WG2 - ITA REPORT DesignSegment PDF
WG2 - ITA REPORT DesignSegment PDF
WG2 - ITA REPORT DesignSegment PDF
Tunnel Linings
TABLE OF CONTENTS.............................................................................................................................4
MAIN AUTHORS, CONTRIBUTORS AND REVIEWERS IN ALPHABETIC ORDER...............................6
LIST OF SYMBOLS AND ABBREVIATIONS.............................................................................................7
PREFACE.................................................................................................................................................10
1 SCOPE AND LIMITATIONS..................................................................................................................11
2 DESIGN PHILOSOPHY .......................................................................................................................12
2.1 Limit state or load and resistance factor design ........................................................................................ ..........12
10 GASKET DESIGN...............................................................................................................................10
10.1 Gasket Materials..................................................................................................................................... ..........40
Main Authors
Mehdi Bakhshi, Aecom
Verya Nasri, Aecom
Contributors
Elena Chiriotti, Incas Partners
Ron Tluczek, Gibb
WG2’s Reviewers
Christoph Eberle, Mott MacDonald
Gernot Jediltschka, Geoconsult
Giovanni Plizzari, University of Brescia
Jon Hurt, Arup
Markus Thewes, Ruhr University
ITA’s Reviewers
Benoît de Rivaz, BM-Underground
Chungsik Yoo, Sungkyunkwan University
Eric Leca, Arcadis
A = effective tension area of concrete around rebar divided by number of steel bars, mm2
Ad = load distribution area inside segment under thrust jack forces, mm2
Ag = gross area of concrete section, mm2
Aj = area of contact zone between jack shoes and the segment face, mm2
As = area of reinforcing bars, mm2
a = distance from edge of vacuum lift pad to edge of segment in the load case of stripping (demolding), or dimension
of final spreading surface under thrust jack forces, mm
al = transverse length of contact zone between jack shoes and the segment face, mm
at = transverse length of stress distribution zone at the centerline of segment under thrust jack forces, mm
b = average width of tunnel segment also known as the ring length, or width of tested specimen, m
C = cos(θ) in the elastic equation method
C2 = cos2(θ) in the elastic equation method
C3 = cos3(θ) in the elastic equation method
Cc = compression force in the concrete section, N
Ct = tensile force in the section due to fiber reinforcement, N
De = external diameters of the tunnel segmental lining, m
Di = internal diameter of the tunnel segmental lining, m
d = thickness of tested specimen, or total width of the segment cross section, mm
d1 = length of load transfer zone for the case of longitudinal joint bursting load, mm
dburst = centroidal distance of bursting force from the face of section, mm
dc = concrete cover over rebar, mm
dk = contact width for the case of longitudinal joint bursting load, mm
ds = distributed width of tension block inside the segment for the case of longitudinal joint bursting, mm
E = modulus of elasticity of concrete, MPa
Ecm = modulus of elasticity of concrete, MPa
Es = oedometer stiffness of the ground also known as Eoed; or modulus of elasticity of rebar, MPa
EI = flexural rigidity of segmental lining considering a unit width, N.m2
e = eccentricity, mm
el = eccentricity of normal (axial) force defined as M/N, mm
ek = eccentricity of joint contact zone considering gasket and caulking recesses, mm
eanc = eccentricity of jack pads with respect to the centroid of cross section, or maximum total eccentricity in longitudinal
joints consisting of force eccentricity and eccentricity of load transfer area, mm
F = forces acting on bottom segment due to self-weight of segments positioned above, N
Fsd = bursting tensile forces developed close to longitudinal joints, N
Fsd,r = spalling tensile forces developed close to longitudinal joints, N
Fsd,2 = secondary tensile forces developed close to longitudinal joints, N
fy = yield stress of required reinforcing bars, MPa
f1 = first peak flexural strength, MPa
fbot = stress at the extreme bottom fiber of concrete section, MPa
f’c = specified compressive strength of concrete segment, MPa
fcd = concrete design compressive strength, MPa
fck = concrete characteristic compressive strength, MPa
fctd = fiber-reinforced concrete design tensile strength, MPa
fct,eff = concrete tensile strength, MPa
f’co = compressive strength of partially loaded concrete surface, MPa
qe1 = horizontal earth pressure at crown of lining applied to the elastic equation method, MPa
qe2 = horizontal earth pressure at invert of lining applied to the elastic equation method, MPa
qw1 = horizontal water pressure at crown of lining applied to the elastic equation method, MPa
qw2 = horizontal water pressure at invert of lining applied to the elastic equation method, MPa
R = radius from centerline of lining, m; or minimum radius of alignment curvature, m; or reaction force in the scheme of
segments stacking for storage and transportation, kN
ro = radius of excavated tunnel, m
s = maximum rebar spacing, mm
S = distance between stack supports and free edge of segments in the load case of segment storage, m; also sin(θ) in
the elastic equation method
S2 = sin2(θ) in the elastic equation method
S3 = sin3(θ) in the elastic equation method
ss = sample standard deviations of test results
sr,max = maximum crack spacing, mmTburst = bursting force, N
w = segment self-weight, kg/m; or maximum crack width, mm
yc = distance from extreme compression fiber to centroid of equivalent compression force in the section, mm
y = distance from extreme tension fiber to the neutral axis, mm
β = dimension of the loaded surface under thrust jack forces according to Iyengar diagram, mm; or in crack width
analysis, ratio of the distance between neutral axis and tension face to the distance between neutral axis and
centroid of rebar
ΔPg, invert = vertical gradient of radial grout pressure between the crown and invert of tunnel, MPa
δ = displacement of lining applied to the elastic equation method; or opening distance of longitudinal joint on one side
due to poor ring build and tunnel deformation, mm
δd = diametrical distortion, mm
εtu = ultimate tensile strain
εcu = ultimate compressive strain
ε’csd = compressive strain due to shrinkage and creep equal to 150×10-6
φ = strength reduction factor; or rebar diameter, mm
φA = outer diameter of the segmental ring
γ = material safety factor
λ = slenderness defined as the ratio between the developed segment lengths and its thickness
θ = angle from crown in the elastic equation method; or rotation of segment joints, radian
ρconcrete = specific weight of concrete, kg/m3
ρeq = equivalent specific weight of grout, kg/m3
σc,j = compressive stresses developed under jack pads because of axial effects of thrust jack forces, MPa
σcm = fully spread compressive stress in method of the Iyengar diagram, MPa
σcx = bursting tensile stresses using the Iyengar diagram, MPa
σp = specified post-crack residual tensile strength of FRC segment, MPa
τ = birdsmouthing or opening of longitudinal joint on one side due to deformation of the tunnel lining, radian or degrees
τyield = shear yield strength of grout, MPa
2.1 LIMIT STATE OR LOAD AND • Service Stages selected (i.e. thickness, width and length) with
RESISTANCE FACTOR DESIGN (h) ground pressure, groundwater respect to tunnel size and anticipated loading
pressure, and surcharge loads cases. Using load factors recommended
Currently, Limit State Design (LSD) also (i) longitudinal joint bursting load by structural codes, the design strength of
known as Load and Resistance Factor (j) loads induced due to additional the segments is compared with the required
Design (LRFD) is utilized to design precast distortion strength for specific load cases in order to
concrete tunnel segments. Limit State (k) other loads (for example, earthquake, determine the required concrete compressive
Design requires the structure to satisfy two fire, explosion, aerodynamic, strength (fc or fcd) and type and amount of
principal criteria: the Ultimate Limit State mechanical, electrical and temperature reinforcement. The final designed geometry,
(ULS) and the Serviceability Limit State (SLS). loads as well as internal loads and the compressive strength, and reinforcement are
A limit state is a condition of a structure effect of adjacent tunnels) specified in order to ensure that the precast
beyond which it no longer fulfils the relevant segmental lining satisfies all service conditions.
design criteria. LSD is a design philosophy In the design procedure, load factors are
that considers the variability in the prediction applied for specific load combinations. ITA WG2 (2000) provides a flowchart
of loads, and properties of the structural Typical load factors are shown in Table 2 of shield tunnel lining design that start
elements and employs specified limit states for various governing load cases. Utilizing with inputs related to “Planning of Tunnel
to achieve its objectives for constructability, these load factors, the resulting axial forces, Project” including alignment, geology,
safety, and serviceability. bending moments, and shear forces may be specification/code/standard to be used
calculated to design the reinforced concrete and function/capacity to be given to tunnel.
