See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/316447130

Cyclic Ratcheting Resistance of Buried Pipelines

Conference Paper · May 2017

DOI: 10.4043/27823-MS

CITATIONS READS
5 2,127

4 authors, including:

Indrasenan Thusyanthan Stuart Kenneth Haigh

University of Cambridge University of Cambridge
58 PUBLICATIONS   776 CITATIONS    184 PUBLICATIONS   2,181 CITATIONS   

SEE PROFILE SEE PROFILE

D.J. Robert
RMIT University
123 PUBLICATIONS   882 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Microbially induced calcite precipitation View project

Numerical and centrifuge modeling of Onshore Wind Turbines under seismic loading View project

All content following this page was uploaded by Indrasenan Thusyanthan on 20 May 2017.

The user has requested enhancement of the downloaded file.

OTC-27823-MS

Cyclic Ratcheting Resistance of Buried Pipelines

N. I Thusyanthan, Saudi Aramco, J. Wang, DNVGL UK, S.K. Haigh, University of Cambridge, D.J. Robert, RMIT

Copyright 2017, Offshore Technology Conference

This paper was prepared for presentation at the Offshore Technology Conference held in Houston, Texas, USA, 1–4 May 2017.

This paper was selected for presentation by an OTC program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Offshore Technology Conference, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the
written consent of the Offshore Technology Conference is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words;
illustrations may not be copied. The abstract must contain conspicuous acknowledgment of OTC copyright.

Abstract
Upheaval buckling (UHB) is a common issue for high pressure and high temperature (HPHT) pipelines. The out-of-
straightness (OOS) of the buried pipelines combined with the high axial compressive forces induced by the extreme operating
conditions causes the pipeline to move upwards till the downward soil resistance is sufficient to resist any further movement.
As the pipeline mobilise upwards, the gap underneath is soon filled up by soil particles from around the pipe. As a result,
when pipelines cool down during temporary suspension in operation, they cannot return to their original position. As
pipelines undergo many thermal cycles of start-up and shut-down during the lifetime, a pipeline OOS feature can
progressively move upwards and eventually lead to failure even if it was designed safely against UHB in the original profile.
This phenomenon of cyclic pipe movements upwards is known as pipeline ratcheting, which is the driving mechanism behind
many UHB pipeline failure cases. This paper presents insight into ratcheting and results of small scale experiments that were
conducted in Schofield Centre, University of Cambridge, to model the ratcheting response of pipelines buried in saturated
sands. The results obtained have shown that the critical upward mobilization at which ratcheting is initiated is a function of
soil critical state friction angle and pipeline diameter. The pipeline downwards force on the soil during shut-down is the other
parameter affecting the initiation of ratcheting. This paper presents valuable insight on ratcheting and mitigating upheaval
buckling of pipelines to the design engineers.

Introduction
High Pressure High Temperature (HPHT) pipelines operate under extreme operating conditions than the surrounding
ambient conditions, causing the pipeline to expand, while the backfill on top of the pipeline restraints its movements. These
pipelines are installed at ambient temperatures and out-of-straightness features are unavoidable under buried conditions. As
the pipeline tend to expand under high axial forces, any OOS features present in the pipeline bedding surface leads to upward
mobilizations, leading to upheaval buckling failures (Fig.1,3) unless the backfill material provides sufficient resistance
against the upward pipeline movement. If the pipeline upwards mobilization is beyond a critical limit, the gap underneath the
pipeline is filled up by soil particles from around the pipeline. Thus, when pipelines cool down during temporary suspension,
it cannot return to their original position. As pipeline undergo many thermal cycles of start-up and shut-down during the
lifetime, pipeline can progressively move upwards in steps during each of the stop/start cycles. This phenomenon is known as
pipeline ratcheting (Fig.2).

Upheaval
buckling (UHB) Buried pipe
Seabed level

Backfill soil

thermal expansion Seabed thermal expansion

Figure 1: Upheaval Buckling (UHB) of a buried pipeline

2 OTC-27823-MS

This incremental upward movement of the pipeline, if not mitigated, would eventually lead to the failure of the pipeline.
Thus, ratcheting is the fundamental driving mechanism behind most UHB pipeline failures cases (Fig.3). Current pipeline
guidelines such as the DNV-OS-F101 and DNV-RP-F110, while stating that ratcheting should be addressed in design, do not
offer a comprehensive criterion for design against ratcheting. Hence the conventional design wisdom is to avoid it completely
by limiting the mobilization, typically 20mm in sands. However, detailed understanding of the ratcheting is still not available.

