Blackpool South Strategy project: analysis of pipe-jacking records

Abstract

In recent years, there has been an increased resort to microtunnelling/pipe-jacking as a means of constructing underground conduits (for water, sewage, gas and other utilities) to avoid on-street disruption in urban areas. In this paper, technical details of two 1200 mm internal diameter microtunnels in silty sand totalling 550 m in length are discussed; the microtunnels were constructed by Ward and Burke Construction Ltd. as part of the Blackpool South Strategy project. A general overview of the tunnelling process is provided, including the separation plant, jacking facilities and the bentonite supply process. The results show that the lubrication system was very effective at maintaining low skin friction, and that the pipe string was almost fully buoyant for the majority of the drive. Stoppages were shown to have a significant but transient effect on the jacking force; high jacking forces upon resumption of jacking after a stoppage return to ‘baseline’ levels after the length of one pipe diameter. Machine deviations did not appear to play a major role in increasing jacking forces for this particular project.

Full text

ABSTRACT: In recent years, there has been an increased resort to microtunnelling/pipe-jacking as a means of constructing
underground conduits (for water, sewage, gas and other utilities) to avoid on-street disruption in urban areas. In this paper,
technical details of two 1200 mm internal diameter microtunnels in silty sand totalling 550 m in length are discussed; the
microtunnels were constructed by Ward and Burke Construction Ltd. as part of the Blackpool South Strategy project. A general
overview of the tunnelling process is provided, including the separation plant, jacking facilities and the bentonite supply
process. The results show that the lubrication system was very effective at maintaining low skin friction levels, and that the pipe
string was almost fully buoyant for the majority of the drive. Stoppages were shown to have a significant but transient effect on
the jacking force; high jacking forces upon resumption of jacking after a stoppage return to ‘baseline’ levels after the length of
one pipe diameter. Machine deviations did not appear to play a major role in increasing jacking forces for this particular project.
Keywords: Microtunnelling; Jacking force; Skin friction; Lubrication; Bentonite; Stoppages; Deviations.
1 INTRODUCTION
The rapid expansion of urban areas worldwide has resulted in
a need to provide new and/or upgrade existing water, sewage,
gas and other utility conveyance networks. Pipe-jacking has
emerged as the preferred method of utility pipeline
construction, as it avoids the on-street disruption arising from
trenches constructed from the ground surface. However, the
difficulty in identifying suitable intermediate shaft locations in
urban projects means that long drives are often necessary;
keeping jacking forces at manageable levels is a challenge in
these drives. For example, excessive stress concentrations can
give rise to spalling at the joints between pipes, potentially
inducing pipe failure [1]. Intermediate jacking stations are
cumbersome, however, and are generally kept to a minimum.
The total jacking load (Ftotal) consists of the resistance at the
face (Fface) of the tunnel boring machine (TBM) and the
frictional resistance (Ffriction) between the pipe train and the
surrounding ground. The frictional force is often the main
contribution to the jacking load, especially in long drives [2].
The introduction of a lubricant into the overcut (the annulus
formed on account of the TBM having a larger diameter than
the pipes) is an efficient means of reducing the jacking force.
The skin friction (τ, force per unit surface area of pipe)
depends on the effective normal stress
!"#
$%&
from the soil on
the pipe, the total effective weight of the pipe string (
'())
)
and the angle of effective interface shearing resistance
between the pipe and the soil, δ:
* +
,
"#
$-./00
123
4
5678
(1)
where D is the pipe diameter and L is the embedded pipe
string length. Lubrication has the effect of lowering δ.
Additionally, if the pipe is buoyant in the lubricant,
'())&
will
be lower than the weight of the pipe
'
, and may be as little as
zero in a fully buoyant condition. A number of authors [1, 3-
6] attest to the benefits of a properly-lubricated overcut. For
example, τ values ranging between 0.1 kPa in well-lubricated
drives to 4 kPa in moderately-lubricated drives were identified
in four different drives in clay and gravel deposits [7].
Stoppages and deviations in steering also influence the
jacking forces along a tunnel string [8, 9]. A study of
microtunnels in glacial till in Ireland [10] found that stoppages
in sands/gravels required a higher frictional force to be
overcome upon recommencement of jacking than in clays.
