Download full ebook of Geotechnical Aspects Of Underground Construction In Soft Ground Proceedings Of The Tenth International Symposium On Geotechnical Aspects Of Cambridge United Kingdom 27 29 June 2022 1St Edition online pdf all chapter docx
Download full ebook of Geotechnical Aspects Of Underground Construction In Soft Ground Proceedings Of The Tenth International Symposium On Geotechnical Aspects Of Cambridge United Kingdom 27 29 June 2022 1St Edition online pdf all chapter docx
Download full ebook of Geotechnical Aspects Of Underground Construction In Soft Ground Proceedings Of The Tenth International Symposium On Geotechnical Aspects Of Cambridge United Kingdom 27 29 June 2022 1St Edition online pdf all chapter docx
https://ebookmeta.com/product/structural-information-and-
communication-complexity-29th-international-colloquium-
sirocco-2022-paderborn-germany-june-27-29-2022-proceedings-merav-
parter/
https://ebookmeta.com/product/intelligent-tutoring-systems-18th-
international-conference-its-2022-bucharest-romania-
june-29-july-1-2022-proceedings-scott-crossley/
Practical Aspects of Declarative Languages 24th
International Symposium PADL 2022 Philadelphia PA USA
January 17 18 2022 Proceedings 1st Edition James Cheney
https://ebookmeta.com/product/practical-aspects-of-declarative-
languages-24th-international-symposium-padl-2022-philadelphia-pa-
usa-january-17-18-2022-proceedings-1st-edition-james-cheney/
https://ebookmeta.com/product/practical-aspects-of-declarative-
languages-24th-international-symposium-padl-2022-philadelphia-pa-
usa-january-17-18-2022-proceedings-1st-edition-james-cheney-2/
https://ebookmeta.com/product/cambridge-igcse-and-o-level-
history-workbook-2c-depth-study-the-united-states-1919-41-2nd-
edition-benjamin-harrison/
https://ebookmeta.com/product/proceedings-of-the-23rd-
international-symposium-on-advancement-of-construction-
management-and-real-estate-fenjie-long-editor/
M.Z.E.B. Elshafie, G.M.B. Viggiani & R.J. Mair, editors
GEOTECHNICAL ASPECTS OF UNDERGROUND CONSTRUCTION IN
SOFT GROUND
PROCEEDINGS OF THE TENTH INTERNATIONAL SYMPOSIUM ON GEOTECHNICAL
ASPECTS OF UNDERGROUND CONSTRUCTION IN SOFT GROUND, IS-CAMBRIDGE
2022, CAMBRIDGE, UNITED KINGDOM, 27-29 JUNE 2022
Geotechnical Aspects of
Underground Construction in
Soft Ground
Editors
Mohammed Z.E.B. Elshafie
Qatar University, Doha, Qatar
Robert J. Mair
University of Cambridge, UK
Cover top photo credit: Tideway
Bottom cover photo: © Copyright Crossrail
CRC Press/Balkema is an imprint of the Taylor & Francis Group, an informa business
All rights reserved. No part of this publication or the information contained herein may be
reproduced, stored in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, by photocopying, recording or otherwise, without written prior
permission from the publisher.
Although all care is taken to ensure integrity and the quality of this publication and the
information herein, no responsibility is assumed by the publishers nor the author for any
damage to the property or persons as a result of operation or use of this publication and/or
the information contained herein.
Applied for
Published by: CRC Press/Balkema
Schipholweg 107C, 2316XC Leiden, The Netherlands
e-mail: [email protected]
www.routledge.com – www.taylorandfrancis.com
Table of contents
Organisation xv
Sponsors xvii
v
Tunnelling in Urban Areas and Pile Interception Challenges – A Case Study: Bank Station
Upgrade Project (BSCU) 98
A. Nasekhian, C. Anthony, B. Haig, M. Dewhirst, J. Ares & C. Barker
Impacts of new development on existing underground assets using greenfield model 108
K. Neaupane & Y. He
Cementitious systems with carbon nanomaterials for underground infrastructure 115
I. Papanikolaou, A. Al-Tabbaa, M. Goisis & T. Embley
Pipe jacking tunnel construction crossing the Bang Pakong River 123
K. Piriyakul, S. Pochalard & E. Rungrueng
Nine Elms Station substructure – implementation of the Observational Method by
progressive modification 129
Nigel Pye, C. Lile, Anthony O’Brien & C. Penh
Evaluation of InSAR data for measuring the surface settlement during shield tunnel
construction of the North-South Line in Amsterdam 137
K.J. Reinders, F.J. van Leijen, R.F. Hanssen & M. Korff
Deep excavations for buildings in the Sabana Formation Bogota 153
J.A. Rodriguez
EPB tunneling and shaft construction in soft deltaic deposits for a railway link to Barcelona
airport 145
S. Sánchez, E. Silva, J. Izquierdo, A. Gens & E. Alonso
Use of jet grouted columns for deepening of basements and underpinning existing structures
at Valkyrien in Oslo 161
T. Sandene, K. Karlsrud & A. Worren
Settlements induced by EPB TBMs tunneling, a case study of theoretical and monitoring
values 169
A. Siemińska-Lewandowska & R. Kuszyk
Monitoring of an existing concrete-lined tunnel at CERN excavated in the molasse rock 177
K. Soga, V. Di Murro, L. Pelecanos, C. Kechavarzi, L. Scibile, J.A. Osborne & R.F. Morton
Achieving sustainability in tunnelling through innovation 184
A.H. Thomas
Measured post-construction ground response to EPBM tunnelling in London Clay 191
M.S.P. Wan, J.R. Standing, D.M. Potts & J.B. Burland
Gilgel Gibe II Hydropower Project in Ethiopia; TBM Tunnelling, when the rock turns into
mud: Analysis of a major collapse, its causes and solutions 199
E. Zoppis & A.M. Baldi
vi
Asymmetric pressure distribution in EPB shields: Evaluation of measurements and numerical
simulations 226
A. Bezuijen, T.S. Dang & G. Meschke
Pore pressures in front of a slurry TBM, the influence of plastering mechanisms 234
A. Bezuijen
Face pressure and volume loss relationships for pressurized tunneling in Granular soils 242
S.J. Boone & J.N. Shirlaw
Albert Embankment: Design of deep excavations in the River Thames foreshore 250
O. Brown, P. Stewart, S. Thomson, B. Patel, A.M. Waller, S. Sismondi & F. Quesada
Centrifuge and numerical modelling of the influence of structural stiffness on basement
heave in over-consolidated clay 259
D.Y.K. Chan & S.P.G. Madabhushi
Influence of TBM geometry on lining loads of deep tunnels 266
V. De Gori, A. de Lillis & S. Miliziano
Development of a new excavation technique for centrifuge testing in sand 275
N.E. Faustin, R.J. Mair & M.Z.E.B. Elshafie
Interpretation of soil parameters used for numerical analysis with small strain model for
deep excavation in loose to medium dense sand 283
H.B. Bin-Chen, P. Khac Hai & C. Hung
Prediction of damage intensity of reinforced concrete tunnels and soil against blast loading 291
K. Senthil, S. Rupali & L. Pelecanos
EPB-TBM tunnel under internal pressure: Assessment of serviceability 300
N.A. Labanda, A.O. Sfriso, D. Tsingas, R. Aradas & M. Martini
A risk assessment of downdrag induced by reconsolidation of clays after upwards pipe
jacking 309
N.A. Labanda, A.O. Sfriso, D. Tsingas, R. Aradas & M. Martini
Small-scale modelling of pile drilling in sand – investigation of the influence on surrounding
ground 317
E.J. Lande, S. Ritter, E.J. Lande, H. Tyvold & S. Nordal
Excavation of an artificial tunnel using compressed air 325
N. Losacco, M. Cafaro & R. Marazzita
The use of adaptive smoothed finite-element limit analysis to seismic stability of tunnels 330
H.C. Nguyen
Upper bound analysis of seismic stability of tunnels using cell-based smoothed finite element 337
H.C. Nguyen
Investigation of the seismic performance of the complicated tunnel sections with
non-uniform heights 343
H. Nitta, S. Ito, T. Otsuka, S. Konishi, K. Tsuno, S. Tsuchiya & K. Maekawa
Influence of the annulus grout on the soil-lining interaction for EBP tunneling 350
M. Ochmański, G. Modoni & G. Spagnoli
vii
Long term additional load on a shield tunnel in soft clay due to clay consolidation with
water leakage 357
S. Oka, J. Saito, Y. Ito, W. Li, S. Kaneko, A. Afshani & H. Akagi
A study of a strut-free excavation system in deep excavations 365
C.Y. Ou, A. Lim, P.G. Hsieh & S.C. Chien
Increasing the passive resistance of deep excavations in very soft soils to mitigate ground
movements through centrifuge modelling 371
J.P. Panchal & A.M. McNamara
Machine learning algorithms applied to the blowout susceptibility estimation around
pressurized cavities in drained soil 379
F. P.-Ramirez & C. Arson
Specifying and Testing fibre Reinforced sprayed concrete: Advantages and Challenges of
some testing methods 385
Benoit De Rivaz
Undrained seismic response of tunnels 395
E.A. Sandoval & A. Bobet
Preliminary evidences on the influence of grains micro-structural features on the TBM tools
wear 401
D. Sebastiani, S. Miliziano, G. Guida & F. Casini
