U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Bridges & Structures

<< Previous Contents Next >>

Technical Manual for Design and Construction of Road Tunnels - Civil Elements

Chapter 6 - Rock Tunneling

6.1 Introduction

Chapters 6 through 10 present design recommendations and requirements for mined and bored road tunnels in all types of grounds. Chapter 6 addresses analysis, design and construction issues for rock tunneling including rock failure mechanism, rock mass classification, excavation methods, excavation supports, and design considerations for permanent lining, groundwater control, and other ground control measures. Chapter 10 addresses the design of various types of permanent lining applicable for rock tunnels.

Because of the range of behavior of tunnels in rock, i.e., from a coherent continuum to a discontinuum, stabilization measures range from no support to bolts to steel sets to heavily reinforced concrete lining and numerous variations and combinations in between. Certainly these variations are to be expected when going from one tunnel to another but often several are required in a single tunnel because the geology and/or the geometry change. Thus, the engineer must recognize the need for change and prepare the design to allow for adjustments to be made in the field to adjust construction means, methods, and equipment to the challenges presented by the vagaries of nature. This chapter provides the engineer with the basic tools to approach the design, it is not a cookbook that attempts to give instantaneous solutions/designs for the novice designer.

The data needed for analysis and design rock tunnels and the investigative techniques to obtain the data are discussed in Chapter 3. The results of the analysis and design presented hereafter are typically presented in the geotechnical/technical design memorandum (Chapter 4) and form the basis of the Geotechnical Baseline Report (Chapter 4). Readers are referred to Chapter 7 for tunneling issues in soft ground. Problematic ground condition such as running sand and very soft clays are discussed in Chapter 8. Mining sequentially based on the sequential excavation method (SEM) principles is discussed in Chapter 9.

6.2 Rock Failure Mechanism

Only in the last half-century has rock mechanics evolved into a discipline of its own rather than being a sub-set of soil mechanics. At the same time there was a "merging of elastic theory, which dominated the English language literature on the subject, with the discontinuum approach of the Europeans" (Hoek, 2000). These two phenomena have also occurred during a time of ever-increasing demand for economical tunnels. Hence, design and construction of rock tunnels have taken on a new impetus and importance in the overall field of heavy construction as it applies to infrastructure.

Understanding the failure mechanism of a rock mass surrounding an underground opening is essential in the design of support systems for the openings. The failure mechanism depends on the in situ stress level and characteristics of the given rock mass. At shallow depths, where the rock mass is blocky and jointed, the stability problems are generally associated with gravity falls of wedges from the roof and sidewalls since the rock confinement is generally low. As the depth below the ground surface increases, the rock stress increases and may reach a level at which the failure of the rock mass is induced. This rock mass failure can include spalling, slabbing, and major rock burst.

Conversely, excavation of an underground opening in an unweathered massive rock mass may be the most ideal condition. When this condition, paired together with relatively low stresses, exists, the excavation will usually not suffer from serious stability problems, thus support requirement will be minimal.

6.2.1 Wedge Failure

Due to the size of tunnel openings (relative to the rock joint spacing) in most infrastructure applications, the rock around the tunnel tends to act more like a discontinuum. Behavior of a tunnel in a continuous material depends on the intrinsic strength and deformation properties of that material whereas behavior of a tunnel in a discontinuous material depends on the character and spacing of the discontinuities. Design of the former lends itself more naturally to analytical modeling (similar to most tunnels in soil) whereas design of the latter requires consideration of possible block or wedge movement or failure wherein the design approach is to hold the rock mass together. By doing so, the rock is forced to form a "ground arch" around the opening and hence to redistribute the forces such that the ground itself carries most of the "load".

To stabilize blocks or wedges, and hence the opening, the first step is to determine the number, orientations and conditions of the joints. The Q system, described in 6.3.4 gives the basic information required for the joint sets:

  • Number of joints
  • Joint roughness
  • Joint alteration
  • Joint water condition
  • Joint stress condition

With these parameters defined, analyses can be made of the block or wedge stability and of the support required to increase that stability to a satisfactory level. For small tunnels of ordinary geometry the initial analysis (if not the final) can be estimated from a simple free-body approach.

For larger tunnels with complicated geometry and/or a more complicated joint system, it is recommended that a computer program such as Unwedge be used to analyze the opening. Once the basic parameters of the problem are input to the program, a series of runs can be made to evaluate the impact of such variations on the calculated support required for the opening. A design practice using Unwedge will be introduced in Section 6.6.2.

