Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Rigging Selection and Lift Point Design for Heavy Lift Y.S. Choo BSc, MSc, PhD, FASCE, FIMarEST, FRINA, CEng, PEng National University of Singapore, Singapore ABSTRACT This paper presents examples to highlight the engineering considerations for rigging selection and associated lift point designs for a number of heavy lift projects. It also highlights results from systematic field instrumentation on doubled slings that show that current industry recommendation for a 55/45 ratio for sling tension distribution is un-conservative for skew rigging arrangement. This has significant implication on the lift point design, especially for plate trunnions, with regards to moments resulting from the uneven doubled sling tension acting on both sides of the main-plate. Selected results on fabricated pipe trunnions establish the superior performance of pipe trunnions with through-brace arrangement. The associated considerations for attachment of lift point to structure are also highlighted. The paper demonstrates that successful solution for heavy lift project can be readily found through expert knowledge, with consistent selection and proportioning of key components of the lift system. Keywords: Heavy lift, knowledge-based system, padeye, trunnion, lifting, strength, design, fabrication, installation AUTHOR’S BIOGRAPHY Dr Y.S. Choo is Associate Professor in Department of Civil Engineering and Director of Centre for Offshore & Maritime Engineering in National University of Singapore. He was President of Singapore Structural Steel Society (in 1992-94) and served in the Board of Directors, International Society of Offshore & Polar Engineers (in 2000-02). He serves as technical consultant on fabrication and lift installation of major marine and offshore projects. His research interests include installation engineering, knowledge-based system development, and strength of tubular and plated structures. INTRODUCTION Heavy lift is one of the major operations in marine and offshore installation, and is also extensively deployed in shipbuilding. Due to the advancement in heavy lift technology, large modularised ship blocks may be fully outfitted, and then lifted and joined to form the entire ship. Similarly, an offshore structure may be fabricated in a yard, transported to the selected offshore location, and then installed by lifting. Sheerleg crane barges (vessels) are suitable for operations in relatively shallow waters and provide requisite capabilities for lifting large ship blocks or offshore modules. This paper presents the engineering and installation considerations with regards to rigging selection and lift point design for heavy lift through projects which the author has been involved in. The lift installations of these structures (or modules as generically referred to in this paper), each weighing more than 1000 tonne, have been successfully carried out using the Asian Hercules II sheerleg crane barge, and the projects are chosen to illustrate different rigging

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

systems and associated lift points adopted. The module structures were lifted near shore, where lift dynamics was observed to be minimal. RIGGING SELECTION The importance of good conceptual and preliminary designs in the development of safe and cost-effective lifting schemes and procedures has been highlighted by Crowle1 and Fern and Griffin2. Offshore hook-up and commissioning costs are very high as compared to those for doing the same work onshore. Thus, it is necessary to focus on the installation phase and to design the structures for ease of construction and installation. Fern and Griffin2 highlighted that the structural design of the Piper B and Saltire A topsides was dominated by the installation considerations. Incorrect selection of rigging arrangement may lead to damage of components, structural failure or personal injury and may thus have major implications to the project cost and schedule. In recognition of the importance of installation engineering, the author and his colleagues have been involved in research and development on various aspects of heavy lift. Choo et al.3 presented a knowledge-based approach for lift installation using case-based reasoning and demonstrated the capability of the C-LIFT software through liftability studies. The knowledgebased approach was further enhanced over the period, and Choo et al.4 presented the heavy lift design system (H-LIFT) that provided computer-based support to engineers involved with heavy lifts using sheerleg crane barges. The knowledge base incorporated the domain knowledge on heavy lift, design criteria5,6 and research results reported by his group on strength of lift point components, including padeyes7,8, fabricated plate trunnions9-11, fabricated pipe trunnions12,13 and sling load distribution14-16. It is recognized that the maximum dimensions of modules are constrained by the crane capacity and reach of the crane barge, and the minimum clearance requirements between the module and the crane boom due to the rapid fall off in the crane capacity with lift radius (Mayfield & Zimmerman17; Corcoran18; Mawer et al.19). The selection of appropriate rigging configuration is thus an integral part of heavy lift engineering, and is illustrated through three projects in this paper. In Fig 1, a single hook-4 lift point rigging arrangement using 3 spreader bars was used to lift and position a 1200 tonne module onto the Laminaria FPSO (Floating Production, Storage and Offloading) vessel in Sembawang Shipyard (Singapore). In the figure, it can be noted that the module was sitting on the transport barge during preparations for lifting. The transport barge was then towed away after the module was lifted from the barge. The module was then placed onto the FPSO vessel. It can be observed that process equipment was positioned above the main deck level and this excluded the use of a simple single hook-4 sling arrangement. In this case, the provision of an appropriate length top bar, and correct sling lengths to suit the centre of gravity of the lift system, helped ensure that the two bars at the lower level were vertically above the lift points. In this configuration, the adoption of plate trunnions at four strong points of the module enabled placement of the doubled slings to be completed very quickly for lift installation (as illustrated in Fig 2).

