Muirwood Janvier 2019 BD Page A Page

Muir Wood Lecture 2018
Monitoring and Controlling Ground Behavior at the Source

Recent Applications to Pressurized Tunneling
Edward J. Cording
N° ISBN : 978-2-9701122-4-2 aPril 2018

Muir Wood lecture 2018 - Monitoring and Controlling Ground Behavior at the Source - Recent Applications to Pressurized Tunneling
N°ISBN : 978-2-9701122-4-2 / april 2018 - Layout : Longrine – Avignon – France – www.longrine.fr
The International Tunnelling and Underground Space Association/Association Internationale des Tunnels et de l’Espace Souterrain (ITA/AITES) publishes this report to, in accordance with its
statutes, facilitate the exchange of information, in order: to encourage planning of the subsurface for the benefit of the public, environment and sustainable development to promote advances
in planning, design, construction, maintenance and safety of tunnels and underground space, by bringing together information thereon and by studying questions related thereto. This report
has been prepared by professionals with expertise within the actual subjects. The opinions and statements are based on sources believed to be reliable and in good faith. However, ITA/AITES
accepts no responsibility or liability whatsoever with regard to the material published in this report. This material is: information of a general nature only which is not intended to address the specific
circumstances of any particular individual or entity; not necessarily comprehensive, complete, accurate or up to date; This material is not professional or legal advice (if you need specific advice,
you should always consult a suitably qualified professional).
Muir Wood Lecture 2018
Monitoring and Controlling Ground Behavior at the Source

Recent Applications to Pressurized Tunneling
Edward J. Cording
Professor Emeritus of Civil and Environmental Engineering,
University of Illinois at Urbana-Champaign
>> Table of contents
Abstract...............................................................................................................................................5
1. Introduction...................................................................................................................................6
2. SHIELD- DRIVEN TUNNELS................................................................................................................7
2.1 THE THAMES TUNNEL 1825 TO 1843: WHERE SHIELD TUNNELING BEGAN...................................7
2.2 TUNNELING WITH OPEN FACE SHIELDS....................................................................................7
2.3 GROUND MODIFICATION, REINFORCEMENT, AND REPLACEMENT FOR OPEN SHIELDS AND
PRESSURIZED FACE TBMS.......................................................................................................8
2.4 PIONEERING INVESTIGATIONS CORRELATING THE SOURCES OF GROUND MOVEMENT WITH
SURFACE SETTLEMENT, CHICAGO: 1939-1941...........................................................................9
2.5 GROUND MOVEMENTS DURING SHIELD TUNNELING ON WASHINGTON D.C. METRO:
1970 TO 1974.............................................................................................................. 10
2.6 SHIELD TUNNELING IN THE CHICAGO CLAY: 2000....................................................................11
3. PRESSURIZED TUNNELING.............................................................................................................13
3.1 PRESSURIZED FACE..............................................................................................................13
3.2 CONTROL OF REGULAR GROUND LOSS BY FILLING AND PRESSURIZING GAPS.........................14
3.3 EPBM TUNNELING AT CAPITOL HILL, SOUND TRANSIT U230 PROJECT: 2011..............................15
3.4 EPBM TUNNELING BENEATH SCHULICH BUILDING, TORONTO: 2012.........................................16
3.5 GROUND CONTROL ON LOS ANGELES METRO TUNNEL PROJECTS..........................................19
4. LARGE DIAMETER PRESSURIZED TBMS AND THE ALASKAN WAY VIADUCT REPLACEMENT
TUNNEL..............................................................................................................................................21
4.1. INTRODUCTION )..................................................................................................................21
4.2. SITE CONDITIONS ON THE ALASKAN WAY VIADUCT REPLACEMENT PROJECT...........................21
4.3. ASSUMED VOLUME LOSS VS ACTUAL SURFACE SETTLEMENT.................................................22
4.4. INNOVATION AND RISK.........................................................................................................23
4.5. COORDINATION PLAN...........................................................................................................24
4.6. TBM ADVANCE BENEATH ALASKAN WAY VIADUCT AND PIONEER SQUARE ...............................27
4.7. TBM ADVANCE WITH INCREASING DEPTH TO BNSF TUNNEL....................................................28
4.8. SETTLEMENT IN SANDS AT NORTH END OF TUNNEL...............................................................29
4.9. SETTLEMENT DUE TO DIFFERENTIAL PRESSURE ....................................................................30
5. CONCLUSIONS..................................................................................................................................31
COORDINATION.........................................................................................................................31
MONITORING GROUND BEHAVIOR AT THE SOURCE .....................................................................31
PREVENTION OF GROUND LOSS.................................................................................................32
6. ACKNOWLEDGEMENTS...................................................................................................................33
7. REFERENCES.....................................................................................................................................34
4 Muir Wood Lecture 2018

>> Abstract
Presented are investigations and guidelines for monitoring, and controlling The benefits have been most dramatic in their application to large
pressurized tunnel boring machines (TBMs) to prevent ground loss, and diameter TBMs and to TBMs driven at shallow depth in urban areas. The
damaging settlement to structures. The process begins in the planning 17.5-m-diameter EPBM selected for the recently-completed tunneling
stages with the owner’s team and continues during construction with for the Alaskan Way Viaduct Replacement project for State Route 99 in
a coordinated program of managing, operating and monitoring the Seattle was the largest pressurized TBM to be driven beneath an urban
tunneling process. area. To assess and mitigate risks, the owner and contractor teams built
on previous experience with large diameter pressurized face TBMs,
Monitoring of ground movements at their source is at the heart of the including tunnels in Porto (Portugal), Barcelona, and Madrid driven with
observational method for geotechnical projects. In large underground 9- to 15-m-diameter earth pressure balance machines (EPBMs). Ground
rock caverns and on slopes, borehole extensometers and inclinometers improvement and reinforcement methods were implemented, but the
have been used for over 60 years to determine the depth of zones of primary and most effective ground control measures were in planning
movements and locate the geologic features affecting stability. In urban and executing the TBM operation. Continuous pressurization of the TBM
tunneling and excavation, these instruments are used to locate the face, shield steering gap, and lining gap at the shield tail prevented ground
sources of ground movement that could cause damaging settlement. loss and prevented damaging settlement throughout the drive beneath
For pressurized TBMs, such observations, made close to the advancing Seattle structures.
TBM, are key to understanding and controlling ground behavior.
Essentially, such a tunneling operation achieves the objective described
Examples are provided of earlier, open face shield tunneling where ground in an 1818 patent application for a tunnel shield by Marc Isambard
behavior could often be observed directly. The process often relied on Brunel of “opening… the ground in such a manner that no more earth
the ability of the ground to stand unsupported, and the operator was shall be displaced than is to be filled by the shell or body of the tunnel”
provided with equipment with which he could not always be successful. (Muir Wood, 1994, Skempton and Chrimes, 1994). For the 17.5-m
EPBM, the settlements immediately above the shield were smallest
The tunneling industry has witnessed a revolution. Pressurized face TBMs- when the tunnel was at shallow depth, and there was no surface
- slurry balance machines and earth pressure balance machines (EPBMs) settlement. The results were the opposite of calculations based
--- have enabled tunneling at greater depths under waterways and at on an assumed percentage of ground loss, which show the largest
shallower depths in urban areas without damaging settlement. Advances settlements and potential damage to structures occur when the tunnel
in recent years have included the increasing capability to coordinate and is at shallow depth. That can lead to prescribing ground modification
control the TBM operation and monitor key machine functions and their measures or adjusting the alignment to satisfy the assumption, rather
impact on the surrounding environment in real time as the TBM advances. than placing primary emphasis on preventing the ground loss by
Maintaining and monitoring a continuously pressurized envelope of controlling the tunneling operation. Recent progress has made reliable
conditioned muck and injected slurries around the face and body of the prevention of ground loss possible with the real time linkage of TBM
shield, and grout around the lining at the tail of the shield have resulted operating parameters and geotechnical observations.
in improved and consistent control of ground movements throughout the It is of critical importance that there be a coordinated effort among
tunnel drive. TBM operation and geotechnical monitoring, supervision, engineering,
data management, safety, and construction management to assess
In this paper, real time records of key TBM operating parameters and mitigate risks, and monitor and control the tunneling operation.
are coupled with real time observations of ground movements and Examples of such efforts are provided. Consistently controlling
porewater pressures immediately around the advancing TBM. Borehole pressures and minimizing settlement throughout the tunnel drive serve
extensometers, piezometers, and directionally drilled horizontal as a demonstration to project participants and the community alike that
inclinometers are used to pin-point the sources of ground movement structures along the tunnel alignment will be protected
and groundwater changes around the TBM to aid in making adjustments
and confirming that ground control is being achieved. The results of such
observations are described for both open face shields and pressurized
face TBMs on tunneling projects in Washington, D.C, Chicago, Toronto,
Seattle and Los Angeles.
Monitoring and Controlling Ground Behavior at the Source - Recent Applications to Pressurized Tunneling 5
1 >> Introduction
Two hundred years ago, Marc Isambard Brunel submitted a patent Close monitoring of the sources of ground movement on pressurized
application, for a tunnel shield with the objective of “opening… the TBM projects will lead to better understanding of their capabilities and the
ground in such a manner that no more earth shall be displaced than is to procedures for consistent operation to minimize risk and prevent damage.
be filled by the shell or body of the tunnel” (Muir Wood, 1994, Skempton Consistently controlling pressures and minimizing settlement throughout
and Chrimes, 1994). Efforts to achieve that goal with open-face shields the tunnel drive serve as a demonstration to project participants and
are described in Section 2. the community alike that structures along the tunnel alignment will be
protected.
Examples of monitoring and controlling ground behavior at the source
include the pioneering observations by Karl Terzaghi and Ralph Peck in
1940 in liner plate tunnels in the soft Chicago Clay. From 1995 to 2000
the Chicago clay was revisited during driving of a 3.6-m open shield with
rotating cutterhead. Sources of ground behavior were located using
extensometers, inclinometers and piezometers. The results showed that
the settlement was primarily caused by loss of ground into the unfilled
overcut gap and by time-dependent consolidation of the clay due to the
stress changes indicated by the piezometers, conditions that are relevant
in pressurized tunneling today.
Current pressurized tunnel boring machines (TBMs), described in Section

3, are capable of controlling not only the stability of the face but also
preventing ground loss into the gaps around the shield and tail of the
advancing TBM. EPBMs in Toronto, and Los Angeles were advanced at
shallow depth in granular soils by filling and pressurizing the gaps around
the shield without using the ground replacement procedures such as
compensation grouting.
Projects with large diameter TBMs, described in Section 4, of necessity,

have led in the development of monitoring and ground control procedures,
as shown by EPBM tunneling on Porto Metro, Portugal, and Barcelona’s
Line 9, and more recently on the 17.5-m-diameter EBPM on the Alaskan
Way Viaduct Replacement Project in Seattle. The developments in Porto
were two-fold: (1) operation of the TBM included injection of bentonite to
positively fill and pressurize the overcut gap around the TBM and injection
to maintain pressures in the cutterhead chamber above groundwater
levels between advances and (2) development of a coordinated program
of managing, operating and monitoring the tunneling process to control
ground movements.
On the Alaskan Way Viaduct Replacement tunnel, monitoring of ground

behavior at the source was conducted with combination extensometers/
piezometers at an average spacing of 16 m, which provided an almost
continuous view of the effect of TBM face and shield gap pressures
on ground displacement and changes in groundwater pressures.
Measurements made in both clays and sands, at depths ranging from 10
to 60 meters, confirmed that the filling and pressurization of the gaps were
preventing ground loss. Residual displacements around the advancing
TBM were related to the stress changes due to the differential between
overburden and face/shield pressures.