Front-face pressure drop during the standstill phase for EPB mechanized tunnelling in
coarse-grained soils 408
D. Sebastiani, S. Miliziano & A. Bezuijen
A miniature EPB TBM for use in a geotechnical centrifuge 415
C.J. Shepheard, A.S.N. Alagha, G.M.B. Viggiani & S.K. Haigh
Urban tunnelling in glacial soils: Tunnel de Champel, Geneva 421
W. Steiner, T. Witschi & A. Ferrari
Simplified stress-strain models applied to data from triaxial and pressuremeter tests on
London Clay 430
P.J. Vardanega, M.D. Bolton, S.K. Haigh, R.W. Whittle, A. Klar & M.G. Williamson
Propped cantilever wall stability design with the ‘What You Design Is What You Get’
method – background and development 438
C.K.S. Yuen
Multi-objective optimisation design for composite tunnel linings using non-dominated
sorting genetic algorithm 444
W. Zhai, D. Chapman, A. Faramarzi, H. Huang & D. Zhang
Support pressure transfer at a slurry supported tunnel face due to time dependent decrease of
soil permeability 451
C. Zhao, Z. Zizka, B. Schoesser, M. Thewes & A.A. Lavasan
SEM deformation prediction and observation by 3D numerical analysis 459
H. Zheng, M. Mooney, M. Gutierrez & C. Bragard
Face stability of slurry shield-driven tunnel in an aquifer 467
T. Xu, W.H. Zhou & A. Bezuijen
viii
Numerical analysis of Double-O-Tube shield tunneling in Shanghai 475
D. Zhou & L. Zdravković
Physical modelling of transient processes at the slurry supported tunnel face during shield
excavation 482
Z. Zizka, B. Schoesser & M. Thewes
ix
Long-term behaviour of ground around tunnel due to groundwater level fluctuations 590
Wei Li, Shigeaki Oka, Alireza Afshani & Hirokazu Akagi
Numerical investigation into time-dependent effects on short-term tunnelling-induced
ground response in London Clay 597
A.R. López, A. Tsiampousi, D.M.G. Taborda, J.R. Standing & D.M. Potts
Semi-coupled modelling of soil-structure interaction during tunnel construction: Two case
studies from Bank Station Capacity Upgrade 605
A. Luciano, M.N. Pascariello, E. Bilotta, S. Acikgoz & R. Mair
Centrifuge tests on tunnel-building interaction in liquefiable soil 613
G. Miranda, V. Nappa, E. Bilotta, S.K. Haigh & S.P.G. Madabhushi
The effect of deep excavation on existing railway tunnel 620
M. Mitew-Czajewska
Simplified modelling of the transient response of underground structures due to dynamic
loads generated from underground tunnels 627
L. Pelecanos, K. Senthil & S. Rupali
Mapping the risk of building damage due to excavation-induced displacements 632
S. Ritter, L. Piciullo, A.O. Kydland Lysdahl, M. Kahlström, J. Langford & F. Nadim
A new approach for compensation grouting in highly permeable gravel 640
M. Sailer, J. Fillibeck & S. Geuder
A case study on the effects of anchor drilling in soft, low sensitive clay and sandy, silty soils 647
T. Sandene, S. Ritter & E.J. Lande
Prediction of long-term settlement in shield tunnel using GA-BP neural network 656
Yi-Ming Shen, Dong-Mei Zhang, Jie Zhang, Dong-Mei Zhang & Jie Zhang
Tunnelling through a piled foundation: Interaction effects 664
Davor Simic & Belén Martínez-Bacas
The use of protective structures to reduce tunnelling induced damage to buildings 673
G. Song, A.M. Marshall & C.M. Heron
Evaluation method on ground movement using continuum ground model 681
M. Sugimoto, J. Chen, P.T. Anh, K. Manabe, L.G. Lam & S. Chaiyaput
The response of the Europier Terminal Building to the excavation of the T2B Basement at
Heathrow Airport 689
G.R. Taylor
Longitudinal structural deformation of shield tunnels induced by overlying excavation 696
H.N. Wu, S. Chen, R.P. Chen, Y. Liu, H.Z. Chen, F.Y. Meng & S.L. Shen
Simulating the water-soil leakage induced deformation around the shield tunnel with
material point method 702
X.C. Xie, D.M. Zhang, M.L. Zhou, S.J. Feng, D.M. Zhang & S.J. Feng
Influencing factors and protection technologies of underground diaphragm wall and deep
foundation pit construction on metro station 710
H.F. Xing, L.L. Liu & H. Zhang
x
A centrifuge modelling study on the effect of foundation configuration on tunnel-frame
interaction 719
J. Xu, A.M. Marshall & A. Franza
Modelling ground response to TBM tunnelling with active face support 727
T. Xu, W.H. Zhou & A. Bezuijen
Tunnel face stability considering drainage and surface settlements 732
C. Yi, S. Senent & R. Jimenez
Impact of subsoil spatial variability on deformations of immersed tunnel 738
X. Zhang, X. Wu & W. Broere
Coupled elastoplastic analysis of the soil-pile foundation interaction induced by deep
excavations 746
C. Zheng, A. Franza & R. Jimenez
xi
Permeability characteristics of coarse-grained soil conditioned with foam for EPB shield
tunnelling 836
S. Wang, S. Huang, Q. Hu & Z. Liu
Stabilisation of Singapore soft marine clay using a novel sustainable binder for underground
construction 836
H. Yu, Y. Yi, R. Liu & N. Jiang
Experimental study of pore water pressure development in soil when foam infiltrates into
saturated sand 848
D. Zheng, A. Bezuijen & M. Thewes
xii
Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9
The 10th symposium (IS-Cambridge 2020) was planned to be held in June 2020 in Cambridge,
United Kingdom. The call for papers drew an overwhelming response and 278 abstracts were
received, resulting in 112 technical papers accepted for publication in the proceedings from
over 25 countries. However, the COVID-19 outbreak in January 2020 and the very rapid
spread of the virus to all corners of the globe, with all the associated health risks, eventually
prompted the symposium organisers to postpone the conference to June 2022. The organisers
also decided to publish two versions of the proceedings. A first version (Version 1), this
volume, which contains only the accepted technical papers (112 in total), is published almost
1.5 years prior to the actual symposium to ensure that the hard and valuable work of the
authors is duly recognised. In 2022, just before the actual symposium, a second version of the
proceedings (Version 2) will be published. This new version will contain all the components
traditionally found in TC204 symposium proceedings including: all papers in Version 1 plus
the written versions of (i) the Fujita Lecture, (ii) the Special Lectures, (iii) the Session Reports,
and (iv) the Bright Spark Lecture. Version 1 will go out of print as soon as the complete Ver
sion 2 is published.
The themes for the two versions of the proceedings are in line with the terms of reference of
Technical Committee TC204 and include:
xiii
The success of publishing this version of the proceedings must be attributed to the authors of
the papers who shared their valuable work, the reviewers who diligently worked through the
papers providing useful technical advice and to the teams at Cambridge, which conducted the
editorial review of the papers and helped manage the complex logistics through very uncertain
times. The work presented here would not have been possible without the efficiency and tre
mendous support of Mrs. Tian Wu, Mrs. Anama Lowday and Dr Njemile Faustin who sup
ported the conference organisers every step of the way. The strong support from the staff at
Taylor and Francis, and from the Chairman and Secretary of TC204 (Prof. Adam Bezuijen
and Prof. David Chapman) is also acknowledged.
xiv
Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9
Organisation
xv
Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9
Sponsors
Conference Sponsor
Session Sponsor
xvii
Field case studies and sensing technologies
Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9
K. Soga
Department of Civil Engineering, University of California, Berkeley, USA
ABSTRACT: The behaviour of existing cast-iron (CI) tunnels subjected to ground movements induced by
new tunnelling works is a much discussed topic. In many cases new tunnels and shafts, particularly in highly
populated urban areas, will need to be constructed adjacent to existing tunnels. This inevitably results in
ground deformations that are transmitted to these structures. This paper investigates the response and the
radial tolerance of CI lining to these imposed movements. This paper highlights the common ‘damage assess
ment’ procedures in practice and emphasizes the need of paying equal attention to historical case studies
when carrying out often complex analyses. A recent case study from the construction of Crossrail Line
(Queen Elizabeth Line) in London has been presented to demonstrate this. In this example the CI lining is
subjected to significant deformations caused by nearby tunnelling and excavation works.
