C., “Suction Caisson Design Issues, Geolog-
load should not be required. However, the mooring and ical, Geotechnical, and Geophysical Properties of Deepwater
anchor design should define a minimum acceptable level of Sedirnents,” Proc. OTRC 2001 lrrt’l Conf. Honoring Prof. W.
test loading. This test loading should ensure that the mooring A. Dunlap, Conducted by the Offshore Technology Research
line’s inverse catenary is sufficiently forrned to prevent unac- Center, edited by C. Aubeny and J--L. Briaud, Houston, May.
ceptable mooring line slacking during storm conditions due
to additional inverse catenary cut-in. Another function of the E.l3. Gilbert, R. E. and Murff, J. D. (2001), “Workshop
test loading is to detect severe damage to the mooring compo- Report--Design Methodologies and Criteria for Suction
nents during installation. Caissons for Deepvvater Mooring Applications,” Prepared by
Offshore Technology Research Center, College Station, June.
E.8 References E.14. Norwegian Geotechnical Institute, “Skirted Founda-
E.1. Doyle, E. H. (1998), “The Integration of Deepwater tions and Anchors in Clay. State-of~the-Art Report,” Joint
Geohazard Evaluations and Geotechnical Studies,” 30th Industry Project. NGI Report 524071-1. 15 December 1997.
Annual Offshore Technology Conference, OTC 8590, Hous- Confidential.
ton, May. E.15. Buitrago, Jaime, and Zettlemoyer, Nicholas, “Fatigue
E.2. Jeanjean P., Andersen K., and Kalsnes, B. (1998), “Soil Design of Critical Girth Welds for Deepwater Applications,”
Parameters for Design of Suction Caissons for Gulf of Mex- Proceedings, Offshore Mechanics and Arctic Engineering,
ico Deepwater Clays,” Proc. Offshore Technology Confer- Paper 98-2004, Lisbon, July 1998.
ence, Houston, OTC 8830,Houst0n, May 1998. E.16. Connelly, L. M. and Zettlemoyer, N., “Stress Concen-
trations at Girth Welds of Tubulars With Axial Wall Mis-
E.3. Andersen, K. H. and Jostad, H. P. (1999), “Foundation
Design of Skirted Foundations and Anchors in Clay,” 31st alignment,” Proceedings, 5th International Symposium of
Annual Offshore Technology Conference, OTC 10824, Tubular Structures, Nottingham, U.K., August 1993.
Houston, May. E.1’7. British Standards Institute, Code of Practice for
Fatigue Design and Assessment ofSz‘eeZ Stmctnre.s, BS 7608,
E.4. Murff, J. D. and Hamilton, J. M. (1993), “P-Ultimate For
1993.
Undrained Analysis Of Laterally Loaded Piles,” Journal of
Geotechnical Engineering, ASCE, Vol. 119, No. 1, pp. 91- E18. Hunt, R. I-, Chan, Jack H-C, and Doyle, Earl H., “Driv-
107. ing Fatigue Estimation for Ursa TLP 96” OD Piles,” Proceed-
 DESEGN ass Answers or-' S'3'ATi{3¥\§KEEPi¥\2G Svstsras roe Froarruo Srsocrosas ‘it?

ings, 9th International. Offshore and Polar Engineering Clay,” ASCE Civil Engineering In The Oceans V, Texas
Conference, Brest, France, May 1999. A&M University, November 2-4.
E.19. Buitrago, J. and Wong, P. C., “Fatigue Design of Driven E.31. Skempton, A. W., (1951): “The Bearing Capacity Of
Piles for Deepwater Applications,” Proceedings, Vol. II, 13th Clays,” Proceedings, Building Research Congress, London,
International Offshore and Polar Engineering Conference, pp. 180-189
Honolulu, Hawaii, May 25-30, 2003, pp. 726-733.
E.32. Lunne, T., Roberston, P.K.., Powell, J. (1997): Corie
E.20. Bogard, D. “Effective Stress and Axial Pile Capacity: Penetration. Testirig in Geotechnical Practice, ISBN 0419
Lessons Learned. from Empire, OTC paper ii 13059, May 23750 X, Spon Press. 312 pages.
2001.
E.33. Newlin, J-A.: “Suction Anchor Piles for the Na Kika
E.21. Bogard, D. and Matlock, H.: “Application of Model
FDS Mooring System. Part 1: Site Characterization and
Pile Tests to Axial Pile Design,” OTC paper ti 6376, May
Design.,” Proceedings, International Symposium “Deepwater
1990.
Mooring Systems: Concept Design, Analysis, and Materials,
E.22. Doyle, E. H., (1998), “The Integration of Deepwater Oct. 2-3, 2003, Houston, TX.
Geohazard Evaluations and Geotechnical Studies,” 30th
Annual Offshore Technology Conference, OTC 8590, Hous- E.34. Newlin, J-A.: “Suction Anchor Piles for the Na Kika
ton, May. FDS Mooring system. Part 2: Installation Performance,” Pro-
ceedings, International Symposium “Deepwater Mooring
E.23. Jeanjean, P., Campbell, K., and Kalsnes, B.: “Use of systems: Concept Design, Analysis, and Materials, Oct. 2-3,
Integrated Study to Characterize the Marlin Deepwater 2003, Houston, TX.
Fiel.d.,” Proc. Society of Underwater Technology Conference,
September 1998, London. E.35. DNV (2002): Recommended Practice DNV~R.P-.E3()2-4
Design and lrzstallation of Plate Anchors iri Clay, Issued
E.24. Dutt, R. N., Rainey, W. S., H.amil.ton, T. K.., Pelletier, J. December 2002.
H., and Doyle, E. H., (1997), “Recent Advances in Deepwa-
ter Gulf of Mexico Geotechnical Investigations,” 29th E.36. Andersen, K., Murff, J. D., and Randolph, M. (2004):
Annual Offshore Teclmology Conference, 8303, Hous- “Deepwater Anchor Design Practice--Phase II Report to
ton, May. API/Deepstar--Volume ll-4 Suction Caisson Anchors“-
Findings and Recommendations,” Report submitted to API/
E.2S. Young, A. G., C. D. H.onganen, A. J. Silva, and W. R.
Deepstar. I
Bryant, 2000, “Comparison of Geotechnical Properties from
Large-diameter Long Cores and Borings in Deep Water Gulf E.37. Andersen, K., Andresen, L., Jostad, H.P., and Clukey,
of Mexico,” 32nd Annual Offshore Technology Conference, E., (2004): “Effect of Skirt Tip Geometry on Set-up Outside
Houston, Texas (OTC 12089). Suction Anchors In Soft Clays,” Paper OMAE2004-51564,
E.26. Borel, D., Puech, A., Dendani, H, and de Ruijter, M., Proceedings, 24th OMAE Conference, Vancouver.
(2002), “High Quality Sampling for Deepwater Engineering: E.38. Chen, W. F., (1975). Limit Analysis and Soil Plasticity,
the STACOR Experience,” Ultra Deep Engineering and
Elsevier Publishing Co., Amsterdam, The Netherlands.
Technology Conference (LHDET 2002), Brest, France.
E.39. Randolph, M. F. and Houlsby, G. T. (1984), The Limit-
E.27. Doyle, E. H. and Kaluza, M. J-, (2001), “A Review on
ing Pressure on a Circular Pile Loaded Laterally in Cohesive
the Use of 3-D Exploration-Level Data for Geohazard. Stud-
Soil, Geotechnique, London, England, 34(4), 613-623.
ies,” Proc. OTRC 2001 Int’l Conf. Honoring Prof. W. A.
Dunlap, Conducted by the Offshore Technology Research E.40. Ruinen, R. (2004): “Penetration Analysis of Drag
Center, edited by C. Aubeny and J--L. Briaud, Houston, May. Embedment Anchors in Soft Clays,” Presented at ISOPE
E.28. Andersen K. H. and Lauritzen, R., (1988), “Bearing Conference 2004.
Capacity for Foundations with Cyclic Loads,” ASCE Jcur E.41. Randolph, M. F. and House, A. R. (2002), “Analysis of
Geotechnical Erigr., V114, no. GT5, pp 540~555.
Suction Caisson Capacity in Clay,” Paper No. OTC14236,
E.29. Andersen, K. H., Kleven, A. and Heien, D, (1988), Proceedings of the Offshore Technology Conference, Hous-
“Cyclic Soil Data for Design of Gravity Structures,” ASCE, ton, Texas, May 6-9.
Joar Geotechnical Erzgra, v l14,GT5, pp 517-539.
E.-42. McClelland, B. and Focht, J. A. (1958), “Soil Modulus
E.30. Dutt, R. N., Doyle, E. H. and Ladd, R. S., (1992), for Laterally Loaded Piles,” Transactions ASCE, Vol. 123:
“Cyclic Behavior of a Deepwater Normally Consolidated 1049.
118 API Racommnmoso PRACTICE ZSK.

E.43. Matlock, H. (1963), “Applications of Numerical Meth- E53. Andersen, K., Murff, J. D., and Randolph, M. (2004):
ods to Some Structural Problems in Offshore Operations,” “Deepwater Anchor Design Pract;ice-»-~Phase II Report to
Journal Qf'Petr0Zeum Technology, 1040- l. 046. API/Deepstar-Vol.urne II-2 Suction Caisson Anchors»--3D
E.44. Matlock, H. (1970), Correlations for Design of Later- Finite Element Analyses and Comparison with Simplified
ally Loaded Piles in Soft Clay,” Proc. Offshore Technology Capacity Prediction Methods,” Report submitted to API/
Conference: OTC Paper No. 1.204, Houston. Deepstar.

E.45. Hamilton J-M. and J-D. Murff (1995), “Ultimate Lat» E54. Andersen, K., Murff, J. D., and Randolph, M. (2004):
eral Capacity of Piles in Clay,” Proc. Offshore Technology “Deepwater Anchor Design Practice--~Phase II Report to
Conference, Houston, OTC 7667, Houston, May 1995. API/Deepstar-»\/olume II~3 Suction. Caisson Anchors---~
Back Calculation of Prototype Installation Behavior,” Report
E.46. Grande, L. (1976) “Samyirke mellom pel og j ord.” Lic.
submitted to API/Deepstar.
Techn. Thesis, NTH, Institute for Soil Mechanics a11d Foun-
dation Engineering. Trondheim, Norway. E55. Clukey, E, Aubeny, C., and Murff, D. (2003): “Corn-
E.47. Colliat, J-L., Boisard, P., Andersen, K.H. and parison Of Analytical And Centrifuge Model Tests For Suc-
Schroeder, K. (1995), “Caisson Foundations As Alternative tion Caissons Subjected To Combined Loads,” Proceedings,
Anchors For Permanent Mooring Of A Process Barge Off~ 22nd International Conference on Offshore Mechanics and
shore Congo,” Proceedings of the 27th Annual Offshore Arctic Engineering, Cancun, Mexico.
Technology Conference, Houston, Texas, OTC 7797.
E.56. Aubeny, C., Han, S., and Murff, D. (2003): “Refined
E.48. Randolph MF, Hefer, PA, Geise Jl\/I, and Watson PG. Model for Inclined Load Capacity of Suction Caissons,” Pro-
(l998b), “Improved Seabed Strength profiling Using T-Bar ceedings, 22nd International Conference on Offshore
Penetrometer,” Proc. Conf. Offshore Site Investigation and Mechanics and Arctic Engineering, Cancun, Mexico.
Foundation, Behaviour ‘98, Society for Underwater Technol-
ogy. E57. Clukey, E., Templeton, J», and Randolph, M. (2004):
“Suction Caisson. Response Under Sustained Loop Current
E.49. Whittle, A.J. & Sutabutr, T., (1999), “Prediction of Pile Loads,” Proceedings, Offshore Technology Conference,
Set-up in Clay,” Transportation Research Record, 1663, 33~ Paper #16843, Houston, TX.
41.
E.58. Andersen, K., Murff, J. D., and Randolph, M. (2004):
E50. Sparrevik, P.: “Suction Anc.h.or—~»»-A (versatile Founda-
“Deepwater Anchor Design Practice---.Phase II Report to
tion Concept Finding Its Place in the Offshore Market,” Pro-
API/Deepstar---Volume III--Vertically loaded Drag
ceedings of the 17th International Conference on Offshore
Anchors,” Report submitted to API/Deepstar,
Mechanics and Arctic Engineering, OMAE 98-3096.
E51. Randolph, M., O’Neill, M.P., and Stewart, D.P: (1998): E.59. Andersen, I<I.., Murff, J. D., and Randolph, M. (2004):
“Performance of Suction Anchors in Fine-Grained Calcare~ “Deepwater Anchor Design Practice-~»-»Phase II Report to
ous Soils,” Paper #8831, Proceedings, Offshore Technology API/Deepstar---Volume III-4--Vertically Loaded Drag
Conference. Anchors~»~»~»~Findings and Recommendations,” Report submit-
ted to API/Deepstar.
E52. Andersen, K., Murff, J. D., and Randolph, M. (2004):
“Deepwater Anchor Design Practice-—~Phase II Report to E.60. DNV (2002): “Recommended Practice DNV-RP-
API/Deepstar--“Volume II--~Suction Caisson Anchors,” E302--~Design and Installation of Plate Anchors in Clay,”
Report submitted to API/Deepstar. Issued December 2002,
 APPENDIX F---DETERMINATION OF AVAILABLE THRUST

F.1 Introduction lated first. This thrust has to be corrected by applying thrust
deduction factors. These factors depend on the following:
This appendix provides guidelines for the determination of
the thrust generated by various types of propulsion devices. a. Propeller/thruster installation geometry and arrangement.
Also addressed is the influence of the installation and b. Inflow velocity into the propeller.
arrangement of the propulsion devices, which often leads to a c. Propeller sense of rotation (ahead or reverse operation).
reduction of the available thrust (net force acting upon the
vessel). F.3 Propeller Thrust at Bollard Pull
The guidelines apply to typical propulsion devices and
The following paragraphs (F.3.1 and F.3.2) provide guide-
installation scenarios for DP vessels supporting offshore
lines for the calculation of the thrust generated by open and
operations. These include the following:
nozzled propellers at bollard pull and certain inflow veloci-
a. Open and nozzled propellers installed in the stern of a ties, if required. Any deduction caused by the factors men-
ship-shaped vessel (conventional main propulsion tioned in Section 3 have to be applied later. The calculation.s
arrangement). of these deductions are discussed in Section 5. The calcula-
b. Azimuthing or flxed-direction, nozzled thrusters installed tion of the thrust produced by tunnel thrusters and its associ-
under the bottom of a hull. ated deductions are described in Section 6.
c. Tunnel thiusters installed in a transverse tunnel in a hull.
F.3.1 OPEN PROPELLERS
Two methods of thrupst evaluation are provided:
Figure F. 1 can be used for quick determination of the pro-
a. Tables and figures for quick and rough estimates that can
peller thrust at zero speed for an open propeller. Required
be used for the design of thruster assisted mooring and input data are propeller diameter and the power applied. The
prelirninary design of a DP system. diagram clearly indicates that, for a given power, the thrust
b. References for more rigorous determination of available increases with increasing propeller diameter. It also indi-
thrust. They be used for the final design of that. for given propeller the specific thrust increases
system. with d.ecreasing load. See C.6 for a list of references [C.l,
The estimated. avail.able thrust as determined by this appen- C.2, C.3, C.4, and C.5] that provide detailed information
dix should be further reduced under certain conditions as and data for the design and performance calculation of open
specified in 5.9. propellers.

F.3.2 NOZZLED PROPELLERS

F.2 Performance Criteria
Figure E2 allows a quick determination of the propeller
The performance of a conventional propeller, designed to
thrust at zero speed for a nozzled propeller. The same basic
power a vessel at a certain speed, is normally expressed by considerations apply as discussed above for open propellers.
the efficiency of the propeller. During stationkeeping, how- The diagram also indicates the considerable increase of thrust
ever, the propeller operates at zero inflow velocity (or at very for a nozzled propeller in comparison with an open propeller
low speeds), and the application of an efficiency expression is of same diameter and power load. See C.6 for references [C1,
not feasible. A popular expression for the performance of a C7, and C8] that provide detailed information and data for the
propeller in stationkeeping application is the specific thrust: design and performance calculation of nozzled propellers.
propeller thrust per horsepower.
Every propeller delivers maximum thrust at zero inflow F.4 Calculation of Thrust Deductions
velocity. Even in the case of a constant power operation
(which is feasible, for example, with controllable pitch pro- F.4.1 PROPELLERS INSTALLED AT THE STERN
pellers, or fixed-pitch propellers driven by certain prime OF A SHIP-SHAPED VESSEL
movers), the propeller thrust decreases with increasing inflow The suction of the propeller creates a low pressure field at
velocity. Inflow velocity is caused by either current speed, the aft body of the vessel, resulting in a reduction of the avail-
movement of the vessel, or the jet from another propulsion able propeller thrust. At zero inflow velocity, and with the
device. For the analysis of the stationkeeping propeller, the propeller turning into ahead direction, this reduction amounts
maxirnnm thmst at zero inflow (or bollard pull condition) to approxiniately 5 percent of the propeller thrust. During
will be considered the benchmark performance. astern operation, the reduction is about 15 to 20 percent.
To determine the available thrust (or net force acting upon Detailed data regarding the propeller/hull interaction for con-
the vessel), the propeller thrust at bollard pull has to be calcu- ventional vessels are included in F.6, [F.l, F.2, F.4, and F.5].
m
Aagfiawgdskogsmaw AW R E C OM M E N D E S ? QA C T E E 2 S K
9“nvRH
A_8V4
no

222Oh?w
iA1
qK _ll i; | Ii|“_£‘¥i!I|figi-' I!
%_dm‘_
‘“iUH‘“d!"5
__m $§ Ii¥|“i€l§-¥i iI_[liI| i_§2i§ i | l| i {

Ti:
iii:EE IEE‘:
E _W éWi___“
£_{5t
O
L
W
WW_:
W_"__
Z
_M

_
m
:_
_A
U[mm

_m
a
u
_
6_

“V M
m__d__w
_a
I__
_w
9M
I
WHW9_‘W
___
_
_ __A_{A Q5‘ fiii ‘|T lNiE{§iL1
W 1t_Y
*L __=
__

=
“
\AMlNEWHU5_‘ ”
_e
H__fiH_
E_
___
I
W
2
W
Y‘
4
W_ W%___wW__

_El_ _m WI¥§¥iIIll
E_I _I IW
}_ __ WW
_
_ H
W_7

5 f_
75 W_ 5i

5
_;

P O W 9 U P O D_ H m r r
D B _K A r 9 8 m p mu’Q fi }

_MN _
6 _"P_ 6_ Y_1p
_9_ nu Y e E F P Y O p Q E r T h F nu S §h O 9 8 “H P m 9 “W8 rS

dr A_
qu
QU O
HO
2
R5
_?_fiWWz‘R
3 Q
__:__‘W
MWw n____
W1*5
“I
‘_
___“ 1 1“hJ_{“$3i
“L
1‘_ ; ‘
\\‘\“[‘_‘"HQ\“khW;\‘i;:(‘;f\: m “ “ “ 1‘i:“ “ ‘ ‘ ‘ ‘ 1‘
JA _Wi_Vfw\flWuAI Z_
‘} } “‘} “ }ku“ ‘hL
_ __
_ is_ _
__ Wm _mW
E
4
Y___‘W MW_1__fif %*W__W%
|A_l
“ “ W“1‘_l l ‘l ‘l l l1‘“ i

_
_ ‘“I ‘ ‘ ‘ W
‘L_ NFN
1 _i‘
?:Q;Hw3:*PO$BmQw
_1“ ‘:‘“\1l}“‘:1gi“H‘k“h‘

A
fl_‘31‘i_r“:_k_\‘Lki“‘_

_ _ii _ _

Q“9“?_22
64
nu8
2
_ “ 1 __ _‘‘
“:15l‘ _ ‘
V‘
7
"“%___"W\ i

_
r~__
W_

T‘ __ _W _a

2S

D‘ G W 8 gfip m “P Q E Y Q ’$ R A r 8 a { h m_ Q W

Q 9 U r9 F2 D_r m E r0 p r T _h r U S t P r O p ‘Q
Mr W m N O ZH 8 S
 Dasrorr sue Anarvsis or Srarrouxsaerno Svsrarvrs son Froarrno Sraucroaes 121

F.4.2 RIGHT ANGLE GEAR THRUSTER F.4.5 THRUST IN REVERSE OPERATION

PROPELLERS
Some of the thrust producing devices applied for dynamic
The presence of the gear hou.sing and support struts in the positioning need to reverse the operation of the propeller to
flow to the propeller causes a reduction of thrust. For a. produce thrust in reverse direction. Azirauthing thrusters typ-
thruster of average design, this deduction is about 10 percent. ically produce thrust in one direction only. They control
In cases where the diameter ratio gear housing to propeller direction of thrust by controlling the azimuth angle.
exceeds 0.45, a deduction of 15 percent may be applied. Some thrusters, such as tunnel thrusters or fixed direction
nozzled thrusters are designed as bi-directional devices and
F.4.3 THRUST DEDUCTION DUE TO INFLOW are capable of generating approximately equal amounts of
VELOCITY thrust in both directions. Propellers optimized for operation in
For a propeller applied for stationkeeping, propeller opera- one direction (the majority of marine propellers) are subject
tion in certain inflow velocities is caused by currents as well to severe deductions while operating in reverse m.ode. Table
as by the wake created by thrusters operated in the vicinity. F.2 indicates values for thrust losses of nozzled propellers
Table F.l indicates approximate deductions of thrust as a [C24].
function of the inflow velocity. An. accurate prediction for the
performance of ducted or open propellers at certain inflow F.4.6 THRUST DEDUCTION DUE TO PROPELLER!
velocities is feasible by a detailed analysis IF. 1, R2, R3, E4, HULL INTERACTION
P15, P17, and R8]. See F6 [F23 and F-.24] for references
including information regarding the thrust losses caused by F.4.6.1 Coanda Effect
the m.utual interference of thrusters. The high-velocity wake from a propulsion. device installed
under the bottom of a vessel may cause areas of low pressure
Table F.1-- Correction Factor for inflow Velocity
at the hull that result in considerable deductions from the
Inflow Velocity in Knots available thrust forces. The magnitude of these deductions
Type Propeller l 2 3 4 depend on the distance of the propeller from the hull, the
location of the propeller relative to the centerline of the hull,
Open Propeller .95 .90 .85 .80 ":§f\1Hf\ 1‘\|'\ *III'\I'1-:'\'\(“\ l"\ l"l'l4\f'1 ‘F Il('\ .f'\ I
1.1u.111 ‘int: tauiuo Of ‘tho 10115-5&5, (Il.1.lU J.1'Oi"i"i l.l’lC draft uf {TED v’€SSCl..

Nozzled Propeller .94 .88 .82 .76 A correction factor from 5 percent to 15 percent should be
applied to account for this hull interaction. Sources of infor-
mation and data regarding the thrust losses due to propellerx’
F.4.4 THRUST DEDUCTION DUE TO OBLIQUE
hull interference are included {E17 and R24}.
INFLOW CROSS-COUPLING EFFECTS
The operation of a propeller in an inflow other than parallel F.4.6.2 Twin-Hulllnteraction
to the propeller axis alters the performance characteristic.
Deductions due to inflow velocity may be reduced. However, This effect occurs at twin-hull semisubmersibles equipped
the creation of crossacoupling forces may cause deduction with rotatable, under the hull mounted propulsion devices. At
from the overall balance of forces. The direction of these certain azimuthing angles, the propeller jet from the thruster
forces are orthogonal to the propeller axis. These effects are is directed towards the neighbor hull, causing a resistance
the least researched subjects in propulsion for dynamic posi- opposite the direction of the thrust. This effect can be ampli-
tioning. See F.6 [F.l'7, F.l8, F.l9, and I-7.24] for sources of fied by the above mentioned Coanda effect. Little informa-
information and qualitative data. tion regarding these effects is available. The magnitude of the
thrust losses depends on the thruster installation geometry
Table F.2--Thrust Losses in Reverse Condition and the configuration of the semisubmersible hulls. Counter-
measures (which apply also to the Coanda effect) include hor-
Nozzle Type Loss iri Percent
izontally tilting the propeller axis downwards or fitting guide
Symmetric nozzle 5-l0% vanes to the exit of the nozzle. Both methods deflect the pro-
peller jet away from the neighbor hull. An indication has been
Non-symmetric nozzle, elliptic blades 10-25% found of an average thrust loss of l0 to 15 percent due to the
Non-symmetric nozzle, cambered blades 25-50% above discussed phenomena, with peak losses of over 50 per»
cent at some positions and in particularly unfavorable condi-
tions {R22}. A discussion of the above interaction effect,
including model test results, may be found in JerDeflec1fiorz
VLU’1(-ZS for Improved Performance cf Rotatable Thrusrers
[F22].
122 API Racoiuiviarvoeo Paacnca ZSK

F.5 Performance of Tunnel Thrusters nominal design submergence of the tunnel. If this submer-
gence is decreased due to a reduction in the draft, or due to
Though there are some similarities, tunnel thrusters differ
motions of the vessel, the thruster impeller will ventilate
in many ways from the other propulsion d.evices. They are
(sucking air) and/or cavitate. Both cause a reduction in impel»
analytically treated as axial flow pumps. As with marine pro-
ler thrust.
pellers, the thrust increases with decreased power load. A
large propeller diameter yields to a high thrust at a given The analytical determination of the losses due to the
power. The tunnel thruster is subjected to thrust deductions motions of the vessel is complex. First, a relative motion
by factors typical for axial flow pumps, restrictions in the analysis has to be performed for the environmental conditions
flow to and from the impeller, as well as tunnel entrance and in which the vessel is expected to operate. With these data--»
exit losses are the major contributors to the reduction in the periodic variations of the subinergence at the tunnel loca-
net thrast output. tion-—-the thrust losses duririg the operation of the impeller at
reduced submergence can be calculated [E24 and l-1125].
F.5.1 SIDE FORCE OF TUNNEL THRUSTERS
F.6 References
Figure F3 can be used for quick determination of the side
force of a tunnel thruster. The figure assumes an optimum F.1 Lewis, E. V. (ed), Principles of Naval Architeczfure,
installation geometry. The tunnel length is about twice the Vol. II, SNAME, 1988.
propeller diameter. The hull is perpendicular at the tunnel
exits. The exits are conically shaped. No protective bars F.2 an Manen, J. D., Fundamentals 0fS/tip Resistance and
restrict the tunnel ends. The impeller/hull interaction losses Propulsion, Parr B: Propulsion, NSMB Publication No.
are included. 132a.

F.3 O’Brian, T. P., The Design ofScrew .Pr0peZlers, Hutch-

F.5.2 THRUST DEDUCTIONS FOR TUNNEL inson & Co., Ltd., 1962.
THRUSTERS
E4 Saunders, Hydrodynamics in Ship Design, SNAP»/IE,
In addition to the thrust losses due to the installation geom- 1957.
etry typically associated with tunnel thrusters, further thrust
losses may occur during certain operational conditions. The F5 Harvald, S. A., Resistance and Propulsion 0fShz}2s,
performance prediction of a tunnel thruster is based on a Wiley, 1983.
26 3 _,____,,,__F,_ _ .. . L .