2 >> Shield-driven tunnels
2.1 THE THAMES TUNNEL 1825 TO 1843 : WHERE SHIELD

TUNNELING BEGAN
Marc Isambard Brunel designed, financed, built, and drove the first
subaqueous shield under the Thames River, recovered from multiple
irruptions of the river into the works, rebuilt the shield under the river,
refinanced the project, was knighted in 1841, and finished in 1843.
The fascinating story is well known to us because of research and
technical papers prepared by Skempton and Chrimes (1994) and by
Muir Wood (1994), the title of which is copied as the heading for this
paragraph.
Such papers are more than history, they record the engineering/
tunneling precedents on which we build. They document the
a.Thames RIver Tunnel Shield.
observations of how ground affects tunneling and how tunneling
methods control ground behavior. The papers describe how the
thinner than expected cover of London Clay led to collapse and
flooding of the tunnel. The ground loss leading to this collapse began
with flowing on lenses of sand and silt in the tunnel face.
The tunnel was large. It was excavated as a single 12-m-wide box

in which a brick-lined double carriage-way was erected under the
protection of multiple roof shields, each having three levels with
a pocket at the face in which a miner would work, incrementally
removing a breast board at a time excavating and setting the board
forward a few tens of cm, bracing it with a screw jack against the
shield frame, then proceeding downward to the next board (Figure b. New York City East River Shield.
1a). It was the classic method of breasting in ground with short
stand-up time, used in open face shields for the past 175 years.
The tunnel is in use today on the London Underground.
2.2 TUNNELING WITH OPEN FACE SHIELDS
Barlow designed a 2.5-m circular shield, which Greathead drove

under the Thames River in 1869. In 1886, Greathead drove twin
3-m-diameter circular shields using compressed air and cast-iron
rings that became the standard for open-face hand-mined shields.
Open face shields were capable of supporting the face with breast c. Open Shield, 1950’s.
boards supported by screw jacks, and later hydraulic jacks bearing
against the frame of the shield so that the shield could be advanced
while maintaining support of the breast boards by retracting the jacks
(Figure 1c). Some open shields were built with shelves (or tables)
so that the excavated soil would form multiple angles of repose to
support the face, such as Figure 1b, a compressed air shield under
the East River in New York City for Pennsylvania RR. (Noble, 1910,
Trans. ASCE).
Efforts to improve productivity led to mechanization in the 20th

century. Digger shields -- open face shields with mechanical
excavators -- typically have a pan in the bottom to allow the muck
to form an angle of repose as the shield is pushed forward, thus
supporting the lower portion of the face. Figure 1d shows a digger
shield with a muck pan designed to provide an angle of repose, and
d. Digger Shield, 1970’s.
Figure 1. Open Face Shields.
a digger and stinger that could be penetrated into the face to loosen the shield advance, dropping a lean mix of cement and sand to fill the
the soils so that the shield could push forward in hard ground and fill voids as they rose to an overlying caliche layer before they could break
the pan. Breast tables and breast flaps were used in the upper face through and reach the surface.
of mechanized shields. Rotating breast flaps such as those shown
for the digger shield in Figure 1d were most effective in supporting In the 1980s, on Sound Transit twin bus/transit tunnels driven in downtown
the upper face during standstill. Seattle with digger shields in glacially overridden tills and outwash sands
and gravels, ground loss at the face occurred at several locations when
Shields with a rotating cutterhead were developed. Lovat shields the shield encountered raveling and running sands, and was of most
had guillotine (flood) doors on the cutterhead that could be closed concern where the tunnel turned corners and passed beneath buildings.
down to prevent inflows and provide access to the muck chamber Backfill grout and compaction grout were injected through holed drilled
behind the cutterhead between advances of the shield (Section 2.3) from the surface, from basements, and from the tunnel to fill the voids and
compact the loosened zones (Robinson, et. al. (1991).
Often, reliance was placed on the stand-up time of the ground to
prevent ground loss into the face of the shields. Stand-up time is On two projects driven in alluvial raveling and running sands, one with a
reduced when the shield is pushed forward due to vibration and digger shield with limited ability to provide an angle of repose in the pan,
shear forces imposed on the ground. and the other with a shield with a rotating cutterhead, chemical grout was
injected from the surface through a pattern of tube-a-manchette pipes
In flowing, running and raveling ground, the operator could not always (TAMs) over long reaches of the tunnel to increase stand up time and
be successful in preventing large ground loss with the shield equipment prevent large ground loss into the tunnel face as the shield passed.
provided. On a project in alluvial sands and clays in Washington, D.C., On another project in Los Angeles, where the tunnel turned 90 degrees
I observed over two shifts the operation of a shield with a rotating from one street to another and passed at shallow depth below building
cutterhead being advanced in dry sands. On the first shift, the operator spread foundations, chemical grout was injected above the tunnel and
balanced the cutterhead rotation with a steady thrust and shield below building footings through holes drilled from the tunnel 25 meters
advance, minimizing ground loss by easing his way through the ground. forward of the shield.
On the second shift, the operator tended to thrust the rams and crowd
the cutterhead, which became torque limited, and then back off on the Extensive permeation grouting or backfilling of voids over long reaches of
thrust, rotate the cutterhead with little forward movement of the thrust a tunnel is an obvious condition where the tunneling method does not fit
rams, causing a night of work to fill the void created above the shield. well with the ground conditions.
Urban tunneling with these shields, such as in the alluvial sand, silt and Pressurized face TBMs have largely eliminated such conditions
clay on the Washington, D.C. Metro in the 1970s and 1980s, was usually Fully pressurized TBMs also prevent ground loss into the gaps
conducted with dewatering wells spaced on 30 - to 50 - m centers, around the body of the TBM shield and reduce reliance on permeation
although inflow of perched water on contacts was a common problem grouting or compensation grouting methods that would otherwise be
and difficult to adequately dewater. Low compressed air pressures (in the required to prevent structure settlements from exceeding allowable
range of 0.3 bars to 0.6 bars) were occasionally used. levels. Fully pressurized TBM tunneling, such as cases described in
Sections 3 and 4, is capable of controlling settlements to values less
2.3 GROUND MODIFICATION, REINFORCEMENT, AND than those achievable with compensation grouting.
REPLACEMENT FOR OPEN SHIELDS AND PRESSURIZED
FACE TBMs. At the same time, ground improvement measures are an important
tool for pressurized TBMs, as well as for sequential excavation
With open shields, in the absence of adequate stand up time in running methods. Decisions are made to use these additional procedures to
and raveling soils, it was necessary in a number of cases to use ground reduce risk and maintain stability. Ground improvement procedures
improvement or backfilling of voids over significant lengths of the tunnel are most likely to be used where the TBM cannot be pressurized, for
drive. Several examples follow. example, at launch and exit when the TBM shield is not fully buried
in the ground. They may be used at start-up in the vicinity of critical
In Phoenix, Arizona, in the 1980s, a digger shield was driven in dry structures where there has been no monitoring to establish a record
alluvial sand-gravel-cobbles that would stand in the face when the of the ground control and there is uncertainty as to the capabilities of
shield was not advancing, but would ravel and run as the shield was the TBM operation to mobilize, pre-check, and test systems before
pushed forward. Difficulty with penetrating the cobbles to form an angle the start. Ground modification may be used in soil or rock with
of repose in the face contributed to the large volume losses and voids open voids where conditioner and slurries cannot be contained and
that formed above the shield in the first drive, which was largely beneath pressurized.
open fields. Prior to the second drive beneath a street in downtown
Phoenix, the decision was made to drill closely spaced holes over the
length of the drive, prior to tunneling. A backfilling operation followed

Ground modification procedures can be used to facilitate interventions In the sequentially excavated liner plate tunnels, Peck’s team conducted
into the cutterhead chamber for repair and maintenance. Cutterheads a series of Squeeze Tests over 24- to 48-hour periods, observing and
may be driven into safe havens constructed of secant walls or jet recording the sequence of excavation and support of the heading and
grout walls for interventions in compressed or free air. Emergency bench, sampling soils and measuring the displacement of rods driven
access under free air to repair a damaged cutterhead was achieved into the clay ahead of face and in side walls and arch as the excavation
on the Port Mann Water tunnel project by ground freezing. The TBM took place.
was at a depth of 60 m in granular tills, below the Fraser River in
the metropolitan area of Vancouver, British Columbia. The contractor, Surface settlements using the excavation sequence shown in Figure 2
McNally/Aecon engaged Moretrench, who drilled and conducted the were 100 mm. The Squeeze Tests measurements revealed that much
freezing operation from a barge. of the ground loss was occurring due to settlement of the steel ribs and
liner plate arch as the bench was excavated and posts were placed
There are many other examples. Georgios Anagnostou in his beneath the arch support.
2014 Muir Wood Lecture “Some Critical Aspects of Subaqueous
Tunneling” describes a case on Zurich Cross-rail where large-
diameter pipe arches were used to stabilize the ground before driving
an 11.3-m-dia slurry shield beneath a structure adjacent to the river,
where, because of the shallow, 9-m-depth below the river bottom,
pressures required to limit building settlement were close to the total
water/soil overburden pressure for the portion of the shield beneath
the riverbed. He also describes a large-scale system of porewater
pressure relief wells below the seabed on the Storebaelt tunnel to
reduce groundwater pressures to levels that could be handled by
pressurization of the face.
In urban tunneling, ground control procedures need to extend to the

drilling of holes for ground improvement, as well as to the tunneling.
It has been observed on several recent projects that the ground
movements caused by the controlled pressurized TBM operation
Figure 2 : Chicago Subway: 1939-1941: Observations in the liner plate tunnel.
were well below those induced by the drilling of multiple holes for
ground improvement when drilling techniques were not properly
controlled and fitted to the ground conditions. An example where On subsequent tunnel contracts, surface settlements were reduced to
ground movements were controlled during drilling of compensation 50 mm by placing wall plate I-beams at the base of the arch to support
grout pipes, as well as being controlled during EBPM tunneling, is it longitudinally as the bench was excavated beneath the arch. Two
described in Section 3.5 for the Regional Connector project in Los small tunnels (“monkey drifts”) were excavated ahead of the top heading
Angeles. in order to install the I-beams before the arch support was placed
(Terzaghi, 1942a)
2.4 PIONEERING INVESTIGATIONS CORRELATING THE • On one of the later tunnel contracts, surface settlements again
SOURCES OF GROUND MOVEMENT WITH SURFACE increased to 100 mm as the tunnel depth increased from 12 to 18 m in
SETTLEMENT, CHICAGO: 1939-1941 order to pass below the Chicago River. In one of the early examples of
Peck’s use of the observational method, he prepared a Squeeze Test
During construction of the Chicago Subway, Karl Terzaghi, Consultant report describing his observations of the construction events affecting
to the City of Chicago, and Ralph Peck, Assistant Subway Engineer, the excessive ground loss. (Cording, 2013). Filling of the gap between
selected by Terzaghi to supervise the soil mechanics laboratory, the clay and liner plate was being delayed several advances behind
conducted some of the earliest investigations of the relationship the top heading excavation. Without any filling, the only support to
of tunnel construction to surface settlements by measuring the prevent the clay from squeezing was the compressed air pressure,
displacement of the clay into the tunnel and observing and recording which could not be increased above the 1 bar maximum to balance
construction conditions. When the tunnel bottom was in soft clays the higher overburden stresses. Settlements were reduced from 100
in the downtown Chicago “Loop”, shields were used with fixed to 50 mm by minimizing and promptly filling the gap when the lining
openings at the front through which the clay squeezed (Terzaghi, was installed, and by reducing the time and distance of the bench
1942b). When the bottom of the tunnel was in stiffer clays, to the excavations and support installation beneath the top heading.
north of the Chicago River, the liner plate method was used. It was a
sequential heading and bench excavation method with compressed
air pressures less than 1 bar (Terzaghi, 1942a). A cast concrete lining
followed approximately 15 m behind the excavation.
Figure 3 : Filling gaps with pea gravel between the excavated clay surface and the liner plate: Large gaps and delays in filling were a major source of ground loss.
Although the Squeeze Tests were for a sequential excavation method, movements around tunnels and excavations. The instrumentation
the approach of assessing the cause of surface settlements by allowed determination of the three-dimensional pattern of ground
correlating ground loss into the tunnel with construction events was movements extending from the source of movements around the tunnel
a forerunner of the process of controlling ground movements at their shield, distributed through the soil mass to the surface (Cording and
source around pressurized TBMs. Hansmire, 1975, Hansmire and Cording, 1985).
Summary: On the first tunnel drive, the level surveys showed large surface
settlements of 150 mm. Without the observations of ground movement
• The Squeeze Test was a concentrated period of measuring ground and the shield operation, there could have been uncertainty as to
loss at the source, and observing the construction and soil conditions source, perhaps the assomption that the ground loss was occurring
affecting ground loss and surface settlement. into the tunnel face.
• All the data was summarized on a single blueprint.
• Delayed filling of gaps was a major cause of ground loss and settlement. However, the inclinometer located ahead of the face showed only 6 mm
• The observations identified the source of ground loss so that of lateral displacement toward the tunnel face (Figure 4).
construction procedures could be corrected.
2.5 GROUND MOVEMENTS DURING SHIELD TUNNELING ON

WASHINGTON D.C. METRO: 1970 TO 1974
The University of Illinois, under contract to Washington Metro, conducted

a monitoring program on Phase 1 construction of braced excavations,
tunnel and station caverns in rock, and a shield tunnel driven in soil.
For the digger shield tunnel, a test section was established in Lafayette
Square, in front of the White House, with multiple position borehole
extensometers and inclinometers concentrated around the tunnel. The
inclinometer torpedo that was advanced in 0.6-m increments down the
casing was equipped with a newly developed servo-accelerometer that
could read inclinations to 1/15000, consistent with current capabilities,
enabling extension of its use in monitoring slope stability to monitoring Figure 4 : Ground movements occurring over the shield body
of advancing digger shield, Washington, D.C. Metro, 1972.

