Permanent anchorages for offshore structures - imcyc.com Permanent Anchorag… · 1.2 Anchorage forces 1.3 Tethered structures ... for floating offshore structures followed by a ...

Permanent anchoragesfor offshore structures

STATE OF ART REPORT - PERMANENT ANCHORAGES FOR QFFSHORESTRUCTURES

FOREWORD

In construction at sea there is a continuous development of new technologies, research workand verification of practical behaviour of structures at their location. A good example ofsame is the foundation aspect of sea structures. During the first FIP Symposium on ConcreteSea Structures in Tiblisi in 1972 some speakers touched on “the foundation” and only Ben CGerwick mentioned problem areas varying from clean rock to low density silts.

Since then much has been done, specifically in relation to the gravity structures in the NorthSea and other areas. In 1979 this brought us so far that a Working Group on Foundations outof the Commission on Sea Structures could present a State of Art Report on Foundations ofConcrete Gravity Structures in the North Sea.

After this massive task the same group, headed by T. Ridley, continued in a developing field“Permanent anchorages for floating structures”. They thereby follow the tendency to greaterwater depths, more hostile environments and exploitation of short-life new resources.

I hope that this report will help to support - and not hold back - this new field of offshoreactivities, and developments are such that at the next congress we can report on a lot ofexperiences.

J H von LoenenChairman

4.1.1 Pile driving

4.1.2 Insert piles bored and grouted

4.1.3 Under-reaming

4.1.4 Pile driving analysis

4.2 Holding capacity of pile anchors

4.2.1 Axial capacity of anchor piles

4.2.2 Effect of tension versus compression

4.2.3 Effect of cyclic loading (axial)

4.3 Effect of lateral deflection

4.3.1 Lateral capacity of piles

CONTENTS

FOREWORD

INTRODUCTION

1 REVIEW OF CONCEPTS FOR OFFSHORE FLOATING STRUCTURES1 .O Introduction

1.1 Structural behaviour

1.2 Anchorage forces

1.3 Tethered structures (tension leg platforms)

1.4 Restrained structures

1.5 Anchored structures

1.6 Design of anchorage foundations

1.7 Summary of the state of art

1.8 Conclusions

1.9 References

2 FLUKE ANCHORS (CONVENTIONAL ANCHORS)

2.0 Introduction

2.1 Field investigations

2.2 Installation of fluke anchors

2.3 Holding capacity

2.4 Summary of fluke anchors

2.5 References

3 DEADWEIGHT OR GRAVITY ANCHORS

3.0 Introduction

3 . 1 installation

3.2.1 Analysis methods

3.2.2 Effect of repeated loading

3.3 Summary of deadweight anchors

3.4 References

4 PILE ANCHORS

4.0 Introduction

4.1 Installation. .

4.3.2 p - y curves

4.3.3 Ultimate capacity

4.4 Summary of anchor piles

4.5 References

APPENDIX 4A: REVIEW OF RESEARCH INTO THECAPACITY OF PILES IN TENSION

5 ROCK ANCHORS5.0 Introduction

5.1 Rock mechanics

5.2 Design considerations

5.3 Conclusions

5.4 References

INTRODUCTION

In August 1979, the FIP Commission on Concrete Sea Structures published a State of Artreport on Foundations of concrete gravity structures in the North Sea. This report wasprepared by a Working Group on Foundations which was formed to assist the Commission inthe understanding of the design and construction of offshore structures of this kind.

The Working Group on Foundations has followed this early work by a further State of Artreview of the permanent anchorages for offshore structures. In this review, the problem hasbeen considered of providing safe tension foundations for the new developments usingfloating offshore structures as platforms for long-term use in exposed offshore locations. Asthis development is still at an early stage, the review represents a useful summary of presentknowledge, and reference to ongoing activity for this type of foundation problem.

It has been considered necessary to limit the State of Art review to types of permanentanchorages which are capable of resisting forces at the seabed in excess of 1 MN. There are anumber of anchorage devices under development which have been proved to a lessermagnitude of loading, but these are not yet believed to offer solutions for the type of offshoreplatform structure under review. The types of anchorage which are dealt with are:

Fluke anchors

Deadweight anchors

Pile anchors

Rock anchors

The report has been prepared by a joint effort of the members of the FIP Working Group onFoundations. In particular, it deals with a general description of the development of designsfor floating offshore structures followed by a review of the different anchor types. Commentsare made about the prediction of anchor holding capacity and the methods of installation.Each section of the report is treated independently and they can be regarded as separatereviews with summaries, references and diagrams.

The Chairman of the Working Group on Foundations wishes to acknowledge the considerablevoluntary effort made by the members. It is hoped this effort will stimulate and supportfurther attention to the exciting growth of interest in the subject of anchorage foundations.

Members of the Working Group are:

Dr. J.B. Burland, Imperial College, London, UK

Ove Eide Norwegian Geotechnical Institute, Oslo, Norway.

T. Kvalstaad, Det norske Veritas, Oslo, Norway.

T. Ridley, Ove Arup & Partners, Scotland, UK

W.J. Rigden, British Petroleum Ltd., London, UK

F.P. Smits, Delft Soil Mechanics Laboratory, Delft, Netherlands.

R. Sullivan, Woodward-Clyde Consultants, Houston, USA

Tom Ridley -Chairman of the Working Group

1 REVIEW OF CONCEPTS FOR OFFSHORE FLOATING STRUCTURES

1 . 0 Introduction

This paper attempts to review the development of proposals for fixed offshore floatingstructures, as a background to further consideration of the methods to be adopted for theirseabed foundations. In 1979, the FIP published a State of Art Report{1 1) on foundations forconcrete gravity structures which highlighted the knowledge obtained from the behaviour ofseabed soils in supporting the very high compression loads these structures imposed. Thefoundation problem for the floating structure is one of dealing with the imposed tension loadson seabed soils, and this review is thus considered to be a logical next step of the WorkingGroup on Foundation’s interest in seabed foundation problems.

1.1 Structural Behaviour

The structural behaviour of a fixed offshore platform, of which the structure bears directly onthe seabed, presents a problem in design which has been fully investigated, i.e., steel jacketand concrete gravity structures. The structural behaviour of a fixed offshore floating platformis still a design problem in its infancy, and is already the subject of a very considerableresearch effort throughout the world. Because of the idea of flotation being the new conceptinvolved, the design problem has been assumed to be separately linked between:

(a) The structure

(b) The cables (or moorings)

(c) The seabed anchorage

It is considered that this approach presents unnecessary possible complications in the essentialquestion of evaluating overall structural behaviour. One obvious difficulty arises in bringingtogether the dynamic behaviour of structure, cables and anchorages into a common designapproach, particularly as they are all critical to the type of overall structural performancewhich can be experienced in the ‘fixed’ offshore condition” 2).

The ability to predict the dynamic behaviour of the structural cables under variousenvironmental and functional loads is vital to the structural and foundation design process.These structures under consideration are contemplated to be fixed in exposed offshorelocations for a lifespan of approximately 10 to 20 years, without structural failure during themost severe storms. The design of cables and anchorages for these structures thus poses newproblems not encountered in traditional marine technology, which has largely been confinedto dealing with the mooring of ships in relatively sheltered waters, or of drilling rigs in theocean for limited periods of time.

In the traditional type of fixed offshore structure, the design problem is governed by the needto resist the forces of wind, waves and currents by establishing structural behaviour withminimum displacements. This creates very large forces in the structural elements andfoundations which increase substantially as offshore locations in deeper waters are considered.The use of a floating structure changes this approach completely, and introduces the questionof allowing considerable displacement of the structure under the action of hydrodynamicforces. This significant displacement has the result of increasing the forces in cables andseabed anchorages.

There are two conditions of structural behaviour for a fixed floating structure from which theoverall design problem is determined, viz:

(a) Tethered structures

Where the structure is restrained in the vertical direction but allowed to be displaced in thehorizontal direction. This structural behaviour is achieved by the USC of vertical cableswhich resist, in tension, the large buoyancy forces frqm the structure to react with suitablydesigned seabed foundations.

(b) Anchored structures

Where the structure is unrestrained except for the structural behaviour of the mooringcables which are suitably anchored to the seabed.

In both cases the question is one of understanding the behaviour of the floating structure, tostudy how the foundation forces are derived, with their magnitude and directions. An estimatecan then be made of forces thus generated from the activity of the applied loadings, theoverall behaviour of the structure and its anchorage cables. The requirements for safefoundations need to be chosen with due regard for the structural behaviour of the seabed soilsat the location site.

Anchorage Forces

Figure 1.1 relates to the Tethered structure which gives a ‘steady state’ of vertical tensionforces in the cables. When exposed to the environment of wind and current, additional forcesare generated, which are considered to be still in the ‘steady state.’ The ‘dynamic’ behaviourof the structure due to the wave regime gives two cases for the extreme forces in cables. Thefirst is when the storm wave is adding to the vertical tensions and its increased buoyancyeffect, giving maximum vertical forces. The second is when the horizontal forces on thestructure are at their maximum due to the storm waves. The excursion of the structure is at itsmost extreme magnitude and this gives maximum horizontal force at the foundationcl.3).

Figure 1.1. Tethered structure.

Figure 1.2 relates to the Moored structure which gives a ‘steady state’ under the action of themooring cables. Conceptually, these are arranged to provide for horizontal forces at theanchorage with small vertical forces only under extreme conditions. During the ‘steady’ and‘dynamic’ state conditions the effects are similar to the tethered structures, in the wayexcursions take place, the difference being the allowance of vertical excursions due to thecompliance of the cables in this direction. In very deep water locations, the mooring cablesare usually made as short as possible which introduces vertical forces in the foundationsduring all states, from the angled cabled inclinations. The maximum vertical and horizontalforces at foundations are thus derived from the reactions of the cables during the ‘steady’ and‘dynamic’ states as for tethered structures”

Figure 1.2. Anchored structure (catenary mooring).

1.3 Tethered Structures (Tension Leg Platforms)

This type of structure uses an excess of buoyancy, provided in the form of integral flotationchambers, to apply tension to a vertical cable system. The concept was first seriouslyconsidered about 1960 for the designs of offshore oil/gas production platforms to be locatedin water depths from about 200 to 1000 m, where traditional platform designs of fixed steel orconcrete structures were proving to be increasingly difficult to justify economically. Thecompliant structure is expected to be capable of use for these operations, with the economicsbeing largely unrelated to water depth. This concept has come to be known as the ‘TetheredBuoyant Platform’ and Figure 1.3 illustrates a typical structure of this kind. The developmentwork on this type of structure has been intensive since 1970, and recently a structure of thistype has been ordered for the North Sea Hutton Field.

There are numerous different concepts, each endeavouring to optimise the design features, butthe serious problem area is related to the development of suitable oil/gas production methodsto suit these conceptsc’ 0.

A typical proposal of a TBP structure for 450 m water depth” 6) caters for a 15 000 t payloadwith a steel structural weight of approximately 15 000 t. For this structure, the excessbuoyancy will be about 10 000 t, which gives a static tension force in the vertical cables of2500 t each if four cable systems are used.

From the foundation point of view, the main feature of the tethered structure is that thecombined static and dynamic tension forces from the tension members are to be resisted and,except in water depths below 200 m, there are very low horizontal forces experienced duringextreme storm wave loadings.

1 . 4 Restrained Structures

Although not strictly a floating structure, the concept of a guyed tower platform has beendeveloped for water depths up to 600 m(1’). The tower is held upright by 16 cables connectedto clump weights which further extend to pile or drag anchors (Figure 1.4). Under normaloperating conditions, only small deck motions will be induced. During heavy storms theclump weights will lift off, permitting the tower and guylines to absorb the environmentalloads. Large horizontal loads will thus have to be taken by the pile or drag anchors. Themaximum tension forces are approximately 350 t in 600 m water depth.

LSeabcd Anchorages.

Figure 1.3. Tethered buoyant platform - general arrangement.

In Tension

AnchorTemplate. Trai,i”q Line..

