246 Guidance Notes On Geotechnical Performance of Spudcan Formations GN E-Jan18
246 Guidance Notes On Geotechnical Performance of Spudcan Formations GN E-Jan18
246 Guidance Notes On Geotechnical Performance of Spudcan Formations GN E-Jan18
GUIDANCE NOTES ON
JANUARY 2017
Foreword
For the last decade, ABS has been involved in numerous joint industry projects on jackup spudcan foundations.
ABS has outlined this knowledge herein to provide guidance covering the design and installation of jackup
spudcan foundations addressing the following topics:
• Site assessment
• Spudcan penetration prediction in a single layer of soil, such as clay or sandy soil
• Punch-through prediction and mitigation
• Foundation stability assessment
• Foundation fixity
• Spudcan-jacket pile interaction
• Spudcan footprint interaction
These Guidance Notes become effective on the first day of the month of publication.
Users are advised to check periodically on the ABS website www.eagle.org to verify that this version of
these Guidance Notes is the most current.
We welcome your feedback. Comments or suggestions can be sent electronically by email to rsd@eagle.org.
Terms of Use
The information presented herein is intended solely to assist the reader in the methodologies and/or techniques
discussed. These Guidance Notes do not and cannot replace the analysis and/or advice of a qualified
professional. It is the responsibility of the reader to perform their own assessment and obtain professional
advice. Information contained herein is considered to be pertinent at the time of publication, but may be
invalidated as a result of subsequent legislations, regulations, standards, methods, and/or more updated
information and the reader assumes full responsibility for compliance. This publication may not be copied
or redistributed in part or in whole without prior written consent from ABS.
GUIDANCE NOTES ON
SECTION 1 Introduction
1 General Comments
For the last decade, ABS has been involved in numerous joint industry and government sponsored projects
on jackup spudcan foundations. The current state of the art in spudcan design and installation represents
worldwide contributions by a large number of investigators from government agencies, marine warranty
survey groups, classification societies, contractors, shipyards, universities and petroleum companies.
ABS has gathered this body of knowledge to provide a guideline on geotechnical performance of spudcan
foundations, including site assessment, spudcan penetration prediction, punch-through problem, foundation
stability assessment, foundation fixity, spudcan-jacket pile interaction and spudcan footprint interaction.
Mobile Jackup Platform: A self-elevating unit with movable legs capable of raising its hull above the
surface of the sea.
Perforation Drilling: A technique performed to perforate the stiff layer with drill holes and remove the soil
as one of the mitigation methods to reduce the punch-through failure.
Prescriptive Preloading: During preloading, the hull is kept at, in or close to water level, with each individual
leg preloaded by sequential filling and discharge of selected preload ballast tanks.
Punch-Through Failure: Unexpected sudden and rapid penetration of the spudcan through soil occurs
when strong soil is over soft soil and is a major risk for the stability and equilibrium of the jackup
structure.
Rack Phase Difference (RPD): The difference in elevations between the chords of any one leg.
Reaming: A technique that involves forcing a leg into position by incremental vertical reciprocation to penetrate
the spudcan into the soil at the required position. Reaming is one of the mitigation methods for reducing
spudcan footprint interaction.
Remoulded Undrained Shear Strength: The magnitude of the shear stress that a disturbed soil can sustain
in an undrained condition.
Simultaneous Preloading: During preloading the hull is held with minimal draft or air gap and the preload
is incrementally increased on all the legs simultaneously.
Soil Sampling: A suitably stored, small amount of soil for visual inspection and laboratory testing for the
determination of the soil unit geological provenance, characteristics and geotechnical engineering design
parameters.
Spudcan: A large inverted cone that is roughly circular in-plan with a shallow conical underside and a
sharp protruding spigot. It is mounted at the base of a jackup’s leg, and is primarily considered to provide
sliding and bearing resistance to the jackup rig when deployed into the sea bed.
Stomping: A process where a spudcan is initially emplaced away from the center of an old footprint and then
the spudcan is used to displace soil towards the old footprint at desired positions to widen the disturbed
region. In soft to stiff clay, stomping is very effective in mitigating spudcan footprint interaction.
Triaxial Test: A common laboratory testing method widely used for obtaining shear strength parameters
for a variety of soil types under drained or undrained condition.
7.1 Symbols
A = spudcan maximum cross sectional area
Aa = area ratio
Ae = spudcan effective bearing area based on cross-section taken at uppermost part of bearing area
in contact with soil
As = spudcan laterally projected embedded area
cul = undrained cohesive shear strength at spudcan tip
cu0 = undrained cohesive shear strength at lowest maximum bearing area
cv = coefficient of consolidation of clay
D = maximum spudcan diameter below mudline
Deff = equivalent spudcan diameter at mudline
Df = diameter of the spudcan which forms the footprint
e = opening ratio
rf = failure ratio
Sua = undrained shear strength averaged from (h – ht) to (h + hplug)
Subs = shear strength intercept at the clay layer’s surface in strong overlying soft soil
Su_footprint = footprint undrained shear strength
SuH = shear strength at the back flow depth hc
Sum = shear strength intercept at the layer’s surface
Su,plugbase = undrained shear strength corresponding to the level of the sand plug base
Sutop = undrained shear strength of the top strong clay
Su_undisturbed = undisturbed undrained shear strength
Su0 = undrained shear strength at the lowest depth of the maximum plan area of spudcan
τ1 = adjusted time factor for soil consolidation during the operational period
τ2 = adjusted time factor for soil consolidation during the elapsed time after a footprint is formed
7.3 Abbreviations
CAUC Anisotropically consolidated undrained triaxial compression test
CAUE Anisotropically consolidated undrained triaxial extension test
CID Isotropically consolidated drained test
CoV Coefficient of variation
DS Direct shear test
DSS Direct simple shear test
MinV Miniature vane
MODU Mobile offshore drilling unit
MV Motor vane
PP Pocket penetrometer
RPD Rack phase difference
SI Site investigation
TV Torvane
UU Unconsolidated undrained triaxial tests
1 General
A clear understanding of the seabed and sub-seabed conditions is critical for the site specific assessment of
a jackup’s suitability during installation, elevated operations and leg extraction. The jackup foundation site
assessment procedure usually comprises:
• Acquisition of regional and local geological and geotechnical data (to include previous jackup foundation
performance if available)
• Geological data review to develop the ground model
• Geophysical site survey to refine the ground model
• Geotechnical Site Investigation (SI) to further refine the ground model
• Generation of geotechnical design profiles with engineering parameters
• Performance of jackup foundation site-specific assessment using the design profiles.
The Geophysical Site Survey report should describe the interpreted ground model; include an explanation
of the geological setting, depositional environment and history together with descriptions of any potential
geohazards which could influence jackup operations.
Intrusive geotechnical SIs are conducted in order to ground-truth the geophysical data and to obtain the required
geotechnical index and strength measurements. The geotechnical SI allows confirmation or further refinement
of the interpreted ground model. Adequate data are required to facilitate detailed engineering characterization
of each soil layer and to provide understanding of the spatial variation of these parameters.
Preferably the geophysical and geotechnical components are planned together as integrated parts of the same
investigation. The soil design profiles with associated engineering strength parameters are then developed
for use in the predictive bearing capacity analyses. The actual scope of work developed will depend upon
the vertical and lateral variability of the soil as well as the presence of any geohazards.
The main requirements of site investigation are to be in accordance with 3/6.3 of the ABS Rules for Building
and Classing Offshore Installations.
7 Laboratory Testing
Soil samples are extruded from the sample tubes on the SI vessel, (in the same direction which they were
taken to minimize disturbance by stress reversal), where they are carefully separated from drill cuttings and
any heavily disturbed material. The samples are described, photographed and catalogued in accordance
with recommended industry practice. Selected undisturbed samples are stored in sealed containers, which
are usually wax-filled, to preserve the moisture content and limit further disturbance.
A range of different soil tests are available from simple laboratory index tests to advanced stress path tests
conducted in specialist onshore soil testing laboratories. Typically offshore tests include moisture content
and density determination, carbonate content, particle size distribution together with simple strength tests such
as pocket penetrometer, torvane, laboratory vane, fall cone, unconsolidated undrained triaxial tests (UU).