Concrete precast tunnel segments should be segmental lining. For illustration, as an These inputs are connected to different
designed using load and strength reduction example the table below provides factored design processes such as load condition,
factors for specific load combinations as load combinations for segmental tunnel lining condition, computation of lining
specified in concrete design codes such linings according to ACI 544.7R, 2016. If internal forces and check of lining safety. All
as ACI 318-14, EN 1992-1-1 (2004), ACI different load factors are provided by local these processes are connected to decision
544.7R (2016), fib Bulletin 83, (2017), codes, these should be used in preference. boxes such as safe and economical design
AASHTO DCRT-1-2010 or the WG2 report and final approval and end with execution
21373-ITA-Report-16-WG2-BD_P “Twenty 2.3 DESIGN APPROACH of construction works, For more details on
years of FRC tunnel segments practice” the design approach for precast concrete
(2016). Care should be taken to ensure that Initially an appropriate segment geometry is tunnel segments, refer to ITA WG2 (2000).
a consistent set of load or strength reduction
factors are utilized and that various codes
are not “mixed and matched”. LOAD CASE LOAD FACTORS
2.2 GOVERNING LOAD CASES AND Load case 1: stripping (demolding) 1.4w
LOAD FACTORS
Load case 2 : storage 1.4 (w + F)
Segmental tunnel linings are installed in the The tunnel intrados for water and Initially a thickness is assumed for the
rear of the TBM shield and are generally in wastewater or Combined Sewer Overflow segmental lining ring which is later
the shape of circular rings. The size of the (CSO) tunnels will be governed by the optimized during detailed design. The
ring is defined by the tunnel internal diameter design volumes of conveyed water or following guidelines are given to assist in
(intrados), thickness, and length of the ring. design storm (e.g. 1 year, 25 years, 100 choosing an initial lining thickness.
years) specified by local authorities or A review of more than 100 projects
3.1 INTERNAL DIAMETER OF THE predicted by system modeling. published in ACI 544.7R (2016), AFTES
BORED TUNNEL (2005), Groeneweg (2007) and Blom (2002)
The minimum tunnel intrados is the smallest indicates that the ratio of internal tunnel
The dimension of the tunnel intrados is tunnel which encircles all these elements. diameter (ID) to the lining thickness falls in
determined by considering the internal space Note that for the spaceproofing of road the range of 18-25 for tunnels with an ID
requirements, which depends on the intended and railway tunnels, it is crucial to consider of more than 5.5m, and 15-25 for tunnels
use of the tunnel and client requirements. the impact of maximum super-elevation with an ID of 4.0m to 5.5m. JSCE (2007)
due to the rotation of clearance envelopes. recommends that the ring thickness should
For railroad and subway tunnels, the inner In determining the tunnel intrados, be less than 4% of the outer diameter of the
dimensions of the tunnel are typically allowance should be made for construction segmental ring, which translates into an ID
governed by the train clearance envelope tolerances. For example, DAUB (2013) to thickness ratio of 23. For tunnels under
(i.e. structural gauge), track structure, recommends a radius tolerance of R = 4m diameter, no correlation could be found
drainage trough, structure of the overhead ±100 mm for TBM-bored tunnels. and the lining thickness generally ranges
catenary, when applicable, and emergency from 150mm to 280mm. Here the lining
walkways (i.e. egress space). For double 3.2 THICKNESS OF THE SEGMENTAL thickness is dictated by construction and
track tunnels, the tunnel intrados is LINING RING design loading requirements.
additionally governed by the distance
between the centers of tracks. Having set
the tunnel intrados, the utility services (i.e.
electrical, water, telecommunication, etc.)
are installed in the unoccupied space.
Sufficient ventilation space is generally
provided if egress space and cross
passageways are provided (RTRI, 2008).
The segment wall thickness should optimized for the efficiency of the tunnel sequence of alternating right-tapered and
provide sufficient space and clear distance works. Depending on the tunnel diameter, left-tapered rings produces a straight drive.
for gaskets and caulking recesses. The the ring length can range between 0.75m Alternatively, a sequence of right-tapered
minimum segment wall thickness must also and 2.50m (DAUB, 2013). On one hand, rings or left-tapered rings results in a curve
be compatible with the bearing surface it is desirable to have a short ring length with the minimum radius of curvature.
area of the TBM longitudinal thrust cylinders for ease of transportation and erection, Up and down directional corrections are
(AFTES, 2005). construction of curved sections, and to achieved by rotating the tapered segment
reduce the length of the shield tail. On the ring through 90° (ÖVBB, 2011). This ring
To achieve a robust design, the segment other hand, it is desirable for the ring length system provides good sealing performance
thickness should be capable of handling all to be larger to reduce production cost, for an impermeable tunnel but the
loading cases and service conditions. The the number of joints, the total perimeter of requirement for different sets of formwork is
lining thickness may be increased in order segments and gasket length, the number a disadvantage.
to cater for unforeseen loads, particularly if of bolt pockets where leakage can occur,
sealing gaskets are installed. as well as increasing the construction Currently, the universal ring system is the
speed (JSCE 2007). Generally, for smaller most conventional system, where both
In addition to structural factors, the final diameters, the available space for segment circumferential faces of the ring are inclined
lining thickness should consider durability supply and handling defines the limitation to the tunnel axis. As indicated in Figure
and if a sacrificial layer is considered for of the ring length, whereas for larger 2, the ring taper is split between the two
the design life of the tunnel, (for instance, diameters, segment weight and production circumferential faces and all curves and
for CSO tunnels) then the sacrificed layer are the limiting factors. Typically, a ring directional corrections can be negotiated
thickness should be added to the required length of 1.5m would be used for tunnel through the rotation of the segmental ring.
structural thickness. diameters of 6m to 7m, increasing to a ring The main advantage of this system is that
length of 2m for tunnels larger than 9m in only one type of formwork is required
The TBM shield outer diameter is diameter. (ÖVBB, 2011). The required ring taper (k)
determined by adding the tail clearance to cater for the designed alignment can be
and shield skin plate thickness, also known 3.4 SEGMENTAL RING SYSTEMS calculated with the following formula :
as overcut, to the segmental lining outer
diameter (RTRI, 2008). The shield outer Different systems exist for tunnel segmental
Eq. (1)
diameter has an influence on the minimum rings, these include parallel rings, parallel
radius of curvature of the alignment. A rings with corrective rings, right/left-tapered
review of the more than 100 tunnel projects rings, and universal ring systems (see Figure
with different sizes (JSCE 2007) indicates 2).
Besides the minimum radius of curvature
that when the shield outer diameter is less
(R), a correction curve radius should be
than 6m, between 6-10m, or more than Parallel ring systems consist of rings with
catered for which assists in returning a TBM
12m, the minimum radius of curvature parallel end faces and with circumferential
back to the designed tunnel alignment. The
can be limited to 80m, 160m, and 300m, faces perpendicular to the tunnel axis.
correction curve radius should be at least
respectively. In practice, however, larger This system is not inherently suitable for
20% less than the smallest designed curve
radii are being utilized and the above- curves and is problematic where packing
radius (DAUB, 2013).
mentioned limits may be considered to between the rings is required to restore line
be lower bound limits for the radius of and grade. These joints cannot always be
Using the universal rings, a straight
curvature. It should be noted however, properly sealed as the packing reduces the
alignment can be achieved by rotating
that the minimum radius of curvature is compression in the gasket.
each alternate ring by 180°. Horizontal and
a function of ring geometry (taper, ring
vertical curves can be negotiated by partial
width), overcut, shield design (articulated The parallel rings with corrective rings
rotation of the rings. In recent years, by
or not) and radial gap between segment system utilizes the corrective rings for
using advanced software for guiding TBM’s,
and tailskin rather than just the shield outer directional corrections. With this system, the
universal rings can produce straight drives
diameter. All these parameters should be requirement for different sets of formwork is
with the key segments always above the
taken into account when determining the the main disadvantage.
springline by adjusting the drive error (of
minimum curve radius.
less than a few millimeters) in two or three
The right/left ring system generally consists
consecutive rings.