-» - v-.~ »

24 - - - - - _ l
+ il

_4_‘__

____;_

_, ---E'— Optimum value

22 - I‘ -
i l
4

D/h
9
flu ma4-a-n
4
1
1
Y1 . ‘.- . .
l —g--— Lowvalue
H . -»~.m».i .»_ i
Forep _HWP
if
1
-M

2Q l_ _ ___. _ _ ,,_,,,____
#——-w“~e~
D4 ..._1... B
‘ fl__
-f
S'de
‘c bl
fl
___... ,1
poof 18 - . ~ - 4» -----------------
S 1 Ii vi 1 '-'-‘ - ‘ 3 -
%n- '

1 ' 1
-iv ‘:
A
-.¢-. ‘i.< ». i ¢
4
1

1 6 --------------------------- - Q‘4
i1 l i - - ______a_: W, . __.______. - -"Ir
.
41
. _ ____... 7 7 _ _._— ,_7______

'1
A
T .
A l
. v 1
1 - V4
,.g,\.
14 ‘ .....,__ ...1._.......-.,.,,,,,,,_1_,_, . . .
25 35 45 if 65 75 as
Povver;/Propeller Disk Area (npfsq ft)
Note: Power delivered to the propeller.
Figure F.3—Side Force, Tunnel Thrusters
 Basie-i\i AND Anarvsis or STA"£'lONi<EEPiN{5- Svsreivts eon Froarrno Srauoroaas 123

F.6 Morgan, M., Dynamic Positioning of Ojjfshore Vessels, R19 Symons, P. J. R., et al, The Design of the SEABED
The Petroleum Publishing Company, 1978. Operations Vessel, RINA Transactions, 198-1.
F.7 Oosterveld, M. W., Wake Adapted Ducted Propellers, E20 Nienhus, U., Simulations ofLow Frequency Motions of
NSNB Publication No. 345. Dynamically Positioned Offshore Structures, RINA Transac-
F.8 Van Manen, J. D. and Oosterveld, M. W., Analysis of tions, 1986.
Ducted Propeller Design, SNAME Transactions, 1.966.
E21 Haslum, K., Thruster Operation Strategies and Fuel
F.9 Oosterveld, M. W. and van Oortmerssen, G., Thruster Economy, Norske S1i.vllingelorers Forening, 1988.
Systems for Improving the Maneuverability and Position-
Keeping Capabilities of Floating Objects, OTC No. 1625, F.22 Bjoerheden, O., Jet .Deflection Vanes for Improved
1972. Performance ofliotatable Thrusters, K.aMeWa, 1986.

F.10 Schneiders, C. C. and Pronk, C., Performance of E23 Lehn, E., Thruster Interaction Effects, NSF1 Report R
Thrusters, OTC No. 2230, 1975. 102.80, 1980.

F.11 Brix, J-, Maneuvering Devices for Offshore Services E24 Lehn, E., Practical Methods for Estimation of Thrust
and their Hydrodynamic Properties, INTEROCEAN No. Losses, Marintek Report No. 51.3003.00.06, 1992.
244, 1976.
E25 Olsen, H., Reduction ofBou/thruster Ejjiciency Due to
E12 Brix, J-, Maneuvering Technical Manual, Schiff & Ship Motions in Waves, Det Norske Veritas, 1978.
Hafen, June 1988.
E26 N. A., Eflect of Waves on Thruster Performance,
F..13 Sjouke, J. and Lagers, G., Development of Dynamic
NSMB Report, 1970.
Positi0ningforIHC Drill Ship, OTC No. 1498, 1971.
E27 Stuurman, A. M., Fundamental Aspects ofthe Effect of
R14 Wise, D. A. and English, J. W., Tank and Windtunnel
Propeller Cavitation on Radiated Noise, NSMB Symposium,
Tests for a Drill Ship with Dynamic Positioning Control,
1975.
OTC N0. 2345, 1975.
rwnn I I ‘|'"\ Tr?“ 1\ I ’ ‘\.T
F.15 English, J. W. and Wise, D. A., Hydrodynamic Aspects Lao Mooring Tanners in .L)(3(:’p i/rater, Marin, Report two.
ofDynamic Positioning, NDL Report Ship 197, 1976. 32, 5-1988.

F.16 Norrby, R. and Ridley, D. E., Notes on thrusters for E29 Mueller-Graf, B., Systematische Untersuchungen von
Ship .Maneuvering and Dynamic Positioning, SNAME Trans» Verstellpropellem, VWS Report No. 568/71.
actions, 1980.
E30 Guide for the Certification of Thrusters, American
F.17 Deter, D. R., SEDCO 445: Propulsion System for Bureau of Shipping, 1984.
Dynamic Positioning, INTEROCEAN No. 243, 1978.
F.31 Safety and Reliability ofDynamic Positioning Systems,
R18 Shatto, H. L. and Van Calcar, H., Improving Dynamic Det Norske Veritas, Symposium, 1975.
Positioning Performance in the Deepwater; High-Current,
Rough Water Environment, OTC No. 4749, 1984. E32 Eek, B., Technische Stroemungslehre, Springer, 1978.
 APPENDIX G—-MOORING STRENGTH RELIABILITY

G.1 Basic Considerations which is often referred to as LRFD (Load and Resistance
Factor Design) or partial safety factor format. Instead of
This Appendix discusses mooring system reliability and
using a total safety factor, this approach employs a number of
two approaches for mooring strength design: Detenninistic
load factors for the load components and a number of mate-
and LRFD (Load and Resistance Factor Design). Mooring
rial factors for the line component strength. Since the load
system reliability can be affected by many factors such as
and material factors are developed from reliability analysis
overloading, component defects/damage, fatigue, corrosion,
for a large number of cases representing different geographi-
maintenance, human. error, etc. The di.scussi.on here addresses
cal locations, water depths, vessel types, mooring types, and
the factor of overloading. Therefore the term “reliability” or
mooring components, this approach may yield mooring
“probability of failure” are related to failures due to overload-
designs with more consistent reliability.
i.ng only.
Up to year 2000, all official. mooring design codes use a
total safety factor to protect against uncertainties in the design
G.2 Deterministic Approach (Total Safety parameters such as environment, mooring line strength, sys-
Factor Format) tem response, and. method of analysis, etc. The lntemational
The strength design procedure specified in this document Standards Organization (ISO), however, recommends that the
is based on a deterministic approach, where the mooring sys- Structural Standards published by the ISO should be level-l
tem responses such as line tensions and anchor loads are eval- limit state codes with a partial safety factor format. This rec-
uated for a design environment defined by a return period. ommendation is not followed in the draft ISO Stati.onkeeping
The mooring system responses are then checked against the standard because of lack of fundamental work in this area. It
mooring component strength to ensure a factor of safety is expected however, that ISO will eventual.ly adopt a partial
against mooring breakage. The return periods and factors of safety factor mooring code as the ISO mooring standard.
safety, which were developed mainly by industry experience
and consensus, are generally applicable to all geographical G.4 DeepStar CTR 4404 Study “Reliability
locations, water depths, vessel types (semisubrnersible, Ship- Based Mooring Design (...3odes”
shaped, or spar), mooring types (catenary, taut leg, spread, or
single point), and mooring components (chain, wire rope, or Because of the ISO recommendation and the introduction
synthetic fiber rope). The only exception is that lower return of partial safety factor code in classification rules, the off-
periods are used for mobile moorings to reflect less severe shore industry is pressed to consider adopting a partial safety
consequence of mooring failure for mobile unit operations. factor format for mooring design. In I997 DeepStar initiated
This approach has the advantage of providing a simple CTR 4404 to investigation the feasibility of developing a reli-
design procedure that can yield mooring designs with reason- ability based mooring design code. This work is necessary to
able effort. It has been used by the offshore industry for more ensure that the future ISO mooring code will not be biased. by
than 30 years, and therefore it has a broad base industry sup- a single study.
port. However, there are also concerns on the shoitcomings of The scope of this study is summarized as follows:
this approach:
l. Develop study matrix. There are 25 cases in the study
1. Industry reliability studies indicate that mooring matrix, including the following parameters:
designs developed by this approach may have signifi- - Environment: GOM, North Sea, West Africa
cant difference in probability of failure.
- Water depth: 70m---3,000m
2. This approach was calibrated by past experience from ~ Vessel: semi, ship, spar
shallow water operations with more conventional ves-
~ Mooring: catenary, taut leg
sels and mooring systems. There -are questions whether
= Mooring line: steel, polyester
this approach will provide sufficient reliability for
deepwater operations with new vessel concepts such as 2. Develop mooring designs for the study matrix accord»
spar and new mooring concepts such as taut leg and ing to API RP 2SK, considering both intact and
fiber rope mooring. damaged conditions.
3. Determine random variables and modeling uncertainty.
G.3 Load and Resistance Factor Design 4. Calculate notional probability of failure. Monte Carlo
(Partial Safety Factor Format) reliability methods were used for evaluating notional
Recently substantial effort has been devoted to the devel- probabilities of failure for intact and damaged
opment of a design procedure based on a reliability approach, conditions.
126 API Racoswusnoso Paacrice ZSK

5. Develop partial factors of safety based on the follow- ing systems in West Africa have a lower probability of
ing principles. failure than mooring systems in the North Sea.
= l\/Iaintain average probability of failure 5. The calculation of probability of failure is affected by
i Minimize the spread in probabilities of failure uncertainty in many parameters, such as environmental
model, reliability analysis method, software for moor-
The conclusions of this study are:
ing and reliability analysis, and mooring design
l. The spread i;n the probabilities of failure for the differ- assumptions. Therefore calculated probabilities of fail-
ent mooring systems cannot be reduced by a single set ure should be considered notional and should be used
of partial safety factors applied to the mean, low, and comparatively. Setting an absolute target for probabil-
wave frequency components of the total line tension. ity of failure can be a dangerous practice.
From this point of view, there is no apparent benefit 6. Taut leg and catenary moorings have similar probabili-
from using a single set of partial safety factors post- ties of failure. The cases for comparison are limited in
applied to mean, low, and wave frequency components the DeepStar study. However, this has been confirmed
of line tensions. by other industry studies.
2. Using low safety factors for mean tension appears 7. Steel and polyester moorings have similar probabilities
inappropriate for the following reasons. of failure due to overloading. Again, the cases for com-
~ A low safety factor can be assigned to a load. corn- parison are limited in the DeepStar study. However,
ponent if the uncertainty for that component is low. this has also been confirmed by other industry studies.
The mean line tension consists of 4 components:
pretension, mean wind force, mean wave drift G.5 Conclusions
force, and current force, among which only the pre-
tension can be considered to have low uncertainty. The deterministic approach recommended in the present
All the other 3 components have significant uncer- revision of API RPZSK has the advantage of providing a sim-
ple design procedure that can yield mooring designs with rea-
tainty from the environmental criteria and load cal-
sonable effort. It has been used by the offshore industry for
culation.
more than 30 years therefore it has a broad base industry sup-
= Mean tension often dominates the moorings of port. However, there are also concerns on the shortcomings
many floating systems, especially the deepwater with this approach:
systems and those subjected to high currents. In this
case a low safety factor for mean tension is equiva- 1. Ind.ustry reliability studies indicate that mooring
lent to a low total safety factor and low reliability designs developed by this approach may have signifi-
for the system. cant difference in probability of failure.
~ Mean tension has a long duration and therefore can 2. This approach was calibrated by past experience from
be more harmful than the peak dynamic tension shallow water operations with more conventional ves-
which often lasts only seconds. sels and. mooring systems. There are questions whether
3. The most important factor affecting the mooring sys- this approach will provide sufficient reliability for
tern reliability is the environment. It has been shown deepwater operations with new vessel concepts such as
that the different levels of uncertainty inherent in the spar and new mooring concepts such as taut leg and
environmental models for the three geographic areas fiber rope mooring.
included in this study lead to different levels of proba- DeepStar and other industry reliability studies h.ave
bility of failure for mooring systems designed to a addressed the second concern. There is no evidence that float-
single set of safety factors. However this need not be ing systems operating in deepwater, using spar, taut leg moor-
the case. Theoretically we can design to similar level ing configuration or fiber rope have lower reliability against
of probability of failure, by accounting for environ- overloading than the more conventional systems. These stud-
mental uncertainty. However, the feasibility of ies, however, confirm the first concern. The DeepStar study
implementing environmental partial safety factors in a further pointed out that the most important factor affecting
stationkeeping code requires further study. the mooring system reliability is the environment. The differ-
4. Because of differences in the uncertainty associated ent levels of uncertainty inherent in the environmental mod-
with extreme environmental conditions, mooring sys- els for the three geographic areas included in the DeepStar
terns in the Gulf of Mexico have a higher probability of study lead to different levels of probability of failure for
failure due to overloading than mooring systems in the mooring systems designed to a single set of safety factors.
North Sea, when the mooring systems are designed to Theoretically we can design to similar level of probability of
the same single set of safety factors. Similarly, moor- failure by accounting for environmental uncertainty. How-
 D§S¥G¥\t mo Anatvsis or STATl£}¥\i§<EE¥>2¥\iG Svsrsns eon Froariuo Srnocrones 1.27

ever, the feasibility off implementing environmental partial reliability. However, other reliability studies indicate that
safety factors in a stationkeeping code requires further study. multiple partial safety factors do provide a more flexible
Other studies have emphasized uncertainty in the low~fre- design format, with a potential for producing designs with
quency response of the platform as a significant factor. more uniform reliability. This potential can be realized in
Uncertainties related to vortex indoced. motions of platforms mooring line design.
in high currents have not yet been taken into account in reli- The total safety factor approach is retained in the present
ability-based calibration of mooring design codes. revision of the recommended practice. However, designers
The DeepStar study indicates that the spread in the proba- should bear in mind various factors that may affect mooring
bilities of failure for the different mooring systems cannot be system reliability. The industry will., no doubt, continue to
reduced by a single set of partial safety factors applied to the improve the design procedures for mooring systems. Reliabil-
components of the total line tension. For mooring systems ity-based calibration is seen as a rational procedure to incor-
dominated by mean tension, using a low safety factor for porate varioas improvements into revised designed equations.
mean tension may actually result in mooring designs of low
 APPENDIX H-—-MOORING DESIGN FOR VORTEX INDUCED MOTIONS (VIM)

H.1 Basic Considerations H..1.2 VIM FUNDAMENTALS

H.1.1 PURPOSE OF DOCUMENT Cylindrical structures exposed to a current create alternat-
ing eddies, or vortices, at a regular period. Figure H.l shows
Many floaters consisting of cylindrical structures such as how these eddies appear in the downstream. wake of a cylin-
spars, TLPs, and semi-submersibles can be susceptible to der.
vortex. induced motion (VIM) when exposed to currents. The frequency of vortex shedding (f_,.) is related to the non-
VIM is most prominent for spar platforms where most of dimensional Strouhal number, S:
industry experience has been acquired. It has been noted
recently that under controlled experimental conditions, multi- fg " -== SVC /D l)
column floaters such as TLPs and semi-submersibles may
experience limited VIM. For these hull types, however, there
where
is no substantiated in-service evidence of VIM. Nevertheless,
designers of multi-column floaters should be aware of the V6 I Current velocity,
potential impact of VIM, particularly if novel concepts are
considered. D K Cylinder Diameter.
There are a number of special issnes for spar VIM, and The eddies create alternating lift and drag forces on the
guidance is needed by the industry to address these issues. cylinder. When a natural period of a structure falls close to
This Appendix presents some high level design guidelines for the Strouhal period, oscillations of the structure can occur.
mooring systems under VIM conditions. More detailed tech- This phenomenon is traditionally known as Vortex Induced
nical information, such as industry experience in model test- Vibrations (VIV), and is well known for risers and tendons.
ing and comparison with full scale data, strength and fatigue VIV is not restricted to long cylinders. Spar platforms, in
analysis procedures currently used by the industry, methods particular, experience vortex induced oscillations when their
to improve mooring design for VIM, and potential future surge/sway or roll/pitch periods are close to the Strouhal
technology development, can be found in the Commentary period. Figure H2. shows an example of the motion trajectory
for Appendix H. of a spar subjected. to a current of slightly over 2 kts. The
It should be noted that technology dealing with VIM is period of the motion in the transverse direction in this case is
advancing constantly. Designers are encouraged to pay atten- about 180 seconds, which is close to the natural sway period
tion to the latest research in this area and consider the use of of the spar. There is also a smaller motion in the in-line direc-
more advanced technology when it becomes available. tion. vvhich is at one-half the transverse period. Because of the

' ' lief-E-5 ' '

'
-:-.f_:.-.§'J 3 2 =;;- 5.
1 1;; -- _;-.-..-
- = .= ‘*4’
.;'-:;_." I.
"
3-
' J . I:

,1 . . _ ... .,, . . - .. _-._ -; § I: -' i ii» .. ..;...-.-;:-==--=-';5:=-55:=;E-i:E=ié-'"";"5I§"‘;‘I"""’“"'"&=E:E1E:i;E1-j;=;:;:;-=-=;.-.._.._. -5 Iii

if '= I iii :5 I" "ii" “‘- " I -' ---Z"? Arm?
wit“
(5.: iii ........-.1-..:.;.;z'-"ii ?""ii5§].§'§E;§=i:! §=-="" "’5"""“"’=‘=5*"-“‘3‘= ‘*5’ ===’-'35'%T§3§"555"‘:’ i i? ;5;§l§;"’§5§&;I;:;:;:.... ' :3 I 5 -:"3.3_§f-

é)
- .- .,._._:::-2-5
-. -_ _--___ . -2 -. €.“§~;¢/I
i\\<J»M>. 2» t . "” w x AA xi \ __11*,.;,@,.,§'4_§-s. -. _ . _:..
g‘. 5 E _.: ._: :-. _
‘-1
, ea
.,%< teexe
_ . I H 2 _ __ I_ _ _ _

‘ Q-\ :i§_\'.‘-as 1 _
{
<:=~| ..,......
1 :__- . . ._ - -- - -I: . E_.::.:_;._
.- _ - ;-_-I2
- - - _- -_:. _- . -. ._I_ . _-;_- -- _: ‘~ * i I “gr: V“ < _.;.;.-4:1:53;;1;1;;g;g:;.;;:1;;;;;,_-.-.;.;,..,._,..___ 2?, :._
__:_L:.555-FEE;3%;I:I:E?;;;_:3;.35E;EEE€?é€fiEEEE;E€E25:: _
-- '~,_ kc -\." " Q»-,_ ;-- -
'=.'}'--. ;j. __.,. Q.~.."'\~“‘v'”\
. -Q-.~w H " -:1-":21; .
- " " '- =1" - .. ._ .., has \‘A*='.='¢s-A-<-t>~“‘\\*"~._. “Tr ..,' -I

‘I

Q‘ ‘O
I U I I
I» I

Figure H.1—Eddies in the Downstream Wake of a Cylinder

130 API Recoumeuoeo Pnacncs 28K

H.1.3 EFFECTS OF VIM ON MOORING DESIGN

1
I(( ( ( ( ( ( WI
I
( ( ( ( ( ( ( ,{{ { { { s s is *******************if
VIM has two primary effects on the mooring design:
4,0 I*******************In*********** _ ,3... I
1. The average in-line drag coefficient in lock-in is
increased over what it is Without VIM, and
‘
I '
I
‘
.T5‘5i=;I=.555?:5=E;-'5-‘ziziziziz;.;.;.
‘L~..........~.......__.__......~....m..................;._.;.;._-.;.;.-.;.;.-_.;.;.;.;.; iziziz
'I'I'IjIjIjIjIjIjI_I
rrrr'7 l
_i ,,,,,,,,,,,,,,,,,,,,,,,,,,,~
l
M ,,,,,,,,,,,,,,,,,,,,, H 2. The motions impose displacements on the mooring
line fairleads which cause additional oscillating moor-
ing line tensions.
20 a—l W????????????~
These effects have to be taken into account for strength and
I
T
1 ¥EI§;'-;
l.-.-'-':3:3:-'.

i -§.%I=§§5I3;"3: 1
fatigue design of the mooring system. The occurrence of the
I---------—--——*E;§;§;§;f-;;'%§;§§;;§;-_E;§;§;5;§.;§;§;§
;I;i:3:3:3:':3:31-.3.-7;i:':I;.3_'§:§I-:
-- **********’—+ —
I
" V V V V~V****—~ ''" .
Loop Current and associated eddies in the Gulf of Mexico
5 make consideration of VIM particularly important. For exam»
1 i EEii i55-i "?Eii i5Eii iféiji -I-5'5f5i §:§&§§§§i§§'§i.€-féi i i 1 i

ple, unlike other extreme events, e.g., hurricanes, the Loop

1l -1]"3:f:§:§:§'{'.§'.§'§:§:§:§:§;'-.=_E-Z!IIE11I;§§I5.'-:3'.3:3:3:3:3:3:§:3
§z§z§z52§z§;;=.§z§z§z§=§z§zj;_5;':5525555iiEEE255;-:1-:;‘I§"§'§§§§§§§E5?;§§§iz:’£i§i.5i=E.=-‘.=-I
—- as( ( ( ( ( ( -F-~ l Current and associated eddies may affect a particular site for
l =E;;;;;;5-j.-;;j;I.;:5;;;j.-_I_-_I.-_-;I:555.5-5-;;_.-j$._=j-';E;;EIE;E;E;I;;;E an extended period of time and can contribute to low cycle
i
= 1i5§E§E'5-;§'§ §E§EfE§E§E5E;5EZ5;: §?_?§=§ I§ _E§j§ §f§5 I fatigue of mooring components.
23:3-1-3:.-: : :-: : : :-z-;=z=z-:-I-:;:=I-3-"£55525;.- Mm __________________W
i l—4—A m...._.
I

H.1.4 CURRENT INDUSTRY PRACTICE

~20 1 — F I Unlike other resonant responses, the amplitude of VIM is
bounded. Transverse motion amplitude (A) is usually given as
l I
i
I
l
I
I
I -‘.-‘Ii ?151:5?-'55."?E5;"E5E5E5E5i5E5E5i
"3'i'§I5I§="'-I-:3'3:3:3:3:3:3:3:3:§:-:-:-:-:-:-: :-:-
:'!-;i_-;P-'-."=3:3:':3:3:323:3:3:-:3:3:3."-."-.'3:':3:3
I
Il a non-dimensional ratio of amplitude to diameter, A/D. The
largest single amplitude transverse motion observed on bluff
mm .................-;1;-__-;-Z:E5;EEE55EE55;EEE§E§E§E§E§E§E§;:;:;§E§ ________ _:__._ :,7W_ _________________ ,,_, ,,,,, ,...,,

I ~":2I-2E2E2iziziziz£&ii zia""""‘*'*'
I _ y
bodies is on the order of A/D w l. Helical strakes have com»

C ((( i -(( ( ( C
‘ '.=i 5i i5E.E5ii5i i{BE55:;i§5i3Iz'?§I3:I?i€§i" .

monly been used on spars (and risers) to reduce this ampli-

I I I l It U tude. Strakes can ideally be 95% effective in eliminating
VIM, however their effectiveness spars depends on vari-
it | ous factors, e.g., the exact layout and size of the strakes,
appurtenances, and current profiles. There is currently no
Figure H.2--Motion Trajectory of a Spar analytical method for determining the motion amplitude,
Experiencing VIM although research is ongoing. The industry currently relies on
model testing for this purpose. There is still a lack of suffi-
long natural periods for floaters, the term “Vortex Induced cient full scale data from which to generalize on the scalabil-
Motions” (VIM) is typically applied to this phenomenon in ity of model testing, particularly for the newer generation of
lieu of VIV, although the phenomenon is the same. spars: the truss spar and the cell spar. It is usual practice, if
Semi-submersibles, TLPs and other types of floaters may current loadings and VIM are determined to be a design
also be susceptible to VIM, as vortex shedding can occur driver, to perform well planned model testing to determine
from the columns or pontoons. motion amplitudes and drag coefficients for purposes of
The occurrence of “lock~in” is related to the non~dirnen- mooring design.
sional Reduced Velocity E:
H.2 EnvironmentalConsiderations
V, = V, T/D (H.2)
H.2.1 CURRENT CONDITIONS
where Current velocity, profile, direction and duration affect ves»
sel VIM, consequently these should be specified in the design
T e Characteristic period, criteria for mooring system strength and fatigue. As discussed
VG K Current velocity. in Section 3.5, the most common categories of currents are:
(l) tidal currents (associated with astronomical tides), (2) cir-
The definition of can vary (see the Commentary Section culation currents (e.g'., the Gulf Stream, the Gulf of Mexico
CH.2.l). If is the natural period in still Water (no current), Loop Current and associated eddies, Brazil current), (3)
lock»-in can typically occur for values of 4-<K:<l0 for trans»- storrn-generated currents, and (4) internal wave generated
verse VIM. The precise range of lock-in depends on parame- soliton currents. Although spar VIM was detected in the
ters such as the structural shape, vortex mitigation devices, GOM under eddy currents and hurricane-generated inertial
appurtenances, current profile, mass ratio, and damping. current, other types of current may also induce VIM.
 Dacron ano Anatvsis or-" Srarzouxsseiuo Svsrsras son Ft_OA3'%NG Smucruass 131

