You are on page 1of 27

Design Of Tunnel Support System In

Soft Geological Formation In NATM

Method Of Excavation

Department of Mining Engineering,

V.N.I.T. Nagpur

Guided By Submitted by

 To develop a complete support solution for a tunnel

excavated by NATM principle in soft formation
 That’s include choice & design of
 Pre-excavation support system/ fac e stabilization
 Primary lining &/or temporary lining
 Secondary lining
 Soil stabilization method ie. Grouting work design
 Other support system design like soil nailing, bolting etc.
 Face stabilization
 Most of the tunnel constructed in cities are at low
depth and so excavated earth and nearby formation are
mostly soft material.
 Limitation of TBM
 Limitation of present available guidelines viz IS:4880,
IS:5878, IS:4756, IS: 15026 - conservative and
uneconomical design and sometime inadequate
 Conflicting idea between NATM design philosophy
and minimization of surface settlement.
 Gap in translation of theoretical idea/ technique
developed in field of numerical and analytical
modeling of geotechnical system and its application in
Support System Design design

Economy Use Safety Maintainability

Short Term
Long Term Resource Ground
Availability Composition

Time Human Depth Of

Water Table
Equipment Friction angle Compressibility
Void Ratio
Technology Modulus
Water Content Directionality

NATM is based on Seven principles.

I. Mobilization of the strength of rock mass
II. Shot-crete protection
III. Measurements
IV. Primary Lining
V. Closing of invert
VI. Rock mass classification
VII. Dynamic Design
Tunnelman's Ground Classification for Soils by Terzaghi (1950) and
Modified by Heuer (1974)
Classification Behavior Typical Soil Types
Heading can be advanced without initial Loess above water table; hard clay,
Firm support, and final lining can be constructed marl, cemented sand and gravel
before ground starts to move. when not highly overstressed.
Chunks or flakes of material begin to drop out
Residual soils or sand with small
Slow of the arch or walls sometime after the ground
amounts of binder may be fast
raveling has been exposed, due to loosening or to over-
raveling below the water tale, slow
----- stress and "brittle" fracture (ground separates
Raveling raveling above. Stiff fissured clays
Fast or breaks along distinct surfaces, opposed to
may be slow or fast raveling
raveling squeezing ground). In fast raveling ground, the
depending upon degree of
process starts within a few minutes, otherwise
the ground is slow raveling.
Ground with low frictional
strength. Rate of squeeze depends
Ground squeezes or extrudes plastically into on degree of overstress. Occurs at
tunnel, without visible fracturing or loss of shallow to medium depth in clay of
Squeezing continuity, and without perceptible increase in very soft to medium consistency.
water content. Ductile, plastic yield and flow Stiff to hard clay under high cover
due to overstress. may move in combination of
raveling at excavation surface and
squeezing at depth behind surface.
Granular materials without Clean, dry granular materials.
cohesion are unstable at a slope Apparent cohesion in moist sand,
Cohesive -
greater than their angle of repose or weak cementation in any
(approx 30o -35o ). When exposed granular soil, may allow the
Running -----
at steeper slopes they run like material to stand for a brief period
granulated sugar or dune sand of raveling before it breaks down
until the slope flattens to the angle and runs. Such behavior is
of repose. cohesive-running.

A mixture of soil and water flows

into the tunnel like a viscous fluid. Below the water table in silt, sand,
The material can enter the tunnel or gravel without enough clay
from the invert as well as from the content to give significant
face, crown, and walls, and can cohesion and plasticity. May also
flow for great distances, occur in highly sensitive clay
completely filling the tunnel in when such material is disturbed.
some cases.