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Hook Rigging

Module Crane Barge

Transport Barge

FPSO Vessel

Site

Fig 1. Lift installation of process module using 3-bar rigging system onto FPSO vessel

Doubled slings

Plate trunnion

Plate trunnion

Fig 2. Rigging up for module lift. Riggers positioning doubled slings onto plate trunnions.

Fig 3 shows the lift installation of the second deck module of the Cajun Express semisubmersible vessel using a single hook-4 lift point rigging arrangement In this case, each of the four lift points adopted to connect the slings to the module structure was a fabricated pipe trunnion, which will be described in detail in later sections of this paper. One key consideration in the lifting of the module was potential interference between the module and the A-frame structure of the crane barge (as can be observed in Fig 3).

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Hook

Rigging

Module

Semi-submersible Vessel Crane Barge

Site

Fig 3. Lift installation of side block using single hook-4 slings rigging arrangement

Another lift installation example is shown in Fig 4, with a multi-tier rigging system involving doubled slings (around special sheaves which enabled sling tension equalisation for the doubled slings) and twelve lift points at the module to ensure minimal deformation of the flexible deck panel during lifting. The plan dimensions of the deck panel measured approximately 90m length and 45 m width, and the lift weight was approximately 1100 tonne. In this project, special techniques and algorithms were developed to enable the engineers to optimise selection of the available slings, and to maintain the stresses in the system to be within allowable limits. With the available strong points at the deck column positions (which were located along reference grid-lines with somewhat different spacings), and known sling angles based on the centre of gravity of the lift system, padeyes (such as one shown in Fig 5) were designed for the lift operation. It can be noted from Fig 5 that due to the high bearing stress imposed by the shackle pin onto the padeye hole, cheek plates were welded on both sides of the main plate to ensure that the stresses were within allowable limits. The main plate was slotted to the circular pipe extension and the weld provided resistance to the sling load through in-plane shear and bending. Reinforcement gusset plates were placed to provide out-of-plane support to the main plate.

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Hook

Rigging

Module

Crane Barge Site

Fig 4. Lift installation of Malampaya weather deck using multi-tier rigging system

Main-plate Reinforcement gusset plate cheek-plate Pin-hole

Attachment to structure Fig 5. Padeye Detail (with main plate slotted to pipe extension for offshore deck structure)

HEAVY LIFT DESIGN CONSIDERATIONS Figs 2 to 4 have shown three different module structures successfully lift installed using different rigging systems, and the associated lift points. In each of the figures, the crane barge, site, module, rigging and hook are highlighted to show the variety of rigging systems selected. It may be mentioned that the objective of a heavy lift operation is to safely lift a module from the start (pick-up) site and accurately install it at the target (put-down) site. In heavy lift design, as illustrated in Fig 6, the basic input data from sites (start and target sites), module, and crane barge are consolidated and processed to derive the lift procedure, rigging arrangement and details for rigging components while satisfying the constraints due to structural behaviour, geometrical arrangement and other contingency requirements. Some of the tasks and considerations for heavy lift design involving the use of a sheerleg crane barge are discussed below. Details on design criteria and recommendations can be found in API5, DnV6, Mayfield

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

& Zimmerman17, Mawer et al.19 , Bunce & Wyatt20, Brown & Root21 and Shell22.