H.2.2 ENVIRONMENT FOR STRENGTH ANALYSIS H.3.2 VIM RESPONSE PREDICTION

Mooring analysis for strength under the VIM condition is Model testing has been the primary tool for VIM prediction
r1orm.all.y condu.cted for an extreme current with associated because of difficulties in obtaining full.-scale response data in
wind and waves. The metocean criteria should specify current a timely fashion to support projects, and the lack of a validated
velocity, profile, and direction as well as the intensity and numerical or analytical approach. Industry studies indicate,
direction. (colinear or non-colinear) of wind and wave condi- however, that model tests are only able to accurately model
tions associated with extreme currents. The guidelines in this
certain effects while compromising others. From this point of
view, confidence in model test results and VIM design criteria
document address mainly mooring design and analysis for
should be established through comparison. with field measure-
this condition. However, recent experience suggests that con-
ment data. The reliance on model testing, the limitations of
sideration should also be given to extreme wind and waves
model testing, and limited validation with full-scale data
with associated current. When the Neptune spar experienced
should be recognized as a potential sources of uncertainty in
VIM und.er hurricane generated inertial current in the Gulf of
the design process. A more detailed discussion on model test-
Mexico, VIM was recorded only after the inertial current had ing can be found in the Commentary Section CH.4.
deepened and the wind and waves had decreased. For this
condition, the application of some of this guideline may need
H.4 Strength Design
careful evaluation. Further investigation and experience is
needed to arrive at a conclusion on this issue. H.4.1 VIM DESIGN CRITERIA
The first step in strength design is to establish suitable
H.2.3 ENVIRONMENT FOR FATIGUE ANALYSIS VIM design criteria. VIM-related design parameters for
mooring strength design include:
For long-term fatigue analysis, current events can be repre-
- In-line and transverse VIM response amplitude (A/D)
sented by a number of discrete current bins, wi.th each current
as a function of reduced velocity (I/1.).
bin consisting of a reference direction, a reference current
* Drag coefficient as a function of VIM response amplitude.
velocity and profile, associated wave and wind conditions,
and probability of occurrence. Industry studies indi.cate that ~ Definition of ranges for lock-in and lock-out.
for some mooring systems, conside;ra.bI.e fatigue damage may ~ VIM response trajectory or envelope.
be caused by a single extreme VIM fatigue event, which These criteria may be based on a combination of project
should also be addressed. Note that the current for the worst- specific model test data and previous VIM design experience,
case VIM fatigue event may not coincide with the 100-year among other aspects. Depending on the approach taken, there
return period current event, but could occur under lower will be varying levels of uncertainty in the VIM criteria spec-
return period current events. For fatigue analysis of single ified for a particular project. The recommended practice is to
VIM events, the current criteria should specify the current develop criteria for a base case (the best estimate) and some
velocity, profile, direction, and duration (build-up and decay) sensitivity cases. Tension safety factors for intact and dam-
for current events spanning a range of return periods. aged conditions should be met for the base case. Sensitivity
cases can be used to check the robustness of the mooring sys-
H.3 Vessel Response tem, i.e. the risk of mooring failure in the event estimates of
certain influential parameters such as mooring stiffness, cur-
H.3.1 VIM RESPONSE MODES rent velocity, drag coefficient, lock-in definition, or VIM
amplitude are inaccurate. One of the important roles of the
The exciting force induced. by vortex shedding on the hull sensitivity check is to detennine if, with some changes in sys-
of bluff body floaters, such as spars, may cause response in tem parameters, the system would enter a VIM lock-in
any of the 6 rigid body modes of response. The response of regime that would not be apparent for the base case design
primary concern for most floaters is the transverse (sway) criteria alone.
response and in-line (surge) response, which are typically An example of V, versus A./"D design criteria is provided in
included in a mooring analysis. However, other response Figure H.3 showing the lock-in transition, locked-in region,
modes should be checked to ensure that forces due to vortex and the locked-out region. This type of curve is typically used
shedding do not excite them or do not significantly affect the to define the VIM response amplitude. It is important to note
mooring system response. For example, for some floaters that, for most spars and other moored offshore vessels experi-
large pitch, roll, or yaw responses or large mean transverse encing VIM, the shape of this curve will likely vary with. cur-
displacements could affect the mooring system. Special rent direction, i.e., the amplitude of response varies with
hydrodynamic effects such as “galloping”, which is due to current direction for the same reduced velocity. Particular
large variations in lift versus vessel heading, are beyond the attention should be applied to defining the VIM response ver-
scope of this document. sus current heading when setting the design criteria.
132 API RECOMMENDED PRACTICE 28K

i" "' " " 1unln|||»q||nfl|nq____‘ nnun¢:_______________’********************'1nlF'" 7 up-can _______________________________%_T:l;rmu' 5 _n~»n-an-an _, ___ ________ Wirrf********"nu: nlnnuwnuuuvuunnmnnununununnnnn ________t;____""""""""""""rm " mu-on-uu____. .________7_ ***********

1 l il
. _ . .. .. ... . . . . . . . . . ..:i:i31....,. ..
- .. . . .. .. . .. . _ _ . .. . ..:::-'-

1 '-:-, ‘1

1 \ Possible locked-out

»- lI Locked-in- I iiiiiii ******** I \ I **************I

1. \
--- l
I transition
I =
1 1
A/D
ole i

1
ii

‘jgr
1

== "*”“ 1 1 .-1”" *
ii _ 1* .
, , 1
\ 11, 1 ‘
3
l -1», —- i l i . -1!.
.¢222.“-* 1 1 1 1
5 .
VMAmp
to 1
i -*- *’*’ --- i --_-__ _-_;___-_-_7_‘L **- ———————————————————— -- - _ ___ j ————————————————————— -1-__-__--_______ i —, -7———————- -7—————————.——.;.:- 1: - W’; 1— -- -- . . i i ’— _____ l. 1-

I1 1
I
l ‘ 3

1
1

O '=;:,I':_.. i i _ . . 1 . . l

Reduced Velocity (Vr)

Figure H.3—Examp|e VIM Amplitude versus Reduced Velocity

H.4.2 STRENGTH ANALYSIS METHOD = Mooring systems experiencing a high mean load and
large tension variation may stress the chain to beyond
Most mooring analysis software was not design.ed to
handle spar VIM analysis, therefore a simplified. analysis the elastic region, vvhere fatigue test data are not avail-
procedure (Cornrnentary Section CH5) is typically used. able. To ensure sufficient fatigue life, the mooring sys-
More accurate approaches can be considered when tern should be designed to avoid this situation.
advanced mooring analysis software that can model VIM ' Fatigue damage of chain at the fairlead requires special
responses in addition to other typical loads and motions attention since additional bending stress is imposed on
becomes available. the chain in this region, and chain typically has the low-
est fatigue life of all the components in the mooring
H.5 Fatigue Design system.
H.5.1 BASIC CONSIDERATIONS
' Sensitivity cases, similar to those used in the strength
Mooring tensions due to VIM are cyclic by nature and con- analysis, may be considered to account for uncertainty
tribute to the fatigue damage of the mooring system. The fol- in the VIM prediction.
lowing factors should be considered when assessing fatigue
due to VIM: H.5.2 LONG TERM FATIGUE DAMAGE
~ The VIM period in the offset position, corresponding to For long-term fatigue analysis under the VIM condition,
the specific current bin under consideration, should be
current events can be represented by a number of discrete
used when calculating the number of tension cycles.
current bins, with each current bin consisting of a reference
This period can vary with current direction and magni-
direction, a reference current velocity and profile, associated
tude and can be different than the still water natural
wave and wind conditions, and probability of occurrence.
period.
Fatigue damage for each current bin is evaluated, and the sum
~ In addition to a long-term fatigue damage evaluation, a of fatigue damage due to VIM from all current bins is added
fatigue analysis for the 100-year VIM fatigue event (or to the fatigue damage due to wind and vvaves to yield total
other single yvorst-case event as noted in Section H.5.3) fatigue damage (see Commentary Section CI—I.6 for more
is also recommended. detailed procedure).
 Dacron AND ANfi=tt.‘{SiS or Srariouaaaeine Srsraivis Foe Ftoariuo Sraucruaas 133

H.5.3 FATIGUE ANALYSIS FOR SINGLE both wind and waves and VIM) and single event fatigue
EXTREME VIM EVENT under the intact condition.
Ind.ustry studies indicate that considerable fatigue damage
H.5.5 CHAIN FATIGUE AND WEAR
can be caused by a single extreme VIM event. Therefore in
addition to the long-term fatigue damage evaluation, a fatigue Fatigue damage of chain at the fairlead is typically higher
analysis for the I00-year VIM fatigue event is also recom- than that away from the fair.lead. For mooring systems where
mended. Since VIM response is largel.y dependent on reduced chain fatigue is critical, it is important to shift periodically the
velocity, the current for worst-case VIM may not coincide links at the fairlead so additional fatigue damage due to bend-
with the 100-year return period loop or hurricane current. The ing can be more evenly distributed. If this operational proce-
highest VIM amplitudes for fatigue consideration could occur dure will be carried out in the field, fatigue damage for the
under a lower return period current. The current direction links around the fairlead can be evaluated based on percent of
should be the worst direction identified in the strength analy- time when the links are located at the fairlead. However, the
sis. Instead of using a constant extreme design current for the links at fairlead should have sufficient fatigue strength to sur-
whole event, current variation. based on field measurements vive at least a single extreme VIM event (for example the
for strong loop currents can be considered. Note the duration 100-year VIM event).
of this event may be different from that obtained from the Mooring systems subjected to VIM may also experience
long term current distribution. increased wear in the links at the fairlead, which is caused by
Fatigue analysis is typically performed for the intact condi- high contact pressure and large movement between links.
tion only. However, a fatigue analysis for the damaged condi- This issue should be carefully evaluated, and the measure of
tion should be considered for the single extreme VIM event periodically shifting the links at the fairlead should be consid-
when progressive line failure due to fatigue is a concern. cred. to alleviate the wear problem. Wear measurement using
go-no-go gage as outlined in API RP 21 can also be consid»
H.5.4 FACTOR OF SAFETY ered for detecting excessive chain link wear.
As stated in Section 7.5, the factor of safety for fatigue
design is 3.0, which should be applied to long term (due to
 APPENDIX CH--COMMENTARY ON APPENDIX H---
MOORING DESIGN FOR VORTEX INDUCED MOTION (VIM)

CH.1 Basic Considerations Because of the above issues, it is important to address VIM
conservatively in the spar mooring design stage. This can be
CH.1.1 PURPOSE OF COMMENTARY achieved. through the following measures:
Many floaters consisting of cylindrical structures such as l. Establish design criteria that recognize the uncertain-
spars, TLPs, and semi-submersibles can be susceptible to ties in VIM behavior, for example checking sensitivity
vortex induced motion (VIM) when exposed to currents. cases in addition to the base case and checking field
VIM is most prominent for spar platforms where most of measurement data as well as model test data.
industry experience has been acquired. Appendix H provides
some high level design guidelines for spar mooring systems 2. Conduct fatigue analysis for the 100-year VIM
response condition in addition to long term fatigue
under VIM conditions. This Commentary, which is not con-
analysis.
sidered to be part of the recommended practice, provides
additional background information describing the developing 3. Select mooring hardware and system design character»
state of the art to supplement Appendix H. istics (strake configuration, mooring stiffness, etc.) that
There are a number of special issues for spar VIM, as dis- can better tolerate or mitigate VIM.
cussed in the section below. The industry has devoted exten-
sive effort to address these issues, leading to significant CH.2 VIM Design Parameters
improvements in the design procedure. Although improved
CH.2.1 REDUCED VELOCITY VR
design procedures have been used in many spar design
projects, there are areas where industry consensus has not For the 6-degree of freedom motions of multi-member
been completely reached. There are also areas that require bodies the definition of reduced velocity is complicated. In
further improvement .in the future. Designers are encouraged general, the definition of V, involves the eigen-periods of the
to exercise judgment and caution in applying the information system under mean load (which depend on the nonlinear
in this Commentary to their designs and to study the refer- mooring system and hydrostatic stiffness, and the full 6-by-6
ences listed in Section CH.9. Results of new studies should mass and added mass matrices), the characteristic dimension
be considered when they become available. of the body (which may vary with the eigen-mode under con-
sideration), and the characteristic velocity of the flow inci-
CH.1.2 SPECIAL ISSUES FOR SPAR VIM dent on the member.
The following discussion applies to translational VIM of a
1. There is no mature analytical tool for the prediction of classic spar transverse to the current direction. In this case,
spar VIM. Currently VIM design criteria are typically transverse VIM occurs when the vortex shedding period
obtained from model testing. Model testing practices (Strouhal period) is close to the natural period of the moored
need to be validated with field measurement data, system transverse to the current direction. For a classic spar
which are quite limited. the relation between VIM response and the natural or
2. Spar VIM is affected by current velocity, direction, observed period of the transverse motion is often given in
profile, hull geometry and appurtenances. terms of the reduced velocities V}.,., or Viobs,

3. The duration for peak current and resulting VIM can

Vrn. x I/cTn/D and mobs 3 I/cT0bs/D (CH-1)
be much longer than peak storm duration.
4. Model tests can only model certain parameters while where
approximating others. Hence care needs to be exer-
cised in the interpretation and use of model test data. V6 I the characteristic current velocity, typically the
highest velocity in the current profile,
5. Where VIM results in large tension cycles at high
mean load, fatigue life can be short for mooring com- T,1 K the still water natural period of the moored ves-
ponents with low fatigue resistance such as chain. sel transverse to the current direction under
6. The calibration of the factors of safety for mooring mean load,
design does not include the spar VIM condition and the TObs K the observed period of VIM,
uncertainties associated with spar VIM. Consequently
sensitivity checks may be warranted. D K spar diameter.
136 API RECOMMENDED PRACTICE 28K

Note that Vmbs is only defined over the range of current of the parameters as discussed above for transverse VIM. The
velocities that induce VIM, whereas I/,..,., is defined for all cur- magnitude of inline A/D is typically much less than the trans-
rent velocities. Model test data indicate that VIM is a func- verse A/D. Field measurement data for a classic spar with an
tion of M.,, or V,,,,;,,, and VIM is negligible when V”, is below a equally spaced spread mooring system indicate inline A/D of
threshold value. For transverse VIM of classic spars a thresh- 10% to 15% of the transverse A/D [CH.20]. However, the
old value of approximately 4 has been observed in the model magnitude can be higher if the natural period for the inline
tests. Field data for classic spars support the finding from motion is about half of the natural period for the transverse
model tests, but no field data are available to date to validate motion (resonance condition). Also, unsymmetrical mooring
this for truss spars. The range of I/},, or I/Q05». where signifi- system stiffness could result in a VIM trajectory for which the
cant VIM can be induced is often referred to as the “lock-in” major axis of the VIM is not transverse to the current direc-
range. tron.
Via, is a function of T,.,, which is a. function of mooring
stiffness and vessel mass. Mooring stiffness at various offsets CH.2.4 DRAG COEFFICIENT
can be significantly different, especially for grouped moor-
Model tests are typically used to detcrrnine drag coefficient
ing patterns. Vessel mass includes added mass, which is typi-
Cd to be used in design. A “base drag” Cdg is assumed for the
cally determined by analytical tools or model testing. If
case where A/D 1 0.0 (no VIM) and amplification factors are
available, field measurement data should be used to calibrate
applied to account for VIM effects. The drag augmentation is
the added mass values. The transverse stiffness used for cal-
a function of A/D and K. and can be expressed as [CH.3,
culating Vm is typically evaluated at the mean offset under
CH.24, CH.29]:
current and associated wind and waves. Since the mean off-
set is dependent on the drag force, which is dependent on the
VIM amplitude, the process of selecting the appropriate off- Cd 2 Cd0[l +k(A/D, I/}.)] (CH.2)
set for VIM calculation is iterative. It should be noted that
the observed period from model tests or field measurements The mean drag force on the cylinder is given by
T05, can be different from the calculated still water natural
period, T,,, which is used in most analyses as it is readily F, = 0,, 5l p Vi/1, (ct-1.3)
available. Calibration of calculated values with available
model test or field measurement data may be desirable when where
such data are available.
Cd I mean drag coefficient (absolute current veloc»
Since En is a function of current velocity and natural
period of the moored vessel, VIM can be generated under rel- ity) in the presence of VIM,
atively mild currents (for example 1 to 2 knots) if the natural p K density of the fluid,
period. of the vessel is long. This may be the case wi.th deep-
water floating systems that have lovv mooring stiffness. V6 e free stream current velocity,
AP K projected area.
CH.2.2 TRANSVERSE (CROSS FLOW) VIM
In the lock-in range the drag coefficient increases almost
Transverse VIM occurs when the vortex shedding period is
linearly with A/D. For a classic spar where the spar diameter
close to the transverse natural period of the moored vessel,
is well defined, the drag coefficient under lock-in condition
and the spar typically oscillates in the direction perpendicular
can be expressed by the following equation:
to the current in a periodic pattern. Transverse motion is nor-
mally expressed as the ratio of single amplitude transverse
Cd 2 Cdo + f(A/D) (CH4)
VIM to spar diameter (A/D). However, transverse VIM some»-
times has an asymmetric pattern. In this case transverse A/D
should be specified for two opposite directions. Transverse Where Cd is the spar drag coefficient with VIM, and is spar
VIM is a function of a large number of parameters such as drag coefficient without VIM. The coefficient f is hull spe-
reduced velocity, spar type (classic or truss), strake configura- cific, and is normally determined by model testing. It also
tion (shape, height and coverage), current characteristics depends on the definition ofA/D and Cd (extreme or mean A/
D, and absolute or relative velocity Cd). Some of the refer-
(profile, speed, and direction), and hull appurtenances (anode,
ence publications demonstrate the variability to drag
chain, fairlead, and pipe), etc.
observed in model tests. Such variability may warrant sensi-
tivity checks on drag predictions as part of the mooring
CH.2.3 INLINE VIM
design.
lnline VIM is typically in the direction of the current, and it For truss spars which consist of a large number of compo»
may affect the transverse VIM. Inline A/D is also a function nents of various diameters, D is defined as the diameter of the
 Dssion AND Anarvsis or STATiONKEEP¥NG Srsrsras roe i;t.OA'¥'iNG Srsccruaas 137

hard tank. Attention should be given to the possibility of - Mooring stiffness characteristics
increased drag on the small truss members d.ue to hull VIM. - Degrees of freedom
* Current direction and profile
CH.3 Effects of Water Depth and Current
Turbulence ' Directional resolution
- Test rig damping
Generally VIM magnitude is not a function of water depth,
but VIM and mooring line tension can be affected. by change ~ Blockage (wall) effect
of stiffness i.n different water depths. Mooring stiffness typi- ~ Length of response record
cally increases with decreasing Water depth. The higher
Industry has devoted significant effort to improve model
mooring stiffness in shallower water may reduce or even sup-
testing methodology to obtain better predictions for spar VIM
press VIM under certain conditions, due to a resulting value
responses. Improved model testing conducted recently
of V, less than the lock-in threshold. However, if higher stiff-
yielded VIM predictions that compare reasonably well with
ness fails to reduce or suppress VIM, the mooring line can
field measurements [CI~I.20]. However, all the model tests
experience a significant increase in line tension. Industry
conducted to date can only accurately model certain parame-
experience indicates that VIM can cause significant line ten»-
ters vvhile approximating others. Different model testing
sion increase for typical steel taut leg moorings in water
methodologies and practices can result in different test
depths of 2,000 to 3,000 ft, where VIM amplitudes may be a
results. From this point of view, confidence in model test
significant fraction of the total offsets. The VIM influence on
results and VIM design criteria should be established through
line tension is much smaller in deeper water, say 5,000 ft
adherence to sound engineering principles and comparison
water depth, because mooring stiffness generally decreases,
with field measurements where available. The reliance on
while VIM amplitude remains similar in magnitude regard»
model testing, the limitations of model testing and limited
less of water depth. Although VIM of the same magnitude is
validation with full-scale data should be recognized as poten-
less damaging to deepwater moorings, the effects of VIM on
tial sources of uncertainty in the design process.
these moorings still need to be considered. Large seafloor
slope may result in significantly different anchor depths for
CH.4.2 MODEL TEST PARAMETERS
different mooring lines, causing directional change of moor-
ing stiffness. This in turn may induce directional VIM CH.4.2.1 Flow Similitude
response.
While. the correlation between the limited available field The basis for hydrodynamic model testing is that geomet-
measurements of spar VIM and model test results does not ric and dynamic similitude between prototype and model
indicate that turbulence in ocean currents influences spar fluid flovv is preserved. Reynolds i}1.1I1’}lI)€l' scaling and Froiide
VIM response, there is evidence from model testing that high number scaling are the two relevant scaling parameters for
levels of turbulence in the model basin can affect VIM hydrodynamic model testing of offshore structures [CH.22,
response. The structure an.d intensity of turbulence in ocean Chapter 9].
currents and the potential impact of current turbulence on The Reynolds number is defined as
VIM remains an uncertainty for further observation and V.D
investigation. R, I (CH5)

where
CH.4 Model Testing Re = Reynolds number,
CH.4.1 BASIC CONSIDERATIONS VG K characteristic velocity (e.g., flow velocity),
Model tests are routinely conducted. to investigate spar D e characteristic length (e.g., hull diameter),
VIM and VIM mitigation methods. Model testing has been
v Z kinematic viscosity of the fluid, v = 1.21 x I05’
the primary tool for spar VIM prediction because of the diffi-
culties in obtaining full-scale measurement data and the lack fig/sec for seawater.
of a sufficiently mature numerical approach. A sound VIM The Froude Number is defined as
model testing practice should pay attention to the following
issues: V.
F,, I ---‘-=- (CH.6)
- Geometric scaling A/gD

' Dynamic scaling where

- Hydrodynamic scaling F}, = Froude number,
* Modeling of appurtenances g = gravity acceleration, 32.2fi/secg.
138 API RECOMMENDED Psacnca 28K

Satisfying the Reynolds and Froude scaling simultaneously ated with the vessel’s rigid body modes, mass ratio, and
for the model and prototype flows, however, is practically reduced velocity.
impossible. For a model dimension D that is substantially Modeling all of the rigid‘-body modes (e.g. surge, sway,
smaller than prototype, either the gravity g needs to be signif- heave, roll, pitch, and yaw) may not be important as long as
icantly increased, or viscosity v of the testing fluid needs to those that are candidates for lock-in are included. For exam-
be significantly decreased. Neither of these changes is practi- ple, a spar niight lock-in to sway at lower velocities and roll
cal in a model basin. at higher velocities [CH.ll]. The two degrees of freedom
For separated flow dynamics that cause VIM, Reynolds might actually couple (lock-in simultaneously). In this case it
number scaling is the governing scaling law. Reynolds scal- is important that the sway and roll modes and periods be
ing is particularly difficult to achieve for an offshore floating properly scaled. On the other hand, if the transverse sway is
structure. For spar hull diameters of 70-140 ft and current the dominant VIM response, then tests with a single~degree
velocities of 2-~5 knots, the Reynolds number for the full of freedom rigid body .mode have shown reasonable agree»
scale structures (prototype) are in a range of 20,000,000 to ment between model test and full scale data [CH.2l].
100,000,000. To match such Reynolds numbers in the model The concept of reduced velocity V}. has been introduced in
basin requires that the model experience the same hydrody- Section H.l. It is an important dimensionless parameter for
namic force as that of the prototype, which is obviously VIM:
impractical. There are currently two basic testing approaches,
supercritical and subcritical Reynolds number, used in the
industry: . VT
D
V=—“- (CH1)
I.
l. Supercritical Reynolds number model testing. It is
important to test at supercritical. Reynolds numbers in The definition of characteristic period T can vary (see Sec-
order to attain a flow regime which is similar to the tion CH.2). If T is defined as the still water system natural
flow experienced in full scale [CH.20, CH.2l, CH.33]. period, VIM ‘lock-in’ for a classic spar typically occurs for
Supercritical model tests conducted at Reynolds num- values of 43 K5] 0. The reduced velocity for model flow must
bers of between 600,000 and 2,000,000 for Genesis correspond to the reduced velocity for the prototype flow in
and Hoover classic spars have shown good agreement order to achieve proper fluid-structure VIM similarity. That
with full scale (l5,000,000 < Re < 40,000,000) is, in addition to selections of proper scaling for VG and D,
responses measured in the field. However, supercritical scaling for period T should also be appropriate.
Reynolds number model testing places significant Mass ratio has a large effect on the range of lock-in, and
demand on the capacity of the model basin and to date possibly the amplitude [CI-l.l0, CH.23, CH.30]. Mass ratio
supercritical model tests have only modeled ll-DOF for a free floating body is by definition equal to 1.0 (displace-
and with a uniform current profile. ment : weight). This mass ratio should. be maintained for
2. Subcritical Reynolds number model testing. For a heli- model tests as well.
cally straked cylinder, in the near field flow separation
is controlled by the sharp edges of the strakes and not CH.4.2.3 Geometric Similitude
by boundary layer effects [CH.4]. In addition, it is pos-
Geometric similitude is achieved when the geometry of the
sible to include the effects of 6-DOF spar motions and
model and prototype bodies is similar. The geometry of the
current profile in the model test. Subcritical model tests
hull and strakes (if appropriate) should be accurately scaled.
conducted at Reynolds numbers of between 50,000
This includes any construction openings in the strakes, brack-
and 400,000 for the Genesis spar were conservative
ets (which might affect the flow along the strakes), chains,
when compared to full scale (30,000,000 < Re <
anodes, external pipes and other appurtenances that may
40,000,000) measurements [CH. 17, CH.20].
affect the flows around the body. Some members, e.g., the
As stated above, all model tests conducted to date can only truss members of a truss spar, may result in viscous damping
accurately model certain parameters while approximating effects that are Reynolds number dependent. Care should be
others. Both approaches have advantages and disadvantages, exercised to size these members to result in a representative
and both show acceptable agreement with full scale data for amount of damping in model tests.
classic spars [CH. l7, CH.20, CH.2l, CH33].
CH.4.2.3.1 Model Scale
CH.4.2.2 Dynamic Similitude
The model has to be geometrically similar to the prototype,
In addition to hydrodynamic scaling, dynamic similitude meaning that the shape of the model is the same as that of the
requires that the rigid body dynamics for the full scale and prototype while the characteristic length of the model is
model scale systems are similar. Dynamic scaling is associ- smaller. Due to hydrodynamic force consideration, the indus-
 Dssiou AND Auatvsrs or Srariounsseiuo $Y8“i"Eiv‘i$ son Ftoarrno Srnucrusas 139

try typically uses smaller (1/100 ratio) model scale for high, CH.4.2.3.5 Current Direction and Profile
supercritical, Reynolds number model testing and relatively
VIM response for spar hull can be sensitive to small.
larger (1/50 ratio) model scale for low, subcritical, Reynolds
changes in current directions. Fine heading resolutions (e.g.,
number model testing.
at 10 to 15-degree increment) may be required to capture the
maximum VIM response.
CH.4.2.3.2 Appurtenances Tow tests simulate a slab current uniform with depth. In
It is important to model all details of the spar hull accu- reality design currents have a profile and current speeds gen-
erally decrease with depth. Efforts have been made to simu-
rately. This includes all appurtenances such as fairlead, pipes,
late shear current profiles in tow, flume and basin tests
chains, anodes, risers and flowlines. Details of strakes includ-
[CH27]. Any attempt to generate shear current profiles in
ing cutouts or holes in strakes should also be modeled cor-
model scale generates excessive turbulence. Careful consid-
rectly. Accurate modeling of appurtenances is particularly
eration needs to be given while interpreting VIM responses in
important in developing VIM directional sensitivity and test- the presence of turbulent flow. Turbulence in laboratory gen-
ing effectiveness of VIM suppression devices such as strakes. erated shear flow can be mitigated by using varying density/
viscosity stratified liquid layers in the model tests.
CH.4.2.3.3 Model Degrees of Freedom
CH.4.2.3.6 Free Surface Effect
Models of single degree and multiple degrees offreedom
have been used. For the single degree of freedom model, The free surface acts as a barrier through which flow can-
which is mainly used in high (supercritical) Reynolds number not pass, and as a means to dissipate energy through wave
testing, only transverse VIM is allowed. For the multiple radiation. Free surface effects can be important when the
degrees of freedom model, which is mainly used in low (sub- Froude number is greater than 0.15. For surface-piercing tow-
critical) Reynolds number tests, the vessel is free to respond ing test of a spar hull model, the towing speed is limited by
in all six degrees of freedom. The relative importance of mul~ wave resistance (Froude number). High speed towing might
tiple degrees of freedom model is determined by the level of result in Froude number that far exceeds the field Froude
coupling between motions of different degrees of freedom. number and exaggerates the free surface effects. One way to
avoid the excessive wave resistance for high Reynolds
ber model testing is to tow a completely submerged, horizon-
CH.4.2.3.4 Mooring Stiffness Characteristics
tally mounted mirror "image of double body with a divider
There are two approaches to modeling the stiffness distri- plate in the center. The divider plate is used to prevent flow
bution of the prototype mooring system. One approach is to communication across the divider plate.
use the reduced velocity (I/1.) as a design parameter. In this
case, the VIM response in the model is related to the design CH.4.2.3.7 Damping
A/D via the reduced velocity. In the model tests, the spar Damping can affect VIM response, therefore the damping
response is measured at different reduced velocities. In the (hydrodynamic and mechanical) generated in the model basin
design phase, the transverse period of the spar (hence the should be consistent with the damping expected in the field.
K.) is calculated at different offsets. At each offset, the A/D Since mechanical damping may be generated by the testing
is based on the V, at that location. In this approach, a linear equipment and is absent in the field, care needs to be taken to
symmetric mooring system can be used for the model test understand the effect of damping on the VIM response and to
set-up. mitigate such effects [CH.33]. Hydrodynamic damping, due
Alternately, the actual spread mooring of the spar is mod» to mooring lines and wave effects, in the model test should be
eled. In this case the current speed is the design parameter given careful consideration when estimating the amplitude of
rather than the reduced velocity. A spar has typically three or full scale VIM.
four groups of mooring lines. Each mooring line or group of
prototype mooring lines is modeled by an equivalent model CH.4.2.4 Length of Response Record
mooring line. The horizontal force-displacement characteris- Sufficiently long response time histories are required to
tic of each mooring line or group is modeled by a bi- or tri-~ provide meaningful statistics such as standard deviation, sig-
linear spring system so as to mimic the nonlinear force»~dis- nificant, and maximum values. The length of time history
placement characteristic of each mooring line or group. This required depends on the periodicity of the VIM response
allows for modeling of the complete nonlinearity and asyrn» [CH.2l]. Where the VIM motion is well developed and sus~
mctry of the stiffness. For some mooring systems such as tained (e.g., fully locked-in), relatively few cycles are
grouped mooring system, the asymmetry may contribute to required to establish the maximum VIM amplitude. If the
highly directional VIM response. VIM response is modulated (e.g., in the lock-in and lock-out
140 API RECOMMENDED Pnncncr.-: 28K