DOI: 10.1201/9780429321559-1
3
of horizontal movements and even more so at the Underground standards (London Underground 2007
longitudinal direction ahead of the face of the tunnel- and 2009). Li et al., (2015a) shows an example of this
ling. Alhaddad et al 2017 and 2020 describe these in envelop (Figure 1) for a typical geometry of LU North
more details. The focus of this paper is not the accur ern Line CI tunnels (Diameter = 3.8 m and Depth = 19
acy of predicting these movements but rather the mbgl at Euston Station area). When the lining is
response of CI tunnels to accommodate them at defected (visually inspected), reduction factors are
radial directions. applied.
(ii) It is often conservatively assumed that CI tun
nels deform freely to follow these calculated ground
2.2 Determine existing hoop forces in the lining
movements from stage (i). In many cases, this has
been shown to be a reasonable assumption along the Hoop forces are determined using the ‘Elastic Con
longitudinal axis, where the cast iron rings are more tinuum’ method after Duddeck & Erdmann (1985)
flexible to slide and bend against each other to accom where the maximum and minimum hoop thrusts are
modate those movements (Alhaddad 2016). However, calculated using Equation 1 and Equation 2, respect
transversely, this is subject to debate for mainly two ively. Existing hoop forces are calculated based on
reasons: 1- the calculated horizontal movements given input parameters. The imposed hoop forces
derived from the empirical methods are not as reliable are assumed to be zero unless it is thought that sig
as the settlement predictions and 2- the CI tunnel rings nificant loading changes occur.
are less flexible to accommodate movements radially
than they are longitudinally. Hence, the importance of
case studies to provide evidence based ‘assessments’.
4
2.3 Determine existing bending moment in the 2.6 Circularity of the lining
lining
Wright (2009) states that typically CI tunnels have
The bending moment within the lining is calculated a squatting of 0.5% to 1%. When the existing circular
using Equation 3 after Morgan (1961). The existing ity of the tunnel is unknown, the LU standard requires
bending moment is measured by assuming an exist the use of a value of 0.6%. If the assessment for 0.6%
ing ovalisation. For LU tunnels this is specified to ovalisation fails, it is recommended to repeat the pro
be 0.6% of the tunnel diameter (the applicability of cedure and to adopt a more realistic approach taking
this assumption is debated in this paper). For into account that some of the existing ovalisation has
example, a 3.56 m diameter tunnel will have been caused by the as-built imperfection of the lining
a maximum radial deflection of 10.69 mm. This during its initial construction. Assuming a correct dir
deflection is assumed to be either constant (Figure 2 ection of this imperfection (squat or elongation) is
right) or variable (Figure 2 left). For the variable fundamental.
case, the assessment considers both scenarios: The TLL Tube Line Report (2005) has carried out
squatting (horizontal egg shape) and elongation a comprehensive and an extensive laser scanning and
(vertical egg shape). FE modelling analysis of 33% of running tunnels. 93%
of these experience around 1% squatting, which is
believed to be caused by the imperfection during
installation.
When looking at a number of CI tunnels that are
Where δ = Deflection (actual radius - nominal built adjacent and one after the other, the data show
radius) that one of these tunnels shows more squatting than
the other (Wright, 2009; 2010). This suggests that at
2.4 Determine increase in bending moment from least this differential squat between the two tunnels
lining distortion due to the new construction (assuming the one that was squatted more was installed
first) might have been caused by the changes in the
The total bending moment due to the construction stresses within the lining rather than by the as-built
work is measured by adding the existing deflection imperfections. This magnitude should not be
value to the maximum imposed value. In tunnelling disregarded.
related cases this value is calculated from assuming Almost all of the linings scanned by TLL had
that the lining follows the imposed ground move experienced squat (Wright, 2010) while FE model-
ments freely and equates to the differential soil ling and soil parameters at typical LU tunnel depths
movements radially along the extrados of the should result in elongation of the tunnels (earth pres
lining. sure coefficient ‘K’ values are thought to be 1-1.25).
This indicates that changes in horizontal stresses
2.5 Check the axial force and bending moment during the excavation and/or the drainage of soil in
remain within the permissible capacity the long-term caused by the higher permeability of
envelope the tunnel has lowered the surrounding pore water
pressure and resulted in the coefficient earth pressure
The envelope is determined at stage 1 and if the ‘K’ values to be lower than 1. If we accept this
imposed stresses calculated at following stages are hypothesis, then it is more likely that the recorded
outside the envelope further analyses (e.g. FE mod non-circularities are largely due to the external stres
elling) and/or assessments will be needed. Mitigation ses and less likely to be caused by the as-built imper
measures are also recommended accordingly. fection. It will be shown in the presented case study
An important input into specifying the tolerable that tunnel linings are more susceptible to imposed
ovalisation on the lining is the existing circularity of squatting than they are to elongation (cracks hap
the lining and the assumptions that are made about its pened where rings experienced squatting).
nature. This indicates that current damage assessment
procedures might need revising, especially where the
CI lining is subjected to additional squatting. Aas
suming that the non-circularity is a result of as-built
imperfection is not completely justified. On the other
hand, assuming that the ring is a rigid ring (con
tinuum theorem) is not perfectly representative too.
Li (2014) carried out a 3D FE modelling paramet
ric study and suggests that bending moment is pri
marily sustained by the contact of the circumferential
flanges with little influence from the joints while
hoop thrust is mainly transmitted through the joints.
Figure 2. Deflection Shape of the Lining; squat (left) and This should be taken into account and a ‘continuum
uniform (right). theorem’ that assumes a rigid ring is therefore a less
5
realistic approach. When modelling a flexible ring,
Li (2014) observes that the ovalisation is first caused
by the pressure of the ring’s self-weight and con
tinues to change by soil loading to a value as high as
1% in the absence of imperfections. A value which is
close to what has been witnessed in the majority of
circularity surveys of tunnels. At complex scenarios
Figure 3. RMT’s Rings Condition and Geometry.
such as cross-passage openings, this modelling
becomes especially less relevant (Li et al., 2015b).
It is noteworthy to mention that the scanned tun
nels have not had concerning signs of damages and of on-going water ingress could be seen. This was
therefore have not necessarily exceeded their stress confirmed by the many visits of the author to the
state envelopes. In other words, assuming an elastic tunnel. A typical schematic of the missing bolts and
continuous ring is not a realistic approach, although tunnel cross-section is shown in Figure 3. Note that
it is surely a conservative one. the location of the key-segment varies and some
However, this becomes less conservative when times staggers between the adjacent rings.
the existing ovalisation is treated to be an as-built The invert is made of an in-situ cast concrete
imperfection. More importantly, the direction of ova block. The quality and consistency of the concrete is
lisation (squat or elongation) becomes more relevant not known and historic drawings and records show
to making a more evidence based decision, as will that there is an underlying layer of gravelly material
be demonstrated in the case study below. Deploying beneath the concrete surface.
mitigation measures, such as monitoring, become
a pragmatic tool to manage the uncertainties and the 3.2 Description of Crossrail work
risks associated with such assessments.
The Crossrail tunnels were constructed using SCL
techniques and excavated in two stages: a pilot
2.7 Monitoring ovalisation tunnel of approximately 6 m diameter followed by an
Monitoring helps with managing the risks during the enlargement of the diameter to approximately 11 m.
construction work but also provides an insight that The invert of the RMT is at an elevation of
can be used for future assessments. Conventionally between +92 to +86 m ATD, falling from west to
systems such as ‘Bassett Convergence System’, east over the extent of the interface with the Cross-
prisms read by automatic total stations and chain of rail works. The RMT is located within London Clay,
tilt meter sensors have been used to monitor the ova while the Crossrail works extend into the top of the
lisation in real-time and more recently new technolo Lambeth Group. The interface between the London
gies such as shape arrays and digital image Clay and Lambeth Group is at around +76 m ATD.
correlation (DIC) techniques have been deployed The new tunnelling work crossed RMT at four
successfully to measure these deformations. Alhad locations (Figure 4). They are described below
dad et al. (2020) describes deployment of a high pre (based on a chainage reference at RMT):
cision DIC system within tunnel environments.