The deep anchor of the borehole extensometers located 0.6 m above the concrete. The 20-m-deep tunnel was driven beneath a street with
tunnel showed no settlement ahead of the face, but unexpectedly large adjacent low-rise business and residential structures, and passed beneath
settlements occurred over the shield, on the order of 50 mm for every 1.2 commuter rail lines of METRA and Chicago Transit Authority. McNally
m advance, for a total of 330 mm. Tunneling and University of Illinois conducted a joint instrumentation
and monitoring program at six test sections along the two alignments,
Measurement with plumb bobs in the tunnel showed that the shield was monitoring ground movements and groundwater pressures during and
plowing (pitch of the shield significantly greater than the grade). The shield after tunneling (Kawamura and Cording, 1989, Srisirirojanakorn, 2004).
was not articulated, and had a large hood extending ahead of the face
to provide protection during anticipated future hand mining of rock in the Extensometers located above the tunnel showed that the primary
invert. In order to maintain the shield on grade, its angle of attack was source of the ground loss was the 19-mm overcut gap. Ground loss
significantly greater than its grade, causing the front of the shield to be also occurred prior to fully expanding the steel ribs behind the tail of the
approximately 300 mm above the rear so that progressive settlement shield. At the 20-m-depth, surface settlements after the shield passed
occurred over its length. were approximately half of the settlement of the deep anchor (Figure
5a). Pneumatic Piezometers and Westbay multiple position piezometers
On the second tunnel drive, the configuration of the hood was changed, (Figure 5b) recorded the undrained response of the soft clay to the stress
reducing the surface settlements from 150 to 50 mm. The volume of the changes occurring during tunneling, including (1) the build-up in pressure
surface settlement trough, as a percentage of the tunnel volume, dropped as the cutterhead approached, (2) the rapid drop in stress over the shield
from 5% to 1.67%. Volume increases occurring above the tunnel caused as displacement occurred into the 19 mm radial overcut gap around the
surface volume to be less than the volume losses around the tunnel. shield body followed by a small increase near the back of the shield. (3)
Behind the shield, pressures dropped as the ground began to displace
Ground losses due to plowing and yawing were larger for shields with into the temporary void created as the steel rib and timber lagging lining
high length/diameter aspect ratios. In subsequent years, shields were emerged from behind the tail and then increased as the lining was
articulated, effectively reducing the aspect ratio, aiding steering and expanded against the ground. Pressures continued to increase as the
reducing ground loss. shield advanced and load was transferred to the lining (Figure 5b).
Summary:
• Three-dimensional pattern of vertical and lateral movements in the

ground mass and at the surface was obtained with a concentration of
extensometers and inclinometers around the tunnel shield.
• Ground movements monitored at the source around the advancing
shield, using borehole extensometers and inclinometers, were correlated
with shield configuration and operation.
• Cause of ground loss was located and corrected.
2.6 SHIELD TUNNELING IN THE CHICAGO CLAY: 2000

a. Surface and deep anchor settlement, Test Section 4.
Before the development of pressurized TBMs, ground was consistently
lost into the overcut gap (the radial gap between the gauge cutters on
the cutterhead and the body of the shield). The gap is required to facilitate
steering and reduce friction. The only way to reduce the volume of ground
loss was to reduce the size of the gap.
This condition was observed during tunneling in Evanston, Illinois in the

soft Chicago clay. McNally Tunneling used a 3.7-m Lovat shield with
a rotating cutterhead to drive tunnels on two projects, at depths of 10
and 20 m. The shield had flood (guillotine) doors that could be closed for
access to the chamber (Figure 5a). The tunnels connected the Evanston,
IL sewer system through drop shafts to the deeper, 10-m-dia storm water
storage and transport tunnels located in Silurian Dolomites on Chicago’s
TARP (Tunnel and Reservoir Project).
Initial tunnel lining was 100-mm steel ribs and timber lagging, expanded b. Piezometric pressures, Test Sections 3 and 4.
against the ground behind the shield, and final lining was cast-in-place Figure 5 : Settlements and piezometric pressures during advance of open shield,
Chicago Clay: 2000.
(4) With time the excess piezometric pressures generated by the stress Summary:
changes drained back to ambient levels, and additional surface settlement
occurred due to consolidation of the lightly over-consolidated clay (Figure • Behavior of the clay with the non-pressurized shield has similarities
6, TS 4, and Figure 7). to a pressurized TBM operation.
• Piezometers record the undrained behavior of clay, and are indicators
Because the initial lining of steel ribs and timber lagging did not restrict of the reduction in total stress in the clay due to the presence of the
drainage from the clay into the tunnel, a drop below ambient groundwater gap.
pressures in the clay resulted in additional consolidation and a settlement • Settlements as the shield passed were concentrated above the
of 25 mm (Figure 6, TS 3). The settlement due to drainage was prevented unfilled shield gap
at the crossings beneath the two commuter rail lines by installing a plastic • Piezometric pressures can be used to monitor pressures and limit
membrane around the steel rib and timber lagging lining as it was erected consolidation. Pressures above ambient porewater pressures
in the tail of the shield (Figure 6, TS 4). (Gasketed segmental concrete after the shield passed caused consolidation and time-dependent
linings will reduce or prevent consolidation due to drainage, depending settlement.
on the sealing and relative permeability of the lining with respect to the • For the permeable lining, additional consolidation occurred due to
ground.) drainage and drop of porewater pressures below ambient
• Consolidation due to drainage was prevented by installing a membrane
around the steel rib and timber lagging. Gasketed segmental linings
will reduce or prevent consolidation due to drainage.
• For a pressurized TBM in soft clays pressurization is a balancing act:
maintain sufficient pressures at face and in overcut gap to keep the
overstress ratio below squeezing [(YH- Pi) /su] < 6] where Y is unit
weight, H is depth to tunnel crown Pi is face/shield pressure and
su is undrained shear strength. At the same time, prevent excess
porewater pressures after the shield passes in order to minimize
settlement due to consolidation.
Figure 6 : Time dependent changes in excess porewater that caused

consolidation of soft Chicago Clay.
Consolidation in TS 4 occurred in a zone around the tunnel, and caused a

decrease in volume in the clay and an additional surface settlement of 20
mm. The profile in Figure 7 shows the volume of surface settlement, which
is equal to the surface settlement times settlement, trough half width, w =
2.5 i = (r + z tan b) where i is the point of inflection on a Gaussian curve, r
is tunnel radius, and b = 38 to 40o, the vertical angle of draw from tunnel
springline to the half width of the settlement trough in the clay (Figure 7).
Figure 7: Settlement profile, before and after consolidation of the Chicago Clay, TS-4.

3 >> Pressurized tunneling
3.1 PRESSURIZED FACE advances in soils permeable to air, is an important part of controlling the
face of an EPBM (Garabagh, et al., 2012). Pressure gradients measured
3.1.1 Monitoring Ground loss at the face by the earth pressure gauges distributed over the height of the cutterhead
chamber are used to identify air bubbles (Mosavat and Mooney, 2005).
In shield tunnels, the potential for over-excavation and large ground loss
into the face was checked by measuring volume or weight of muck cars. Examples of face pressure measurements with time are shown for two
In EPBMs, the primary means is to measure the weight of the muck on tunnel projects (Figures 9 and 10).
the conveyor belt using two weight scales (Figure 8). The weights are
reconciled to the calculated weight of the volume of soil excavated during 3.1.2 Face pressures at York University test section, Toronto: 2012
the advance, with adjustments for the weight of all fluids injected into the
face, chamber, and screw conveyor during the advance. Continuous Figure 9 shows the upper, middle, and lower face pressures (measured
monitoring and display of the volumes with respect to the cutterhead by earth pressure gauges mounted on the back wall of the chamber) for
advance has proven to be useful information for the operators. an EPBM driven through a 140-m test section in sandy glacial soils in
Toronto. Very consistent pressures were maintained throughout the drive
Monitoring the weight scales throughout the advance is important for of approximately 92 1.5-m advances, which was accomplished in five
checking for large local large ground losses. If there are indications of days, on a 24- hour basis;
possible voids, secondary grouting through the segments is used to
check for and fill voids. Grout pressures should be high enough and The drop in pressures after each advance demonstrates that face
drill set-up on the trailing gear should allow drilling far enough above the pressures were maintained above the static groundwater pressures
segments to intercept rising voids. during the advance. The face pressures were also consistent, without
rapid excursions that could cause them to locally drop below the dynamic
Extensometers are used to measure the regular ground losses occurring groundwater pressures around the cutterhead and allow inflow of soil
around the shield but large local ground losses in the face will usually not (Figure 9).
occur at the extensometer location.
The primary control of ground loss into the face of a pressurized face
TBM is proactive control and monitoring of pressures to balance both the
dynamic groundwater pressures and the effective active pressure during
and between advances of the TBM.
Conditioning of the muck with additives, including foam, polymer and
Figure 9 : Face (chamber) pressures during 2nd EPBM drive through test
section, York University, Northern Tunnels, Toronto.
3.1.3 Face pressures and ground loss at Beacon Hill tunnel, Seattle
Face pressures rising at the end of the shove provide evidence that they
Figure 8 : Earth Pressure Balance Machine (EPBM). are not high enough to balance the ground water pressures. Figure 10
is a profile of a tunnel section in sandy glacial soils on the Beacon Hill
Project, Sound Transit, in Seattle, Washington (Robinson, et. al. 2012,
bentonite that fit the soil conditions, is a critical part of an EPBM operation 2013). The plot shows two locations where pressures increased at
in order to provide constant pressurization of the face throughout the the end of each 1.5-m advance, indicating that the EPBM was being
tunnel drive and to make the muck a viscous fluid that will flow through operated at face pressures below the static groundwater pressure. The
the chamber and screw conveyor, and be transportable on the conveyor conveyor belt weights of the excavated muck at those locations showed
belt, Automatic injection of bentonite slurry into the chamber and in the high values which were apparently thought to be erratic, however, they
overcut gap around the shield body to maintain pressures between were accurate. Both belt weight scales showed values significantly
advances is described for the EPBMs in Section 4. Checking for and greater than the target weights, which corresponded with the increasing
venting air bubbles that form in the chamber, both during and between face pressures.
As a result, sands were flowing into the cutterhead and large volume Pressurized TBMs address all these sources of ground loss.
losses were occurring. The voids migrated upward in the sands
to a hard clay till layer above the 40-m-deep EPBM. Months later 3.2.1 Pressurizing the face
voids were discovered and holes were drilled from the surface at the
locations of the high muck weights to locate and backfill the voids The primary purpose and benefit of the pressurized face TBM has
with a low cement content sand mix. Higher pressure compaction been to prevent inflows and large ground loss into the face, but
grouting was conducted to complete the filling and densification of pressurized face tunneling also reduces regular ground losses-the
the voids. Total volume placed was within approximately 2% of the smaller elastic displacements at the tunnel face. Face pressures
excess volumes calculated from the belt scale weights. also control the pressures in the shield overcut gap when it is filled,
which controls settlement over the shield as described in section
In the central portion of the figure, the upper face pressures balanced 3.2.3.
groundwater pressures, as indicated by the groundwater pressures
dropping slightly at the end of the advance, and the excessive belt 3.2.2 Filling and pressurizing the tail gap around the tunnel
weights did not occur. lining
The injection of grout through the tail skin to fill the gap between the
perimeter of the shield and the installed lining has become a standard
in pressurized tunneling. The gap is large enough, on the order of 150
to 200 mm, that partial filling over several rings would cause very large
settlements. The industry has switched from injecting grout through
ports in the segments behind the shield to continuous injection of
grout through the tail of shield during the advance. A lock-out is
used to prevent advance if there are insufficient operational grout
ports to fill and pressurize the gap. Two--component grouts with
rapid gel time are being widely used. On many projects, including
the recent 17.5-m-diameter EPBM in Seattle (Section 4), there was
little to say about ground loss into the tail gap, because there was
none. Extensometer anchors may show slight upward displacement
immediately above the ring being grouted, but the effect is local
because the pressures are acting over the narrow 1.2- to 2-m width
of the ring. Full grouting around the ring during the advance also
prevents squatting of the ring so that there is very little settlement
due to lining deflection.
Two--component grouts with rapid gel time have also been specified
Figure 10 : Beacon Hill Tunnel: Face pressures and large conveyor belt weights. on recent tunnels driven with non-pressurized TBMs in rock where
the concern is to prevent flow of grout forward into the open shield
gap,. The system was required by Northeast Ohio Regional Sewer
3.2 CONTROL OF REGULAR GROUND LOSS BY FILLING District for the Euclid Creek tunnel in Cleveland, Ohio, a 7 m diameter
AND PRESSURIZING GAPS tunnel driven in the Chagrin shale and supported with a segmental
concrete lining. McNally-Kiewit, the tunnel contractor, successfully
3.2.1 Potential ground loss around a tunnel shield tested and operated the system throughout the tunnel drive. The
system provides a more positive and complete filling of the tail gap,
Experience with open face shields has shown that the regular ground replacing the often uncertain and difficult procedures for filling the
losses that create the Gaussian-shaped settlement trough can occur gap that had previously been used in rock tunnels, such as delayed
(1) at the face, (2) over the shield due to ground loss into the overcut grouting through the segments with a sloping grout surface behind
gap or due to plowing or yawing (The advent of articulated shields the shield that leaves the ring initially unsupported, or the multi-step
aided steering and effectively reduced the length/diameter aspect process of injecting pea gravel and then later injecting cement grout
ratio and preventing ground loss due to pitching and yawing of the into the pea gravel.
shield), (3) at the tail during the time taken to grout the gap or expand
the lining, and (4) with deflection of the lining.