Structural Behaviour

Site Module

Heavy Lift Design

Contingency Requirements

Barge

Lift Operation

Geometrical Constraints

Rigging Arrangement

Component Design

Fig 6. Heavy Lift Considerations and Associated Design Tasks

The start and target sites should be investigated for the necessary barge access and manoeuvring. The start site may be the location where the module is fabricated, while the target site may be the location for final system integration. The water depth at the location, geometrical details of the module and other adjacent objects, and relevant characteristics at both start and target sites will affect the accessibility, movement and gesture (that is, barge orientation, boom and jib angles) of the crane barge, and the associated rigging configurations should be considered. The inclination angles of main boom and fly jib of the chosen sheerleg crane barge should be appropriately selected to accomplish the lift task. The selection of these angles is subjected to constraints such as the lift capacity, outreach, module weight and geometry, rigging arrangement and the clearances between various objects involved. Hook height limitation, which is a function of crane boom angle (with associated outreach tied to the boom length) and draft of the crane barge are also important considerations. Lift points are generally located at the available strong points in the module to prevent excessive structural deformation or damage during the lifting operation. These lift points should be selected to allow the lifting forces to flow smoothly into the main structural members. In addition, the compatibility of the rigging selected and the associated lift point type (padeye, plate trunnion or pipe trunnion) requires detailed considerations. In certain cases, reinforcement may be required to strengthen the module and to maintain the geometric dimensions to ensure tight tolerances for assembly. Special attention needs to be given to the local deformation and stresses of the lifted module, as well as the assembled blocks during the assembling operations to ensure that contact and impact forces are minimised. The rigging arrangement to be selected should consider the available strong points in the module and other installation requirements. A spreader bar or frame, with appropriate rigging arrangement, may be used to prevent physical interference and protect the exposed equipment from damage (as seen in Fig 1). A greater sling angle, with respect to the horizontal plane, generally results in proportionally smaller compressive forces acting on the module structure. For the proposed sling angle, the strength and associated capacity of the crane hook prongs needs to be checked.

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

KNOWLEDGE-BASED SYSTEM FOR HEAVY LIFT (H-LIFT) Based on the research and development efforts and knowledge generated over the last decade, the author and his colleagues have developed a knowledge-based system for heavy lift (H-LIFT) which served as an assistant for engineers to evaluate possible rigging schemes and check the suitability of designed components. Choo et al.4 presented details on the H-LIFT software and illustrated its features through design examples. The multi-tier rigging system adopted for the lift installation of the Malampaya weather deck was first developed and implemented within H-LIFT. This enable pre-planning and visualisation of the lift sequence, and Fig 7 shows the crane barge lifting the module from the pick-up location. The actual lift installation using the multi-tier rigging scheme was successfully carried out, as shown in Fig 4.

Fig 7. Pre-planning of the Malampaya weather deck using the H-LIFT knowledge-based system

The use of more lift points helps to prevent the module from local over-stress and excessive deformation during lifting, as shown in Figs 4 and 7 for the lift installation of the Malampaya weather deck. As reported by Selvakumar and Choo23, the flexible module required twelve lift points to be provided such that a rigging system with multiple tiers of doubled slings minimised the distortion during lifting. It is also pertinent to re-iterate that each of the doubled slings for the Malampaya deck lift passed through a smooth sheave system which minimised the frictional effect on the doubled sling. FORCE SPLIT RATIO FOR DOUBLED SLING Doubled slings are commonly used in lift installation of structures, as illustrated in Figs 2 and 4. Current industrial practice recognises that the frictional effect at the hook on the doubled sling may result in a force split ratio α (i.e. ratio of larger to smaller tension experienced by the two arms of the doubled sling, for example T1/T2 as shown in Fig 7) of 55/45=1.22. The author and his colleagues14 found that the force split ratio for un-symmetrical rigging arrangement was