,\vTmx”7

Anchor UngroutecPile. Piles.-

PretensionC h a i n . \

Figure 1.4. Schematic of prototype guyed tower for 350 m water depth.

1 . 5 Anchored structures

As future developments are considered for the location of large floating structures in exposedoffshore locations, to capture the energy resources of the ocean, the various concepts arechiefly concerned with maintaining their positions during the most severe seastates experiencedduring their useful life. From the point of view of structural behaviour, these structures mayhave considerably different design parameters for the extent to which displacements may betolerated to suit functional needs. The common feature of these concepts is that theirstructural behaviour in resisting hydrodynamic forces is conditioned by their use of catenarytype heavy weight cables, to provide foundation connection to the seabed.

A// ;gFigure 1.5. Schematic of Conprod, floating production platform.

For offshore oil and gas exploration, floating drilling rigs, mainly semi-submersibles havebeen used for many years. The experiences with these structures are among the mostimportant bases for developing more permanent installations. In recent years, floatingplatforms for production of petroleum have also attracted considerable interest, particularlyfor the North Sea. An example of such a platform is the Conprod shown in Figure 1.Y.Q.Some of the main advantages of floating production platforms are the quick installation andthe moderate cost increase with increasing depth. Such platforms are thus often considered formarginal fields, for which the simple removal at the end of the production period is an addedadvantage.

For the application of floating structures to harness the energy from waves, a number ofconcepts are presently being studied in the UK and Norway(l.y). The location of thesestructures is being investigated at approximately 60 m water depth off the Western coast ofthe Outer Hebrides. Figure 1.6 gives the outline arrangement of the structural system for thistype of structure, and for a 100 m length of structure the maximum anchorage forces fromthe cables is approximately 3000 t horizontally.

incident Waves.

I Converter Units

Transmitted AndDiffracted Waves.

I - -D

contours.

-4--_----

Schematic View Of A Converter Array.

End Elevation.

Figure 1.6. Schematic of wave energy concept.

Another concept for a large floating offshore structure is the power plant for OffshoreThermal Energy Conversion (OTEC) being chiefly developed in the USA and Japan. Anumber of different types of structural systems are currently being studied for location inwater depths up to 6000 rn(l,lo) for which catenary cable systems are to be adopted. Figure 1.7indicates the type of structural arrangement proposed and gives the approximate maximumloading for 45” inclined tension forces as 900 t vertically and horizontally.

Figure 1.7. Schematic of OTEC - concept for deep ocean environment.

1.6 Design of Anchorage Foundations

From the foundation point of view, the anchored structure gives a design problem for whichthere has been a considerable amount of marine technology experience in traditional anchortechnology. However, this experience is only useful to a limited degree, as the problems oftraditional anchorage of large structures have been usually only dealing with temporarylocations in relatively sheltered and shallow water depths. New foundation solutions arehaving to be found, to deal with the application of this type of floating structure in the verydeep and/or severely exposed locations where they are needed for safe long term installation.

The requirement of providing safe foundations for a floating offshore structure is notdissimilar to that for other types of fixed offshore structures, i.e., the reliance upon theseabed soils to carry the imposed loads without instability and intolerable deformations. Forthe anchorage foundation, there are the following factors to be taken into account, viz:

(a) The applied tensions can vary in magnitude and direction requiring large resistancegenerally in excess of 1 MN (100 t).

(b) Only small movements can be tolerated at the anchorage points

(c) The time to install anchorage foundations can be considerable, depending on availableequipment and lift capacity.

In order to be able to predict the structural performance of the anchorage foundation, it isclearly necessary to obtain precise information about the mechanical properties of the seabedsoils. This will involve investigation of geological stratification of the various strata, togetherwith sampling and testing of loadbearing materials. The information obtained can be used topredict the structural behaviour of the anchorage foundation under the applied loadings. Anapproach to the selection of the appropriate type of anchorage foundation can thus bemadecl 11).

Other constraints in the design of anchorage foundations arise from problems associated withtechniques for installation, inspection, maintenance and repair (Reference 1.12) As floatingoffshore structures are considered for use in the very great water depths of the world’soceans, these practical considerations can dominate the design problem. For such deepwaterapplications, the anchorage forces do not increase significantly, and dynamic effects arereduced due to the greater dampening from the length of moorin&c&b.

1.7 Summary of the State of Art

The permanent anchorages to resist forces in excess of 1 MN currently being proposed for usewith floating offshore structures are basically of the following types, viz:

Fluke anchors

Deadweight

Pile anchors

Rock anchors

and the general features of these are discussed in the following sections of the report. Asummary of the state of art review is now given.

Fluke Anchors (Section 2) It is difficult to predict the behaviour of a drag embedment anchorby theoretical methods, due to the complex soil behaviour during installation. For long termuse, there is also the added problem of limited knowledge of the influence on anchorbehaviour from large values of cyclic loading variations. There is also the restriction of usefor this type of anchor, where the seabed soils may be of shallow thickness overlying rock,thus causing lack of soil stability when resisting large anchorage forces. Due to low resistanceto vertical forces, lengthy mooring lines are needed to achieve nearly horizontal loading atseabed.

Deadweight (Section 3) The efficiency of the deadweight anchorage system is lower than otheranchorage types, as the seabed soils are mobilised to a limited degree. The size of thedeadweight anchorage is very large and, in shallow water depths, may need to be partiallyburied in the seabed to avoid unwelcome hydrodynamic effects on the floating structure. Fortethered floating structures, where the horizontal anchorage forces are much lower than forthe moored structures, the deadweight anchorage foundation is appropriate from a safetypoint of view. Holding capacities of up to 100 MN are being considered for possibleapplication. Short mooring cables are possible due to the large uplift force resistance.

Pile Anchors (Section 4) A considerable research effort is already under way with the aim ofdeveloping tension pile design criteria for the anchorage of tethered floating structures. Theeffect of combined tension and lateral loading is also being studied, as there is a need toappreciate this type of behaviour in more detail to improve prediction methods. In particular,the combination of static and dynamic tension loads is poorly covered by existing knowledge.Nevertheless, tension pile anchorage foundations can, with present technology, be designedand installed to take very large vertical tension and horizontal forces. Possible applicationsalready being studied involve anchorage forces of the order of 50 MN vertical a’nd 8 MNhorizontal for tethered floating structures.

Rock Anchors (Section 5) The use of rock anchors has been widely adopted for temporaryand permanent use on land-based structures, but there is little evidence of their use as seabedfoundations. There is a lack of knowledge of the engineering properties of the types ofsubmerged rock likely to be encountered, so that safe design parameters can only be basedupon very conservative judgement at the moment.

Nevertheless, the use of rock anchors in the form of drilled piles is a very attractive solutionwhere the combination of high anchorage forces and/or deepwater locations are consideredfor new developments in offshore structures. This is because the drilled pile can be easilyadjusted to meet seabed irregularities and, for a rock foundation, this makes it the mostpractical alternative solution. The disadvantage is the high cost and limited availability of thespecialised drilling and grouting equipment.

1.8 Conclusions

The development of floating structures as feasible concepts for deep water locations is still atan early stage. Much research work is currently being devoted to solving the different designproblems posed by their possible use as an extension of traditional offshore platformengineering technology. The design problem also covers the widening of knowledge needed onthe extreme conditions of wind, wave and currents to which these structures are likely to beexposed during their useful life.

Structural behaviour of floating structures is governed by the wave dynamic excitation and theinteraction forces set up in the structure and its foundation soils. From the point of view offatigue behaviour, the consideration of sea states, other than the extreme conditions, is ofgreater importance to the overall structural behaviour. Where the structure consists of cableelements, the problem of dynamic response is more complex and long term structuralperformance is very difficult to predict by the designer.

In view of this situation, the full range of interaction forces between cables and their seabedanchorage foundations is still difficult to establish for accurate design purposes. The State ofArt position, with regard to the selection of foundation systems suitable to meet therequirements of floating structures, is largely based on the background of traditional marinetechnology. This is based upon the use of anchorage foundations whose holding capacity isjudged without detailed reference to the mechanical properties of seabed soils.

One of the chief factors which will influence the performance of seabed foundations is theefficiency which can be achieved during their installation in the deep water offshore locations.The precise positioning of the seabed foundation structure will depend on the use of veryadvanced marine techniques of surface vessels and equipment. The costs of installation maywell be as much as 25% of the total cost of the completed structure.

1.9 References

1.111.12

FEDERATION INTERNATIONALE DE LA PRECONTRAINTE. Foundations ofConcrete Gravity Structures in the North Sea FIP State of Art Report, August 1979.PANTING, J.R. and HORTON, E.E. Analysis of the tension-leg stable platform.Paper No. 1263 presented to the Offshore Technology Conference, 1970, Houston.GIE. T., and de BOOM W.C. The wave induced motions of a tension leg platform indeep water. Paper No. 4074 Presented to the Offshore Technology Conference 1981,Houston.WILHELMY. V., FJELD, S. and SCHNEIDER, S. 1981, Non-linear response analysisof anchorage systems for compliant deep water platforms. Paper No. 4051 Presented tothe Offshore Technology Conference 198 1, Houston.OLBJORN, E. and FOSS, I. Certification of new Concepts. Oceanology InternationalConference, 1980LANG and HEDLEY. BP Development of TRP Production System. EEC Symposium1979.SYNDER, WARDELL, and LOFTIN. The guyed tower as a platform for integrateddrilling and production operations. Paper presented at Marginal Fields Conference,Norway, 1977.KURE, G. Conprod, a Floating Concrete Oil Production Platform. Paper presented atConcrete Ships and Floating Structures Conference, Amsterdam, November 1979.HMSO. Wave Energy - A Review Paper Energy Paper No. 42, HMSO, November1979.ATTURIO, J.M., VALENT, P.J. and TAYLOR, R.J. Preliminary selection of anchorsystems for OTEC. C.E.L. Technical Report R 853, Port Hueneme, California, March1977.MCCORMICK, M.E., (Editor) Anchorage Systems, Pergamon Press, Oxford, 1979.DOVE, P.G.S., ABBOT, P.A., and HARVEY, D.G. Deepwater high capacitymoorings. Paper No. 4050 presented to the Offshore Technology Conference, Houston,USA. 1981.

2 FLUKE ANCHORS (CONVENTIONAL ANCHORS)

2.0 Introduction

This group of anchors includes the commonly used ship anchors as well as the modern riganchors developed-during the last 10 years. The fluke anchor generally consists of a shankthrough which theload is applied and flukes which are constructed to make the anchorpenetrate. Additional parts, such as tripping palms and stabilizer bars, have been used onseveral anchor designs to encourage embedment and improve the stability against rotation andpull-out (see Figure 2.1).

Figure 2.1. Anchor terminology.

TRIPPINGPAL MS

(a) NAVY STANDARD STOCKLESS ANCHOR

STABlLI2E.RI

(b) NAVY STOCKLESS ANCHOR WITH STABILIZERS

The basic shape of conventional anchors (stock anchors) (Figure 2.2f) remained almostunchanged until the Stockless anchor (Figure 2.2g) was patented in 1821. This anchor hadhinged flukes attached to the shank. Then again, very little improvement took place until 1939when Danforth invented his anchor (Danforth anchor or lightweight type anchor LWT) withsharpened flukes and streamlined shank. The efficiency (i.e. holding power/dry weight ofanchor) was improved considerably by this new deisgn. The Danforth or LWT-anchorrepresents the first efficient burial anchor (Figure 2.3a and b).

Further work was conducted by the British Admiralty during the 1940s FarelP1j, Dove(2.2).Extensive series of model and full scale tests were carried out and, during the 195Os, the USNavy conducted tests in order to define and improve the capabilities of conventional anchors.Towne and Stalcup@J). This work resulted in the STATO-anchor with large tripping palms,Figure 2.3d. The information gathered from these experiments may be summarized as:

(a) Longer, narrower flukes tend to give better holding capacity.

(b) For a given fluke area, the best fluke angle (i.e. angle between fluke and shank) decreasesas fluke length increases.