3.2.3 Filling and pressurizing the steering (overcut) gap

around the body of the shield
The overcut gap has not received as much attention as the tail gap.
Extensometer measurements above open face shields have shown that
the overcut gap was often the primary cause of the regular settlements,
as illustrated for tunneling in the Chicago clay in Figure 6.
With EPBMs, there is evidence that conditioned muck is more likely to

flow into the overcut gap in fine-grained soils than in coarse-grained
soils. Extensometer measurements on one EPBM project, without the
use of bentonite injection in the overcut gap, showed small settlement in
the hard clays throughout the drive but settlement over the shield in the
alluvial sands. The configuration at the cutterhead/shield interface and the
conditioning of the muck will have an impact on the ability of the muck to
flow from the face into the overcut gap. Figure 11 : Monitoring around EPBM, Capitol Hill, Seattle: 2011.
Loss of ground into an unfilled or partially filled overcut gap will result in Sampling through ports in the shield showed that the muck was fluid
large settlements for both shallow tunnels and large-diameter tunnels. and filling the gap and that the 100-mm gap was being maintained. The
Sections 3.4 and 3.5 describe the use of bentonite injection to limit instrumentation showed that the fluid pressures in the overcut tracked
settlement for two EPBM tunnels at shallow depth in sands. with the face pressures. Face and shield body pressures were balanced.
Piezometric pressures in the sand lenses built up and tracked below the
On larger diameter tunnels, increase in the perimeter of the annular gap face and shield pressures as the TBM approached and passed. The deep
results in a large increase in its volume. Even if the radial depth of the gap anchors of the extensometers showed no settlement as the TBM passed.
is not increased significantly from that used for smaller diameter tunnels,
loss of ground into the gap can result in excessive settlement. Assuming Extensometer E202 was located above the shield body when the
that such ground loss will occur can lead to decisions to increase the advance was stopped and face pressure was intentionally dropped at
depth of the tunnel or provide ground improvement measures. Section Ring 49 to check conditions for future cross passage excavation (Figure
4 describes the first use of bentonite injection in the overcut gap on the 12). The extensometer anchor settled 10 mm, indicating that face/shield
Porto, Portugal project, and its use on the 9-m and 12-m EPBMs on pressure was holding the gap open and preventing ground loss, but that
Barcelona Line 9, and on the 17.5-m-diameter EPBM in Seattle only a portion of the 100 mm gap was lost.
3.3 EPBM TUNNELING AT CAPITOL HILL, SOUND TRANSIT

U230 PROJECT: 2011
The contractor, Jay Dee, Coluccio, Michels Joint Venture (JCM JV)
selected a Hitachi Zosen EPBM and proposed using a large 100-mm
radial overcut to reduce wear on hard facing and cutters and reduce
interventions for tool changes for the two 1.5-km drives in glacial
lacustrine clays and tills with granular soils and boulders on Sound
Transit U230 project in Seattle. The gauge cutters creating the 100- Figure 12 : Extensometers and Piezometers in Test Section, Capitol Hill.
mm radial gap were shell cutters that extended beyond the perimeter
of the cutterhead body. The large overcut and configuration of the
gauge cutters facilitated flow of conditioned muck from the cutterhead Summary:
into the overcut (DiPonio, et al: 2012).
• Piezometers were combined with borehole extensometers.
Sound Transit was concerned for the ability to control ground •T he large (100-mm) overcut gap and cutterhead configuration facilitated
settlement with such a large overcut. The contractor proposed a test flow of conditioned muck around the cutterhead to fill the overcut gap.
section at start of tunneling to monitor ground movements around the • Fluid pressures in the overcut gap on the shield body tracked with face
shield with borehole extensometers. Piezometers were also placed pressures, and piezometric pressures in the sand lenses built and tracked
in the borehole to monitor dynamic groundwater pressures. Fluid below the face/shield pressures as the TBM approached and passed.
pressure cells were placed on the shield body to monitor the pressure •E xtensometers showed no settlement as the TBM passed.
in the overcut for comparison with face and groundwater pressures •E xtensometer located above the shield settled 10 mm when face
(Figure 11). pressure was intentionally dropped.
3.4 EPBM TUNNELING BENEATH SCHULICH BUILDING,

TORONTO: 2012
3.4.1 Plan for controlling settlement using compensation

grouting
On the Northern Tunnels Contract for Toronto-York Spadina

Subway Extension Project (TYSSE), the contractor was OHL/
FCC, and the twin tunnels were driven with two Caterpillar EPBMs
(Figure 13). The tunnels were to be driven through the future
York University Station site and 6 m below the adjacent Schulich
School of Business main building, a prestigious facility on the York
University campus consisting of a reinforced concrete frame on
spread footings (Figure 14).
Figure 14 : York University Schulich Building.
The plan was to perform compensation grouting from three shafts,

drilling over 100 tube-à-manchete (TAM) grout pipes to limit the
maximum change in vertical building displacements to 15 mm during
the sequence of compensation grouting and tunnel advance and to 10
mm after tunneling, thereby reducing damage to very slight or negligible
values (Figures 15 and 16).
Imposing the 31 mm ground movement on the structure would result in

angular distortion of 3x10-3. Building stiffness reduces the lateral strain
and distortion (Figure 16).
3.4.2 Relationship between building distortion and damage
The relationship in Figure 16 represents the state of strain at a point and

can be used to represent the average state of strain in the bottom and
in the top of a structural bay. The boundary between damage zones is
a constant value of the maximum principal tensile strain (Cording et. al.
2001, 2010). Boscardin and Cording (1989) developed the relationship
in terms of the maximum angular distortion and lateral strain of a beam
with length/height ratio of one, but it is equivalent to the relationship
for the state of strain at a point shown in Figure 14, with only slightly
different shape of the boundaries for the damage zones.They built on the
relationships developed by Burland and Wroth (1974) for bending and
shear strains in beams with window penetrations producing high E/G
values). The relationship for angular distortions was correlated with the
Figure 13 : Northern EPBM Tunnels, Toronto. work of Skempton and MacDonald (1956) for settlement of buildings.
The lateral strains were correlated with the lateral displacement vs
damage criterion developed by the United Kingdom National Coal Board
The special measures developed during design and the measures for subsidence over deep mines. Damage level descriptions in the figure
employed during tunneling are presented in Kramer, et al. 2015. were developed for masonry structures by Burland et. al. (1977).
During design, surface settlement volumes were estimated at
0.5% ground loss with a maximum of 1%, which would result in an The relationship has general applicability to different types of structures.
estimated maximum greenfield settlement of 31 mm (Figure 15). It is used for bearing walls behaving as a single beam and is also used
for assessing average state of strain and damage to the infill or finishes
within the bays of a frame structure.

Stiffness of the ground/structure can reduce both angular distortion and The tunnel designer, Hatch Mott MacDonald, in assessing the impacts
lateral strain estimated from the greenfield settlements. A relationship of tunneling on the Schulich building, performed numerical analyses
showing the reduction in lateral strain due to the relative soil/grade beam to determine the effect of building stiffness on the distortions using the
stiffness is provided in Boscardin and Cording, 1989. Professor Robert distortion/damage relationship of Figure 16.
Mair, in his 2013 Muir Wood lecture, provides relationships and examples
of the effect of building stiffness in reducing settlement and the shape and
distortion of the settlement profile.
Figure 15 : Profile of ground conditions and predicted settlements at Schulich Builidiing, Northern Tunnels, Toronto.
Figure 16 : Damage criterion based on average state of strain in a bay due to lateral strain and angular distortion.
3.4.2. Test section prior to EPBM tunneling beneath Schulich provide a positive means of filling the overcut gap in the anticipated more
Building granular soils. Fluid pressure cells were also recommended and installed
by Caterpillar on the shield body to monitor pressures in the overcut gap.
The three shafts for compensation grouting had been excavated, but the
compensation grout holes had not been drilled as the two TBMs reached Static ground water level in the granular soils was 10 m above the tunnel,
the future York University Station, 140 m from the adjacent Schulich a pressure of 1 bar on the upper face of the EPBM. On the first drive the
building. The contractor requested to advance tunnels beneath the upper face pressures were at 1.9 bars and settlements 6 m above the
building without installing compensation grout pipes. shield were in the range of 1 to 5 mm. On the second drive through the
test section, the contractor elected to increase face pressure to 2.2 bars
During the drive of the two EPBMs toward York University, settlement of resulting in a reduction of settlements to 1 to 2 mm (Figures 9 and 17).
the surface,12 m above the tunnels had been small, typically in range of 1
to 5 mm. However, there were limited data from borehole extensometer At the end of the first drive through the test section, upon reaching
data showing displacements above the tunnels and at the level of future and embedding the TBM cutterhead in the future station headwall, the
building foundations the tunnel. Further, there was concern for the ability face pressures were dropped for maintenance of the cutterhead. Face
to control ground movements and fill the overcut gap in the increasingly pressures are coupled with overcut pressures, so that the pressures
granular glacial soils at the York University Schulich building. in the overcut gap also dropped resulting in 13 mm of settlement of
the extensometer located 6 m above the shield, a confirmation of the
The Toronto Transit Commission (TTC), and consultants recommended effectiveness of pressurization of the overcut gap in preventing settlement.
that a test section be installed prior to tunneling beneath the Schulich
building to determine if settlement could be consistently controlled within Control of the TBM operation by the contractor, OHL/FCC, and the
the 10-mm criterion established for the compensation grouting program. consistency of face pressures, conditioning, and filling of the overcut gap
was evident in the two drives through the test section. The settlements
TTC coordinated the test section program and conducted daily reviews of 1 to 2 mm were far below the maximum 10-mm criterion (Figure 17).
with the designer, contractor, construction manager, geotechnical
consultant, and instrumentation group. Instrumentation included 33 The passage beneath the building was accomplished using the same
multiple position borehole extensometers, each with four anchors, 0.7 m ground control procedures as in the test section. Vertical displacements
above the tunnel and at the level of the future building foundations, as well as measured by liquid level lines, robotic total station measurements
as precise level surveys of the surface and the extensometer heads, and and precise level surveys showed a maximum heave of 1.9 mm and
piezometers. (Kramer, et al, 2015). settlement of 1 mm over the 81 and 62 ring pushes.
The 6.1-m Caterpillar EPBMs had been built with a system for bentonite
injection through the shield body but it had not been regularly used by the
contractor during the drive toward York University in the predominantly
clayey tills. TTC consultants recommended that the system be used
throughout the test section and passage beneath the Schulich building to
Figure 17 : Settlement 6 m above tunnels after passage of EPBMs through test section, 9/18/2012.