Highly preconsolidated clay with

Ground absorbs water, increases
plasticity index in excess of about
in volume, and expands slowly
Swelling 30, generally containing
into the tunnel.
significant percentages of
Tunnel Behavior: Sands and Gravels
Tunnel Behavior
Designation Degree of
Above Water Below Water
Table Table
Loose, N<10 Cohesive Running Flowing
Very Fine Clean Sand
Dense, N>30 Fast Raveling Flowing
Loose, N <10 Rapid Raveling Flowing
Fine Sand With Clay
Firm or Slowly
Binder Dense, N>30 Slowly Raveling
Rapidly Raveling
Loose, N <10 Rapid Raveling or
Sand or Sandy Gravel
with Clay Binder
Firm or Slow
Dense, N>30 Firm
Running ground. Uniform
(Cu <3) and loose (N<10)
materials with round grains
Sandy Gravel and combined with
run much more freely than
Medium to Coarse Sand extremely heavy
well graded (Cu >6) and
discharge of
dense (N>30) ones with
angular grains.
Excavation And Support For Different Soil Types
 Excavation Sequence - Top heading, bench &
invert; drifts
 Initial Shotcrete Lining- 50- 450 mm
 Installation Location- Installation of shotcrete
support immediately
 Pre support measures & ground improvement-
grouting, soil mixing and ground freezing, spiling
methods (pipe arch canopies, grouted solid bars,
perforated steel pipes), face doweling, umbrella
 Support Installation- Support installation dictates
Control of surface settlements with umbrella arch method in second
stage excavations of Istanbul Metro by Ibrahim Ocak, 2007
• Author use empirical model to prove that use of umbrella arch reduce max.
settlement by 66%.
• Other advantages as marked by author are elimination of application of soil
nails to tunnel face, obligation of leaving unexcavated middle section at
tunnel face. Also faster excavation is possible.
• Cost analysis must be done to measure advantage of this additional cost
though author resonate that long term advantage would be more
Design of sequential excavation method for large span urban
tunnels in soft ground – Niayesh tunnel By Mostafa Sharifzadeh,
Farshad Kolivand, Masoud Ghorbani, Shahaboddin Yasrobi , 2013

• FEM model (Plaxis 3D Tunnel) was used

evaluate different excavation sequence to
minimize settlement.Consequently SD method
was selected
• Factors affecting the tunnelling performance
were identified for different sequential schemes,
including ring closure time, number of
excavation stages, subdividing area, and
excavation step of the central gallery.
• Among these factors, the rapid closure of the supporting ring and excavation step
of the central gallery (drift) are the most important factors in controlling tunnel
deformations and surface settlements in soft ground tunnelling.
• The central gallery (middle drift) has an important role in maintaining the tunnel
crown and reducing the displacements and surface settlement, and it must be
excavated at the final stages.
Physical modeling of tunnels in soft ground: A review by
M.A. Meguid, O. Saada, M.A. Nunes , J.Mattar, 2007
Stability factor of Cohesive Soils and Silty Sand Above Water Table by Peck’s
Cohesive (clayey) soils behave as a ductile
Ncrit is the stability factor,
plastic material that moves into the tunnel in a
Pz is the overburden pressure to the tunnel
theoretically uniform manner. Cohesive (clay)
materials or materials with sufficient cohesion
Pa is the equivalent uniform interior
or cementation to sample and test for
pressure applied to the face (as by
unconfined compression strength, an estimate
breasting or compressed air), and
of ground behavior in tunneling can be
Su is the undrained shear strength (defined
obtained from the equation:
for this purpose as one-half of the
Pz − P a
Ncrit = unconfined compressive strength).

Critical support pressure required to prevent failure of rock mass surrounding

the tunnel by Hoek (1999)
2𝑃𝑜 − 𝜎𝑐𝑚 Pcr = Critical support pressure
𝑃𝑐𝑟 =
1+𝑘 Po = Hydrostatic stresses
σcm = Uniaxial compressive strength of rock
1 + sin ɸ mass
1 − sin ɸ φ = Aangle of friction of the rock mass
The increase in unit thrustΔTA is increase in unit thrust capacity of the rock arch,
φ is effective friction angle of the rock mass,
capacity of the reinforced
Tb is stress at yield of the reinforcement steel (fully
zone of tunnel arch bygrouted bolts),
Bischoff and Smart (1977) A is cross-sectional area of the reinforcement steel,
S is spacing of the reinforcement steel, in both directions,
𝑛 t is effective thickness of the rock arch (= L – S)
𝑇𝑎𝑛2 45 + 𝑇𝐴 𝑡
Δ𝑇𝐴 = 2 𝑏 𝑏 L is length of the reinforcement steel