(c) A fluke angle of approximately 30 to 35” seems to give optimum holding capacity insandy seabed.

(d) A fluke angle of approximately 50” seems to give optimum holding capacity in mudseabed (mud = soft clay).

(e) The angle between the shank and the seafloor is critical. Holding capacity can bedecreased by up to 50% when the shank angle approaches 6” to 12”. For larger angles,the holding capacity becomes marginal. This means that vertical load components tends todecrease the holding capacity considerably.

Stone anchor

Stone anchor withwooden teeth

One fluke anc!;Lrwith stone

Four fluke anchorwith stone

Ro!nan woodenanchor

Admiralty anchorwith stock

HALL-anchor

SPEK-anchor

POOL-anchor

AC-14 anchor

Figure 2.2. Development of ship anchors (from Reference 2.17).

a) Danforth anchor b) L.W.T. anchor

cl L.S.T. anchor

d) STAT0 anchor

e) BOSS anchor

f) Delta-double shank anchor

Figure 2.3. Anchor development from ship to rig anchors (from Reference 2.17).

(f) The use of heavy chain will influence the anchor penetration due to the penetrationresistance of the chain itself.

(g) For conventional anchors, a stabilizer element is necessary to prevent rotation of theanchor. The length of the stabilizer bar can influence considerably the anchorperformance.

Based on the knowledge gained from the above mentioned investigations, several anchordesigns have been introduced on the ship and rig market. The efficiency of these designs was,however, not always too impressive and during the 1970s new anchor designs have beenpresented which often deviate considerably from the traditional anchor shape. Bruce, Hook,

DIGGER STEVIN STEVD IG H O O KHK3 (* AUXILIARY) HOOK (SOFT HW) HOOK (MU0 TYPE 1

a ) V A R I O U S STEVIN A N D H O O K ANCHORS

, khos

xx)750

l o w15002 0 0 025mJloo0r o o o5 o o otoa3

!Oooo135ca!.9ooo2 2 5ca2 7 2 5 0

wan FWe snmh

7200 9 6 0 rm1 5 0 0 ‘200 !3501 xa 1360 15.301670 15cKl 168021.0 ,710 1 9 2 52360 !B90 2 1 2 52 5 . 0 2 0 3 5 2290

2 7 0 0 2’60 Z.302 9 7 5 2380 2680

3 2 5 0 26W 2 9 2 53560 2860 3 2 2 04100 3280 3 6 9 01.50 3560 m o or900 3 9 2 0 1.10

5300 12.0 ,110

5 6 5 0 4520 5005

n mdkmetres

(ha TL48 ,3805 2 1 i2556 19%6 5 2150‘ 0 2 . 6 0-5 211580 2 9 2 59 0 3 1 0 5

rca 3 . 2 5110 3600125 41201.0 a15150 5 1 2 0160 5 6 3 51 70 w90180 6 5 0 0

b) FLIPPER DELTA ANCHOR

Figure 2.4. Modern rig anchors (from Reference 2.16 and Reference 2.18).

Boss, Stevin, Doris, and Delta Flipper are examples of these new designs, see Figures 2.3 to2.6. Test data are limited, but the manufacturers claim efficiencies in the range of 10 to 40,dependent on anchor weight and soil conditions, and provided with a correct installation. Asthe weight of these anchors presently available goes up to range 100 to 500 kN, the holdingpower may be expected to lie in the range 1 to 5 MN as the lower efficiencies will apply to theheavier anchors.

The most frequently used anchor types on semi-submersible drilling platforms are, however,still the LWT/Danforth anchor or anchors of similar improved designs, such as MOOREASTand STATO. For work in the North Sea area, anchor weights, of 130 to 180 kN are commonpractice and each semi-submersible is generally equipped with 8 to 10 anchors.

6500 2.53

NOMINAL DIMENSIONS

9ooo 2.79 5.M) 3.36 115

Other Sizes Geometrically to Scale

Edm F G H JEdm F G H J RR RRmm mm mm mm mm mmmm mm mm mm mm mm mm- - - ..-- - - ..-

46 35346 353 335335 64 53 20164 53 201 72

60 50060 500 403403 92 70 29292 70 292 105

75 66675 666 6326321125 92 383125 92 383 137

7575 734 697734 697 140 92140 92 419419 151

R Rm m

a) BRUCE ANCHOR

b) DORIS MUD ANCHOR

Figure 2.5. Modern rig anchors (from Reference 2.15).

Ibs_--

22cixI

33300Ad030

5x03- -

it=IIlkc2

loo0lOcKI

150032cK.m

A- -2222s3

23 158 1w 71 172

357 199 250 89 216

AC3 223 287 1’32 248

452 2 5 2 317 126 282

570 317 400 481 355

652 3L!L3 458 SSl rob717 399 SCM co6 447

771 429 542 652 481

mzasuremants In cm.

D I,h+:>- K

Figure 2.6. STEVIN anchor.

On more recently built platforms and work barges, the newer anchor designs are gaining anincreasing part of the market. It may be expected that, for future anchorage of floatingoffshore structures with catenary mooring systems, these anchors will dominate.

2.1 Field investigations

2.2 Installation of fluke anchors

Prior to installation of an anchored structure, a site survey will be required. Common practicetoday is to carry out an acoustic (boomer-sparker) interpretation of the seabed at theproposed platform area, including the proposed anchor positions. Additionally, sidescan sonarmay be used and, to some extent, grab sampling or gravity coring of soil material from thetop metre(s) could be required.

Based on this information from the site survey, it is normally possible to describe the seabedsoil in the area as sand, clay or mud and the approximate thickness and horizontal extensionof the different soil types.

There are mainly two possible ways for installation of fluke anchors for a floating structure.

(a) Anchors, chain/wire and such like are transported by the floating structure and installedimmediately after arrival by means of an anchor handling tug.

(b) The anchors are transported and installed by anchor handling tugs or other vessels, priorto the arrival of the floating structure, which then has to be hooked up after arrival.

Which of the two alternatives is to be preferred depends on several factors such as expectedduration of weather windows, anchor size and design, soil conditions, permanent ortemporary installation, type of structure, and such like.

For semi-submersible drilling platforms, which move several times a year to new locations,alternative (a) is certainly more practical than (b), but for wave power devices alternative (b)may probably be more attractive.

Anchor handling tugs can, today, normally handle fluke anchors with weights up to about

kN. The stern winch capacity is generally between 1 and 2 MN for larger vessels and thebollard pull lies around 1 MN.

The installation process can be described as:

Anchor handling tug hooks up anchor and chain from floating structure, alternative (a) andgoes to a position which is normally marked previously.

Anchor attached to pennant wire is lowered to the seabed together with the main chain/wire.

Pennant wire with marker is buoyed off.

Pretension is applied.

(a) by the winches of the floating structure

(b) by the bollard pull of the anchor handling tug after payout of a sufficient length ofchain/wire, say a scope of about 10.

If the prescribed pretension (test load) cannot be achieved and the anchor drags, the aboveprocedure is repeated. If the anchor fails again, a second and eventually a third anchor (back-up or piggy-back anchor(s)) will be applied. Some typical failure patterns are shown in figure2.7).

When determining the drop position of the anchor, the expected dragging distance should betaken into account.

In sandy soils, the peak holding capacity will normally be reached within 10 to 30 m dragging

._- - - _--- - - -_-. - _- ---- - - _._- - ----.-(A) ANCHOR PERFORMING PROPERLY

--; A- -..__ __ _ - -- -- ‘-/$m/Fz - ---- - - --

/ --/--/- ---. - -

- _- - s.<-_r--- __ -____ --~-___ - -- -- -zL~--~~~ - --~ - -_

(Cl ANCHOR BALLING-UP AND PULLING OUT

(DI USE OF EXCESSIVE FLUKE ANGLE IN SAND

Figure 2.7. Different types of anchor behaviour (from Reference 2.14).

distance while in softer soils (mud, soft clay) increasing capacity has been reported even after50 to 100 m dragging distance.

The anchor design (shape, weight, fluke angle, additional structural elements, etc.) is of greatimportance for the anchor’s stability as well as for its ability to penetrate. In sand, thepenetration depth of burial anchors (fluke tip penetration) will generally be restricted to a fewmetres, whereas in softer clays burial depths of more than 10 m have been reported.

Theoretical prediction of stability and penetration ability is a complex matter. Saurwalt’2.d-Z,‘)presented a comprehensive work on theoretical treatment of anchor penetration, stability andholding capacity. Due to the complex shape of the different anchor types, he ended up withvery complicated expressions to establish equilibrium equations even after considerablesimplifications had been included.

This, together with a general lack of information about the soil properties, may be the reasonwhy model and medium scale testing combined with full scale experience has been given moreattention and still forms the basis for prediction of anchor behaviour.

Further work on anchor stability has been presented by Ura and Yamamoto@l”).

The holding capacity of fluke anchors is normally determined by the following expression,which dates both to Leahy and FarrirP’):

where W = anchor weight

C, b = constants dependant on soil and anchor type

This relationship plots as a straight line on a log F vs log W plot, with C being the interceptat W = 1 and b the slope of the line. When b # 1.0, C will depend on the units used forforce (weight). The values of C and b are normally determined from medium scale field tests.The above expression will then allow an extrapolation to heavier anchors. The extent ofpossible extrapolation is at present unknown.

Typical values for b lie in the range 0.7 to 0.9 for sand as well as clay, while values for C varyconsiderably for different soil types and different anchor designs.

For the most efficient anchor designs in clay, it seems that the following values will giveconservative estimates, at least for anchor weight above 50 kN.

Soft mud

& < 10 kN/m2b = 0.8 c = 15 to 25

10 kN/m2 < S, < 25 kN/m2 b = 0.8 c = 25 to 40

S, < 25 kN/m2 b = 0.8 C = 40 to 60

In sandy seabeds the following values for b and c may be used for an estimate of possibleholding capacities

Loose sand b = 0.8 C = 1 0 to 2 0

Medium dense b = 0.8 C = 20 to 3 0

Dense b = 0.8 C = 30 to 4 5

These numbers are given merely to give an indication of the working range of modern flukeanchors optimally designed and correctly installed under horizontal short term loading andapply for holding. capacity and anchor weight expressed in kN.22

The values for b and C were evaluated from reported holding capacity measurements, fromfield tests and from advertising material from anchor manufacturers which have been plottedin Figure 2.8. There is a considerable scatter in the values for different anchor types, evenwhen tested at the same location and the soil data from the different fields are generally nottoo well defined.‘2.‘2.2,21’.

;;OOK u)FT CLAyb Boss, SANDrr

ODANF S A N D

BOSS, MUD

i//y / ~BRUCE,FIRH CLAY”i,-BRUCE, SAND I‘/

’ / BRUCE; S5FT MUD I- O A N F F I R M C L A Y

STAT’l.CLAY I AN0 SAND I

8STATO,SAND O A N F S T I F F C L A Y

‘3&9 0 ‘

. . BOSSJDENSE S A N D

1 10 100 loo0ANCHOR WEGHT,W, kN

Figure 2.8. Holding capacity vs. anchor weight; data form field tests, model tests andmanufacturers’ data.

The criterion of weight alone should not be taken as the main parameter for prediction ofholding capacity. The ability to dig in and penetrate and then be stable during the penetrationphase is a function of anchor shape, weight distribution, fluke angle and soil parameters aswell as the anchor weight. Even slight modifications of fluke angles and stabilizing elementslike stabilizer bars (see Figures 2.1 and 2.3) may have a significant effect on the stability andability to penetratel2 19-221). Mobilization of holding capacity is then a function of thepenetration depth, the fluke area and the soil parameters.

It would certainly, at least from a geotechnical point of view, be more satisfactory if holdingcapacity predictions could be carried out, based on soil mechanics principles. Proof loading offuture fluke anchors designed for much higher holding capacities than today’s rig anchorsmay be impractical or even impossible within the economical limits of a project.