3.4.3 Summary Regional Connector. TBM tunneling on the Regional Connector

light rail line was completed in February, 2018. The first 150 m of the
• Toronto Transit Commission and contractor, OHL/FCC, conducted twin tunnel drives was at shallow depth in alluvial sandy soils beneath
a coordinated and cooperative program for monitoring ground structures as described in the following sections.
behavior and controlling the TBM operation.
• Consistent control of TBM operating parameters, including face Purple Line. On the Purple Line, which extends west along Wilshire
pressures, conditioning, and filling of overcut and tail gaps was Boulevard through Century City to University of California at Los
demonstrated in the 140–m test section. Angeles and the Veterans Administration Hospital, three heavy-
• Settlements of 2 to 5 mm on the first drive were reduced to 1 to 2 mm rail sections with twin tunnels and seven station excavations are in
on the second drive by increasing face/shield pressures. Settlement design or underway. The Purple Line is predominantly in alluvial soils
was below the 10 mm maximum, and below values achievable with and in the Fernando formation of very hard silt and clay. Sections of
compensation grouting. the tunnel pass through soils with high concentrations of methane
• During the drive beneath the Schulich Building settlements ranged and H2S. One section passes through soils infused with naturally
from 1.9 mm heave to 1 mm settlement. occurring asphalt, in the vicinity of the La Brea tar pits.
3.5 GROUND CONTROL ON LOS ANGELES METRO TUNNEL 3.5.2 Regional Connector Project: EPBM start-up at shallow
PROJECTS depth beneath structures
3.5.1 Pressurized Tunnel Projects On the Regional Connector Light Rail line, the twin tunnels were driven
with a single Herrenknecht EPBM. The second tunnel holed through
After LA Metro experience with large ground losses using open face in February 2018, and construction began on sequentially excavated
digger shields in sandy alluvial soils in the early 1990s, a Tunnel cross-passages and a cross-over cavern. Detailed descriptions of the
Advisory Panel was convened in 1994 and concluded that controlled project and the start of tunneling are provided by Hansmire, et al.,
tunneling could be conducted in Los Angeles and that pressurized 2017.
face TBMs should be utilized.
The first 150 m of the tunnel drive beyond the braced excavation launch
Gold Line East Side Extension. The next LA Metro tunnel project shaft passed beneath a reinforced concrete frame parking structure
was the Gold Line East Side Extension, driven in 2006 with two on spread footings and then shops and a mall at a depth increasing
Herrenknecht EPBMs at a depth of 15 to 20 m in Old Alluvium consisting from 6 m to 12 m in medium to medium dense clean uniform sand
predominantly of silty clays and silty sands. (Choueiry et. al., 2007, interbedded with silty sand, gravel and well-graded sand. Near the
Robinson and Bragard, 2007). Compensation grout pipes, although end of the 150-m section, the Fernando Formation, a weak siltstone/
installed beneath structures located near the start of tunneling, were claystone (very hard silty clay), occupied most of the tunnel face, with
not used during tunneling. Ground settlements throughout the drive the granular alluvium remaining at and above the tunnel crown.
were typically in the range of zero to 8 mm. There was no injection of
bentonite in the overcut gap in these predominantly dense silty clays The contractor, Regional Connector Constructors (RCC), drove the twin
and silty sands. tunnels with a rebuilt Herrenknecht EPBM that had been used to drive
one of the twin tunnels on the LA Metro Gold Line East Side Extension
Current Metro projects. On current Metro projects, the maximum in 2006 and one of the twin tunnels on Seattle’s Sound Transit U220
surface settlement criterion is 13 mm. A fully pressurized envelope University Link project in 2011. For the Regional Connector project, a
around the TBM face, shield and tail is required, including filling system was added for filling and pressurizing of the steering (overcut)
and pressurization of the steering (overcut) gap and monitoring with gap, monitored with six pressure cells on the shield perimeter.
pressure cells installed on the shield perimeter.
Contract documents required installation of a compensation grout
Currently five underground light and heavy rail projects are underway. system in the first 120 m of the tunnel drive and permeation grouting for
a 30-m length of tunnel beyond to protect a building and a 4-m-wide
Crenshaw. The tunneling portion of the Crenshaw light rail line storm drain. Hayward Baker used horizontal directional drilling (HDD)
was completed in 2017. The twin tunnels were driven with a single to install horizontal compensation grout pipes extending 62 to 127 m
Herrenknecht EPBM in sand and gravel alluvium, with the tunnel crown from the braced excavation, 2 m above the crown of the twin tunnels.
above the water table in much of the drive In these sandy alluvial soils. Drilling of the pipes produced no measureable surface settlement and
once the gaps were filled and ground loss prevented, the remaining pre-conditioning of the ground through the tube-à-manchette (TAM)
settlement was a function of the pressures applied at the face and pipes prior to tunneling was controlled to heaves less than 4 mm.
around the shield body with respect to the overburden stresses.
It was recognized that settlements during tunneling could be controlled

without compensation grouting if the ground control capabilities of the
EPBM were fully mobilized at start-up. This proved to be the case. Surface
settlements throughout were well below the 6-mm surface settlement
level planned for initiation of compensation grouting as the TBM passed.
Recognizing the shallow depth of the tunnel and the limited tunnel
distance to monitor ground movements and confirm that TBM ground
control procedures were in place before passing beneath structures, the
contract required a “tunneling performance demonstration zone” that
was approximately 18 m of tunnel construction prior to starting under the
structures. This zone is actually an unexcavated berm below the deck
beams for the station excavation.
Figure 18 : Settlement above EPBM at start of first drive.
RCC prepared a plan for sequential pressurization of the cutterhead,
and the overcut and tail gaps as they passed through the portal seals.
Bentonite was injected into the overcut gap between the 1.5-m advances
and a two-component grout was injected through the tail skin as the
shield was advanced.
Limited surface access for installing and monitoring extensometers

and settlement points led to the installation of horizontal inclinometers
in horizontal directionally drilled (HDD) holes extending 117 m from the
launch shaft. The inclinometers were a Shaped Accelerometer Array,
SAA, consisting of a string of measuring points located at 1-m spacing.
They were located 1 m above the tunnel and below the compensation
grout pipes so that they measured ground displacements around
the advancing TBM, largely unaffected by the stiffness of the overlying
compensation grout pipes and the pre-conditioning grout.
Figure 19 : Horizontal Inclinometer displacement, 1 m above EPBM, First Drive.
During monitoring of the horizontal inclinometers, adjustments were made
to correct for drift of the gauges, and each point in the array was zeroed 8
m ahead of the advancing TBM in order to show only the displacements
due to tunneling. Beyond the shaft and berm, tunnel depth was 6 m and overburden
pressure was 1.2 bars, higher than the 0.7-bar face/shield pressure. Thus
Settlement of the horizontal inclinometer points compared closely with settlement 1 m above the shield was 10 mm.
the settlement of the deep anchors of extensometer 1 (MPBX 1), located
just beyond the berm in the braced excavation and adjacent to the Once the shield was beyond the shaft wall and the shallow berm, upper
parking structure where tunnel depth was 6 m. A total settlement of 9 face/shield pressures were increased to 1 bar, and then incrementally
mm occurred over the shield (Figure 18). After the shield passed beyond increased from 1 to 2.5 bars as the depth of the tunnel increased from
the berm, the settlement 1 m above the tunnel was less than 5 mm as 6 to 12 m. As a result settlement of the horizontal inclinometer located 1
shown by the continuous record provided by the horizontal inclinometer m above the tunnel decreased to less than 5 mm and then to near zero
(Figure 19). levels (Figure 19).
Initially, as the TBM cutterhead advanced beneath the berm in the shaft, Beyond the first 150 m of the tunnel drive, the hard clay/silt (Fernando
upper face pressures, and shield overcut pressures as well, increased Formation) occupied the full face of the tunnel. Pressure gauges on the
from 0.4 to 0.7 bars, which were close to the overburden pressure applied perimeter of the shield showed that pressures were tracking with the face
by the total height of the 3- to 3.5-m berm and sandbags. The 0.7-bar pressures indicating that the conditioned clayey, silty muck was flowing
pressures were maintained until the body of the shield had passed beyond around the cutterhead and filling the overcut gap. The tunnel passed 1.5
the berm and shaft, and MPBX 1. As a result the first three horizontal m below existing Metro Red Line tunnels with settlements measured by
inclinometer points, located beneath the berm showed heave but there seven extensometers immediately above the tunnel that ranged from 0 to
was no blow-out of muck or bentonite (Figure 19). 6 mm after driving both tunnels.

4 >> Large diameter pressurized tbms
and the alaskan way viaduct replacement tunnel
4.1 INTRODUCTION 4.1.2 Barcelona Line 9
Marc Isambard Brunel not only built the first subaqueous shield Barcelona’s Line 9 was driven with 9-m EPBMs for double track
tunnel, but it was also large, even by today’s standards. It was a box sections and 12-m EPBMs for double track with station platforms in
structure with a width of 12 m to accommodate a twin carriageway the tunnel. EPB shield control parameters with target, warning, and
with masonry side and center walls. It was unprecedented, innovative alert levels were provided for tunnel reaches and jointly monitored by
and risky. Contractor and Engineer field representatives. The ability to fill the
overcut gap with the conditioned muck flowing from the cutterhead
New records are being set for the size of shielded tunnel boring was not considered a systematic ground support measure, and EPB
machines; the largest shielded TBM is now 17.6 m. (However that shields were equipped with a dedicated system to execute volume-
only represents an average increase in width of 31 mm/year over the controlled slurry bentonite injection in the annular gap around the shield
175 years since Brunel completed the Thames tunnel!) Although (Bono, 2008). On at least one of the Line 9 contracts, a combination
It will always be difficult to match the unprecedented conditions bentonite/lime slurry was injected in the shield gap (Escoda, 2017).
encountered and the innovative measures used by Brunel, projects
with large diameter TBMs have dealt with new challenges, have 4.2 THE ALASKAN WAY VIADUCT REPLACEMENT TUNNEL,
developed innovative procedures for mitigating risk, and have SEATTLE: 2017
demonstrated the capability to control the ground to levels far below
those that could cause damaging settlement. Following are ground Washington State Route 99 extends along Seattle’s Elliot Bay
control measures developed and used for urban tunneling with large- waterfront on the double deck Alaskan Way Viaduct, a reinforced
diameter earth pressure balance TBMs. The experiences on these concrete structure built in 1950 on piles extending through 10 m of
and other large-diameter tunnel projects, were an essential part of the hydraulically placed fill and recent alluvium into glacially overridden
design and construction planning and risk assessment for the Alaskan soils. The structure will be replaced by a single, 2.8-km-long,
Way Viaduct Replacement Tunnel. 17.5-m-diameter tunnel accommodating a double deck structure,
each with two traffic lanes and a breakdown lane, with longitudinal
4.1.1 Porto Metro, Portugal: Mechanized tunneling in urban ventilation ducts, and emergency passenger egress at the sides. The
areas design-build contractor, Seattle Tunnel Partners (STP), a Joint Venture
of Dragados USA and Tutor Perini, selected a Hitachi-Zosen earth
On the Porto Metro, Portugal, a sinkhole and building collapse led to pressure balance machine (EPBM).
review and implementation of recommendations for coordinated effort
and measures for control of the 9-m EPBMs. Descriptions of the tunnel operation and ground behavior are included
in papers by Cording, et. al. 2015, 2017, Escoda, et. al. 2017,
The investigation and lessons learned are described in Diez and Fernandez, et. al. 2017, and Mosavat and Mooney, 2015.
Williams, 2003, in Raleigh, P., 2006, and in Guglielmetti, V., Grasso,
P., Mahtab, A., Xu, S., 2008. The TBM holed through in April, 2017. The double-deck roadway
Measures included: structure was being built as the tunnel advanced and was completed
• Development of a Protocol for Advancing the Tunnel (PAT) involving in 2018.
- Rigorous management control and monitoring of the TBM
operation, combined with analysis of expected tunneling At tunnel depth beneath the Alaskan Way Viaduct are glacially
conditions and verification of ground structure behavior by overridden mixed soils consisting of outwash sands and lacustrine
geotechnical monitoring clays. At tunnel depth beneath Pioneer Square are hard lacustrine
• Key machine operating targets provided to the tunnel operators clays, underlain by a continuous, pro-glacial outwash sand and gravel
• Project engineering and construction management more involved in at invert level. To the north, the sand and gravel rises into the tunnel
the tunneling process. face and dense granular tills and till-like soils are present in the tunnel
• Requirement for closed face, pressurized tunneling throughout entire face and crown (Figure 20a).
alignment, in the unweathered granite as well as in the variable and
weathered, soil-like granite. The granular pro-glacial outwash had two major effects. In Pioneer
• Automatic injection of bentonite into the cutterhead chamber to Square, its presence at or near invert level below the lacustrine clays
maintain pressures between advances, and bentonite Injection to required face pressures that balanced the 5-bar water pressure at
smear the tunnel face before daily interventions to change cutters in the invert. Over the length of the tunnel, its high hydraulic conductivity
the abrasive granite. and continuity resulted in a flat groundwater table near high tide level,
• Primary grouting of the lining through the tail skin. Secondary grouting even though the ground surface increased to elevation +50 m to the
through segments to check for volume loss north. At its low point, the tunnel crown dropped 30 m below the
• Filling of the overcut gap with bentonite. groundwater level, thus the maximum groundwater pressures were
approximately 3 bars at the crown and 5 bars at the invert of the TBM.
4.3 ASSUMED VOLUME LOSS VS ACTUAL SURFACE For illustrative purposes, Figure 20a shows the surface settlements along
SETTLEMENT the length of the Alaskan Way Viaduct Replacement tunnel for a volume of
the surface settlement trough of 0.5%, a value that was used in one of the
Ground loss is commonly reported as a percentage of the volume of the estimates during preliminary design. In the figure, the surface settlements
settlement trough with respect to the volume of the tunnel per unit length. are determined for a trough width based on a vertical angle of 40 degrees
The relationship is used to summarize and compare ground control from the springline to the half width, w, of the settlement trough. These
achieved on projects but often information on the causes of the volume calculated settlements exceed the 25-mm criterion at both the shallower
loss are not provided so that the relationship between percent ground loss north and south ends of the tunnel.
and TBM parameters required to control the ground loss are not known.
In contrast to Figure 20a, the actual surface settlements during the
Using a percentage of ground loss obtained from experience on small- passage of the TBM were smallest and not measurable when the tunnel
diameter tunnels will tend to overestimate the ground loss for a larger was shallowest (crown less than 30 m deep). They were in the range of
tunnel. For example, the gaps around a large-diameter tunnel have 2.5 mm when the tunnel was deepest (crown 45 to 60 m deep), a value
dimensions close to those used for smaller diameter tunnels, so the that was still an order of magnitude less than the maximum settlement
volume of ground lost into an unfilled gap will increase more closely in criterion (Figure 20b).
proportion to the diameter rather than the square of the diameter, which is
the relationship for the percent ground loss
a. Surface settlement criteria and calculated surface settlement for an assumed ground loss of 0.5%.
b. Observed surface settlement.
Figure 20 : Surface settlement calculated from percent ground loss compared to observed settlement on AWVR Project.

