Finite Element Analysis of Spreader Bar by Utilizing The Arrangement and Connection of Padeyes
Finite Element Analysis of Spreader Bar by Utilizing The Arrangement and Connection of Padeyes
Finite Element Analysis of Spreader Bar by Utilizing The Arrangement and Connection of Padeyes
ABSTRACT
KEYWORDS: Spreader bar, heavy lifting, Euler stress, padeye, finite element
1.0 INTRODUCTION
Nowadays, major drilling and production facilities of oil and gas industry are located
offshore and thus installing this equipment necessitate efficiency and safety. The
industry has developed a number of ways to overcome heavy lift challenges through
experience and innovation. Heavy lifting is among main methods of marine installation
for heavy equipment. The conventional way to install major facilities, such as topsides
and production equipment, is through heavylift vessel (www.rigzone.com). Heerema's
Thialf and Saipem's S7000, (by then renamed) were upgraded such that the combined
lifting capacity of two cranes on each vessel is 14200 tonnes and 14000 tonnes
respectively. The Balder and Hermod semi-submersible crane vessels (SSCV) were
each fitted with two enormous cranes. S7000 is well-known for holding world record
for an actual lift of 12150 tonnes and for lifted 9500 tonnes jacket in 2007 for Pemex in
dynamic positioning mode. Also, Thialf has a staggering lifting capacity of 14200 tonne
or equivalent to weight of more than 1180 fully laden London buses (offshore-
technology.com).
During heavy lifting, the structure is subjected to higher stress due to its self-weight and
dynamic load from variation of hoisting speeds, crane, vessel motions, cargo barge
movements, object movements and others. (DNV VMO, 2014). Thicker and higher
strength of steels are used in designing installation aids for heavy lifting. The use of
slings in bridle arrangement configuration will induce stress in the structure. This
additional stress can be eliminated using SB. By utilizing SB to vertically line up the
sling on top of structure’s lifting point, this will allow a straight pull movement. SB is a
structure designed to resist compression forces induced by angled slings, by altering line
of action’s force on lift point into the vertical plane. (GL Nobel Denton, 2013).
Upper Sling
Upper padeyes
Spreader bar
(SB)
Lower Sling
Delta
Plate
Tubular pipe is commonly used in designing SB, as shown in Figure 1, due to its
constant properties in any sectional direction if compared to I beam which have a
combination of strong and weak axis. There are two types of SB padeye design usually
found in offshore lifting (Figure 2) where the differences are on bottom padeye’s
design. Where, bottom part of type 2 padeye is fabricated in form of delta plate whereas
bottom of type 1 padeye is sharing same plate with its top.
Type 1 is using less material and welding filler because of simplicity of lifting point
(padeye) connection design. Though by sharing a same plate, bottom shackle’s need to
be de-rated (depend on manufacture) since side loading applied to the shackle will
caused larger size of shackle (Figure 3). Increased size of shackle will lead to larger
plate needed and stimulate weight gain of lifting system, whereas type 2 will require
more material to be used in fabricating delta plate. By introducing delta plate, top
padeye will need to be shifted away from bottom padeye to avoid clashes between top
shackle and delta plate. This arrangement will result in a bigger in-plane moment hence
thicker pipe need to be used.
Upper shackles
Lower shackles
The plate used to fabricate padeye is made from through-thickness (Wang, et al., 2015)
property material to avoid failure due to lamellar tearing during lifting. Lamellar tearing
is a separation in parent or based metal caused by through-thickness strains. Such
strains are induced primarily by weld metal shrinkage under conditions of high strain.
When detected, lamellar tearing can result in often difficult and costly repairs and
subsequent construction delays (Ship Structure Committee, 1979). Furthermore,
required type of steel is barely available in the market therefore require to be procure
within at least 3 months in advance, depending on its availability. Therefore, if any
defect detected during fabrication, schedule of offshore installation campaign will be
drag forward until procurement of material is completed.
Also, the limitation on rigging weight due to operational/material handling issue or as
per client’s requirement on “Not to Exceed Weight” need to be considered. Hence,
designing SB using relatively thinner material for weight reduction and lessen welding
work to reduce risk of defect in welding is essential.