Additionally the short term proof loading as practised today may not be relevant as averification of the holding capacity under long-term permanent loads (pretension), combinedwith repeated load variations due to waves, wind, and current. One reason for this is thedevelopment of suction forces under short term testing of holding capacities. Negative porepressure will develop underneath the anchor flukes and, possibly, along the sliding surface inOC clays and dense sands.

The dragging velocity or pulling rate during testing, normally in the range of 1 to 2 m perminute, may thus influence considerably the results.

Another factor is the holding capacity of the part of the anchor chain or wire partly, or fully,embedded in the seabed. Tests performed by CEL’2.1Y) showed that the anchor chain canprovide a considerable portion of total anchoring resistance and that chain resistance canincrease significantly as an anchor is dragged into deeper, stronger soil.

The effects of earthquakes and scour on sandy seabeds as well as seabed instabilities in clayeyseabeds (slope and wave induced) represent potential sources of risk, which at least to someextent can be incorporated in a geotechnical analysis of the anchor/soil/environmentinteraction.

The effect of repeated loading combined with high permanent loading is mainly restricted topermanent deformations. A totally stable fluke anchor will thus most probably movegradually into a new position. Experience from laboratory triaxial tests on clay and sandindicates that the deformation velocity will depend on the permanent stress level as well as thecyclic stress level, and that the latter dominates. This will result in displacement of theanchored structure, changes in pretension of the other anchor lines, transfer of load to otheranchors which again may be over stressed and eventually damage to other connectionsbetween the structure and seabed (for example risers) or to nearby structures (bridges).

A dragging fluke anchor may destroy pipelines, communications cables, wellheads, and otherequipment.

The tolerable anchor displacement will thus vary for different structures and locations, andthe safety factors to be applied on holding capacity should be related to the consequences ofexcessive displacements.

It is not possible within the scope of this contribution to go into a detailed presentation of ageotechnical analysis method of fluke anchors.

As mentioned in Section 2.2.2, Saurwalt (2.5-2.y) has presented a comprehensive contribution onthis matter. However, a brief look at some simplified calculation models is given and theresults, are compared with the empirical constant b and c.

For a shallow (not deeply penetrated) fluke anchor, the resistance against horizontaldisplacement can be expressed as the sum of passive earth pressure in front of the anchor,friction or adhesion developed between the anchor and the underlying soil as well asfriction/adhesion along the sides of the moving anchor/soil mass as indicated in the figurebelow.

Fh = Eph + Sh + Sside

L cosa - L cosa

Assumptions: Horizontal pull

a = 30” in sand

a = 45” in clay

+ = 35 to 45” in sand

S, = variable in clay, but constant with depth

Shallow embedment in sand (i.e. shank on seabed or slightly below seabed):

Eph = %d2by’.Kph

sh = ( Wanchor + wsoil) .cos2a-tan+

Sside = 2dlcosa.Kh.c my ‘tan+

wsoil x ‘/zd*lcosa*b*y’

Shallow embedment in clay:

Erph = 2d b-S,

sh z 0.7 -S,-1.b-cosa

Sside x 2dlcosa.0.7S,

For several of the anchor types, the dimensions are proportional to I+‘%. Evaluation ofStevin, Flipper Delta, and Bruce anchors gave the following approximate idealizedexpressions:

d-l = (0.2 to 0.3) W%

d-b = (0.2 to0.3)Ws

Lb = (0.4 to 0.5) W%

(Win kN)

(d and I in m)

The coefficient of passive horizontal earth pressure may be expressed as:

Kph = tarP(45 + O/2)

For friction angles in the range 35 to 45”, this corresponds to Kph values in the range 4 to 6.Submerged unit weight of soil may be assumed to be approximately 10 kN/m3, and thecoefficient of lateral earth pressure normal to the sides of the wedges, Kh is assumed to be1 .o.

For sand, the expression for Fh will take the following form for sandy seabed:

Fh = (6 to 8). Ws.d + (0.5 to 0.6)W

For clayey seabed we find

Fh = (0.8 to 1.2)Su W%

The above approximate expressions have been plotted in Figures 2.9 and 2.10 as log F/vs log Wdiagrams for different tip penetration depths (6) in sand and different undrained shearstrengths in clay.

The general trend is that these expressions give lower holding capacities than reported frommodel and full scale tests (Figure 2.8).

1 160 1ANCHOR WEIGHT, W, kN

Figure 2.9. Holding capacity vs. anchor weight, simplified geotechnical evaluation formodern rig anchors in sand.

For deeply embedded fluke anchors in clay, these anchors will behave similarly to embeddedplate anchors. The holding capacity may be evaluated by a bearing capacity equation:

F = Su~Nc~Ap~Sc

where S, = undrained shear strength

NC = bearing capacity factor

Ap = projected area in pulling direction

SC = shape factor

For deeply embedded anchors NC values in the range 12 to 15 have been evaluated for shortterm loading including suction effects. The projected area of several anchor designs can berelated to the weight of the anchor by the following approximate expression:

Ap x (0.2 to 0.3) W 2/3

The shape factor will be close to unity. This leads to the following expression

F = (2.4 to 4.8) S,. W’s

which gives 3 to 4 times higher values than evaluated for shallow embedded anchors.

Figure 2.10. Holding capacity vs. anchor weight, simplified geotechnical evaluation formodern rig anchors in clay.

2.4 Summary of fluke anchors

Available designs of fluke anchors represent an adequate type of anchorage with ultimatehorizontal capacities up to the range 1 to 5 MN for a wide range of soil conditions. Futureenlarged anchors may be capable of taking even higher loads, say 10 MN, but thecorresponding weights will then most probably exceed 1 to 2 MN.

The main restrictions are as follows:

only horizontal or close to horizontal loading

unidirectional loading

requires sufficiently thick layer of soil over rock

The choice of a special fluke anchor for a catenary mooring system will depend on severalfactors, and at least the following should be taken into account:

dynamic and permanent forces

pulling angle

soil conditions (stratification, soil type, strength characteristics)

seabed topography (unevenness and slope)

anchor installation procedure

stability of anchor when dragging on seabed

ability to penetrate, steel wire rope and chain resistance

stability when dragging in penetrated condition

10 000

1 10 100ANCHOR WEIGHT, W,

weight of anchor and weight distribution

shape of anchor: fluke angle, fluke shape, stabilizer elements, tripping palms, shank shapeand length.

Prediction of fluke anchor behaviour during installation and under operational conditions istoday mainly based on experience and empirical relationships, derived from medium to smallscale field and model tests as well as proof loading of installed anchors.

Extrapolation of available data into holding capacity ranges exceeding today’s range ofexperience should be done with great care as there are several sources of uncertaintyconnected to the test data.

There is thus an increasing need for analysis methods based on geotechnical principles. Thiswill, at least to some extent, require refined field investigations. In this way fluke anchorfoundations could be designed in a similar way to any other foundation.

2.5 References

2.152.162.17

2.182.19

FARRELL, K.P. Improvements in mooring anchors, Royal Institution of NavalArchitects, Quarterly Transactions, Vol. 92, No. 4, 1950. pp. 335-350.DOVE, H.L. Investigations on model anchors. Royal Institution of Naval Architects,Quarterly Transactions, Vol. 32, No. 4, 1950.TOWNE, R.C. and STALCUP, J.V. Development of 15 OOO-pound STAT0 anchor,Technical Report Z58, Naval Civil Engineering Laboratory, Port Hueneme, California,6 November 1961.SAURWALT, K.J. Movements and equilibrium of anchors holding on an impervioussea bed. Section I. Schip en Werf, Nr. 25, Rotterdam, 1971.SAURWALT, K.J. Movements and stability of anchors on an impervious inclineduneven sea bed. Section II. Schip en Werf, Nr. 9, Rotterdam, 1972.SAURWALT, K.J. Stocked anchors holding on an impervious sea bed. Section III.Schip en Werf, Nr. 26, Rotterdam, 1972.SAURWALT, K.J. Anchors penetrating and holding on a soft planar sea bed. SectionIV. Schip en Werf, Nr. 16, Rotterdam, 1973.SAURWALT, K.J. Anchors digging in and holding in a soft planar sea bed. Section V.Schip en Werf, Nr. 25, Rotterdam, 1974.SAURWALT, K.J. Explanatory anchor experiments. Section VI. Schip en Werf, Nr. 26,Rotterdam, 1974.URA, T. and YAMAMOTO, Y. A basic study on the stability of anchors. J.S.N.A. ofJapan, Vol. 143, 1978.LEAHY, W.H. and FARRIN, I.M. Determining anchor holding power for model tests.S.N.A.M.E., Vol. 43, London, 1935.COLE, M.W. and BECK, R.W. Small anchor tests to predict full scale holding power.Society of Petroleum Engineers, SPE 2637, 1969.BECK, R.W. Anchor performance tests. OTC-paper No. 1537, presented to the FourthAnnual Offshore Technology Conference, Houston 1972.PUECH, A., MEUNIER, J., and PALLARD, M. Behaviour of anchors in differentsoil conditions. OTC-paper No. 3204, presented to the Tenth Annual OffshoreTechnology Conference, Houston, 1978.BRUCE, Bruce anchors. (Publicity literature.)ANON. New hook anchors. (Publicity literature.)KRAMER, U. Zugwiderstande und Eringdringverhalten von Schiffsankern innichtbindigen Boden. Mtteilungen, Heft 9, 1975. Institut fur Grundbau undBodenmechanik der Technischen Universitat Hannover.ANKER ADVIES BUREAU, Flipper delta anchors. (Publicity literature.)TAYLOR, R. Conventional anchor test results of San Diego and Indian Island.Technical Note 1581, Civil Engineering Laboratory, Port Hueneme, California.TAYLOR, R and ROCKER, K. Conventional anchor test results at Guam. TechnicalNote 1592, CEL, Port Hueneme, California, 1980.TAYLOR R. J. Performance of conventional anchors. Technical Memorandum,42-81-02, CEL, Port Hueneme, California, 1981.

DEADWEIGHT OR GRAVITY ANCHORS

Introduction

Deadweight or gravity anchors consist commonly of concrete or steel blocks, scrap metal orany other material of relatively high density.

Uplift holding capacity is mainly dependent on the weight of the anchor, while horizontalcapacity is a function of the shear strength at the anchor/soil interface. For drainedconditions, (i.e. sand/gravel or long-term static loading on clay) this means that the maincomponent of the horizontal capacity is equal to tan + times the weight minus uplift force.Additional uplift capacity will result from suction effects at the anchor soil interface andfriction forces on embedded parts like skirts. However, uplift forces exceeding the weight ofthe anchors is generally to be treated with care.

For clay and short term loading the horizontal capacity will depend primarily on theundrained shear strength of the soil and on the shape and design of the anchor. Additionalhorizontal capacity will be generated by passive earth pressure against penetrated parts of theanchor like skirts or ribs (see Figure 3.1).

Pi’PV

$I P”I

‘ExX$-------- ----

1 1.--w----u.\ RI,\

‘\ ’--------Y!Figure 3.1. Deadweight anchor with skirts and inclined load.

Large deadweight anchors will be required for anchorage of floating offshore structures. Acircular or rectangular caisson-like structure will be a possible design. After setdown at theseabed, ballasting with gravel, sand, mud or heavier minerals (drilling mud, ore) can becarried out. The experience gained from installation and operation of gravity platforms in theNorth Sea has clearly shown that well controlled positioning, installation and operation oflarge caissons is possible in water depths of 200 m. Extension to deeper water should, inprinciple, not be a major problem.

3.1 Installation

The installation problems for gravity anchors will generally be very similar to the problemsexperienced with installation of gravity platforms. Smaller anchors may be transported onbarges and lowered to the seabed with a crane, while large gravity anchors may have to beconstructed as floating structures and lowered to the seabed by controlled ballasting. Afurther possibility is free-falling gravity anchors, hydrodynamically shaped to be stable and hitthe seabed with the bottom downwards.29

At least for larger anchors, skirts and ribs should be provided to increase the horizontalcapacity of the anchor. Penetration capacity and soil reaction stresses against the basestructure will have to be predicted with sufficient accuracy to prevent installation problems.