Figure 2: Schematic cross section view - Pipeline ratcheting leading to pipeline incremental movement upwards

Literature Review
Compared with classical pipeline upheaval buckling concept, the understanding of pipeline ratcheting is in early phase and
available research in pipeline ratcheting is limited (Thusyanthan et al.2011). It is previously known, in offshore literature, as
“upheaval creep” due to its apparent time-dependent nature. Over the last few decades, the term “ratcheting” has gradually
replaced “creep”. This is rather inappropriate as “creep” refers to continued strain increments under constant stress, whereas
the hypothesis behind the ratcheting mechanism involves a cyclic process. Figure 3 presents a classical upheaval buckling
pipeline which clearly shows the soil infilling below the pipeline.
Nielsen, et al.(1990) presented a detailed case study of the first recognized subsea pipeline upheaval buckling failure: the
Mærsk Olie of Gas AS' inter-field pipelines in the Danish Sector of the North Sea in 1986. The paper reported a “time-
dependent upheaval behaviour” on one particular pipeline segment where the upheaval took place within 2 months during
which period the line was subjected to 4 additional shut-down situations, including 2 major ones. The paper also reported on
a 0.5 m imperfection growth prior to failure after 17 major shutdowns varying between 1 and 17 hours (average duration was
5 hours), representing an average imperfection-growth rate of 30 mm per temperature cycle. Based on these observations, the
paper suggested that the classical structural-side upheaval buckling analysis applied during the design phase was insufficient,
as it could neither model the detrimental effects of plastic deformation of the pipeline in combination with lack of
straightness, nor explain the geometric imperfection amplitudes growing with time. It was also in this paper that the term
“cyclic ratcheting” was first proposed.
Finch (1999) supported the work by Nielsen, et al. (1990) with experimental evidence. The paper reported one centrifuge
test at 20g with reconsolidated clay slurry as backfill. Results indicated that the initial stiffness during the reload (2nd pull-
out) curve was much the same as initial pull-out, whereas the final peak mobilization (δf) was higher, indicating potential
progressive failure if cycling continued. Finch, 1999 also reported full-scale cyclic ratcheting tests in loose saturated sand (e
= 0.91, γ’ = 8.5 kN/m3) with a PVC model pipeline (D = 220 mm) at H/D = 2.13. The paper argued that these results
illustrated that ratcheting was occurring, and that a large number of such cycles might result in eventual upheaval buckling
failure. However, a closer look at data of Finch (1999) suggested that, during the first 5 cycles where the positive maximum
load was reasonably constant, the incremental displacement after each cycle was not constant, and such few cycle numbers
could not tell whether the trend of incremental displacement versus cycle numbers was worsening or diminishing. If the
residual displacement per cycle continues to reduce, a stable situation may still eventually occur, preventing ratcheting from
doing any harm. Despite its limitations, (Finch, 1999) was still one of the very few full-scale experimental attempts on
pipeline ratcheting. As (Guijt, May 7-10, 1990) correctly stated, the lack of cyclic loading data in sand prevented further
understanding towards pipeline ratcheting. Thusyanthan et al.(2011), based on the limited available data on ratcheting and
angle of repose of typical sands (32°), proposed criteria for initiation of ratcheting based on mobilization as percentage of
pipeline diameter.
OTC-27823 -MS 3

Figure 3: Typical Upheaval buckling of a buried pipeline Figure 4: Forces on pipeline

Uplift Resistance Models

When the pipeline is starting to mobilize upwards, there are three vertical forces considered in UHB pipe design (Figure 4),
1. Effective Pipe weight (Pipe weight – Buoyancy)
2. Pipeline uplift force due to operating temperature and pressure
3. Uplift resistance of the soil R

It shall be noted that the effective pipe weight is constant and does not change with mobilization. Buoyancy force on
pipeline, whether from water or soil (slurry clay), is a constant and it is accounted for within the effective pipe weight and
hence buoyancy is not part of uplift resistance.