In this study, an advanced SB padeye design (Type 3) and comparison analysis with
other two common SB padeye design is introduce where main plate of padeye (Figure
4) is slotted through tubular pipe in order to get maximum lifting capacity by
transferring load through weld at joint. Tubular joint without stiffener inside the chord
will cause chord to experience punching stress thus reducing its capacity. Therefore, slot
in connection is ideal for designing connection between lifting point and tubular pipe.
Lifting point on SB is installed right above lifting point of module to ensure that vertical
pull conditions can be achieve without overstress due to sling arrangement occurred in
lifted structure.
Wall thickness
Outer diameter
Tubular pipe
A
Length of SB, l
View A-A
(Cross Sectional
Figure 2. New spreader bar padeye design Area)
2.1 Material
For the purpose of this study, SBs are design to lift the 19620N structure. Lifting
arrangement is shown in Figure 1. The length of SB is fixed to 800mm based on
distance between two lifting point of structure and dimensions of SB’s geometry is
detailed in Table 1 below.
Material properties selected for fabrication of plates and tubular are mild steel which in
compliance to ASTM A36 for plate and ASTM A106 Grade B for pipe. Calculations are
based on minimum yield strength of relevant material grade as specified in Table 3.
𝑀(𝑥) − 𝑃𝑦 − 𝐹𝑎 = 0 𝑎 ≤ 𝑥 ≤ (𝐿 − 𝑎) (2)
𝑀(𝑥) − 𝑃𝑦 − 𝐹𝐿 + 𝐹𝑥 = 0 (𝐿 − 𝑎) ≤ 𝑥 ≤ 𝐿 (3)
Where:
M = Total Moment, P = Axial Load, F = Vertical Load
P and F are derived from calculated sling load based on rigging arrangement as shown
in Figure 1.
From equation (1), (2) and (3), it shows that SB is needed to resist axial compressive
force and bending moment. Thus, Figure 5 shows the eccentricity on geometry that
generates moment in SB where the bending moment created will reduced the capacity
of SB.
Any force that applied to SB at neutral axis resulted in a purely compression force in
tubular pipe. Therefore, forces that are not lined up with neutral axis generate bending
force or bending moment. SB that is subjected to bending forces and/or bending
moments is more difficult to be properly design and will no longer be simple and light
weight construction as preferred (heavyliftnews.com).
To eliminate eccentricity, padeye’s top is moved to CL of tubular pipe and left Euler
buckling as an only mode of failure for SB. This concept is illustrated in Figure 6.
According to API RP2A WSD 22nd Edition, allowable Axial Compressive stress, 𝐹𝑎
must be determined from the following AISC formulas for members with D/t ratio or
less than 60. In the (AISC ASD, 9th Edition) equations for allowable compressive
stresses, various imperfections such as effect of residual stresses, actual end restraint
conditions, crookedness, and small unavoidable eccentricities are empirically taken into
account.
The API code which is based on AISC 9th Edition requirement assumes arbitrarily that
the elastic buckling holds valid when stress in the column is not greater than one-half of
the yield stress (Fy/2) (J.S.Arora, http://user.engineering.uiowa.edu).
For column having effective length less than Cc, it is assuming the failure by crushing
of the material induced by predominantly axial compressive stresses. Failure occurs
when stress over cross-section reaches yield or crushing value for the material.
𝐹𝑎
𝑘𝑙 2
( )
[1 − 𝑟 2 ] 𝐹𝑦
𝐹𝑦 2𝐶𝑐
0.5𝐹𝑦
𝐸𝑞. (7)
𝐸𝑞. (6)
𝐸𝑞. (8)
𝑘𝐿
𝐶𝑐
𝑟
Figure 5: Variation of critical stress and allowable stress as specified by the API code
Finite Element Analysis is widely used in offshore industry to design offshore and
subsea structure. All structures were modelled using Space Claim (ANSYS Package).
In SpaceClaim, spreader bar components i.e tubular pipe, main plate of padeye, cheek
plate and stiffener are group together and Shared Topology is activated. Shared
Topology occurs is triggered when bodies are grouped into multibody parts. It allows
for a continuous mesh across common regions where bodies touch, instead of having to
define Contact Regions in the ANSYS Workbench. These bodies share topology in the
region where they are in contact with, so the mesh is continuous across part as shown in
Figure 8. It is often, but not always, more desirable for analysis to have a continuous
mesh across parts than to use contact (ANSYS 14.5 User Guide).