6
1) Chainage 223 to 243: Construction of the SCL
westbound platform tunnel (PTW) with 6m
clearance.
2) Chainage 263 to 288: Construction of the SCL
central concourse passageway with 5m
clearance.
3) Chainage 298 to 313: Construction of SCL
Finsbury Circus access shaft excavated around
2 m to the south of the RMT. The shaft was
approximately 40 m deep; varying between
12.5 m and 15.7 m in diameter.
4) Chainage 343 to 485: Construction of the SCL
eastbound platform tunnel. To the east of chai
nage 343 the clearance between RMT and the
enlarged Crossrail platform tunnel is approxi
mately 2 m decreasing to less than 0.5 m Figure 5. Propping Installed over 15 m of RMT.
towards the end. From around chainage 460 to
480, the RMT gradually enlarges towards a step After construction had finished, an inspection by
plate junction. The new tunnel was constructed the author and Crossrail’s Asset Protection team as
parallel and directly beneath these chainages for well as the limited monitoring data in place con
more than 100 m. The photograph in Figure 4 firmed that the props did not get engaged (they did
shows the enlargement directly beneath the not take any load meaning elongation of all of the
RMT at these chainages. rings were less than 10 mm or 0.4% ovalisation).
Other relevant sections to note are as follows:
3.4 Monitoring layout
1) RMT Enlargement Chamber: This chamber,
bordered by chainages 243 and 263, consists of Figure 6 shows the conventional monitoring that was
13 rings of 3.7 m diameter with an opening at carried out inside RMT. This included the use of auto
the northern side to an unused shaft. The con mated total stations (ATS) monitoring 3- and 5-point
nection between the larger rings and the smal radial prism arrays at generally 20 m centres. In add
ler diameter rings is a concrete head wall ition, levelling points (LP) were installed in the invert of
which is typical for such transitions. the tunnel (track-bed) and monitored manually on
2) Blomfield Box: From chainage 440 to 475, a weekly basis. Looking west the arrangement of the
a 36m deep box-retained cut excavation was prisms were; P1 at Crown, P2 and P4 at right-axis and
formed around 5m to the south of the RMT. right-invert and P3 and P5 at left-axis and left-invert,
3) Propping: In accordance with the Damage respectively. LP at invert level was monitored manually.
Assessment for the POT, propping was Visual inspection of the RMT was also carried out
installed to mitigate radial deformation; along generally on a weekly basis to check for any signs of
a 15 m section between chainage 375 to 390. distress or change in tunnel condition.
Over a short section of the RMT above the con
struction of eastbound platform tunnel, further moni
3.3 Damage assessment and mitigation measures toring was installed by Cambridge Centre for Smart
7
Infrastructure and Construction (CSIC). This work is 5) Stage 5: Enlargement of the influencing cross-
described in Alhaddad et al. (2014). The results of this passages and the start of Blomfield Box exca
monitoring are not relevant to this study except that vation – May 2014.
no damage or cracks were seen in this section of the 6) Stage 6: End of the majority of the construction
tunnel. The lining there was mainly subjected to including Blomfield Box and the arrival of the
elongation (0.4%) and longitudinal curvatures. Eastbound TBM – February 2015.
The settlements of the crown level and invert level,
3.5 Construction progress and monitoring results the horizontal convergence of the RMT measured from
Figure 7 shows sketches of the construction progress points 2 and 3 and the vertical convergence measured
from points 1 and LP for each of these phases are
at each of these stages. The significant movements to
shown in Figure 8. Vertical convergence is measured
the RMT occurred at the following times:
from differential values collected from two different sys
1) Stage 1: Excavation of the Finsbury Circus tems (ATS and manual monitoring) with an estimated
Shaft (FCS) – Completed before June 2012. measurement error of up to 2 mm. Also, LPs only meas
2) Stage 2: Excavation of the Central concourse ure the settlement and are installed on the track bed and
pilot tunnel and enlargement – Completed by are assumed to follow the lining at the invert level.
early April 2013. A maximum settlement of approximately 100 mm
3) Stage 3: Excavation of the Eastbound and was measured for the construction of the westbound
Westbound pilot tunnels – Completed in platform tunnel and approximately 90 mm for the
October 2013. construction of the eastbound platform tunnel. This
4) Stage 4: Excavation of the Eastbound and
Westbound enlargement and influencing Cross-
passages pilot tunnels – Completed by mid-
March 2014.
8
exceeded the preconstruction assessment calculating
a maximum of 70mm settlement and further assess
ments had to be carried out during the works. This
included carrying out a circularity survey to measure
the tunnel’s existing circularity (Crossrail 2013).
The horizontal and vertical convergences give an
indication of the magnitude of the ovalisation
imposed on the lining. Concluding from the direction
of the movements, RMT experienced up to 10 mm
of elongation (with no signs of damage) and up to
10mm of squat (where cracks were reported). These
were above the 7 mm value calculated during the ini
tial damage assessment procedure.
9
When looking at the convergence history of the deformation shape. This trend of movements implies
closest monitored ring at CH243 (7 rings away from that the cracks or some of them may have happened
the nearest crack – where the furthest crack is one around the peak (the invert cracks had been spotted
ring away from the headwall and is roughly at CH on a visit on 7th of February 2014, hence the cracks
250), it may be seen that there is a rapidly increasing must have happened before this date).
trend of movements when PTW enlargement excava
tion approaches the tunnel (Figure 11). The values
are maximum on 28th of January 2014 when the tun 4 CONCLUSION
nelling face was recorded to be directly beneath the
RMT. The trends start to decrease as the tunnelling CI tunnels deform longitudinally and/or transversely
face crosses RMT and proceeds to the other side. to accommodate external movements (such as those
Data are baselined on 15th of July 2013, a few caused by nearby construction activities). The focus
days before the chainage was influenced by the PTW of this paper is the response of CI tunnels to those
pilot excavation (the construction of the pilot tunnel movements transversely and along their radial rings.
then was paused until 1st of October 2013). This paper demonstrates the ‘damage assessment’
It should also be noted that the divergence of axes stages that are often followed to assess whether CI
1-5 and the convergence of axes 3-4 start to return to tunnels can accommodate those movements radially
their original shape after a peak corresponding to the and briefly describes the mitigation measures where
PTW crossing underneath. The trend of returning to the deformation are assessed to be excessive; namely
the original shape appears to be taking on a plastic the use of props inside the tunnel, when possible.
Mitigation measures are not often practical (for
examples props are not possible within running tun
nels) and it is important to understand the likely
response of the tunnels to be able to evaluate the
risks in advance of the start of any given construc
tion work. Numerical assessments are complex, and
results of such assessments are not guaranteed to be
correct. Historical case-studies provide an important
insight to enable evidence-based decision makings.
This paper presents a case study where the CI
tunnel was subjected to significant radial movements.
It is argued that where constructed within London
Clay, typical CI linings (uniform tunnels not influenced
by other structures such as existing cross passages) are
likely to be more susceptible to squatting than they are
to elongating deformations. This is because the evi
dence suggests that CI linings have experienced squat
ting over their lifetime and as a result they would have
less bending tolerance to accommodate squatting than
elongation.
The cracks in this case study have evidently been
the result of radial deformation when the rings were
squatting with about 0.4-0.5% ovalisation. The rings
had elongated with similar ovalisation values with
out any sign of damage. This reinforces the argu
ment that CI rings experience squat over their
lifetime. In this case the alignment of the segments
and the staggering of the keys appear to also contrib
ute to the radial tolerance of CI rings (by perhaps
concentrating the loads over the weak axes).
Designers must be more cautious and should not
downgrade the existing circularities as ‘as built
imperfections’ especially when the new constructions
are likely to impose more squatting on the tunnel.
This is, for example, when excavating shafts nearby
or when carrying out tunnel excavations with
a parallel and an offset axis from the existing linings.