As indicated earlier, except for a small tunnel in very massive rock, the concept of "solid rock" is usually a misconception. As a result, the behavior of the ground around a rock tunnel is usually the combination of that of a blocky medium and a continuum. Hence, the "loads" on the tunnel support system are usually erratic and nonuniform. This is in contrast to soft ground tunnels where the ground may sometimes be approximated by elastic or elastic-plastic assumptions or where the parameters going into numerical modeling are significantly more amenable to rational approximations.

In its simplest terms the challenge to supporting a tunnel in rock is to prevent the natural tendency of the rock to "unravel". Most failures in rock tunnels are initiated by a block (called "keyblocks" by Goodman, 1980) that wants to loosen and come out. When that block succeeds, others tend to loosen and follow. This can continue until the tunnel completely collapses or until the geometry and stress conditions come to equilibrium and the unraveling stops. Contrarily, if that first block can be held in place the stresses rearrange themselves into the ground arch around the tunnel and stability is attained. Figure 6-1 illustrates how detrimental blocky behavior propagates while Figure 6-2 shows how holding the key block in place can stabilize the opening (After Deere 1969).

6.2.2 Stress Induced Failure

As the depth of a tunnel becomes greater or where adjacent underground structures exist and the ground condition becomes less favorable, the stress within the surrounding rock mass increases and failure occurs when the stress exceeds the strength of the rock mass. This failure can range from minor spalling or slabbing in the rock surface to an explosive rockburst where failure of a significant volume of rock mass occurs.

Progressive Failure in Unsupported Blocky Rock

  • Step 1- Block A drops down
  • Step 2- Block B rotates counterclockwise and drops out
  • Step 3- Block C rotates counterclockwise and drops out
  • Step 4- Block D drops out followed by block E
  • Step 5- Block E drops out followed by block F
  • Step 6- Block F rotates clockwise and drops out

Figure 6-1 Progressive Failure in Unsupported Blocky Rock

Prevention of Progressive Failure in Supported Blocky Rock

  • Step 1- Block A and C are held in place by rock bolts and shotcrete
  • Step 2- Block B is held in place by Blocks A and C
  • Step 3- Block D is held in place by Blocks A, B, and C
  • Step 4- Blocks E and F are held in place by Blocks A, B, and D assisted by rock bolts and shotcrete

Figure 6-2 Prevention of Progressive Failure in Supported Blocky Rock

The stress induced failure potential can be investigated using the strength factor (SF) against shear failure defined as (σ1f - σ3)/(σ1 - σ3), where (σ1f - σ3) is the strength of the rock mass and (σ1 - σ3) is the induced stress, σ1 and σ3 are major and minor principal stresses, and σ1f is major principal stress at failure. A SF greater than 1.0 indicates that the rock mass strength is greater than the induced stress, i.e., there is no overstress in the rock mass. When SF is less than 1.0, the induced stresses are greater than the rock mass strength, and the rock mass is overstressed and likely to behave in the plastic range.

6.2.3 Squeezing and Swelling

Squeezing rock is associated with the creation of a plastic region around an opening and severe face instability. From a tunnel design point of view, a rock mass is considered to be weak when its in-situ uniaxial compressive strength is significantly lower than the natural and excavation induced stresses acting upon the rock mass surrounding a tunnel. Hoek et. al. (2000) proposed a chart to predict squeezing problems based on strains with no support system as shown in Figure 6-3. As a very approximate and simple estimation, Figure 6-3 can be directly used to predict squeezing potential by comparing rock mass strength and in-situ stress. If finite element analysis results are available, one can simply predict the squeezing potential based on the calculated strains from the FE analysis. For example, the squeezing problems, if a tunnel is excavated at the proposed depth, are severe when the calculated strains from FE analysis is 2.5% or higher. It should be noted that strains in Figure 6-3 are based on tunnels with no support installed.

A Relationship between Strain and Squeezing Potential of Rock Mass

Figure 6-3 A Relationship between Strain and Squeezing Potential of Rock Mass (Hoek, et. al., 2000)

Swelling rock, in comparison, is associated with an increase in moisture content of the rock. Swelling rock can sometimes be associated with squeezing rock, but may occur without formation of a plastic zone. The swelling is usually associated with clay minerals, indurated to shale or slate or not, imbibing water and expanding. A relatively simple swell test in the laboratory will allow prediction of the swell and will also provide the "swelling pressure", where the swelling pressure is defined as that pressure that must be applied to the rock to arrest the swelling. Obviously, the support system has to resist at least the full swelling pressure to arrest the swelling movement. Montmorillinitic shales, weathered nontronite basalts, and some salts found in evaporate deposits are typical swelling rocks. Chapter 8 provides more detailed discussions about problematic squeezing and swelling ground.