Index tests are conducted to assist with the soil type classification and also to provide preliminary parameters
for initial jackup foundation assessment.
While these undrained soil strength tests are comparatively quick and relatively easily conducted in the offshore
laboratory they measure shear strength by different failure mechanisms, and may provide significant strength
data scatter and may not provide an accurate indication of sample quality. Some offshore soil laboratory
tests, such as the fall cone and the motor vane, can be used to provide an indication of the soil sensitivity.
Advanced onshore soil laboratory testing is encouraged as these strength tests [e.g., anisotropically consolidated
undrained triaxial compression test (CAUC), anisotropically consolidated undrained triaxial extension test
(CAUE), direct simple shear (DSS) test] provide data for calibration of the in-situ penetrometer results.
Oedometer tests provide information about the stiffness and drainage behavior of the soil. Simple offshore
laboratory strength tests may provide adequate information for relatively simple soil conditions which lend
themselves to modeling by simple bearing capacity formulations. However, for more complex soil conditions,
simple soil test results with the application of simple models become less reliable and advanced soil testing
is advisable. The relative reliability of various tests in measuring the undrained strength of clay is presented
by [Ref.1] in Section 2, Table 1.
TABLE 1
Relative Reliability of Tests Measuring the Strength of Clay Soils
Intact Su(1)
Test Type Soil Profiling Remoulded Su(1)
<40 kPa 41-80 kPa >80 kPa
Piezocone 1 2 2 2 4-5
T-bar & ball 1 1-2 1-2 1-2 1-2
penetrometers (with pore pressure
measurement)
In-situ vane --- 1-2 1-2 1-2 3
UU(2) --- 4-5 3-5 2-5 2-3
Motor vane(2) --- 3-5 3-5 4-5 2-3
Torvane(2) --- 3-5 3-5 4-5 ---
Pocket penetrometers(2) --- 4-5 4-5 4-5 ---
CIU/CAU/DSS --- 1-2 1-2 1-2 2
Notes:
1 Rating ~ 1 High reliability; 2 High to moderate reliability; 3 Moderate reliability; 4 Moderate to low reliability;
5 Low reliability.
2 The test result reliability is dependent on the sample quality (or degree of sample disturbance) and soil homogeneity.
TABLE 2
Index Properties for Jackup Foundation Site Specific Assessment
Index Properties and Additional
Soil Type Strength Properties
Parameters
Clay Undrained shear strength Su Water content
Remoulded shear strength and soil Plastic limit
sensitivity St Liquid limit
Submerged unit weight of clay
Coefficient of consolidation
Over consolidated ratio
Sand Critical state angle of friction Particle size distribution curve
Crushing strength Relative density
Submerged unit weight of sand
Over consolidated ratio
9.3 Clay
The undrained shear strength, Su, can be established from a number of commonly conducted strength tests
as follows.
• In-situ testing: field penetrometers (cone), vane shear
• High quality laboratory testing: CAUC, CAUE, DSS
• Simple laboratory testing: unconsolidated undrained triaxial test (UU), miniature vane (MinV), motor
vane (MV), torvane (TV), pocket penetrometer (PP)
The reliability of the interpreted strength parameters relies heavily on the quality of test and sample conditions.
The simple laboratory tests typically should be conducted immediately after the samples are extruded in
the laboratory. However, the test results may tend to show significant variability both within a given type
of test and between different tests, and are ‘operator dependent’. Tests of this type may be sufficient where
reasonable knowledge of the ground conditions and knowledge of successful jackup operations in the
locality are available; the data may also be used to extend or interpolate between results from previous
high quality laboratory tests. As samples are not reconsolidated prior to testing, the reliability of these tests
suffers from unquantifiable sample quality.
The knowledge of remoulded shear strength and sensitivity is necessary for consideration of strain softening
effects, particularly for highly sensitive clays. Strain softening reduces the average operational strength
during spudcan penetration, thus increasing the penetration. Partial remoulding of the soil will also affect
the magnitude of strength recovery or enhancement with time and this is relevant for spudcan breakout
force assessment and bearing capacity evaluation for jackup revisits.
In-situ cyclic full-flow penetrometer tests (using T-bar or Ball) with 10 cycles of penetration and extraction,
will generally show a well-defined remoulded penetration resistance. The ratio by which the penetration
resistance decreases between initial and post-cyclic penetration resistance will be less than the actual sensitivity
at the elemental test level, largely due to the partial remoulding that occurs during initial penetration. Such
tests do, however, provide an appropriate measure of remoulded shear strength that is directly applicable to
spudcan performance.
Where the sediments may be susceptible to significant strength loss during cyclic loading under environmental
or seismic loading conditions, cyclic shearing tests should be undertaken in the laboratory in addition to
monotonic tests. In the depth range of expected spudcan penetration, sufficient cyclic tests should be
undertaken to establish a “cyclic fatigue” curve, showing how the normalized shear stress to cause a given
magnitude of strain varies with the number of cycles. This information may then be used to assess an
appropriate cyclic shear strength for use in quantifying the jackup’s performance during installation and
under ultimate cyclic loading conditions.
9.5 Sand
The most common laboratory tests for determining effective strength parameters in sand are CID and DS
tests, which represents triaxial and plane strain condition, respectively. Sample disturbance is inevitable
when sampling cohesionless material from the seabed. The samples are reconstituted to their approximate
in-situ state, with the relative density generally estimated from the cone resistance. Appropriate effective
stresses are then applied before the shearing stage. During shearing, it is important to verify that the shearing
rate applied is slow enough to prevent the development of excess pore pressure. Theoretically, under a given
stress level, it is expected that the friction angle φ' from DS tests will be greater than that from CID tests.
To better account for the stress level effect on φ', the design value may be estimated from the value of the
relative density, ID, and the critical state friction angle, φcv, using an appropriate strength-dilatancy
relationship that takes account of the mean effective stress p' during bearing failure.
Since the value of critical state friction angle φcv lies within a small range, at least for silica sand, it is possible
to estimate the in-situ directly from the cone resistance, qc. The following expression has been applied
widely to sandy sites in the North Sea by [Ref.2].
qc σ ′ (1 + 2k0 )
ID = a1ln – a2ln v 0 – a3
pr 3 pr
where
a1 = 0.336
a2 = 0.154
a3 = 1.91
Qcrushing = particle crushing strength on a natural log scale, in kPa (kgf/mm2, lbf/in2)
p' = mean effective stress which in turn depends on the value of φ', in kPa (kgf/mm2,
lbf/in2).
As an approximation, the recommendation is to treat p' as the maximum preload pressure.
The value of φcv may be obtained from direct shear tests on disturbed sand, from the “steady state” friction
angle in the later stages of the test. Some of the reported values for φcv and Qcrushing by [Ref.4] are given in
Section 2, Table 3. The above procedures provide an estimate of the peak angle of friction, including the
effect of the average stress level in the soil. It is important to note, however, that as a spudcan continuously
penetrates the soil, the peak strength is not mobilized simultaneously throughout the deforming soil. As a
result, calculations of spudcan resistance based solely on peak strength of a rigid-plastic soil can result in
overestimates of resistance. [Ref.5] addresses this issue by employing reduced friction angles. An alternative
approach is suggested in Section 3 where a mobilization factor is applied to the calculated resistance.
TABLE 3
Values of φcv and Qcrushing Derived from Triaxial Compression Tests
Sand Mineralogy Qcrushing φcv (°)
Ticino Siliceous (containing comparable amounts of 10.8 33.5
quartz and feldspar grains)
Toyoura Quartz 9.8 32
Hokksund Siliceous 9.2 34
Mol Quartz 10 31.6
Ottawa Quartz (with varying fine content from 0 to 20%) 9.8 to 10.9 30 to 33.5
Antwerpian Quartz & Glauconite 7.8 to 8.5 31.5
Kenya Calcareous 8.5 40.2
Quiou Calcareous 7.5 41.7
∑ (zi − z )(Su ,i − Su )
n
ρ= i =1
∑i=1(zi − z )2
n
Sum = S u – ρ z
where the mean values of depth , z , and undrained shear strength, S u , are calculated as:
n
∑z
1
z = i
n i =1
∑S
1
Su = u ,i
n i =1
These formulations simply fit a straight line through n data points with all the raw data points assumed to
be correlated and each data point in the layer assumed to be of equal importance.