transition regions), long records are required to establish sta- in the model test (uniform) and in the field (may have been
tistical values. While these regions don’t produce as large of a non-uniform).
VIM response, they could be important for computing moor- A benefit of this approach is that motions in all 6 degrees-
ing line fatigue. Hence sufficient time record lengths should of-freedom can be modeled. This allows for responses in the
be obtained. Note that the start up transient response should roll and pitch degrees of freedom to be identified. and incor-
be excluded from the record for statistical analysis. porated in the design. It also allows for the hydrodynamic
coupling effects between the different degrees of freedom.
CH.4.3 CURRENT INDUSTRY PRACTICES The ability to use larger models also facilitates more detailed
As mentioned above, there are two practices prevalent in modeling of the hull details and appurtenances. The vertical
the industry today for spar VIM model tests, mainly centered moored. set-up also gives the ability to model the spatial vari-
on testing at either supercritical or subcritical Reynolds num- ation (nonlinearity and asymmetry) of the prototype mooring
bers. The former tests are performed using a horizontal, sub- system. One additional benefit is that such approach has
merged cylinder in a high speed towing tank [CI-L20, CH.2l, much less requirement for model basin capacity and can be
CH.33], the latter being performed on a floating, surface- carried out in model basins without high-speed tow capabil-
piercing vertical cylinder with external spring lines simulat- ity. Examples of the low, subcritical Reynolds number model
ing the mooring system [CH.2, CH.ll, CH.l6, CH. 17, testing of spars can be found in [CH.ll, CH.l7, CH.20,
CH. 19, CH.27, CH.28]. The former approach has so far been CH.27, CH.28].
limited to classic spars.
Model tests have not been performed for all spars. VIM CH.4.4 FIELD MEASUREMENT DATA COMPARED
response itself is self-limiting, and if a bounding analysis WITH MODEL TEST DATA
indicates that the mooring system will not be governed by
Field measurements of VIM response have been recorded
high current or VIM responses then VIM tests have not been
for 3 classic spars---Genesis, Hoover, and Neptune [CH.20,
performed [CH. l 9].
CH.2l, CH33]. Note that in the field the current profiles vary
in speed and heading with depth, as opposed to the slab cur-
CH.4.3.1 High, Supercritical, Reynolds Number rent described in the tow tests earlier. Hence, adjustments to
Modei Testing the model test values may be required to account for such
In this approach, model tests are conducted for model Rey- variations. Although motion measurement systems have been
nolds numbers in the supercritical range beyond Re = installed on a number of truss spars, no significant VIM
600,000. The basis for testing the hull model at the supercriti- responses have been recorded [CH9].
cal R8 regime is the assumption that, once beyond the transi- Of particular interest is the Genesis spar, for which field
tion range, model and prototype flow similitude is preserved. measurement data, supercritical and subcritical Reynolds
Mod.el tests at supercritical Reynolds number for classic spar number test data are available ICH. 17, CH.20, CH.33].
VIM show relatively good agreement with field measure-
ments [CH.2l, CH.33]. CH.5 Strength Analysis Procedure
Note that high Reynolds number model testing places sig-
nificant demand on the capacity of the model basin and is Mooring analysis for high current/VIM conditions may
available only at a few test facilities worldwide. An example require special computer software that is capable of modeling
of the high, supercritical model testing of a classic spar can be VIM in time domain. A simplified analysis procedure can be
found in [CH.2l, CH.33]. The described rig has been used to used if the waves associated with the current are mild, result-
tow the spar hull model at up to R8 12,000,000. ing in low mooring line dynamics. Once the VIM design cri-
teria are established, the following simplified analysis
CH.4-.3.2 Low Reynolds Number Model Testing procedure can be used. An example of strength analysis can
be found in [CH.l6].
In this approach, the model is either towed at low speed or
in-place tested in a flume or wave basin with current generat- l. Select a current direction
ing capability. Froude scaling is not explicitly required. How- 2. Determine the mean vessel offset under the design cur-
ever, the Froude number is typically chosen so that it is less rent with associated wind and waves based on an
than that of the full scale. The model test Reynolds number is
estimated Cd. Spar set-down should be considered to
typically in the subcritical range. Model tests at low Reynolds
yield realistic results.
numbers for a classic spar has shown conservative results
compared with field measurements [CI~I.20]. The conserva- 3. Calculate in-line and transverse VIM and Cd based on
tism may be due to the difference between the current profiles the design criteria established according to Section
 i3ESiG~N auo Analysts or Srarronxeanavo Svsrar/is son FLOA”i"it\i(3 Srnucrunas Mt

H.4. If this Cd value is significantly different from the 5. Determine in-line and transverse VIM amplitude coef-
estimated Cd in Step 2, iteration may be required. ficient C,,, which is a function of reduced velocity, and
is equal to 1.0 at peak VIM under lock-in condition.
4. Determine the envelope of possible maximum vessel
offsets including the effects of current/vvind/*.vave ves- 6. Calculate the reduced velocity for the current bin and
sel offsets (Step 2), and in-line and transverse VIM further modify the mean in-line and transverse A/D
(Step 3). (Step 4) by C,,.

5. Determine line tensions and anchor loads correspond- 7. Determine drag coefficient Cd for the current bin based
ing to the envelope of possible maximum vessel offsets on the modified mean transverse A/D (Step 6). If this
calculated in Step 4 by static mooring analysis. Cg value is significantly different from the estimated
Cd in Step 3, iteration may be required.
6. Evaluate additional line tensions and anchor loads due
to line dynamics, which is superimposed to the quasi- 8 Perform VIM mooring analysis based on the modified
static values obtained in Step 5. mean in-line and transverse A/D (Step 6), and Cd (Step
7), using the procedure for strength design. Determine
7. Repeat Steps l-6 to obtain line tensions and anchor average tension ranges R,-, and corresponding average
loads for other current directions. response period T} from the time trace of line tensions
8. Identify the worst direction for design check. for a fevv VIM cycles. Note the average response
period T} may vary due to the relative orientation of the
mooring line and current.
CH.6 Analysis Procedure for Long Term
Fatigue Damage 9 Determine number of cycles to failure N, correspond;-
ing to R, for the mooring component of interest using
The recommended procedure for long term fatigue damage
an appropriate T-N equation. Chain usually has the
evaluation is provided below. An example of fatigue analysis
shortest fatigue life, and chain fatigue life at fairlead is
can be found in [CI-I. 16].
further reduced because of additional stress concentra-
ll. The long-term current events can be represented by a tion from bending. Stress concentration factor
number of discrete current bins. Each current bin con- accounting for bending at fairlead should be deter-
sists of a reference direction and a reference current mined by testing or finite element analysis. A factorjg,
velocity with associated wave and wind conditions. which is defined as ratio of chain stress concentration
The probability of occurrence of each current bin must factor at fairlead to that avvay from fairlead, can be
be specified. The number of reference directions used for calculating fatigue life of chain links at fair-
depends on the directionality of the current at the site, lead. The factorff, can vary significantly depending on
and the specified directions should include those for the number of fairlead pocket and fit between chain
which significant VIM is predicted. The required num- and fairlead. This factor can be as low as 1.2 for a 7-
ber of reference current velocities normally falls in a pocket tight fit fairlead, but it can be higher for a loose
range of l0 to 50. Fatigue damage prediction can be fit fairlead. The value of N; is reduced by a factor of
fairly sensitive to this number for certain mooring sys- j§,.M at the fairlead, vvhere M is the slope of the T-N
tems, and therefore it is best determined by a equation.
sensitivity study.
l0. Calculate annual fatigue damage for the ith current
2. Select a current bin and calculate the duration ti for the bin:
current bin in a year based on the probability of occur-
rence for the current velocity and direction.
D;:(I;/Ti)/N;

3. Determine the natural period Tn of the moored spar

under the current bin without VIM based on an esti- ll. Repeat Steps 2 to 10 for other current bins
mated Cd.
l2. Determine cumulative fatigue damage for all current
4. Specify extreme in-line and transverse A/D values for bins, which is combined with the fatigue damage from
the current bin based on available model test or field wind and Waves to obtain total fatigue damage D,-
measurement data. The mean A/D for fatigue analysis (Refer to Section 6.3.2 for methods to combine fatigue
can be evaluated by multiplying the extreme A/D by a damage). The predicted fatigue life is l /’ D; (years),
coefficient g, which should be determined by available which should be greater than the service life times a
model test or field measurement data. factor of safety.
142 API Rscouusunso PRACTICE 28K

CH.7 Methods to Improve Mooring or reduce the adverse impact of these appurtenances may
Design for VIM reduce VIM. A rigorous model test program is required to
evaluate the effectiveness of these measures.
CH.7.1 POLYESTER ROPE FOR MIDDLE SECTION
Spiral strand wire ropes are commonly placed at the middle CH.7.6 TIGHTENED MOORING
section of mooring l.ines for spars. The use of polyester ropes
in this section may sometimes reduce the line tensions due to VIM is not observed in the model basin when V} is below a
VIM because the lower rope stiffness makes the polyester threshold value. This condition can be achieved in some cases
mooring more compliant for large vessel movements. Also by tightening the mooring system, for example using higher
the use of polyester rope may reduce V}, which in turn may initial tensions, or tightening the mooring system in advance
prevent lock-in. Also tension variati.on. due to dynamic loads of high current events, thus reducing the natural period. of the
on the floating vessel can be lower for polyester mooring. moored vessel, and eliminating VIM for current speeds
This will result in lower fatigue damage to all mooring com- below the maximum design value. The adoption of this mea-
ponents including chain, which has the lowest fatigue resis- sure should be based on rigorous model testing and analysis,
tance. A sensitivity study investigating the effects of using and on addressing sensitivity to higher current and lower
polyester ropes instead of spiral. strands can be found in threshold IQ. An operational procedure to ensure a tight
[CH.l6]. mooring during high current events should also be developed
[CH. l 8].
CH.7.2 SPIRAL STRAND FOR TOP AND BOTTOM
SECTIONS CH.7.7 SOFTENED MOORING

Chains are commonly placed at the top (vessel) and bottom Softened mooring lines may significantly reduce the line
(anchor) sections of mooring lines for spars. The use of spiral tensions due to VIM because the lower mooring stiffness
strand in these sections may significantly reduce fatigue dam- makes the mooring system more compliant for large vessel
age due to VIM because spiral strand has much higher fatigue movements. A mooring system. can be softened by different
resistance than chain. This option requires significant hard.- methods:
ware modification, which includes replacing chain jack and 1. Use catenary mooring instead of taut leg mooring.
chain fai.rlead with linear winch and bending shoe. The indus-
try has good experience with mooring systems using spiral 2. Increase the length of the bottom chain segment in a
strand, linear winch, and bending shoe. taut leg mooring.
3. Slacken all mooring lines of ea taut leg mooring.
CH.7.3 IMPROVED CHAIN FAIRLEAD
4. Slacken the leeward mooring lines of a taut leg
The chain section in contact with the fairlead is more sus-
mooring.
ceptible to fatigue failure because of the presence of bending
load in addition to tension. Chain fairleads with 7 pockets are For methods l to 3, the risers must be able to tolerate the
commonly used for spar moorings. The use of chain fairleads larger vessel offset resulting from lower mooring stiffness.
with 9 pockets can reduce chain bending, thus reducing chain Although Method 3 will reduce line tension for the same
fatigue damage in this section. Also chain fairlead design VIM, it may cause earlier initiation of VIM due to higher
resulting in a tight fit between the chain and the fairlead reduced velocity at lower current velocities. Method 4 may
pocket can yield a much lower stress concentration factor and require a sophisticated mooring operational procedure
longer fatigue life. Alternatively bending shoes that yield low because of the possibility of changing current velocity and
stress concentration in chain can be used. direction. Implementation of this method requires careful
investigation and planning.
CH.7.4 STRAKE DESIGN
VIM can be reduced by improved strake design. Options CH.8 Future Technology Development
include improving strake shape, increasing strake height, and CH.8.1 MODEL TESTING AND FIELD
eliminating discontinuities and holes in strakes. To evaluate MEASUREMENT
the effectiveness of these options, a rigorous model test pro-
gram is required. To minimize uncertainties in VIM prediction and provide
data for calibrating analytical tools, advancement in model
CH.7.5 HULL APPURTENANCES testing technology and acquisition of field data are essential.
Currently the focus is on measuring the motion and line ten-
Hull appurtenances such as anodes, chain, fairleads, and sion responses. However, more attention should be given to
pipes may affect spar VIM response. Measures to eliminate the measurement of the exciting forces in the future, since
 Dacron ans Auarvsis or Srarronxssszno Svsrsrvis son Froaravo Srnucrusss M3

this may lead to development of reliable analytical tools for dure is often used. To improve accuracy, the industry needs
spar VIM prediction. advanced mooring analysis software that can model current,
inline and transverse VIM in addition to other typical loads
CH.8.2 PEAK VALUE STATISTICS and motions.
Currently mooring strength and fatigue design are typically
based on criteria established for extreme VIM values. This is CH.9 References
different from the traditional approach that is based on stan- CH.1 Allen, D. W., and Henning, D. L., “Surface Rough-
dard deviation and peak value statistics, which is in turn a ness Effects on Vortex-Induced Vibration of Cylindrical
function of the duration of the extreme environmental event. Structures at Critical and Supercritical Reynolds Numbers,”
The traditional approach is not used here because the peak OTC 13302, 2001.
value statistics have not been well established for transverse
and inline VIM, and the duration for the extreme environ- CH.2 Bangs, A.S., Miettinen, J. A., Mikkola, T. P. J-, Sil-
mental event, for example the 100-year current, is difficult to vola, I, and Beattie, S. M., “Design of the Truss Spars for the
estimate for many locations. Nansen/Boomvang Field Development,” OTC 14090, 2002.
Preliminary investigation of some full scale and model test
data for VIM of classic spars in the l.ock-in range (where the CH.3 Blevins, R. D., “Flow-Induced Vibration,” 2nd Ed,
motion is well developed and sustained) indicates the maxi- Kreiger Publishing, Malabar, PL, 2001.
mum to standard deviation ratio for inline VIM is about 85%
CH.4 Cowdrey, C. P. and Lewes, J. A. “Drag Measure-
to 90% of that determined by the Rayleigh distribution. For
ments at High Reynolds Numbers of a Circular Cylinder Fit-
transverse VIM in the lock-in range, the ratio of maximum to
ted with Three Helical Strakes,” NPL/Aero/384, Nation.al
standard deviation of VIM amplitude can vary from 1.6 to
Physical Laboratory (UK), July, 1959.
over 2.0 for duration of a few hours to a few days, respec-
tively. Note these values are derived from classic spars and CH5 Deep Oil Technology, Inc., “Vortex Induced Motions
are given for illustration only, and therefore should not be of Large Floating Structures,” Joint Industry Project Report,
used for a specific project without further investigation. Fur- R-900501, 1990.
ther data is required to develop recommendations on peak
value statistics and extreme current duration so the traditional CH.6 DeepStar Joint Industry Program, “Conceptual Spar
approach can be adopted. Facility for Ultra-Deepwater GOM, Phase 1C, Hull VIV
Issues,” Report DeepStar CTR 6405, 2003.
CH.8.3 ANALYTICAL TOOL FOR VIM PREDICTION
CHI? DEEPSTAR VI PROJECT, “DeepStar Ocean Turbu-
Currently there is no reliable analytical tool for the predic- lence Measurement Program,” Deepstar Report 6801, 2004.
tion of spar VIM. Computational Fluid Dynamics (CPD) is
being studied as a means of performing VIM assessment, but CH.8 Ding, Z. J-, Balasubramian, S., Lokken, R. T., and
three dimensional codes necessary for analysis of spars with Yung, T-W., (2004) “Lift and Damping Characteristics of
strakes are still under development. Attempts have been made Bare and Straked Cylinders at Riser Scale Reynolds Num-
to conduct time domain simulations for single or multiple bers,” OTC 16341, 2004.
degrees of freedom spar models with forcing functions deter- CH.9 Edwards, R., Shilling, R. Theti, R. and Karayaka,
mined from model testing, field measurement, or CPD.
M., “BP Horn Mountain Spar—--Results of Comprehensive
Although this approach is promising, significant effort is still
Monitoring of Platform and Riser Responses,” Deep Offshore
required to advance it to a mature stage, especially in the area
Technology Conference and Exhibition, Marseille, 2003.
of forcing function. Since coupling of surge/pitch and sway/
roll can be significant, a 4 degree-of-freedom (surge, sway, CH.10 Feng, C. C., “The Measurement of Vortex-Induced
pitch, and roll) forcing function based on strip theory may Effects on Flow Past Stationary and Oscillating Circular and
provide satisfactory results. Once reliable analytical tools D-Section Cylinders,” M. A. Sc. Thesis, University of British
become available, standard deviation and peak response sta- Columbia, 1968.
tistics can readily be generated, and traditional mooring anal-
ysis methods can be used. CH.l1 Finn, L.D., Maher, J-V., and Gupta, I—I., “The Cell
Spar and Vortex Induced Vibrations,” OTC 15244, 2003.
CH.8.4 MOORING ANALYSIS TOOL FOR SPAR
CH.12 Gartshore, I. S., Khanna J. and Laccinole, S., “The
VIM
effectiveness of vortex spoilers on a circular cylinder in
Most mooring analysis software was not designed to han- smooth. and turbulent flow,” Proc. 5th Intl. Conf. on Wind
dle spar VIM analysis therefore a simplified analysis proce- Engineering, Fort Collins, Colorado, 1978.
144 API RECOMMENDED Psacnca 2SK

CH.13 Hafen, B. E., Meggitt, D. J-, and Liu, F. C., “Strum- Mooring,” Offshore Technology Conference, OTC 2491,
min.g Suppression--An Annotated Bibliography,” Technical 1977.
Note N-1456, Naval Civil Engineering Lab, Oct., 1976.
CH.25 Smith, D. W., Kokkinis, T., Thoinpson, H. M. and
CH.14 Halkyard, J-, “Evolution of the Spar Floating Drill- Greiner, W. L., “Hi.ndcasting VIM-Induced Mooring Fatigue
ing, Production and Storage System,” SNAME (Texas) for the Genesis Spar,” OMAEZOO4-51547, 2004.
Annual Meeting, Offshore Technology: Evolution and Inno-
vation, Houston, Texas, 2001. CH.26 Soreide, M., “Experimental Investigation of In-Line
and Cross-flow VIV,” Proc. 13th Intl. Offshore and Polar
CH.15 Holmes, S., Applied Research Associates, Personal Engineering Conf., Honolulu, 2003.
Communication, Nov. 7, 2003. See link: http:/7
www.arasvo.comfplatfoirn.hnn. CH.2'7 van Dijk R. T., Voogt A., Fourchy P. and Saadat M.,
“The Effect of Mooring System and Sheared Currents on
CH.16 Huang, K., Chen, X. and Kwan, C.T., “The Impact of
Vortex Induced Motions of Truss Spars,” Proceedings 22119
Vortex-Induced Motions on Mooring System Design for
International Conference on Offshore Mechanics and Arctic
Spar-based Installations,” OTC 15245, 2003.
Engineering, OMAE ‘O3, Cancun, Mexico, 2003.
CH.17 Irani, M. B. and Finn, L. D., “Model Testing for Vor-
tex Induced Motions of Spar Platforms,” OMAEZOO4-51315, CH.28 van Dijk R. T., Magee Allan, Perryman Steve,
2004. Gebara Joe, “Model Test Experience on Vortex Induced
Vibrations of Truss Spars,” OTC 15242, 2003.
CH.18 Kokkinis, T., Sandstrom, R. E., Jones, H. T., Thomp-
son, H. M., and Greiner, W. L. “Development of a Stepped CH.29 Vandiver, I-K. “Drag Coefficients for Long-Flexible
Line Tensioning Solution for Mitigating VIM Effects in Cylinders,” Offshore Technology Conference, OTC 4490,
Loop/Eddy Currents for the Genesis Spar,” OMAE2004- 1983.
51546, 2004.
CH.30 Williamson, C.H.K. and Govardham R., “Vortex
CH.19 Magee, Allan, Anil Sablok, Joe Gebara, “Mooring Induced Vibrations,” Annual Review of Fluid Mechanics,
Design for Directional Spar Hull VIV,” Proceedings, OTC 36:16.1-16.42, 2004.
15243. 2003.
1 Woodgate, L. and Matbrey, J . F. M. “Further Experi-
CH.2l) OTRC/MMS Workshop, “Spar Vortex-Induced ments on the Use of Helical Strakes for Avoiding Wind
Motions,” Proceedings of Workshop, Navasota, Texas, Octo- Excited Oscillations of Structures of Circular or Nearly Cir-
ber 22-24, 2003. cular Section,” Natl. Phys. Lab. (UK) Aero. Report 381,
CH.2l Sandstrom, R. E., Yung, T-W., Slocum, S. T. and 1959.
Ding, Z. J-, “Advances in Prediction of VIV for Spar Hulls,” CH.32 Wu, Z.-J-, “Current Induced Vibrations of a Flexible
Deep Offshore Technology Conference, Marseille, France, Cylinder,” Ph.D. Thesis, Division of Structural Engineering,
19-21 Nov 2003. The Norwegian Institute of Technology, The University of
CH.22 Sarpkaya, T. and Isaacson, M., “Mechanics of Wave Trondheim, Norway, 1989.
Forces on Offshore Structures,” Van Nostrand Reinhold
CH.33 Yung, T. W., Sandstrom, R. E., Slocum, S. T., Ding,
Company, New York, 1981 .
J. Z. and Lokken, R. T., “Advancement of Spar VIV Predic-
CH.23 Sarpkaya, T. “A Critical Review of the Intrinsic tion,” OTC 16343, 2004.
Nature of Vortex-Induced Vibration,” Journal of Fluids and
CH.34 Zdravkovich, M. M., “Review and Classification of
Structures, 19 389~447, 2004.
Various Aerodynamic and Hydrodynamic Means for Sup-
CH.24 Skop, R.A., Griffin, O.M. and Ramberg, S.E. pressing Vortex Shedding,” Journal of Wind Engineering and
“Strumming Predictions for the SAECON II Experimental 1ndusir'ialAer0dynanzics, Vol. 7, 145-189, 1981.
 APPENDIX I--GLOBAL ANALYSIS GUIDELINES FOR DEEPWATER
FLOATING SYSTEMS