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

higher than 1.22. As a result of the preliminary findings, the author worked with Heerema to conduct systematic field measurements on the effect of sling friction on force split ratio15. The sling friction measurement was conducted on board of the semi-submersible crane vessel, Hermod, in Singapore. The objective of the field measurement is to obtain realistic values of the force split ratio α for doubled slings. Three series of tests for the doubled sling arrangement, including different sling diameters and lengths, and hook blocks with greased and un-greased hook surfaces, were conducted. Fig 8 is a snap-shot of one of the test series B in which each of the two slings is doubled at the crane hook prong. Fig 9 shows the schematic arrangement of the test setup, in which two specially designed padeye assemblies (with rotating feature to cater for different combinations of sling lengths and sling angles) were welded to the Hermod deck. At each end of the doubled sling, one of the two shackles was instrumented with strain gauges attached to the two arms of the shackle and pre-calibrated with given sling tension. Continuous data logging was carried out during the load cycle as the hook load was incrementally increased to the planned magnitude. Post-processing was subsequently performed for each load level, and the force split ratio corresponding to the ratio of TA/TB or TC/TD (as indicated in Fig 9) was computed.

T1 T2

Fig 8. Doubled-sling tension measurement on-board heavy lift vessel, Hermod. (Tension on two arms of doubled sling indicated as T1 and T2 respectively.)

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Hook Load

TD T TA

TB Z

X

Y

Rotating padeye assembly

Fig 9. Schematic arrangement of doubled slings connected to instrumented shackles attached to rotating padeye assembly.

For series B-6, the loading cycle consisted of two stages, in which stage one corresponded to symmetrical rigging configuration when the hook was first positioned vertically above the geometric centre of the test set-up (as shown in Position 1 in Fig 10a) before the hook load was incrementally increased, and subsequently unloaded. For stage two, the crane boom was lowered such that the boom tip had a horizontal offset of 1m (to Position 2 in Fig 10b). The hook load was then incrementally increased, resulting in an un-symmetrical (or skew) rigging configuration.

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Position 2

Position 1 Hook

Hook

Contact angle

T1

T2

T’1

before boom down

T’2 after boom down

Z (a)

(b)

Y

Fig 10. Doubled sling arrangement (shown in Y-Z plane) to measure sling tensions T1 and T2 for symmetrical rigging arrangement (before boom down) and skew rigging arrangement (after boom down).

The force split ratio observed in series B-6 is plotted against time for the two different stages in Fig 11. The measured results show that under symmetrical rigging configuration, the assumed force split ratio of 1.22 provides a reasonable basis for design. However, when the inclined hook load or an unsymmetrical rigging arrangement is adopted, a higher force split ratio (α>1.5) may exist in the two arms of the doubled sling. Based on the other measurements, a larger contact angle (i.e. longer contact length around the hook prong) is found to result in a higher force split ratio which is significantly larger than the value of 1.22 as assumed by industry. The details of the test arrangement, procedure and results are reported by Choo and Ju15, and Choo et al.16. The measured results have direct implication to plate trunnion design with un-symmetrical rigging arrangement as the higher force split ratio for the doubled sling will introduce additional out-of-plane moment to the main plate of the trunnion and may therefore overstress the plate trunnion. 1.8

Force split ratio = (T1/T2)

1.6 1.4 1.2 1.0 0.8

Time to pick ratio before boom down

0.6 0.4

Time to pick ratio after boom down

0.2

Time (s)

0.0 0

50

100

150

200

Fig 11. Variation of force split ratio for Test B-6 for symmetrical rigging configuration (before 100 seconds) and ratio for subsequent skew rigging configuration . Curves with solid squares and circles correspond to doubled slings AB and CD respectively

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

FABRICATED PIPE TRUNNIONS FOR PLATED STRUCTURES In this section, the concept of fabricated pipe trunnions which the author has extended to lift installation of plated structures is presented, as indicated in Figs 12 and 13. In Fig 12, the lift installation of the 2400 tonne centre block using a dual hook – four lift point system was adopted for the Cajun Express project. It can be readily observed in Fig 12b that the deck longitudinal bulkhead has been utilised in the installation of the side blocks onto the four column tops of the semi-submersible. In order to ensure compatibility in the deck deflections for the centre block and the side block (with the longitudinal bulkhead imparting the structural stiffness), a temporary truss consisting of circular hollow section members and two support girders were provided on each of the open sides of the centre block. Upon final placement of the centre block onto the four column top supports, the support girders rested on top of the main deck, thus resulting in a total of eight support points and ensured good matching of the deck plating which was essential for integration of the deck blocks. It is appropriate to highlight that the four locations selected for attaching the pipe trunnions to the centre block were the most appropriate locations due to the intersection of the primary transverse and longitudinal bulkheads. If additional lift points were selected, this might have resulted in relative distortion of the deck during lifting and thus would be counterproductive. The flexibility in proportioning the dimensions of the fabricated pipe trunnions and the more efficient load transfer to the plated structure have demonstrated its distinct advantages over the more conventional padeye design (which would have required significantly thicker main plates and associated reinforcement around the padeye area in view of the diagonal sling configuration). The lower portion of the pipe trunnion, which was pre-slotted, facilitated the placement onto the longitudinal-transverse bulkhead cruciform. After welding, the resulting structural arrangement enhanced the load carrying capacity of the pipe trunnion.