Detailed knowledge of the local soil characteristics can be incorporated into the soil strength derivation process
if the data can be presented in a quantitative form such as by test reliability (see, for example, Section 2,
Table 1). This can then be incorporated into the statistical method used to determine the strength profile
and be achieved with the use of weighting factors to modify discrete data values. However, manipulation of
raw data in this fashion has to be justifiable and conducted with caution.
Further information provided by the discrete set of strength points includes an estimate of the variation of
the undrained shear strength measurements. This is commonly expressed in terms of standard deviation σ
and coefficient of variation (CoV) as follows.
∑ [S ]2
n
− (S um + ρz i )
1
σ= u ,i
n i =1
σ
CoV =
Su
The approach of adopting a conservative estimate of the shear strength profile, as is common practice in
other assessments of bearing capacity, is not appropriate for the jackup installation calculation, where an
accurate estimate of the actual penetration is required. However, the use of confidence bands on soil strength
profiles, making reasonable allowance for uncertainties in soil strength measurement, is beneficial for
assessment purposes.
Confidence bands placed on the strength profile should reflect the consequence of the final bearing capacity
analysis. For instance, in situations of soft clay where final penetration depth is critical so that the jackup does
not run out of leg length, it is essential that the strength profile is lower than the best estimate. However,
worst case scenarios, for cases with punch-through potential, may be to consider upper bound strengths in one
layer and lower bound strengths in the next. It is indeed possible that many different profiles may require
analysis, each reflecting bounds on the problem and consequence.
Consideration of uncertainty in parameters other than strength may also be required. For example, the depth
of a layer interface may be a critical factor for punch-through calculations. If there is uncertainty in its
depth (or in the thickness of a layer), this should be accounted for in the final bearing capacity analysis.
Uncertainties may arise due to spatial variability or uncertainty in interpretation of the test data.
For relatively homogeneous soil layers confidence bands can be determined in a meaningful way by a
proportion of the standard deviation of the measurements above and below the mean profile, where the
chosen proportion should reflect the uncertainty and severity of consequences. Calculating at ±1 standard
deviation from the mean undrained shear strength is often employed in industry practice.
1 Background
Before a jackup is installed at a site, a prediction of the spudcans’ penetrations into the seabed as a function of
the imposed loads should be made. The actual load-penetration response is recommended to be monitored on
site, and this response should be compared with the predictions.
The purposes of this load-penetration prediction are to:
• Evaluate whether the rig may be able to operate at the site (e.g., whether the leg length available is
sufficient).
• Identify any potentially hazardous conditions (e.g., the possibility of punch-through), so that plans can
be made to mitigate risks.
• Provide a benchmark against which the actual load-penetration performance can be compared. Deviations
from the predictions may indicate an inadequate understanding of ground conditions. In this latter case
the consequences depend critically on the nature of the deviations, and whether there are possible
implications of hazards.
To make prediction the following information is required:
• Geometry of the spudcans
• A ground model and design soil profile
• Expected light-ship load and maximum preload on each spudcan
Recommended best practice is that a complete load penetration curve should be developed, normally extending
to a depth at least, the greater of:
• The predicted penetration at 1.5 times the maximum preload value
• 0.5 times the spudcan diameter below the predicted penetration at the preload value
Analysis and design tools to determine the spudcan penetration and resistance curve can be classified as
two general methods (e.g., advanced numerical analysis and simplified prediction methods).
3 Numerical Simulations
Lagrangian finite element analysis of large deformation problems such as spudcan penetration often
encounters numerical problems arising from excessive element distortion causing degradation of accuracy
and convergence difficulty. The Eulerian formulation does not face this problem as it permits the mesh to
deform independently of the material. It is widely used in fluid mechanics applications, and has also been
applied to other problems like deep penetration analysis involving geological media. A numerical simulation
method is convenient for simulating single layer or double layer soil with complete soil parameters. The
simulation should be assessed and verified by centrifuge test result or field data.
The spudcan is normally simulated as a rigid body using Lagrangian element, which means that the deformation
of the spudcan during penetration is neglected. A void mesh layer allows the soil to heave and flow into the
initially empty Eulerian elements. The soil domain is simulated as Eulerian elements because of severe soil
distortion. Typically, commercially available software employs explicit Eulerian analysis, which uses a total
stress approach and provides no information on excess pore water pressure.
The undrained, effective stress Eulerian finite element analysis of spudcan penetration in clay was
developed, which allows excess pore pressure to be computed [Ref.6]. Three elastic–plastic soil models, namely
the modified Cam-Clay, Mohr-Coulomb and Drucker-Prager models, are implemented in ABAQUS and
applied to the spudcan penetration problem. Comparisons with centrifuge model data indicate that provided the
undrained shear strength profile of the ground is well-replicated, so will the penetration resistance and pore
pressure response. Both measurement and computation show significant levels of excess pore pressure being
generated during preload. It is, therefore, reasonable to expect that the subsequent dissipation of this excess pore
pressure will lead to significant changes in the strength of the soil around the embedded spudcan. Hence,
computing pore pressure build-up is an essential step in assessing the long-term performance of spudcan
foundations.
5.1 General
The following Subsection deals with the calculation of the load-penetration curve for a deposit that can be
treated as a single layer of clay or sand. Multi-layers of soil will be discussed separately in the next
Section.
Penetrations in carbonate sands are highly unpredictable, and they may be small in strongly cemented materials,
or large, in un-cemented materials. Extreme care should be exercised when operating in these materials.
For silts, it is recommended that upper and lower bound analyses for drained and undrained conditions are
performed to determine the range of expected penetrations. The upper bound solution is modeled as loose
sand and the lower bound solution as soft clay.
This does not give hc/D directly in terms of Sum, and some iteration is needed, making use of:
FIGURE 1
Embedded Spudcan with Open Soil Cavity
mudline
Sum SuH Su0
Su
hc
h
z ρ
ym
Su avg
D z
The ultimate vertical bearing capacity of the spudcan foundation Qv in clay can be expressed as follows.
Qv = Su0NcAeff + γ'Vc
where
Su0 = undrained shear strength at the lowest depth of the maximum plan area of
spudcan, see Section 3, Figure 1
Nc = bearing capacity factor at shallow spudcan penetration prior to any backflow
πDeff
2
Aeff = , in m2 (mm2, in2)
4
Vc = volume of the conical spudcan below mudline, in m3 (mm3, in3)
FIGURE 2
Definition of Equivalent Cone
Mudline
ym
z
Volume Vc
Deff
The above calculations make the approximations that, until the critical depth, there is no backflow; but
after the critical depth, there is full incremental backflow around the spudcan. This is clearly an idealization,
as in reality a more gradual transition will occur.
Nc is estimated using the tabulated values suggested [Ref.8], see Section 3, Table 1. The bearing capacity
factor is a function of the cone angle β, the dimensionless embedment depth h/D, the roughness factor α,
(0 ≤ α ≤ 1, a value of 0.5 is recommended in the absence of evidence that supports any other value), and
the dimensionless measure of the rate of increase of strength with depth ρD/Sum. Alternative published
values of bearing capacity factors may be used if they can be justified.