I.1 Basic Considerations tems, a complete scaled model of the deep and ultra deep
water floating system is simply too large to test in present
The design and analysis of mooring and riser systems for
model basins. The model’s mooring and riser systems are
floating drilling and production systems require global analy-
often truncated, and a numerical model is validated and
sis, which evaluates responses of the floating vessel under
then used to interpret and extend the model test results to
specific environmental conditions. Traditionally global analy-
predict the prototype floating systern responses at the full
sis involves th.e use of both numerical analysis and physical
design water depth. This approach is called truncation or
model tests. Numerical analysis models are used to predict
hybrid verification method. Alternatively ultra-small scale
the responses for the floating vessel, and model tests are used
testing (1 :l50 -- 1:170) can be used. The industry has lim-
to calibrate the numerical models and validate their predic-
ited experience with the truncation and ultra-small scale
tions. The robustness of calibrated and validated numerical
testing methods. This causes additional uncertainty in glo-
models to capture the physics and complex interactions at
bal response predictions.
both model scale and prototype scale is critical to the overall
design process. This procedure has provided the basis for Another method that has been used by the industry is to
many successful floating operations in the past. model test components of the full system to gain understand-
ing and confidence in the behavior and properties of the com-
As the industry progresses into deep (to 6,000 ft) and
ponents. For example, in model testing CALM buoys, large
ultra deep (to 10,000 ft) water depths, current technology in
predicting global responses of floating vessels faces new scale forced oscillation tests of the buoy alone may be used to
challenges. First, slender bodies such as risers and mooring develop the hydrodynamic properties of the buoy, with the
lines have much higher impact on the floating vessel in deep risers and mooring lines being tested separately, then numeri-
and ultra deepwater. Also the number of slender bodies cal tools are used to model the entire system. Similarly, for
tends to increase as large deepwater fields are developed. truss spars the hard tank and truss members may be model
The traditional approach of de-coupling the floating vessel tested separately, to get the best possible scaling, and com-
and the slender bodies may not be sufficient for deep and bined in the model test setup and in the numerical modeling.
ultra deepwater operations. To address this issue, sophisti- However, this approach may not fully capture the coupling
cated computer programs have been developed recently to between the components or may introduce new uncertainties
perforrn coupled analysis where the vessel and the slender for the whole system at the expense of greater knowledge of
bodies are integrated. in a single model to fully capture their the individual components.
interaction. Industry studies indicate, however, some of the The uncertainties in numerical analysis and model testing
computer programs may not be efficient tools for design may result in either overly conservative design or unrecog-
because of the large computational effort required for the nized risks, and therefore may have safety and/or cost impli-
complicated models. Also they may not yield consistent cations for a project. To address this issue, several studies
results because different approaches, such as time domain have been sponsored by DeepStar to better understand the
and frequency domain, are used. Some designers still prefer uncertainties in the predicted responses of deepwater systems
to use the more efficient de-coupled or semi-coupled and to provide a basis for reducing these uncertainties. A
approaches for conducting a large number of global analy- summary of the DeepStar studies is presented in Section I.2.
ses required for floating system design. The use of various The global analysis guidelines provided in this Appendix
approaches increases the discrepancy in global response were derived from the DeepStar studies and industry experi-
predictions. ence, and have gone through a lengthy process of consensus
The second challenge is rnodcl testing for deep and ultra building. The objective of this Appendix is to provide general
deepwater operations. In relatively shallow water, a com- design and analysis principles. It is not the purpose of this
plete scaled. model can be tested in a model basin at reason- Appendix to dictate a specific approach for global analysis.
able model scales (1:50 - 1:100). Test results can be Various approaches have been used in the past and this trend
directly scaled using Froude Number scaling to predict the is expected to continue. Instead this Appendix points out the
prototype performance. This approach has been success- important parameters and advantages and limitations of vari-
fully used in many model basins to study floating systems ous approaches for global analysis so the designers can make
in depths to 3,000 ft or more. Testing floating systems in intelligent decisions for their analysis and model testing. This
deep and ultra deep water depths, however, is more diffi- document also points out directions for future technology
cult. Due to the depth and scope of mooring and riser sys- development.
146 API RECOMMENDED PRACTICE 28K

|.2 Summary of DeepStar Studies urnbilicals. Coupling of the floating vessels and slender bod-
ies can be divided into 2 categories:
A series of studies on global analysis were conducted in 3
phases of th.e DeepStar JIP from 1999 to 2003. In DeepStar
|.3.1 SINGLE VESSEL COUPLING
Phase IV [Ref I. 1] three theme structures--MFPSO, Spar, an.d a
TLP--were selected for the stud.ies. Realistic designs were In this case a single floating vessel is connected with moor-
developed for each structure for water depths ranging from ing lines (or tendons in TLP) and risers, which affect the ves-
3,000 to 10,000 ft, and design environments for Gulf ofMex- sel responses in the following terms:
ico hurricanes and loop currents were used. Global perfor-
mance analyses of these structures were performed by a l. Mooring and riser system stiffness»--l\/lootring stiff-
number of organizations, and the predicted design responses ness provides majority of the restoring force to keep
were compared to characterize the differences and uncertain- the vessel on station and therefore must be properly
ties among the different analyses. Since the differences and accounted for. The importance of riser stiffness
uncertainties were found to be significant, model tests to depends on number and type of riser. For drilling ves-
measure design responses for these three structures were then sel equipped with a single top tensioned riser, the riser
cond.ucted to provide the basis for further comparing and stiffness is often neglected. On the other hand, for
characterizing the uncertainty between predicted and mea- floating production vessels equipped with a large num-
sured responses. ber of risers, riser mean load and stiffness can make
In DeepStar Phase V [Ref 1.2-I. 12] a study was undertaken significant contribution to the total restoring force and
to compare the predicted responses of three theme structures therefore should be properly accounted for. In particu-
with model test in order to assess the accuracy of and differ- lar, mean riser loads may result in significant
ences between various numerical models. In this study, ana- asymmetry in the stiffness of combined riser and moor-
lysts from a number of organizations conducted detailed ing system. In the case of CALM buoys, mooring and
analyses of data from the floating vessel model tests to vali- riser system forces may be large compared to first
date their analytical tools. Following the validation process, order wave forces, requiring wave frequency responses
predicted results from the validated analysis models were to be calculated in the presence of the riser and moor-
compared to model tests for each floating structure to assess ing system.
the overall accuracy and scatter between the predicted
2. Mooring and riser system damping--Mooring and
response and measured values from the model tests. Included
riser system damping affects mainly the low frequency
in DeepStar V is also a first attempt to develop guidelines for
motions of the vessel. The importance of this parame-
global performance analysis and verification of deepwater
ter depends on the number, type, and size of the
structures.
mooring lines and risers, water depth, type of vessel,
In DeepStar Phase VI a review was first carried out on the and the metocean environment. For a semisubmersible
guidelines for global performance analysis and verification to drilling vessel equipped with a single riser and 8 moor-
identify areas for further development [Ref L13]. Then a
ing lines operating in shallow water depths, mooring
study was conducted to further develop global analysis guide-
and riser damping is often neglected. As the water
lines that can be used to guide global response analysis and
depth, number of mooring lines and risers increase,
m.odel testing and can be directly incorporated into API RP
mooring and riser damping will become more and
ZSK. In addition, methods for model testing floating struc-
more important, especially for vessels such as FPSO
tures with truncated mooring lines and risers were further
where low frequency motions dominate the design. In
investigated.
general, low frequency damping from mooring lines
These studies represent a major effort by the industry to and risers is larger for catenary systems than for taut
advance the global analysis technology for deep and ultra- systems, and increases as the magnitude of the vessel's
deep water operations. The most important piece of work is wave frequency motions increase. Neglecting mooring
the model testing for 3 theme structures, which provides a
and riser damping in global analysis is always conser-
sound basis for bench marking different analysis models. vative and may be justified where the impact of the
Important findings from the DeepStar and other industry
damping is small. However, the cost impact of this
studies have been incorporated in the following sections.
practice should be carefully evaluated to avoid signifi-
cantly over-sizing the mooring and riser system.
|.3 Coupling of Floating Structures
3. Current load on mooring and riser--Current load on
Floating drilling and production systems typically consist mooring and riser imposes additional loading on the
of single or multiple floating vessels such as spar, TLP, floating vessel. The importance of this parameter
FPSO, or semisubmersible, which are connected with slender depends on the number, type, and size of the mooring
bodies such as mooring lines, risers, fluid transfer lines, and lines and risers, the water depth, and the relative mag-
 Dssioiv use ANAt_YS%S or Srarsonxsssino Svsraus son Fi.C>ATiN£3 Srnucrusss 147

nitude of the current loads compared to wind and wave 3. Semi-coupled analysis--In this approach the floating
loads. Since neglecting this loading is always non-con- vessel, the mooring and riser system are integrated into
scrvative, it is good practice to include it in global a single model. The vessel is analyzed dynamically, but
analysis, especially in deep water or high current area. the slender bodies are analyzed quasi-statically. This
Wave load on mooring and riser»--Wave load on approach can accurately account for the current load
mooring and riser has not been found to play an impor- and nonlinear stiffness of the mooring and riser system
tant role on global response of large floating vessels bu.t cannot capture the damping and inertia of the slen-
and therefore is often neglected. However for CALM der bodies, which are either estimated separately or
buoys, offloading and mooring lines may have a signif- simply neglected. By neglecting slender body dynam-
icant impact on the wave frequency motions of the ics, the analysis can be much faster, but the accuracy
buoy. will depend largely on the slender body damping
estimation.
Inertia of mooring and riser--Inertia of mooring and
riser provides additional contribution to the inertia of
l.3.2 MULTIPLE VESSEL COUPLING
the floating vessel. This parameter may not be very
important for large floating vessels, but can be impor- In this case two or more floating vessels are connected
tant for small floating structures such as buoys. with fluid transfer lines (FTL) and therefore the response of
one vessel affects the other vessels. An example of multiple
The industry has used a number of approaches to analyze a vessel coupling is an FPSO connected to a spar or TLP dry
floating vessel connected with mooring and riser system: tree unit on one side and an offloading buoy on the other side.
De-coupled analysis--In this approach the floating Another example is a moored drilling tender vessel operating
vessel and the mooring or riser system are analyzed. alongside a floating dry tree unit (spar or TLP), with both
separately and therefore called de-coupled analysis. vessels inter-connected by mooring hawsers. There are two
The floating vessel is analyzed first, and the slender types of multiple vessel coupling:
body parameters such as stiffness, damping, wave and 1. Environmental load coupling--This coupling arises
current load, and inertia are either estimated and input because each vessel influences the incident wave kine-
to the floating vessel analysis or simply neglected. matics field in its vicinity due to diffraction and
Mooring stiffness is most important and always radiation effects. Also there are shielding effects for
required by the floating vessel analysis. This has not wind and current loads. This coupling can be negligi-
been a problem since mooring stiffness can easily be ble if the vessels are far apart from each other.
estimated. Inclusion of other parameters often depends
on the judgment of the analyst and availability of 2. Structural coupling--The response of one vessel can
appropriate parameter values. The responses of the be affected by another vessel through the FTLs, and
vessel from the analysis are then input to the mooring the stiffness, damping, inertia, wave and current loads
or riser analysis to obtain the responses of the slender of the FTLs can influence both vessels connected to its
bodies. ends.

This approach was routinely used in the early stage of Similar to single vessel coupling, the industry has used a
the offshore industry because coupled analysis tools number of approaches to analyze multiple floating vessels
were not available. The accuracy of the analysis is connected with FTLs.
uncertain, depending largely on the skill and experi- l. De-coupled analysis--In this approach the floating
ence of the analyst. vessels are analyzed. separately. The environmental
Coupled analysis——In this approach the floating ves- load coupling is neglected, and only the mean load of
sel, the mooring and riser system are integrated into a the FTLs is accounted for in the structural coupling.
single model and the vessel and slender bodies are ana- The accuracy of this approach depends on the distance
lyzed dynamically. This approach can fully account for between the vessels, the size of the FTLs and the ves-
the interaction between the vessel and the slender bod- sels, the relative stiffness of the FTLs and the vessel's
ies and therefore allows capture of all the slender body mooring systems, and the severity of the environment.
effects such as stiffness, damping, wave and current Industry experience indicates that if the FTL stiffness
load, and inertia. Some software may select to ignore is much smaller than the stiffness of the floating ves-
minor effects such as wave load on the slender bodies. sels, this approach can provide good results.
This approach is most accurate, but the analysis can be 2. Coupled analysis—-In this approach the floating ves-
very time consuming unless efficient frequency sels and the FTLs are integrated into a single model
domain solutions are used. and the vessels and FTLs are analyzed dynamically.
148 API RECOMMENDED Pnacrics 28K

This approach can fully account for the interaction current load spectrum. These results are directly applicable to
between the vessels and the FTLs and therefore can frequency domain coupled analysis. For time domain coupled
capture all FTL effects such as stiffness, damping, analysis, however, the results from the frequency domain
wave and current load, and inertia. This approach is hydrodynamic analysis is normally transformed to time
most accurate, but the analysis can be very time con» domain by convolution. This practice results in generally
suming unless efficient frequency domain solutions are consistent environmental loads for both frequency domain
used. If the floating vessels are close to each other, the and time domain analysis. Industry studies indicate that this
coupling effects of waves, wind, and current should approach gives satisfactory solutions for typical floating sys-
also be accounted for in a fully coupled analysis. tems [Ref I. 14].
3. Semi-coupled analysis--Itn this approach the floating It should be noted that low frequency vessel motions are
vessels and the FTLs are integrated into a single caused, in part, by nonlinear second order drift forces. If the
model. The vessels are analyzed dynamically, but the natural period of the moored floater is high, say more than 25
FTLs, mooring lines, and risers are analyzed quasi» seconds, a linear frequency domain solution can be achieved
statically. This approach cannot capture the damping by Newmarfs approximation, which eliminates the off-diago-
and inertia of the FTLs, mooring lines, and risers, nal terms in the QTF (Quadratic Transfer Function) matrix.
which are either estimated separately or simply Newmarfs approximation generally gives satisfactory results
neglected. By neglecting slender body dynamics, the for low frequency motions in the horizontal plane where the
analysis can be much faster, but the accuracy will natural periods are much larger than the wave periods. For
depend on distance between the vessels, size of slender low frequency motions in the vertical plane, for example the
bodies and the vessels, the severity of the environment, pitch motion of a spar, Newman’s approximation may u.nder-
and the estimate of slender body low frequency damp» estimate the second order drift forces. If response of this type
ing used. is important for the design, time domain solution of a full
QTF matrix may be required.
i.4 Time Domain and Frequency Domain Second-order wave forces in a random sea oscillating at
Dynamic Analysis the sum-frequencies may excite resonant response in heave,
roll, and pitch of a TLP, which is often referred to as spring-
Two rnethods, frequency domain and time domain analy- ing response. Also deepwater TLPs may experience very
sis, are commonly used for predicting dynamic responses of a large resonant high frequency transient ringing response.
floating vessel. In the time domain method, all nonlinear Time domain analysis is typically performed to evaluate these
effects can be modeled. The term time domain implies recal- high frequency responses for TLP. Cylindrical hull forms
culation of each mass term, damping term, stiffness term, and such as spar may be subjected to highly nonlinear vortex
load at each time step. Hence the computation can become induced excitation, which is also typically analyzed in time
complex and time consuming. The frequency domain domain.
method, on the other hand, depends on the linear principle of
superposition. Hence, all nonlinearities must be eliminated, I.4.2 DYNAMICS OF SLENDER BODIES
either by direct linearization or by an iterative linearization. it
should be noted that linearization in frequency domain analy- There are 4 types of nonlinearity for slender bodies such as
sis is required only for dynamic response about the mean risers and mooring lines, which can be directly modeled by
position. Nonlinear properties such as nonlinear mooring time domain analysis but must be linearized by frequency
stiffness to determine the mean position and amplitudes of domain analysis:
low frequency motions can be handled approximately by fre-
1. Line stretching»--The load versus elongation relation-
quency domain analysis. ship of slender bodies must be linearized for the
dynamic response. The stiffness cannot be a function
l.4.1 ENVIRONMENTAL LOADS ON FLOATING of dynamic line tension but can vary along the line.
VESSELS This is usually not a difficult requirement even in the
Dynamic environmental loads such as wind, current, and case of synthetic material and in most cases, a suitable
wave loads are normally analyzed in frequency domain using linearization can be achieved.
a wind/current excitation and wave radiation/diffraction 2. Geometry cha.nge-In the frequency domain method
solver. The average wetted area of the vessel is used in the it is assumed that the dynamic displacements are small
analysis. The results of the analysis gives first order excita- perturbations about a static position. The static shape is
tion forces, hydrostatics, potential damping, added mass, first fixed and all geometric quantities are computed based
order RAOs, second order drift force coefficients, and wind/ on this position. The mass, added mass, stiffness, etc.
 Desiou AND Answers or $TATiONKEEPiN(3 SYSTEMS son Fi.OATtNG Sraucruass ‘$49

are computed only once. Changes in catenary shape lyzed dynamically using a time domain solver, but the slender
due to the dynamic motion contribution are generally bodies are analyzed quasi~statically (using tension/offset
not severe, especially for deepwater systems. Hence, a table) to allow fast analysis. The damping of the slender bod-
linearization about the position under mean load is ies is typically neglected, resulting in conservative motion
generally acceptable. predictions. The time history ofthe floating vessel response is
3. Fluid loads---The nonlinear term in the l\/lorrison then input to a dynamic mooring or riser analysis software to
equation m.ust be linearized. The quadratic relationship obtained mooring or riser responses. To further simplify the
in the relative velocity must be replaced by an equiva- analysis, dynamic amplification factors are often derived
lent linear relationship. The linearization should take from the dynamic mooring analysis, which are then used to
into account the frequency content of the line motion modify the quasi-static tensions from the semi-coupled time
spectrum. The development of effective stochastic lin- domain analysis. This approach has been extensively used for
earization or energy dissipation method allows spar floating production systems. ,
accurate linear approximation for fluid loads on slen-
der bodies. I.5.3 SEMI-COUPLED TIME DOMAIN PLUS
4. Bottom effects--The frictional behavior between the COUPLED TIME DOMAIN ANALYSIS
grounded line and the seafloor cannot be represented In this approach a large number of design cases are ana-
exactly in the frequency domain. Only the average or lyzed by the semi-coupled time domain analysis with damp-
equivalent behavior of the line can be postulated and ing of slender bodies estimated from model test or energy
included. This simplification. should be adjusted to the dissipation method. A few critical load cases are identified
design objective. Different models may be required for and analyzed by coupled time domain analysis. This
the fatigue and the extreme tension. evaluations. approach has been used for FPSOs.

I.5 Current Industry Practice I.5.4 COUPLED FREQUENCY DOMAIN ANALYSIS

As discussed in Sections I.3 and 1.4, various analysis tech-
In this approach coupled frequency domain dynamic anal-
niques such as de-coupled, coupied, semi-coupl.ed, time
ysis software that captures all slender body effects is used in
domain, and frequency domain analysis have been developed
all design phases. This is the most efficient approach but
to evaluate global responses of floating systems. A survey of
requires the software to be extensively verified by model test-
current industry design practice indicates that designers are
ing and/or coupled time domain analysis. This approach has
using various approaches, depending on vessel type (FPSO,
been used for FPSOs.
spar, TLP, semisubmersible, etc.), loading type (extreme,
fatigue, etc.), design stage (preliminary, final, etc.), and oper-
ation type (drilling, production, offloading, etc.). Following l.5.5 COUPLED FREQUENCY DOMAIN PLUS
are the approaches identified by the survey, which is focused COUPLED TIME DOMAIN ANALYSIS
on single vessel coupling: In this approach a large number of design cases are ana-
lyzed by coupled frequency domain. analysis. A few critical
l.5.'I DE-COUPLED FREQUENCY DOMAIN load cases are identified and analyzed by coupled time
ANALYSIS domain analysis. This approach has been used for FPSOs,
In this approach, dynamic responses of the floating vessel spars, and semisubmersibles.
alone are calculated by a frequency domain wind/current
excitation and wave radiation/diffraction solver. Typically I.5.6 DE-COUPLED FREQUENCY DOMAIN PLUS
stiffness of the slender bodies is accounted for but the damp- COUPLED TIME DOMAIN ANALYSIS
ing is not, resulting in conservative motion predictions. The
In this approach, responses of the floating vessel alone
responses of the floating vessel are then input to mooring or
are calculated by a frequency domain radiation/diffraction
riser analysis software to obtain mooring or riser responses.
solver such as WAMIT. The responses of the floating ves-
This approached has been used for various types of floating
sel are then input to finite element models for slender body
vessels.
responses. Coupled time domain analysis or model test is
then used to derive calibration factors for the de-coupled
I.5.2 SEMI-COUPLED TIME DOMAIN ANALYSIS
frequency domain analysis and to analyze critical cases
In this approach the floating vessel, the mooring and riser and fatigue loading. This approached has been used for
system are integrated into a single model. The vessel is ana- TLPs.
150 API RECOMMENDED PRACTICE ZSK

|.6 Assessment of Current Industry ing lines are few. It is not an accurate and efficient approach
Practice for major deepwater floating production operations, if cali-
bration with model test or coupled time domain analysis is
l.6.1 CRITERIA FOR EVALUATING GLOBAL not carried out.
ANALYSIS SOFTWARE
In order to assess current industry practice, the following l.6.3 SEMI-COUPLED TIME DOMAIN ANALYSIS
criteria are established for evaluating global analysis soft- This approach is more accurate than the de-coupled fre-
ware: quency domain analysis in that the non-linear stiffness and
l. Accuracy--The software should yield accurate pre- current load of the slender bodies can be better accounted for.
diction for responses of the floating vessel and slender However, this approach cannot calculate the damping of the
bodies. This is essential to achieve a good balance slender bodies, which can be important for most deepwater
between reliability and cost of the floating system. floating production systems. By using quasi-static sorlntion
Many designers tend to bias on the conservative side for slender bodies, the efficiency is much better than the cou-
when faced with uncertainty in global response predic- pled time domain analysis. However, this cannot be consid-
tion. For example some software cannot predict ered an efficient analysis tool since dealing with random time
damping of slender bodies, and the damping is simply history still requires substantial computational effort.
neglected by some designers. This is obviously conser-
vative, but the cost impact of this practice can be l.6.4 COUPLED TIME DOMAIN ANALYSIS
significant, especially for deepwater systems with a This approach is considered the most accurate for its capa-
large number of mooring lines and risers. The software bility to capture all system nonlinearities, but least efficient. It
is required to yield accurate results not only for maxi- should be emphasized that the accuracy of this approach is
mum responses, but also for high and low frequency not given because mishandling of certain time domain analy-
standard deviation responses that are important for sis parameters can lead to significant errors in response pre-
fatigue analysis. For fiber rope moorings, the mini- dictions. lndustry studies indicate that to achieve good
mum responses can also be important. accuracy, the time step, length of simulation, number of repli-
2. Efficiency--To properly design a deepwater floating cates, number of frequencies to represent the loading spec-
system, a large number of load cases are often ana- trum, and method to derive the maximum response must be
lyzed. This is necessary to address a large number of carefully chosen. This often results in long simulation times
seastates and environmental directions required in the of hours to hundreds of hours for each load case even with
strength and fatigue design of mooring, riser, and struc- modern high speed PCs. The engineering time to investigate
tural components of the vessel. Optimization of a number of time domain simulations can also be significant.
mooring and riser systems may further increase the Because of its low efficiency, this approach is not suitable for
number of load cases. The number of load. cases typi- routine design of floating systems. However, it is a valuable
cally ranges from a few hundred. to over a thousand analytical tool for systems of high nonlinearity or large move-
using the deterministic design approach. Using ment, for checking critical cases in the design and verifying
response based design approach; however, the number the more efficient frequency domain approach.
of load cases may exceed 10,000. Therefore it is essen-
tial to have efficient global analysis software to ensure I.6.5 COUPLED FREQUENCY DOMAIN ANALYSIS
a proper design can be completed on schedule and
This approach is most efficient and can be accurate for
within budget.
typical deepwater floating systems. The frequency domain
method uses linear principle of superposition. Most slender
l.6.2 DE-COUPLED FREQUENCY DOMAIN
body nonlinearities such as line stretching, geometry
ANALYSIS
change, and bottom effect are very minor for deepwater sys-
This approach camiot accurately account for the interaction tems. The only true nonlinear term is fluid load, which can
between the floating vessel and the slender bodies and there- be effectively linearized by well established technology.
fore is not considered very accurate without calibration with From theoretical point of view, coupled frequency domain
model test or coupled time domain analysis. Also the analysis analysis should yield accurate response predictions for typi-
is conducted in 2 steps: first the floating vessel and second the cal deepwater systems if slender body fluid load is properly
mooring or riser system. The efficiency of this approach linearized..
depends on whether there is an efficient link between the two The above view has been verified by DeepStar and other
steps. This approach can be used for temporary operations industry studies, which compared frequency domain solu-
such drilling operations using MODUs where riser and moor- tions with model testing and time domain solutions. The
 Desron AND ANAi_YSi$ or Srarionueeeino Svsraus son FLGATiNG Sraucrunas 151

comparison was conducted for different types of floating several minutes. A 3-hour simulation may contain less than
structure operating in GOM and West Africa environments. 50 cycles, which is insufficient to provide a good statistical
Frequency dom.ain solutions show good agreement compared confidence for standard deviation. The requirement for simu-
with DeepStar model test results and time domain simula- lation length may even be higher for extreme responses. For
tions for all 3 DeepStar theme structures--spar, turret moored example to obtain statistically meaningful wave frequency
FPSO, and TLP. Other industry studies show good agreement extreme responses, several 3-hour simulations may be
between frequency and time domain solutions for a semisub- needed. Required length of simulation depends on a number
mersible and a spa.r floating production. system operating in of factors, such as periods of wave and low frequency
GOM and a spread moored FPSO operating in West Africa responses, contribution of wave and low frequency responses
[Ref l.l4-l.1'7]. It should be noted that turret moored FPSO to total response, degree of nonlinearity, and system damping.
with large low frequency yaw may require special treatment It should be determined by sensitivity check. Recent studies
in frequency domain solution, as discussed in Section 1.6.8.3. indicate that five to ten 3-hour simulations (15 to 30 hours) or
Coupled frequency domain analysis is very efficient. With equivalent with different seed numbers may provide standard
a high speed PC, the run time for a load case ranges from sec- deviation and extreme responses of good confidence for typi-
onds to minutes, depending on software efficiency, which is cal deepwater floating systems [Ref l. l4-I. 17].
several orders of magnitude faster than coupled time domain
analysis. It is very suitable for analyzing a large number of l.6.7.3 Frequency Discretisation
load cases required by the deepwater floating systems. How-
ever, it may not be able to handle systems of hi gh nonlinearity The number and range of discrete frequencies representing
or large movement, and it requires significant verification or floater transfer functions should be carefully chosen. to cover
calibration by model testing and/or time domain analysis. the peaks in the transfer functions and area of significant
wave excitation. Also it should be clarified how the actual
l.6.6 COMBINATION OF APPROACHES computer program handles possible excitation outside the fre-
quency range of the floater transfer function since this can be
As discussed in Section 1.5, designers often use a combina- a source for erroneous prediction. Small frequency spacing
tion of 2 to 3 approaches to achieve the accuracy and effi- may be required to avoid repeating time history within the
ciency required for d.eepwater system design. For example simulation length. This problem can be alleviated by using
semi-coupled time d.omain analysis or coupled frequency variable frequency spacing, but the repetition period of the
domain analysis is used for large number of load cases and time history is more difficult to assess. The number, range,
coupled time domain analysis is used for a few critical load and spacing of frequency should be determined by sensitivity
cases. This practice has served the industry well in the past check. Industry studies have shown that a few hundred
and is expected to continue in the future when the industry equally spaced frequencies yield satisfactory results for typi-
moves into even deeper water. However, from the accuracy cal deepwater floating systems. To determine proper fre-
and efficiency point of view, coupled analysis is more supe- quency spacing, the following equations can be considered
rior to de-coupled and semi-coupled analysis, and therefore [Ref 1.15]:
should be more emphasized for future applications. ~ To capture the resonant response:

l.6.7 TIME DOMAIN ANALYSIS GUIDELINES

Ate £n~zin(Qto,,) (Ll)
l.6.7.1 Time Step
Choice of time step is crucial for the stability and accuracy ~ To avoid repetition of time history:
of time domain solution, and is often dependent on the peri-
27:
ods of the responses, degree of nonlinearity, and analysis for- w ,.
A 5 -————-------
- .rz'muZai:'01t
r1.2>
mulation. Time step should be determined by a sensitivity
check. Recent studies for deepwater floating systems typi- where
cally used time steps ranging from 0.1 sec. to 0.5 sec.
0) I natural frequency of mode ri,
|.6.7.2 Length of Simulation
Q e is the model damping as ratio to critical damping.
Traditionally 3-hour duration is often used for model test-
ing and time domain simulation. This duration is generally
l.6.7.4 Initial Transient Response
sufficient for the standard deviation of wave frequency
responses because it represents about l,()00 cycles of The length of time domain simulation must allow for tran-
response with a period of 10 sec. Low frequency responses sient response during the initial part of the simulation. The
for deepwater systems, however, typically have periods of time for transient response is normally a function of the
152 API RECOMMENDED PRACTICE 28K