Based on the experience from installations in the North Sea, the following expressions havebeen found applicable for evaluation of penetration resistance, R, of steel skirts (i.e thinskirts, t < 0.1 m). See Reference 3.2.

R = kp(dMpqcW + As : k&z) qc(z) d,0

d = depth of tip of penetrating member

kp(z) = empirical coefficient relating qc to end resistance

kf(z) = empirical coefficient relating qc to skin friction

4cw = cone resistance, MPa

*P = tip area of penetrating member, mz

AS = side area of penetrating member, per unit penetration depth, m2/m

The following values have been proposed for KS and Kp in Appendix F4 to the VERITASRules(3.ll.

Most probable (R,,,b) Highest expected (Rmax)Type of soil

kP kf kP k f

0.4 0.03 0.6 0.05

0.3 0.001 0.6 0.003

A prediction method for local soil reaction stresses is presented as well. The analysis methodis based on a non-linear semi-empirical relationship and contains an elastic solution and aplastic (bearing capacity) solution as the outer limits3 jl).

The evaluation of the holding capacity of a deadweight anchor is similar to the stabilityanalysis of a gravity platform. A review of these problems has been given in the FIPpublication Foundations of concrete gravity structures in the North Sea which also covers soilinvestigations, installation and instrumentation procedures” Q.

There are several failure modes which will have to be investigated among which:

bearing capacity failure or sliding due to self weight

uplift failure

lateral sliding under inclined loading

tilting, overturning

seem to be the most important (Figure 3.2).

Bearing capacity failure due to self weight may be a problem during and shortly afterinstallation. It will mainly be restricted to locations with soft clay.

Sliding of anchor blocks downslope due to the anchor’s own weight has been observed. Slopeangles exceeding 10” will have to be considered as a real problem for almost any type ofanchor.

BEARING CAPACITY FAILURE

UPLIFT FAILURE

SLIDING FAILURE

TILTING FAILURE-

Figure 3.2. Various failure modes for gravity anchors.

Uplift failure may result from underestimation of uplift forces, overestimation of unit weightof anchor material, overestimation of uplift capacity of embedded anchor parts and fromsubsidence of the seabed/settlement of anchor for vertically moored platforms with a fixedwire length.

Lateral sliding can result from an unfavourable combination of vertical and horizontal loadcomponents. The maximum horizontal force occurs generally together with the maximumuplift force. The net submerged weight to resist sliding and overturning will simultaneously beat its minimum. Combined with the deteriorating effect of repeated loading on the shearstrength of soil materials, this will generally be the most critical failure mode.

Tilting/overturning can generally be avoided by a correct anchor design. The attachment pointfor the anchor line should be kept as low as possible. The design requirement is:

Overturning moment < Stabilizing moment

Scour, earthquakes and wave-induced instabilities of the seabed are effects that will have to beconsidered in the design. The degree of attention to be paid to these environmental effects willdepend on the location, water depth, soil type, slope of seabed, and such like.

3.2.1 Analysis methodsThe uplift capacity, Fv, is a function of the anchor weight, W and the stability criterion isthus:

F, < W

However, under inclined loading, which will be the design load case for every offshoreanchorage, additional weight may be required to prevent sliding of the anchor.

The horizontal capacity is composed of the sliding resistance along the anchor/soil interface,or at a deeper-seated failure surface under the skirt/rib tips (see Figure 3.2) and the resistanceacting on penetrated parts of the anchor, such as passive earth pressure in front and sidefriction along the sides of the skirts or ribs.

With increasing angle of load inclination from the vertical (catenary mooring, guyed towerconcept) deeper seated failure surfaces may develop and so-called bearing capacity failure maybe critical.

The horizontal capacity can be evaluated by means of a plastic equilibrium analysis such asbearing capacity formulae (Brinch-Hansen(3.5), MeyerhoP36)) or generalized methods of slices(Janbu’3-7), Morgenstern and Price’3.8) and others). When using the latter method, infinitelylong strip foundations are assumed. The shear forces on the ends of the sliding body shouldthus be added.

Effective or total stress analysis methods can be applied, depending on the type of loading(static or dynamic) to be considered. Storm duration and drainage conditions will influenceheavily the degree of possible pore pressure generation and the resulting strength and stiffnessreduction. The effective stress method, combined with pore pressure generation models andconsolidation in a time step analysis, will be the best solution, provided the soil and load datadescribing the complex process can be defined sufficiently accurately. An approach in thisdirection has been presented by Rhaman’3-Y).

However, for most practical cases, either fully drained or fully undrained conditions will beassumed. Whichever analysis method is chosen, one has to consider the deteriorating effectsof repeated loading on the soil shear strength.

Model testing performed in a centrifuge (Rowe’3.10’) may give the designer an improvedunderstanding of the complicated physical mechanism which governs the behaviour of afoundation under cyclic loading and may reveal critical failure modes. See also Andersen etalO.11).

3.2.2 Effect of repeated loading

The effect of repeated loading will depend on the magnitude of the cyclic load components(wave loading) relative to the static or slowly varying loads (pretension, drift forces) as well asthe absolute load level.

As the behaviour of soil is primarily friction dependent, both strength and stiffness are closelyrelated to the effective normal stress, i.e., the average stress carried by the grain structure of asoil material. Repeated shear stress reversals tend to loosen the grain to grain contact andbring the grain structure into a denser configuration. This will induce a gradual transfer ofstress from the grain structure to the pore water, which then will be squeezed out of the soil.If drainage is prevented or partly prevented, due to low permeability (clay) and long drainagedistances, the pore pressure generation process can be faster than the consolidation process.The effective stress will then gradually decrease and so will the strength and stiffness of thesoil.

In the extreme case, liquefaction or a nearly total loss of strength and/or stiffness may resultfrom cyclic loading, causing a total collapse of the foundation or excessive plasticdeformations.

Experience from laboratory investigations indicates that below a certain cyclic load level(TC/& Chy Or Tc/Ovi sand: ~~ = cyclic shear stress amplitude, S, = undrained shear strength

Of clay, Uvi = initial effective overburden pressure) the effects of repeated loading will beinsignificant. The critical stress level depends heavily on the density of the soil.

Another problem which has to be dealt with when analyzing a foundation under wave (orearthquake) loading is the random time history of the shear stress reversals. Simplifyingassumptions will have to be made to allow an evaluation of pore pressure build-up andstiffness and strength reductions due to a random load history. See also Reference 3.12.

A comprehensive review of the state of art regarding cyclic loading of sand was given bySeed(-‘,“). Pore pressure ‘build-up’ models have been presented by Martin3.14) Finn et a1’3.15)which allow the designer to take random time histories into account. See also Kvalstad andDahlberg(3,‘8).

An industry-sponsored research project ‘Repeated Loading of Clay’ carried out in 1974 to1975 (Andersen’3-1h) and Andersen et a1(3.17)), resulted in improved knowledge of the behaviourof clay under repeated shear stress reversals.

The effect of combined static and cyclic shear stress on clay has been investigated withinthe above mentioned research project and further studies have been carried out byVERITAS’3.12, 3.1R). The results of these investigations indicate clearly that two-way loading hasa stronger deteriorating effect than one-way loading. With increasing permanent stress level,the permanent deformation rate increased. However, the magnitude of the cyclic shear stressdominated the process.

3.3 Summary of deadweight anchors

The efficiency (holding capacity/dry weight) will necessarily be considerably lower fordeadweight anchors than for other types as the weight and strength of the surrounding soilis mobilized only to a limited extent.

Efficiencies will lie in the range 0.4 to 1.0 for vertical as well as lateral loading. The anchormass required will thus be large compared with other anchor types. However, ballasting withcheap material (gravel, sand, mud, etc.) will increase cost efficiency (holding capacity/costs)considerably.

The safety against uplift failure is independent of the soil conditions which makes the gravityanchor attractive for tension leg systems with high permanent vertical forces from a safetypoint of view, as there is no risk of a gradual pull-out.

Soil-structure-load interaction is probably better understood for gravity anchors than for anyother anchor type as the stress conditions can be defined with more confidence than for theother anchor types.

The capacity range is nearly unrestricted, and with today’s technology gravity anchors willoffer a safe high capacity anchoring alternative provided that thorough soil investigation iscarried out to produce the pertinent soil properties as outlined in Reference 3.4

3.4 References

3 .1 VERITAS. Rules for the design, construction and inspection of offshore structures;Appendix F: Foundations, Det norske VERITAS, 1977. H$vik, Norway.

3.2 LUNNE, T. and St. JOHN, H. The use of cone penetration tests to computepenetration resistance of steel skirts underneath North Sea gravity platforms,Proceedings of the Seventh European Conference on Soil Mechanics and FoundationEngineering, Brighton, 1979.

3.3 KVALSTAD, T.J. and DAHLBERG, R. Soil reaction stresses on the base structure ofgravity platforms during installation. Proceedings of the Seventh European Conferenceon Soil Mechanics and Foundation Engineering, Brighton, 1979.

3.4 FIP. Foundation of concrete gravity structures in the North Sea. FIP State-of-ArtReport 1978.

BRINCH HANSEN, J. A revised and extended formula for bearing capacity. TheDanish Geotechnical Institute, Bulletin, No. 28, 1970.MEYERHOF, G.G. Some recent research on bearing capacity. Canadian GeotechnicalJournal, Vol 1, No. 1, 1963.JANBU, N. Slope stability computations, embankment dam engineering. CasagrandeVolume. (Editors: R.C. Hirschfeld and S.J. Poulos) John Wiley and Sons Inc., NewYork, 1973.MORGENSTERN, N.R. and PRICE, V.E. The analysis of the stability of general slipsurfaces. Geotechnique, Vol. 15, No. 1, 1965.RAHMAN, M.S. Analysis for wave induced liquefaction in relation to off-shoreconstruction. Ph.D. thesis, University of California, Berkeley, 1977.ROWE, R.W. Displacement and failure modes of model offshore gravity platformsfounded on clay. Conference Paper 01-75, 218.1, presented to Offshore Europe 1975,Aberdeen, Scotland.ANDERSEN, K.H., SELNES, P.B., ROWE, R.W. and CRAIG, W.H. Prediction andobservation of a model gravity platform on drammen clay, Proceedings of SecondInternational Conference on the Behaviour of Offshore Structures, BOSS ‘79, London.KVALSTAD, T.J. and DAHLBERG, R. Cyclic behaviour of clay as measured inlaboratory. International Symposium on Soils under Cyclic and Transient Loading,Swansea, 1980.SEED, H.B. Evaluation of soil liquefaction effects on level ground during earthquakes.A State-of-art Paper, Proceedings ASCE, National Convention, September 1976MARTIN, P.P. Non-linear methods for dynamic analysis of ground response. Ph.D.thesis, University of California, Berkeley, 1975.FINN, W.D., LIAM, LEE, K.W. and MARTIN, G.R. Effective stress model forliquefaction, Journal of Soil Mechanics and Foundations Division, AXE, Vol. 103,GT6, June 1977.ANDERSEN, K.H. Behaviour of clay subjected to undrained cyclic loading, BOSS ‘76,Proceedings of the First International Conference on Behaviour of Offshore Structures,Vol. 1, pp 392-403, Trondheim 1976. Also in Norwegian Geotechnical Institute,Publication No. 114.ANDERSEN, K.H. BROWN, S.F., FOSS, I., POOL, J.H and ROSENBRAND, W.F.Effect of cyclic loading on clay behaviour. Paper presented to the Conference onDesign and Construction of Offshore Structures, London 1976. (Also published in:Norwegian Geotechnical Institute) Publication No. 113.FOSS, I., DAHLBERG, R. and KVALSTAD, T. Design of foundations of gravitystructures against failure in cyclic loading, Proceedings of the Tenth Annual OffshoreTechnology Conference, 8-11 May, 1978, Paper No. OTC 3114, Houston, Texas.