On the other hand, accepting that there is the risk
of cracks happening on the radial flanges will open
Figure 11. History of Monitoring of CH 243: Settlement (a), the conversation between the stakeholders early on
Horizontal Convergence (b) and Vertical Convergence (c).
and will provide a more pragmatic approach to
10
mitigating the risks. The cracks on their own are not Alhaddad, M. et al., 2020. Cast-Iron Tunnels’ Tolerance to
likely to impose a critical risk to the operation of the Imposed Longitudinal Settlement Curvature Geotechni
rail-tunnels and at extreme scenarios they can be que Symposium in Print: Linear infrastructure.
fixed/replaced after the construction work is com Crossrail, 2007. London Underground Interface, Case His
plete. Hence providing more extensive and better tar tories, Prepared by Geotechnical Consulting Group for
geted (locally targeted) monitoring solutions for Crossrail Project, 1D0101-G0G00-01022A.
such cases could become a more favourable Crossrail, 2010. Civil Engineering Design Standard – Part
7, Ground Movement Prediction, Version 6.0, CR-STD
approach. -303-7 (CEDS 7).
Crossrail, 2011. Assessment of ground movement effects
on the Post Office tunnel at Liverpool Street Station
ACKNOWLEDGEMENT (PO/05), Rev 3.0, C122-OVE-C2-RAN-CR101-00016.
Crossrail, 2013. RESULTS FROM LASER SCANNING
This work was carried out in CSIC and would not THE POST OFFICE TUNNEL AT LIVERPOOL
have been possible without the support of Arup, STREET STATION (PO/05), Rev 2.0, C122-OVE-C2-RG
CH2M Hill (now Jacobs), London Underground and -50031.
Royal Mail Group and thanks to Engineering and Devriendt, M. & Alhaddad, M., 2015. Construction impacts
Physical Sciences Research Council (EPSRC) for of Crossrail Liverpool Street Station on the Royal Mail
their financial support. The authors would like to Tunnel. XVI ECSMGE Geotechnical engineering for
infrastructure and development 1: 831–836.
namely acknowledge Robert Mair, Mohammd Elsha Duddeck, H. & Erdmann, J., 1985. Structural design models
fie, Michael Devriendt, Frances McDonnell, Mathew for tunnels. International Journal of Rock Mechanics and
Wilcock and Chang Ye Gue for their contribution Mining Sciences & Geomechanics 20(1): 83–91.
throughout this work. Li, Z., 2014. Long-term Behaviour of Cast-iron Tunnel
Cross Passage in London Clay. PhD thesis, University
of Cambridge, Department of Engineering.
REFERENCES Li, Z., Soga, K. & Wright, P., 2015a. Behaviour of cast-iron
bolted tunnels and their modelling. TUNNELLING AND
Alhaddad, M. 2016. Photogrammetric monitoring of cast- UNDERGROUND SPACE TECHNOLOGY 50: 250–269.
iron tunnels and applicability of empirical methods for Li, Z., Soga, K. & Wright, P., 2015b. Long-term performance
damage assessment. PhD Thesis, Department of Engin of cast-iron tunnel cross passage in London clay. Tunnel-
eering, University of Cambridge, Cambridge, UK. ling and Underground Space Technology 50: 152–170.
Alhaddad, M. et al., 2014. Multi-Suite Monitoring of an London Underground, 2007. Manual of Good Practice,
Existing Cast Iron Tunnel Subjected to Tunnelling Civil Engineering – Deep Tube Tunnels and Shafts,
induced Ground Movements. Tunneling and Under G-055, Version A1, (LU-G055).
ground Construction - Geo-Shanghai. American Society London Underground, 2009. Standard 1-50 Civil Engineer
of Civil Engineers: 293–307. ing Common Requirements, Version A1, (LU-1-50)
Alhaddad, M. et al., 2017. Imposed Longitudinal Settle Morgan, H.D., 1961. A contribution to the analyses of
ment on a Cast-iron Tunnel from the Excavation of stress in a Circular Tunnel. Geotechnique 11(1): 37–46.
a New Tunnel Beneath. 9th International Symposium on Wright, P., 2009. Assessment of London Underground tube
Geotechnical Aspects of Underground Construction in tunnels investigations, monitoring and analysis. Smart
Soft Ground, IS – Sao Paulo: 343–353. Structures and Systems 6(3): 239–262.
11
Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9
ABSTRACT: The impacts of the lubrication upon the pipejacking works are significant and necessary
to be considered in tunnelling design. This study described a method that can be used to evaluate the
lubrication performance by the reduction in the frictional coefficient μ as a function of injection type,
soil and lubrication natures, and misalignment. The results of an application of the proposed method to
a pipejacking project in soft alluvial deposits were presented. The effect of misalignment increased the
frictional stress τld to 12.5 kPa, reducing the reduction in the μ value to 71%. The combined effects of
misalignment and varying face resistance contributed to the τld value of 4.0 kPa and were deemed as
the main cause leading the low reduction of 84%. The misleading reduction of 60% was attributed to
the inability of the occasional gravel to develop lower face resistance. The effects had greater influence
on the lubrication performance than the injection type.
DOI: 10.1201/9780429321559-2
12
recommended by Stein et al. (1989) for soil-pipe
(unlubricated) interface. The described method that
can be used to evaluate the lubrication performance
can be briefed in short as follows: (1) preparing the
baseline of jacking loads, (2) calculating the fric
tional stress τld for lubricated drives, (3) calculating
the normal contact pressure σz, (4) evaluating the
frictional coefficient μld for lubricated drives, and (5)
assessing the percentage reduction in the μ value.
3 PROJECT BACKGROUND
3.1 Background
The slurry shield was adopted to perform excavation
of the two pipejacking drives in soft alluvial deposits.
The alignment for all the drives is straight. During
pipe ramming, the bentonite slurry with unit weight
of 10.6 kN/m3 was used to stabilise the excavation
face and transport tunneling soil spoils to decantation
chambers. Since the tunnels were excavated via the
cutter wheel of 1500-mm in diameter, an overcut
annulus of 30 mm was formed by using the smaller
concrete pipe of 1440 mm in diameter. The two
drives characteristics are detailed in Table 1.
Figure 1. Distribution characteristics of injected lubricant
for two-point injection. 3.2 Engineering geology
The stratigraphic profile is established with reference
to five 15-m deep geological boreholes penetrating
4 m and 11 m distances from the face is saturated through the 6-m thick silty sand layer into the
through Injectors 2 and 1 by injecting 0.13 m3 7.5 m thick poorly-graded to well graded sand and
(20.9 %) and 0.50 m3 (79.1 %) lubricant, respectively. gravel layer. Based upon the standard penetration
Figure 1 also provides the details about the ramming of test and triaxial test results, the soil physical and
RCPs 15 and 16. Since effective lubrication cannot be mechanical properties are listed in Table 2.
easily sustained for long-distance pipe ramming, multi-
point injection may be used to prevent inadequate
lubrication from occurring. 4 ANALYSIS AND DISCUSSIONS
13
Table 2. Summary of soil physical and mechanical F 1962 (ASTM 2011), and GB 50332 (MOC 2002) are
properties. deemed to be the well-developed models that can be
used to calculate the normal contact pressure σz. The
Thickness SPT-N γ c’ ϕ’ qu models were modified with reference to the Terzaghi
arching model founded upon active trap-door experi
Layer m kN/m3 kPa ° kPa ment (Terzaghi 1936) where the shear bands arise from
the outside of tunnel cross sections along oblique lines,
Backfill 2.0 17.6 with a horizon included angle equal to 45°+ϕ/2, and
Silty sand 2.8 2-9 18.1 0 28 then they turn to vertical lines after passing the tunnel
Gravel/Sand >4.5 >100 20.1 0 35 6.7e3 crown’s level and finally arrive at the ground surface,
Silty sand >3.0 9-18 19.1 0 30 as illustrated in Figure 4.
Equation 4 in the modified models was derived
Note: γ = unit weight, c’ = cohesion, ϕ’ = friction angle, from the limit equilibrium of a horizontal slide (Ter
qu = uniaxial compression strength.
zaghi 1943):
Figure 3. Pipejacking activities at Drive D. Figure 4. Terzaghi arching model (after Terzaghi 1943).
14
Table 3. Parameters used in the modified models.
Silo width δ K c’
Parameters m ° kPa
Note: δ = friction angle in shear plane, α = 45°-ϕ/2, De = outer pipe diameter, Db = tunnel bore diameter, ϕ = soil friction
angle, c = soil cohesion, K0 = soil pressure ratio at rest.