Present understanding on uplift resistance of buried pipelines is based on analysis (Randolph and Houlsby, 1984; Pedersen,
P.T. & Jensen, J.J. 1988,) and experimental work by numerous researchers (Vesic, 1971; Rowe and Davis, 1982; Hobbs,
1984; Randolph and Houlsby, 1984; Trautman et al., 1985; Palmer et al., 1990; Schaminée et al., 1990; Dickin, 1994; Croll,
1997; White et al., 2001; Bransby et al., 2001; and Cheuk et al, 2005; Finch, 1999; Finch et al.2000; Moradi & Craig, 1998,
Wang et al.2009, Thusyanthan et al. 2008, Thusyanthan et al.2010). Thus, there had been several analytical models proposed
for pipeline uplift resistance in sands. The main three models that are commonly used in the industry are summarized below.

Model 1

One of the early models to be used for prediction of peak uplift resistance,R , is given by Equation 1 (Schaminée et al., 1990).
R H
 1.0  f s   Equation 1
  H  D D

where the H is cover to top of pipe and fs is uplift factor (Schaminée).

Model 2

DNV-RP-F110 recommends the use of the following uplift model (Equation 2) to predict the peak uplift resistance, R, in
cohesionless soils.
2
R D  H  D 
 1.0  0.1   f p  1   Equation 2
  H  D H  D  2 H 

where H is cover to top of pipe and fp is uplift factor. It should be noted that DNV-RP-F110 recommended uplift model uses
weight and shear components of the soil regions just above the pipe surface but below top of pipe as shown in the Figure 5
(a)(K is horizontal earth pressure coefficient at rest). The model in Equation 1 (Schaminée et al., 1990) does not account for
weight and shear components below top of pipe. Hence, it should be noted that the uplift factor in Equation 1 and 2 are not
interchangeable between the models and should always be used with model from which it was calibrated.
4 OTC-27823-MS

Model 3

ASCE (1984) and ALA (2001) uplift resistance model is based on uplift factor Nqv and cover height from center of the
pipeline to surface, Hc, as provided below in Equation 3.

R    N qv H c D Equation 3

Uplift resistance R

Peak Uplift
Resistance

Uplift resistance
of soil when pipe
mobilisation is
limited

Upwards Displacement

Limit on mobilisation to Mobilisation Displacement

limit the pipeline stresses for Peak Uplift
(a) (b)
Figure 5: (a) Uplift resistance model (drained), (b) Soil uplift resistance vs pipe upward displacement

As the uplift factor for each model is associated with that model, it is important that if uplift model is changed in design then
the uplift factor value is also changed accordingly as per Equations 4 and 5.

2 2
D  H  D 
f s  0.1   f p  1   Equation 4
H  D  2 H 

π D H
Nqv = 1 − [ ] + fp [ c] Equation 5
8 Hc D

Uplift mobilization distance

In order to prevent upheaval buckling, the pipeline must be buried deep enough such that the soil cover is sufficient in
providing adequate uplift resistance. The required upward movement, or mobilisation, of the pipeline to achieve the desired
uplift resistance is a vital design parameter, in that pipeline integrity under operating conditions relies upon its value. An
important point to note is that the mobilisation of the pipe often needs to be limited to limit the stresses in the pipeline, thus
the available uplift resistance from the soil in not the peak uplift resistance but could be much lower as shown in Figure 5(b).
The DNV states (Pg 44 of DNV RP F110), “The uplift resistance Rmax is assumed to be fully mobilised at a vertical uplift
displacement f, where f is 0.005-0.01 times the height H. Note that f seems to be independent of the ratio of H/D”.
However, experimental full scale experiments (Thusyanthan et al. 2010) have demonstrated that the mobilisation is actually a
function of (H/D) and DNV guideline underestimates the mobilisation distance. The effect of this under-estimation of
mobilisation when combined with the use of tri-linear uplift resistance model, which is recommended by DNV, can lead to
unconservative UHB designs. This has been demonstrated in Thusyanthan et al.2010 using FE results. Based on the data
from literature and full scale testing, Thusyanthan et al. 2010 proposed a model (Eq. 6) to predict the peak mobilisation
distance (f) in loose sands in terms of H and D.

f 1 H 
2 D 
 0.02 e  
Equation 6
D
It should be noted that any effects of soil saturation (dry, moist & submerged) and soil density on mobilisation was not
distinguished in this equation. Robert and Thusyanthan (2014) has provided further insight and validity of this equation and
explained the effect of high dilatancy at low cover depths on peak mobilization.
OTC-27823 -MS 5