The observed surface settlements were based on the automatic Figure 21 summarizes ground improvement measures considered
structure monitoring points (ASMP) (total station) for structures located and used to mitigate risk by WSDOT and by STP. The specified
above the tunnel. The ASMPs showed variations over a period of a compensation grouting beneath two buildings in Pioneer Square
few days or weeks that were greater than the displacements due to was eliminated based on the evidence from the SESMP test sections
tunneling, but the continuous time plots made it possible to compare showing that surface settlements were well below the 13 mm
points above the advancing tunnel with those further away in order criterion for compensation grouting. However, building repairs were
to separate out the displacement due to tunneling. The settlements needed. Large differential settlements and distortions and cracking
over most of the alignment were too small to be precisely determined had occurred over the past century in the two structures, which were
from the level surveys. on timber pile foundations extending through the loose hydraulic
sand fill and recent alluvial deposits. STP stabilized and stiffened the
The results summarized in the following sections provide a foundations, in one building with micro piles and in the other with
perspective on why the constant ground loss assumption did not grade beams.
model the observed ground movements.
The primary ground control measure, in both WSDOT’s contract
4.4 INNOVATION AND RISK requirements and STP’s tunneling plan and operation, was to monitor
and control the TBM operation. Measures included:
Tunneling for replacement of the Viaduct almost did not happen due • Overcut gap pressurized and filled with bentonite.
to political and economic pressures. A major earthquake in 2001 that • Earth pressure gauges on the shield body to monitor pressures in
damaged the Alaskan Way Viaduct and Seawall led to the realization the overcut gap.
that the Viaduct had to be replaced. • Closely spaced extensometers, at an average of 16 m centers,
over the entire alignment.
A tunnel was briefly considered in 2001 but the options brought • Piezometers added in each of the extensometer borings.
forward were to replace the viaduct or travel by surface streets.
The community, recognizing the disruption and traffic impacts, 4.4.1 South End Settlement Monitoring Plan
campaigned for tunneling. A coalition of 300 organizations was
formed and proactively engaged and educated state and city officials The South End Settlement Monitoring Plan (SESMP) section, was
early in the planning process (Donegan, 2017). Washington State located beneath the project construction yard and immediately
Department of Transportation (WSDOT) re-examined the tunneling adjacent to the Alaskan Way Viaduct. STP proposed extending
option. Twin tunnels driven on separate streets were considered, but project contract limits further south to have additional distance for
ultimately a single large-diameter tunnel was specified. controlling and monitoring TBM performance and gaining tunnel
depth prior to passing beneath the Viaduct.
Dick Robbins, in his 2013 Muir Wood Lecture described the Alaskan
Way Viaduct Replacement Tunnel with the heading “MODERN Monitoring of ground movements and TBM performance in the
INNOVATION AT WORK” and noted: “Despite all of the current first 300 m of the SESMP section in November 2013 showed that
challenges for creative development work in the industry, examples settlements were controlled to small values as the tunnel crown
can still be found where innovation is encouraged particularly where passed into the glacial soils (Cording et al., 2015).
project conditions are exceedingly difficult or unprecedented. The
SR 99 Viaduct Replacement Tunnel in Seattle, Washington, USA is Tunneling was halted in December 2013 and an access shaft was
an example of both modern-day risk sharing and innovative design installed to remove and repair the cutterhead and main bearing. In
work.” January 2016, the reassembled TBM was advanced out of the shaft
through the last 100 m of the SESMP section to the hold point at
To mitigate such risks, a series of preliminary design decisions and Safe Haven 3. TBM operations and observations in two test sections
contract requirements were prepared. Of particular concern were confirmed that ground control measures were in place and that
conditions that would have a different or even unique effect on a settlements were being controlled as the TBM approached Safe
larger-diameter tunnel. Risks of tunneling settlement and how the Haven 3. A compressed air intervention was conducted once the
tunnel construction contract was established are described in cutterhead was in the jet-grouted Safe Haven. The advance was
Hansmire et al. (2011). The tunnel contract had explicit requirements continued beyond Safe Haven 3 on April 29, 2016
for geotechnical instrumentation and a Construction Monitoring Task
Force (CMTF) that were implemented by the design-build contractor,
STP.
Figure 21 : Profile of Alaskan Way Viaduct Replacement Tunnel Alignment and ground control measures.
4.5 COORDINATION PLAN 4.5.2 Coordination of the TBM operation
Prior to advancing beyond the hold point at Safe Haven 3 and beneath
the viaduct, a coordinated program of monitoring of ground and
TBM performance and control, review, and adjustment of the TBM
operation became fully organized that drew together the capabilities
of the STP managers, TBM engineers, operators, quality control,
safety, geotechnical monitoring teams and WSDOT engineering and
construction management (CM) (Figure 22). An environment was created
in which information was shared and information-based decisions were
timely executed.
4.5.1 Construction Monitoring Task Force (CMTF)
The Task Force effort included:

• Daily meeting with STP and WSDOT teams, chaired by tunnel manager,
to review previous day’s operation.
• Review of Daily Tunnel Parameter Log to be used by TBM operators
showing target and green range for key operating parameters affecting
Figure 22 : Coordination among owner, contractor,
ground control (Figure 23). tunnel operation and ground monitoring.

bar below target pressures. The increase in lower face pressure between
advances resulted from an increase in density of the conditioned muck as
bentonite slurry replaced air coming out of the foam. A vent line to the top
of the chamber was used for automatic purging and venting of air onto
the conveyor belt.
Bentonite slurry injected into the overcut gap was volume controlled and
automatically injected throughout the advance to fill the 30-mm gap.
Presssure in the gap tracked with the face pressure. Between advances,
pressure in the overcut dropped toward groundwater pressure, and
bentonite was injected to maintain pressures above the groundwater
pressures.
Figure 23 : Example of Daily TBM Parameter Log: 12/9/2016 - 12 /10/2016.
4.5.3 TBM ground control operation
The diameter of the TBM was large enough that the spokes were designed
to free-air access to the cutterhead for changing rippers and disk cutters
(Figure 24). Disk cutters on alternate spokes prior to launch were only Figure 24 : EPBM cutter head in launch shaft,
used for a short distance in the SESMP section and were replaced with Alaskan Way Viaduct Replacement Tunnel.
additional rippers. The cutterhead and ribbon screw conveyor were
designed to accommodate 900-mm boulders. Scrapers on the edge of
the spokes were replaced during compressed air interventions. Ports on
the shield body shown in Figure 24 were designed for probing over the
cutterhead and for bentonite injection into the overcut gap, but were only
used for the latter.
Recipes for conditioning the muck include anti-clay polymer and foam for
the clays. For the sandy soils, polymer, foam and bentonite were used. Up
to two tons of bentonite were added to the injected solution per 2-meter
advance.
Figure 25 is an example of upper and lower face pressures and upper

shield body pressure measured over a two-week period. There were 13
earth pressure cells in the chamber for monitoring face pressure and 6
earth pressure cells on the upper arch at the front and middle sections of
the shield body for measuring pressures in the overcut gap.
The upper pair of face pressure cells served as the reference for controlling
face pressures. Between advances, bentonite was aultomatically injected Figure 25 : Example of TBM face and shield pressures, 12/1/2016 –
into the chamber to maintain the upper face pressure at a constant 0.2 12/13/2016, Station 255+50 to 260+00.
4.5.4 Geotechnical Monitoring Monitoring at the Source. Primary emphasis was placed on reporting
the ground movements at their source, providing continuous time plots of
After every 2-m advance of the TBM, the geotechnical team provided extensometer displacements and the piezometric pressure immediately
a report on ground conditions and time plots of monitoring data from above the advancing TBM. Results were correlated with the time plots of
SolData instrumentation within a 60-m zone around the advancing face and shield overcut gap pressures.
TBM The information was provided to the operations team and
inspectors in the tunnel and emailed to WSDOT and STP team Monitoring at the Surface. Tme plots were provided of surface and
members (Figure 26). structure displacements confirming that settlements were within allowable
criteria.
Figure 26 : Ground and structure monitoring from Safe Haven 3 to Alaskan Way Viaduct and Pioneer Square.
Figure 27 : Alaskan Way Viaduct Replacement tunnel alignment under Seattle.

4.6 TBM ADVANCE BENEATH ALASKAN WAY VIADUCT AND

PIONEER SQUARE
4.6.1 Alaskan Way Viaduct Crossing
The TBM advanced from Safe Haven 3 at Station 210+00 on April 29,
2016, passing 4 to 6 m beneath pile foundations for four Alaskan Way
Viaduct piers (Figure 26 and 27).
The viaduct was shut down by Washington State Department of

Transportation for a period of two weeks, but was reopened to Monday
rush hour traffic a week early as the TBM cutterhead passed beyond the
last viaduct foundation, with no negative effects.
During the passage beneath the Viaduct, pressurization of the overcut

gap on the shield was close to overburden pressures causing 5 mm
heave of the Viaduct foundation.
The large pressurized area of the shield (a cylinder 17.5 m in diameter Figure 28 : Settlement and pressures in clay above advancing TBM,
and 20 m long) can lift a large area and recover settlements further Pioneer Square.
above the tunnel than is possible with grouting behind the tail, which
only pressurizes over the width of the 2-m tunnel lining ring. The ability to
recover displacements above the shield at pressures less than overburden
pressure was also demonstrated at one of the extensometers in Pioneer
Square in sands, as the dynamic groundwater pressures above the shield
dropped and shield pressures provided an increased effective pressure
that supported the wedge of soil above the shield body, recovering
displacement over the shield with no measurable surface settlement.
4.6.2 Displacements and Pressures in Clay beneath Pioneer

Square.
During tunneling beneath buildings in Pioneer Square, there was no

settlement of structures or the ground surface (as measured by the ASMP
points on the structures, the liquid level lines within the structures, and the
survey leveling).
Most of the piezometers located 1.5 m above the crown in this section
were in clay. The undrained behavior of the clay resulted in a reduction
in porewater pressure over the shield due to the difference between the
overburden pressures and the pressures around the shield. Figure 29 : Horizontal displacement toward TBM face Inclinometer # 53,
Pioneer Square .
Figure 28 illustrates the response that typically occurred throughout
Pioneer Square when clay was in the tunnel crown. Piezometric pressure
in PZ 41 began dropping to a value 1 bar below ambient pressure over
the front portion of the shield and recovered over the back half of the
shield and tail. At the same time, the deep anchor, 1.5 m above the shield,
displaced downward approximately 1 mm.
Inclinometers placed ahead of the TBM recorded small lateral

displacements toward the cutterhead. Inclinometer 53 showed 2 mm
when the TBM approached within 3 m of the inclinometer (Figure 29).
4.7 TBM ADVANCE WITH INCREASING DEPTH TO BNSF TUNNEL differential between overburden pressure and the face/shield pressures (12
bars versus 4 bars), the porewater pressures in the clay began dropping
North of Pioneer Square along First Avenue, cover over the tunnel 15 m ahead of the shield, rather than over the shield, and settlements over
increased to 60 m while groundwater levels remained flat at elevation the shield were higher.
+1.5 to 3 m, near tide level in Puget Sound. With the increasing cover,
the differential pressure between overburden pressure and the face/ 4.7.2 Pressures and displacements In Sand at 50-m-depth
shield pressures increased (12 bars overburden vs 4 bars face pressure).
Approaching the BNSF tunnel (Figure 30) the higher differential between Settlements and piezometric pressures at MPBX/PZ 69, in sands at and
the overburden pressures and TBM pressures resulted in deep anchor above the tunnel crown, are shown in Figure 33. Groundwater pressures
settlement between 8 to 15 mm (Figure 31). 1.5 m above the shield were 2.2 bars and began increasing 35 m ahead
of the TBM, to values temporarily approaching within 0.6 bars of the 4-
Surface settlements of 2.5 mm is added to the differential settlement of bar upper face and shield pressure. Deep anchor total settlement of 13
the deep extensometer anchors to obtain the total settlements shown in mm occurred over the shield, the same range as those measured in the
the profile, Figure 31, and time plots in Figures 32 and 33. clayey soils in the high cover section.
For a hydrostatic distribution of pressures, piezometers 1.5 m above

the shield would be 0.15 bars below the shield crown pressure. The
piezometric pressure must be at least 0.3 bars below the shield pressure
in order for the soil in the 1.5 m interval to be supported. As shown in
Figure 30, the hydraulic gradient between the bentonite slurry and the
piezometer was high enough to create a differential pressure of 0.6 bars.
Over a four-day period between advances, penetration and caking of the
bentonite slurry increased the hydraulic gradient and piezometric pressure
dropped to 1 bar below the shield pressure in the overcut gap.
Figure 30 : TBM crossing at BNSF Tunnel and Pike Street Adit.
Figure 32 : Extensometer displacements and piezometric pressures above TBM

in Clay, 55 m depth.
Figure 31 : Settlement 1.5 m above TBM approaching BNSF tunnel.
4.7.1 Pressures and displacements in Clay at 55-m-depth
Extensometer/piezometer, MPBX/PZ 74, in clay at 55 m depth (Figure

32) shows similar behavior to that observed at shallower depth in Pioneer
Square (MPBX/PZ 41, Figure 28). However, because of the greater Figure 33 : Extensometer displacements and piezometric pressures above
advancing TBM in sand, 50 m depth.