3.3 LENGTH OF THE RING of tapered rings with one circumferential
face perpendicular to the tunnel axis and
The length of the lining ring needs to be the other one inclined to the tunnel axis. A
Rhomboidal or parallelogrammic-trapezoidal
systems are assembled from ordinary
segments in the shape of a parallelogram with
key and counter key segments in the shape
of a trapezoid (Figure 4d). The assembly
procedure is carried out by initially installing
the counter key trapezoidal element and then Figure 4c : Trapezoidal system
placing alternate parallelogrammic segments
(left and right) around the ring. Ring assembly
is completed by inserting an often smaller d
trapezoidal key segment. This system is
currently the most common system as it
eliminates crucifix joints, has improved sealing
performance and allows for continuous ring
erection. Other major advantages are that the
angled segment joints prevent early rubbing
of the gaskets during segment insertion and
also facilitate the use of fast connecting dowels
in circumferential joints. Bolted connections
are usually required for longitudinal joints but
in certain cases, longitudinal bolts may be
replaced by guiding rods (see later). Figure 4d : Rhomboidal system
Figure 7 : Segment handling: (a) using lifting devices; (b) using vacuum lifter; and (c) diagram of pullout capacity of concrete during handling.
Construction loads on the segmental lining of Mines (CSM) (Rostami 2008) can be used to account for the increased loading
include the tunnel boring machine (TBM) used to evaluate the rock thrust based on on the convex side is to double the jacking
jacking thrust loads on the circumferential rock strength, tunnel diameter, and cutter loads. Different jacking loads on individual
ring joints and the pressures exerted characteristics (Bakhshi and Nasri 2013a). segments should also be taken into
against the exterior of the completed For soft ground tunneling applications, the account. When the machine characteristics
rings during grouting operations. Precast method presented by JSCE (2007) can be are unknown, a load factor should be used.
concrete segments are designed to used to calculate penetration resistance For illustration, as an example, ACI 544-7R
resist significant bursting and spalling based on earth or slurry pressure that acts (2016) recommends a value of 1.2 for this
tensile stresses that develop along the at the cutting face. load factor. However, when the machine’s
circumferential joints due to advancement characteristics and total thrust are known,
of the TBM. The segments must also be the maximum jacking forces cannot be
able to resist the axial forces and bending exceeded and a load factor of 1.0 should
moments that develop when the annular be used.
space between the segment and the
ground is pressure-filled with grout, initially Different analytical and design methods are
during backfilling of the tail skin void and available which include simplified equations
then during secondary grouting that may for bursting forces (ACI 318-14, DAUB,
be required to ensure complete contact is 2013), the Iyengar (1962) diagram, and
achieved with the ground. two- and three-dimensional finite element
simulations. These are discussed in the
5.1 TUNNEL BORING MACHINE following sections.
THRUST JACK FORCES
5.1.1 Simplified equations
After assembly of a complete ring, the
tunnel boring machine (TBM) advances For post-tensioned anchorage zones in
by thrusting against the most recently pre-stressed concrete sections, structural
assembled ring (a partially assembled concrete codes such as ACI 318-14 permit
ring is shown in Figure 8a). As part of this the use of simplified equations (Eq. 2) to
process, the TBM jacks bear against the determine the bursting force, Tburst, and the
jacking pads placed along the exposed centroidal distance from the face of the
circumferential joint. High compression section, dburst. These equations determine
stresses develop under the jacking pads the forces and stresses that develop in
which result in the formation of significant the circumferential joints due to TBM
bursting tensile stresses deep within the advancement. DAUB (2013) recommends
segment (see Figure 8b). Furthermore, disturbance area disturbance similar equations (Eq. 3) specifically for the
spalling tensile forces are generated for strut stress in area for design of tunnel segments.
transverse direction strut stress
between adjacent jack pads along the
in radial
circumferential joint. direction
Figure 8(a) : Thrust jacks pushing on circumferential
Due to the various geologic materials that joints; and (b) schematics of a simplified disturbance
can be encountered, different methods area of strut under TBM jack shoes (Groeneweg 2007). (Eq. 2)
are used for estimating TBM thrust. For
tunnel excavation through rock, TBM
thrust can be estimated by summing the Once the required machine thrust has
forces required to advance the machine. been estimated for the ground conditions,
These forces include the forces necessary the average thrust force per jack pair is (Eq. 3)
for boring through the rock, the friction determined. On sharp curves, the machine
between the surface of the shield and the thrust is higher on the convex side of the The schematic representation of these
ground, and the hauling resistance of back- curve than on the concave side and the equations is shown in Figure 9. If no specific
up systems. Methods proposed by Fukui different jacking loads on the segments value has been provided for eanc, then the
and Okubo (2003) or the Colorado School should be considered. A simple technique eccentricity of the jacking forces is generally
assumed to be 30mm. Figure 9 and Eqs. 14 recommends a formula to design the 5.1.3 Finite element method simulations
2 and 3 represent the radial bursting bearing strength of concrete (Eq. 7) with
stresses in the circumferential joints, but a partially loaded segment face. DAUB As shown in Figure 11, due to the
these equations are equally applicable to (2013) recommends a similar formula which concentration of the jacking forces, in
the tangential bursting stresses developed is specifically used in the design of tunnel addition to the bursting stresses under the
in the circumferential joints provided any segment faces. jacking pads, spalling stresses develop in
eccentricity of the jack shoe relative to the the areas between the jacking pads and
joint is accounted for. (Eq. 7) in the areas between the jacking pads
and the end faces of the segments. This
problem has been analyzed with the three-
5.1.2 The Iyengar diagram dimensional finite element method (FEM),
(Groeneweg 2007, Bakhshi and Nasri
The analytical method utilizing the Iyengar 2013b; 2014d).
(1962) diagram may also be used to
calculate the bursting tensile stresses in While both linear elastic and non-linear
the design of tunnel segments (Groeneweg FEM simulations can be performed for
2007). Again, the extent of load spreading strength design, the latter is considered
and the resulting magnitude of the tensile more suitable for service design as the
Figure 9 : Bursting tensile forces and corresponding stresses depends on the dimensions of the non-linear analyses capture the response
parameters recommended by ACI 318-14; and DAUB loaded surface (β), and the final distribution of the materials after failure with respect
(2013). surface (a), as shown in Figure 10. Using to crack opening. As shown in Figure 12,
this approach, the bursting tensile stresses this load case is simulated by modeling
Reinforcing rebar or fiber reinforcement is (σcx), which vary significantly from the face segments from two adjoining rings.
designed to accommodate the significant that the TBM jacks bear against to the Factored jacking forces are applied
bursting stresses developed by the jacking centerline of segment, are determined as a along the contact area between the
forces. Equations 4 and 5 may be utilized fraction of the fully disbursed compressive jacking pads and the segment face.
to determine the required area (As) of stress (σcm = F/ab). Reinforcing bars are Recesses for the gasket and stress relief
reinforcing bars with a yield stress of fy for a designed to accommodate the total grooves are modeled between the two
reinforced concrete segment. bursting tensile stresses which can be segments to simulate the transfer of
obtained from the integration of stresses force through a reduced cross section.
for radial direction or the area under the curve. The area of Just after installation, the compressive
(Eq. 4) reinforcing bar required can be determined forces within the gasket are simulated
for tangential from Eqs. 4 and 5. by applying a reaction force using solid
direction (Eq. 5) elements in the FEM program. With this
approach, the translational degrees
High compressive stresses can develop of freedom are fixed in all directions
under the jacking pads due to the TBM between the two installed segments.
thrust jacking forces. Assuming constant Results from the analysis are illustrated in
stress, these compressive stresses, σc,j, can Figure 13 which indicates the transverse
be estimated using Eq. 6. and radial bursting loads under the jack
pad and the spalling stresses in the
(Eq. 6) areas between the jacking pads. The
compressive stresses developed in this
load case are illustrated in Figure 14
Because only part of the circumferential (Bakhshi and Nasri 2013b; 2014d).
segment face is actually in contact with the
pads, the allowable compressive stresses
(f’c or fcd) can be factored to account for the Figure 10 : Iyengar (1962) diagram for determining
bursting tensile stresses (Groeneweg 2007).
strength of a partially pressurized surface.
For illustration, as an example, ACI 318-
Figure 11 : Spalling and bursting stresses in segment joints due to jack thrust forces (Groeneweg 2007). 5.2 TAIL SKIN BACK-GROUTING
PRESSURE
Figure 13 : Bursting and spalling tensile stresses Figure 15 : Backfilling of tail-skin void (Guglielmetti et al. 2007).
developed in segments due to TBM thrust jack forces
and gasket pressure: (a) transverse stresses; and (b)
radial stresses (Bakhshi and Nasri 2013b; 2014d).
using radial springs with the segments 5.4 TBM BACKUP LOAD axial forces and punching shear for all
supported radially. Linear translational critical cases (e.g. shallow cover, deep
springs have been used to represent this The TBM backup load is applied on the tunnel, etc) using axial force-bending
type of interaction. The method described segmental lining behind the shield. In order moment interaction diagrams. This load
by USACE EM 1110-2-2901 (1997) is one to control buoyancy, additional weight may case is often a governing load case in sub-
method that can be used to determine the be provided by the backup system inside sea and river crossing projects.
spring stiffness per unit of exterior tunnel the tunnel after ring installation and before
surface. Using the same grout pressure placing ballast or installation of any precast A consideration for the TBM backup load
in the crown as for cavity filling and with buoyancy unit. TBM drawings as in Figure is the early stage strength (setting time) of
the radial spring stiffness, the bending 19 indicate that this backup load is applied the tail void grout and the TBM advance
moments and axial forces developed at specific locations on the tunnel lining rate. The TBM advance rate may have to
within the lining can be determined. This intrados with two wheels on each ring. be limited to allow the tail void grout to
loading case results in small incremental For this load case, punching shear should achieve the strength required to support
axial forces with large bending moments. be considered with the assumption that the first and second gantry.