Experimental Results on Pipeline Ratcheting

A series of plane strain uplift testing was undertaken using two model pipes, 100mm and 258mm diameter pipes. A plane
strain chamber with internal dimensions 1000mm width, 750mm height and 75mm thickness (L) was used in the
experiments. The chamber’s front face comprises of a 25mm thick perspex sheet that enables usage of the particle image
velocimetry (PIV) technique so that the displacement field of the backfill can be measured throughout the pullout event. The
pipeline uplift was displacement controlled at a rate of 0.005 mm/s. A pore pressure transducer (PPT) was embedded in the
bottom of the model pipes. During all tests, very small negative excess pore pressure was developed, thus confirming that
drained conditions were achieved. Experiments were undertaken in saturated condition with Fraction E sand, which is a well-
graded (with D50 of 0.18mm), sub angular laboratory sand commonly used for geotechnical physical modeling tests. Its
material properties and particle size distribution resemble soil samples from many near-shore seabed sites. The sand was
prepared at a relative density I D of 35% giving a saturated unit weight γsat of 18.5 kN/m3. Two types of tests are reported here,
A. Monotonic uplift tests in loose, fully saturated Fraction E sand using a 100mm and 258mm diameter model
pipes; both the model pipes were 75mm in length (L).
B. Cyclic uplift test in loose, fully saturated Fraction E sand using a 100 mm diameter model pipe;

A. Monotonic uplift loading

Figure 6(a) and (b) illustrate how a gap is formed beneath the pipe as the pipe move upwards. When the two sides of infilling
sands slope intersect, the angle of the slope is angle of repose of the sand. Based on the measured angle of reposition, it is
possible to derive an upper bound on δ where upward movement of the pipe becomes definitely “unrecoverable”, i.e. the pipe
cannot return to its original burial depth without exerting significant bearing pressure on the trench. This “upper bound” may
be defined by the moment when the two miniature slope failures meet at the centre of the void, so that any additional pipe
upward movement will lead to elevation of the lowest point of the triangular trench formed underneath. From Figure6(c) and
(d), the upper bound limit can be theoretically derived as:
1
𝛿𝑈𝐵 = 𝑅([1/𝑐𝑜𝑠 𝜙𝑐𝑟𝑖𝑡 ] − 1) = 2 𝐷(𝑠𝑒𝑐 𝜙𝑐𝑟𝑖𝑡 − 1) Equation 7

ϕcrit = 32.5°

ϕcrit = 32.5°

(a) (b)
Figure 6: (a) pipe (D= 100mm) uplift at 10mm mobilization, (b) pipe (D=258mm) uplift at 22mm mobilization

Soil
Pipe

ϕcrit = 32.5° R ϕcrit = 32.5° D/2

δ = R (sec ϕcrit -1)

Upwards movement  Angle of repose a

2 × ϕcrit
(a) (b)
Figure 7: (a) Angle of repose below pipeline, (b) Pipeline upwards movement and angle of repose
6 OTC-27823-MS

For the Fraction E sand used in the current research, δUB is approximately 9% D given a measured ϕcrit value of 32° which is
typical for most marine sands. This means a δUB value of 9 mm for the 100 mm-diameter model pipe or 23 mm for the 258
mm-diameter model pipe. Compared with the current industry practice of limiting δ arbitrarily to 20 ~ 30 mm in order to
prevent ratcheting, δUB offers a much more scientific approach.

Most onshore and offshore sands have a ϕcrit range of between 32° and 38°, leading to a δUB range of between 9%D and
13.5%D. In addition, δUB calculated depends purely on the pipe geometry and independent of the uplift resistance vs
displacement. This poses an additional design consideration: if the mobilization distance required to achieve the design uplift
resistance is greater than δUB, pipeline ratcheting will almost certainly occur due to irreversible accumulation of soil particles
underneath the pipe. Hence, δUB has to be an independent design limiting criterion if cyclic ratcheting is of concern.