TABLE 1
Bearing Capacity Factors Nc for Conical Shaped Footings
ρD/ β = 30 β = 60 β = 90 β = 120 β = 150
Sum
h/D α=1 α=0 α=1 α=0 α=1 α=0 α=1 α=0 α=1 α=0
0.0 0.0 8.79 4.61 6.68 4.45 6.17 4.64 6.05 4.95 6.06 5.32
0.1 8.95 4.80 6.89 4.68 6.40 4.90 6.29 5.22 6.31 5.59
0.25 9.18 5.05 7.17 4.98 6.71 5.22 6.61 5.57 6.61 5.94
0.5 9.50 5.41 7.56 5.41 7.13 5.68 7.04 6.03 7.05 6.41
1.0 10.03 5.98 8.18 6.07 7.78 6.37 7.71 6.73 7.72 7.12
2.5 11.10 7.12 9.39 7.33 9.04 7.64 8.98 8.06 9.00 8.46
1.0 0.0 14.47 7.53 8.87 5.81 7.53 5.56 7.09 5.68 6.96 5.93
0.1 14.13 7.45 8.88 5.92 7.65 5.74 7.25 5.88 7.15 6.16
0.25 13.72 7.38 8.91 6.04 7.79 5.93 7.45 6.11 7.35 6.40
0.5 13.24 7.28 8.94 6.2 7.97 6.16 7.65 6.39 7.59 6.70
1.0 12.68 7.21 9.03 6.43 8.20 6.49 7.97 6.79 7.94 7.12
2.5 12.19 7.34 9.38 6.97 8.77 7.24 8.61 7.52 8.60 7.90
2.0 0.0 20.10 10.45 10.98 7.14 8.82 6.46 8.03 6.37 7.73 6.50
0.1 18.40 9.65 10.50 6.92 8.64 6.41 7.97 6.41 7.74 6.59
0.25 16.72 8.89 10.02 6.74 8.64 6.40 7.93 6.46 7.76 6.68
0.5 15.08 8.20 9.60 6.59 8.34 6.40 7.91 6.56 7.81 6.83
1.0 13.54 7.60 9.29 6.55 8.32 6.53 8.03 6.80 7.98 7.10
2.5 12.35 7.37 9.37 6.99 8.71 7.15 8.53 7.42 8.52 7.80
3.0 0.0 25.71 13.36 13.09 8.49 10.08 7.35 8.93 7.04 8.43 7.03
0.1 22.00 11.51 11.84 7.77 9.44 6.99 8.57 6.83 8.20 6.93
0.25 18.85 9.98 10.81 7.24 8.93 6.69 8.27 6.71 8.03 6.87
0.5 16.18 8.75 10.00 6.82 8.56 6.54 8.07 6.65 7.93 6.90
1.0 13.98 7.79 9.42 6.61 8.38 6.55 8.06 6.80 7.99 7.23
2.5 12.42 7.40 9.36 6.99 8.68 7.11 8.50 7.38 8.49 7.76
4.0 0.0 31.32 16.27 15.18 9.83 11.33 8.22 9.81 7.69 9.09 7.54
0.1 25.08 13.10 12.99 8.51 10.13 7.48 9.03 7.20 8.58 7.22
0.25 20.44 10.83 11.41 7.61 9.29 6.94 8.52 6.87 8.22 7.01
0.5 16.91 9.11 10.26 6.97 8.71 6.63 8.17 6.72 8.01 6.95
1.0 14.23 7.91 9.50 6.64 8.42 6.56 8.07 6.80 8.00 7.09
2.5 12.46 7.40 9.35 6.86 8.67 7.04 8.48 7.38 8.47 7.71
5.0 0.0 36.92 19.18 17.26 11.17 12.56 9.11 10.66 8.34 9.73 8.04
0.1 27.75 14.48 13.99 9.14 10.74 7.87 9.45 7.52 8.90 7.46
0.25 21.68 11.46 11.86 7.90 9.56 7.12 8.71 7.01 8.37 7.12
0.5 17.43 9.37 10.44 7.08 8.81 6.69 8.25 6.76 8.06 6.98
1.0 14.40 7.98 9.55 6.66 8.44 6.57 8.08 6.80 8.00 7.08
2.5 12.48 7.40 9.35 6.85 8.66 7.03 8.47 7.34 8.45 7.70
• If z ≤ ym, then there is partial penetration and no backfill can occur and the following bearing capacity
expression should be used:
1
Qv = γ'DeffNγFmobAeff + γ'Vc
2
• If z > ym, then the following bearing capacity expression should be used:
1
Qv = Fmob γ′DN γξ hγ + γ′hN qξ sqξ hq A + γ'(V – Vsoil)
2
where
Vsoil = volume of the backfill soil that rests on the spudcan, in m3 (mm3, in3), see
Section 3, Figure 3
Nγ, Nq = bearing capacity factors
FIGURE 3
Backflow in Sand
Mudline
φcv
Volume Vsoil
The bearing capacity factor, Nγ, is obtained in Section 3, Table 2 [Ref.9]. These are calculated directly for
conical shaped footings, and are presented in terms of the cone apex angle, β, the roughness factor, α, (0 ≤
α ≤1, a value of 0.5 is recommended in the absence of evidence that supports any other value), and the
friction angle, φ'. The values [9] are only presented for surface footings (h = 0). For h > 0, the value ξhγ =
1.0 is recommended [Ref.10] (i.e., no adjustment should be made to this term to account for depth of
embedment).
The bearing capacity factor, Nq, is only relevant once the lowest maximum bearing area is below the mudline
(h > 0). The following equation was suggested for a plane strain surface footing [Ref.10]:
π φ′
Nq = eπtanφ'tan2 +
4 2
The empirical shape factor ξsq = 1 + tan φ' is applied to convert the plane strain condition to a value
appropriate for axial symmetry.
A depth factor ξhq = 1 + 2tanφ'(1 – sinφ')2tan-1(h/D) allows for an increase in the contribution of this factor
with depth. However, no account is taken of the influence of cone angle or roughness on the Nq factor.
If a calculation is based on realistic values of the peak angle of friction, for a penetration process such as
that of an approximately conical spudcan, the soil resistance will be significantly overestimated since the
peak strength is not mobilized simultaneously through the deforming soil. In the SNAME 5-5A, this issue
is addressed by artificially reducing the angle of friction used in the bearing capacity calculation. This
procedure involves the use of unrealistically low angles of friction that may bear little resemblance to
measured values. Here a mobilization factor, Fmob, is used to the calculated resistance. Based on back
analysis of 10 case records [Ref.1], of which eight are from the North Sea Region, and the others from the
Gulf of Mexico and offshore Australia, a value of the reduction factor of 0.25 to 0.5 is suggested. It would
be expected that lower values of Fmob would be applicable to more compressible materials (e.g., carbonate
sands) and higher values for stiffer materials, but little quantitative evidence exists. If alternative values of
the mobilization factor are better established they should be used. It is arguable that, for consistency, a similar
procedure should be used in clays, but experience shows that for clays a calculation that does not employ any
such factors provides sufficient accuracy, so that the added complication is not justified.
Note that the above procedures are intended solely for the estimation of the load-penetration response of
the spudcan during preloading. Calculation procedures are set out in SNAME 5-5A and ISO 19905-1 for
determining fixity of spudcans under combined horizontal and moment loading, and these require selection
of appropriate bearing capacity factors. However, calculation procedures in SNAME 5-5A and ISO 19905-1
are not consistent with the above procedure, and the different modeling methods should not be combined.
TABLE 2
Bearing Capacity factors Nγ
Cone Apex Angle
β = 30 β = 60 β = 90 β = 120 β = 150
Friction Roughness
Angle φ' Factor
(degrees) α=1 α=0 α=1 α=0 α=1 α=0 α=1 α=0 α=1 α=0
5 2.63 1.87 1.07 0.86 0.62 0.52 0.39 0.33 0.22 0.19
10 5.29 2.83 2 1.3 1.17 0.82 0.78 0.56 0.55 0.37
20 20.86 6.62 7.33 3.08 4.54 2.11 3.37 1.69 2.73 1.43
30 89.8 16.26 31.99 7.95 21.12 6.22 17.58 5.77 15.93 6.27
40 504.1 45.24 209.2 23.75 142.8 22.13 129.4 25.84 128.1 34.36
50 6504.3 145 2650 108.8 1923.3 115.2 1905.4 180.2 1981.3 347.2
In comparison to spudcan penetration without a lattice leg, backflow is found to be partially inhibited; and
a lattice leg with a smaller opening ratio tends to induce a deeper cavity formation and yield larger transient
vertical bearing capacity. The cavity induced by spudcan penetration is more likely to develop in stiff clay
than in soft clay. However, cavity formation appears to account for only a part of the bearing capacity
improvement. The other part of the increase is likely to be due to the change in the back flow pattern due to
the presence of the leg itself. Moreover, the simulated leg displaying the same radius with the spudcan footing
(Aa = 1) is shown to yield the largest vertical bearing capacity compared to one with smaller leg radius
(Aa = 0.61). A transient vertical bearing capacity improvement of approximately 30% can be inferred compared
to the spudcan without the leg effect considered.