Precast segments should be designed for the backup load is applied uniformly. The
this load combination using axial force- precast segments should be designed for
bending moment interaction diagrams. the resulting factored bending moments,
Figure 18 : Modeling localized grouting pressure applied Figure 19 : Typical TBM details for a section behind the shield where backup loads are
over one-tenth of the lining (The grouting pressure applied on segmental ring.
of 250 kPa presented in this figure is indicative. The
appropriate value will be project specific).
USACE
Using the beam-spring method (AASHTO
Kr = E/(R.(1+ν)) kt = 0.5 kr/(1+ν) (DCRT-1-2010), JSCE (2007), and ÖVBB
EM 1110-2-2901 (1997)
(2011)), the lining can be modeled in cross-
Austrian (ÖVBB, 2011) Kr = Es/R kt =0
section as a series of beam elements that
JSCE (2007) Kr = not reported kt = 1/3 Kr span between the longitudinal joints of
the segments. The interaction between
AFTES (1993) Kr = E/(R.(1+ν)) kt = not reported the ground and the lining is generally
DAUB (2005) Kr = Es/R kt = 0 modeled using linear translational springs
in the radial, tangential, and longitudinal
Table 5 : Subgrade Reaction Modulus directions. Because the lining and ground
are represented by a series of beams and
springs, this method is commonly referred
to as the beam-spring method. Various
LOAD BENDING MOMENT AXIAL FORCE SHEAR FORCE two-dimensional approaches are used to
evaluate the effect of the segment joints,
including solid ring models with full bending
rigidity, solid ring models with reduced
bending rigidity (Muir Wood 1975), ring
models with multiple hinged joints, and ring
models with rotational springs.
Figure 21 (c) illustrates the various 6.1.3 Finite element method (FEM), strength; and where relevant, properties of
parameters used in the Janssen model. finite difference method (FDM,) and discontinuities such as joint spacing, joint
discrete element method (DEM) apparent dip direction and joint apparent
Two rings are required to evaluate simulations dip, peak and residual joint friction angle,
the coupling effects; however, in this peak and residual joint cohesion, joint
method, only half of the segment width In soft ground, loose rock, and partially normal stiffness, joint shear stiffness,
is considered from each ring for the homogeneous solid rock, ÖVBB (2011) Eintact/Emass, GSI and Mi (based on
longitudinal and circumferential joint and AFTES (1993) recommend using Hoek-Brown).
zone of influence. This approach utilizes the FEM and FDM methods to calculate
symmetry to remove complex support the forces in the tunnel lining. The DEM Precast segments are designed using an
conditions on both rings. Considering the method is generally considered more axial force-bending moment diagram.
self-weight of the lining, and distributing appropriate for tunnels in fractured
the ground, groundwater, and surcharge rock. A two-dimensional approach is 6.2 LONGITUDINAL JOINT BURSTING
loads along the beam, lining internal forces generally sufficient for continuous linear LOAD
can be calculated using a conventional structures that do not contain sudden
structural analysis package. changes in cross sectional geometry or Normal (hoop) forces developed in the
high concentrations of loadings. Three- lining are transferred through a reduced
a
dimensional techniques are generally used cross sectional area along the longitudinal
with more complex geometry and loadings joints where gaskets and stress relief
such as at crosscuts that intersect the grooves are present. Bursting tensile
main tunnel (ÖVBB 2011), or in soft stresses can develop along the longitudinal
ground with 3D arching. joints in a comparable manner to the
tunnel boring machine (TBM) thrust jacking
FEM is used to model the ground loads on the circumferential joints. The
surrounding the lining (Bakhshi and Nasri maximum normal force obtained from
2013c). The advantage of this method is the ground pressure, groundwater and
that one is able to determine the ground surcharge loads plus the gasket pressure
deformations and the post yielding should be applied to the longitudinal
behavior of the segmental lining materials, joints in order to obtain the maximum
including any redistribution of stress that ULS design compressive force. General
b result from deformation of the lining and equations from ACI 318-14 and DAUB
excavation of the tunnel (ÖVBB 2011). (2013), the Iyengar (1962) diagram, and 2D
FEM analysis techniques can also be used finite element method (FEM) simulations
to represent non-uniform and anisotropic are various methods utilized for the design
stresses such as when nonsymmetrical of longitudinal joint bursting (Bakhshi and
features are present in the ground. This Nasri 2014b).
can be the case when several different
geologic formations or external loads 6.2.1 General equations
are present within close proximity of an
existing structure (AFTES 1993). Using The general equation of post-tensioned
c FEM techniques, complex underground anchorage zone by structural codes
conditions, and tunnel characteristics can such as ACI 318-14 (Eq. 2) is used to
be analyzed and the large axial forces analyze this load case, where Ppu is the
and bending moments developed in the maximum normal force, and eanc is the
segments can be determined. maximum eccentricity comprising the
normal force eccentricity (M/N) and the
For DEM modeling, engineering properties eccentricity of the load transfer area.
for analysis of segmental linings in rock Alternatively, the simplified equation by
formations include properties of intact rock DAUB (2013) (Eq. 3) is used for evaluating
such as unit weight, modulus of elasticity, bursting stresses in the longitudinal joints.
UCS, internal friction angle, tensile DAUB (2013) presents more detail on
Figure 21: (a) Double ring beam-spring model with radial springs simulating ground, and joint springs simulating
longitudinal and circumferential joints; and (b) scheme of ring joint (Plizzari and Tiberti 2009), (c) presentation of parameters
used in Janssen model for determining rotational spring stiffness.
this specific load case using an approach tensile stresses, if necessary, are placed misalignment during erection of the ring.
that transfers force by means of a tension at 0.1ds and 2/3d from the face of the Such load cases are referred to as ring
block as shown in Figure 22. DAUB (2013) segment respectively. General equations, ovalization (due to the out of round ring
endorses placing additional reinforcement which include Eq. 6 and Eq. 7, can be build) or birdsmouthing (where, due to
for spalling and secondary tensile stresses used to determine the compressive stress rotation, only the inner or outer surface
when there are high eccentric normal and the required strength of the partially of the longitudinal joint is in contact). The
forces (e > d/6). Bursting, spalling, and loaded surface. design procedure assumes that the ring
secondary tensile stresses are calculated is initially built in the shape of an ellipse,
using the following equations. 6.2.2 The lyengar diagram and FEM and that the chord length of the displaced
simulations segment, as shown in Figure 23(c) does
(Eq. 14) not change. An assumption is made on the
The Iyengar diagram method (Figure 10) out of round build allowance over diameter,
and FEM simulation can be used as an say 15mm. The joint rotation causing
(Eq. alternative approach to determine the birdsmouthing (τ in Figure 23c) and
15) stresses within the longitudinal joints. opening distance due to poor ring build
Two-dimensional FEM models can be (δ in Figure 23c) are calculated. The load
used to simulate the longitudinal joint and required to close the gap under minimum/
The total eccentricity, e, consists of gasket using appropriately shaped ends maximum embedment loads is assessed
the normal force eccentricity and the to represent the recess for the gasket and and compared to the hoop force under the
eccentricity of hinge joint considering the stress relief grooves (curvature of the embedment loads. Depending on whether
gasket and caulking recesses (e = el + ek = elements is neglected in this analysis). the joint remains open or close, one of
M/N + ek), d1 = dk – 2e, and ds = 2e’ = d - Figure 23 illustrates generalized analytical the two diagrams shown in Figure 23 (d)
2el. These parameters are shown in Figure results, including bursting tensile stresses is used for calculation of birdsmouthing
22. Where appropriate, the build tolerance and compressive stresses, in the area eccentricity.
and system deflection should be taken into around longitudinal joints.
account. The above discussion only considers flat
Besides considering the normal force joints as they are the most conventional
DAUB (2013) further endorses placing eccentricity (M/N) and the eccentricity joint shape. For convex joints, the design
bursting tensile reinforcement at of the load transfer area, it may be approach is similar except a line load is
0.4ds from the face of the segments. necessary to consider the eccentricity assumed instead of a distributed load at
Reinforcement for spalling and secondary due to ovalization of the segment and the joint locations.