Furthermore, the stoppage duration had an effect on this
jacking force in clay but not in gravel. Long stoppages in soft
ground may allow the TBM to settle, resulting in deviations
from the intended path. Deviations, irrespective of the cause,
can increase the required jacking force. An analysis of data
from microtunnels in alluvium and glacial till in Ireland [11]
found that for two drives of similar length and ground
conditions, the drive with the greater deviations overall
required much greater jacking forces.
This paper provides an overview of some tunnelling aspects
of the Blackpool South Strategy project, U.K. Following a
brief description of the tunnelling process, data recorded
(jacking force, deviations, bentonite injections) for two 1200
mm internal diameter tunnel drives are presented and
analysed.
2 THE BLACKPOOL SOUTH STRATEGY PROJECT
The project site is located in Blackpool, U.K. The over-
arching purpose of the project is to improve bathing water
quality along the seafront and mitigate the risk of flooding.
This necessitated (i) an increase in the capacity of the sewer
network, (ii) a reduction in the volume of surface water
entering the network and (iii) an upgrade to the wastewater
pumping station situated at Lennox Gate. The latter element
Blackpool South Strategy project: analysis of pipe-jacking records
Kevin G. O’Dwyer1, Bryan A. McCabe1, Brian B. Sheil2, David P. Hernon3
1Civil Engineering, National University of Ireland, Galway, University Road, Galway, Ireland
2Department of Engineering Science, University of Oxford, United Kingdom
3Ward and Burke Construction Limited, Unit N, Bourne End Business Park, Cores End Road, Bourne End, United Kingdom
email: k.odwyer2@nuigalway.ie, bryan.mccabe@nuigalway.ie, brian.sheil@eng.ox.ac.uk
Cite as:
O’Dwyer, K.G., McCabe, B.A., Sheil, B.B. and Hernon, D.P. (2018) Blackpool South Strategy Project: analysis of pipe-
jacking records, Proceedings of Civil Engineering Research in Ireland (CERI 2018), pp. 265-270.

included the provision of a new storm-water holding tank to
hold excess storm water until pumping back into the sewer
network is possible. New pipework will connect this tank to
the existing pumping station. A new surface water pumping
station will pump surface water straight to sea.
The ground conditions comprise a layer of peat (1 − 2 m
thick), overlying medium dense sand (2.6 − 4.4 m thick), in
turn overlying silty sand. The water table was at a depth of 1.2
m in a borehole close to the reception shaft of Drive A.
Figure 1. Location of tunnels and shafts at Blackpool site
(adapted from Google Maps).
The locations of the two drives, A and B, referred to in this
paper, are shown in Figure 1. These provide a sleeve for 700
mm internal diameter ductile water pipes which will transport
excess stormwater from the holding tank to the outfall pipe at
Harrowside, which will then feed the water out to sea.
3 TUNNELLING PROCESS
TBM, pipe and general details
A schematic of the tunnelling process, including TBM,
jacking frame, separation plant, slurry feed and return lines
and control unit, is provided in Figure 2.
Figure 2. Schematic of the microtunnelling process
(courtesy of Herrenknecht).
The TBM used at Blackpool was a Herrenknecht AVN
1200 (slurry shield), with a cutterhead diameter of 1515 mm
and a machine lining outer diameter of 1505 mm. Each
concrete pipe was 2.5 m long with an outer diameter of 1490
mm, providing an overcut 25 mm thick into which lubrication
may be pumped.
A laser positioned on the back wall of the launch shaft and
aimed at a target at the rear of the machine helped the operator
to direct the TBM. In general, ground conditions dictate how
far the machine deviates from the intended line and what
steering interventions are needed. For instance, the machine
can change direction more easily in sandstone bedrock or
cobble formation than in sand or clay deposits [12].
Drive A was 272 m in length, constructed in an east north-
easterly direction with a gradient of -0.154 % and an initial
launch invert depth of 7.61 m. Drive B was 295 m in length,
constructed in a west south-westerly direction with a gradient
of 0.347 % and an initial launch invert depth of 7.59 m (see
Figure 1).