Tunnel Behavior for Clayey Soils and Silty Sand (after Bickel et al., 1996)
(Modified from Peck 1969 and Phienwaja 1987.)
Stability Factor, Ncrit Soft Ground Tunnel Behavior
Cohesive Soils
1 Stable
2-3 Small creep
4-5 Creeping, usually slow enough to permit tunneling
May produce general shear failure. Clay likely to invade tail space
too quickly to handle
Silty Sands Above Water Table (with some apparent cohesion)
1/4 - 1/3 Firm
1/3 - 1/2 Slow Raveling
Elastic closed form models for ground-lining interaction by Ranken,
Ghabo11ssiand Hendron, 1978
Flexibility ratio is defined as
F =(Em/EJ )(R/t)3[2(1-vl2)/(1+vm)]
Em is the modulus of the surrounding medium, or ground,
Ej is the modulus of the lining,
t is the lining thickness, and
vl and Vm are the Poisson ratios of the lining and ground,

Compressibility ratio
C = (Em/EJ)(R/t)[(l - vl2)/((l+ vm) (1- 2 vm))]
K = support stiffness
Pmax = maximum support pressure;
Analytical Solutions for Support Stiffness and Ec = Young’s modulus of concrete;
Maximum Support Pressure for Various Support tc =lining thickness;
ri = internal tunnel radius;
Systems by (Brady & Brown, 1985) σcc = uniaxial compressive strength of
concrete or shotcrete;
W = flange width of steel set and side
Support System Support stiffness (K) and maximum support pressure (Pmax) length of square block;
X = depth of section of steel set;
As = cross section area of steel set;
Ec |ri2 − ri − t c 2 | Is = second moment of area of steel
Concrete K=
1 + vc 1 − 2vc ri2 + ri − t c 2 set;
/Shotcrete Es = Young’s modulus of steel;
lining σys = yield strength of steel;
σcc ri − t c
Pmax = 1− S = steel set spacing along the tunnel
2 ri2 axis;
θ= half angle between blocking points
in radians;
1 𝑆𝑟𝑖 𝑆𝑟𝑖3 𝜃 𝜃 + 𝑠𝑖𝑛 𝜃 𝑐𝑜𝑠 𝜃 2𝑆𝜃𝑡𝐵 tB = thickness of block;
Blocked steel = + +
𝐾 𝐸𝑠 𝐴𝑠 𝐸𝑠 𝐴𝑠 2𝑠𝑖𝑛2 𝜃 𝐸𝐵 𝑊 2 EB =Young’s modulus of block
sets material;
3𝐴𝑠 𝐼𝑠 𝜎𝑦𝑠 l = free bolt or cable length;
𝑃𝑚𝑎𝑥 = db = bolt diameter or equivalent cable
2𝑆𝑟𝑖 𝜃 3𝐼𝑠 + 𝑋𝐴𝑠 𝑟𝑖 − 𝑡𝐵 + 0.5𝑋 1 − 𝑐𝑜𝑠 𝜃
Eb= Young’s modulus of bolt or cable;
1 𝑠𝑖 𝑠𝑙 4𝑙 Tbf = ultimate failure load in pull-out
mechanically or = +𝑄 test;
𝐾 𝑟𝑖 π𝑑𝑏2 𝐸𝑏
chemically sc = circumferential bolt spacing;
anchored rock sl =longitudinal bolt spacing;
bolts or cables Pmax = Q = load-deformation constant for
𝑠𝑖 𝑠𝑙 anchor and head.
Analytical Solutions for Soil – Liner Interaction
Wu & Penzien (1997)
Analytical Solutions Thrust Moment

crown = Pd+Ps+Pw

Springline = Pd+Pw-Ps

Relaxation Pd= -0.5(1+k0)(1/(1+C))(hm-hww)(d/2) = -1/4(1-k0)(1/(1+F))(h m-hw w)(d/2)

Ps= 0.5(1-k0)(1/(1+F)(hm-hww)(d/2)

Pw= -(1/(1+C))hww(d/2)

crown = Pd+Ps+Pw

Springline = Pd+Pw-Ps

Pd= -(1+k0)(1-vm)(1/(1+C))(hm-hw w)(d/2) = -(1-k0)(1-vm)/((3-4vm)/(1+F))*(hm-

Ps= 2(1-k0)(1-vm)/((3-4vm)(1+F))*(hm-

Einstein & Schwartz (1979)