Usually the opening ratio, e, of a lattice leg is about 0.8 to 0.9. To reduce the opening ratio one possible
way is to employ a top-mounted skirt on a spudcan. In this arrangement, the spaces between chords of a
conventional lattice leg are closed by installing a plate between each of two adjacent chords. The plates
extend upward from the top of the spudcan to the soil surface, see Section 3, Figure 5. The area ratio is
about 0.5. In comparison with conventional spudcans without skirts and spudcans with downward skirts
extending to the cone tip, a top-mounted skirted spudcan has improved global bearing capacity, spudcan
fixity, and resistance against punch-through, without compromising hydrodynamic performance [Ref.15].
If the top-mounted skirted spudcan also has downward skirts, the effect of the above three benefits will be
amplified.
FIGURE 4
Lattice Leg View
83
83
Spudcan
Lattice
Opening
leg
83 60
Lattice
R60
leg
FIGURE 5
Top Mounted Skirt on Spudcan
1 General
Punch-through potentially occurs in ground exhibiting post-peak reduction in bearing resistance, in which
exceeding the peak bearing resistance results in excessive and uncontrollable spudcan settlement and loss of
hull trim. Punch-through failure is possibly the most hazardous of all the geohazards for jackup foundations,
and is responsible for more than half the events accounting for loss of life, injury, structural damage, etc.
Prediction methods, as indicated below, may be used to evaluate such risk. Other methods can be used if
they are justified.
3 Prediction Methods
FIGURE 1
Nomenclature for Spudcan Penetration in Sand over Clay
Qv = NcSubsA +
2
(
γ ′sand 2
)
hlayer − h 2 πDKstanφ' + γ ′sand (hA + V – Vsoil)
where
Ks = punching shear coefficient. For cases with 25° < φ' < 35° and
0.6
Su Su
0.05 < < 0.5, Kstanφ' ≈ 2.5
γ ′sand D γ ′sand D
Su = undrained shear strength at h + D/4 below mudline, in N/m2 (kgf/mm2,
lbf/in2)
Vsoil = volume of the backfill soil that rests on the spudcan, in m3 (mm3, in3), see
Section 3, Figure 3
Other parameters are as defined in Section 4, Figure 1. For determination of bearing capacity
factor, see Section 4, Figure 2 and for the selection of φ', see 2/9.5 [Ref.8].
For dense sand overlying soft clay, back-analysis of centrifuge test results indicates that the above
method underestimates peak resistance. Calculation for such soil profiles can be improved by using
methods in [Ref.18] and [Ref.19]. These methods provide estimates of bearing capacity when the
spudcan is at different depths, taking into consideration the change in shear strength of sand, and
accounting for the progressive change of failure mechanism.
• For h = hlayer
4S ua h plug
Qv = N c S u , plugbase + σ v′ + A – γ'Vsoil
D
• For h ≥ hlayer +ht
4S ua (h plug + ht )
Qv = N c S u , plugbase + σ v′ + A – γ'Vsoil
D
where
Nc = bearing capacity factor
Su,plugbase = undrained shear strength corresponding to the level of the sand plug base, in
N/m2 (kgf/mm2, lbf/in2)
= effective stress at the level of the sand plug base, in N/m2 (kgf/mm2, lbf/in2)
Sua = undrained shear strength averaged from (h – ht) to (h + hplug), in N/m2
(kgf/mm2, lbf/in2)
hplug = height of the sand plug, in m (mm, in)
ht = height of the spudcan widest diameter, in m (mm, in)
FIGURE 2
Values for Spudcan Bearing Capacity Calculation when h ≥ hlayer [Ref.18]
FIGURE 3
Nomenclature of Spudcan Penetration in Sand over Clay when h ≥ hlayer
φcv
SAND
hlayer Qv [γ′sand, φ′ or (ID, φcv)]
h
Subs
Sub
CLAY 1
[Sub or (Subs, ρ), γ′clay] ρ
Trapped soil
plug
z
FIGURE 4
Nomenclature of Spudcan Penetration in Strong Clay over Soft Clay
Qv
h
Strong CLAY
hlayer [γ′clay,top, Sutop]
h′layer
D
Subs
Sub
Soft CLAY 1
[γ′clay,b, Sub or (Subs, ρ)] ρ
To date, perforation drilling has been achieved using the rotary open hole water flush method. If insufficient
flush rates are used then the disturbed soil may not be removed from the hole. On the other hand, if excess
drilling pressures are applied then this may lead to hydraulic fracturing and linkage between adjacent
perforations. Both scenarios significantly reduce the effectiveness of the operation.
By using reverse circulation (i.e., using an airlift to remove cuttings from the base of the hole up through
the center of the drill string to surface) the effectiveness of the excavation, and hence overall process, may
be increased as the material removal efficiency will improve. Since there is little experience with this
method for perforation drilling, it is not possible to confirm its efficiency and reliability on vessels so far
used for this operation. Experimental data suggest that factors such as perforation distribution pattern in
particular, perforation depth, drilling methods, etc., may significantly affect the reduction of peak bearing
resistances.
1 Approach
The overall foundation stability assessment may follow SNAME 5-5A and ISO 19905-1. There are three
steps recommended in the order of increasing complexity and reducing conservatism (See Section 5,
Figure 1):
i) Step 1
a) Preload check of leeward leg for pinned spudcan
b) Sliding check of the windward leg for pinned spudcan
ii) Step 2
a) Foundation capacity and sliding check for pinned spudcan
b) Foundation capacity and sliding check for spudcan with moment fixity and vertical and
horizontal stiffness
c) Foundation capacity and sliding check for spudcan assuming non-linear foundation stiffness
iii) Step 3
Foundation displacement check
3 Acceptance Criteria
The adequacy of a foundation to resist bearing and sliding loads, and when considered rotational stiffness,
should be verified using the criteria specified in SNAME 5-5A, ISO 19905-1 or the ABS Rules for Building
and Classing Offshore Installations.
FIGURE 1
Foundation Stability Checks
Perform structural analysis
Step 1a
Preload check OK
Step 1b
Sliding check
Not OK
Step 2a OK
Foundation capacity and
sliding check (pin)
Not OK
Step 2b OK
Foundation capacity and
sliding check
Not OK
Step 2c
Foundation capacity and
sliding check
Not OK
Step 3 OK
Displacement check
on all legs
Not OK
Foundation NOT acceptable Foundation acceptable
1 Introduction
Foundation fixity, so called spudcan soil rotational stiffness, is the rotational restraint offered by the soil
supporting the foundation. Traditional self-elevating unit (SEU) design usually assumes a pinned condition
to represent spudcan and soil interaction, this being equivalent to zero foundation fixity as shown in
Section 6, Figure 1. Since 2003, the MODU Rules permit consideration of foundation fixity for cases
involving dynamic response. ABS Rules allow consideration of a range of fixity values up to the “fully
fixed” condition. An SEU Owner is to verify that the conditions for which the SEU has been approved are
satisfied. Actually, the spudcan/soil rotational supports are partially fixed, see Section 6, Figure 1.
FIGURE 1
Spudcan Soil Rotational Stiffness
1 2 = 0 3 < 1
At present, to assess the suitability of an SEU for installation at a specific site, and hence the adequacy of its
spudcan foundation, the industry relies extensively on SNAME 5-5A. The value of SNAME defined rotational
stiffness is approximately 20%-40% of the “fully fixed” value, while the measured dynamic fixity ranges
from 30% to 80% of the “fully fixed” value [Ref.24]. The key parameters affecting spudcan fixity include
soil permeability, footing embedment, lattice legs, spudcan skirt, time lag between installation and
operation, etc.