6.3 Rock Mass Classifications

6.3.1 Introduction

Rock mass classification schemes have been developed to assist in (primarily) the collection of rock into common or similar groups. The first truly organized system was proposed by Dr. Karl Terzaghi (1946) and has been followed by a number of schemes proposed by others. Terzaghi's system was mainly qualitative and others are more quantitative in nature. The following subsections explain three systems and show how they can be used to begin to develop and apply numerical ratings to the selection of rock tunnel support and lining. This section discusses various rock mass classification systems mainly used for rock tunnel design and construction projects.

6.3.2 Terzaghi's Classification

Today rock tunnels are usually designed considering the interaction between rock and ground, i.e., the redistribution of stresses into the rock by forming the rock arch. However, the concept of loads still exists and may be applied early in a design to "get a handle" on the support requirement. The concept is to provide support for a height of rock (rock load) that tends to drop out of the roof of the tunnel (Terzaghi, 1946). Terzaghi's qualitative descriptions of rock classes are summarized in Table 6-1.

Table 6-1 Terzaghi's Rock Mass Classification
Rock ConditionDescriptions
Intact rockContains neither joints nor hair cracks. Hence, if it breaks, it breaks across sound rock. On account of the injury to the rock due to blasting, spalls may drop off the roof several hours or days after blasting. This is known as a spalling condition. Hard, intact rock may also be encountered in the popping condition involving the spontaneous and violent detachment of rock slabs from the sides or roof
Stratified rockConsists of individual strata with little or no resistance against separation along the boundaries between the strata. The strata may or may not be weakened by transverse joints. In such rock the spalling condition is quite common
Moderately jointed rockContains joints and hair cracks, but the blocks between joints are locally grown together or so intimately interlocked that vertical walls do not require lateral support. In rocks of this type, both spalling and popping conditions may be encountered
Blocky and seamy rockConsists of chemically intact or almost intact rock fragments which are entirely separated from each other and imperfectly interlocked. In such rock, vertical walls may require lateral support
Crushed but chemically intact rockHas the character of crusher run. If most or all of the fragments are as small as fine sand grains and no recementation has taken place, crushed rock below the water table exhibits the properties of a water-bearing sand
Squeezing rockSlowly advances into the tunnel without perceptible volume increase. A prerequisite for squeeze is a high percentage of microscopic and sub-microscopic particles of micaceous minerals or clay minerals with a low swelling capacity
Swelling rockAdvances into the tunnel chiefly on account of expansion. The capacity to swell seems to be limited to those rocks that contain clay minerals such as montmorillonite, with a high swelling capacity
6.3.3 RQD

In 1966 Deere and Miller developed the Rock Quality Designation index (RQD) to provide a systematic method of describing rock mass quality from the results of drill core logs. Deere described the RQD as the length (as a percentage of total core length) of intact and sound core pieces that are 4 inches (10 cm) or more in length. Several proposed methods of using the RQD for design of rock tunnels have been developed. However, the major use of the RQD in modern tunnel design is as a major factor in the Q or RMR rock mass classification systems described in the following sub-sections. Readers are referred to Subsurface Investigation Manual (FHWA, 2002) for more details.

6.3.4 Q System

On the basis of an evaluation of a large number of case histories of underground excavations, Barton et al. (1974) of the Norwegian Geotechnical Institute proposed a Tunneling Quality Index (Q) for the determination of rock mass characteristics and tunnel support requirements. According to its developer: "The traditional application of the six-parameter Q-value in rock engineering is for selecting suitable combinations of shotcrete and rock bolts for rock mass reinforcement, and mainly for civil engineering projects". The numerical value of the index Q varies on a logarithmic scale from 0.001 to a maximum of 1,000 and is estimated from the following expression (Barton, 2002):

Q is equal to the product of the Rock Quality Designation, RQD divided by J_n, J_r divided by J_a, and J_w divided by the Stress Reduction Factor, SRF.
6-1

Where RQD is Rock Quality Designation, Jn is joint set number, Jr is joint roughness number, Ja is joint alteration number, Jw is joint water reduction factor, and SRF is stress reduction factor. It should be noted that RQD/Jn is a measure of block size, Jr/Ja is a measure of joint frictional strength, and Jw/SRF is a measure of joint stress.