Analytical Solutions Thrust Moment


Excavation =(0.5(1+k0)(1-a0)-0.5(1+k0)(1+2a2))m hd/2

= -(0.5(1+k0) (1+2a2) h(d/2)
full slip

=(0.5(1+k0)(1-a0)+0.5(1+k0)(1+2a2))m hd/2

Thrust at Crown

Excavation =(0.5(1+k0)(1-a0)+0.5(1-k0)(1+2a4))m h d/2

= -(0.25(1+k0) (1-2h+2b2) h(D/2)
no slip
Thrust at Springline

=(0.5(1+k0)(1-a0)+0.5(1-k0)(1+2a4)) mhd/2
a0=CF(1-vm)/(F+C+(CF(1-vm))) b2=[C-(1-vm)]/[2C(1-vm)+4vm-6-3C(1-vm)]
a2=(F+6)(1-vm)/2F(1-vm+(6(5-vm))) C=[(d//2)Em(1-vL2)]/[EL(AL/wL)(1-vm2)]
a4=b2 F=(d/2)3Em(1-vL2)/ [EL(ALI)(1-vm2)]
Peck, Hendron & Moharaz (1972)

Analytical Solutions Thrust Moment

Thrust at Crown

=0.5[(1+k0)b1-(1/3)(1+k0)b2] s h d/2
Overburden = (1+k0)b2 s h (d/2)2
Thrust at Springline

=0.5[(1+k0)b1-(1/3)(1+k0)b2] s h d/2

b1=1-a1 b2=(1+3a2-4a3)
a1=(1-2vm)(C-1)/((1-2vm)c+1) a2=(2F+1-2vm)/(2F+5-6vm) a3=2F/(2F+5-6vm)
Ranken, Ghaboussi and Hendron (1978)
Solutions Thrust Moment

Thrust at Crown Moment at Crown

Overpre = h(d/2)(((1+k0)(1-Ln))- ((1-k0)(1+Ln))) =( h(d/2)2/2)[(1+k0)(1-vm)(C/6F)(1-Ln) -((1-k)/2)(1-Jn-2N)]
(no slip) Thrust at Springline Moment at Springline

= h(d/2)(((1+k0)(1-Ln))+((1-k0)(1+Ln))) =( h(d/2)2/2)[(1+k0)(1-vm)(C/6F)(1-Ln) +((1-k)/2)(1-Jn-2N)]

Thrust at Crown Moment at Crown

ssure = ((1+k0)(1-Lf)-(1-k0)(1-Jf))  h(d/4) =( h(d/2)2/2)[(1+k0)(1-vm)(C/6F)(1-Lf) -(1-k) (1-J)]
Thrust at Springline Moment at Springline
= ((1+k0)(1-Lf)+(1-k0)(1-Jf))  h(d/4) =( h(d/2)2/2)[(1+k0)(1-vm)(C/6F)(1-Lf) +(1-k) (1-J)]

Thrust at Crown Moment at Crown

on = ((1+k0)(1-Lnɸ)-(1-k0)(1-Jnɸ))  h(d/4) =( h(d/2)2/2)[(1+k0)(Lnɸ/6F)-(0.5(1-k0)(1+J-N))]
Thrust at Springline Moment at Springline
= ((1+k0)(1-Lnɸ)+(1-k0)(1-Jnɸ))  h(d/4) =( h(d/2)2/2)[(1+k0)(Lnɸ/6F)+(0.5(1-k0)(1+J-N))]
Thrust at Crown Moment at Crown

= ((1+k0)(1-Lfɸ)-(1-k0)(1-2Jfɸ))  h(d/4) =(h(d/2)2/2)[(1+k0)(Lfɸ/6F)-((1-k0)(1-J))]

(Full Slip) Thrust at Springline Moment at Springline

= ((1+k0)(1-Lfɸ)+(1-k0)(1-2Jfɸ))  h(d/4) =(h(d/2)2/2)[(1+k0)(Lfɸ/6F)+((1-k0)(1-J))]

Ln = (1-2vm)(C-1)/(1+(1-2vm)C) Lnɸ = (1-2vm)C/(1+(1-2vm)C)