4.7.3 Settlement distribution in BNSF Tunnel and Pike Street Adit
The TBM passes at a skewed angle beneath the BNSF tunnel. Automatic
structure monitoring points (ASMPs) precise survey prisms) were installed
on the walls and floor of the BNSF tunnel.
Small settlement of 0.5 mm began to be observed in the BNSF

tunnel as the TBM approached within approximately two diameters
(35m) of the BNSF tunnel and were skewed to the east (right) due to
the angle of TBM approach. After the TBM passed the BNSF tunnel,
total settlements reached 5 mm at the TBM centerline (Figure 34).
The precision of the instrumentation in both the BNSF tunnel and the Figure 36 : Observed settlements compared with pre-construction analysis,
Pike Street Adit permitted the measurement of the shape of a classic Pike Street Adit.
Gaussian settlement trough despite the small settlement magnitude.
Prior to tunneling, three-dimensional FLAC analyses were conducted
to assess the effect of the TBM passage on the BNSF tunnel and the
Pike St Adit. In the analysis, the ground loss was assumed to be one
half of the volume of the overcut gap, which resulted in an estimated
settlement of 15 mm (Figure 36). This conservative estimate showed
that distortions would be below damage levels in both the adit and
BNSF tunnel. Although the analysis used a small-strain constitutive
model, it showed a wider distribution of settlement than the observed
settlement trough.
Measurements as the BNSF Tunnel was approached indicated that

settlement volumes and displacements would be smaller than those
assumed in the analysis, because the overcut gap was filled.
Figure 34 : Settlement distribution measured with ASMP prisms in BNSF tunnel. 4.8 SETTLEMENT IN SANDS AT NORTH END OF TUNNEL
Beyond the BNSF crossing, granular soils were increasingly concentrated

The Pike St. Adit is immediately beyond the BNSF crossing and 22 m at and above the tunnel face, consisting principally of lacustrine silts and
above the crown of the AWVR tunnel. Settlement within the adit was outwash sand and gravel and sandy till-like soils including sand and gravel
measured with a tilt beam (essentially a horizontal in-place inclinometer). with less than 5% fines.
The settlement trough was similar to that observed in the BNSF tunnel.
(Figures 35 and 36). Polymer was added to the bentonite injected in the shield gap in order to
increase viscosity and limit penetration in the sandy soil. As a result the
outward hydraulic gradient, was sufficient to allow the pressures in the
shield gap to support groundwater pressures and the effective pressure
from the wedge of soil above the 17.5-m-width and 20-m length of the
shield.
The monitoring results confirmed the effectiveness of pressurizing and

filling of the overcut gap. Settlements 1.5 m above the shield were the
same for both sandy and clayey soils over the entire tunnel drive. They
were less than a millimeter when the differential between overburden and
face/shield pressures was less than 2 bars in both the clays at the south
end and the sands at the north end (Figure 37).
Figure 35 : Settlement distribution measured with tilt beam (horizontal

inclinometer) in Pike Street Adit.
4.9 SETTLEMENT DUE TO DIFFERENTIAL PRESSURE At the shallower depths at the north and south ends, for differential
pressures less than 2 bars, deep anchor settlements were 1 mm and
In Figure 38, the deep anchor settlements are plotted in the opposite volume loss was on the order of 0 .01 %. The small displacements appear
direction, to illustrate their correspondence with difference between to be due to higher stiffness of the ground at small strains. The settlements
overburden and face/shield pressures. Differential pressures were were a function of the difference between overburden pressures and face/
particularly high at depths of 60 m because the groundwater shield body pressures and inversely proportional to the stiffness, Young’s
pressures and required face/shield pressures were low. Even with modulus, E, of the ground.
the differential of 8 bars, the high stiffness of the glacially-overridden
soils resulted in relatively small deep anchor settlements of 8 to 11 It should be noted that in more compressible soils, once the gaps are
mm (a volume of less than 0.1%), not due to ground loss but due to pressurized and filled, settlements can be reduced, as needed, by
elastic displacement. increasing TBM face/shield pressures above the minimum values required
to balance ground water and effective active pressures.
Figure 37 : Observed settlements at the surface and 1.5 m above TBM.
Figure 38 : Settlements above TBM compared with overburden minus face + shield pressures.

5 >> Conclusions
We are witnessing a revolution, and it is not over. Progress in Measurements made in both clays and sands, at depths ranging
pressurized TBM operation has reached the point where consistent from 10 to 60 meters, confirmed that filling and pressurazing of the
control of ground movements is being achieved, preventing damaging gaps was preventing ground loss. The small displacements around
ground movements. the advancing TBM were related to the stress changes due to the
differential between overburden and face/shield pressures. Once the
Consistently controlling pressures and minimizing settlement throughout gaps are filled, displacements can be reduced further, if needed, by
the tunnel drive serve as a demonstration to project participants and increasing the upper face and shield pressure, The condition would
communities alike that structures along the tunnel alignment will be be most likely a concern in soft soils, where there are significant
protected. pressure differences. Displacements due to the pressure differential
on the Alaskan Way Replacement Project were not measurable
COORDINATION when the tunnel was at shallow depth and were at least an order of
magnitude below allowable settlements throughout.
The ground control achieved for EPBM tunneling at shallow depth
and for large diameter TBMs has benefited from and depended on a Monitoring of the piezometric pressures around the TBM and
coordinated, collaborative effort between owner and contractor teams. comparing them to the face/shield gap pressures provided
The Protocol for Advancing the Tunnel instituted on the Porto Metro information needed to determine the effectiveness of the face and
and the Construction Monitoring Task Force in Seattle formalized the shield gap conditioning, injections, and pressures in creating a
process. In these cases, the events and changes on the projects hydraulic gradient sufficient to support the groundwater and the
resulted in the parties seeking special, cooperative, interactive efforts. effective active forces in the face and above the shield.
Reach by reach tunnel logs of key operating parameters were used on

the Porto Barcelona line 9, and Toronto Northern Tunnels Projects. Daily
tunnel logs used on the alaskan Way Viaduct project provided targets
and green range for key TBM operating parameters and automatic
settings. The Construction Monitoring Task Force reviewed the values
obtained for each of the ring advances from the previous day.
The pressurized tunnel boring machine employs complex systems,

produces a plethora of electronic data, as well as some critical non-
electronic observations, and engages a multitude of engineering
disciplines, underscoring the importance of a coordinated effort among
owner and contractor teams, regardless of the contracting procedures
and contract form.
Extensive electronic and non-electronic data is also obtained from

monitoring the ground and structures. It is most helpful to have a
geotechnical monitoring team embedded within the TBM operation to
interpret and correlate interpret and correlate the information with key
TBM operating parameters.
MONITORING GROUND BEHAVIOR AT THE SOURCE
The cases presented in the paper highlight the importance of

monitoring ground behavior at its source in order to understand the
causes of ground movement and changes in groundwater pressure
and to confirm that control is being achieved with the pressurized TBM
operation.
On the Alaskan Way Viaduct Replacement tunnel, monitoring of

ground behavior at the source was conducted with combination
extensometers/piezometers at an average spacing of 16 m, which
provided an almost continuous view of the effect of TBM face and shield
gap pressures on ground displacement and changes in groundwater
pressures. Figure 39 : Alaskan Way Viaduct Replacement EPBM in in exit shaft.
5 >> Conclusions
PREVENTION OF GROUND LOSS Tunneling at shallow depth requires special investigations and special
control, but the model that assumes a given percentage of ground loss
The pressurized TBM is more than a pressurized face TBM, it has does not represent the behavior of a well-controlled pressurized TBM
a presssurized envelope that extends from the cutterhead, around operation and can lead to the conclusion that the tunnel should be
the body of the shield to the grout being injected through the tail of deeper or smaller or the damage will be moderate to severe and ground
the shield to fill the gap around the lining, throughout the tunnel drive. improvement procedures are required, rather than focusing on controlling
Positive filling and pressurization of gaps prevents ground loss. the TBM operation. The ability to control ground movements with the
tunneling process reduces the need for such procedures.
Estimates of ground loss percentages based on small diameter shields
will result in overestimation of the ground loss for a large diameter At the same time, there are many situations where ground improvement
shield. For example, the perimeter increases in proportion to the methods are essential to the pressurized TBM operation. They are used
diameter but the gap dimension remains relatively small, so that the on most projects, particularly in conditions where the TBM cannot be fully
gap dimension increases closer to the diameter than the square of pressurized. Drilling of multiple holes for ground improvement requires the
the diameter. use of control measures that fit the ground and prevent ground loss. In
some cases, the pressurized TBM was far better preventing ground loss
than the drilling operation.
Figure 40 : Alaskan Way Viaduct Replacement EPBM in exit shaft.