Figure 22 : Force transfer recommended by DAUB (2013) in longitudinal joints using the tension block concept.
c d
Figure 23 : Developed bursting tensile stresses around longitudinal joints under hoop (normal) forces due to service loading condition and gasket pressure: (a) no eccentricity; and
(b) eccentric contact stresses (Francis and Mangione 2012), (c) ring ovalization due to out of round ring build, (d) stress diagrams for joint closure (Case 1) or non-closure (Case 2)
6.4.3 Design for fire incident and depth is calculated and the resulting
explosion reduction in concrete and reinforcement
properties (modulus and strength) is
A load case specific to road and railway determined based on available data. This
tunnels is tunnel fires. One of the decisive will determine of the amount of the lining
factors for the design of tunnel structures thickness lost during the fire. If using basic
in the load case of fire is the heat release structural analysis programs, applying
rate. Rates from 10 MW up to 200 MW non-linear temperature gradients are not
have been proposed at different projects allowed for, an equivalent temperature
with the majority ranging around 100 MW. load can be established that has the same
Another important factor is the duration impact on the equivalent section as the
of the design fire with values ranging original temperature gradient has on the
from 30 minutes to 3 hours, sometimes original section (Neun, 2012). This can be
with an added cooling phase. Both heat followed by adopting a layered section
release rate and duration will depend on analysis where the normal force (N) and
intended use of tunnel and project specific bending moment (M) can be determined
conditions such as the type of traffic (train, by integration/summation of the stresses in
cars, heavy goods or dangerous goods the individual layers.
transport) and the required safety level
(with stricter requirements in case of a Explosions, on the other hand, are
possible impact on structures above the simulated by increasing the internal
tunnel, or the risk of inundation) (Neun, radial pressure on the tunnel lining at the
2012). The actual time-temperature curve service condition, by a representative
resulting from such a design fire which value such as one atmosphere (14.5
depends on the individual cross section psi or 1 bar) (Caan et al. 1998). This
and wind speeds inside the tunnel can internal radial pressure results in reduced
be used for simulation of a temperature axial forces without significant change
gradient between the intrados and in the bending moments. Recently an
extrados of the tunnel lining. However, advanced and detailed design procedure
since adequate Computational Fluid for tunnels subjected to internal explosion
Dynamics (CFD) modelling is very time and possibly preceded by fire accidents
consuming and the results are not easily was developed (Colombo et al. 2015).
transferred to a structural model, it is Simplified FE model and dynamic analyses
common to use standard deterministic were carried out to study the tunnel’s
time-temperature curves for the projects response under internal blast loads in the
(Neun, 2012). Most common standard form of pressure-impulse (p-i) diagrams
design fire curves for the structural design and an ultimate limit state criteria based on
of tunnel structures include RABT-ZTV eccentric flexural capacity (M-N interaction
(Railways) also known as EBA, RABT-ZTV diagram) was generated. Also a limit state
(Highways) also known as ZTV, ISO 834 criterion taking into account the fire-blast
Cellulose, HC Hydrocarbon Eurocode interaction was introduced through the
1, HCM Modified Hydrocarbon (HCinc) modification of the M-N diagram. This
and Rijkswaterstaat also known as RWS procedure is suggested for an advanced
(ITA WG6, 2004). The ISO 834 curve is blast-fire analysis.
recommended up to an expected fire
heat release rate of 50MW, above which
the hydrocarbon curve (up to 100MW)
and thereafter the RWS curve (up to the
stoichiometric limit) should be applied
(ITAtech Vol1, 2016). Accordingly, one
of the standard design fire curves can
be selected (temperature versus time)
and applied on the tunnel intrados. The
increase in lining temperature versus
AASHTO (DCRT-1) 34 to 48
Several guidelines are available which Not provided
(2010) (5000 psi to 7000 psi)
recommend the compressive strength of Japan’s Railway
precast concrete tunnel segments and Technical Research Not provided 42 to 60
these are summarized below. Institute (RTRI 2008)
Full-scale point load tests, as shown Figure 29 : Point load test setup and measurement instrumentation
in Figure 29, simulate the TBM thrust simulating TBM thrust jack force (Caratelli et al. 2012).
jack forces on the segment during the
excavation process (Caratelli et al. 2012),
as well as the force transfer through the a
reduced cross section in longitudinal joints.
Flexural stresses in the joints are evaluated 9.4 CRACKING VERIFICATION Fib Model Code 2010 (2013), Italian
using the maximum bending moments and standard CNR-DT 204 (2006), RILEM
corresponding axial forces obtained from Cracking in segments is a major TC 162-TDF (2003) recommendation,
analysis such as beam-spring modelling, contributor to reduction in serviceability and German DAfStb guideline (2012) can
considering the joints as reinforced due to reduction of watertightness and be used to calculate the crack width in
concrete sections with bolts acting as reinforcement corrosion. In particular, concrete sections reinforced by fibers with
tension rebar (where the bolts comply with cracking has a significant effect on the and without conventional reinforcement.
the durability and design life requirements). durability of the tunnel in an environment The flexural crack width of FRC segments
Developed stresses in the concrete at with frequent freeze-thaw cycles. Analyses is well documented in fib Bulletin No. 83
segment joints are limited to the allowable using appropriate methods should be (2017) which considers analytical sectional
compressive stress of concrete. Developed carried out to ensure that cracking in approaches as well as finite element
stresses and forces in the bolts are limited segments does not impair the serviceability, methods.
to the allowable stress of the connecting durability or intended purposes of the tunnel
bolts as published by the manufacturer. lining. Cracks induced in segments under 9.4.2 Maximum allowable crack width
service loads are mainly caused by bending
9.3 DEFORMATION VERIFICATION moments and axial forces and the designer Cracking in tunnel segments is controlled
should ensure that the flexural crack width by limiting the crack width to prevent
Segment deformations are obtained from is not greater than the allowable crack durability issues as a result of increased
the different analytical models presented width (see Table 11 below). The flexural permeability, excessive water leaks, and
above. However, joint gap and joint offset crack width calculation for reinforced reinforcement corrosion. Allowable crack
are only obtained from models that simulate concrete (RC) and fiber reinforced concrete widths are recommended by standards
joints between segments and rings. For (FRC) segments is presented below. Note and guidelines which consider the function,
SLS verification, these deformations are that in the SLS, the maximum shear force importance, service, life span, purpose,
limited to allowable values recommended developed at segment joints as a result of surrounding environment, and surrounding
by standards, guidelines and often project modeling with joint simulation should be soil conditions of the tunnel (JSCE, 2007).
specifications. As an example, the Austrian limited to the shear crack capacity. General guidelines for allowable crack
Society for Concrete and Construction widths are given in the Table 11.
Technology (ÖVBB, 2011) recommends 9.4.1 Flexural crack width in segments
allowable deformations, shown in Figure As a most comprehensive guideline,
33, for both segments and joints for tunnels The flexural crack width in reinforced the Austrian Society for Concrete and
with diameters up to 8 m (26 ft). concrete tunnel segments due to bending Construction Technology (ÖVBB, 2011)
moment and axial force is calculated using specifies the allowable crack width in
ACI 224.1R (2007), JSCE (2007) and EN segments based on the tunnel function,
1992-1-1 (2004) formulas as shown in Eqs. and corresponding watertightness
17, 18 and 19, respectively. requirements, as illustrated in Table 12.