Table 6-2 (6-2.1 through 6-2) gives the classification of individual parameters used to obtain the Tunneling Quality Index Q for a rock mass. It is to be noted that Barton has incorporated evaluation of more than 1,000 tunnels in developing the Q system.

Table 6-2 Classification of Individual Parameters for Q System (after Barton et al, 1974)

Table 6-2: Classification of Individual Parameters for Q System

Table 6-2 (Continued) Classification of Individual Parameters for Q System (after Barton et al, 1974)

Table 6-2: Classification of Individual Parameters for Q System

Table 6-2 (Continued) Classification of Individual Parameters for Q System (after Barton et al, 1974)

Table 6-2: Classification of Individual Parameters for Q System

Evaluation of these Q-parameters and the use of Table 6-2 can be illustrated considering a reach of tunnel with the following properties:

ParameterDescriptionValueTable
RQD75 to 90RQD = 806-2.1
Joint SetsTwo joint sets plus random jointsJn = 66-2.2
Joint roughnessSmooth, undulatingJr = 26-2.3
Joint alterationSlightly altered joint walls, non-softening mineral coatings, sandy particles, clay-free disintegrated rock, etc.Ja = 26-2.4
Joint water reduction factorMedium inflow with occasional outwash of joint fillingsJw = 0.666-2.5
Stress reduction factorMedium stress, favorable stress conditionSRF = 1.06-2.6

With the parameters established, Q is calculated:

Q is equal to the product of the Rock Quality Designation, RQD divided by J_n, J_r divided by J_a, and J_w divided by the Stress Reduction Factor, SRF equals 80 divided by 6 times 2 divided by 2 times 0.66 divided by 1 equals 9

Refer to Figure 6-25 for guidance in using Q to select excavation support. It should be noted, however, that "the Q-system has its best applications in jointed rock mass where instability is caused by rock falls. For most other types of ground behavior in tunnels, the Q-system, like most other empirical (classification) methods has limitations. The Q support chart gives an indication of the support to be applied, and it should be tempered by sound and practical engineering judgment" (Palmstream and Broch, 2006). The Q-system was developed from over 1000 tunnel projects, most of which are in Scandinavia and all of which were excavated by drill and blast methods. When excavation is by TBM there is considerably less disturbance to the rock than there is with drill and blast. Based upon study of a much smaller data base (Barton, 1991) it is recommended that the Q for TBM excavation be increased by a factor of 2 for Qs between 4 and 30.

6.3.5 Rock Mass Rating (RMR) System

Z.T. Bieniawski (1989) has developed the Rock Mass Rating (RMR) system somewhat along the same lines as the Q system. The RMR uses six parameters, as follows:

  • Uniaxial compressive strength of rock
  • RQD
  • Spacing of discontinuities
  • Condition of discontinuities
  • Groundwater condition
  • Orientation of discontinuities

The ratings for each of these parameters are obtained from Table 6-3. The sum of the six parameters becomes the basic RMR value as demonstrated in the following example. Table 6-9 presents how the RMR can be applied to determining support requirements for a tunnel with a 33 ft (10 m) width span.

Determination of the RMR value using Table 6-3 can be demonstrated in the following example:

ParameterDescriptionTable 6-3Value
Rock Strength20,000 psi = 138 MPaA112
RQD75 to 90A217
Spacing of Discontinuities4 ft –- 1.2MA315
Condition of DiscontinuitiesSlightly rough, slightly weatheredA425
GroundwaterDrippingA54
Discontinuity OrientationFairB-5
Total RatingClass II, Good RockC68

Bieniawski, Barton and others have suggested various correlations between RMR and other parameters. For the purpose of this manual, the most applicable correlation between Q and RMR is given in:

Q is equal to 10 to the power of RMR minus 50 divided by 15.
6-2
6.3.6 Estimation of Rock Mass Deformation Modulus Using Rock Mass Classification

The in situ deformation modulus of a rock mass is an essential parameter for design, analysis and interpretation of monitored data in any rock tunnel project. Evaluation of the stress and deformation behavior of a jointed rock mass requires that the modulus and strength of intact rock be reduced to account for the presence of discontinuities such as joints, bedding, and foliation planes within the rock mass. Since the in situ deformation modulus of a rock mass is extremely difficult and expensive to measure, engineers tend to estimate it by indirect methods. Several attempts have been made to develop relationships for estimating rock mass deformation modulus using rock mass classifications.