Support girder Pipe trunnion

Support seating Fig 12. Installation of centre block using dual hook – four sling scheme with pipe trunnions. (a) Crane vessel approaching semi-submersible (b) Structural arrangement on centre block to ensure compatible deformation of deck plates

Fig 13 shows the structural arrangement for one of the pipe trunnions designed for lifting the two side blocks of the Cajun Express. For this pipe trunnion, the lower portion of the circular chord was first welded to the cruciform (consisting of the side shell, transverse bulkhead and additional plate). Based on the rigging arrangement which was consistent with the known

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

module centre of gravity and other requirements, the upper pipe chord axis was aligned along the sling direction, with full penetration weld provided to connect the lower chord section (with the line of full penetration weld visible in Fig 13). Additional gusset plates connecting the pipe chords to the side shell provided reinforcement to the overall lift point connection.

Fig 13. Pipe trunnion use for side block lift of Cajun Express. Note provision of gusset plates to strengthen trunnion-to-structure attachment

Another effective use of fabricated pipe trunnions is for overturning of a module, which involves the sliding of the doubled sling (connected to either side of the pipe chord) around the circumference of the side braces of the trunnion. Fig 14 shows the arrangement for a dual hook-four lift point rigging scheme which relies on trunnions 1 and 2 to serve as pivot points during the complete overturning cycle while trunnions 3 and 4 (connected to slings to Crane B) are only used during half the overturning cycle. These two slings are then disconnected from trunnions 3 and 4, before attachment to trunnions 5 and 6 to complete the overall lift cycle.

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Fig 14. Schematic view of flip-over procedure for offshore module through use of pipe trunnions.

STRENGTH OF FABRICATED TRUNNIONS Although fabricated pipe or plate trunnions have been used in many offshore engineering projects, the recommendations (such as Brown & Root21 and API5) do not provide sufficient guidelines for design. In view of the growing importance of fabricated trunnions for heavy lift, the authors and his colleagues9-13 have conducted research into the behaviour and strength of trunnions. This section highlights the superior performance of a structural scheme, termed the through-pipe trunnion, as compared to a standard X-joint arrangement through systematic tests and nonlinear finite element studies. Choo et al.12 presented experimental and numerical results for seven large-scale tests on fabricated pipe trunnions with different chord diameter-to-thickness ratios (with 2γ0= d0/t0=24.5, 33.4 and 40.6) and three structural arrangement: chord with side braces (X-joint), chord with through pipe brace, and chord with through shear plate and side braces. Fig 15 shows the test arrangement within the test frame with the 10,000 kN displacement-controlled actuator. The actuator incrementally applied a downward displacement onto the pipe chord, and the two saddles supporting the pipe braces provided the associated reaction within the self-reacting test frame.

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Fig 15. Test arrangement for fabricated trunnions using the 10000 kN computer-controlled actuator in NUS.

Table 1 summarises the geometric dimensions and material properties, and maximum load, Pu,test applied to specimens CT5 (with side braces) and CT7 (with through pipe brace). It can be noted that for CT5, both the chord and brace wall thicknesses (t0 and t1) and the chord yield stress (σy0) are larger than those of CT7. In Table 1, the chord and brace diameter-tothickness ratio is referred as 2γ0=d0/t0 and 2γ1=d1/t1, while the brace-to-chord wall thickness ratio is τ=t1/t0. After the tests, the observed failure modes are shown in Fig 16, with significant chord plastification for specimen CT5 and brace yielding for specimen CT7. From the tests, it is clearly shown that the through-brace trunnion specimen CT7 attained higher strength than CT5, despite thinner chord and brace wall thicknesses. This is due to the direct load path for the continuous brace pipe which places less demand on the pipe chord for CT7, while the side braces of CT5 transfers the brace shear load through chord wall bending, which is less efficient. Similar observations were made for other specimens as reported by Choo et al.12 Table 1 Sectional Properties and Strength of Fabricated Trunnions Specimen CT5 CT7