3
x
1 1
4 Legend:
x x
x 1-Tubular Pipe
2- Padeye-main plate
2 4 3. Padeye-cheek plate
x x 4. Stiffener plate
3
x
Figure 6. Typical details of SB components and continuous meshing across the parts
The model then exported to ANSYS for assignment of materials and contact details.
The meshing is performed with element size of 5mm as shown in Figure 9 and
SOLID186 Element is assigned to the geometry. SOLID186 is a higher order 3-D 20-
node solid element that exhibits quadratic displacement behaviour where the element is
defined by 20 nodes having three degrees of freedom per node: translations in the nodal
x, y, and z directions. The geometry, node locations, and element coordinate system for
this element are shown in Figure 9. A prism-shaped element may be formed by defining
the same node numbers for nodes K, L, and S; nodes A and B; and nodes O, P, and W.
A tetrahedral-shaped element and a pyramid-shaped element may also be formed
(ANSYS 14.5 User Guide) as shown in Figure 9.
Summary of mesh statistic for each type of SB is as shown in Table 4. The static
structural analysis is performed to determine the stress level for imposed load. Then the
data is exported to linear buckling module to find critical buckling load for each type of
SB.
Bearing load is then assigned to the surface of upper padeye’s pinhole representing
dynamic sling load (DSL) applied to SB as shown in Figure 10, 11 and 12 for SB type
1, 2, and 3 respectively. As defined by ANSYS, the bearing load simulates radial forces
only and applied on interior of cylinder in the radial direction by using a coordinate
system (ANSYS 14.5 User Guide).
In addition, torsion is imposed to the surface of pinhole for upper and bottom padeye to
distributes moment "about" (the vector of) an axis curved faces where right-hand rule is
applied to determine sense of moment.
Note:
Loadings are derived from rigging arrangement as shown in Figure 1. The maximum loading in upper
sling is separated into horizontal (parallel to longitudinal axis of SB) and vertical component. Latera load
is assuming 5 percent of DSL.
Figure 10(b). Detail of loading condition and boundary condition applied to the SB
type 1
Figure 11(b). Detail of loading and boundary condition applied to the SB type 2
Figure 12(b). Detail of Loading and Boundary Condition applied to the SB type 3
Once stress result is found, linearization of stress is performed. The linearized stress
results calculate membrane, bending, peak, and total stress along a straight-line path in
the ANSYS workbench. When result is evaluated (stress linearization), component
stress values at the path points are interpolated from appropriate element's average
corner nodal values. Stress components through the section are linearized by a line
integral method and are separated into constant membrane stresses, bending stresses
varying linearly between end points, and peak stresses (defined as differences between
actual (total) stress and membrane plus bending combination). The details view shows
membrane, bending, membrane + bending, peak, and total stresses. The bending
stresses are calculated such that neutral axis is at midpoint of the path. Principal stresses
are recalculated from the component stresses and are invariant with the coordinate
system as long as stress is in the same direction at all points along the defined path
(ANSYS 14.5 User Guide).
4.0 EXPERIMENT
The objective of the testing program is to determine critical buckling loads for tubular
under compressive force which conducted according to ASTM E9. Only tubular pipe
for SB type 3 is used in the experiment. In total, 5 specimens are used in this testing.
The tubular size selected as in Table 1 is based on loading applied to the SB structure.
Due to huge moment for SB type 2, bigger diameter needs to be selected. Total weight
of SB type 1 and 2 are more than 50 percent heavier than SB type 3. The weight ratios
(SB’s weight/weight of lifted structure) are 0.30, 0.41 and 0.15 percent for SB type 1, 2
and 3 respectively. The differences in sling loads for SB type 1 and 2 are due to the
eccentricity of padeyes (Figure 5) where eccentricity is varying on padeye’s
arrangement (Figure 2). For SB type 1 and 2, the eccentricities are 10mm and 30mm
respectively. No eccentricity should be considered for SB type 3 since the CL of top
padeye lay at similar line with CL of tubular pipe. SB type 2 having a biggest
eccentricity is since delta plate arrangement requires it to be installed far away from
upper padeye to ensure that it will not clash with shackle that was installed at the top of
padeye. Therefore, padeye type 2 produce highest moment among other types of SB
thus require thickest material in design.