Table 4. Soil properties for 17th pipe (after Zhang et al. 4.3 Effect of lubricant injection type
2016). Two drives (Drives C & D) were analysed because
of their good data completeness. The cumulative vol
Thickness γ c’ ϕ’
umes of injected lubricant for Drives C & D were
retrieved from Figures 2 and 3 and estimated relying
Parameters m kN/m3 kPa ° only upon lubricant injections into overcut annulus
of the analysed baseline section (Tables 6-7). For
Medium to coarse sand
with clay
9.3 17.6 0 30.2 instance, the analysed baseline section for Drive
Silty clay 17.6 18.3 10.6 6.9
C was 2-8 m (Section 1), 8-21 m (Section 2), 21
40 m (Section 3) and 40-75 m (Section 4). While the
15
Table 5. Summary of reduction in μ value against each Table 7. Summary of cumulative volume at Drive D.
baseline section.
RCP RCP RCP RCP RCP RCP RCP
favg τld σz μld Red. μ 02 09 19 29 38 48 58
Note: Each drive has four baseline sections and values Table 7. (cont’d). Summary of cumulative volume at
shown here are associated to the sections. favg = average Drive D.
jacking force, τld = frictional stress, σz = calculated normal
contact pressure, μld = backanalysed frictional coefficient, RCP RCP RCP
Red. μ = percentage reduction in μ value. 68 78 88
16
of the theoretical overcut annulus of 0.138 m3/m. times larger than 0.4 kPa of the 2-8 m section in the
The excessive injection volumes made the percent same gravel. While traversing through the 11
age reductions to reach to 88% (0.03 vs. 0.25), 84% 24 m section of gravel at Drive D, the jacking force
(0.04 vs. 0.25), 60% (0.1 vs. 0.25), and 97% (0.008 increased by 637 kN to 1617 kN (Figure 3). The
vs. 0.25), respectively. The enough buoyancy and increase of 637 kN reduced the percentage reduction
the lubricant-saturated overcut caused an 84% reduc to 84% from 98%, most likely because of the exag
tion from 0.25 suggested by Stein et al. (1989) for gerated frictional stress τld of 4.0 kPa, which is 5
clay-pipe interface to 0.04. The varying face resist times larger than 0.8 kPa of the 31-102 m section in
ance induced by jacking into the clayey gravel and the same clayey gravel. The main cause to lead to
the excessive pipe deviation were deemed to be the the reduction of 84% was attributed to the combined
main cause to lead to the reduction of 84%. The μld effects of misalignment and varying face resistance,
value equal to 0.04 was far less than the lower limit, resulting from tunnelling into the clayey gravel.
which also indicated adequate lubrication. It is worth
to note that jacking into gravel at 31 m distance
caused the reduction of 60% (the lowest in this 5 CONCLUSIONS
study). This phenomenon was most likely because of
the gravel not being long enough to develop lower The method that can be utilised for assessing the lubri
face resistance. The use of the excessive injection cation performance using the percentage reduction in
volumes accompanied with the justified buoyancy the frictional coefficient was described. The effect of
made the percentage reduction far less than 0.1 for lubricant injection mode, soil and lubricant natures and
the other two sections, which also indicated adequate misalignment on the lubrication performance was
lubrication. To short, the effect of misalignment and investigated. Some main conclusions can be drawn as
the varying face resistance contributed to the low follows:
percentage reductions despite the excessive injection
volumes. The occasional gravel led to the misleading (1) The excessive volumes of injected lubricant
reduction. The effects on the lubrication perform made the overcut full of lubricant and led to
ance outweighed the effect of injection type. enough buoyancy. Despite the excessive vol
umes, the significant pipe deviation and/or the
4.4 Effect of soil and lubricant natures varying face resistance had implications on the
percentage reduction in the frictional coefficient
The permeation of lubricant into the surrounding μ. The occasional gravel could lead to the mis
geology would mitigate the effort of establishing leading reduction. The said effects on the lubri
a lubricating layer at soil-pipe interface. This is most cation performance outweighed the effect of
likely because of the inability of the lubricant to injection type.
develop a filter cake of low permeability. Such (2) The excessive volumes of injected lubricant were
a permeable overcut could also result in injection either attributed to permeable ground or to the
volume in excess of the theoretical overcut. The phe inability of the injected lubricant to develop
nomena discovered in this study resulted in the rela a filter cake of low permeability. The highly vis
tively large injection volume of 0.552 m3/m at Drive cous lubrication with Marsh cone viscosity of 38
C (including mostly the gravel) than 0.534 m3/m at mins reduced the friction resistance to viscous
Drive D (including mostly the clayey gravel). On the resistance, leading to the reductions greater
other hand, the continuous injection of the lubricant than 88%.
with Marsh cone viscosity of 38 mins into the over- (3) The types of injection and the distribution charac
cut annulus effectively reduced the friction resist teristics of lubricant may vary between pipejack
ance to viscous resistance. This led the reductions in ing projects. Notwithstanding that, this study
excess of 88%, which is consistent with Staheli et al. provides an access of evaluating the lubrication
(2006). performance for pipejacking works and the pre
sented results would be useful in managing the
4.5 Effect of misalignment lubrication performance for upcoming pipejack
ing project.
The effect of misalignment significantly increased
the friction resistance and had implications on the
lubrication performance. There was an increase of
the jacking force of 771 kN while spanning between REFERENCES
8 and 21 m distance at Drive C (Figure 2). This
ASTM. 2011. F1962-11 Standard Guide for Use of Maxi-
increase in the jacking force caused the percentage Horizontal Directional Drilling for Placement of Poly
reduction in the μ value to reduce by 27% from ethylene Pipe or Conduit under Obstacles Including
98%. The main cause was not because of the overcut River Crossings. West Conshohocken: PA.
not full of lubricant, but because of the exaggerated British Standards. 2009. BS EN 1594-09 Gas Supply
frictional stress τld of 12.5 kPa induced by the pipe System-Pipelines for Maximum Operating Pressure
deviation in excess of 60 mm, which is almost 30 over 16 Bar-Functional Requirements. Brussels: UK.
17
Cui, Q.L., Xu, Y.S., Shen, S.L., Yin, Z.Y. & Staheli, K. 2006. Jacking Force Prediction: An Interface
Horpibulsuk, S. 2015. Field performance of concrete Friction Approach Based on Pipe Surface Roughness.
pipes during jacking in cemented sandy silt. Tunneling Ph.D. Thesis. Georgia Institute of Technology.
and Underground Space Technology 49: 336–344. Terzaghi, K. 1936. The shearing resistance of saturated
German ATV Rules and Standards. 1990. ATV-A 161 soils and the angle between the planes of shear. In Pro
E-90. Structural Calculation of Driven Pipes. Hennef: ceeding of the 1st International Conference on Soil
Germany. Mechanics and Foundation Engineering, June 1936.
Japan Microtunnelling Association (JMTA). 2013. Micro Cambridge: MA.
tunnelling Methods Serious II, Design, Construction Terzaghi, K. 1943. Theoretical Soil Mechanics. New York:
Management and Rudiments. Tokyo: Japan. USA.
Pellet-Beacour, A.L. & Kastner, R. 2002. Experimental and The Ministry of Construction of the People’s Republic of
analytical study of friction forces during microtunneling China (MOC). 2002. GB 50332-02 Structural Design
operations. Tunneling and Underground Space Technol Code for Pipeline of Water Supply and Waste Water
ogy 17(1): 83–97. Engineering. Beijing: China.
Stein, D., Möllers, K. & Bielecki, R. 1989. Microtunneling: Zhang, H., Zhang, P., Zhou, W., Dong, S. & Ma, B. 2016.
Installation and Renewal of Nonman-Size Supply and A new model to predict soil pressure acting on deep
Sewage Lines by the Trenchless Construction Method. buried jacked pipes. Tunneling and Underground Space
Berlin: Germany. Technology 60: 183–196.
18
Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9
Md Mizanur Rahman
University of South Australia, Mawson Lakes, Australia
ABSTRACT: A recent water leak incident taken place throughout the parallel tunnels LUO09 construction in
the soil alluvial deposits in Kaohsiung, Taiwan was analysed and discussed in this study. Dumping quick-set
cement was intended to ease the water leak incident but in vain. The water leak was initiated by the piping and
the associated ground loss caused two large surface cave-ins. The existing underpass caused the jet-grout col
umns installed not exactly in plumb. Their overlapping was estimated to be less than the design value of 60 cm,
developing seepage-prone weak zones. The hydraulic gradient being equal to 12.1 and existence of seepage-
prone weak zones were deemed as the main cause initiating the water leak incident. The pinhole test results
highlighted not only the nonplastic nature of the Kaohsiung silt but also its vulnerability to piping under large
hydraulic gradients. Some bullet points that indicate engineers should do or avoid were summarised.