(Eq. 17)
Leakage to be limited
fib Model Code 2010 (2013) 0.20 mm
with some surface staining
JSCE standard (2007) 0.004dc dc is the concrete cover over the rebar
Watertightness of service tunnels (i.e. utility, ASTM D2240), whereas STUVAtec (2005) Figure 37 indicates how the hydrophilic
rail and road) must be ensured during and AFTES (2005) call for a maximum insertion improves the sealing performance
design and construction in order to prevent hardness of 85. Other important properties of a composite EPDM gasket in terms of
water infiltration, minimize maintenance include tensile strength and elongation resisting higher water pressure after several
and repair costs, maintain operational (as determined by ASTM D412) which are days of immersion in water. The hydrophilic
safety, and protect mechanical and recommended to be greater than 12 MPa insertion swells under water pressure and
electrical equipment inside the tunnels. In (1700 psi) and 300 percent, respectively. acts as an extra backup to the EPDM
the one-pass segmental lining system, the Regardless of the design parameters, the profile. About 50% of the swelling occurs
watertightness of the tunnel is guaranteed ultimate performance of the gasket will within 7 days with nearly 100% of the
by the individual components of the support be tested when the lining is installed. The swelling occurring within 30 days.
system, namely the precast concrete technical solutions that engineers need
segments and the segment gaskets which to implement to achieve the required A second solution is to install two sealing
are placed between segments in the watertightness depend on specific project gaskets, one near the extrados and
longitudinal and circumferential joints. As circumstances. The important factors for one near the intrados of the segment,
shown in Figure 34, gaskets are positioned sealing gaskets are the water pressure, thus providing double security for the
around the individual segment like a frame safety factor, size of tunnel and segments, waterproofing performance. When used
and primarily near the lining extrados to gap and offset between segments, and in combination with sealing bars installed
provide the joint watertightness. tolerances. between the extrados and intrados
gaskets, isolation chambers can be created
10.2 WATER PRESSURE AND GASKET that confine any localizing leakages thus
DESIGN permitting precise repairs by grout injection
methods. The connecting gasket bars are
An important consideration in the design generally glued in place at the segment
of gaskets is the maximum ground water precast plant. Note should be taken that
pressure. Depending on the expected the watertightness of a double gasket
ground water pressure, different gasket system is defined by the higher capacity of
profiles may be selected. The first the two gaskets, not by the sum of both
generation of gaskets could only withstand gaskets’ capacity (BSI PAS 8810, 2016).
a maximum water pressure of 3 bar. Today,
with the advance of technology and limited 10.3 GASKET RELAXATION AND
Figure 34 : Segment gaskets positioned near the lining offset between adjacent segments due to FACTOR OF SAFETY
extrados for joint watertightness.
more accurate segment erection inside
the TBM, water tightness of up to 10 bar In addition to the expected ground water
10.1 GASKET MATERIALS is often achievable with a standard mono- pressure, the design has to define the
extrusion EPDM gasket profile. As shown watertightness performance of a sealing
The gasket material must be suitable for in Figure 35, water tightness between gasket and include a safety factor that
the encountered ground and groundwater segments is created through compression takes rubber relaxation effects into account.
conditions. EPDM (Ethylene Propylene of the gaskets during the assembly process It is crucial that the gasket profile and
Diene Monomer) has been established of the segments. rubber compound uphold the designed
as the preferred material, however, it is reaction force to withstand the applied
not resistant to hydrocarbons. Alternative To resist higher ground water pressures, ground water pressure years after its
materials such as Chloroprene Rubber/ two main solutions are available. The first installation. The majority of the relaxation
Styrene Butadiene Rubber (CR/SBR) are solution consists of a composite seal occurs within months after installation. The
available for such situations, but they do not which combines the two different sealing relaxation can be tested with so-called
perform well in acidic environments. technologies of an EPDM compression aging tests using an accelerated procedure
gasket and a hydrophilic seal. As shown in with elevated temperatures in order to
For the gasket to perform as intended, Figure 36, this may consist of co-extruded get results within a reasonable timeframe
several material-specific requirements gaskets with a hydrophilic layer, composite (Figure 38). As the relaxation behavior of a
must be met. One of these properties is profiles with hydrophilic cord, or designing sealing gasket is mainly influenced by the
the hardness of the rubber compound. a composite solution with a separate geometry of the gasket profile, such aging
BSI PAS 8810 (2016) requires a maximum hydrophilic seal next to the standard EPDM tests should be carried out for each profile
shore hardness of 75 (as determined by gasket. type.
Figure 35 : Water tightness between segments created through gasket compression The width of the gasket profile is a function
of the segment thickness which is a
function of tunnel diameter. The following
gasket profile widths are commonly used
with regard to the tunnel diameter:
• Tunnel Diameter < 4m,
Gasket Width = 20mm
• 4m < Tunnel Diameter < 7m,
Gasket Width = 26mm
• 7m < Tunnel Diameter < 11m,
Gasket Width = 33 or 36mm
• 12m < Tunnel Diameter,
Gasket Width = 36 or 44mm
Figure 36 : Standard, composite and double gasket solutions for projects with different watertightness requirements
Figure 37 : Effect of hydrophilic swelling cord on improving sealing Figure 38 : Typical long term relaxation test results according to ISO 11346
performance of a composite EPDM gasket in terms of resisting higher water
pressure after several days of immersion in water
Some tunnel project specifications specify time consuming and prone to failure so in Figure 41. From the watertightness
a 5 mm gap and 10 mm offset for segment practically all tests are currently carried diagram, the gasket resisting pressure
gaskets. Dowels can easily provide this out on steel specimens. Test approaches corresponding to designed gap and offset
requirement, and the segment installation vary, but following STUVAtec’s (2005) should be higher than the maximum
process is easier than when using bolts. recommendations and as shown in Figure factored working pressure for the project.
Bolts are generally the preferred connection 40, the geometric situation is simulated on
system in longitudinal joints, and in order a T-joint in the laboratory, whereby, as on
to avoid additional time and labor costs to the circumferential joint, a straight piece of
achieve the above tolerances, contractors sealing profile is pressed against the end of
tend to change to higher gap and especially a longitudinal joint. Gaskets must guarantee
higher offset values. While engineers design the water tightness under all possible gaps
gaskets for a specified gap and offset, they and offsets. Therefore, it is necessary to run
are encouraged to consider possible larger the watertightness test with different gaps
gap and offset tolerances. Although this will and off-sets. For every offset setting (0 – 20
result in a larger, more expensive gasket, mm), the test has to run through a range
it will result in potential cost saving due to of different gaps. For every gap, the water
faster erection and construction time. pressure is built up in steps of 1 bar and
is held there for 5 minutes. In this manner,
Watertightness tests using gaskets can be every combination is tested until leakage
performed on steel or concrete specimens. occurs. Plotting all “failure points” results
Working with concrete specimens is in a watertightness-gap diagram as shown Figure 41 : Typical watertightness-gap diagram
11.1 BOLTS, DOWELS AND GUIDING The bolts are metallic while the embedded In the dowel type connection, the dowels
RODS threads are generally plastic. Figure 47 are inserted into the segment during ring
shows a typical arrangement for a straight assembly and are either mortise inserted
The connections between segments within bolt. Note should be taken of the following or dove-tailed into the segment of the last
a ring and between rings can be divided geometrical details: assembled ring. Dowel connections require
into three categories; joint connections with • The pockets should be large enough for less work for the construction of the mold
bolts, dowels and guiding rods. head of the bolt and pneumatic wrench to and less manpower in the tunnel. The
be easily inserted dowels and sockets are made of plastic
In the bolt type connection, the segment • The slot side of the pocket should have a and sometimes have a core of steel. Figure
is first placed in position and then the conicity of at least 1°. 48 shows a typical arrangement of a dowel,
bolts are inserted and tightened. Bolt • The bolt slot in the segment that houses which is placed on the neutral axis at the
connections require more effort in the the plastic bolt socket should have a middle point of the segment. Because of
construction of the mold as it is necessary compatible conicity. the kinematics of the assembling process,
to create pockets and grooves into which • The bolt axis should pass through the this type of connection is only used
the bolts are inserted. It is also necessary center of the segment. between the rings in circumferential joints.
to have more personnel in the tunnel to • The distance between the end of the nut In most cases, the dowel connections
insert the bolts. This type of connection and the extrados of the segment should are used with rhomboidal and trapezoidal
is traditionally associated with rectangular be sufficient. segments to avoid early crawling or
segments and is generally used between creeping of the gaskets when the segments
rings and between segments within a ring. are being inserted.
a b c
Figure 51 : Issues with post-installed anchors: a) drilling, b) installation quality, c) fixing failure
One application of cast-in channels is in Figure 52 : Post-installed framing channels fixed to segments using connection bolts and mounting plates at the back
railway tunnel and especially high-speed
rail tunnels because of the requirement to
Figure 54 : Curved cast-in channels used as fastening system in railway and high-speed rail tunnels
Figure 55 : Modern fastening systems with tension rods for supporting intermediate slab
spalling. Circumferential length (arc) and to temperature, shrinkage and creep lack global geometrical certification.