The buckling load as specified in Equation (6) is depending upon geometry and elastic
modulus of column and not upon the strength of it. However, in AISC ASD, 9th Edition,
the equations for allowable compressive stresses, various imperfections such as effect of
residual stresses, the actual end restraint conditions, crookedness, and small unavoidable
eccentricities are empirically taken into account. To get a buckling stress, buckling load
is divided by area of cross section. Considering that buckling stress is found as above, it
is noted that allowable compressive stress is fall beyond elastic region for type 1 and 2
of the SB therefore, buckling load did not applied since excessive yielding occurs
before reaching the buckling.
For type 1, maximum moment allowed is 5.49E+05N.mm which equivalent to 75
percent of tubular yield strength. Design moment is 2.24E+05N.mm which at 41 percent
of allowable moment (based on API 2A Stress Criteria). The maximum moment
allowed is 1.19E+06N.mm which equivalent to 75 percent of pipe strength. Design
moment is 9.93E+05N.mm which at 84 percent of allowable moment. Therefore, by
using SB type 1, 31 percent of the pipe strength is used to resist bending moment while
for SB type 2, 63 percent of the pipe strength is used to resist bending moment.
FEA is performed to obtain the Von-Mises stress. The stress linearization is performed
for each type of SB to separate primary (structural) and secondary
(geometry) stress.
As shown in Table 6, mathematical modelling for value of stress is consistent with FEA
analysis.
The experiment results for 5 specimens of pipe used to design SB type 3 shows that the
pipe will be buckled at an average load of 23447.12N.
Buckling
Specimen
Load (N)
S1 23781.80
S2 22128.31
S3 24368.38
S4 26203.11
S5 20754.02
Result as shown in Table 9 indicate that for SB type 1, the utilization of axial stress is
only 26 percent of allowable while type 2 is only 12 percent. Remaining tubular pipe
strength is used to resist design stress due to bending moment. For SB type 3, the
utilization of axial stress is at 90 percent capacity.
6.0 CONCLUSIONS
Since SB type 1 and 2 are not categorised as long column as shown in Figure 7, its
failure is mainly caused by excessive yielding instead of buckling. Therefore, SB type 1
and 2 is considered as not buckling sensitive. SB type 1and 2 use respective 31 and 63
percent of the pipe strength to resist bending moment which resulting in heavier section
needed to be used. Weight of SB type 1 and 2 are more than 50 percent heavier
compared to SB type 3. Due to highest moment induced for SB type 2, it is not
recommended for used on heavy lifting since thicker material and relatively more
welding works and Non-Destructive Test) NDT are required that may increase chances
for cracks in the weldment or plate. The moment induced in SB type 1 and 2 resulted in
increased of length slot that require more filler material for welding and therefore
increase chances of having crack in the structure.
7.0 ACKNOWLEDGEMENTS
The authors would like to thank Universiti Putra Malaysia for providing facility to
perform this study.
8.0 REFERENCES
GL Nobel Denton, Guideline for Marine Lifting & Lowering Operations, June 2013
API RP 2A-WSD 22nd Edition - Recommended Practice for Planning, Designing and
Constructing Fixed Offshore Platforms-Working Stress Design, December 2014.
Ship Structure Committee, Significance and control of lamellar tearing of steel plate in
the Shipbuilding Industry, 1979.
Rigzone. Training. How Does Heavy Lift Work?. Retrieved from http://www.rigzone.
com/training/insight.asp?insight_id=321&c_id=20 (Access on 15-Jun-2015)
Wang Yuan-qing, Liao Xiao-wei, Zhang Yuan-yuan & Shi Yong-jiu (2015).
Experimental study on the through-thickness properties of structural steel thick
plate and its heat-affected zone at low temperatures. Journal of Zhejiang
University- SCIENCE A (Applied Physics & Engineering),16(3), 217-228.
J.S.Arora,http://user.engineering.uiowa.edu/~design1/structuraldesignii/compresiondesi
gn.pdf (Access on 15-Jun-2015)