PILE ANCHORS

Introduction

Pile anchors develop their vertical holding capacity by mobilizing the side friction/adhesion atthe interface pile/soil and by the weight of the pile (Figure 4.1). Under-reamed (belled) pileswill behave as a combination of a deep embedment anchor and a pile (Figure 4.2).

The resistance against lateral loading is a function of the shear strength of the upper metres ofthe soil as well as the stiffness and bending moment capacity of the pile itself. The horizontalcapacity can be increased considerably by adding special elements such as skirts or wings tothe pile top (Figure 4.3).

With today’s installation equipment, huge piles can be installed in almost any kind of soil.Pile diameters exceeding 2 m have been installed and pile lengths towards 150 m have beenreported. Design axial capacities in the range 30 to 50 MN are now common in the North Seafor jacket foundations.

Recently, pile anchors with a diameter of 2.13 m designed to take 8 MN horizontal static loadhave been installed in dense sand in the North Sea.

Large diameter piles with the attachment point for the mooring line below seabed willprobably be able to take lateral loads exceeding 20 MN in dense sand and overconsolidated

Figure 4.1. Simple pile anchor. Figure 4.2. Belled pile.

Figure 4.3. Special elements to increase lateral capactiy.

Figure 4.4. Pile anchor with attachment point below seabed level.Pile anchor with attachment point below seabed level.

Figure 4.4.

-v-v-&

currenl

ifa-,--- --

SL -SL-0

Stage 1: Pile ]ust above mudline Stage 2: Pile touches mudlIne and an up-stream posltton error 01 vessel IS assumed.

&__-_ _- -_- -

H o i s t s l a c k e n e d

d e c r e a s i n g

P increasino II -

P U P P E T S Y S T E M

Stage 1: Hammer begms operatmg.Sage 3: Selfwelght makes Ihe pile penetrates e a f l o o r .

Subsequent stages from pile landing to first hammer blow.

H y d r o b l o k h a m m e r

Figure 4.5. Underwater driving of piles without lateral support (from Reference 4.2).

clays (Figure 4.4). A smaller version of this type was installed for a permanent mooringsystem for the Piper A Platform, see Reference 4.17.

For normally consolidated clays, the lateral capacity will be smaller, say in the range 1 to2 MN unless wings or skirts are adopted.

4.1 Installation

Installation of offshore piles is a comprehensive theme which can only be touched to a limitedextent within this Report.

There are different methods of installation which all have shown their feasibility under variousconditions such as:

driving

drilling and grouting

jetting

or a combination of these, like combined drilling and driving, or jetting and driving.

4.1.1 Pile driving

Driving is the most straightforward and generally the most economical method, if therequired penetration depth can be reached, i.e., suitable soil conditions and efficient drivingequipment.

As no lateral support will exist when installing anchor piles, underwater driving will be thebest solution. The efficiency of underwater hammers is still increasing and the technique hasalready developed considerably (see Figure 4.5). The successful installation of piles on theCognac platform was performed with underwater hammers in 300 m water depth.

Pile driving should be carried out in a nearly continuous process. If driving is interrupted fora certain length of time, the driving resistance may increase considerably. This phenomenonknown as ‘set up’ is commonly observed when driving in clay and is probably caused byconsolidation of the remoulded material surrounding the pile or thixotropy processes.

SETTING AN UNDERWATER STRUCTURE

b(cl Secton. v,ew lllustratang t h e

cementmg around the ex-ter,or of the anchor structure.

(d) Sectional VIEW lllustratmg thecement,ng of the nnterlor ofthe anchor structure dungremoval of the drill string.

(e) Vkw 5~rn~l.x t o lllustratlngthe pos,t,on of the drillstmg when a t IS free ofthe anchor str”ct”W andthe pos,t~o” of the cementI” and around the upper

Elevat,on v,ew partly in sectlon Slmllar view illustrating theof the anchor structure and drilling of the hole with the

drill strmg being lowered In anchor structure moving into

the water and ready to drill. the drilled hole while drilling1s progressing.

port,on of the anchor Structure

Figure 4.6. Installation of a drilled anchor pile (from Reference 4.1).

4.1.2 Insert piles/bored and grouted

If the predicted driving resistance exceeds the capacity of available hammers, the insert pile orthe bored and grouted pile may be an alternative. Experience from exploratory drilling of oilwells and conductor installations has been used on several occasions to install pile anchorswhen conventional anchors fail under proof loading.

The floating installation vessel equipped to perform drilling lowers the drill string to the sea-beds, with the anchor pile releasably connected at a position above the drill bit and theexpansible under-reamer (see Figure 4.6). The drilling progresses in the usual manner withmud circulation and the under-reamer drilling a hole to a diameter greater than the diameterof the pile anchor. The anchor is following the drill string into the hole.

When the predetermined depth has been reached the anchor is grouted by pumping cementthrough the drill bit while the released drill string is lifted out of the anchor pile.

4.1.3 Under-reamingAn economical alternative might be to construct a bell footing which will increase the verticalcapacity considerably. After driving the pile to a predetermined depth, a hole is drilled beyondthe pile tip and an under-ream constructed as shown in Figure 4.2. By using drilling muds, abell may be established in almost any type of soil material. A reinforcement cage can beinserted to increase the strength of the concrete bell. This is more important for tension pilesthan for compression piles due to the low tensile strength of the concrete.

Under-reams with a diameter of more than 5 m have been constructed in a hard clay layer atapproximately 40 m depth at the Ekofisk field.

4.1.4 Pile driving analysisDriven piles will probably be the most economic alternative for pile anchors, unless theinstallation process can be combined with drilling of production wells or other favourablecircumstances which make the drilled pile an interesting alternative.

A prediction of driveability will be required for choice of driving equipment, and to check ifthe pile can be driven to the design depth.

Today the most commonly used prediction method for evaluation of pile driveability/hammerefficiency is the wave equation method. (Smith) tJ2). Numerous computer programs using thewave propagation theory have been developed. The hammer-pile-soil system is divided into aspring-mass system as shown in Figure 4.7.a. The hammer impact characteristics can bemodelled in different ways and the generated force wave is followed down the pile length in atime step analysis.

The soil model commonly used is shown schematically in Figure 4.7.b. The model requiresinput of the maximum static side and tip resistance, the displacement (quake) necessary tomobilize maximum resistance and the viscous damping factor J.

A common assumption is that the maximum soil resistance during driving is equal to the staticpile resistance. As remoulding effects and pore pressure build-up will be induced during piledriving the above assumption tends to overestimate the driving resistance.

Recent developments include the effects of an internal soil plug in the analysis and research isbeing carried out to establish better values for soil damping and quake.

Full scale measurements of pile behaviour during driving and redriving after set-up deliversvaluable data which may improve our understanding of the various factors influencing thedriving resistance.

4.2 Holding capacity of pile anchors

Pile anchors may be subjected to a combination of vertical and horizontal loads and acombination of static and dynamic forces. In the following, a short review of current design

Figure 4.7a.

r-O(m) +I

SIDEFRICTIONAL

RESISTANCE

POINTRESISTANCE

Idealization of soil-pile-hammer system.

STATIC

\ nh(m) JV(m,l)

REFORMATION

OYNAMIC

Figure 4.7b. Static and dynamic soil modelcommonly used in pile analysis.

methods with comments on some of the most significant effects which may influence theholding capacity is given.

4.2. I Axial capacity of anchor piles

The axial capacity is a function of the shear strength along the pile-soil interface. As cohesiveas well as cohesionless soil can be treated primarily as friction materials, the shear strengthwill depend on the lateral (horizontal) stress acting at the pile-soil interface and the frictioncoefficient tan 6.

The lateral stress is mainly a function of the overburden pressure, the density of the materialand the installation method. A driven pile (open, plugged or closed ended) will displacematerial as the pile tip penetrates and increase the total lateral stress.

In soils with a low sensitivity, this will result in an increased shear strength afterconsolidation. In sensitive clays and cemented sands, the opposite effect may occur and

inserted and grouted piles may give higher capacities. High pressure grouting may increase thecapacity further.

The shaft friction, fs, can be expressed as:

fs = K.p,+tan&

K = coefficient of earth pressure

PO = effective overburden pressure

tand = coefficient of friction at pile-soil interface

As K-p, is equal to the normal effective stress the value of K should reflect the abovementioned influence of initial density and installation procedure.

For loose to medium dense sand and normally consolidated clay a reasonable conservativevalue for K would be K, and based on the expression

K = K, = 1 - sin+

typical values for (KetanQ will be 0.25 to 0.30.

For overconsolidated clays and dense sand higher values will apply. The determination ofK-values for design is a delicate matter, however.

Side friction will not increase linearly with depth for long piles. Experiments show that alimiting value offs will be reached at a certain depth (critical depth). The limiting value ishighly dependent on the installation procedure.

Evaluation with the above described method, often called the p-method (j3 = Katand) hasmainly been used for piles in cohesionless soil but has in the later years found increasing usefor piles in clay as well. Continuous research is going on to improve effective stress analysis ofpiles in clay’4.s-4.10).

The method most widely used for determination of side friction in clay has been the a-methodor total stress method

The empirical factor a varies in the range 0.2 to 1.5 depending on the undrained shearstrength, S,. The higher values apply for lower shear strengths. The scatter in measureda-values is considerable, reflecting the difficulties in determining S, from more or lessdisturbed soil samples, the effect of different installation methods, soil sensitivity, etc.

Another ‘popular’ method is the A-method(4-4?

fs vav = A (C’ VaV + 2SUaV)

A = empirical factor

C’VaV = mean effective overburden pressure between mudline and pile tip

SUav = average undrained shear strength along the pile shaft

which makes the shaft friction dependent upon both undrained shear strength and effectiveoverburden pressure.4 0

- - - -

tLOCATIO” SYMBOL SOURCE___ - -

DLTIDIT

YDlt*NI.

C L E V E L A N D

DRAYTON

NORTN S E A

LEYOORE

ST.NYOIE

NEW O R L E A N S

V E N I C E

ALLIANCL

DONALDSONVILCE

“SC “ousTo*

5.11 Fll.NtlfCD

e.I)ITISH CoLVYnl.

8URNSIDE

0 NOUSEL

:YAM””

CECX0 PECXA FOX

WOOD14

IOYLIUS

A OLESSL’

m UC C L E L

0 MC &EL

0 D.““.GI

a YC CLEL

YC C.“”

0 l ECX

IIDQW,LAND

L A N D

Figure 4.8. Frictional capacity coefficient, y, vs. pile penetration (from Reference 4.4).

Values of A from a large number of pile tests are shown plotted versus pile length in Figure4.8, including also a few tension tests which seem to fall into the same range as thecompression tests.

Regulating authorities such as the American Petroleum Institute (API), Department ofEnergy, UK (DEn), and Det norske Veritas (DnV) have issued regulations specifying designcriteria and limit values for side friction’” Y ?I.

4.2.2 Effect of tension versus compression

Most of the available data have been evaluated from compression tests and various authorshave shown that the shaft friction is often considerably lower under tensile test loading thanunder compressive loading. Some of the reasons for this might be:

reduction of vertical stress around tension pile under static load, due to pile-soil interaction,tends to reduce also the lateral stress at the pile soil interface, while compression piles tend toincrease the normal stress beyond the overburden pressure.

pile tension tends to decrease the pile diameter and thus the normal stress while compressiveforces will increase the diameter.

4.2.3 Effect of cyclic loading (axiaoThe main effects of cyclic loading are as follows:

build-up of pore water pressure and reduction of the normal effective stress at the pile-soilinterface, i.e., a sort of temporary degradation.

densification (consolidation) of material surrounding piles which may reduce the normal stressat pile-soil interface (arching), i.e., a sort of long term, permanent degradation

increasing permanent deformations when combined with static tension, (a sort of cyclic creep)which eventually may lead to pull-out of a tension pile.

To predict the degree of capacity reduction to be expected due to cyclic loading is a complexmatter which is further complicated by the flexibility of the pile itself, and eventually by theadjacent piles of a pile group.