B. Cyclic uplift loading

The same testing apparatus used for the monotonic uplift testing was used in the cyclic tests with 100mm pipe and 350mm
soil cover. The ratcheting test must enable load-controlled movement of the actuator driving the model pipe. This was
achieved by passing the unfiltered signals from the load cell directly to the A3200 servo drive that was accessible by the
NViewHMI software on the controlling computer. Load-controlled cyclic movement was hence possible by programming the
actuator to change direction whenever the force reading from the load cell reached the prescribed Rmob value or zero for a
specific number of cycles before increasing Rmob to the next threshold. The actuator speed was kept constant at 0.005
mm/sec. An additional displacement-measuring laser (supplied by Baumer Group) was mounted to the connecting rod of the
model pipe. With the deformation of the connecting rod negligible for the force range considered, the laser enabled accurate
measurement of the pipe vertical displacement during the cyclic ratcheting action.
The model preparation process and test procedure for the cyclic testing were almost the same as for monotonic testing,
except that load-controlled cyclic loading rather than displacement-controlled monotonic loading was implemented.
A total of 6 different Rmob/L values were selected: 850 N/m, 970 N/m, 1120 N/m, 1270 N/m, 1380 N/m, and 1550 N/m,
corresponding to 54%, 62%, 71%, 80%, 88%, and 98% of the monotonic Rpeak/L value (Fig. 8 and Table 1). The model pipe
was cyclically loaded between zero and each Rmob/L value for 10 cycles before increasing the Rmob/L threshold. In order to
allow ratcheting displacement to fully develop, the model pipe was also programmed to undergo a miniature cyclic loading
process, simulating a quasi-constant-uplift-force condition on the model pipe, during each cycle whenever the particular
Rmob/L is reached. This miniature cyclic loading process comprised 5 additional cycles for each ratcheting cycle with peak-
to-peak range of 50 N/m. After all 60 ratcheting cycles completed, the model pipe was then pulled monotonically for an
additional short distance until the new Rpeak was clearly identifiable.
Series 5

Series 6
Series 4
Series 3
Series 2
Series 1

Figure 8: Cyclic uplift loading of model pipe of diameter 100mm at H/D of 3.5.
OTC-27823 -MS 7

Table 1 Summary cyclic ratcheting test data

Rmob / δcycle Rmob / δcycle Rmob / δcycle
Cycle Cycle Cycle
Rpeak (mm) Rpeak (mm) Rpeak (mm)
1 0.939 21 0.727 41 0.900
2 0.462 22 0.729 42 0.939
3 0.432 23 0.648 43 0.897
4 0.365 24 0.585 44 0.875
54% 71% 88%
5 0.323 25 0.574 45 0.821
Series 1 Series 3 Series 5
6 0.312 26 0.555 46 0.800
fp=0.4 fp=0.59 fp=0.76
7 0.307 27 0.600 47 0.767
8 0.292 28 0.523 48 0.742
9 0.281 29 0.456 49 0.741
10 0.257 30 0.453 50 0.715
11 0.452 31 0.874 51 2.96
12 0.465 32 0.908 52 2.83
13 0.442 33 0.807 53 2.74
14 0.411 34 0.759 54 2.28
62% 80% 98%
15 0.391 35 0.721 55 2.12
Series 2 Series 4 Series 6
16 0.373 36 0.708 56 1.99
fp=0.49 fp=0.69 fp=0.91
17 0.374 37 0.810 57 1.85
18 0.37 38 0.664 58 1.77
19 0.348 39 0.580 59 1.78
20 0.325 40 0.572 60 1.67

(a) (b)
Figure 9: (a) Accumulative upwards displacement vs cyclic loading, (b) upwards movement vs cyclic loading

It is evident from Figure 9(a) that the upward movement per cyclic loading is mostly below 1mm for all the cycles except the
last set of cyclic loadings (series 6). Hence, the rate of upward movement is greatest for this set of loadings. If the pipeline
downward movement during shutdown is up to 1mm, then there will not be ratcheting in any of the cyclic loadings steps
except in the last set of cyclic loading. It shall also be noted that during the last cyclic loading, the pipe’s accumulative
upward movement exceeded 9mm (9%D) at which point the infilling from both sides of the pipe meets. This experimental
results confirm the fact that ratcheting is to be expected once the pipe upward movement has exceeded 9%D. Figure 9(b)
shows the accumulative upwards movement in each of the series.
8 OTC-27823-MS