d0 (mm) 508 508

Chord t0 (mm) 15.2 12.5

σy0 (MPa) 467 350

d1 (mm) 406 406

Brace t1 (mm) 17.0 12.5

σy1 (MPa) 250 376

Geometric Ratios 2γ1 τ 2γ0

Pu,test (kN)

33.4 40.6

4260 5160

23.9 32.5

1.1 1.0

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Through Brace

Brace

Chord

Chord Fig

16.

Front view of fabricated trunnion test specimens CT5 and CT7, showing (a) failure due to chord plastification (b) failure due to brace shear failure

The left-hand side of either Fig 17a or 17b shows the view of one quarter of specimen CT5 or CT7 (after flame cutting and grinding). The right-hand side of either Fig 17a or 17b shows the deformed shape and stress distributions of the specimen obtained through nonlinear finite element analysis. It can be observed that the computed deformed shapes match those of the tests closely. The thick bearing plate which was provided to transfer the support reaction to the brace can be observed.

Side Brace

Through Brace

CT5

CT7

Fig 17. Comparisons of cut-away views of deformed shapes of test specimens CT5 and CT7 with nonlinear finite element predictions: (a) significant yielding and distortion of chord wall, with brace remaining elastic (b) significant shear yielding of brace, and minimal distortion of chord wall

For CT5, it is seen that the brace remained relatively un-distorted while there is significant chord wall bending at the top position of the brace-chord intersection. In addition, the pipe chord section below the brace is seen to be pulled away from the chord longitudinal axis. For CT7, the through pipe brace is seen to have significantly deformed through shear yielding outside the chord pipe region, while there is no noticeable pipe chord deformation. Thus, the through pipe brace arrangement does not require through-thickness property of the chord material whereas there may be more stringent requirements for the side brace arrangement (as indicated in Fig 17a).

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

As summarised in Table 1, CT7 with the through pipe brace arrangement sustained higher load than CT5 which has thicker chord and brace wall thicknesses. The through pipe is observed to provide more efficient load transfer with no direct through-thickness requirement on the pipe chord material. With proper proportioning, this structural scheme can be readily adopted for heavy lift installation of structures. DISCUSSION The previous sections have covered project examples which illustrate various rigging arrangement selected, field instrumentation for doubled sling arrangement, and research into fabricated pipe trunnions to highlight recent efforts which aim to better understand the engineering issues related to heavy lift. In this section, the considerations for sling load distribution, lift point attachment to structure and fabrication aspects are discussed. Sling Load Distribution A single hook-4 point lift is statically indeterminate and is thus significantly influenced by sling length inaccuracies, sling stiffness and structure stiffness5,6,20-22. The variations of skew load factors versus sling length misfit are presented by Mawer et al.19. A tandem crane lift is generally statically determinate, and variations in sling lengths within normal tolerances give insignificant deviations in sling loads. For a doubled sling, the tension on each arm of the sling will be different due to frictional effects over the hook or the trunnion brace. These frictional effects are supposed to be taken into account by a force split ratio of 55/45=1.2 in the sling tensions T1 and T2 on the arms of the doubled sling21. However, recent studies14-16 and selected results presented in this paper have indicated a larger force split ratio for skew rigging configuration. There is therefore a need to ensure that a padeye or plate trunnion is designed for the additional out-of-plane moment resulting from the uneven load distribution. For fabricated pipe trunnions, the inherent bending and torsional strengths of the pipe chord is able to sustain larger force split ratio for doubled sling arrangement, and thus offer distinct advantages over padeyes or plate trunnions. There is an associated requirement to ensure that the attachment of the lift point to structure transfers the sling tensions efficiently onto the module structure. Lift Point Attachment to Structure It is strongly recommended by Shell22 that lifting points (such as padeyes and trunnions) are designed to transfer load in shear rather than tension. The concern on through-thickness load through the plate thickness relates to possible problem with lamellar tearing of the plate. The lift point (which may be padeye, plate or pipe trunnion) should be attached to locations on the structure that are capable of resisting the lift point design load. The orientation of the padeye and plate trunnion is also significantly dependent on the rigging configuration adopted for lift installation, and the connection of the main-plate to the adjacent structural elements should be designed appropriately.