Two-way cyclic loading will be predominant in the upper part of a long flexible tension pileand may induce a very significant degradation in side friction. This loss in capacity will haveto be transferred down along the pile and cause a gradual increase in the load level in thelower part of the pile which will be subjected to one-way loading. Although experience showsthat this type of loading will not induce significant degradation, a steady increase in load levelmay lead to increased creep rates.

A considerable research effort will be required to establish rational and safe design criteria forpermanent tension piles with additional cyclic loading. Research projects have already beeninitiated both in USA, UK, France and Norway with the aim of developing tension pile designcriteria for anchorage of vertically moored structures (tension leg platforms).

The working group have carried out a survey to determine the research and eight projectswere identified. The results of this survey which was carried out by W.S. Rigden andR. Sullivan are summarized in Appendix 4A of this section.

(a) LOADING (b) DISPLACEY?NT (c) SOIL PRESSURE DIS-CONDITIONS LACErlENT CHARACTERISTICS

Figure 4.9. p-y curves for analysis of laterally loaded piles.

4.3. I

Effect of lateral deflection

The effect of combined tension and lateral loading is another point where our understandingneeds to be improved. Lateral loading will influence the axial capacity, at least to a certaindepth below mudline, depending on the soil strength, the flexibility of the pile and the loadlevel. Repeated load reversals and change in load direction may have a deteriorating effect onthe contact between pile and soil. In clays (especially stiff clays) a permanent gap may resultwhile in sands a continuous fill-up with surface material will probably take place. This effectfrom lateral loading could be taken into account by reducing or neglecting the axial capacityof the pile to some depth below seabed.

Lateral capacity of pilesThe lateral capacity of a pile or a pile group is a function of the soil strength and stiffness,the pile dimensions and spacing flexibility and bending moment capacity and the load type;static, cyclic, multidirectional, and such like.

The ultimate capacity of the soil may exceed the bending capacity of the pile, and generally asoil-pile interaction analysis will be required where the soil and pile deflection characteristicsare modelled as shown in Figure 4.9. Here the soil is represented by a so-called Winkler-foundation where the soil is replaced by springs having deformation properties expressed asso-called p-y-curves, Figure 4.10. No interaction between the soil elements is incorporated inthis model.

Other analysis methods exist such as the elastic half-space theory and finite element methodswhich take account of element interaction.

Q.z / klNITIAL

PILE DEFLECTION, y

Figure 4.10. General shape ofa p-y curve.

b/B) y 3b/BO

Figure 4.11. p-y curve for sand.(from Reference 4.12).

Figure 4.12. Typical p-y curve for stiff clay(from Reference 4.13).

Figure 4.13. p-y curves for extreme load (a) and subsequent repeated loading (b)(from Reference 4.6).

Figure 4.14. Modification of p-y curve for cyclic loading in clay (from Reference 4.14).

PIPu --

/ (4-w - w - e - - - - - - - - - - -

Figure 4.15. Modification of p-y curve for reloading (from Reference 4.14).

4.3.2 p-y-curves

For practical purposes the p-y-curve approach seems to be sufficiently accurate and has foundwide use in the design of offshore pile foundations.

Various authors have presented different shapes of p-y-curves based on test experience (modeland full scale). Some of these curves are shown in Figures 4.11 and 4.15 based on References(4.12-4.14 and 4.6).

For further details about the shape of the curves and determination of the curve parametersfrom soil investigations we refer to the above references and to the rules of API, DEn andDnV.

4.3.3 Ultimate capacity

The ultimate capacity near the surface can be evaluated by a wedge stability analysis (seeFigure 4.16). Broms’4.1S) proposed:

pa = 3Kpy’z

for cohesionless soils and other authors have proposed more complex expressions(412)

For shallow failure in clay Gill and Demars’4.11) proposed

pa = 0.25 Su ’

z = depth below mudline

b = pile diameter

A more general expression is

pu = N,.S,

NC = bearing capacity factor dependent on relative depth z/b

/passive ,’

(0) (bl

Figure 4.16. Wedge analysis of ultimate lateral capacity near seabed level.

1I fIIa1 t4 II I

P r a n d t l ( 1 9 2 0 )

Nr = 5 . 1 4

T e r z a g h i ( 1 9 5 5 )

N = 5 . 7'r

S k e m p t o n ( 1 9 5 1 )

Nr * 7 . 5

With side friction

Nr % 9 . 5

T e r z a g h i (1955)

Nr = 11 .4

Figure 4.17. Lateral flow of soil, idealized examples for clay.

For larger depths a local failure rather than a wedge failure will be the case. Some typicalflow patterns around a pile in clay are shown in Figure 4.17.

Corresponding to the findings for deeply embedded plate anchors NC factors in the range 6 to12 will apply.

For cohesionless soils the expression

pu = Nq-y ‘z

may apply. Again referring to deeply embedded plate anchors Nq - values may be extractedfrom Figure 3.

4.4 Summary of Anchor Piles

Anchor piles can, with present technology, be installed and designed to take very large tensileand lateral loads and represent an interesting alternative for permanent anchorage of largefloating structures.

However, there is need for improved methods for prediction of holding capacity both laterallyand vertically. In particular, the combination of static and dynamic tension loads, which is animportant loading case for anchoring piles, is poorly covered by most existing investigations.Effective stress analysis methods will probably be further developed in the near future, andcombined with the information from small, medium and large scale, well instrumented modeltests with static and cyclic loadings it is expected that this may lead to a more rational andsafe design of anchor piles.

As for the other anchor types, a soil investigation will be required to produce the soilparameters necessary for design calculations; shallow borings and CPT’s for evaluation oflateral capacity and deeper borings eventually combined with down-the-hole CPT’s or othertests for evaluation of the axial capacity.

A very broad review of pile design and installation was presented by H. St. John’4.16). Thisreport goes in depth on most aspects regarding piles and contains more complete referencelists than could be included in this Report.

4.5 References

4 .14.24.3

4.164.17

CARMICHAEL, F.R. Offshore Drilling Technology, Noyes Data Corporation, 1975.Hydroblok, Publicity Material.SMITH A.E.L. Pile driving analysis by the wave equation. Proceedings AXE, Civ.Eng. Div., Vol. 127, No. 1, 1962.VIJAYVERGIA, V.N. and FOCHT, J.A. Jr. A new way to predict capacity of piles inclay. Proceedings of the Eighth Annual Offshore Technology Conference. Vol. 2. PaperNo. 1718, Houston, Texas, 1972.API. API RP 2A, Recommended practice for planning, designing and constructingfixed offshore plat orms,f American Petroleum Institute, Eighth Edition, Dallas, Texas,1977.Det norske VERITAS: Rules for the design, construction and inspection of offshorestructures. H#vik, Norway, 1977.DEPARTMENT OF ENERGY. Guidance on the design and construction of offshoreinstallations. London, 1977.KIRBY, R.C. and WROTH, C.P. Application of critical state soil mechanics to theprediction of axial capacity for driven piles in clay. Proceedings of the Ninth AnnualOffshore Technical Conference. Paper No. 2942, Houston, Texas, 1977.ESRIG, M.K., KIRBY, R.C., BEA, R.G. and MURPHY, B.S. Initial development of ageneral effective stress method for the prediction of axial capacity for driven piles inclay. Proceedings of the Ninth Annual Offshore Technical Conference. Paper No. 2943,Houston, Texas, 1977.CARTER, J.P., RANDOLPH, M.F. and WROTH, C.P. Stress and pore pressurechanges in clay during and after the expansion of a cylindrical cavity. InternationalJournal for Numerical and Analytical Methods in Geomechanics, Vol. 3, 1979pp 305-322.GILL, H.L. and DEMARS, K.R. Displacement of laterally loaded structures in non-linearly responsive soil. Technical Report, R 670, Naval Civil Engineering Laboratory,Port Hueneme, California, April 1970.REESE, L.C., COX, W.R. and GRUBBS, B.R. Analysis of laterally loaded piles insand. Preprint, Offshore Technology Conference. OTC paper 2030, pp 743-483,Houston, Texas, 1974.REESE, L.C., COX, W.R. and KOOP, F.D. Field testing and analysis of laterallyloaded piles in stiff clay. Preprints, Offshore Technology Conference, OTC paper 2312,pp 641-690, 1975.MATLOCK, H. Correlations for design of laterally loaded piles in soft clay. Preprint,Offshore Technology Conference, OTC paper 1204, Vol. I, pp 544-595, Houston,Texas, 1970.BROMS, B.B. Lateral resistance of piles in cohesionless soils. Proceedings AXE, I.Soil Mechanics and Foundation Engineering, Vol. 90, SM3, 1964.ST. JOHN, H. CIRIA, Offshore Technology Paper 5 1980, OT-R-8002.LOWE, E.A. AND YIN, V.C. Permanent mooring system for large construction andsupport vessels in the Piper Field. Journal of Petroleum Technology, April 1980.

APPENDIX 4A: REVIEW OF RESEARCH INTO THE CAPACITY OF PILES INTENSION

Information has been obtained on eight programmes, five of which have been completed, andthree are still underway. Brief details of each programme are given together with a contactaddress for further information.

1. ‘Library’ of Pile Test Data

Contacts: Professor R.E. Olson9202 Knoll Crest LoopAustin, Texas, 78759

Sponsor: American Petroleum Institute.

This project collects data for published pile load test cases, for all types of pile in all soiltypes. The data is summarised in a library for ready access. Pertinent summary data are in acomputer bank and a program has been written to use the data bank. Data may be selected toaccord with pile types, pile size, soil strengths and various other variables for similarsummaries or for analyses. Pile capacities under axial compressive or tensile loads may beanalysed using a variety of methods on the computer, and then the measured and computedpile capacities are automatically plotted against each other. The program can also performrelevant statistical calculations to determine the relative accuracy of existing methods ofprediction and to help develop improved methods.

2. Cyclic Loading of Piles and Pile Anchors at Haga

Contacts: Norwegian Geotechnical InstitutePostboks 40,Taasen Oslo 8, Norway

Sponsor: Ten oil companies

The programme presented here consists of about 20 field model pile tests. One axial and onelateral static test will first be carried out as ‘reference tests’. Afterwards static loads of varyingintensities will be applied together with sequences of cyclic loads in order to examine theeffect of cyclic loading on the original static capacity. Laterally loaded piles will be exposed totwo-way cyclic loading in order to examine the soil behaviour; development of deformationand strength degradation.

A. Pile type

All piles are of the dimensions:

Diameter = 0.15 m

Wall thickness = 0.0015 m

Length = 5.0 m axial loads4.0 m lateral loads

B. Soil typeThe test site consists of moderately to heavily over-consolidated clay (OCR = 2 - 20). Theclay is medium stiff (undrained shear strength of the order of 45 kN/m2 measured by in-situvane) and of medium plasticity (PI = 15 - 20%).

C. Instrumentation details

The following parameters will be monitored:

Pore pressures at pile surface and in surrounding soil

Earth pressures at pile surface

Pile strains

Applied loads

Deformation at pile head

Inclination at pile head.

3. The BRE Anchor Pile Programme

Contacts: Dr. H. St. JohnBuilding Research EstablishmentGarston,Watford, WD2 7JR

Sponsor: Department of Energy (UK)British PetroleumBritish National Oil Corporation

This programme is the most comprehensive under way at the present time. Driven piles,closed and open ended, and drilled and grouted piles will be tested in tension and

compression, under static and cyclic loading and under the action of combined axial andlateral loading. The piles are installed in glacial till and gault clay at the BRE Test Site atCowden.

The instrumentation is comprehensive, and includes displacement measurements down thepile, measurements of ground heave, and pore pressure measurements in the surrounding soilas well as the usual pile head load and strain gauge measurements.

This programme is due for completion by 1983.

4 . The Arcolprod Foundation Programme

Contacts: Taylor Woodrow345 Ruislip Road,Southall,Middx., UBl 2QX

Sponsors: Taylor W oodrowDepartment of EnergyEEC Directorate General for Energy

This was part of a large research programme into the performance of a compliant structure.Fully instrumented steel piles were jacked into the glacial till at the BRE test site at Cowden.The piles were tested in tension and compression under cyclic and static loading conditions,and under combined axial and horizontal loading.