Uplift Ratcheting Model and Criteria for Initiation of Ratcheting in Cohesionless Soils
The out-of-straightness (OOS) features in a buried HPHT pipeline which has mobilized upwards during operating conditions
will tend to move downwards during shutdowns as the pipelines tend to shorten axially. In the absence of axial pipe-soil
friction and soil migration underneath the pipe, the pipeline should return to its original as-laid position, creating an average
downward bearing pressure on the soil equivalent to the submerged weight of the pipe (with contents) divided by the contact
area. In reality, however, both axial friction and infilled soil in the gap beneath the pipeline may restrict the pipeline and
prevent it from returning pre-operation profile. It is difficult to evaluate the exact bearing pressure of the pipeline on the base
soil when it has completely cooled down. Therefore, it is reasonable to assume that net soil resistance always returns to zero
during cyclic ratcheting as shown in Figure 10(a). However, when the pipeline cools down during a shut-down, due to
pipeline contraction there could be a downwards force more than the submerged weight of the pipe (with contents) per unit
length. This means that the pipe could push down into the soil during every cycle. This means that the ratcheting would only
occur if the pipeline is unable to push down into the soil to eliminate the infilling effect of the soil. This model is
demonstrated in Figure 10(b). Thus, proposed ratcheting framework in Figure 10 (a) and (b) are fundamental for ratcheting
mitigation designs.

(a) (b)
Figure 10: (a) Pipeline ratcheting with zero downwards force to soil during shut-down. (b) Pipeline ratcheting with
downwards force to soil during shut-down.

Theoretically, the effect of cyclic ratcheting in loose sandy backfill conditions on the available uplift soil resistance is two-
fold: cumulative δ gradually reduces the available soil cover height and hence reduces the Rpeak, whereas the cyclic
movement of the pipe compacts the backfill soil, leading to a stiffer response and higher R peak. Strictly speaking, in order to
identify the operational limit on the allowable mobilized uplift resistance, Rmob expressed as a percentage of monotonic Rpeak,
where cyclic ratcheting between 0 and Rmob will not lead to a diverging cumulative δ, independent experiments must be
conducted using identical physical models but varying R mob values only. However, this would lead to a large number of tests
for any prototype geometry, and would require almost perfect repeatability amongst all the physical models. Given the time
and financial constraints on this research component, limited experiments performed have been reported in this paper. Further
ongoing research will be published in a future paper.
The literature on pipeline ratcheting is very limited. Upheaval failure attributed to ratcheting was reported by Nielsen et al.
(1990). Finch (1999) provided some experimental data of ratcheting behaviour from full scale testing of a 0.22m diameter
pipe buried to a depth of 0.58m with fine sand (submerged unit weight 8.8kN/m 3). Experimental uplift results from dry sands
(fine and coarse sand) were presented by Cheuk et al. (2008) for a pipe diameter of 0.1m and cover of 0.25m. The test results
of Cheuk et al. (2008) showed that a cavity with sloping sides forms beneath the pipe during the mobilization of peak
resistance. For fine sands, particles stated to infill this cavity after 3mm (~10D50) of pipe movement, leading to unrecoverable
upward pipe displacements. For the coarse sand, this infilling mechanism occurred at a pipe displacement of 9 mm (~4D50).
The onset of ratcheting is related to the stage when soil particles start to fill the void beneath the pipe. This onset depends
on soil particle size, friction angle of the soil, dry/partially saturated or submerged soil and pipeline diameter. For clay soils,
shear strength and time duration of the uplift cycle will also influence the onset of infilling. It should be noted that the onset
OTC-27823 -MS 9

of infilling below the pipe is one of the key requirements for ratcheting but it does not on its own lead to ratcheting, i.e. the
pipeline downward force during shutdown, if sufficient enough, could move the infill soil and revert to original pipeline
profile.
Infilling of soil is fully confirmed (in uncemented sands and drained conditions in clay) when both sides of the infill have
met at the centre when the pipe has moved upwards such that it is above the angle of repose lines as shown in Figure 6. Thus,
based on the geometry shown in Figure 6 & 7 and assuming angle of repose (a) to be 32 (critical friction angle of sand), we
can calculate the mobilisation () which corresponds to the stage at which the infilling from both side of the pipe meet at the
centre. This value is 9%D. If we assume that when both sides of the infilling meeting at the centre is a clear criterion for
ratcheting, then we can expect ratcheting to occur for mobilisation greater than 9%D in sands with angle of repose of 32
(note that it will be 13.5%D if the angle of repose is 38).
Initiation of the infilling of soil can occur much earlier than the 9%D mobilisation, but this may not lead to ratcheting and
currently there is limited data to verify this. The results from Cheuk et al. (2008) on initiation of infilling in fine sand suggests
that 3%D may be the lower limit for the onset of infilling and hence ratcheting.
Based on the limited data available on ratcheting and the presumption that ratcheting is certain to occur at 9%D
mobilisation for sand with critical state friction angle of 32o, we could propose three different regimes for ratcheting in
relation to mobilisation and pipeline diameter as shown in Figure 11.
1. Mobilisation of more than 9%D - Zone A- Ratcheting
2. Mobilisation in-between 3%D to 9%D - Zone B- Ratcheting possible
3. Mobilisation less than 3%D - Zone C- Ratcheting not likely