gasket groove tolerances may have a deformations after casting. However, due Measurement systems using theodolite
major impact on the sealing performance to the fact that the temperature, relative and photogrammetry are alternative
of gaskets as a result of excessive gap or humidity and other ambient conditions methods which lack the necessary speed
offset. are maintained relatively constant in the and accuracy (to within few tenths of
production plant, any dimensional deviation a millimeter) needed for regular quality
Some project specifications do not due to temperature and shrinkage can be control. 3D industrial measurement prevails
allow for the use of segments that have neglected. Therefore, it is not necessary as the only accurate measuring system
been produced outside the tolerance to wait until the concrete segments have that can meet such high demands at a rate
requirements. DAUB (2013), however, gained full maturity (i.e. at 28 days) to carry required for a high speed quality control/
permits these segments to be used in out the dimensional control measurements. quality assurance program. As shown
areas with low projected tunnelling forces, However, it is highly recommended to in Figure 57, 3D measurements using a
providing gasket performance is not conduct these measurements at one laser interferometer system facilitates the
compromised and calculations demonstrate specific phase of the production cycle, accurate digitization of surfaces by direct
adequate capacity of reinforcement in for instance after segment stripping polar coordinate measurement. A skilled
the joints in the event of exceeding joint (demolding). operator can comprehensively measure
distortion and segment width tolerances. the full profile of over 20 segments per shift
The dimensional control program should (Clarke-Hackston et al. 2006). In addition
12.2 Measurement and Dimensional be implemented on individual segments to measurements taken from a single
Control as well as segmental systems in the standpoint, a laser tracker system can be
form of test rings. The most traditional employed where a second instrument is
A proper quality control system is measurement methods are manual utilized (see Figure 57b) in order to perform
essential to ensure the production of measurement by means of steel templates, a complete measurement of form. Figure
high-precision segments with tolerances micron rods, caliper squares, precision 58 indicates typical output from the spatial
which fall within the limits imposed by measuring tapes and measurement arms analyzer 3D graphical software platform
contract documentation. Segments, like (ÖVBB, 2011). However, conventional with tolerances, best-fit of external surfaces
any other concrete element, are subject instruments like micron rods or templates and deviations from referenced dimensions.
a b
c d
Figure 57 : 3D measurement: (a) laser Interferometer schematic, (b) laser tracker system, (c) instrument location for mold, (d) stacked segments for rapid measurement
a b c
Figure 58 : Spatial analyzer 3D graphical software platform output: (a) typical segment tolerances, (b) best-fit of external surfaces and typical deviations from referenced dimensions,
(c) tolerances on virtual (test) ring build.
12.3 TEST RING AND DIMENSIONAL place, in order to avoid recovery forces. As
CONTROL FREQUENCY indicated in Table 14, reference dimensions b) Dimensional Control Frequency
to be verified on the assembled test ring Two major factors are considered to ensure
Dimensional control is carried out on include outer and inner diameters (on at dimensional control during the production
formwork or segments to measure tolerances least two axes), outer circumference (to phase, i.e. the testing objects and testing
on reference dimensions, and on the test be measured in three planes) and joint frequency. Forms, individual segments and
rings for controlling system tolerances. assembly misalignment. (ÖVBB, 2011), test rings are the three main testing objects
in addition, recommends joint opening that can be measured. Standard practice
a) Test Ring Assembly and Tolerances and joint misalignment as other system is to measure the tolerances of every
tolerances to be verified on the test ring. segment for the first 10 castings, and then
One of the original goals of early test ring measure every 50th segment after that.
assembly was to check for alignment of A 10-measurement system as illustrated Some projects, however, call for a more
bolting systems and general ring assembly in Figure 59 has recently been presented frequent controlling program. A dimensional
including segment to segment and ring to as an alternative formwork tolerance controlling program based on monthly
ring connections. However, with current 3-D measurement method as opposed to the measurements is not recommended.
computational modelling, the risk of bolting individual reference dimensions given in Some tender design documents require
or assembly location errors within the forms Figure 56 and Table 14. Both systems are assembling an extra ring on the test ring
is minimal. acceptable as they both assure segment after a specific number of castings. This is
dimension accuracies compatible with the not recommended considering the degree
In a test ring system, the recommendation reference dimensions given in Table 14. of difficulty, the high cost and little value of
is to build a single ring or a double ring With this alternative method, if there are any such a practice. For segment dimension
where the bottom test ring is measured issues with the 10 measurements, then the control, testing should be resumed at the
completely. Generally, the test rings are segment direct measurements are to be initial frequency soon after detection of any
assembled without the elastomer gasket in followed as per Table 14. inadmissible deviations.
Figure 59 : Alternative tolerance measurement method for formwork dimension control Figure 60 : Typical construction tolerances in contract documents
AUTHORED DOCUMENTS
Alder, A.; Dhanda, D.; Hillyar, W.; and Runacres, A., 2010, “Extending London’s Docklands Light Railway to Woolwich”, Proceedings of the
ICE-Civil Engineering, V. 163, No. 2, pp. 81-90.
Bakhshi, M.; and Nasri, V., 2014a, “Review of International Practice on Critical Aspects of Segmental Tunnel Lining Design”, Proceedings of
the 2014 North American Tunneling (NAT) Conference, Los Angeles, CA, pp. 274-282.
Bakhshi, M.; Nasri, V., 2014b, “Guidelines and Methods on Segmental Tunnel Lining Analysis and Design – Review and Best Practice
Recommendation”, Proceedings of the World Tunnel Congress 2014. Iguassu Falls, Brazil.
Bakhshi, M.; and Nasri, V., 2014c, “Developments in Design for Fiber Reinforced Concrete Segmental Tunnel Lining”, 2nd FRC International
Workshop (1st ACI–fib Joint Workshop) on Fibre Reinforced Concrete, Montreal, Canada, pp. 441-452.
Bakhshi, M.; and Nasri, V., 2014d, “Design Considerations for Precast Tunnel Segments According to International Recommendations,
Guidelines and Standards”, TAC 2014, Vancouver, Canada.
Bakhshi, M.; Nasri, V., 2013a., “Latest Developments in Design of Segmental Tunnel Linings”, Proceedings of the Canadian Society for Civil
Engineering General Conference (CSCE 2013), Montréal, Québec, Canada.
Bakhshi, M.; and Nasri, V., 2013b, “Practical Aspects of Segmental Tunnel Lining Design”, Underground—The way to the future:
Proceedings of the World Tunnel Congress 2013., G. Anagnostou, and H. Ehrbar, eds., Geneva, Switzerland.
Bakhshi, M.; and Nasri, V., 2013c, “Structural Design of Segmental Tunnel Linings”, Proceedings of 3rd International Conference on
Computational Methods in Tunnelling and Subsurface Engineering: EURO: TUN 2013. Ruhr University, Bochum, Germany, pp. 131-138.
Bakhshi, M.; Barsby, C.; and Mobasher, B., 2014, “Comparative Evaluation of Early Age Toughness Parameters in Fiber Reinforced
Concrete,” Materials and Structures, V. 47, No. 5, pp. 853-872.
Barros, J. A. O.; Cunha, V. M. C. F.; Ribeiro, A. F.; and Antunes, J. A. B., 2005, “Postcracking Behaviour of Steel Fibre Reinforced
Concrete,” Materials and Structures, V. 38, No. 1, pp. 47-56.
Blom, C.B.M., 2002, “Design philosophy of concrete linings for tunnels in soft soil,” PhD dissertation. Delft University Press, December
2002.
Caan, C.P.; Jansen, J. A. G.; Heijmans, R. W. M. G.; and van der Put, J. L., 1998, “High Speed Line – South: The Groene Hart Tunnel-
Lining Design”, Reference Design. Report No. 9G4 0001 981028, 52p.
Caratelli, A.; Meda, A.; Rinaldi, Z., 2012, “Design According to MC2010 of a Fibre-Reinforced Concrete Tunnel in Monte Lirio, Panama”,
Structural Concrete, V. 13, No. 3, pp. 166-173.
Cavalaro, S. H. P., Blom, C. B. M., Aguado, A., & Walraven, J. C., 2011, “New design method for the production tolerances of concrete
tunnel segments”. Journal of Performance of Constructed Facilities, 26(6), 824-834.
Çimentepe, A.G. 2010, “Evaluation of Structural Analysis methods used for the Design of TBM Segmental Linings”, MSc Thesis, Middle
East Technical University, Ankara, Turkey.
Clarke-Hackston, N., Messing, M., Loh, D., & Lott, R., 2006, “Modern high precision high speed measurement of segments and moulds”,
Tunnelling and Underground Space Technology, 21(3), 258-258.
