Large Diameter Pipe Roof Box Excavation For Passenger Linkway Tunnel
Large Diameter Pipe Roof Box Excavation For Passenger Linkway Tunnel
Large Diameter Pipe Roof Box Excavation For Passenger Linkway Tunnel
net/publication/328126747
Large diameter pipe roof box excavation for passenger linkway tunnel
CITATION READS
1 1,092
2 authors, including:
Senthilnath G T
The University of Queensland
21 PUBLICATIONS 17 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Senthilnath G T on 06 October 2018.
Abstract:
This paper presents a case study about the construction of a 6m x 8.5m cross
sectioned and ~120m long Bukit Panjang Underpass. The project consists of a
pipe roof box system that was constructed using 2.0m diameter steel pipes with
steel frame supports. The various aspects associated with the large diameter pipe
roofing system including the challenges encountered during construction have
been discussed in this paper. This was the first time large diameter pipes had
been used to form the pipe roof box in Singapore.
Keywords: Pipe roof support, Pedestrian Linkway, Pipe jacking, excavation methods
1.0 Introduction
Located in the western region of Singapore, the LTA contract 9057 Bukit
Panjang Underpass project was introduced to construct an additional MRT
Station entrance and underpass connecting to the existing Bukit Panjang
MRT Station. The underpass spans approximately 120m undercrossing Bukit
Panjang Road together with several utilities including a 13m wide canal and a
large diameter water main pipe. The overview and layout plan of the project
are shown in Figure 1 and Figure 2. The tunnel has a cross section of 8.5m
width x 6m height while the overburden above tunnel crown varies from 6m
to 13m.
A typical pipe roof system is formed by jacking several small diameter steel
pipes from one end of the tunnel (jacking shaft) to the other end of the tunnel
(receiving shaft) using either a micro-TBM or the soil-
displacement/hammering method. Upon completion of the pipe roof,
excavation of the tunnel takes place in small round lengths / spans (limited by
the bending capacity of the small diameter pipes). During the staged
excavation of the tunnel, the pipe roof is supported using a temporary or
permanent structure support system. A long linkway tunnel constructed using
2
the typical pipe roof system would therefore result in longer project duration
(due to short excavation spans). Moreover, installation of steel pipes within
allowable tolerances using pipe jacking for a tunnel of ~100m would require
an intermediate shaft.
Having considered the setbacks of a typical pipe roofing system, the decision
to use large diameter pipes was made. The selection of large diameter pipes
enabled using a more powerful TBM that not only facilitated a longer tunnel
length without the need of an intermediate shaft but also allowed the
introduction of a retractable machine. The number of pipes required to be
driven was also reduced to 13 nos (compared to 42nos using 0.813m dia
pipes).
3
Figure 3: Geological profile along the tunnel
Field permeability test indicated permeability for the weathered granite in the
range of 1x10-7 m/sec at invert level.
Certain stretches of the linkway tunnel was expected to have F1 / sandy GVI
at the tunnel crown which introduced the risk of face instability within the
pipe roof during tunnel excavation. Although this would not cause any global
4
failure, the possible collapse of a tunnel face would result in increased
ground surface settlement.
The proposed temporary support for tunnel excavation consists of steel pipes
with temporary steel frame as shown in Figure 4. The pipe roof box was
constructed using 13 numbers of 2m diameter steel pipes that were jacked
surrounding the tunnel boundaries using a micro-TBM. 2D Plaxis [1] with an
approximated plane strain using Mohr-coulomb material model was used to
analyze and design a typical cross section. In order to account for ground
movement due to pipe jacking, a negative volume strain was considered for
each pipe in the FE model. The mesh used around the tunnel is illustrated in
Figure 5.
The pipe roof system was not closed at bottom as it would be resting on GIV
or GIII. However, in order to control the water ingress during the linkway
tunnel excavation, post grouting was recommended. In addition, allowable
water ingress during excavation was established using a 2D flow analysis
model and was set to be less than 25 l/min, for a maximum round length of
10m.
Ground settlement was expected to happen during 2 stages (a) during pipe-
roof installation and (b) during excavation of the linkway tunnel and support
frame installation. The latter would be controlled by round length and the
stiffness of the pipe roof element while the former would be dependent on
various factors such as face stability, overcut and water ingress during cutter
tool replacement etc.
5
4.1 Factors influencing face stability
Once the pipe roof system forming the outer boundary of permanent works
was installed, the design intent was to carry out excavation from both ends of
the tunnel. An intermediate temporary shaft was proposed by the contractor
to expedite the excavation works giving four working fronts for the tunnel
excavation. Taking advantage of the higher bending moment capacity of
bigger roof pipes, steel support frames were designed for a typical spacing of
6m. 800 series and 900 series UB steel members were used as support
6
frames. Typical support systems for a similar excavation would consider
support members for invert location as well. However in this case, since the
tunnel invert was in competent ground (GIV and better), the permanent RC
slab was used as the temporary tunnel invert support. A typical tunnel
linkway section with RC base slab is shown in Figure 4. Casting the
permanent RC slab during excavation and installing the temporary steel
frames at longer spans facilitated to carry out permanent works faster.
The TBM setup which takes place before commencement of each drive
consists of setting up the jacking frame, the launching seal and the back
plate. 2 numbers of 700 ton capacity hydraulic jacks were used for the pipe
jacking operation. Figure 9 shows the typical jacking set up used in this
project.