Jn = (1-2vm)(1-C)F-(0.5(1-2vm)C+2)/[{(3-2vm)+(1-2vm)C}F+{0.5(5-6vm)}(1-2vm)C+(6-8vm)]
Jnɸ = {(1-2vm)C+2vm}F-(1-vm)(1-2vm)C/[{(3-2vm)+(1-2vm)C}F+{0.5(5-6vm)}(1-2vm)C+(6-8vm)]
Nn = {(1-2vm)C+1}F-(0.5(1-2vm)C-2)/[{(3-2vm)+(1-2vm)C}F+{0.5(5-6vm)}(1-2vm)C+(6-8vm)]
Nnɸ ={2(1-2vm)C+3}F-(0.5(1-2vm)C)/[{(3-2vm)+(1-2vm)C}F+{0.5(5-6vm)}(1-2vm)C+(6-8vm)]
Lf = (1-2vm)(C-1)/(1+(1-2vm)C) Lfɸ = (1-2vm)C/(1+(1-2vm)C)
Jf = {2F+(1-2vm)}/{2F+(5-6vm)} Jfɸ = {F+(1-vm)}/{2F+(5-6vm)}
Nf = {2F-1}/{2F+(5-6vm)} Jfɸ = {4F+1}/{2F+(5-6vm)}
Analytical Solutions Thrust Moment
Curtis , 1976
Thrust at Crown = Nconstant - Nmax

Excavation Thrust at Springline = Nconstant + Nmax

Full Slip
Nconstant = [(σv+σH)d/2]/[2+(1-k0)2k]

Nmax = ((σv-σH)d/4)(3-4vm)/ (5-6vm+4Q1)

Thrust at Crown = Nconstant - Nmax

Excavation Thrust at Springline = Nconstant + Nmax

=[1/4(σv-σH)2(d/2)2] /(1+(3-2vm)Q2)
No Slip
Nconstant = [(σv+σH)d/2]/[2+(1-k0)2k]

Nmax = ((σv-σH)d/2)[(1+2.vmQ1)(3-4vm)(1+Q1)
Muir & Wood, (1975)

Thrust at Crown

Excavation = 4/3(σv+σH)d/2+4/3..deflection(d/2)
Full Slip
Thrust at Springline

= 2/3k0(σv+σH)d/2+2/3..deflection(d/2)+σHd/2
k = (1-vm)/((1-2vm)(1+vm))  = 3Em/((1+vm)(5-6vm)d/2 Rs = 9EI/(3(d/2)4)
 = (Em/EL)(d/2)/(AL/wL) Q1 = (Em/EL)(1/(1+vm))(d/2)3/(12I) σH = σvk0+Hww(1-
 = (ɸID+d)/(d/2) Q2 = (Em/EL)(1/(1+vm))(d/2)3/(12I) σv = mh+S
ɸID = 2(d/2-tL) I = (wLtL3)/12

v : Poisson's ration for ground

m AL : Cross-Sectional Area of Liner

F : Flexibility v : Poisson's ration for Liner


 : Ground Unit Weight

m S : Surcharge
Em : Young's Modulus for ground  : Water Unit Weight

h : Depth to Springline EL : Young's Modulus for Liner

d : Diameter of Tunnel hw : Depth from Water Table
tL : Thickness of Liner I : Moment Inertia per Unit Length
Rs : Stiffness Factor wL : Width of Liner
C : Compressibility k0 : Coefficient of Lat. Earth Pressure
Time dependent deformations during tunnelling and stability of tunnel
faces in fine-grained soils under groundwater by Roger Ho¨fle Æ Jochen
Fillibeck Æ Norbert Vogt, 2008
• FEM modeling software Plaxis 3D Tunnel, was applied to check influence of
the permeability on tunnel face stability in hypothetical situation
• only qualitative but no quantitative statements regarding the tunnel face
stability can be made
• It was not possible to determine a realistic stability ratio for the tunnel face for
the hypothetical problem with the typically applied phi-c-reduction.
• It shows limitation of Plaxis 3D Tunnel software

Analytically-based simplified formulas for circular tunnels with two

liners in viscoelastic rock under anisotropic initial stresses by Fei
Song, Huaning Wang, Mingjing Jiang, 2018
• A general analytical solution is proposed for circular tunnels with two liners
in viscoelastic rock.
• The anisotropic initial stresses, any viscoelastic models and installation times
of two liners can be considered.
• Analytically-based simplified formulas for final liner pressures and tunnel
convergence are provided.
• A tunneling design method for the installation times of second lines is
Conclusion And Future Project Orientation

 Empirical formulation are outdated

 Analytical solution are also not considered good enough
 Numerical solution of geotechnical problems gives solution which
matches the observed behavior
 But numerical method developed till now are based on certain
 Among these are
• Geological formation properties doesn't change with time
• Consolidation/ self compaction doesn't occur
• Static soil properties are considered, whereas soil water content
have huge impact on its properties.
• Sesmic consideration like ground liquifaction are not considered