3 Foundation Fixity
For a rigid circular footing resting on an elastic half-space, the vertical, horizontal and rotational spring
constants K1, K2, and K3, are given by:
2GD
Vertical stiffness K1 = k d1 in N/m (kgf/mm, lbf/in)
1 −ν
16GD(1 − ν )
Horizontal stiffness K2 = k d 2 in N/m (kgf/mm, lbf/in)
7 − 8ν
GD 3
Rotational stiffness K3 = kd3 in N-m/rad (kgf-mm/rad, lbf-in/rad)
3(1 − ν )
where
G = soil shear modulus, in N/m2 (kgf/mm2, lbf/in2)
ν = Poisson’s ratio of the half space medium
D = footing diameter, in m (mm, in.)
kd1, kd2, kd3 = depth factors for the vertical, horizontal and rotational spring stiffness respectively, as
shown in Section 6 Figures 2, 3 and 4. Case (a) in the Figures represent an open hole
above the spudcan, case (b) in the Figures represent a back filled hole [Ref.25].
h = spudcan embedment depth from mudline to spudcan maximum lowest bearing area,
in m (mm, in.) see Section 3, Figure 1
3.1.1 Selection of Shear Modulus G for Clay
The value of initial, small strain shear modulus G for clay may be calculated as follows.
I rNC
G = Gmax = Su in N/m2 (kgf/mm2, lbf/in2) with G < SuIrNC
OCR 2.25
where
Su = undrained shear strength measured at the depth of spudcan lowest maximum
bearing area plus 0.15 times spudcan diameter, in N/m2 (kgf/mm2, lbf/in2)
OCR = over consolidation ratio
IrNC = rigidity index for normally consolidated clay. For extreme loading
conditions, and in absence of other data, IrNC should be conservatively
limited to 400 based on over-consolidated clay sites with plasticity index of
up to 40% [Ref. 26]. For regions with low OCR and plasticity index less than
40% like the Gulf of Mexico, IrNC could be up to 600.
FIGURE 2
Depth Factors kd1 for Vertical Spring Stiffness
FIGURE 3
Depth Factors kd2 for Horizontal Spring Stiffness
FIGURE 4
Depth Factors kd1 for Rotational Spring Stiffness
• For sand:
HL0 = 0.12 VL0
ML0 = 0.075VL0 D
• For clay:
HL0 = cu0 A + (cu0 + cul)As
ML0 = 0.1VL0 D
where
As = spudcan laterally projected embedded area, in m2 (mm2, in2)
cul = undrained cohesive shear strength at spudcan tip, in N/m2 (kgf/mm2, lbf/in2)
cu0 = undrained cohesive shear strength at lowest maximum bearing area, in N/m2
(kgf/mm2, lbf/in2)
The yield envelopes for a spudcan in soil are illustrated in Section 6, Figure 5(a) & (b).
FIGURE 5
Yield Envelopes for Conical Footings
–1 0 FH/HL0 1
0 FM/ML0
2
FV/VL0 = 0.5
FV/VL0 = 0.25
1 1 FH/HL0
0.5 0.5 0
–2 0 2
FM/(FHB) = 0
1 –2
FV/VL0
When site specific or regionally applicable information is sufficient to establish a value of n, then
that value should be used. Finite element analysis for Gulf of Mexico clay indicates the range of n
is between –0.25 to –1.0, with n = –0.5 providing the best overall representation [Ref.27]. When
no such information is available, a default value of n = 0 could be used. In special cases where an
extremely conservative stiffness reduction is desired a value of n = 1.0 could be used, which leads to:
fr = 1 – rf
5 Consolidation Effect
The zone of clay being remoulded during spudcan penetration will reconsolidate after completion of the
spudcan installation. The extent of consolidation depends on the period between spudcan installation and
operation, soil consolidation characteristics, penetration depth, spudcan geometry etc. The effect of the
consolidation on excess pore pressure response and settlement of spudcan was studied, as well as foundation
fixity by means of centrifuge tests and numerical simulations in Kaolin clay [Ref.28].
In “no dissipation” tests spudcan rocking commenced immediately after installation, since there is no
consolidation. In “full dissipation” tests the excess pore pressure was completely dissipated before
commencement of rocking, which is full consolidation. Compared with the settlements in all the “no
dissipation” tests that show a non-linear increase, the settlements in all the “full dissipation” tests show
much smaller settlements.
The yield envelope, Section 6, Equation (1), is considered to be relatively conservative which cannot
accommodate a long-term combined force locus in “full dissipation” test [Ref.28]. Actually, the penetration
resistance at a specific depth is changing with the excess pore pressure due to pressure dissipation. A ratio
of 1.7 between the long-term and short-term resistance is obtained from the test results. The yield envelop
based on long-term penetration resistance is found to accommodate the entire force locus.
The initial foundation fixity for the “full dissipation” tests is about double that from the “no dissipation”
test since consolidation of the surrounding soil leads to a significant increase of bending moment resistance in
“full dissipation” tests. With an increase in the number of cyclic loadings, foundation fixity in “full dissipation”
tests shows an approximately constant value, while the “no dissipation” tests show an increasing tendency.
This increase is partially caused by the dissipation of excess pore pressure in the soil surrounding the
foundation, and also arises because of the relatively large settlement deeper into the soil specimens. After
one thousand loading cycles the magnitude of the foundation fixity in the “no dissipation” tests is gradually
close to that in the “full dissipation” tests. Hence the post-installation, long-term consolidation can substantially
enhance the loading capacity and rotational fixity of spudcan footing. If such beneficial effects can be
reliably obtained and accounted for, they can benefit the design.
1 Introduction
A jackup rig is often used to do drilling or work-overs adjacent to existing pile supported platforms. Also
the jackup may be sited adjacent to a pile supported structure to provide additional accommodations, power
generation or fabrication space. Furthermore, some modular packages, such as those for drilling apparatus
or construction cranes, are sometimes transferred from a jackup rig to a fixed platform for construction,
drilling or reworking of an existing well. Prior to these works, the jackup should first be positioned
adjacent to the fixed platform. Depending on the platform’s footprint and the location of the jackup rig, the
spudcan foundation of the jackup rig may be close to the permanent pile foundations of the fixed platform, as
shown in Section 7, Figure 1. The proximity of the spudcan to the existing piled platform would induce
stresses, and may affect the performance of the pile foundations, and subsequently causes distress to the
superstructure, see Section 7, Figure 2. If the piles have not been designed to withstand these stresses, the
structural integrity of the piles may be threatened.
According to SNAME 5-5A, if the foundation materials comprise either a deep layer of homogeneous firm
to stiff clay or sand, and if the proximity of the spudcan to the pile is greater than one spudcan diameter, no
significant pile loading is expected. When the proximity is closer than one spudcan diameter, then analysis
by the platform owner is recommended to determine the consequences of the induced pile loading.
The induced loading on the pile is the function of:
• Spudcan-pile clearance
• Spudcan and pile diameters
• Spudcan penetration depth
• Pile length for a floating pile
• Pile socket length for a socketed pile
• Upper clay thickness for a dual layered soil profile
• Soil rigidity index
• Pile bending stiffness
FIGURE 1
Potential Soil Loading Effects on Jacket Platform (after [Ref.29])
Jacket
Translation
Remolded Displaced
Zone Pile
Pile
NOT TO SCALE
Conductors
FIGURE 2
Spudcan-Pile Interaction
Operation
Induced Soil Movement
Removal
Pile Response
Bending Deflection Axial Settlement
Moment Force
Distress to Superstructure
Assessment Improvement
FIGURE 3
Incremental Soil Movement Trajectories at Different Distances from Spudcan
0.25D 0.5D 0.75D 1D 0.25D 0.5D 0.75D 1D
5 Dimensionless Charts
Systematic, dimensionless charts may be used to perform preliminary estimates on the severity of spudcan-
pile interaction problems in various field conditions [Ref.31]. However, a detailed three-dimensional finite
element analysis should be performed in situation where higher accuracy is required.
The maximum bending moment is estimated by multiplying the bending moment for a case of a socketed
pile having spudcan-pile clearance of 1D with correction factors to account for the effects of spudcan-pile
clearance, pile socket length and thickness of upper clay layer for a socketed pile, clay rigidity index and
pile bending stiffness, where appropriate, see Section 7, Figure 4. Each of the above parameter is represented by
an adjustment factor.
The bending moments in the charts are for a fixed-headed pile. It is postulated that this would result in a larger
induced pile head bending moment when compared to the actual situation where the pile head’s rotational
degrees of freedom are considered but the translational degrees of freedom are partially fixed due to the
fact that the pile head can deflect together with the platform leg. As such, the measured induced pile bending
moments should generally be on the conservative side. The bending moment responses of piles were
compared with free head and totally fixed head, which represents for the two extreme cases [Ref.32].