7
Figure 9: Pipe Jacking Setup
The collapsible cutter head shown in Figure 8 was used in this project so as
to retract the TBM to the launching shaft without the completion of a
receiving shaft. During the TBM driving mode, the cutter head extends to a
diameter of 2.082m while during the retraction mode, the cutter head is
folded to diameter of 1.810m. The outer shield of the TBM is sacrificed to
allow for the retraction and the TBM is pulled back to the jacking shaft using
a winch system.
The TBM comprises of an inbuilt hydraulic conveying system which includes
a feed and slurry pump together with the respective pumping lines. During
the TBM operation, clean suspension (water or bentonite) is transported to
the tunnel face through the feed pumping system maintaining the required
face pressure at the tunnel front. Concurrently excavated soil is transferred to
the slurry treatment plant which includes a centrifugal and de-sanding system
(shown in Figure 10). Bentonite is injected to the voids between the steel pipe
and soil to stabilize the surrounding ground until annulus grouting is
complete.
8
Figure 10: Muck management and slurry treatment system
During the pipe jacking operation, ground water drawdown and surface
settlement was controlled by maintaining a target face pressure at the tunnel
front close to hydrostatic pressure (0.7 to 1.4 bars). During excavation, the
ground water drawdown was controlled through the water tight system
formed by the inter-connected steel pipes. In addition, the following
measures were taken to minimize the ground water drawdown and ground
surface movement.
• Jet Grouting Piles (JGP) within the tunnel – At the initial stages, soil
improvement to treat the sandy F1 and GVI layers within the tunnel
was considered to be impossible due to the major utilities along Bukit
Panjang Road. Introducing large diameter steel pipes enabled to carry
out JGP from within the steel pipes. JGP blocks were formed at 6m
intervals along the tunnel to hold the tunnel face at every excavation
span and prevent face instability / ahead-of-face ground relaxation. A
view of the grout block formed is shown in Figure 11. The micro JGP
9
machine which was used to carry out the ground improvement work
from within the pipe roof is shown in Figure 12.
10
plates or concrete packing. This minimized the settlement expected
during the tunnel excavation.
In order to monitor the impact of the pipe roofing and tunneling work on the
surroundings, several instruments including ground settlement markers,
utility settlement markers, water stand pipes, piezometers, extensometers,
strain gauges and 3D prisms were installed. As it could be seen from Figure
14, majority of the ground settlement and ground water drawdown was
observed during the steel pipe jacking activity. As an exception, a single
settlement marker recorded a maximum settlement of up to 76mm during the
steel pipe installation. The settlement plot versus time for the settlement
markers along the tunnel alignment is show in Figure 14. Incremental
settlement after pipe jacking activity was in the range of 20mm and some of
this settlement bounced back when the excavation was completed and the
ground water table stabilized to initial values. The ground was recharged
during the pipe jacking operation to maintain the water table within alert
levels. Figure 15 shows the time history of piezometer reading during the
project duration. As it could be inferred from the plot, the piezometer
remained stable during initial pipe jacking operation. However, there was a
sudden drop in piezometer (and hence associated settlements) when the cutter
head encountered mix face conditions and during cutter tool change in full
face rock. The recharge wells helped to restore the water table once the pipe
jacking was complete. After the installation of pipes, a water tight structure
was achieved and there was no further significant fluctuation of water table.
11
Figure 14: Ground settlement progress during pipe jacking & excavation
Currently, the tunneling work for the Bukit Panjang Underpass project has
been completed remaining permanent structure construction work. The
installation of 13 drives of 2m dia steel pipes using the pipe jacking was
completed within approximately 8 months from March to November 2016.
Pipe jacking at the mixed face/ soil-rock interfacing area took relatively
longer than pipe jacking in rest of the ground condition.
12
The 120m long linkway tunnel excavation was completed within 2.5 months
from February to April 2017 while the permanent structure casting and
removal of steel frames which was the final stage of the tunnel work was
completed within the 2.5 months between May and July 2017. Since the
concerns of water ingress and face stability were addressed before
commencement of the tunnel excavation work, there were no major
challenges during the tunnel excavation itself.
The Bukit Panjang Underpass project has demonstrated the use of large
diameter pipes for pipe-roofs and its advantages. The project has established
that by the use of larger diameter pipes, improved speeds of tunnel
excavation and high construction productivity can be achieved while
ensuring safe tunnel excavation.
Experience gained from this project confirmed that a robust pipe roofing
system can be formed using large diameter pipes while JGP within the tunnel
effectively improved loose soil (F1) at the crown proving to be excellent
material to provide a stable face during tunnel excavation.
References:
[1] R. B. J. Brinkgreve and P. A. Vermeer, “Plaxis users manual,” Balkema,
Rotterdam (The Netherlands), 2002.
[2] S. R. Kramer, W. J. McDonald, and J. C. Thomson, “Pipe Jacking and
Microtunnelling,” in An Introduction to Trenchless Technology, Springer,
1992, pp. 86–120.
[3] R. Röhner and A. Hoch, “Calculation of jacking force by new ATV A-161,”
Tunnelling and Underground Space Technology, vol. 25, no. 6, pp. 731–735,
2010.
13