However, even in Zone A and B, ratcheting does not occur if the pipeline downward force during shutdown is sufficient
enough to move the infill soil and revert to original pipeline profile. Thus, the amount of possible downwards movement of
pipe during shut-down is a critical value. If the pipe mobilisation is below this value, the ratcheting would not occur. This
line is named “Pipeline downwards Line (PDL)” and it is shown in Figure 11 with a typical value of 20mm (this value is not
fixed). This value is dependent on pipeline properties, operating conditions and OOS feature. Ratcheting does not occur if the
mobilisation is below PDL. Thus, the fourth region where ratcheting cannot occur is region below PDL.

4. Region below PDL line - Ratcheting is not possible.

Figure 11: Proposed regimes for ratcheting based on pipeline diameter and upward mobilization of pipe
10 OTC-27823-MS

Conclusion
Pipelines undergo thermal cycles during start-up and shut-down stages. As the pipeline temperature and pressure increases,
the axial force in the pipeline increases and pipeline OOS features mobilise upwards till equilibrium is reached with the uplift
resistance from the cover soil. When oil/gas supply is shutdown, the pipeline cools down and the pipeline tries reverting to its
original shape. This downwards movement of the pipeline feature is resisted by the downward resistance of soil if the void
below the pipe has already infilled with soil particles. If mobilisation (upward movement) of the pipe during operational
stage is not the same as downward movement during the shutdown stage, the pipe can experience incremental upward
movement which can ultimately lead to upheaval buckling failure. This phenomenon is named pipe ratcheting.

This paper has provided insight into the ratcheting fundamentals and has proposed ratcheting frame work (Figure 10(a) and
(b)). It can be concluded that the onset of ratcheting depends on “upwards mobilisation”, “pipeline diameter”, “particle size
(D50)”, “critical state friction angle of soil” and “pipeline downward force during shutdown”. The design limit on upward
mobilisation to avoid ratcheting (and hence to avoid UHB failures) can be obtained using chart provided in Figure 11.
Below are the four key regions for ratcheting in Figure 11.

1. Mobilisation of more than 9%D - Zone A- Ratcheting

2. Mobilisation in between 3%D to 9%D - Zone B- Ratcheting possible
3. Mobilisation less than 3%D - Zone C- Ratcheting not likely
4. Region below PDL line - Ratcheting is not possible.

Further research on ratcheting in ongoing and future publications will provide further insight into ratcheting.