period and damping of the response. Industry studies indicate |.6.7.6 Application and Limitation of Time Domain
that typical deepwater fl.oating systems may require approxi- Analysis
mately a couple thousand seconds for transient response. To
As can be seen from above discussion, time domain analy-
determine a proper time for initial transient response, the fol-
sis places a heavy burden in terms of computer and engineer-
lowing equation can be considered [Ref]. 15]:
ing time, skill, and experience on a designer. Mishandling of
certain analysis parameters can lead to erroneous predictions.
Tn Its inefficiency makes it unsuitable to handle a large number
Tiransz'emf 2 n/la-x(5’é)
of load cases, which unfortunately are often required for
modern deepwater systems. The advantage of time domain
where analysis is its capability to model all system nonlinearity.
Ty, e natural period of mode n, This advantage may be limited for typical deepwater floating
systems since most nonlinearities are either minor or can be
5', K model damping as ratio to critical damping. effectively linearized. Nevertheless, time domain analysis can
be valuable for the following applications:
|.6.7.5 Extreme Response 1. Critical Load Cases--Time domain analysis is often
The determination of extreme value from stochastic time performed for a few critical load cases to ensure a reli-
history is always a challenge, which can be addressed by the able design after a large number of load cases have
following 3 methods: been analyzed by the more efficient frequency domain
analysis.
l. Probability Density Function (PDF)---In this 2. Verification and Calibration Factor for Frequency
approach a PDF for the extreme response is con- Domain Analysis---Time domain analysis can be used
structed, and the Most Probable Maximum (MPM) is to verify frequency domain analysis software or
the response where the PDF is a maximum. This is not develop calibration factors to modify frequency
a practical approach since construction of a smooth domain analysis results.
PDF may require a large number; say a few hundred,
of responses different realizations [Ref 3. Large Vessel Movement--Examples of this condition
1.15]. include turret m.oored FPSO subjected to large low fre-
quency yaw and floating vessel equipped with DICAS
2. Average of Maximum Responses--In this classical (Differentiated Compliance Anchor System)
approach the average of the maximum responses from
a number of realizations of different random seeds is 4. Transient Condition--Exarnpl.es of this condition
taken as design maximum response. The average of include broken line transient condition and floater sub-
maximum responses from five to ten 3-hour simula- jected to West Africa squall.
tions is typically used for deepwater floating systems. 5. Highly Nonlinear Hydrodynamic Loading--»-Struc-
tural components subjected to highly nonlinear
3. Fitted Probability Distribution Models-~In this
hydrodynamic loading may require time domain analy-
approach a peak probability distribution model, such as
sis. Examples of this condition include green water
Rayleigh, Normal, Gumbel, Weibull, and Exponential,
effect, ringing and springing of TLP tendons, wave
is selected and the parameters in the selected model are
loading on a buoy, and cylindrical components sub-
determined using available response time histories.
jected to vortex induced excitation.
Then the expected extreme response can be computed
from the fitted model. This approach may require 6. Highly Nonlinear Bottom Effect--An example of
fewer realizations then Method 2. However, in practi- this condition is a long length of mooring line or riser
cal applications the fitted parametric model often fails moving up and down on the seafloor u.nder a severe
to describe the “true” upper tail behavior, resulting in storm.
biased extreme response prediction. Some analysts use 7. Dynamic Positioning--Floating vessels equipped
special techniques such as fitting the upper tail or tak- with DP system or DP assisted mooring may require
ing average of predictions from several realizations to time domain stationkeeping analysis.
improve accuracy. Nevertheless this is an approach 8. Highly Nonlinear Mooring Material--Examples of
that requires substantial skill and experience. this condition include certain fiber ropes such as nylon
Method l is most rigorous but is not a practical approach and rubber mooring fender.
for design. Method 2 and 3 are commonly used and consid- 9. Impact or Contact Loading--Exarnples of this con-
ered acceptable approaches, but Method 3 requires more skill dition is collision of floating vessels and loads on
and experience. rubber fender or riser guide.
 Dusters awn Auatvsis or Srariouassrnne svsrsias son Froarino Srnucruaas 153

l.6.8 FREQUENCY DOMAIN ANALYSIS loading, highly nonlinear mooring material, and impact or
GUIDELINES contact loading, etc., as discussed in Section 1.6.7.6. It should
be noted that some of these cases can still be solved in fre-
Cl.6.8.1 Analysis Parameters
quency domain, but special treatments may be required to
The frequency discretisation requirement is similar to that yield acceptable or conservative approximations.
for time domain analysis as discussed in Section 1.6.7.3. One An example case that requires special treatment is turret
exception is that the number of frequencies can be much less moored FPSO, which may experience large low frequency
because repeating time history is no longer an issue. lf a finite yaw due to turret located. at a significant distance from the
element solution is used, the user needs to pay attention to the bow. The large low frequency yaw causes significant change
number of elements representing the floating vessel and slen- in vessel heading, wind, wave, and current loading, resulting
der bodies, but this is the same for time domain analysis. in non-stationary vessel and slender body responses. The tra-
The primary responses from frequency domain analysis ditional frequency domain approach of fixing the vessel at the
are the standard deviation responses for wave and low fre- stable equilibrium heading will not yield accurate response
quency components. The extreme responses are derived from predictions, and the Rayleigh distribution may not be a good
the standard deviation responses and a suitable peak probabil- fit for the response peaks. Several special treatments have
ity distribution model. For floating systems where frequency been used by the industry to provide better frequency domain
domain analysis yields good approximations, most of the solutions. Section 5.6.2 of the main document describes a
wave and low frequency responses can be represented by a conservative special procedure, which uses a design heading
narrow band Gaussian process with Rayleigh distributed defined as the stable equilibrium heading plus or minus sig-
peaks, and Equations 5.5 to 5.8 in Section 5.5 in the main nificant low frequency yaw. In another special procedure, fre-
document should apply. This does not rule out the use of quency domain analyses are performed at different headings
other peak probability distribution models if they can be that bound the expected range of low frequency yaw. It
shown to better represent the responses of a floating system. should be emphasized that special treatments of this nature
Time domain simulation or model testing are often used to require significant verification and calibration by model test-
identify the most suitable model to determine the extreme ing and/or time domain solutions.
responses from the standard deviation responses.
I -u Inn J54 I54
|. I ueepatar bIUdl9S for Global Analysis
l.6.8.2 Verification and Calibration of Frequency Guidelmes
Domain Analysis Software
l.7.1 STUDY SCOPE
Since frequency domain solution is a linear approximation
to nonlinear problems, frequency domain analysis software DeepStar Phase VI has conducted a series of studies for the
should be tho-roughly verified or calibrated before use for development of global analysis guidelines. The studies
design. Model test data should be used for this purpose if include:
good model test data are available. Time domain solutions
l. State of technology review for global analysis of float-
can also be used to bench mark frequency domain solutions.
ing vessels
It is common industry practice to check a few critical load
cases by model testing and/or time domain analysis after the 2. Comparison of time and frequency domain approach
floating system is designed by the more efficient frequency for coupled analysis
domain analysis. This is a prudent practice to ensure a reli- 3. Sensitivity study on mooring line fatigue
able floating system.
4. Parametric study on effects of slender body and other
important parameters to global responses of floating
l.6.8.3 Application and Limitation of Frequency
vessels. The following parameters have been
Domain Analysis
investigated:
DeepStar and industry studies have shown that frequency ~ Vessel type: spar, TLP, FPSO, sernisubmersible.
domain coupled analysis yields accurate response predictions
- Water depth: 3,000, 6,000, and 10,000 ft.
for typical deepwater floating systems including spar, TLP,
FPSO, and semisubmersible floating systems. In fact fre- - Environment: GOM hurricane and loop current
(Spar VIM in loop current is not considered).
quency domain analysis has been used for the design of many
major floating operations around the world. There are cases, ' Number of slender bodies: small, median, and large.
however, that are difficult for frequency domain analysis to ~ Mooring line material: steel, polyester.
handle, and time domain analysis may be more appropriate. ' Drag coefficient of slender bodies: low, median, and
These cases include large vessel movement, transient condi- high.
tion, dynamic positioning, highly nonlinear hydrodynamic ~ Wave period: base value plus and minus l0%.
154 API RECOMMENDED Psaci-"ice 28K

' Significant wave height: base value plus and minus The uncertainties in global analysis tend to decrease
10%. with increasing water depth, as the dynamic response
' Wind spectrum: NPD and API. in terms of total response becomes smaller in deeper
water. This indicates that extending the technology and
The spar is a classic spar of 122 ft diameter and 650 ft
experience in global analysis for shallow and deep
draft. It is moored by a symmetric l4-point chain/wire rope/
water operations to ultra deep water operations is gen»
chain mooring, and the risers are vertical self standing risers.
erally acceptable. (Refer to Section 1.7.4.)
The TLP is a 4-column, ring pontoon semisubmersible type
vvith a displacement of 59,000 short tons. It is moored by 12
tendons, and the risers are vertical top tensioned risers. The l.7.3 MAJOR CONCLUSIONS ON TIME AND
FPSO is a 200,000 DWT, turret moored vessel. It is moored FREQUENCY DOMAIN COUPLED
by a symmetric 12-point chain/wire rope/chain mooring, and ANALYSIS FOR SINGLE VESSEL
the risers are catenary type. The semisubmersible is a 4-col- Coupling between floating vessel and slender bodies is
umn, ring pontoon floater with a displacement of 64,000 important for deepwater operations. Slender body
short tons. It is moored by a symmetric 16-point chain/wire parameters affecting global responses include stiff»
rope/chain mooring, and the risers are catenary type. All the ness, damping, current load, wave load, and inertial
moorings were designed for an offset limit of 5% and tension load. Some parameters can be neglected under various
limit of 60% under intact condition. The change in the num- conditions.
ber of slender bodies was achieved by changing the number
of risers. Analytical approaches addressing the coupling effects
can be classified in 3 categories: de-coupled, coupled,
Major conclusions from these studies are provided below.
It should be emphasized that the response values from the and semi-coupled analysis, which can be conducted in
parametric study were generated for the 4 specific vessels by time or frequ.ency domain. Coupled analysis is more
a frequency domain coupled analysis program to indicate superior to the others and therefore should be empha-
som.e trends. Any use of these values outside this context may sized in the future.
not be appropriate. Time domain coupled analysis can model. all. non~1in-
carities directly but is time consuming. The accuracy
I.7.2 MAJOR CONCLUSIONS ON GENERAL often depends on handling of a number of parameters
DESIGN PRACTICE such as time step, length and number of realizations. It
is suitable for final design check, calibration of fre-
- Modern design process requires accurate and efficient
quency domain approach, and some highly non-linear
analysis tools for global analysis. The industry has used
problems.
various approaches, often a combination of different
tools, to meet this requirement. This practice is
O
Frequency domain coupled analysis is efficient but
expected to continue as the industry moves into deeper requires linearization of non-linear slender body prop-
and deeper water. erties, which include line stretching, geometry change,
fluid load, and bottom effect. Industry experience indi-
~ Differences in global response predictions from differ-
cates that properly formulated and calibrated frequency
ent approaches are inevitable. The uncertainties due to
domain solutions can be accurate for most applications,
these differences have been addressed by conservative
especially for deepwater operations. It can meet both
analysis procedures and factor of safety in the strength
accuracy and efficiency requirements. The industry
design.
trend is increasing use of frequency domain coupled
~ Differences in standard deviation responses are much analysis.
more significant than extreme responses, which com- G
Most software developments are focused on time
bine mean and maximum dynamic responses. The dif-
domain coupled analysis software. The industry needs
ferences in standard deviation responses are magnified.
also reliable, user friendly, and affordable frequency
exponentially in fatigue analysis. Typically model tests
domain coupled analysis software.
are conducted for extreme seastates, which may not
provide sufficient data for verifying the lower fatigue G
All coupled analysis software, time domain or fre-
seastates. This adds more uncertainty to fatigue life quency domain, needs verification and calibration.
prediction, in addition to the uncertainties known to Because full scale measurement data are not available,
the industry due to T-N curve, factor of safety, and the DeepStar model test data for spar, TLP, and FPSO
method to combine wave frequency and low frequency and similar industry test data provide the best basis for
damages. bench marking. Software developers should fully use
 Desiou AND ANAL‘*{StS or Stariouxsssiuo Svsrsivis son Ft.OA'i'tN(3 Sruucruass 156

this resource. Field measurement program to obtain full l.7.5 EFFECT OF WATER DEPTH ON LOW
scale global response data should be considered in the FREQUENCY NATURAL PERIOD
future.
Natural period of a floating system is an important parame-
ter since it affects the number of low frequency cycles in a
I.7.4 EFFECT OF WATER DEPTH ON DYNAMIC
time domain simulation or model test record. In the past the
OFFSET AND TENSION industry typically used 3-hour duration, which yields approxi-
Most of the offshore industry experience was accumulated mately 1,000 wave frequency cycles and 100 low frequency
from relatively shallow water (below 3,000 ft) operations. cycles for shallow water systems. The 3-hour duration may
Only in the last few years floating production operations have result in significant under estimation of standard deviation and
moved to the deepwater region (up to 6,000 ft), but there is no maximum low frequency response when we move into deeper
and deeper water because the number of low frequency cycles
permanent units installed in the ultra deepwater region (up to
will be much less due to longer natural periods of deepwater
10,000 ft). There is a concern that we may face higher uncer-
systems. As show in Figures 1.3 and 1.4, translational natural
tainty in global response analysis when we move to deeper
periods increase with increasing water depth for all types of
and deeper water where the experience base is small and
floaters studied, and some floaters may have natural periods
model testing is difficult. To address this concern, the para»
exceeding 400 sec. in 10,000 ft water depth. The low fre-
metric study examined the effect of water depth to global quency yaw natural period is not an important parameter
response. The study found that the dynamic offset and tension except for FPSO where the low frequency sway-yaw natural
(max minus mean) in terms of total offset and tension period can be over 1000 second (Figure 1.5). The traditional
decreases with increasing water depth for all floaters studied approach of using 3-hour duration for time domain simulation
(Figure 1.1, 1.2). Since most uncertainties in global analysis or model testing will only provide ll to 27 low frequency
are related to dynamic response predictions, this trend implies cycles for natural periods of 400 to 1000 sec., which will not
the design uncertainties tends to decrease as water depth be adequate to capture the standard deviation and maximum
increases. The dynamic mooring line tension in terms of the responses. For the specific floaters studied, FPSO has the
total tension for FPSO appears to be less sensitive to change longest surge natural period whereas the semisubmersible the
in water depth than other structural types (Figure 1.2). shortest.

so —---——-------- -~ ------—-------—
FPSO

ll
II——A— SPAR
ll . HeavY Line: LooP Current ,1 1 U
50l ‘ SEI\/It
T T 1 T T Light Line = Hurricane ‘ .
__ J 1_ _. I, _..9%_.
TLP

FPSO
40 __ - ---------- ....W __ - - - -
/"‘“\

Q qu- SPAR

°/»a>To
V li
, I -5- SEW
-i It
S9 30
l
in.......-............................................
1
i- __................ .- 1 ; _ ______h ________________I
I —-re- TLP

Off
camc
Dy
20 ------------------ - -------------------- — .
H I I fi __i_A_r Ar at

10% . ... .

- ' ' ‘ .- ggggggg ‘. J Mwmwgw

0 A
v
soc tooo isoc 20cc 2500 3000 3500

Water Depth (m)

Figure 1.1-Dynamic Offset in Terms of Total Offset
 APi RECOMMENDED PRACTICE 28K

so ~ ~~~~~~~~—~————~--~~ ***********W ———h* ii iw ::r;"" WW "7 *;-pg-#1 _________ "1 ______ __

........i_.... FPSO

i ...._A__ SPAR

50 _ __ _ _ _ How/y Line = Loop Current SEW

i i
i i_ighti..ine = Hurricane
i 1 ————————————————*—— # ~ TLP

FPSG
/""~\
40 “
SPAR

‘\--»'
I N 10 i A U SEM!

°/son
Jota TLP
30?jjjjjjjj~ 777”""""""""~ ___"
, ".‘ . ________________________ ______W _ __ ____ _ _ A07 Q______________
‘ - “Q1 _
._ ____ _:_____________________________ i

Dynam
Ten
c 20
‘i W ‘ "{{{{{{{{{{{{in
1 H :.__, 1

10 i1_,*__,_ ________________ ________ _ __ i ____________________ '_ __ _____

__ _ _____ ~ ~
A i

D'- - A
ii i i
0 i ___ W 1
500 1000 1500 2000 2500 3000 3500
Water Depth (m)

Figure I.2—Dynamic Tension for l\/lost Loaded Mooring Line or Tendon

_ _ ________ W7 1

0 I —-I-— FPSO
. I:'11; %— SPAR
400

—'3'€—' TLP’
350 t i A iiiiiiiiiiiiiiiiiiiii M 0 _ M
i

800
/"--

-\.._,_/-

Cl
S60
-1
250 0 Z: *7;::;:"*" 3'1""W: ‘ ' ’* **' '
Per
o

200 ~-»»ooooooooooooooo~ »» ~ *
Surge
Natura
150 1 0 X 7;, ,,,,,

$00

A E
50 W11» —***** 0 ; 0

i
0 7’ _ W 7 _* ""'"' " " " " ' hf‘ —\~":;:::::;____:_::r _ 7 -_ 1

0 1000 2000 3000 4000

Water Depth (m)
GOM Hurricane

Figure i.3—Surge Natural Period Increases with Water Depth--Hurricane

 DESiGN ANS Answers or $TATiONi<EEPiNG Systems FOR Fi_OATiNG Srnucruaes

350 T,__—_;__—_—_ _________I;_______ _ iii _ ,

--Q-— FPSC1
_ _ _ _ _ _‘

--o- SPAR
300 0 ____ ii i ——-Q-- SE08
*-~)'é-=- TLP

250

4"“-~

SEC
~\_./

Ci 200 _ _ _ ___i___,,__i,_____,,,.,,_,,,_ 7* 7* 7* M __ _ __ _ ‘i _

Pare * U
j .1. i

150

Surge
Nature
L

100
P i . _ i _ _______ J W”
‘i

1‘

[
i . L i

E
i
i
50 ;i_- 0. . . -

0
O 1000 2000 3000 4000
Water Septh {m}
GOM Loop Current

Figure l.4—Surge Natural Period Increases with Water Depth--Loop Current

11 50 “F i ”””””””””””””””””””””””””””"F ”
‘ i

--Q-— Hurricane
i
—£&— Loop Currant
1050

950
.¢-"*-~

my

d
sec
QM) 850 11

-..-

T50 ;,: w¢. _4 ._4,

650

SwayFPSO
NatLFu-rYaawP
550 i iiiiiiiii ii
iiii ii iiii ii i i
T
i i 1 “4"‘?‘1a_

>4 ;4."-7 ‘. . . a
‘;il$4””;

450

300
0 i 000 2000 3000 4000
Water Depth (m)

Figure |.5—FPSO Low Frequency Sway-Yaw Mode Natural Period

158 API RECOMMENDED PRACTICE 2SK

l.7.6 EFFECT OF WATER DEPTH AND FLOATER ~ Slender bodies such as mooring lines and catenary ris-
TYPE ON DAMPING ers provide significant damping as a result of line
motions. An exception is TLP where the vertical ten-
Evaluation of low frequency damping is a difficult task for
dons and risers contribute little damping. This is also
global response analysis and commercial software often
true for spar’s vertical risers.
depends heavily on analyst’s experience an.d judgment to pro-
~ Hull viscous damping due to current drag is an impor-
vide a damping value. The parametric study included an
tant source of damping except for FPSO where the cur-
investigation on how different sources contribute to the low
rent drag is low.
frequency darnping and how damping changes with water
depth and type of offshore structures. The results for the hur-
ricane condition, where damping is important because of its l.7.7 EFFECT OF NUMBER OF SLENDER BODIES
high dynamic responses, are presented in Figure L6 and 1.7. ON DAMPING AND CURRENT LOAD
The results indicate the following trends: As shown in Figure 1.8, damping increases with increasing
~ For all four types of floaters, total damping increases number of risers for FPSO, SEMI and TLP. However, damp-
with increasing water depth, primarily due to the ing for SPAR is not affected by the number of risers because
increase in current drag on slender bodies. its vertical self standing risers contribute little damping. Most
' Semisubmersible is the most heavily damped, with its spar floating production units were designed using the semi-
damping level close to or above the critical damping. coupled time domain. approach, which neglects mooring and
Spar is the least damped, with total damping in the riser damping. To evaluate the accuracy of this approach,
range of 20% to 40% of critical. analysis was performed for the cases with and without line

140 ~ aaaaaaaaaaaaaa~ — P
Q-“— FPSO
, -Q-- SPAR
" R F r * " r r F E * f"“"“““'""”'"”'"""—"i 5 '-'|:]- SEMI

T i i '_'§!-' TLP
i i 1

. . . \. (:3 CI)

/"""*~

8.
“-
“-..

Cr't‘c
._
/2» .
so i
a a aas
i

Damp
°rig
e 60

LFSurg
.- .A
40 ~ P -W _. .

‘K .

20 _ aaaaaaaaaaaaaaaaaaaaaaaaa~ i - -

i
i .
ii

<3 .......who......... ...i...-.... - ______________..i.............. __......._.i___._.._l

500 1 500 2500 3500
Water Depth (m)
GOM Hurricane

Figure l.6-“Low Frequency Surge Damping---Hurricane

 Desiree AND /3‘\NAt.YSiS or Srarionxeseme Svsraras FOR Ftoarruo Sraucruees tee

120 nnnunnuut _______ __|n ___________________________________

1
____ *** WW ***** W--pnqqmmuq _"_ypnrW W W gr WW W W W
.
»p»m~q-gggq|_qg“w¢-gnu“-vg __________________________________________________________ __ _______________________ __1.
1
‘:7
1
:W ':_ nnnnnr_W;W_»uuuunu'

l E] Still water

100 Wind
/‘—'\
--

C53.
---
Current
...»-
i 1 WW f'j'.'f:...._. '
80 W Wave drift T
‘hum’

line motion
"_-

_,_____ ..'_,.r.n _,_ J »___

G). C)
Nonlinear
1, Q Total

CSurgeDamp
LF
%ng
r't
-55-C.)

l
j i _

§\-\ -.\
-.'-.‘-“-
_ _ ____________

"'::.::::....

._m,........_.
_______,_
.
‘

‘
__
— ————————W—W
_ _ _
——————— —— W —
.
__lr
‘
i

.
_7___7______ '-

'
‘
1
- -.

W =
-

20 l
I mm Ila"
..
0 iilllll , pg g i lmmmll
FPSO SPAR Sliivil TLP
Water Depth = 1828.8m, GOM Hurricane

Figure l.7—-Low Frequency Surge Damping Contributions in Hurricane

110 ~~~~~—--~~eeeeeeeeeee ----------— — —eee~—ee

FPSO
1 i _ __ . _. . I 1 i
SPAR
1 00 W W SEMI

‘-9? TLP k

90 ”T""” "T T T e iiiiiiiit is W‘W* PW” Tel;

80

‘W-III’

%Crgtca)
it

70 c WW W WW * ********WWW

60 W. ,,,,,,,,,,,,,,,,_, i _
SurgeDamp
i_Fn

50 _ i D u i ________________________________________________ ___ _

1
i
40 rrrrrrrrrrrrrrfie —e W e .......................................................W_______W "Hi ___.. _..F_;:w~~_~m__ ......... ........... .......... i
A i

30 ._-......e_.___. _...................... ; to .................................................

0 10 20 30 410
Number of Risers
Water Depth 1828.8m, Hurricane
Figure l.8—-Effect of Risers on Low Frequency Surge Damping
160 API Recomivisuoso PRACTICE 28K

darnpiing. As shown in Figure 1.9, the differences in dynamic are illustrated in 1.13 an.d 1.14, where the low frequency wind,
offset are small for all water depths and for both hurricane high and low frequency wave components are presented for
and loop current conditions. Similar trend was observed for the SD (standard deviation) dynamic vessel offset. For the
dynamic line tensions. This indicates that current spar analy- hurricane environment, low frequency wind is dominating for
sis approach should yield good approximations for spar all floating vessels except for FPSO, which is dominated by
design, assuming vertical self standing risers are used. This low frequency wave. For the loop current environment, all
may not be true if catenary risers are used. However, neglect- three components have significant contribution except for
ing line damping is always conservative. FPSO, which is again dominated by low frequency wave
The parametric study also investigated the impact of num- components.
ber of risers on slender body current load. As expected, cur- Figure 1.15 shows that the dynamic mooring line tension of
rent load increases with increasing number of risers, as shown the most loaded line is heavily dominated by wave frequency
in Figure I. l 0. The impact is highest for FPSO and the lowest component in the Hurricane condition for all floaters. In the
for spar. loop current condition, the dynamic line tension is almost
equally contributed by both low frequency and wave fre-
|.7.8 EFFECT OF SLENDER BODY DRAG quency motions of all floaters except for TLP. Since its heave
COEFFICIENT natural period is near the wave period, TLP is highly
As expected, the current drag coefficient of the slender impacted by wave frequency motions in both
bodies has a noticeable effect on global. response under the environments.
loop current condition (Figures 1.11 and 1.1.2). Increasing cur»- Contributions to the mean environmental load are also
rent drag coefflcient of slender bodies will generally increase investigated. In the Hurricane environment (Figure 1.16),
vessel offset and line tension due to the increase of current wind dominates all floating structures except for FPSO,
load from the slender bodies. Change of drag coefficient has which is dominated by mean wave drift force. It is apparent
no significant impact on offset and line tension under the hur- that FPSO is dominated by waves in both mean and dynamic
ricane environment. responses. In the Loop Current condition (Figure 1.17), cur-
rent load (hull and slender bodies) demonstrates its superior
I.7.9 EFFECT OF FLOATER TYPE ON dominance over the total mean environmental load across all
ENVIRONMENTAL LOADS types of floaters. The turret moored FFSO has the highest tine
drag load from current due to the large number of mooring
Different types of structures have different responses to line and catenary risers. SPAR has the highest current load on
wind, wave, and current excitations. The dynamic responses the hull.

__:jur WWWW WW '"lQ—uw=W'WWWWWWWWWWWWWWWWWWWWWWWWWW**

I I‘

|:| Line damping

,. ., _ - --

iiiiiiiiiiiiW T FT T W ii I No tine damping

SE3 E!
., I
E - - it ,,,,,,,,,,,,,,,___ ,,,,,_. _ _ __.__._ =_,,_____ _ _ll

I
10, ' . " \ _uI
c MaT%x-Moean
SPARDynam
Off

\\\R i , ,
I--»—---4

\\\ ”"”'l """""""""""T

ti
I [

I
T T T if i F‘"""""""""""W *1

&
Water depth (m) 914.4 I 1828.8 ' 304s 914.4 1 828.8 I 3048

Total offset (m) 25.74 44.06 52.26 43.01 90.4 148.58

I
Hurricane Loop Current
Figure |.9—-Effect of Neglecting L'ne Damping for SPAR
 E?ES¥GN AND /3\¥\§AE..YS¥S OF STAT¥ONi"(EEPiNG Svswms FOR FLOAETNG STRUCTURE$

1 8 11-;_________ 1
Base Case
Mean Load
(tonne)
4 sY C e r

rl I
i -—I— FPSO (305 r)
..¢.. seam (459 t)
14%?
—{:}-- SEM|(786’£)

—-)%~ TLF’(1008t)
»¥ 2 C.-__-:,_W_’_~~~ e e ,z,z;e:;~~E‘1,7:---:::;~,~ _ 1, WT. _:;~, W”:
r: 1
r
/""\

ad ‘r

Lo
._...... 1 1 _........uC_e,.4
C3

Mean
ase
__.83 __ 17 T: — _— — )——:——— — — __ _ _ _ _ _;T_—_—:7———:—— — ~—~

i
i
EfiedyCur
ncier rent{Drag
ir
Se (% ToteE3
6 Wmmmm,M______Mfl_CW I. "___ mm__u_____u____eefl W C Cu,‘ 1
I

4;
E
I
Z
E
E

2 _,_ ‘r

Q 1

O 10 20 30 40
Number of Risers
Water Depth 1828.8m, Hurricane

Figure l.10—-Effect of Risers on Current Load from Slender Bodies

W eeeeC 7_ an-an __ 1'7'"7T_TLli1'IIIfIiITTLf*****' '

E C<:i=Low
[:| Cdefvied
Z Nemwai Medium Lew High
Riserffenden é 0.6 1.5 l Cd=§-iigh
Wire Rope 1 .2 0.7 1.8
Chain 2.45 1.4 3.7
11 __

r""-»

6 \ M
%W£)
et J __ C

%
r

' /’_,// {J

MexOffs /.