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

Fabrication Aspects Based on the results of the plate trunnion tests10, failure occurred at the heat-affected zone of the side brace or shear plate after significant material yielding. For fabricated pipe trunnions, the test results have been reported by Choo et al.12. In accordance with industry recommendations5,6,22, compatible welding consumables and processes should be adopted and the trunnion components should be 100% non-destructively tested. In addition, the accurate orientation of the main plate of the padeye or plate trunnion (depending on the chosen design) to the sling direction needs to be checked to ensure proper load transfer from the sling. CONCLUSIONS This paper has presented examples to highlight the considerations for rigging selection and associated lift point designs for a number of heavy lift projects. It can be observed that proper selection of rigging scheme for given module structure and design of lift points is essential for successful lift installation. The systematic field instrumentation of doubled slings with various combinations of sling lengths and diameters, sling angles and hook blocks for symmetric and skew arrangement has shown that current industry recommendation for force split ratio of 55/45 is un-conservative for skew rigging arrangement (if no special design features to minimise friction between sling and hook prong, or sling and lift point is provided). This has significant implication on the lift point design of padeye or plate trunnion type which is not efficient in transferring out-of-plane loads. Selected results on fabricated pipe trunnions have demonstrated the superior performance of pipe trunnions with through-brace arrangement. The sling load transfer from the brace to the pipe chord is efficiently transferred with no detrimental effect to the chord. The associated considerations for attachment of lift point to structure are also important design issues that require detailed assessment and solutions. The paper has demonstrated that successful solutions for heavy lift projects can be readily found through expert knowledge on rigging selection and consistent proportioning of the key components of the lift system. ACKNOWLEDGEMENTS The author acknowledges the significant support of Mr Prem Raj of Sembawang Marine & Offshore Engineering Pte Ltd, Mr John Chua of Asian Lift Pte Ltd, Mr K. W. Kwan of PPL Shipyard Pte Ltd, and Mr M. Ripping of Heerema Marine Contractors, and their permission to present selected project details. The important contributions of his colleagues: Dr F. Ju, Professor J. Y. R. Liew, Professor N. E. Shanmugam, Mr C. K. Quah and Professor K. H. Lee in the National University of Singapore are gratefully acknowledged. The author appreciates the helpful suggestions from the reviewers in improving the technical aspects of the paper. NOMENCLATURE d0 d1 Pu,test

Outer diameter of chord (mm) Outer diameter of brace (mm) Ultimate capacity of test specimen (kN)

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

t0 t1 β 2γ0 2γ1 σy τ

Wall thickness of chord (mm) Wall thickness of brace (mm) Diameter ratio d1/d0 Chord diameter to thickness ratio, d0/t0 Brace diameter to thickness ratio, d1/t1 Yield stress of material (N/mm2) Brace wall to chord wall thickness ratio, t1/t0