6 >> Acknowledgement
I have been privileged to work with individuals and organizations that are committed to excellence
in building underground projects and in communicating and advancing the state of the tunneling
art. Some of the projects on which we have worked are presented in this lecture. I am particularly
indebted to Harvey Parker, William Hansmire, Anthony Stirbys, Hossein Bidhendi, Michael DiPonio,
Jason Choi, William Edgerton, Randall Essex, Glen Frank, Brian Fulcher, and Peter Raleigh for their
perspectives and contributions on the projects and topics of this lecture, and am most appreciative
of the opportunities we have had to work together over the years.
Monitoring and Controlling Ground Behavior at the Source-Recent Applications to Pressurized Tunneling 33
7 >> References
Anagnostou, G. 2014. Some critical aspects of subaqueous tunneling. Muir Wood Lecture, International Tunneling Association.
Bono, R., Ortu, M., Valdermarin, F. 2008. Surface settlement minimization in soft soil when excavating with an earth pressure
balance shield. Jornada Tecnica: Tuneles con EPB Simulacion y Control de la tuneladora, Barcelona.
Boscardin, M., and Cording, E. J. 1989. Building response to excavation-Induced settlement. Journal of Geotechnical
Engineering, ASCE, Vol. 115, pp. 1-21.
BTS/IOM3 Joint Meeting. 2006. Tunnelling the Porto Metro – Under Pressure in Porto Granite.
Burland, J.B. and Wroth, C.P.1974. Settlement of buildings and associated damage. State of the Art Report. Conference on
Settlements of Structures. Cambridge.
Burland, J.B., Broms, B.B., and De Mello, V.F.B. 1977. Behavior of foundations and structures. State of the Art Report. Proc.
9th Intl. Conf. on Soil Mechanics and Foundation Engineering II, Tokyo, Japan, pp 495-546.
Burland, J. B., 2008. The Assessment of the risk of damage to buildings due to tunnelling and excavation. Jornada Tecnica 16-
12-08, Movimientos de Edificios Inducidos por Excavaciones: Criterios de dano y gestion del riesgo. Escola Tecnica Superior
d’Enginyers de Camins, Canals i Ports de Barcelona, Universitat Politecnica de Catalunya.
Choueiry, E., Elioff. A., Richards J., Robinson, B. 2007. Planning and construction of the Metro Gold Line Eastside Extension
Tunnels. Rapid Excavation and Tunneling Conference Proceedings, Society for Mining, Metallurgy and Exploration, ed. M. T.
Traylor and J.W. Townsend, pp 472-494.
Cording, E. J. and Hansmire, W.H. 1975. Displacements around soft ground tunnels. General Report, Session 4, 5th Pan-
American Congress on Soil Mechanics and Foundation Engineering, Buenos Aires.
Cording, E. J., Long, J.H., Son, M., and Laefer, D. 2001. Modelling and analysis of excavation-Induced building distortion and
damage using a strain-based damage criterion. London Conference on Responses of Buildings to Excavation-Induced Ground
Movements, Imperial College, London.
Cording, E. J., 2013. Tunneling in Chicago Clay: Pioneering work in ground control. 7th International Conference on Case
Histories in Geotechnical Engineering.
Cording, E., Nakagawa, J, Painter, C., McCain, J., Vazquez, J., Stirbys, A., 2015. Controlling ground movement on the SR 99
Alaskan Way Viaduct Replacement Tunnel. Rapid Excavation and Tunneling Conference Proceedings, ed. M. Johnson and S.
Oginski, Society for Mining, Metallurgy and Exploration, 2015, pp 493-508.
Cording, E., Nakagawa, J., McCain, J., Stirbys, A., Sowers, D., Vazquez, J., Painter, C., 2017. Managing ground control
with earth pressure balance tunneling on the Alaskan Way Viaduct Replacement Project. Rapid Excavation and Tunneling
Conference Proceedings, ed. C. Lawrence and A. Del Vescovo, Society for Mining, Metallurgy and Exploration, pp 82-98.
Diez, R. and Williams, R.D. 2003, TBM Tunneling under very low cover – Approach to Trinidade Station, Porto Metro (Portugal).
DiPonio, M., Frank, G., Gharahbagh, E., Cording, E. 2012. Settlement risk of extended overcut in dense soils. Proceedings,
North American Tunneling Conference, pp 54-64.
Donegan, B., 2017. Presentation at Cutting Edge Conference, Seattle, November.
Escoda, R. 2017. Personal communication.
Escoda, R., Magro, J. L., Vazquez, J. 2017. SR 99 bored tunnel in Seattle: Performance and challenges of “Bertha” the largest
TBM ever. Rapid Excavation and Tunneling Conference Proceedings, ed. C. Lawrence and A. Del Vescovo, Society for Mining,
Metallurgy and Exploration, pp. 472-487.
Fernandez, E., Houser, G., Gonzalez, F., Herranz, C., Herten, A. 2017. The crossing under the Alaskan Way Viaduct. Rapid
Excavation and Tunneling Conference Proceedings, ed. C. Lawrence and A. Del Vescovo, Society for Mining, Metallurgy AND
Exploration, pp 226-237.
Guglielmetti, V., Grasso, P., Mahtab, A., Xu, S., ed. 2008. Mechanized tunneling in urban areas, CRC Press.
Garahbagh, E. A., DiPonio, M. A., Raleigh, P., Hagan, B. 2012, Keeping the chamber full: managing the “air bubble in EPB
tunneling, Proceedings North American Tunneling Conference, pp. 1065 – 1073.

7 >> References
Hansmire, W.H. and Cording, E. J. 1985. Soil tunnel test section: Case history summary. Journal of Geotechnical Engineering
Division, ASCE, Vol. 111, No. 7, pp. 1301-1320.
Hansmire, W. H, Sowers, D. l., Everett, S.L., Phelps, D. J. 2011. Settlement Considerations for Designing the Alaskan Way
Tunnel. Rapid Excavation and Tunneling Conference Proceedings, Society for Mining, Metallurgy and Exploration, ed. S
Redmond and V. Romero, pp 751-764
Hansmire, W., McLane, D., Frank, G., Bragard, C. 2017. Challenges for tunneling in downtown Los Angeles for the Regional
Connector Project, Proceedings of the International Tunneling Association World Tunneling Congress
Kawamura, N. and Cording, E.J. 1999. Long-term behavior of tunnels in Chicago Clay. Proceedings, 3rd National Conference
of the Geo-Institute, Geo-Engineering for Underground Facilities, University of Illinois at Urbana-Champaign, Geotechnical
Special Publication No. 90, pp 866-878.
Kramer, G., Bidhendi, H., Cording, E., Walters, D., Poon, E. 2015. TYSSE tunneling test section and excavation beneath
Schulich Building. Rapid Excavation and Tunneling Conference Proceedings, ed. M. Johnson and S. Oginski, Society for
Mining, Metallurgy and Exploration, pp 690-706
Mair, R. 2011. Tunneling in Urban Areas and Effects on Infrastructure; Advances in Research and Practice, Muir Wood Lecture,
International Tunneling Association.
Mosavat, K and, Mooney, M. 2015. Examination of excavation chamber pressure on a 17.5-m diameter earth pressure balanced
tunnel boring machine. International Conference on Tunnel Boring Machines in Difficult Grounds, Singapore.
Muir Wood, A. M. 1994. The Thames Tunnel 1825 to 1843: Where shield tunneling began. Proceedings of the Institution Civil
Engineers, Civil Engineering 102, August.
National Coal Board, Subsidence Engineers Handbook, National Coal Board Production Department, U. K.
Peck, R. B. 1969. Deep excavations and tunneling in soft ground. State of the art, Proceedings of the 7th International
Conference on Soil Mechanics and Foundation Engineering, Mexico City, pp 225-290.
Raleigh, P. 2006, Tunneling the Metro do Porto - Under pressure in Port Granite, Proceedings Tunneling Association of Canada.
Robbins, D. 2013. A Tradition of Innovation: The next push for machine tunneling. Muir Wood Lecture, International Tunneling
Association.
Robinson, R.A., Kucker, M.S., Parker, H.W. 1991. Ground behavior in glacial soils for the Seattle transit tunnels. Proceedings
Rapid Excavation and Tunneling Conference, Vol. 2, Ch. 7, pp. 93-117.
Robinson, B. and Bragard, C. 2007. Los Angeles Metro Gold Line Eastside Extension – Tunnel Construction Case History,
Rapid Excavation and Tunneling Conference Proceedings, Society for Mining, Metallurgy and Exploration, ed. M. T. Traylor and
J. W. Townsend, pp 472-494.
Robinson, R.A., Sage, R., Cording, E., Clark, R., Raleigh, P., Wiggins, C. 2012. Conveyor belt weigh scale measurements, face
pressures and related ground losses in EPBM tunneling. Proceedings, North American Tunneling Conference, 2012, pp 65-72.
Robinson, R., Sage, R., Cording, E, Clark, R. 2013. Lessons learned from EPBM tunneling on the Sound Transit Beacon Hill
Project, Seattle, Washington. International Tunneling Association, World Tunnel Congress.
Schmidt, B. 1969. Settlements and ground movements associated with tunneling in soil. Ph.D. thesis, University of Illinois,
Urbana.
Skempton, A. W. and MacDonald, D. H. 1956. Proc. Inst. Of Civ. Engrs., Vol. 5, Part III, pp. 727-784.
Skempton, A.W. and Chrimes, M. M. 1994. Thames Tunnel Geology, Site investigation and geotechnical problems. Geotechnique
44, No. 2, pp 191-216.
Srisirirojanakorn, T. 2005. Pore pressure response and ground displacements in Chicago Clay during tunneling and over long
term. Ph.D. thesis, University of Illinois at Urbana-Champaign.
Terzaghi, K.1942a. Liner plate tunnels on the Chicago, Illinois Subway, ASCE Proceedings, V. 68, No. 6
Terzaghi, K. 1942b. Shield tunnels of the Chicago Subway. Journal of the Boston Society of Civil Engineers, V 29, No. 3,
longrine 04 90 14 48 48 - (21791 - 01/19)
ITA Secretariat - c/o MIE2 – Chemin de Balexert 9 - CH-1219 Châtelaine (GE) - Switzerland
Tel. : + 41 21 693 23 10 - Fax : + 41 21 693 41 53 - Email : secretariat@ita-aites.org - Web : www.ita-aites.org

Muirwood Janvier 2019 BD Page A Page

Uploaded by

Copyright:

Available Formats

Muirwood Janvier 2019 BD Page A Page

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Muirwood Janvier 2019 BD Page A Page

Uploaded by

Copyright:

Available Formats

Muir Wood Lecture 2018

Monitoring and Controlling Ground Behavior at the Source

N° ISBN : 978-2-9701122-4-2 aPril 2018

Monitoring and Controlling Ground Behavior at the Source

2.2 TUNNELING WITH OPEN FACE SHIELDS....................................................................................7

PRESSURIZED FACE TBMS.......................................................................................................8

2.4 PIONEERING INVESTIGATIONS CORRELATING THE SOURCES OF GROUND MOVEMENT WITH

SURFACE SETTLEMENT, CHICAGO: 1939-1941...........................................................................9

2.5 GROUND MOVEMENTS DURING SHIELD TUNNELING ON WASHINGTON D.C. METRO:

2.6 SHIELD TUNNELING IN THE CHICAGO CLAY: 2000....................................................................11

3.2 CONTROL OF REGULAR GROUND LOSS BY FILLING AND PRESSURIZING GAPS.........................14

3.4 EPBM TUNNELING BENEATH SCHULICH BUILDING, TORONTO: 2012.........................................16

3.5 GROUND CONTROL ON LOS ANGELES METRO TUNNEL PROJECTS..........................................19

4.2. SITE CONDITIONS ON THE ALASKAN WAY VIADUCT REPLACEMENT PROJECT...........................21

4.3. ASSUMED VOLUME LOSS VS ACTUAL SURFACE SETTLEMENT.................................................22

4.4. INNOVATION AND RISK.........................................................................................................23

4.5. COORDINATION PLAN...........................................................................................................24

4.7. TBM ADVANCE WITH INCREASING DEPTH TO BNSF TUNNEL....................................................28

4.8. SETTLEMENT IN SANDS AT NORTH END OF TUNNEL...............................................................29

4.9. SETTLEMENT DUE TO DIFFERENTIAL PRESSURE ....................................................................30

MONITORING GROUND BEHAVIOR AT THE SOURCE .....................................................................31

PREVENTION OF GROUND LOSS.................................................................................................32

4 Muir Wood Lecture 2018

Current pressurized tunnel boring machines (TBMs), described in Section

Projects with large diameter TBMs, described in Section 4, of necessity,

On the Alaskan Way Viaduct Replacement tunnel, monitoring of ground

6 Muir Wood Lecture 2018

2.1 THE THAMES TUNNEL 1825 TO 1843 : WHERE SHIELD

The tunnel was large. It was excavated as a single 12-m-wide box

The tunnel is in use today on the London Underground.

2.2 TUNNELING WITH OPEN FACE SHIELDS

Barlow designed a 2.5-m circular shield, which Greathead drove

Efforts to improve productivity led to mechanization in the 20th

Figure 1. Open Face Shields.

8 Muir Wood Lecture 2018

In urban tunneling, ground control procedures need to extend to the

2.5 GROUND MOVEMENTS DURING SHIELD TUNNELING ON

The University of Illinois, under contract to Washington Metro, conducted

10 Muir Wood Lecture 2018

• Three-dimensional pattern of vertical and lateral movements in the

2.6 SHIELD TUNNELING IN THE CHICAGO CLAY: 2000

This condition was observed during tunneling in Evanston, Illinois in the

Figure 6 : Time dependent changes in excess porewater that caused

Consolidation in TS 4 occurred in a zone around the tunnel, and caused a

12 Muir Wood Lecture 2018

14 Muir Wood Lecture 2018

3.2.3 Filling and pressurizing the steering (overcut) gap

With EPBMs, there is evidence that conditioned muck is more likely to

3.3 EPBM TUNNELING AT CAPITOL HILL, SOUND TRANSIT

3.4 EPBM TUNNELING BENEATH SCHULICH BUILDING,

3.4.1 Plan for controlling settlement using compensation

On the Northern Tunnels Contract for Toronto-York Spadina

Figure 14 : York University Schulich Building.

The plan was to perform compensation grouting from three shafts,

Imposing the 31 mm ground movement on the structure would result in

3.4.2 Relationship between building distortion and damage

The relationship in Figure 16 represents the state of strain at a point and

16 Muir Wood Lecture 2018

2.4 PIONEERING INVESTIGATIONS CORRELATING THE SOURCES OF GROUND MOVEMENT WITH

2.5 GROUND MOVEMENTS DURING SHIELD TUNNELING ON WASHINGTON D.C. METRO:

2.6 SHIELD TUNNELING IN THE CHICAGO CLAY: 2000....................................................................11

• Three-dimensional pattern of vertical and lateral movements in the