%
/1» 1 __,_._... \\
/ _ \x
\\
\\
\

/"
¢

\
4444p_

=2:iEII2;| r:'2: //~ >'E:'!-=-=:'.z|:5

E ;:21:2;s:e.2\!.-5
:::5! iE21 2?;2iii2%
==
:aiiif
- E5:
iEl
5:55
= :zs;
::
:..
-=.

.\\\\\\ W\\\\§
lil:I

23. 55-i i:H3i:i_-'e:|

=55 iz %_:
25%i \\_\\\
\
‘Cir\
//j
________ W ___ /

FPSO SRR SEMI TLP

Water Depth = 1828.8m, Loop Current

Figure l.11--Effect of Current Drag Coefficient on Vessel Offset--“Loop Current

4; 6 2 A D, R E C O M M E N D E D HP RA C T m E 2 S K
_
432
65
8?
O
GOO
UO
0O

:13 ‘“5“\‘M“_FU‘M““‘IM_f CC dd L Wlw

M
2“M

I O__ /M
%‘///V
O
_KHM/_II2;/;$§ AIEfi
,/f//7
_/I/ av 1A;k W __
__
_ 4;
H‘A1\é _

_:
T /_A/I
/ “M/V",
N”!/“V?/N1_¢’J§_i%§i%/WM/hfwy/_7y
___

Eo@5602
mwcmwt flmmoé
X5_§2&
fi_jwgtFFt :
;i_
:5
:§:;
_
if
:_g_3
_
___fl
_ ___ M:M_M_W_W__M_M_M_ _
___H‘m_“h“__":_@_"_
it
Tm m
”um”:
;_"Wm: __M
mMWm_ _“m_HMH"”HH_”'h_H _M_ _“ _ H______"_mh\_"m__"__'__:_m_:fl__"__
_:“L_
%;2
_
WM
_;__S

gm:W"__ mm_
__ _m:__ _;“hm
:2“m
MHm
m“___
‘F:L
__
______
____
: m _ h_ W_
__
__
_“_
_
_m
___:_Hm____
_____
v_"w_““_m___"
_nm:_
hm _U///7 Wm fi
/ /H

LM 5%:;
T
_m
__:__
__
_____
_m_
g______i_
______ _ m_M_i "_ u_ _

3 M:

F
_P_ iM
S R E W TL P
B S H0 n n 9 9O0 1 J 22 1 1. 34 8 B’ 9 O
I

c‘
_ mu
w U mm 12 "___MM d C
“H U r ___‘ 6 m D r a g C O 6“H m ‘E H O fi L _m 6 T t w S m fi % O0 p C U __fie m

1 42
8
OU3O
6
O
GON‘\
_
A% ___:IT
___IWmm_TWw
3:Im:_m__W:_5mM1;m:M:
__mMm___m
m___:
W;:W__m: WWWW_:_"mWm_m_m_" m
I: __ _g___ __ _ _ _ _i

1\
NJ1L\ 1\ \
ZED LLHFFFiWWWWNW

Wk%%ZW/%&%?7/;/
i A§
_WQA 6
_ ji% N

gé//
_I
/ ’/
,V/
/IK //
//I/7
//I/M1?//_
//J//_
__/I
/_
//’///_‘ _ //F7
/4
/
/_
X’,
_X’
X
Iix
’f_/I
I
‘ __W
//I
/J/J‘,/
/_
//1’
/Z
J?_ _

mwgmwfig
gm
wwwxo awmwgm
fig ‘l_
AM
AQ__:“iAEA‘“

Tm Q S U rg 9 F P SO S P6
AR S E8 M2 TLP
S D O 5SMm W 4 B2 1 9 1 2 26
ME
_

ya’ U 8 9_ ‘E ,8
__ _1_ _ _ 8 2 8 B m _L G 9 Q C 8 H 8 fi ‘Z
T‘W tR ‘Mq“ 1 P 6m 6n age t mY n O t my q U 1w O Q 0 m G w p h % 6 d
_[ M E H C G‘

H 9 U r9 k
1_ D Y H 8 m m OanS 9 C O H __m U m H S _m O M H U rfim 8 H 8 It tr it
 i

Desieu AND A1\11111.Ys1s OF SmT101\11<EEP1r~1e Svstems FOR Ftomue Sreucwess 6

30 77*****77 F 7 7 77 77 1 W
LF Wind
LF Wave
-,..................i 1

U HF Wave
ifi
1
60 1
E
'1
W, _ T7 W 1 1

f.'I.'Jl§Il'.IL'___T'_'.Z'I.. 1 /’
_ Iz /

/
,
7* __...WW.
...-------......._\..

//F /
r I -.:'."::_'.::--—::
41>CD __._.___.
so) 1 1

/
’/
1/
1/ / /
1/
if’// \\\
\‘Cb
\\\\\\
\\\‘\
\\\\\
_-
//’ . '.1.";,",.".;;;;.;;:.;;;.i
/'
/
,1
8DOffset
11111
/

H 1 /
/ H11 ‘1 /
, 1, / ._'..._..__;'_LT_T
\"t
1\\;~;‘“"C‘;§ "_iW1‘\"§\X<fi§\
/ .....__._....v........ 1
11 1
1 1’ ............______
Surge T%otaSurge 20% 7 111
i 11

‘-1-1.-*
,7 .., ._____........_ , ....u_. 1
2/ — M
1 , /:7~:/:/
f//// <’/
1 ‘ / ' ...................
1 ///*' /,/
1 1/ / _.__.__.._..u.,.
1 |/A 7" 7-_-
"T"
1 1
,/
/.»

,
//’
,/
1..._.._........1
i 1 #77-77%
// I. .....__..._ __.._,..... 1 1
i 1
/ // / iv 1‘11 1i
C1 ,7 / \ 3
1’ 4/ ---W---~-———
’ ".'_"__'“"':.“_T'
11
1 V
3 1

/"
\
\\\\‘\\iI:1.\ :25“:
....................._....._.
1
;
_../ »“lT----_—- 1
i

1111\
r_..- .//,
/
.1’ /’ 1‘ I
//
_ __ 1,

W\1
/ -GT7-mi /

if
1
1/, _
/ _. \\
W1‘\
. \\'1\_\\
‘\‘\§=\_\
X\\_\-
\\\‘1x .,.\.....____...._... /
S?\\1\I
\?\
\j\E?‘\l1*
:\1
1\-\\\
\\\1
\ T

Total Surge FPSO SPAR SEMI TLP

SD Offset (m) 1.63 0.28 0.58 0.6‘?
Water Depth = t828.8m, Loop Current
*The tetai percentage may net. eqeei to 186 due te phase differeztce.

Figure |.14--Dynamic Offset Contributions in GOM Loop Current

120
[:1 LF
i
w1=
1
i1_
K30 77 77777 77;7i7;7_7_77_77l...7 7 7 7 7 7
i
i 1

80

n /"'~\

SD
H80
Wee 1

ii
11;1~1i
i
E_E
'-I

SDTeT%ote
"'1--1..-*

i ‘

40 ...... ......--‘i

'1.
I i
i
1
L 1
i
ti
1 1_§E
1.z1;_i~2§%"1
1: 1 111 I1

t 1
29
i i1

13 [
gi
1,1
11=5.i
ii'i Z'-i
5U7
F1
11!_
_I
1?"
‘ it
i
i
i
1
1
'“'“""'1
1
1
.........1 _ 1 ‘I—‘~1c:_: 4

:15? if
{E51 ii
151£:'i
l\

;lt iii

0 ___ —"“""—e“—“—‘¥"'—‘”l ,1 . _ _ _ _ __ _____ __ _ in .

Total SD FPS O SPAR SEMI TLP FPSO SF’AF% SEMI TLP i

Tension (KN) 330 281 471 729 130 "I" 5 1 34 489
Hurricane Loop Current
Water Depth I. t828.8m
*Th.e tetal percentage may not equal to .100 due to phase difference.

Figure |.15—-Dynamic Tension Contributions for the Most Loaded Line

 API Reoommenoeo PRACTICE 28K

70 1-an

_ 1 Wind
‘V Wave Drift
60 5III
5 1 1I1 '1I 1i1 1
'51)
7!

1 1:ggg 1 ol;Lo
I
iii
ii" Current (Vessel)

56 V
I:
/
2

F
1
I
I
11
?%§§ I I§§§§
{:1 rrr
I iéié
I§§§§ W“““~~1~
[I] @“”@“‘I*@@

U3 I /
M I
I"
F-‘~
'\-|-/
4OI1 1

"cs ;
1 1
.................... 1 /,/,
O
CU L //’

...]

‘E
I §%s*2zi¢
/5/47»
I171
CI
so //,¢’
06
ED
2 I IT.:.';;7;:" 1/
P"tt'ri'.'::."".:t"; ;// /
I

20‘ I
// I
1 it If

%a
E
I
1
k

.........................~... /

**;;i7§§§§ / ) 11'1
111
1 1§i1I1111*111

"Iota! Mean
O W ’ --

FPSO
\§, - - "7; I ,,,T:'it:':i,,, one-
1

SPAR
. N%%I
%“” / / ;1(§§§§§§§
Load (tonne) 306 459 786 1,008
Water Depth = 1828 8m, Hurricane
That t Ip tg 31 t<,qualto1OOt’o1"1~1P‘§()Iet t I ea n tiheat

Figure I116--I\/lean Environmental Loads II"! GOM Hurricane

90
Wind
so1i or -pwuuuunu-I-n‘n
iii Wave Drift
Currenflvessel)
79 1--»-7777-77w U Current (Lines)
1-an-1 __ _______:i::1;:;:i:;**** "

Q‘
so ) l—
{ti (1
*6
I-»
1:250 ///////
<11
U 1
' 1

-I
LU 40)
//// DI
I
1 1
I
1

2 30 /% I
1
--I
I
1

20 //%/ 1
1
I
I
I
I
I
1
I@1
1

‘I
I 1 1
I E I I
‘Q .1 2 I

IF“ ease
ms
i I‘ I

—1
2
1 i

I111
CD '~1iIII‘
iIII. 1 i§.'i.‘¢#' -s. §§iE,sos§§§§
Total Mean
Load (tonne) 1O PAFI
1,848
SEMI
1,579
Water Depth e 1828.8m, Hurricane
TLP
‘I 883

Ti e total percentage may not equal to 100 for FPSO due to its non-oo1iI1e'n' env'iron1"11enta1 I dg"

Figure I.17--Mean Environmental Loads In GOM Loop Current

 Dsssou AND ANAi_YSiS or $TAT¥O¥\3KEEPtNG Svsrsras sou Ftoariue Sraucruaas 165

l.7.10 SENSITIVITY OF WIND SPECTRUM - Response to the NPD spectrum is always higher than
the response to the API upper bound spectrum. The dif-
As discussed. in Appendix. B, currently the API and NPD ference tends to increase with increasing water depth
spectrum are commordy used by the offshore industry. The and natural period.
API spectrum, which was published in earlier editions of API - The trends are similar for the wind only and the wind/
RP 2A, has much smaller empirical data base than the NPD wave cases except for the FPSO, which shows almost
spectrum. The uncertainty of the API spectrum is addressed no difference between the NPD, API upper and lower
through specifying a range instead of a single value for the bound spectrum when the total offsets due to wind/’
dimensionless peak frequency. This results in a spectrum wave are plotted. This is due to the fact that FPSO is
defined by upper and lower bound values. In the latest edition dominated by wave dynamics, and wind dynamics is
of API RP 2A, the API spectrum was replaced by the NPD insignificant.
spectrum, which was also specified by the draft ISO standard. ~ Since the API spectrum provides only bounds, the
As illustrated in Figure L18, the NPD curve was fitted. to the response prediction often depends on the judgment of
data with lower periods and extrapolated to higher periods, the analyst. Many analysts select a middle value
resulting in potential over estimation of wind energy as the between the upper and the lower bound, this may sig-
period becomes longer. Since low frequency translational nat- nificantly under estimate the response.
ural periods tend to increase with increasing water depth, a - Since the natural periods are in a range of l00—-400 sec-
question was raised whether NPD spectrum would be too onds for all 4 structures in 3 water depths, the over esti-
conservative for ultra deepwater operations. mation by the NPD spectrum appears not too serious,
To answer this question, three spectra (i.e., API lower especially when the maximum response (mean plus
bound, API upper bound and NPD wind spectra) were inves- maximum dynamic) is considered.
tigated for the wind dominating hurricane environment for 4
structures and 3 water depths. The SD (standard d1eviation) l.7.11 EFFECT OF WAVE HEIGHT AND WAVE
PERIOD
vessel offsets due to wind only are "plotted in Figure I. 19, and
the total SD vessel offsets due to wind/wave are plotted in As shown in Figure II.2l, increase i.n wave height always
Figure 1.20. These plots show the following trends: results in increase of dynamic response for all floaters.

uerm, tom) E 41 .1 s4 mls (so knots)

Period, T = “I/f, seconds
1 E—5 1 E—4 1 E~3 1 E—2 I E~I 1 E+0
5909 *********************** " W ******** ********* ************ ******** 1

‘; 1 1 1iiiiiiiif 1
I 1 1 I 1 1 I 1
I I 1
___.-- 1 I I _
40°‘? ****" I 1 I -Q-— API Upper bound (aIpha=0.0t) I 1 1
I I I —><—- APILowerb0und(aIpha=O.t) 1 1
1 I 1 I 1 I I1 1I 1I 1 11
I
1
1‘
11 1
1
1 1
1
1
I
1 I
1
1
1
1
I
I
1
1
1
1
I
I
1
1 1
I1
1
I1 I
1
1
1
I I 1 1 I 1 1I 1 1 1 I 1 I 1 I I I
/'"- II1 I I I I 11 ‘ I I I 1 1 I I 1’ I I I I
I I. I1 I ' I I I- II1 1
1 t 1
1 1 1 1 I 1 1 1 1{- -1 1 1 1 I1
2/HZ
rn/S
-...../ soon 1'
77 __ 1 71771_1 1 1
_ 1 1 1
I II
1
1 1 1
111II.1I1i
11 1
I 1

H 1 1% 1 1
1 7 1_
1.111
1 1

I I
I 1

I
1
11iI1 ;

11
so I
E
1
I

1
1
I11‘1
1....1_.1
1
I

I
WI-M
“--...r
1
1
I
1
1
1 1
1
11
1
1
1
1
-_
1
1 1 11 I I I
1
I 1 1

I 1 III I1 I1
I
I 1 1 I 1
1I II I 11 1I 1;1
I I

coo 1:I 1 111 1 11 - is 1 77777 1 1 I 77 1 _ I1 1 .1

1 1 1

Spectrum F0
1 1 I I I 1
11 1I 7
11,
1I I 1 1
I
i 1 1 1 I I1 1 I 1- 1 Ii 1
dS 1 1 I
I II1 I
I1 II I11I I
I I I1
I 1 I
1 1
I
1
I1
1;
11
1
I
1
II
1
I I I1
I1 I I
W'n
I1 I1
1
I
1 I
I 1‘ I
1 1 1.1 11 1 111

1 7. ................ ..' 7 . .
II I I
I I 1 I1
I I‘ 1
I 1 1 I I I 1I I I
I I I I I I I 1 I I 1 1
I
1t
‘E
1~
I
1
I
1
I
1
I
1
I I
1
I
I
1
1
I
I I 1 I 1 1 I I 1
I I I
i I 1 1I I‘ III I1 I 1I1 11 I1 I1
1 1 11I111111 1 II1 II1111 I _ 1 I I
I

I I I " 0I I I 1 '1 1 I I - I I '11.-"~.1.= .1-'.1...a.. 1 11.1-111A. 4. 14.. 1.,‘ .s...4- 11- I'IIII.II Ii5f"IIII “III 1 I I I
0 ,,,,,,, ,,,,,1,............. j f'j.j.j1 1. .;;j ‘j 1
IE-5 1E-4 tE-3 tE~2 1E-1 1E+0
Frequency, I e IIT, Hz
Figure I.18--Difference In Wind Spectra Increases with Increasing Natural Period
 0 API RECOMMENDED PRACTICE ZSK

___ """"'"""" __;;**""'W"'""""""""'-~r"--fun»;-r—————i************'mm__'_" """""'pm ' 'mnmum|qi————u-_Lr:r:1mrunnr"""""""'pur;;__ —1¢—————————————————"'~W

6 444 4 4 API Lower Bound -4 444jjjjjjjjj4‘iiiiiiiiii

....‘

API Upper Bound

I {{{{{ Q{{ {NPD
{ { { W H {{W W — l _

‘"»n\uIuI»\u|r)uunn-10:10:11

Wrimd
___________ 4 ____ __ __ __ ___4 _ ____ ________

-f7"""""""' ’ ’ if ‘ f i We ~ ** """""""""""" " ""111 '1"I< ...“...

I Ii
'!s§!:I'
'ie1?§§%
l;i2P='is
§1|;5421.545
LFSDOfise
om 4 444 _ _______4 iiiiiiii 44 4
i ‘E I ......... W 3

_:_T_-__ ........... .-.4._.

t
5 :__
I :1ii1..T...
.'.'.'If
11 ‘
_ _
—”""“
............
lflmll
'1'“ Q

Waterdepth (m)
0 814.4 1828.8 8048
9 8I I I I H— Z
814.4 1828.8 8048 i
_4 WWW
28.8 8048 5 8 288
Surge Tn (880) 242 828 408 247 882 885 102 151 200 142 181
1=1===8o 8848 821111 TLP
Figure |.19—Sensitivity of Wind Spectrum on SD Offset from Wind-~H urrioéne Condition

_[1_' ""Iu-annual __ 1ut________7_ 7

API Lower Bound 9

_ I
1
API Upper Bound _ _ _ E 7 _

NPD i
IT}

fise 9\u

..1;
15gal;
Toe.SDO
mm
m §\\\ iii '1
i1':4
\§
Mil!
4»-.4

i 9 1 .W
Eflfl lfllfl
—1 —‘ 111%
:1u»u w_

= 1 -‘ 3
mini
.§- HEM“
£1k§\\8 ......
Water depth (m) 914.4 1828.8 8048 K<0 t5
um 3948 914.4 1828.8 3948 914.4 1828 8
Surge Tn (sec) 242 828 409 247 332 395 102 1 51 ZOO 142 181

FPSO SPAR 1 SEMI A TLP

Figure |.20—-~Sensitivity of Wind Spectrum on Total SD Offset-“Hurricane Condition

 Dasion AND ANALYS¥S or 3TAT¥C}Ni<EEPiNG Srsrsrvis FOR Promise Srnucmass 167

50 ‘“Wii'L'i""'7""""""' "_'i"_‘_‘_ _ - I __ ___i"' "" '1""""" "1i‘_".‘T?i"‘_‘_‘i““"""1""“"$-T_‘_TIL7”"""5""” """' IiXCil ‘l""""""'""""" "' i7"""""""" ‘Q’

1 i i 1

45 - 8 8 8 8 8 8 8 8 8 8 ---44 4 4 4 4 4 4 4 . . 1-18 = 11111 ,‘

i _ _p

. 8 8 .1 D-— t-is=t2ro

1 - 8888888888 i
l
l
1g 85 444444444444444444444444444444444- \ 4 - .444 4444444444- 4 444444444-4. .. 1;. 41
Toa 5 _ .::;;;;:.:.:; ':.".:.;;i;:;;::. 1 ::;:;:::;:. 1
:2 1 _ _ l _ 11__ , . __ 1l:,1 _ _

ms \\\\
1 L 1
1 ___ _

28 ,1 :
\\§ "\‘_\‘“\‘_
it i 1; 111_1 __.___________4 —:_fl.____g__...1
.8, . .

---
‘

1 2
\\\
.
:
_—.-l.
-T-——-—
i
i

,
. l
:
-
\\ I—----— I
____________11____.... 1 ___. —e 1 ""W _ 1 1 : '.';;:;;;;;;j_ 7"" -
IM 1 1 2" '2 ' """""""~

Dyoar
Mcn
\‘ \
11111 81 5 *1 ct E

FPSO s\\ , -ms. . EM!

PAR TLP FPSO st-ma 1. meme. TLP 1
1 Ctitset Tension
4

Water Depth = 1828.882, Hurricane

Figure l.21--Sensitivity of Wave Height for Hurricane Condition

Increase in wave peak period, however, does not show the L3 DeepStar V Project, CTR 5401-2, “Guideline for
same consistent trend. For example response of FPSO Global Performance and Verification of DeepWater Struc-
decreases with increasing wave peak period, as shown in Fig»- tures,” DNV, March 2003.
ure 1.22.
I.4 DeepStar V Project, CTR 5401-3, “Global Perfor-
l.7.12 POLYESTER MOORING mance of DeepWater Structures-Post Model Test Study,”
OTRC, April 2003.
The parametric study investigated also the impact of using
polyester mooring instead of steel mooring. Study results 1.5 DeepStar V Project, CTR 5401-4, “Global Perfor-
indicate that in addition to generally lower maximum tension mance of Deepwater Structures-FPSO Analysis,” MARIN,
and offset, polyester mooring yields much lower wave fre- April 2003.
quency tension than steel mooring (Figure L23). The much
lower wave frequency tension may reduce fatigue damage by I.6 DeepStar V Project, CTR 5401-5, “Global Perfor-
an order of magnitude. This can be a good choice for West mance of DeepWater Structures»-FPSO Analysis-Marintek,”
Africa or GOM spar operations where fatigue d1ue to swell or April 2003.
VIM can be a dominating design factor.
1.7 DeepStar V Project, CTR 5401-6, “Global Perfor-
Another observation is polyester lines provide mach lower
mance of Deepwater Structures-FPSO Analysis-FMC
damping than steel mooring lines because of their lack of cat-
enary change under wave frequency motions (Figure 1.24). SOFEC,” April 2003.
1.8 DeepStar V Project, CTR 5401-7, “Global Perfor-
l.8 References mance of DeepWater Structures-FPSO Analysis-SBM IMO-
L1 DeepStar IV Project, DeepStar CTR 440113, Deep- DCO,” April 2003.
Star Theme Structure Specifications, Sept. 8, l999.
L9 DeepStar V Project, CTR 540l-8, “Global Perfon
I.2 DeepStar V Project, CTR 5401-l, “Coupled Analy- mance of DeepWater Structures-Spar Analysis—Tecl1nip Cof»
sis Theme Structures Model Tests,” MARIN, April 2003. lexip,” April 2003.
V
A
E
w
fig5E
OMEHNEMQ gS_§S_V A H RE C U M M E N P 2S K

32
154
22
4 0O
00
GO
“‘1‘ “_‘1‘ ‘_1‘“ ‘“““‘‘ ‘ii1‘ ¥‘ ‘“I

5BMW
5
5D
UNaw
5_
my
41; 433 W__
1‘it“
“ i “, 7éFj
W_
O%1; Wjg WI
MINI:
_PHg
“_ M_R
"f:M_Z_
_,_ SN\I\ N \5Q

h
WVq1A,W
2W_
mamq %y
%__ I
k_ jA HQ_q i
M
_AW
AWSi‘>qAW,‘1\,_\_qA,
nM_j qpfiiikiim
5%g%piiq,q1! qQ W
___
y
Q_W
W
_

%
w
pnjW
jgq_W

p Wp7%
WMFH __
_ H_an
__ fi
WrWmg
‘m
Fm
_ A
W_ _ MW
_WQjW BW
H Wm mWcum mTT;mMLgw_UW;um_mm:_WM€:§m£__wmm_mmg_mj_
W_:m MH__m__"
_W_MH_:mM“_mm“HM_d_

“ ‘e
‘,1 i
“_z_‘ ,‘,,,,‘_,_t
W,““ ‘“ 1

2 HWr M _m_

% ~N P: _

A;_V
hWWWAmW

/
_S

i
G nu nu 8 92 2 ___}2 T 2 ‘___, G 3 5 922 *2’ 2 _____ 2 J); 635

F _9 23 JV m p W F Q YT f a
 DESEGN AND ANALYSIS or STATtONKEEPiNG Svsrar/is FOR Froartne Sraocroaas 169

7'0 --------------jj j j j j j j j j j j j j j j j j j j j j j j j j j - \
; E Polyester
i i I Ste-at
so - - i

\\§
\\
50- -- - -
/""-a

\\§
PE
1.... t

40 ,--- ----

30 - - fffffif — — ----+

20 __ __

i esI
it %
gsE Hurricane Water Depth 2 1828.8!“
i_r Loop Current
.,|
Figure l.24--Comparison of Line Damping for Steel and Polyester Mooring

I.10 DeepStar V Project, CTR 5401-9, “Global Perfor- posium, Deepwater Mooring Systems: Concepts, Design,
mance of DeepWater Structures-Spar Analysis-OTRC,” Analysis and Materials, Oct. 2-3, 2003, Houston, Texas.
April 2003.
1.14 Garret, D.L., et al, “Global Performance of Floating
L11 DeepStar V Project, CTR 5401-10, “Global Perfor- Production Systems, Stress Engineering,” OTC 14230, Off-
mance of DeepWater Structures-TLP Analysis-ABB,” April shore Technology Conference, 2002, Houston, Texas.
2003. DeepStar V Project, CTR 5401-ll, “Global Perfor-
L15 Garret, et al, "Mooring~And~Riser-Induced. Damping
mance of DeepWater Structures-TLP Analysis-Marintek,”
in Fatigue Seastates, Stress Engineering,” OMAE2002-
April 2003.
28550, Proceedings of OMAE’02, 21“ International Confer»
1.12 DeepStar VI Project, CTR 6401A, “Guideline Anal- ence on Offshore Mechanics and Arctic Engineering, Jone
ysis and Verification-G"aideline Development,” Noble Den- 2002, Oslo, Norway.
ton, July 2003.
1.16 Garret, et al, “Integrated Design of Risers and Moor-
I.13 Garret, D..L., “Coupled Analysis of Floating Produc- ings,” Stress Engineering, 2003 International Symposium,
tion Systems,” Stress Engineering, 2003 International Sym- Deepwater Mooring Systems: Concepts, Design, Analysis
and Materials, Oct. 2»-3, 2003, Houston, Texas.
 APPENDIX J—-MOORING STRENGTH AND FATIGUE ANALYSIS EXAMPLES

J.1 Strength Analysis Example b. 31/2 in. K4 chain 5.100 rt outboard, break test load z 1838
The following problem illustrates the procedures for the kips‘
analysis of a mooring system using three analysis methods: ¢_ Initial tension m 230 kipS_

3' Quasifiatic ana]ySiS' d. Mooring line description.

b. Time domain dynamic analysis. 1' Diameter X 31/2 in
c. Frequency domain dynamic analysis. 2‘ Elastic Stratch (AB) K 1234 lbs/ft
The example is intended to illustrate the principal steps in 3_ Weight in air 3 1072 lbs/ft
the analysis and to give guidance particularly on the dynamic
analysis procedure. The method described here is not unique 4. Friction coefficient K 1.0 (with mudline)
and it may be necessary to modify some of the steps in the 5. Break test load 1 1,838 kips
procedure to accommodate specific software. 6. Line mass E 3.84 Slugs/ft

4.1.1 MOORING svsnzm DESCRIPTION '7» Tddsddtidl ddddd ness I 025 Shed/ft
The system to be analyzed is: 8. Normal added mass K 0.51 slugsfft

a. A semi-submersible with a 10 point, 36° symmetric pat- 9. Drag coefficient I 1.2

tern, as shown in Figure J. 1. 10. Drag diameter I 7 in. (2 5 nominal diameter)

6‘
0- an
""a<>,,)\\/ 1. 10.