References
ASCE-Guidelines (1984). “Guidelines for the seismic design of oil and gas pipeline systems”, committee on gas and liquid fuel
lifelines of the ASCE technical council on lifeline earthquake engineering. ISBN 0-87262-428-5.
American Lifelines Alliance (ALA) (2001), “Guidelines for the Design of Buried Pipelines”
Bransby, M.F., Newson, T.A., Brunning, P., and Davies, M.C.R. (2001). Numerical and centrifuge modelling of upheaval
resistance of buried pipelines. Proc. OMAE, Rio de Janeiro, June 2001.
Cheuk C. Y., White D. J. and Bolton M. D. (2008), Uplift Mechanisms of Pipes Buried in Sand, Journal of Geotechnical and
Geoenvironmental Engineering, Vol. 134(2),154-163.
Cheuk, C.Y. (2005). Soil pipeline interaction at the seabed. PhD thesis. University of Cambridge.
Croll, J.G.A. (1997). A simplified model of upheaval thermal buckling of subsea pipelines. Thin-Walled Structures 29 (1-4): 59-
78.
Dickin, E.A. (1994). Uplift resistance of buried pipelines in sand. Soils and Foundations 34 (2): 41-48.
DNV-RP-F110, Global buckling of submarine pipelines – structural design due to high temperature / high pressure. Det Norske
Veritas, Norway, 2007.
Finch, M., (1999), Upheaval Buckling and Floatation of Rigid Pipelines: The Influence of Recent Offshore Geotechnical
Research on the Current State of the Art. OTC 10713
Finch, M., Palmer. A and Baumgard, A. (2000), An Integrated approach to pipeline burial in the 21st Century. Deep Offshore
Technology 2000.
Hobbs, R. (1984). In service buckling of heated pipelines. ASCE Journal of Transportation Engineering 110 (2): 175-189.
Moradi, M. and Craig, W.H. (1998). Observations of upheaval buckling of buried pipelines. Centrifuge 98, Kimura, Kusakaba &
Takemura (eds), ISBN 90 5410 986 6
Nielsen, N-J.R., Lyngberg, B. and Pedersen, P.T. (1990), “Upheaval Buckling Failures of Insulated Buried, Pipelines : A Case
Story”, paper OTC 6488 presented at the 1990 Offshore Technology Conference, Houston, pp 581-592.
Palmer, A.C., Ellinas C.P., Richards, D.M., and Guijt, J. (1990). Design of submarine pipelines against upheaval buckling. Proc.
Offshore Technology Conf., Houston, OTC 6335: 551-560.
Pedersen, P.T. & Jensen, J.J. (1988). Upheaval creep of buried pipelines with initial imperfections. Marine Structures 1:11-
22, 1988.
Randolph, M. F., & Houlsby, G. T. (1984). The limiting pressure on a circular pile loaded laterally in cohesive soil.
Géotechnique, 34 (4): 613-623.
Rowe, R.K., and Davis, E.A. (1982). The behaviour of anchor plates in sand. Géotechnique 32 (1): 25-41.
Robert, D. J. and Thusyanthan, N. I., (2014), Numerical and Experimental Study of Uplift Mobilization of Buried Pipelines
in Sands’, vol 6, ASCE Journal of Pipeline Systems Engineering and Practice.
Schaminée, P.E.L., Zorn, N.F., and Schotman, G.J.M. (1990). Soil response for pipeline upheaval buckling analysis: Full-
scale laboratory tests and modelling. Offshore Technology Conference, Houston, OTC 6486
Thusyanthan, N. I., Ganesan S. A & Bolton M.D. and Peter Allan (2008). Upheaval buckling resistance of pipelines buried in
clayey backfill. Proceeding of ISOPE 2008, The Eighteenth (2008) International Offshore and Polar Engineering
Conference, Vancouver, Canada, July 6 - 11, 2008
 OTC-27823 -MS 11

Thusyanthan, N.I., Mesmar, S., Wang J., and Haigh, S.K. (2010). Uplift resistance of buried pipelines and DNV-RP-F110
guideline. Proc. Offshore Pipeline and Technology Conference. Feb 24-25, Amsterdam, Netherlands.
Thusyanthan, N.I, (2012), "Seabed Soil Classification, Soil behaviour and Pipeline design", published in OTC2012, USA
Thusyanthan, N.I, Sultan M, Robert D.J, Wang.J. & S.K, Haigh (2011), “Upheaval Buckling Assessment Based on Pipeline
Features”, published in OTC2011, USA.
Trautman, C.H., O'Rourke, T.D., and Kulhawy, F.H. (1985). Uplift force-displacement response of buried pipe. ASCE
Journal of Geotechnical Eng. Division 111 (9): 1061-1075.
Vesic, A.S. (1971). Breakout resistance of objects embedded in ocean bottom. ASCE Journal of the Soil Mechanics and
Foundation Division.97 (9): 1183-1205.
Wang, J., Haigh, S.K., and Thusyanthan, N.I. (2009). Uplift resistance of buried pipelines in blocky clay backfill. Proc.
International Offshore (Ocean) and Polar Engineering Conference. ISOPE 2009 TPC 564.
Wang, J., Haigh, S.K., and Thusyanthan, N.I. (2010). Uplift resistance design of shallowly buried pipelines. ASCE Pipelines
Journal (currently being drafted).
Wang, J., Haigh, S.K., Thusyanthan, N.I., and Mesmar, S. (2010). Mobilisation distance in uplift resistance modeling of
pipelines. Proc. International Soposium on Frontiers in Offshore Geotechnics. Nov. 8-10, 2010, Perth, Australia.
White, D.J., Barefoot, A.J., Bolton, M.D. (2001). Centrifuge modelling of upheaval buckling in sand. International Journal of
Physical Modelling in Geotechnics, 2 (1):19-28.

View publication stats