Jacking frame and intermediate jacking station
The jacking frame consisted of four hydraulic cylinders that
push the machine and the concrete pipes through the ground.
The hydraulic cylinders have a total stroke length of 3.52 m.
The operator controls the speed at which the hydraulic
cylinders advance the tunnel through the ground; this is
dependent on torque, jacking forces, ground conditions and
the steering of the machine. Figure 3 displays the main setup
for the pipe jacking process with a concrete pipe in place for
the recommencement of jacking.
Figure 3. Concrete pipe installed for jacking.
In the event that the jacking forces become excessive,
recourse is made to intermediate jacking stations (interjacks),
pre-installed partway through the drive. The interjack reacts
off the pipes towards the launch shaft to advance the pipe train
on the side of the reception shaft. The pipe string is therefore
advanced in an ‘inchworm’ manner [13]. Each interjack
consists of 10 hydraulic rams which are placed inside the
tunnel at 100 m intervals. In this project, an interjack was
placed in both tunnels but was not used as the total jacking
force remained sufficiently low.
Separation plant
As the TBM advances, the revolving cutterhead excavates the
soil material. During this process, water is pumped at high
velocity to the head of the machine, where the soil material
and water mix to form a slurry. This slurry is then pumped to
the separation plant above ground, where the solid material is
recovered from the slurry, before recirculation to the face of
the machine in a closed system.
The separation plant consists of a primary shaker, a
secondary shaker and a centrifuge. The slurry returning from
the face of the machine passes through the primary shaker
initially. The primary shaker comprises coarse screens that

only permit material finer than 4 mm to pass. The secondary
shaker removes fine particles (greater than 20 µm). Finally,
the mixture is pumped into a centrifuge which spins between
700 rpm and 2000 rpm. A flocculant is added to bind the fine
silt particles together thereby aiding their removal; particle
sizes greater than 1 µm are removed at this stage. The solid
material that emerges from the shaker screens and centrifuge
is subsequently dried by adding lime and removed from site.
Lubrication
Lubricant is pumped into the overcut to maintain tunnel
stability and to reduce friction at the pipe-soil interface. For
the Blackpool project, the first station was located in the pipe
directly behind the TBM, with a further 19 stations positioned
in ensuing pipes (one every fifth pipe) and one on the launch
shaft wall. Each station comprised three lubrication ports
(separated by 120°) situated at the midpoint of the pipe. A
lubrication station arrangement is shown in Figure 4.
Figure 4. Lubrication station.
The bentonite lubrication system is volume-controlled; the
volumes required for each station are calculated from the
TBM advance rate and ground conditions [14]. The bentonite
solution comprised Hydraul-EZ and water in the ratio 22.7kg
to 400 l. Other additives included (i) soda ash to balance pH,
(ii) MX polymer to prevent additional groundwater
penetrating the mix, and (iii) torque reducer to promote
lubrication and to reduce the potential for the pipeline to
become jammed due to soil pressures exceeding that of the
bentonite lubrication acting on the pipeline. When tunnelling
in fine sands and silts, more lubricant is used than is required
to fill the overcut. The extra lubricant seeps into the ground
creating a filter cake that serves as a membrane or zone of low
permeability to transfer the support pressure acting in the
annular gap into the grain structure of the ground [15]. The
typical volume administered in these ground conditions is 2.5
times the overcut volume [15], based on experience of
monitoring on numerous pipe-jacking projects. The formation
of a filter cake in sands and gravels requires more bentonite
than in clays, due to differences in permeability.
4 MONITORING
Overview
The output data from the TBM was recorded at 200 mm
intervals of jacked distance. Output provided by the TBM
included jacking force, steering deviations, water circulation,
feed and slurry line pump details, advance speed, cutting
wheel revolution, slurry pressure in excavation, interjack
cylinder forces and bentonite injection volumes and pressures.
The results presented in the following sections relate to Drive
A only (with the exception of Figure 9), as findings are
generally consistent between the two drives. Due to a
technical issue with the data acquisition system, data were
only recorded for Drive B beyond a jacked distance of 62 m.