Colombo, M.; Martinelli, P.; di Prisco, M., 2015, “A design approach for tunnels exposed to blast and fire”, Structural Concrete, V. 16, No.
2, pp. 262-272.
Curtis, D.J.; Hay, M.; and Croydon, A., 1976, “Discussion on ‘The Circular Tunnel in Elastic Ground’” Géotechnique, V. 26, No. 1, pp. 231-
237.
Deere, D. U.; Peck, R. B.; Monsees, J. E.; and Schmidt, B., 1969, “Design of Tunnel Liners and Support Systems”, Highway Research
Record 889, Final Report to Urban Mass Transit Administration, U.S. Department of Transportation, Washington, D.C.
Duddeck, H.; and Erdmann, J., 1982, “Structural Design Models for Tunnels”, Tunneling ’82-Proceedings of the Third International
Symposium. Inst. of Mining and Metallurgy S, London, UK, pp. 83-91.
Francis, O.; and Mangione, M., 2012, “Developments in Joint Design for Steel Fibre Reinforced Concrete Segmental Tunnel Linings”,
Proceedings of the World Tunnelling Congress 2012. Bangkok, Thailand.
Fukui, K.; and Okubo, S., 2003, “TBM Cutting Forces with Particular Reference to Cutter and Tunnel Diameters”, ISRM 2003–Technology
Roadmap for Rock Mechanics, South African Institute of Mining and Metallurgy.
Groeneweg, T., 2007, “Shield Driven Tunnels in Ultra High Strength Concrete: Reduction of the Tunnel Lining Thickness”, MSc Thesis, Delft
University of Technology, The Netherlands.
Guglielmetti, V.; Grasso, P.; Mahtab, A.; and Xu, S., 2007, Mechanized Tunnelling in Urban Areas: Design Methodology and Construction
Control, CRC Press, 528p.
Handke, D., 2012, “High-Precision Segments: Prerequisite for a high-quality monocoque tunnel”, Tunnel, 8: 42.
Iyengar, K. T., 1962, “Two-Dimensional Theories of Anchorage Zone Stresses in Post-Tensioned Beams”, Journal of the American Concrete
Institute (ACI), V. 59, No. 10, pp. 1443–1466.
Kirschke, D.; Schälicke, H.; and Fraas, D., 2013, “Finne Tunnel: Innovative targeted resealing of segment bore joints – Part 2”, Tunnel,
4/2013. pp. 30-40.
Mashimo, H.; Isago, N.; Yoshinaga, S.; Shiroma, H.; and Baba, K., 2002, “Experimental Investigation on Load-Carrying Capacity of
Concrete Tunnel Lining”, Proceedings of AITES-ITA 2002 - 28th General Assembly & World Tunnelling Congress 2002, Sydney, Australia,
pp. 1-10.
Mendez Lorenzo, M.G., 1998, “Reliability analysis of a steel fibre reinforced concrete tunnel lining”, Master Thesis. Delft University of
Technology, Delft, the Netherlands.
Mobasher, B.; Bakhshi, M.; and Barsby, C., 2014, “Backcalculation of Residual Tensile Strength of Regular and High Performance Fiber
Reinforced,” Construction & Building Materials, V. 70, pp. 243-253.
Moccichino, M.; Romualdi, P.; Perruzza, P.; Meda, A.; and Rinaldi, Z., 2010, “Experimental Tests on Tunnel Precast Segmental Lining with
Fiber Reinforced Concrete”, ITA 2010 World Tunnel Congress, Vancouver, Canada.
Morgan, H.D., 1961, “A Contribution to the Analysis of Stress in A Circular Tunnel”, Geotechnique, V. 11, pp. 37-46.
Muir Wood, A. M., 1975, “The Circular Tunnel in Elastic Ground”, Géotechnique, V. 25, No. 1, pp. 115–127.
Neun, E., 2012, “Structural Fire Safety Engineering. Eurocode Approach and Practical Application”, Proceedings of the World Tunnelling
Congress 2012. Bangkok, Thailand.
Peck, R. B., 1969, “Deep Excavation and Tunnelling in Soft Ground, State of the Art Report”, 7th International Conference on Soil
Mechanics and Foundation Engineering, Mexico City, Mexico, pp. 225-290.
Plizzari, G.; Tiberti, G., 2009, “Tunnel Linings Made by Precast Concrete Segments”, Construction Methodologies and Structural
Performance of Tunnel Linings, G.A. Plizzari, ed., Brescia, Italy, pp. 136-131.
Poh, J; Tan, K. H.; Peterson, G. L.; and Wen, D., 2009, Structural Testing of Steel Fibre Reinforced Concrete (SFRC) Tunnel Lining
Segments in Singapore”, Proceedings of the World Tunnelling Congress (WTC) 2009, Budapest, Hungary.
Rostami, J., 2008, “Hard Rock TBM Cutterhead Modeling for Design and Performance Prediction”, Geomechanics and Tunneling, V. 1, No.
1, pp. 18-28.
Sinha, R. S., 1989, Underground Structures: Design and Instrumentation, Elsevier Science, New York, NY, 500p.
Soranakom, C., and Mobasher, B., 2007, “Closed Form Solutions for Flexural Response of Fiber Reinforced Concrete Beams,” Journal of
Engineering Mechanics, V. 133, No. 8, pp. 933-941.
Vandewalle, L., 2000, “Cracking Behaviour of Concrete Beams Reinforced with a Combination of Ordinary Reinforcement and Steel Fibers,”
Materials and Structures, V. 33, No. 3, pp. 164-170.
Zhong, X.-C.; Liu, Q.-W.; and Zhao, H., 2011, “Study on the Grouting Pressure of Shield Tunnel”, ASCE Geotechnical Special Publication,
V. 215, pp. 183-190.
Annular gap – space between the surrounding ground and the outer surface of the segments.
CIP – Cast-in-Place.
Circumferential joint – joint between two adjacent segmental rings.
Connections – devices for temporary or permanent connection of two segments or segment rings in the longitudinal and circumferential
joints.
Cover – vertical distance to nearest ground surface from the tunnel crown.
Crosscut – connecting structure between two tunnel tubes or between a tunnel tube and the ground surface or a shaft
Crown – the highest part of a tunnel in cross section.
CSO – Combined Sewer Overflow.
DEM – Distinct Element Method.
Demolding – process of removing a precast concrete segment from the form in which it was cast.
Extrados – the exterior curve of an arch, i.e. the outer surface of the segment or the segment ring
FEM – Finite Element Method, the model representation of a structure as a combination of a finite number of two-dimensional or three-
dimensional components.
Gasket – sealing strips placed in one or more layers around individual segments to ensure permanent sealing of the tunnel tube against the
ingress of water from the surrounding rock mass.
Grout – cement mortar which is injected into the annular gap behind segments to fill the void between segments and excavated ground
profile.
Intrados – the inner curve of an arch, i.e. the inner surface of the segment or segment ring
Invert – the lowest part of a tunnel in cross section.
Longitudinal joint – joint between adjacent segments in a ring.
One-pass lining – segmental lining is considered a one-pass lining when all static and structural requirements of the tunnel lining are
catered for and no further internal lining is required.
Ovalization – deformation of an initially circular segmental ring to an oval shape due to earth pressure, grout pressure, segment self-weight
or uplift.
Ring width – dimension of the segment ring in the longitudinal direction of the tunnel.
Segment – curved prefabricated element that forms part of a ring of support or tunnel lining; commonly precast concrete.
Segment thickness – distance between the inner and outer surfaces of the lining segment, i.e. the distance between the lining intrados
and extrados.
Segmental lining – a set of precast segments assembled in a ring to form the tunnel lining
Springline – opposite sides of the tunnel at center line level.
Stripping – see demolding.
Tail void – annular space between the outside diameter (extrados) of the shield and the extrados of the segmental lining.
TBM – Tunnel boring machine which consists of a cutterhead in front of a shield which excavates tunnels with a circular cross section
through different rock and soil strata.
Test ring – complete segment ring, usually assembled in horizontal position at segment precast plant, for test purposes.
Thrust jacks – hydraulic jacks serving to transmit the thrust forces of the tunnel boring machine to the segment ring, facilitating installation,
or both.
Two-pass lining – tunnel lining consisting of two shells with different structural and constructional requirements which are produced in
independent operations and with different construction methods (e.g. outer shell as a segmental lining, inner shell as a CIP).
ITA Secretariat - c/o MIE2 – Chemin de Balexert 9 - CH-1219 Châtelaine (GE) - Switzerland
Tel: + 41 22 547 74 41 - Email : [email protected] - Web : www.ita-aites.org