2 Q Q 9

<04
Oébz
/2;,6‘
3 . . . 5‘ F .. 8

Ed

.. ..

C I
5 6

Figure .J.1—|\/Iooring Configuration

171
i72 API Racormuaanso Pnacnce 28K

The mooring properties for chain include provision for a reference point, usually the vessel’s center of gravity, can be
tangential added mass. For wire or synthetic ropes, no tan- used to define the motions.
gential added mass is required. The nominal drag diameter is The low-frequency motions can be computed from hydro-
increased by a factor of 2 for chain. For wire rope or synthetic dynamic computer programs, model test data, or design
rope this is not required curves. The computed root mean square (rms) low-frequency
motion for this example is 0.97 ft.
J.1.2 THE ENVIRONMENT The vessel motions at the reference point m.ust be trans-
The design example environrnent is as follows: formed to the end or fairlead of the line to be analyzed. The
procedure required varies with the type of analysis.
a. Water depth K 1,233 ft
b. Significant wave height e 55.8 ft J.1.4.1 Quasi-Static Analysis

c. Peak spectral period K 17.49 sec In a quasi-static analysis, only surge in the quartering
direction is considered. Heave is ignored. The vessel motion
d. JONSWAP spectrum with a peakedness factor 3.3 RAOs and phases can be transformed into the quartering
e. Wind velocity (1 min.) K 1.00 kt direction and convoluted over the sea-state spectrum to pro-
duce rrns line end motion. The computed wave frequency rms
f. Surface current velocity 1* 3.1 kt motion in this example was 8.6 ft. To establish the dominant
frequency response, compare the maximum wave frequency
g. Quartering direction (See Figure J.1)
motion with the maximum low-frequency motion as follows:
h. Design storm duration K 3 hours
maximum wave frequency motion:
i. Wind, wave, current colinear
3.72 X 8.6 e 32.0 ft
Wind loading on the vessel can be accounted for in two
different procedures: maximum low-frequency motion:
a. Wind load is considered to be applied statically to the ves-
sel. In this procedure the one minute average wind speed 3.03 X 0.97 K 2.9 ft
is applied on the vessel as a static load.
where the factors 3.72, 3.03 represent a 1 in 1,000 and l in
b. The dynamic effects of wi.nd are considered by combining 100 wave maximum respectively, typical for 3-hour storm
a steady wind force with a fluctuating wind component. conditions. Wave frequency dominates. Hence the design
The one-hour average wind speed is applied statically to condition is:
the vessel. A wind gust velocity spectrum is defined.
The approach described in item a has been used in the maximum wave frequency motion + significant
present example. low-frequency motion K 32.0 + 2 X 0.97 I 33.94 ft

J.1.3 MEAN LOAD COMPUTATION Note the factors 3.72 and 3.03 are approximate values typically used
for a 3-hour storm. More precise values can be obtained by Equa-
The mean loads can be derived from model test data or com- tions 5.5 to 5.8.
puted. The computed mean loads on the vessel are as follows:
Wind (1 minute average) 680 kips J.1.4.2 Frequency Domain Analysis
Current 378 kips
Steady wave drift 70 kips To compute line end motions in the frequency domain, the
Total mean load 1,128 kips vessel RAOs and phases in the six degrees of freedom must
be translated to the fairlead location. In general, only the
J.1.4 VESSEL/FAIRLEAD MOTIONS motions in the plane of the line are of interest. The line end
motions in the horizontal and vertical directions in the plane
Both wave frequency and low-frequency motions are of the line, and the phases, between them are computed for
required in all analyses. In general the response amplitude each frequency. It is extremely important to retain all phase
operators, and phases, in the three linear directions (heave, information to this point as the dynamic behavior of the line
surge, sway) and the three angular directions (roll, pitch, yaw) is heavily influenced by the tangential motion or stretching of
must be derived. The derivation of these data requires hydro- the line. The line end motions used in this example are given
dynamic computer programs or model test data. Any suitable in complex form in Table J.1 for each frequency. The stan-
 fission AND ANALYSIS or Srarionxassimo Svsrsias son Froariwe Sraucruaas 173

dard RAO value is the square root of the sum of the squares At least 50 frequencies are required to develop a realistic
of the real and imaginary points, in the case of each motion. time series. Care should be taken in generating the wave ele-
vation, that the time series does not repeat itself prematurely.
J.1.4.3 Time Domain Analysis This is normally achieved by using varied frequency spacing.
The fairlead motion RAOs of Table J.1 are used to transform
Time domain analysis requires a further step beyond fre~
the wave elevation time series into a horizontal and vertical
quency domain analysis. A time history of the motions is
motion at the line end, in the plane of the line. Because phas-
required. The following procedure was used in this example:
ing has been properly maintained in Table J.1, the resulting
a. A three hour time series, representing the wave elevation horizontal and vertical motions are correctly phased.
was generated from the sea-state spectrum. The procedure The analysis of a full three hour time series is generally not
is illustrated in Figure J.2. practical in standard design practice. A simplified procedure
is illustrated in Figure J.3. The maximum tangential motion is
computed. A 120-second segment is selected centered on the
h(r) I EA}-cos((oJ,-1+ <|)J,-) maximum tangential motion and the time domain analysis is
performed for the 120-second segment.
where

Aj a ./2.5(toJ,)r&co

rigj K Random Phase [0,2rc],

S(coj) =1 Spectral Density at Frequency coy.

Table J.1—Fair|ead Motion Complex RAO

Line No. Z l
Line Heading = 54.00 Degrees
Wave Heading From +X = -~ 45.00 Degrees
Line End Motions with X in Line from Fairlead to Anchor and Z Vertical
***Location (X m ll0.00 Y I 104.00 Z e -35.00)***
Wave Frequency Low Frequency X Real X lmag Z Real Z Imag
(Rad/Sec) (Sec) (AMP/AMP) (AMP/AMP) (AMPMMP) (AMP/AMP)
0.20 31.42 ~0.0036 M9928 1.0976 0.0518
».26 24.32 L9085 0.7092 40.2909
>.32 19.84 L8496 0.5965 0.2052
L37 16.76 £7840 0.5820 0.3278
>.43 .1450 ~ 0 120 C‘_*IJL‘_J(Q"JC_‘JC_7'.7032
:T) 0.5158 0.3992
£49
F6
6'F1C)
Cl)
737'" 1.2.78 sis C1CBC1
sss Pee 4\OcmU1
@_1 CD
Q3 C3
QT) 1> [\-J
L»)rP-—-~* 0.6080 0.4273 0.4386
0.55 1.1.42 - 152 £5014 0.3289 0.4419
0.61 10.33 -»- 1 0.3899 I.23l5 0.4110
.67 9.42 . C15:9(:3 CDU0
gistpm.{J}
-§ L28 '15 I.l452 P3489
.72 8.67 40.1115 0.1849 .0773 M2685
L78 8.02 -— I.0083 0.0997 .0249
C3C-§C‘)
C_J L 1788
I.84 7.47 -» I.0048 0.0542 0.0090 L1050
.90
C-3
{'5
(:3
{:3cl‘)L 6.98 .0014
CL;
Q3Q)
c_)c—3c_3 7C"0.0l2l
7‘T—) 7)C“; -0.004"/' L0340
6:1
C5
A L-‘JL—)

0.96 6.56 ~0.0004 0.0143 0.0025 0.0145

1.02 6.18 0.0006 0.0165 0.0096 —0.005l
 API Racorvavisnoen PRACTICE 28K

Table J.1—Fairlea0l Motion Complex RAO (Continued)

Line No e l
Line Heading I 54 00 Degrees
Wave Heading From +X 1 -~ 45 00 Degrees
Line End Motions with X in Line from Fairlead to Anchor and Z Vertical
***Location (X e 110.00 Y e 104.00 z 4 -35 00)***
Wave Frequency Low Frequency Tm X lmag ZReal 7 Z lmag :
(Rad /Sec) (Sec) (AMP/AMP) (AMP/AMP) (AMP/AMP) (AMP/AMP)
5.84 C1 .0000 C? 1179
F) C1 0114 T -0 0120 7
5.54 D
'""lL.J 1186
C“; 0. 0 W0 0051
5.27 C31 #193
C.3 0.
5.03 21:;1200 .43

4.80 6:3Q3CT) 1173

C5 .013 C

4.60 5:1 1.0142

3
0
4.41 L:J C-5
CD 1.0111
(:3
c_) Z13 U
(‘L 5...?
‘Ti...?6-‘6"
C‘)

4.24 “Q1 1.0079 F0

1.. =1
‘<-1
4.08 »t 10048
3.93 ~10 '1
<1-0Q
‘5-I->
Q . '-11>55>3;‘
(W3Q1)
c_3c_)c_3c_3c_3
“---J“--3
"---II‘--J*--J*--1
'*-JU1
DJU3
*"--5-ll
"‘F5 0.10033 -~i C)(“*3 <2.:><%<.J£.Jc.>c.><j?-1 1-» 1-~1l—=1\_). ,¢.)L_;. >\Q

\\\\\
A0)

(Dn (0

1
Area e S({.§}{}) Au:

11(1) 1 03]) A00 Cos ((Djt + 0;)

1.549
Figure J.2---Spectral Decomposition
 Desien AND ANAt.YSiS o1= Srarionxassino Systems FGR Fioarnso Sraucruaas 175

Maximum crest

Wave Elevation

vir '|’11'yrr]!11]' 1

Tangential Fairlead Motion

///' |'\/|E1XilTlUlTl mOilOl’l Xm, Xrmg

./
z 4/ R : Xm/Xrmg

, '7 T ' V 7 '7 l V Y 7 ' '

1 V 7 7 U

i if77 7
20 sec

Selected 120-Second Window

Typical horizontal fairlead motion

(vertical fairlead motion similar)

_j-__’~m“~li i 7 i i iii _7_ W7 ______ __ ; - -_ _ _ _ -- ):: __-‘___. - - -

§ ‘ 1* 71"” " "’"'““""‘*sssssssssssssssssssss** '""”“"'“‘"'i""‘

Fiampecl transient area

Figure J.3---Selection of Input Motion

176 API Recoivavisnnao F’RACT1CE'2SK

The tangential motion is used because tangential motion Table J.2---Tension Spectrum
greatly influences the line stretch. The use of a limited time
segment based on the maximum wave elevation is not in gen- Frequency Tension
Rad/Sec Spectrum
eral recommended. Alternative methods, which ensure a sta-
tistically realistic response level can be used. The 120 0.200 0. 10070021240191
seconds of data is adequate and will give results of compara- 0.258 0.3302278
ble magnitude to those developed in a full three hour simula-
0.317 s0.279s9
tion for the computer programs used in this analysis.
1.375
C4 5s6.976s
However, the length of the data should be determined after a
1.433
C4 2172.754
sensitivity study for a particular computer program has been
1.492 1909.006
1.550
DC‘?
c—_~ s10.s9s7
conducted. 0.608 1s7.6015
31.667 60.40002
J.1.5 MOORING ANALYSIS RESULTS $1.725 21.31249
0.783 13.28411
J.1.5.1 Quasi-Static Analysis
0.842 5.786316
The mooring system was analyzed using a computer pro- 0.900 1.439414
gram based on catenary equations modified for elastic stretch 0.958
. . c_._ 0. 194769313-01
and bottom friction. The computed line tensions were for the 1.017 2.250558
most loaded line No. 2 (54 degree direction). 1.075 4.2703 70
1.133 2.902684
Mean Tension I 643 kips (35 percent BTL). 1.192 1.725605
1.250 0.8388341
Mean -+ Low»-Frequency I 650 kips (35 percent BTL).
1.308 0.5401919
Maximum Tension K 779 kips (42 percent BTL). 1.367 0.3427716
1 .425 0.2073264
Other derived parameters are: 1.483 0.1237301
~ . :.-......... ....-1.. . ,1 ... f. .542 10.7565913E-01
iV1&X1111u111 A11u1101‘ LO&u - us U1 U-J E.--....
"II 9"
1.600 10.5461 835E-0l
Maximum Suspended Line Length I 3,986 ft
The line tensions are acceptable (below 60 percent BTL)
J.1.5.2 Frequency Domain Analysis and suspended line length ensures adequate grounded length.
Anchor loads are substantially higher then those produced by
The most loaded line No. 2 was analyzed. Initially, a quasi- quasi-static analysis.
static analysis is required under mean plus low-frequency
tension. From a review of the data it was concluded that wave J.1.5.3 Time Domain Analysis
frequency tensions would dominate. Hence, the line is ini-
tially analyzed under mean plus significant low»-frequency Only the most loaded line, No. 2, was analyzed. The proce~
conditions: dure is similar to the frequency domain in that a quasi-static
analysis was first carried out under mean plus significant low»-
Mean Tension I 643 kips. frequency tension. The applied end motions were then com~
bined with the resulting tensions.
Mean + Low Frequency = 650 kips (35 percent BTL).
Mean Tension I 643 kips (35 percent BTL).
The line was analyzed under a 650 lop tension with the
specified end motions. The output tension spectrum is given Mean + Low-Frequency K 650 kips (35 percent BTL).
in Table J.2. The computed rms tension was 111.5 kips. The Maximum Tension I 1,101 kips (60 percent BTL).
maximum wave frequency tension was 3.72 X 111.5 -»~ 415 Other derived parameters are:
kips. The total tension is:
Maximum Anchor Load m 886 kips.
Maximum Design Tension I 1,065 kips (58 percent BTL).
Maximum Suspended Line Length =1 3,903 ft.
The other derived parameters are: The peak quantities are compared to allowables as before.
Maximum Anchor Load I 870 kips. Tensions are acceptable (60 percent limit); suspended line
length ensures adequate grounded length; and a suitable
Maximum Suspended" Line Length I 4,284 ft anchor must be designed.
 Dssiorv AND Anarvsis or STAT¥ONKEEP¥NG Svsrazus son Fronrnvo Srnucrunss 17?

J.2 Fatigue Analysis Example d. Wave heights, spectral shapes, spectral peak periods or
The following example illustrates the computation of life- equivalent.
time fatigue damage on a mooring line and the estimation of e. Probability of occurrence of the a.bove.
allowable fatigue life. A frequency domain method of
dynamic analysis is used. The mean forces associated with wind, wave and current
are first computed. The low-frequency motions associated
J.2.1 MOORING SYSTEM DESCRIPTION with each environmental condition can be obtained from
computer programs, model tests, or design curves.
J.2.1.1 A scmisubmersible system with a 12 line, 25
degrees/45 degrees/65 degrees mooring is considered. The In the usual case, eight directions at 45 degree intervals are
sufficient to define the environment. In this example we will
system is shown in Figure J .4.
consider one such direction, 225 degrees, in detail. The envi-
J.2.1.2 Mooring parameters used in the analysis for wire ronmental conditions to be analyzed are given in Table J.3.
rope: The sea-states, cumulative probability of occurrence, mean
loads and rms low-frequency motions are given. The follow-
Elastic stiffness (EA) 94,355 kips
ling additional parameters are used:
Air weight 22.7 lbs/ft
Submerged weight 19.3 lbs/ft a. Water depth 1,476 ft.
Mass 0.705 slugs/ft
Normal added mass 0.133 slugs/ft b. Pierson Moskowitz Spectral form.
Drag diameter 3.5 in.
c. Environment is in the analyzed direction 16 percent of the
Drag coefficient 1.2
time.
T-N curve NR4-09 e 731
(For example only) In this example, wind, wave, and current are assumed
Reference breaking strength 1,110 kips colinear. All lovv»frequency motions are assumed to be from
wave effects and are based on a reference stiffness of 18 kips/
J.2.1.3 Mooring parameters used in the analysis for chain:
it. Wind effects are included in the mean load. For a mooring
Elastic stiffness (EA) 147,074 kips system with a stiffness It kips./ft at the mean position, the
Air weight 123 lbs/ft low-frequency motions are computed as:
Submerged weight 107 lbs/ft
Mass 3.73 slugs/ft rms(k) 1 rms(l8) ><
Tangential added mass 0.25 slugs/ft
Normal added mass 0.50 slugs/ft where
Drag diameter 7 in.
rms(k) = the motion at stiffness k,
Drag coefficient 1.2
T»-N curve NR336 e 370 rms(l 8) I the motion at the reference stiffness 18 kips/ft.
(For example only. See
Table 3 for recommen- If low-frequency wind effects were included, a square root
ded equation.) relationship between the actual and reference stiffness values
Reference breaking strength 1,383 kips cannot be used. It is necessary to define the rms motions for
the actual mooring stiffness.
Fatigue analysis are provided here for the wire rope at the
fairlead and for the chain at the chain/wire rope intersection. J.2.3 FATIGUE ANALYSIS
The fatigue damage is computed as follows. The annual
J.2.2 ENVIRONMENTAL CONDITIONS
fatigue damage is initially computed. Low-frequency effects
A fatigue analysis consists of multiple individual line ten- are minor but are included. A year is assumed to have
sion analyses as described in 7.5. The environment is defined 3.15576 x 10 seconds. By Narrow Band Theory, the fatigue
as a set of the following: damage in a given sea—state is:
a. Directions.
b. Wind speeds.
D = N., (./2 n,,,..,,,_,)M - r (1 + M/2)/K
c. Current speeds. "KN; (./5 n,,.,,,,)M - r (1 +114/2)/'14 (11.1)
178 API Rscorvrrvrsuoeo Paacrzcs ZSK

189°

135°
225° (Anlyzed) 1 12

2 A I
.. . 6,30 - - . 11

I I we 5 10

to --1‘Q

.. . 9

5 O8

315°

0°

Figure J.4---Mooring System and Environmental Directions

 Dssreu awe ANAl_YS1S or-" Sraraouusssruo Svsrssvis son Froarnvo Sraucrunas 1 ?9

where natural period. In the usual case, low-frequency effects are

minor and the natural period is adequate. The wave frequency
D I annual fatigue damage,
zero up-crossing period or frequency should be computed
K I the intercept parameter of the T/N curve (731 directly from the tension spectrum generated in a line
for wire rope and 370 for chain, these values dynamic analysis, as follows:
are for example only),
v K ./M2/M, hz,
Mi K the slope of the T/N curve (4.09 in this example
for wire rope and 3.36 for chain, these values
are for example only), M2 =1: Inf’ S(/)df,
R.,,,,.,,,, 1- the ratios of wave frequency and lovv-freq_uency
and 1%;,.,.,,,. rms tension range (twice the single amplitude Ms e la Sr/1.
value) to reference breaking strength,
f K Frequency (hz),
NW K the numbers of wave frequency and low-frc- S(f) I Tension spectrum (kipsz/hz).
and N1 quency tension cycles per year.
N.,. can be computed as: J.2.4 FATIGUE DAMAGE
A detailed computation of the fatigue damage per year
.NWe=vx.3.l5576xlO7xPdxPS (11.2) associated with a 225 degree direction (wave heading
towards), for line 3 (65 degree spreading angle) is given in
where Table J.4 for wire rope and Table J.5 for chain. The total
v X zero up-crossing frequency of the tension spec» annual accumulated fatigue damage is:
trurn (hz),
D = 0.689 x 104 (chain).
Pd 2 probability of the direction (16 percent for this
it _CD
P-._l Ln (.11 >-< |,-1 CD
1"») I“
(wire
It
rope).
example),
PS I probability of the sea-state given the direction. In this example, low-frequency motions have essentially
no effect on the fatigue life, being about three orders of mag-
(Table J.3).
nitude less severe than wave frequency effects. This, how-
Low~frequency fatigue damage and wave frequency ever, may not be true for the cases where lo‘~.v-frequency
fatigue damage are computed independently. The low-fre~ motions are dominant. The fatigue damages for chain and
quency zero crossing period is estimated to be the natural wire rope associated with all directions are given in Table J .6.
period of the vessel/mooring system as a function of applied The total annual fatigue damage on the line, counting all
mean load. The appropriate number of low-frequency cycles directions is:
per year can be computed as:
D A 0.218 x 104 (chain).
N31: (3.15576 x 107 x Pd x PS)/Tn D 1 0.418 x l0"'3 (wire rope).

where The useful fatigue life, computed as:

Tn I vessel/mooring system natural period and the Life == l/(3D) years.

remaining quantities are as previously defined. =»~" 15 years (chain).
If procedures are used that incorporate a general lovv-fre-
x 80 years (wire rope).
quency (from wind or wave or both) tension spectrum, the
actual zero up-crossing period can be used, rather than the The safety factor applied. is 3, based on 6.8.
‘180 AP! RECOMMENDED Pnacnca ZSK

Table J.3—Environmenta1 Condition, Mean Loads and Low-Frequency Motions for the Analyzed Direction
Sig.Wave
(ft)
1¥.;aiii>i......i
(sec)
Probabi lityoo o
(percent)
it M...
(kips)
9 LovvFreq.
(ft)
3.60 8.4 16.96 28.7 0.40
8.30 9.2 36.29 61.4 0.66
1.3.01 1.0.4 26.07 1.31.4 ;.07
17.71 11.6 13.05 212.2 1.22
22.42 12.7 5.31 309.3 1.35
27.12 13.6 1.64 418.8 Q43
31.83 1-4.4 0.52 552.1 .1 .50"
36.53 15.3 0.13 714.9 1.60
41.24 16.1 0.02 891.7 1.68
45.94 17.7 0.01 1239.1 1.75

2* Low frequency motions based on a reference stiffiiess of 18 kips/fftt

Table J.4—Wire Rope Damage

Annual Accumulated Fatigue Damage by Sea-State for Direction 6, 225.0
(A11 Periods and #’s of Cycles Refer to Tensions)
A11 Data is Per Year
RMS RMS Zero Cross. Wave
Tension Tension # of Cycles # of Cycles Period Zero Cross. Frequency Low Frequency
Mean Ten. Sig. Height (Wave) (Low) (Wave) (Low) (Wave) Period (Low) Damage Damage
153.0 3.60 2.2 0.4 0ci§i i:§7ii0i§§i;1i:»s 0.7801-3+104 3.wc.
»9 76 if
3 196E~»~196 0. 1 72E--I 0
162.0 8.30 4.4 1.4 C>.243E~eS>6 0.168E+O5 7.56 1619.14 0 .748E-40 *5 0.52613-08
175.7 13.01 7.2 2.6 OI. 163 E+=[1+6 0.122E+05 8.09 108.22 01.03713-i 04 0.438 E-07
192.0 17.71 10.8 3.3 O.723E+IC+5 0.616E+04 9.11 106.94 G».863E-1 *4 0.599E-07
215.5 22.42 16.3 4.2 0.263E+05 0.258E+04 10.19 103.97 I1. 169E-~I £13 O.6l9E~07
242.4 27.12 24.0 4.9 0.734E~+~IC»4 0.8 1 6E+03 1.23 100.89 B».226E-| D13 |Cr.38C|E-~07
278.2 31.83 34.7 5.5 6.2 1 2E+G14 1G#.265E-+03 12.25 97.79 S#.296E-03 0.200E—07
324.8 36.53 49.3 6.5 0.4-42E+03 0.629E+1{)2 13.34 93.75 3I.258E-iU13 0.900E-08
376.9 41.24 67.7 7.3 0.799E+02 0.13OE+02 14.60 89.52 3. 170E—1 .13 0.300E~»08
480.2 45.94 96.1 8.4 0.325E+02 0.63 3E-H01 1.6.08 82.44 S1.290E-C '3 0.300 1E-08
"Fetal 0. 154E»-CP2 0.243E-06

Total damage measure for this direction: Total number of tension cycles per year for this direction:
Direction (6) 225.0 Wave frequency tension cycles 0.63-4E+06
Probability percent for direction 0.160E+02 Low frequency tension cycles 0.466E+05
First order damage 0.154E--02 Average wave tension period 0.795E+01 (zero crossing)
Second order damage 0.243E-06 Average low tension period 0.108E+03
 DESIGN AND ANALYSES OF STATIONKEEPING SYSTEMS FOR FLOATING STRUCTURES

Table J.5--Chain Damage

Annual Accumulated Fatigue Damage by Sea-state for Direction 6, 225.0
(A11 Periods and #’s of Cycles Refer to Tensions)
All Data is Per Year
RMS RMS Zero Cross.
Tension Tension it ofCyc1es T4 of Cycles Period Zero Cross. Wave Frequency Low Frequency
Mean Ten. Sig. Height (Wave) (Low) (Wave) (Low) (Wave) Period (Low) Damage Damage
A 591) 3.60 2.0 0.4 6.1 571611616 (11 780E+04 7.08 09.76 0.507E~0 5 I. 1 82E-1
162.0 8.30 4.3 1.3 0.243E+06 01 1681-3+05 7.55 09.14 0.1l9E-»0 3 >.229E4~1
175.7 13.01 7.1 2.7 0. l 62E+06 Q1 122E+05 8.11 08.22 0.437E~»~0 3 1.12313-4
192.0 17.71 10.8 3.4 0.721E+05 0| 6 16E+04 9.14
.-

06.94 0.789E~0 2 I. 143E»-1

215.5 22.42 16.4 4.2 0.262E+05 0 258E+0-4
.4

0.22
.-

03.97 0.12013-0 2 .1. 122E--I 0

242.4 27.12 24.2 4.9 0.733E+04 0t 8 l 6E-903
--.
3.1. 1.24
T’:
00.89 0t.l23E-0 2 t.670E-1
27s.2 31.83 35.0 5.6 0.21 l E+04 (Er .265E+0 3 2.26 97.79 0.12313-02 0.330E-1
324s 36.53 49.8 8.5 0.442E+03 Q .629E+0 2 '1'
L. 3.33 93.75 0.850E~0 3 I. 130E-I
376.9 41.24 68.4 7.4 0. 80 l. E+02 0 1 30E-+02 4.57 89.52 0.440E-0 3 |.400E-I
4s0.2 45.94 97.1 8.5 0.326E+02 Q1 633E+01 16.04 82.44 0.590E—0 3 I.300E~»~|
Total 0.689E~0 2 I.53lE-~l
Total damage measure for this direction: Total number of tension cycles per year for this direction:
Direction (6) 225.0 Wave frequency tension cycles 0.634E+06
Probability percent for direction 0. 1 60+02 Low frequency tension cycles 0.466E+05
First order damage 0.689E-02 Average wave tension period 0.795E~%-01 (zero crossing)
Second order damage 0.53 1E-05 Average low tension period 0. 108E-i-03

Table J.6--Annua| Fatigue Damage as a Function of

the Environment
Wlsine 3 (65 degrees)
Probability of Annual Fatigue Damage
Direction ' c ********* 9*
Direction (percent) Chain Wire Rope
0 6.0 o61;4aaa';6a 0.412 5 04
45 8.0 0.535 5 I 0-3 0.436 5 0-H4
.-

90 3.0 0.125 5 1 0"? 0.109 5 1 OT3

135
..-

2.5 0.375 5 I ,-3 0.337 5 I. |Q|"4

180 0.1.33 5 0T2 0.151 5 L03
1-

4.0
225 1 6.0 0.689 5 F2 0.1555 09
270 Tl
.1
.11. 8.0 0.100 5 0*? 0.21351 042
315 01.131 5 06
‘.‘
2.5 0.983 5 101-3
Total 10 0.0 0.218 5 1.0"} 0.41s 5 104