Jacking force
As already mentioned, minimizing jacking forces is an
important consideration during pipe jacking. From Figure 5, it
is clear that the jacking force remained relatively constant
throughout this drive, with an average of ~380 kN. Towards
the end of the drive, the jacking force rose to over 1000 kN, as
the TBM approached the concrete wall at the reception shaft.
Figure 5. Development of total jacking force during Drive A.
Two separate methods of calculating Ffriction are compared:
a) Method A: The face pressure was calculated over the
first 3 m of the drive (the length of the TBM) on the
assumption that Ffriction was negligible over that length
[16]. This face pressure was assumed constant for the
entire drive, enabling Ffriction to be inferred.
b) Method B: Based on the work of Pellet-Beaucour and
Kastner [8], Ffriction can be approximated from a trendline
joining the minimum points on the total jacking load
envelope, while Fface is taken as the difference between
the minimum and maximum envelopes (data plotted at
10 m intervals in Figure 6).
Figure 6. Maximum and minimum jacking force envelopes
and face resistance for drive A

Skin friction is calculated by dividing Ffriction by the
developed surface area of all embedded pipes. Skin friction
values calculated using these two methods are plotted against
jacked distance in Figure 7. Using Method A, it takes ~35 m
for the skin friction calculated to drop below 1 kPa, while it
takes ~70 m for the skin friction to drop below 1 kPa using
Method B. Average values beyond 100m are 0.27 kPa and
0.48 kPa for Methods A and Method B respectively.
It is interesting to note that if the slurry pressure recorded by
the TBM is assumed to be numerically equal to the face
resistance, the inferred skin friction is almost identical to that
derived using Method B (Figure 7). Similar observations were
made for Drive B. In practice, the slurry pressure is chosen to
be slightly higher than hydrostatic ground water pressure. The
match between slurry pressure and face pressure in this drive
is perhaps fortuitous and/or specific to the silty sand, but
suggests that the water pressure contributes significantly more
than the active earth pressure to the face pressure. Had the
machine been in clay, this slurry pressure would be lower than
the face pressure, as the clay material at the face would not
need the same level of support as the silty sand.
Figure 7. Methods of evaluating skin friction for drive A.
The constant face pressure method (Method A) is not a robust
approach as the face resistance is likely to change throughout
the drive. Fface calculated using Method B (also plotted on
Figure 6), shows great variation with jacked distance, with an
average value of 70 kN when the data are plotted at 1m
intervals. This suggests that the value of Fface adopted (200
kN) at the start of the drive was too high, possibly due to
careful driving style of the operator soon after launch.
Frictional resistances have been reported for sand of 2.8
kPa to 4 kPa without the use of lubrication [8] and 0.5 kPa –
2.5 kPa with lubrication [17]; the measured values reported
here are at the lower end of these ranges, suggesting effective
lubrication practice, which is explored further in Sections 4.3
and 4.4.
Lubrication
The volume percentage of bentonite pumped into the annulus
is plotted (on a log scale) against jacked distance in Figure 8
for a selection of the 21 stations noted in Section 3.4. The
distance of these stations from the TBM face is shown in the
legend (the position of Station 25 was fixed at the launch
shaft).
Station 1 (immediately behind the TBM) is responsible for
greater bentonite volume than any other individual station for
most of the drive. The trailing stations pumped smaller
volumes as their purpose was merely to maintain bentonite
levels. For example, at the midpoint of the tunnel (135 m),
Station 1 had produced 44% of the total volume of bentonite,
Station 25 had contributed 13% and Stations 2 and 3 supplied
10.9% and 6.4% respectively. Therefore, these four stations
contributed 74.3% of the total volume of bentonite at this
position.
Figure 8. Percentage of total bentonite volume pumped from selected
stations.
The development of bentonite volume normalised by the
volume of the overcut during the drive is considered in Figure
9. It can be seen that the actual normalised bentonite volume
for Drive A (~6.5 for most of the drive length) far exceeds the
target of 2.5 recommended for filter cake formation [15]. The
corresponding normalized volume for Drive B is lower at ~4.5
although the average skin friction values are virtually the
same for both drives. Further research is required to assess
whether there is a minimum or threshold normalized bentonite
volume which enables minimum friction values to be
achieved.
Figure 9. Normalised bentonite volume against jacked distance for
both Drives A and B.