DOI: 10.1201/9780429321559-3
19
Figure 2. Scenario for progressing the water leak incident.
20
(3) Diaphragm walls of 1.5 m in thickness to retain
the excavation-induced lateral loads were con
structed at the depth of 60 m;
(4) Dewatering wells screened at the bottom of the
tunnels were constructed to secure the excavation
face dry;
(5) The removal of the undermined vehicle under
pass was conducted upon completion of Region
“II” excavation;
(6) New tunnel liners were erected following com
pletion of Region “I” excavation;
(7) Controlled low strength materials were back
filled to the bottom of new vehicle underpass,
allowing for new vehicle underpass installation; Figure 5. Layout of monitoring instruments and locations
of surface cave-ins.
(8) Controlled low strength materials were back
filled to surface to resume traffic.
21
observation wells OW-S1, OW-S2, OW-04, and OW for SM-01 to 07. While the pit excavation appeared
10 gradually declined from 1.2 m/day in the begin to have minimal influence on the ground surface
ning of the recovery phase to 0.1-0.2 m/day in the settlements. It is worth to mention that when the
end of the recovery phase. Except the duration of ground was unfrozen prior to the installation of new
rainfall, the groundwater level in the observation well vehicle underpass, the ground surface settlements
OW-14 remained almost constant, indicating that the increased very quickly, especially for SM-01 to SM
group-well pumping appeared to have negligible 07. Despite the increases in the surface settlement,
influence on the groundwater level outside the walls, the surface settlements then soon reached a steady
which also indicated good watertight effectiveness. state condition.
Figure 7 shows the variations of the ground sur
face settlements while progressing the rehabilitation.
The surface settlement point SM-T1 measured the 5 ANALYSIS AND DISCUSSIONS
smallest settlement, compared with the surface
settlement points SM-T2, SM-T3, SM-T4, SM-A, 5.1 Triggering mechanism
SM-B, and SM-C. Similarly, the surface settlement
points SM-01, SM-02, SM-04, SM-06, and SM-07 The piping phenomenon was responsible for initiating
measured the relatively small settlements, compared the water leak incident, with reference to the testi
with the surface settlement points SM-03 and SM mony of workmen. Notwithstanding that, the initiation
05. This phenomenon is due to the fact that the indi of piping was conceived variously. The existing
cated surface settlement points were all close to the vehicle underpass, in fact, led to some difficulties in
short side of excavation pit that possesses larger installing the jet-grout columns exactly in plumb.
bending stiffness than the long side, exhibiting smal Thus, the overlapping of jet-grout columns was esti
ler lateral deflection and surface settlements. Despite mated to be less than the design value of 60 cm and
the nonnegligible contribution to the ground surface seepage-prone weak zones were thus developed. On
settlements from the effects of the diaphragm wall the other hand, as the groundwater level was at 5-
construction and the strut removal, the associated 6 m depths below the ground surface, the head differ
increases were typically 4.8-9.4 mm and ence between the groundwater level and the
5.5-7.5 mm, respectively, for SM-A to C and SM-T1 34 m depth where the piping initiated was 29 m. The
to T4 and 1-13 mm and 8.2-13.1 mm, respectively, length of shortest path for groundwater seepage was
from the bottom of the soilcrete body to the point of
water ingress and measured at about 2.4 m. This
came out with the hydraulic gradient i being equal to
about 12.1 (i=29 m/2.4 m=12.1). The critical
hydraulic gradient icr for triggering piping within gap-
graded sand-gravel mixture could be as low as 0.2-0.3
as compared to icr=0.9-1.0 for clean sands (Skempton
& Brogan 1994). In spite that this might not be true
for the Kaohsiung soil with the finer particle size
range, the soil particles could easily be detached or
washed away as subjected to such high hydraulic gra
dient i of 12.1. It is evident that the seepage-prone
weak zones allowed the water to flow through the jet-
grout columns and into the sump pit and that this high
hydraulic gradient made the water inflow even greater.
The combined effects of high hydraulic gradient and
existence of seepage-prone weak zones were deemed
as the main cause to initiate the water leak incident.
22
and consequent erodibility of clay soils as described that there is no necessity to perform the test through
in ASTM standard D4647-93 (ASTM 2006). There to 1020 mm of head. The purpose to further increase
were seven 33-mm diameter, 25-mm height speci the head is to classify dispersive soils, which is not
mens tested in the pinhole test for which their proper within the scope of this study. The effluent turbidity
ties are summarised in Table 2. The average grain and size of pinhole at the end of each test were
size D50 for the specimens was about 0.075 mm. recorded. If the effluent remains clear and the pinhole
A steel nipple was pushed into the specimen and the size unenlarged, then the soil is non-dispersive. Con
hole was punched through the nipple as a guide hole trarily, the effluent is turbid and the pinhole size is
using a 1-mm diameter steel needle. Wire screen and enlarged for dispersive soil.
pea gravel were placed in the pinhole cell on either Table 3 summarises the results of the pinhole test.
side of the specimen, as shown in Figure 8. As the For the specimens which were categorised as D, they
pinhole cell assembled, distilled water was allowed to were dispersive soils. There was also a common char
percolate through the specimen under constant heads acteristic. Most of the specimens primarily included
of 50, 180, 380, and 1020 mm, in accordance with silt with a fines content varying from 52 to 96% (Lu
ASTM standard D4647-06. The principal differenti et al. 2019) and their Unified Soil Classification
ation between dispersive and non-dispersive soils, System (USCS) symbols were ML and SM-ML. The
however, is given by the test results under 50 mm of pinhole size enlarged in a range of 2-4 mm after the
head, as suggested by Sherard et al. (1976), indicating test. The specimen S-19 categorised as ND was non-
dispersive soil with a fines content of 94% and
mainly consisted of clay. The USCS classified this
specimen as CL, which was an indication of low plas
Table 2. Summary of soil properties in the pinhole test. ticity clay. The pinhole size for the specimen S-19
remained unenlarged after the test. The test results
Sand Silt Clay Water content LL PI
revealed a fact indicative of the tendency of the silt
from Kaohsiung, Taiwan to possess high dispersion,
Specimen % % % % % or very low strength, or both. The particle-size distri
bution analysis indicated that the Kaohsiung soil
S-14 19 76 5 22.4 27 4.1
belonged neither to the gap-graded gravelly sands sus
S-15 48 52 0 25.6 21.4 0.8
ceptible for segregation piping nor to the dispersive
S-16 15 82 3 24.4 27.5 1.2
clay prone to internal erosion. In fact, the particle size
S-17 4 86 10 19.5 35.2 6.4
for the Kaohsiung soil lay midway in the range of silt
S-18 20 74 6 21.7 26.5 3.3
and sand. The pinhole tests were regarded as an
S-19 6 56 38 42.9 43.9 20
effective means to identify not only the nonplastic
S-20 15 84 1 25.1 26.1 1.2
nature of the Kaohsiung silt contained in the sand but
Note: LL = liquid limit, PI = plasticity index. also its high vulnerability to piping or internal ero
sion. It may be conclusively mentioned that the seep-
age-prone weak zones and high hydraulic gradient
initiated the water leak incident and that water from
the two undermined water mains transferred the Kao
hsiung silt into more flowable debris and its nonplas
tic nature aggravated the collapse even further.
Despite an apparent inadequacy in dealing with simi
lar incident in other sites, the local sandy silt was
found to be prone to the piping at deep depths under
large hydraulic gradients and the two undermined
water mains enhanced its flowable nature, thereby
enlarging the scale of the water leak incident.
23
Table 3. Summary of the pinhole test results.