Different from the fixed headed pile, the elevation of maximum bending moment for free-headed pile is
located around the center of the pile rather than at the pile head. This is because the free restraint at both ends
causes the maximum moment to be focused at the center. However, the magnitude of the maximum
bending moment for a free-headed pile is about one third that for a fixed-headed pile. Therefore, the
dimensionless charts lead to a conservative estimation of pile bending moment due to spudcan installation.
On the other hand, as the soil moves continuously upward, the fixed-headed pile is found to be in
compression due to the restraint at the pile top; while the free-headed pile is in tension. However, the
magnitude of axial force is comparable for the two kinds of piles.
The charts are derived from specific soil conditions consisting of normally consolidated kaolin clay with
typical strength gradient of 1.6 kPa/m depth overlying a layer of sand which represents a hard soil. In cases
where multiple soil layers or soil with irregular shear strength are present, additional care should be exercised
to evaluate the validity of the procedure and to decide whether a detailed analysis should be conducted.
FIGURE 4
Nomenclature for a Socketed Pile
Clay
Spudcan
Penetration
Depth
Clay
Thickness
Pile
Spudcan-Pile Length
Clearance
Pile Socket
Length
Sand Sand
Thickness
1 Introduction
After a jackup is removed from a site, spudcan footprints are left in the seabed, as shown in Section 8,
Figure 1. The spudcan footprints are ground conditions that experience:
• Changes in physical profile of the seabed (existence of depression); and
• Changes in soil properties.
The soil beneath a depression may be highly non-uniform due to back flow of remoulded soils during and after
spudcan penetration and extraction, and reconsolidation of the soil. If the subsequent positioning of another
jackup is very close to or partially overlapping the footprints, the slope of the footprint and the varying soil
strength inside and around the footprint results in an eccentric/inclined soil reaction on the spudcan. This can
cause a spudcan to slide towards the footprint and hence leads to overloading the leg. Thus pre-loading of
jackups near existing footprints can result in uncontrolled penetration, slewing of the rig and even excessive
structural stresses in the legs, which might even lead to catastrophic failure. The footprint feature is dependent
on various factors such as footing shape and size, soil types and strength, previous spudcan size and penetration,
and the elapsed time of previous spudcan operation and after extraction.
FIGURE 1
Bathymetry of an Established Site
SNAME 5-5A suggests that a jackup identical in footing geometry is installed and located in exactly the
same position as that of previously installed unit. It is important to verify that the footing geometry and
position of the rig are exactly similar to the previous jackup. If it is not possible to locate exactly in the
previous location, a minimum distance of one spudcan diameter from the edge of bearing area to the edge
of the footprint is recommended.
A strength ratio of less than 0.5 is categorized as heavily remoulded, which is basically confined within the
spudcan area Rd/Df ≤ 0.5 (Rd is the radial distance from spudcan center to footprint center and Df is the
diameter of the spudcan which forms the footprint). The soil with strength ratio of 0.5 to 0.7 is classified as
moderately remoulded, whereas a ratio range of 0.7 to 0.9 is less remoulded. The extent of radial soil
disturbance varies with depth. At depths of up to 0.5Df, the radial disturbance is found to extend an Rd/Df
ratio of 1.5. Below this depth, the major radial disturbance (r < 0.7) is found to be confined within Rd/Df of
0.75. In terms of vertical extent, it is found that the soil is heavily remoulded to a depth of up to 0.9de and
moderately remoulded from 0.9 to 1.1de, where de is the penetration depth to spudcan base level. Below
this depth, a minor soil strength reduction (0.7< r <0.9) extending up to 0.3Df is observed.
Based on the observations from centrifuge tests, a simplified soil failure mechanism at different penetration
depths is postulated [Ref.33]. For the penetration depth above the crater depth, the horizontal forces and
moments are caused by the non-uniform soil bearing resistance where the resultant soil reaction inclines at
an angle of θ to vertical and an eccentricity of e from the spudcan center, as indicated in Section 8, Figure 3(a).
This inclined eccentric soil reaction tends to push and rotate the spudcan from the stronger soil side towards
the weaker side. On the other hand, when the spudcan penetrates further below the crater, the soil bearing
capacity failure along the sliding surface is initiated that drives the spudcan towards the footprint center. At
the same time, the resistance from the soil outside the sliding plane tends to push the spudcan towards the
footprint center, as shown in Section 8, Figure 3(b). If the horizontal and rotational movements are restricted,
the tendency of horizontal movement and rotation are then transmitted as horizontal forces and rotational
moments acting at the spudcan. A combination of these two forces has the effect of magnifying the magnitude
of induced horizontal force (as both forces tend to move the spudcan towards the footprint center) while
diminishing the magnitude of induced rotational moment (caused by rotation in opposite direction).
FIGURE 2
Generalized Soil Condition of a Footprint
(for τ1 < 0.002; τ2 < 0.2)
Rd/Df
0 0.25 0.50 0.75 1.00 1.25 1.50
~0.2D Crater
0.4~0.5D
0.8~0.9de
1~1.1de
0.2~0.3D
Legends Where
r < 0.5 Su_footprint
r=
(Heavily remoulded zone) Su_undisturbed
0.5 < r < 0.7 de = penetration depth to spudcan
(Moderately disturbed zone) base level
Df = diameter of spudcan used
0.7 < r < 0.9 to create footprint
(Less disturbed zone)
Rd = radial distance
In normally consolidated clay the soil strength variation is the key factor influencing spudcan footprint
interaction. The spudcan will experience relatively high horizontal and moment loadings below the previous
rig’s penetration depth. Hence one may consider a rig with lower required preload pressure where the rig
installation can be terminated at a sufficient distance above the previous penetration depth.
In over consolidated clay the physical profile of the depression can be the dominant factor in spudcan-
footprint interaction. Hence the spudcan will likely experience high horizontal and moment loadings above
the previous penetration depth. If these forces are found to be hazardous to the new rig installation,
avoiding the footprints should be considered or use an identical rig to re-install at the same position of the
previous rig as recommended in SNAME 5-5A. If the spudcan-footprint interaction is unavoidable due to
site constraints, one may consider to carefully position the new rig in such a way that the critical center to
center spacing between the initial penetration and re-penetration of 0.5 to 1.0 times spudcan diameter can
be avoided. If that fails an effective mitigation measure to overcome the spudcan-footprint interaction
should be pursued.
FIGURE 3
Simplified Diagram of Probable Soil Failure Mechanisms
During Penetration at 0.5D from Footprint Center
CL Footprint
θ
e
Rs
CL Footprint
Model Ground
Mbearing
Hbearing
Mslide Hslide
5 Mitigation Methods
The Rack Phase Difference (RPD) system is an effective way to monitor the sliding of a spudcan into an
old footprint and hence prevent the leg damage [Ref.35]. RPD is the difference in elevation between the
chords of any one leg, and it is a direct measure of the inclination of the leg with respect to the hull. RPD
monitoring during the jacking process enables the operator to monitor that the RPD limits are not exceeded,
and to take necessary steps otherwise by stopping the jacking operation. Thus monitoring gives real time control
of the operation.
Infilling of old footprints formed in a coarse grained sea bed with the granular material has been carried
out successfully, and it is well understood that infilling will not pose problems if the material characteristics
are similar. The studies confirmed that the difference in the stiffness between the footprint and the infill
material greatly affects the effectiveness of infilling and the infilling of footprints in layered cohesive
materials, and therefore is not an advisable solution for the footprint interaction problem [Ref.36].
In firm to stiff clay, the physical profile of the depression whose profile is steep and deep is significant in
influencing spudcan footprint interaction. To minimize the effect of a footprint, evening out the depression
by excavation might be done. Leveling the seabed profile is unlikely to be effective in soft clay as the critical
horizontal and moment loadings are due to the variation of soil strength well below the crater depth.
In soft to stiff clay, “stomping” is reportedly very effective in mitigating spudcan footprint interaction. It is
a process where the footings are initially emplaced further away from the old footprint center and then used
to displace soil towards the old footprint at desired positions to widen the disturbed regions by controlled
additional disturbance (Section 8, Figure 4). Reaming, also known as leg working or leg reciprocation is
another possible mitigation method, although its effectiveness is not as good as stomping. To be effective,
reaming should be executed with small penetration-extraction increments [Ref.37]. The major limitations
for these options are the significant rig time involved, and careful planning is also needed for the program.