Pipe buoyancy
It is highly desirable for a train of pipes to be buoyant within
its overcut, to help minimize the
'())&
term in eqn (1).
However, to the knowledge of the authors, pipe flotation has
not been demonstrated using measured data. Making
"#
$
the
subject of Eqn (1) gives:
"#
$+*
9:;8<'())
=>? &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&!@%
A value of 28.20 was assumed for
8
, interpolated from data
measured by Reilly and Orr [18] for a sand/rough concrete
interface. In Figure 10, two extreme scenarios are presented:
(i) full string weight (including TBM and power pack)
assumed, i.e.
'())
=
'
, and (ii)
'())
=0, i.e. the string is
fully buoyant. It is clear that scenario (i) is incorrect as
(impossible) negative values of
"#
$&
arise after a jacked distance
of 37 m. The correct normal stress lies somewhere between
the scenario (ii) data on Figure 10 and the
&&"#
$+ A
axis. This
suggests that the pipe string is almost fully buoyant (on
average over its length), if not fully buoyant.
Figure 10. Demonstration of pipeline buoyancy
Stoppages
As the tunnel is advanced, small peaks in jacking force arise
as a result of overnight and weekend stoppages, an example of
which can be seen in Figure 11.
Figure 11. Influence of stoppages on total jacking force.
It appears that the initial peak in force upon resumption of
jacking following a stoppage is reversed quickly. It can be
seen in Figure 12 how the jacking force (normalized by the
initial force upon recommencement of jacking) has reached
baseline values over a length of little more than one pipe
diameter.
Figure 12. Jacking force ratio over one tunnel diameter for overnight
stoppages
The increase in jacking force after different stoppage
durations, t, are shown in Figure 13. Stoppage durations are
categorised as follows: t < 3 h (pipe change and miscellaneous
minor breaks), 10 h < t < 20 h (overnight stoppages) and t >
20 h (weekends). The jacking force increase is determined
using the initial force after a stoppage minus the average
jacking force calculated between one and two pipe diameters
upon resumption of jacking. Results show that additional
friction to be overcome is 0.22 kPa for short breaks, 0.30 kPa
for overnight stoppages and 0.75 kPa for stoppages greater
than 20 h. This suggests that the length of a stoppage dictates
the increase in jacking force. Although the trend is similar to
that reported by Curran and McCabe [10], values presented
here are much lower in comparison.
Figure 13. Influence of jacking force after stoppages
Steering deviations
The vertical and horizontal deviations from the laser line are
recorded from both the back of the machine and at the drill
head tip. With the articulated joint in the TBM used for
steering, it is important to note that the drill head tip and rear
of machine may not be on the same alignment (see Figure 14).
0
40
80
120
160
200
240
050 100 150 200 250 300
Jacking force increase (kN)
Jacked distance (m)
t < 3 h
t = 1 0h - 20h
t > 2 0h
22h 15m 14h 10m

Horizontal deviations remain relatively low throughout the
drive (<10 mm); only vertical deviations are shown in Figure
14 (over a selected length).
Figure 14. Deviations and total jacking force along tunnel.
For this silty sand site, steering deviations did not play a
major role in the development of jacking forces. The small
spikes in jacking forces, which align with deviations between
40 mm and 50 mm, are actually due to stoppages at these
times.
5 CONCLUSIONS
This paper describes the microtunnelling process in the
context of a recent UK project, including the jacking,
lubrication and slurry separation processes. The results show
that for the drives considered, the lubrication system proved
very effective at maintaining low frictional forces, and the
data suggest that the string was at least partially buoyant for
the majority of the drive. The volume of lubrication used to
achieve this exceeded minimum recommended amounts and
efficiencies may be possible in this regard. Stoppages were
shown to have a significant but temporary effect on the
jacking force; high jacking forces upon resumption of jacking
after a stoppage return to ‘baseline’ levels after advancing
only the length of one pipe diameter. Machine deviations did
not appear to play a major role in increasing jacking forces for
this particular project.
ACKNOWLEDGEMENTS
The first author is funded by an Irish Research Council
Enterprise Partnership Scheme (IRC-EPS) Postgraduate
Scholarship, with Ward and Burke Construction Limited as
the industry partner. The third author is supported by the
Royal Academy of Engineering (U.K.) under the Research
Fellowship Scheme.