REFERENCES 1. Crowle, A.P. Heavy Lift – From Concept to Installation. Proc. Offshore 93: Installation of Major Offshore Structures and Equipment, 1993, London. 2. Fern, D.T., and Griffin, C. Piper B and Saltire A – Topside Design for Installation. Proc. Offshore 93: Installation of Major Offshore Structures and Equipment, 1993, London. 3. Choo, Y. S., Lim, C.K. and Bok, S.H. A Knowledge-based Approach to Design for Heavy Lift. Proc. Offshore 93 - Installation of Major Offshore Structure and Equipment, 1993, London. 4. Choo, Y.S., Ju, F., Li, L. and Li, M. A Decision Support System for Heavy Lift. International Maritime Technology, Transactions of The Institute of Marine Engineers, London, 2001; 112(2), pp. 43-51. 5. American Petroleum Institute. API RP 2A-LRFD. Recommended Practice for Planning, Designing and Constructing Fixed Offshore Platforms – Load and Resistance Factor Design. 1st ed., 1993, Dallas. 6. DnV. Rules for Planning and Execution of Marine Operations. Det Norske Veritas, 1996, Oslo, Norway. 7. Choo Y.S., Choi, K.C. and Lee, K.H. The Effect of Eyebar Shape and Pin/Hole Tolerance on its Ultimate Strength. Journal of Constructional Steel Research, 1993, 26, pp.153-169 8. Choo, Y.S., Choi, K.C. and Lee, K.H. A Comparative Study on Theoretical and Experimental Strengths of Padeyes Under In-Plane Loads. Proc. 4th Int. Offshore & Polar Engineering Conference, 1994, Osaka. 9. Choo, Y.S., Padmanaban, K., Shanmugam, N.E. and Liew, J.Y.R., 1995. Behaviour of Plate Trunnions Subjected to Shear Loads. Proc. 5th International Offshore & Polar Engineering Conference, 1995, The Hague. 10. Choo, Y.S., Quah C.K., Shanmugam N.E. and Liew J.Y.R. Static Strength of Plate Trunnions Subjected to Shear Loads – Part I. Experimental Study. J. Constructional Steel Research, 2002, Vol 58, pp 301-18. 11. Choo, Y.S., Quah C.K., Shanmugam N.E. and Liew J.Y.R. Static Strength of Plate Trunnions Subjected to Shear Loads – Part II. Computational Study and Design Considerations. J. Constructional Steel Research, 2002,Vol 58, pp 319-332. 12. Choo, Y.S., Quah, C.K., Shanmugam, N.E. and Liew, J.Y.R. Fabricated Trunnions for Heavy Lift. Transactions of Institute of Marine Engineers, London, 2001; 113 (2): pp. 64-76. 13. Choo, Y.S., Quah, C.K., Shanmugam, N.E. and Liew, J.Y.R. Experimental Study on Pipe Trunnions Subjected to Shear Loads. In Tubular Structures IX, Puthli & Herion (Eds), 2001, pp. 503-509. 14. Choo, Y.S., Ju, F. and Lee, K.H. Static Sling Tensions in Heavy Lifts with Doubled Sling Arrangement. Int. J. of Offshore and Polar Engineering, 1997, Vol. 7, pp 311-320. 15. Choo Y.S. and Ju, F. Sling Friction Measurements for Doubled Slings – Observed Force Split Ratios. Technical Report for Heerema Marine Contractors, 1998, 100 pp. 16. Choo, Y.S., Ju, F., Ripping, M. and van Herel, H. Sling Friction Measurement for Doubled Slings – Observed Force Split Ratio. Submitted to Journal of Marine Structures, Special Issue on Heavy Lift.

Invited paper for World Maritime Technology Conference, San Francisco October 2003 Revised after review comments – 18Apri2003

17. Mayfield, J.G. and Zimmerman, M.E. Notes on Heavy Lift Design. Proc 18th Annual Offshore Technology Conference, Houston, 1986, Paper OTC5346. 18. Corcoran, M.J. Structural Design for Offshore Heavy Lift Operations, in Steel Structures: Advances, Design and Construction, Elsevier Applied Science, R. Narayanan (Editor), 1987. 19. Mawer, S.J., Hamilton, J. and D T Blanken, D.T. Assessment of Lift Criteria. Proc. Offshore 93: Installation of Major Offshore Structures and Equipment, 1993, London. 20. Bunce, J.W. and Wyatt, T.A. Development of Unified Design Criteria for Heavy Lift Operations Offshore. Proc. 14th Annual Offshore Technology Conference, Houston, 1982, Paper OTC4192. 21. Brown & Root. Joint Industry Project: Heavy Lift Criteria. Brown & Root Vickers Technology Ltd, 1991. 22. Shell U.K. Ltd. Guidelines for Lifting Points and Heavy Lift Criteria. Engineering Reference Document No. EM/039 Rev. 3, 1991. 23. Selvakumar, K. and Choo, Y.S. Fast Track Construction of Offshore Structures. Proc. 2nd Offshore & Marine Technology Conference, 2000, Singapore. 24. HKS Inc. ABAQUS User’s and Theory Manuals. Rhode Island, U.S.A., 2000, Version 6.1.