The two year programme was completed in 1980.

5. Shell Coil Pile Test

Contacts: Shell Oil CompanyPO Box 2099Houston, Texas 77001

Sponsors: A number of oil companies.

A single 30 in. diameter x 1.5 in. wall thickness pile was driven deep into a silty clay. A 74 fttest section stratum was instrumented, and tested statically and cyclically to failure in tension,immediately after driving and again 45 days and 90 days after installation, testing beingcompleted mid 1980.

6 . Static and Cyclic Tests in Stiff ClaySponsor and contact: Heereme Engineering Service BV

LeidenHolland

This programme investigated the performance of 24 in. o.d. piles driven into a stiff clay.Most piles were tested in compression but two were tested in tension statically and under asmall number of cycles. The piles were instrumented with strain gauges. Testing wascompleted in early 1975 and the results reported in OTC 3490 Pile driving and static load testson piles in stiff clay by E.P. Heereme (1979).

7. The Empire Pile Load Tests

Contacts: Chevron Oil Field ResearchPO Box 446La Habra, California 90631

Sponsors: Fourteen Oil Companies.

Test sections of 14 in. o.d. up to 50 ft long pipe piles were installed at depths of up to 358 ft.in firm to stiff highly plastic clay. Numerous different axial loading cases were carried out,including static and cyclic tension and compression. The test sections were comprehensivelyinstrumentated.

Testing was completed by the end of 1975.

Reference: The results have been summarised in two papers:

Cox, Kraft, and Verner Axial load tests on 14 in. pipe piles in clay, OTC 1979.

Kraft, Cox and Verner Pile load tests; cylic loads + varying load rates.

ASCE, Geotechnical Engineering Division 107GT. 11. 1-19 January 1982.

8. The ESACC ProgrammeContacts: Amoco Production Co.

Research Centre,PO Box 591Tulsa, OK 74102Attn. Mr Murphy

Sponsors: Group Oil Companies and Government Organisation.

This programme included theoretical studies, laboratory model testing, and small scale fieldtests. The field trials involved 4% in. o.d. driven and jacked, open and closed ended pipepiles, with up to 40 ft penetration in San Francisco Bay Mud. The loading was axial in staticcompression and tension.

The theoretical aspects of the study were probably the most interesting recent insight into thebehaviour of a pile under load. The programme was completed in January 1980.

ROCK ANCHORS

Introduction

Rock anchors develop their vertical holding capacity from the effective transfer of loading tothe weight of rock in an inverted cone shaped volume located by the depth of penetration

Anchoruge device

Inverted Rockcone

BondedLength

Figure 5.1.

Rock-anchor.

achieved by the anchorage device. The transfer of loading can be provided by a method ofbonding the length of anchorage device to the rock (Figure 5.1). Horizontal loads aretransferred by the direct bearing of the anchorage device against the face of the upper rocklayers.

A rock anchor may fail in one or more of the following modes(5.1* 5.2):

(a) by uplift failure of the rock mass

(b) by failure of the rock/grout bond

(c) by failure of the grout/anchorage device bond

(d) by failure of the anchorage device

(e) by lateral crushing of the upper rock layers

At the time of writing (1980), it is not possible to predict theoretically the geometry of thefailure zone for vertical or inclined anchorages installed in rock. Empirical methods ofanalysis are given in References 5.1 and 5.2 but there is little experimental or practicalevidence or theoretical data to substantiate the methods. For the larger anchorage forcesabove 1MN there is very little data available on the engineeing properties of types of seafloorrock in submerged conditions to the depths needed for resistance to this magnitude of force.

Where groups of rock anchors are closely spaced the interaction of inverted cones has to beconsidered in the overall stability analysis (Figure 5.2). This has the effect of introducing aflat vertical plane at the interface of adjoining cones, and the depth of anchorage needs to begreater for the mobilisation of the same rock weight per anchor as for the single anchor.

Deepwater techniques for anchoring in rock are limited to tension piles (Figure 5.3) whichrequires a previously drilled hole in the seabed before installation and grouting in place”-“.The type of drilling involved has been successfully undertaken in water depths up to 600 ft,and the technique is very similar to that used for offshore oil-drilling methods. There is asevere limitation on the type of floating drilling equipment available for this work in deepwater, and the costs of drilled pile anchors is thus very great indeed. Other types of rockanchors are generally of low order uplift resistance and require installation by divers whichlimits their application to water depths of up to 150 ft. The most generally adopted anchor

Common I nterf as

Figure 5.2. Group anchors.

Drilled Hole

Figure 5.3. Drilled pile rock anchor.

Figure

Rock bolt anchor.

for this type of work is the expanding rock bolt (Figure 5.4). These are located in previouslydrilled holes and by driving into the rock, the expanding end of the bolt obtains the requiredanchorage. It has not been usual to group up this type of anchorage, and little evidence existson their uplift load capacity for long term use’5.“.

Rock Mechanics

Application of the friction cone concept in rock mechanics is used in the analysis of stabilityproblems. The semi-apical angle of the rock cone is the friction angle of the rock material

neglecting the influences of geological discontinuities. The angle of internal friction rangesbetween 35 to 45” in hard igneous rocks (granite and basalt) and 25 to 35” in soft sedimentaryrocks (chalk and sha1e)(5-6).

The friction cone concept can be extended to include cohesion as the additional materialproperty. In most practical stability problems, cohesion is assumed to be a zero since it is adifficult parameter to evaluate and since by ignoring it is to err on the side of safety. Withanchorages resisting permanent tension loads, cohesion should be disregarded.

In rock mechanics, the stability of shallow rock below the ground surface is stronglyinfluenced by the structural conditions and the degree of weathering of the rock mass whereasrock stability below about 100 ft. depth depends more upon the response of the rock mass tothe stress field induced around an opening-. .(( ‘) Consequently the stability of underwater rockanchorages should consider both the degree of weathering of the rock mass and the geologicaldiscontinuities. An investigation should be made of rock conditions at a potential anchor siteby:

(a) Studying the regional geology from sub-bottom geophysical survey and securing rock coresby diamond drilling for identifying and mapping geological discontinuities.

(b) Laboratory testing of intact rock samples to evaluate the shear strength of the rock mass.

5.2 Design Considerations

It is necessary to consider the methods in current use to estimate the anchorage depth requiredto ensure that the anchorage loads will be resisted safely without failure by one of the modes(a) - (e) described in 5.0.

The vertical uplift capacity is usually evaluated as the weight of the rock cone using thesubmerged density of the rock. For groups of anchorage cones a combined action of thewedge of rock weight thus mobilised is taken into account. For conceptual design prior toconducting a site investigation, the angle of inclination between the faces of the inverted conemay be taken as:

(1) 60” when the rock mass is soft, heavily fissured or weathered

(2) 90” in all other cases.

It is uncertain where the position of the apex to the inverted cone should be assumed to act inrespect to the total depth of the installed anchorage device. A conservative estimate is to takethis point to be the middle of the bonded length of the anchorage device. Where there is amethod of achieving any wedging or other lateral holding (i.e. under-reamed) at the end ofthe device then the full depth could be assumed to act.

A formula for calculating the depth of cone needed for overall stability has been presented’5,1)for irregular submerged fissured rock:

Depth of cone (metres)

One anchor Group anchors

’’ 3FxP J(Ifl:sP_((v - 1) n tan 4

where F = factor of safety against failure(usually 2)

S= spacing of anchors (metres)

P = design load of anchorage (MN)

4= angle of friction across fracturesin rock mass

y = bulk density of rock (kN/m-‘)

In considering the possible failure of the rock/grout bonded interface the designconsiderations are based on the following main assumptions:

transfer of the load from the anchorage device to the rock occurs by a uniformly distributedstress acting over the whole of the perimeter of the depth of anchorage

the diameter of the bonded perimeter is the same as that of the anchorage device.

However, these assumptions are open to serious doubt as for high capacity anchorages themethod of developing bond between the rock and the grouted hole is more complex. Muchhigher bond stresses are likely to arise at the upper layers of the anchorage depth, which willhave an influence on overall stability of the foundation. Depending on the rock characteristicsthe working bond stress has been suggested{‘-I) to be in the range of 0.35 (weak) to 1.4 (strong)N/mm2. These values have been shown to provide safe designs from the result of experiencewith regard to the few failures which have been recorded at the rock/grout interface.

In general it is assumed that the failure of the grout/anchorage device bond is not critical ifthe rock/grout interface has been designed to the present very conservative methods. Thisassumes that the anchorage device has been properly inserted into the drilled hole to the samedepth. The embedment of the anchorage device will usually bring into action the type ofsteel/grout bonding characteristics which have been observed in reinforced and prestressedconcrete.

Clearly the avoidance of bond failure will be substantially improved if some method ofsurface roughness or deformation can be provided on the anchorage device.

With the high value of horizontal forces experienced in the anchorage foundation of floatingoffshore structures this aspect of the possible rock failure needs to be carefully considered. Inthe case of drilled piles the individual pile will exert horizontal bearing loads through thegrout to the rock interface at the upper layers. This could cause crushing of the rock leadingto lateral deflection of the pile and related bending stresses. The loss of bond with the rockwhich would also arise could also weaken the vertical uplift bond resistance between thegrout/rock interface. With a knowledge of the rock compression strength in the upper 1 to 3metres below the seafloor the safe lateral resistance of the rock can be calculated andsufficient piles of the necessary diameter can be provided for this purpose.

5.3 Conclusions

The use of rock anchors has been widely adopted for temporary and permanent use on land-based structures, but there is little evidence of their use as seabed foundations. There is a lackof knowledge of the engineering properties of the types of submerged rock likely to beencountered, so that safe design parameters can only be based upon very conservativejudgement at the moment.

Nevertheless, the use of rock anchors in the form of drilled piles is a very attractive solutionwhere the combination of high anchorage forces and/or deepwater locations are consideredfor new developments in offshore structures. This is because the drilled pile can be easilyadjusted to meet seabed irregularities and for a rock foundation this makes it the mostpractical alternative solution. The disadvantage is the high cost and limited availability of thespecialised drilling and grouting equipment.

5.4 References

5.1 LITTLEJOHN and BRUCE. Rock anchors: state of art. Ground Engineering, May1975.

5.2 FEDERATION INTERNATIONALE DE LA PRECONTRAINTE. FIPRecommendations on prestressed ground anchors. (In preparation.)

5.3 TOMLINSON, M. J. Foundation design and construction. Pitman Publishing Co. 1971.pp. 422-425.

5.4 BRITISH STANDARDS INSTITUTION Draft Code CSB/22, Ground Anchors.(Section 3.1.9.)

5.5 MCCORMICK, M.E. (Editor) Anchoring systems. Pergammon Press, 1979 pp. 236-241.

5.6 HOCK, E. and BRAY, J.W. Rock slope engineering. Institution of Mining andMetallurgy, 1977.

5.7 HOCK E and BROWN E.T. Underground excavations in rock. Institution of Miningand Metallurgy, 1980.

Permanent anchorages for offshore structures - imcyc.com Permanent Anchorag… · 1.2 Anchorage forces 1.3 Tethered structures ... for floating offshore structures followed by a ...

Documents

Z - Offshore Structures Overview (Concrete Structures)

Forces on Offshore Structures

15A 1 Offshore Structures

Halbach Permanent Magnet Structures

OFFSHORE STEEL STRUCTURES SPECIFICATION

Support Structures for Offshore Wind Turbines -...

Offshore Concrete Structures

(Offshore Structures) - Wordpress site of PulPong ·...

Introduction to offshore structures

General Specification Offshore Platforms Offshore Structures...

Offshore Structures Versus Land-Based Structures

Inroduction to Offshore Structures

Offshore Structures

Offshore Structures Engineering - EF 2006

Fatigue of Offshore Structures

102229261 Offshore Steel Structures