REFERENCES
[1] Chapman, D. N. and Ichiola, Y. (1999) Predictions of jacking forces for
microtunn elling operations, Tunnelling and Underground Space
Techno logy, Vol. 14, No. 1, pp. 31-41.
[2] Cheng, W., Ni, J. C., Shen, J. S., and Huang, H., (2017) Investigation
into factors affecting jacking force: a case study, Proceedings of the
Institution of Civil Engineers - Geotechnical Engineering, Vol . 170, pp.
322-334.
[3] Marshall, M. (1998) Pipe-jacked tunnelling: jacking loads and ground
movements. PhD Thesis, University of Oxford.
[4] Shou, K., Yen, J. and Liu, M., (2010) On the frictional property of
lubricants and its impact on jacking force and soil-pipe interaction of
pipe-jacking, Tunnelling and Underground Space Technology, Vol. 25,
pp. 469-477.
[5] Borghi, F. X. (2006) Soil conditioning for pipe-jacking and tunnelling,
PhD Thesis, University of Cambridge.
[6] McGillivray, C.B. and Frost, J.D., (2010) Influence of bentonite slurry
on interface friction, Characterization and Behaviour of Interfaces –
Proceedings of Research Symposium on Characterization and
Behaviour of Interfaces, IOS Press, Atlanta, pp. 11 9-124.
[7] Cheng, W., Ni, J. C., Arulrajah, A. and Huang, H., (2018) A simple
approach for characterising tunnel bore conditions based upon pipe-
jacking data, Tunnelling and Underground Space Technology, Vol. 71,
pp. 494 -504.
[8] Pellet-Beaucour, A.L. and Kastner, R., (2002) Experimental and
analytical study of friction forces during microtunnelling operations,
Tunnelling and Underground Space Technology, Vol. 17, pp. 83-97.
[9] Barla, M., Camusso, M. and Aiassa, S., (2006) Analysis of jacking
forces during microtunnelling in limestone, Tunnelling and
Underground Space Technology, Vol. 21, No. 6, pp. 668-683.
[10] Curran, B. G. and McCabe, B . A., (2011) Measured jacking forces
during slurry-shield microtunnelling in a boulder clay at Kilcock,
Ireland, Proceedings of the 15th European Conference on Soil
Mechanics and Geotechnical Engineering, Athens, Vol. 3, pp. 1627-
1632.
[11] Reilly, C. C., McCabe, B. A. and Orr, T. L. L., (2012) Analysis of
microtunn el jacking forces in alluvium and glacial till in Mull ingar,
Ireland. World Tunnelling Conference, Bangkok.
[12] Ni, J. C. and Cheng, W., (2012) Steering characteristics of
microtunn elling in va rious deposits, Tunnelling and Underground Space
Techno logy, Vol. 28, pp. 321-330
[13] Staheli, K. (2006) Jacking Force Prediction: An Interface Friction
Approach Based On Pipe Surface Roughness, PhD Th esis, Georgia
Institute of Technology.
[14] Ulkan, A. (2013) Development in Trenchless Technologies at
Herrenknecht, GSTT 1st No Dig Berlin 2013, Berlin, pp. 2A-1-1 - 2A-1-
6.
[15] Praetor ius, S., Schößer, B ., (2017) Bentonite Handbook: Lubrication for
Pipe Jacking, John Wiley and Sons.
[16] Curran, B. G. (2010) Analysis of Jacking loads and surface settlements
for Irish microtunnelling contracts, MEngSc Thesis, Nati onal Universit y
of Irela nd, Galwa y.
[17] Reilly, C. C. and Orr, T. L. L. (2012) Analysis of interface friction
effects on microtunn el jacking forces in coarse-grained soils,
Proceedings of the Bridge and Concrete Research in Ireland
Conference. Dublin. pp. 121-126.
[18] Reilly, C. C. and Orr, T. L. L. (2017) Physical modelling of the effect of
lubricants in pipe-jacking, Tun nelling and Underground Space
Techno logy, Vol. 63, pp. 44-53.
Stoppage