This may only be possible under mild environmental conditions, and the operation also depends on the
clearance between the jackup and the adjacent fixed structure.
FIGURE 4
Stomping Process
APPENDIX 1 References
1. Osborne, J.J., Teh, K.L., Houlsby, G.T., Cassidy, M.J., Bienen, B. and Leung, C.F. (2010). “InSafeJIP"-
Improved Guidelines for the Prediction of Geotechnical Performance of Spudcan Foundations During
Installation and Removal of Jack-up Units, Joint Industry-Funded Project. http://insafe.woking.rpsplc.co.uk/
2. Jamiolkowski, M.B., Lo Presti, D.C.F. and Manassero, M. (2003) Evaluation of relative density and shear
strength of sands from cone penetration test (CPT) and flat dilatometer (DMT), Soil Behaviour and Soft
Ground Construction, Eds. J.T. Germain, T.C. Sheahan and R.V. Whitman, ASCE, GSP 119, 201-238.
3. Bolton, M.D. (1986). The strength and dilatancy of sands. Géotechnique, 36 (1), 65-78.
4. Randolph, M.F., Jamiolkowski, M.B. and Zdravkovi, L. (2004). Load carrying capacity of foundations.
Proc. Skempton Memorial Conf., London, Vol. 1, 207-240.
5. SNAME (Society of Naval Architects and Marine Engineers). (2008). Recommended practice for site specific
assessment of mobile jack-up units, SNAME Technical and Research Bulletin 5-5A, Rev. 3, New Jersey.
6. Yi J.T., Lee F.H., Goh S.H., Zhang X.Y. and Wu J.F. (2011). Eulerian Finite Element Analysis of Excess
Pore Pressure Generated by Spudcan Installation into Soft Clay. Computers and Geotechnics.
7. Hossain, M.S., Randolph, M.F., Hu, Y., White, D.J. (2006). Cavity Stability and Bearing Capacity of Spudcan
Foundations on Clay. OTC 17770.
8. Houlsby, G.T and Martin, C.M. (2003). Undrained bearing capacity factors for conical footings on clay.
Géotechnique, 53 (5), 513-520.
9. Cassidy, M. J. & Houlsby, G. T. (2002). Vertical bearing capacity factors for conical footings on sand.
Geotechnique 52, No. 9, 687–692.
10. Vesic, A. S. (1975). Bearing capacity of shallow foundations. In “Foundation Engineering Handbook” (ed.
H. F. Winterkorn and H. Y. Fang), 121-147. New York: Van Nostrand.
11. ISO 19905-1:2012. Petroleum and natural gas industries – Site-specific assessment of mobile offshore unit.
Part 1: Jack-ups. International Organization for Standardization (2012).
12. Springman S.M. & Schofield A.N. Monotonic lateral load transfer from a jack-up platform lattice leg to a
soft clay deposit. Proc. Conf. Centrifuge, 1998.
13. Menzies, D., and R. Roper, (2008), Comparison of Jackup Rig Spudcan penetration Methods in Clay, Offshore
technology conference, Houston, Texas, U.S.A.
14. Li Y.P., Zhang XY, Lee F H, Goh SH, Wu JF & Yi JT&. (2013). Effect of lattice leg on penetration
resistance of spudcan foundation-Physical and Numerical modeling. The 8th International Conference on
Physical Modelling in Geotechnics (ICPMG2014), Perth, Australia.
15. Zhang X.Y., Li Y.P., Yi J.T., Lee F. H., Tan P.L., Wu J.F. & Wang S. Q., (2015). A Novel Spudcan Design
to Enhance Foundation Performance. ISFOG 2015, Norway.
16. Yu, L., Kumar A., Hossain M. S. and Hu Y.X. (2012). Mitigation of Punch-through Failure of Spudcan
using Skirted Footing on Sand over Clay Soils. Proceedings of the Twentieth (2010) International Offshore
and Polar Engineering Conference, Beijing, China, June 20-25.
17. Hanna, A.M. and Meyerhof, G.G. (1980). Design chart for ultimate bearing capacity of foundation on sand
overlying soft clay. Can. Geotech. J. 17(2), 300–303.
18. Teh K. L. (2007). Punch-through of spudcan foundation in sand overlying clay. Ph.D thesis, National University
of Singapore, Singapore.
19. Lee, K. K. (2009). Investigation of potential spudcan punch-through failure on sand overlying clay soils.
PhD. Thesis, The University of Western Australia, Perth.
20. Craig, W.H. and Chua, K. (1990). Deep penetration of spudcan foundations on sand and clay. Géotechnique,
40(4), 541-556.
21. Hossain, M.S. and Randolph, M.F. (2009). New mechanism-based design approach for spudcan foundations
on stiff-over-soft clay. Offshore Technology Conference, OTC19907.
22. Hossain, M.S. and Randolph, M.F. (2010). Deep-penetrating spudcan foundations on layered clays: numerical
analysis. Géotechnique, 60(3), 171-184.
23. ABS Rules for Building and Classing Mobile Offshore Drilling Units (2014). Part 3 Hull Construction and
Equipment.
24. Temperton I, Stonor RWP & Springett CN. (1999). Measured spudcan fixity: analysis of instrumentation
data from three North Sea jack-up units and correlation to site assessment procedures, Marine Structure 12:
277-309.
25. Bell R.W. (1991). The Analysis of Offshore Foundations Subjected to Combined Loading. MSc. Thesis
presented to the University of Oxford.
26. Noble Denton Europe and Oxford University (2005). The Calibration of SNAME Spudcan Footing
Equations with Field Data. Report No L19073/NDE/mjrh, Rev.4, dated 21st Nov. 2005.
27. Templeton, J.S. (2009). Spudcan Fixity in Clay, Further Results from a Study for IADC. Proc. 12th
International Conference, The Jack-Up Platform, City University, London. 2009.
28. Lee F.H., (2013). Final Report for Research Project: Spudcan Fixity for Realistic Design of jack-Up Units.
Project No. 082 135 0042. National University of Singapore.
29. Mirza, U.A., Sweeney, M. and Dean A.R. (1988). Potential effects of jack-up spud can penetration on jacket
piles. Offshore Technology Conference. OTC 5762, 147-157.
30. Leung C.F., Chow Y.K., Somsak, S., Tho K.K. & Xie Y. (2009). Spudcan-Pile Interaction Joint Industry
Project. National University of Singapore.
31. Leung C.F., Tho K.K., Chow Y.K., Xie Y, Wong P.C. & Purwana O.A. (2012). Experimental and Numerical
Studies of Spudcan-Pile Interaction. OTC 23053.Houston, Texas, USA.
32. Xie Y. (2009). Centrifuge Model Study on Spudcan-Pile Interaction. Ph.D Thesis. National University of
Singapore.
33. Gan C.T. (2009). Centrifuge Model Study on Spudcan-Footprint Interaction. Ph.D Thesis. National University
of Singapore.
34. Stewart, D.P. and Finnie, I.M.S. (2001). Spudcan-footprint interaction during jack-up workovers. International
Society of Offshore and Polar Engineers (ISOPE), Cupertino, California. Vol. 1. pp. 61-65.
35. Foo, K.S., Quah, M.C.K., Wildberger, P. and Vazquez, J.H. (2003). Spudcan footprint interaction and rack
phase difference. Proc. 9th Int. Conf. Jackup Platform Design, Construction and Operation, City University,
London, UK
36. Jardine, R.J., Kovacevic, N., Hoyle, M.J.R., Sidhu, H.K. and Letty, A. (2001). A study of eccentric jack-up
penetration into infilled footprint craters. Proc. 8th International Conference the Jack-up Platform, eds:
C.D’Mello and L. Boswell, City University, London.
37. Hartono, C.F. Leung, K.K. Tho and Y.K. Chow. (2013). Reaming as a mitigation measure for jackup
reinstallation close to existing footprint. International Conference on the Jack-up Platform, City University,
London, UK. 17th & 18th Sep. 2013.