Final head Colour cloudiness Pinhole size Specific surface area Dispersion classification
Specimen mm mm m2/g
Note: Pinhole size = Pinhole size measured after the test, D = dispersive soil, ND = non-dispersive soil.
grouting should be performed prior to the seepage was from the bottom of the soilcrete
formal one to verify the design grouting body to the point of water ingress; that is,
parameters. 2.4 m, corresponding to the hydraulic gradi
(2) Seepage-prone weak zone: Except for trial ent i of 12.1.
grouting, additional soilcrete columns, result (3) The existing vehicle underpass led to some dif
ing from chemical grouting, should be con ficulties in installing the jet-grout columns
sidered during design phase and constructed exactly in plumb and some seepage-prone weak
next to jet-grout columns, which not only zones were thus developed. The seepage-prone
increases the length of path for the ground weak zones allowed the water to flow through
water seepage, but also prevents formation of the jet-grout columns into the sump pit and
the seepage-prone weak zone. such high hydraulic gradient made the water
(3) Local soil nature: The Kaohsiung silt contained inflow even greater. The combined effects were
in the sand has proved to possess high disper deemed as the main cause initiating the water
sion, or low strength, or both. There is leak incident.
a necessity to investigate the nature of local (4) The results of the pinhole test identified not
soils while progressing upcoming subway con only the nonplastic nature of the Kaohsiung silt
struction project if they are found to exhibit but also its vulnerability to piping or internal
peculiar behaviour differentiated from common erosion under large hydraulic gradients. Water
sense in soil mechanics. from the two undermined water mains and the
nonplastic nature of the Kaohsiung silt aggra
vated the collapse even further. More detailed
investigation is considered to be necessary to
6 CONCLUSIONS
address the explored issue with different aspects
of view.
The main cause to lead to the water leak incident
was investigated. The pinhole test results highlighted
not only the nonplastic nature of the Kaohsiung silt
but also its vulnerability to piping under large REFERENCES
hydraulic gradients. Some main conclusions can be
drawn as follows: ASTM. 2006. D4647-93 Standard Test Method for Identifi
cation and Classification of Dispersive Clay Soils by the
(1) The piping was responsible for initiating the Pinhole Test. West Conshohocken: PA.
water leak incident. Dumping sand bags and Bell, F.G. & Walker, D.J.H. 2000. A further examination of
quick-set cement failed to ease the water leak. the nature of dispersive soils in Natal, South Africa.
Mud water carried away fine particles in the Quarterly Journal of Engineering Geology and Hydro-
geology 33: 187–199.
ground and together flowed into the tunnels Chen, R.P., Li, Z.C., Chen, Y.M., Ou, C.Y., Hu, Q. &
through the ripped liner joints leading to serious Rao, M. 2015. Failure investigation at a collapsed deep
ground loss. Such ground loss caused two sur excavation in very sensitive organic soft clay. Journal of
face cave-ins. The surface cave-ins not only Performance of Constructed Facilities 29(3): 04014078.
impeded traffic, but also caused damages to Hou, Y.J., Fang, Q., Zhang, D.L. & Wong, L. 2015. Exca
adjacent properties. vation failure due to pipeline damage during shallow
(2) The hydraulic pressure at the point of water tunnelling in soft ground. Tunnelling and Underground
ingress was estimated to be about 300 kPa. Space Technology 46: 76–84.
Jo, Y.S., Cho, S.H. & Jang, Y.S. 2016. Field investigation
The length of shortest path for groundwater
and analysis of ground sinking development in
24
a metropolitan city, Seoul, Korea. Environmental Earth Sherard, J.L., Dunnigan, L.P. & Decker, R.S. 1977. Some
Sciences 75: 1353. engineering problems with dispersive clays. In:
Lu, J., Wang, T.H., Cheng, W.C., Yang, T. & Luo, Y. 2019. Sherad, J.L. & Decker, R.S. (eds.), ASTM STP 623; Pro
Permeability anisotropy of loess under influence of dry ceedings of Symposium on Dispersive Clays, Related
density and freeze-thaw cycles. International Journal of Piping, and Erosion in Geotechnical Projects, Chicago,
Geomechanics 19(9): 04019103. 27 June–2 July 1976.
Sherard, J.L., Steele, E.F., Decker, R.S. & Dunnigan, L.P. Tan, Y. & Lu, Y. 2017. Forensic diagnosis of a leaking acci
1976. Pinhole test for identifying dispersive soils. ASCE dent during excavation. Journal of Performance of Con
Journal of Geotechnical Engineering Division 102: 69–85. structed Facilities 31(5): 04017061.
25
Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9
W. Bilfinger
Vecttor Projetos Ltda., São Paulo, Brazil
H.C. Rocha
Companhia do Metropolitano de São Paulo - Metrô-SP, São Paulo, Brazil
ABSTRACT: Major tunneling works in the city of São Paulo have started in the 1970´s, with the construc
tion of the 1st Metro Line. Since then, tunnels with different cross sections have been constructed regularly,
using either mechanized or conventional tunneling methods. From a geological point of view, the city is
mainly located in a sedimentary basin, with its subsoil partly composed by residual soils overlaying bedrock.
Experience has shown that the induced settlements in the residual soils are higher and more variable. Historic
registers of tunneling induced settlements measured for different constructive techniques and soil types are
presented on this paper. Focus is given to the measurements during the construction of Metro Line 4, where
approximately 5 km of line tunnels were excavated using NATM in residual soils, saprolite and rocks, and
around 8 km were excavated in tertiary sediments, using an EBM TBM. Typical values for volume losses on
this context are proposed.
DOI: 10.1201/9780429321559-4
26
Another random document with
no related content on Scribd:
freqüent pero enérgica y brusca. En los segons, la acció del Estat pert
en intensitat lo que guanya en extensió. Es que la lluyta s’ha
generalisat entre tots los membres á la mida de la generalisació y
aument de las necessitats, perdent al meteix temps lo carácter de
guerra oberta pera pendre lo de la competencia. La idea abstracta de
justicia, traduhida en dret positiu al punt que la defineix una
agregació social, está subjecta á la lley de evolució que presideix al
desenrotllo aixís de la vida moral com de la vida material. Per aixó,
en un Estat avansat en cultura, la lley ha de ser complicada, fundada
en sos preceptes, y suau en la sanció.
Lo fins aquí indicat basta pera demostrar que la missió altíssima del
Estat no es pas suprimir la lluyta, sinó regularla, al objecte de que sos
resultats siguin civilisadors y progressius. Aquestos están en relació
directa de la activitat y energía de la lluyta, per qual motiu l’Estat,
lluny de restringirla ó contrariarla, ha de procurar per tots sos medis
fomentarla y exténdrela. Per medi de la direcció de las relacions
exteriors, ha de darli camp franch; per medi de la definició, fixació y
aplicació del dret, ha d’evitar que degeneri en batalla, mantenintla
dins dels límits de la competencia; per medi de las atribucions que se
li concedeixen pera la producció y aument de la cultura, ha de pendre
en ella part activa, provocantla, avivantla y extenentla á tots los
terrenos civilisadors.
Los procediments son sempre fills del método, de manera que pot
ben assegurarse que sent aquells particularistas, particularista ha de
ser tambe aquest. Y en efecte, lo método científich tendeix avuy
directament cap al particularisme. La observació directa ha vingut á
ser la base de las ciencias, aixis de las que buscan las lleys dels sers
materials, com de las que estudian lo món metafísich. Las hipótesis
sens fonament real en fets demostrats no captivan avuy las
intel·ligencias. Los llibres de filosofía en que l’autor no hi va vuidar
més que’ls productes d’una admirable gimnástica de cervell, han
perdut quasi tota la autoritat de que han gosat en altras épocas. Tals
esforsos d’intel·ligencia admiran pero no convensen.
Tots los sistemas filosófichs que fins ara ha concebut l’home y tots
los que concebirá mentres sols disposi de las facultats que
actualment lo posan en relació ab los móns exteriors, poden ben bé
calificarse en dos grans grupos: idealisme y positivisme. Quan se
precindeix de la realitat, y las qüestions se resolen en virtut de
teorías preconcebudas y fillas de generalisacions autoritarias, los
resultats han de ser purament convencionals, y s’imposan més á la
imaginació que á la rahó. Quan s’atén á la realitat y no se avansa un
pás que no pugui fundarse en la observació directa y comprobada, las
conseqüencias, menys brillants que sólidas, s’imposan á la rahó més
que á la imaginació. Los sistemas idealistas produheixen entusiasme,
pero llurs efectes per lo meteix que son molt intensos, no tardan á
decaure. Los sistemas positivistas no arrivan quasi jamay á
entusiasmar, pero en cambi son d’efectes mes durables, puig que
satisfán á la rahó y donan convenciment. Uns y altres sistemas tenen
un gran perill en l’exclusivisme. L’idealisme arriva moltas vegadas á
prescindir completament dels fets, aixis com lo positivisme
precindeix moltas altras de las abstraccions. Reduhintnos als
sistemas basats en la observació, afegirem que sols es llegítim lo
positivisme, quan pera sas resolucions té en compte totas las
impressions de que son capassas las variadas facultats del home,
aixis las reflexivas com las imaginativas.