The modulus reduction method using RQD requires the measurement of the intact rock modulus from laboratory tests on intact rock samples and subsequent reduction of the laboratory value incorporating the in-situ rockmass value. The reduction in modulus values is accomplished through a widely used correlation of RQD (Rock Quality Designation) with a modulus reduction ratio, EM/EL, where EL represents the laboratory modulus determined from small intact rock samples and EM represents the rock mass modulus, as shown in Figure 6-4. This approach is infrequently used directly in modern tunnel final design projects. However, it is still considered to be a good tool for scoping calculations and to validate the results obtained from direct measurement or other methods.

Correlation between RQD and Modulus Ratio

Figure 6-4 Correlation between RQD and Modulus Ratio (Bieniawski, 1984)

Based on the back analyses of a number of case histories, several methods have been propounded to evaluate the in situ rock mass deformation modulus based on rock mass classification. The methods are summarized in Table 6-4.

Table 6-4 Estimation of Rock Mass Deformation Modulus Using Rock Mass Classification
Rock Mass Deformation Modulus (MPa)Reference
equationSerafim and Pereira (1983)
equationBarton et. al. (1980, 1992), Grimstad and Barton (1993)
equation*Hoek and Brown (1998)
equation**Hoek and Diederichs (2006)
Em = 2RMR - 100 for RMR≥50Bieniawski (1978)
Em = Ei/100[0.0028RMR2 + 0.9 exp(RMR/22.82)]Ei = 50GPaNicholson and Bieniawski (1990)
Em = 0.1(RMR/10)3 Read et. al. (1999)

* GSI represents Geological Strength Index. The value of GSI ranges from 10, for extremely poor rock mass, to 100 for intact rock. (GSI = RMR76 = RMR89 - 5 = 9LogeQ + 44)
** D is a factor which depends upon the degree of disturbance due to blast damage and stress relaxation. It varies from 0 for undisturbed in situ rock masses to 1 for very disturbed rock masses. Guidelines for the selection of D are presented in Table 6-5.

6.4 Rock Tunneling Methods

6.4.1 Drill and Blast

When mankind first started excavating underground, the choices of tools were extremely limited - bones, antlers, wood and rocks, along with a lot of muscle power. Exactly when and where black powder was first used has been lost in history but it is generally agreed that progress was quite slow until the early 1800's. Then, in the mid 1800's, Alfred Nobel invented dynamite and we began to make significant progress in excavations for mining, civil, and military applications. For the reader who wants to pursue this interesting topic, see Hemphill (1981).

Modern drill and blast excavation for civil projects is still very much related to mining and is a mixture of art and science. The basic approach is to drill a pattern of small holes, load them with explosives, and then detonate those explosives thereby creating an opening in the rock. The blasted and broken rock (muck) is then removed and the rock surface is supported so that the whole process can be repeated as many times as necessary to advance the desired opening in the rock.

Table 6-5 Estimation of Disturbance Factor, D
AppearanceDescription of Rock MassSuggested Value
Tunnel surrounded by rock mass

Excellent quality controlled blasting or excavation by Tunnel Boring Machine results in minimal disturbance to the confined rock mass surrounding a tunnel.

D = 0

Tunnel with a temporary invert

Mechanical or hand excavation in poor quality rock masses (no blasting results in minimal disturbance to the surrounding rock mass.

Where squeezing problems result in significant floor heave, disturbance can be severe unless a temporary invert, as shown in the photograph, is placed.

D = 0

D = 0.5
No invert

Tunnel with local damage

Very poor quality blasting in a hard rock tunnel results in severe local damage, extending 2 or 3m, in the surrounding rock mass.

D = 0.8

Rock face after blasting

Small scale blasting in civil engineering slopes results in modest rock mass damage, particularly if controlled blasting is used as shown on the left hand side of the photograph. However, stress relief results in some disturbance.

D = 0.7
Good blasting

D = 1.0
Poor blasting

Open pit mine slope

Very large open pit mine slopes suffer significant disturbance due to heavy production blasting and also due to stress relief from overburden removal.

In some softer rocks excavation can be carried out by ripping and dozing and the degree of damage to the slope is less

D = 1.0
Production blasting

D = 0.7
Mechanical excavation

By its very nature this process leaves a rock surface fractured and disturbed. The disturbance typically extends one to two meters into the rock and can be the initiator of a wedge failure as discussed previously. As a minimum this usually results in an opening larger than needed for its service requirement and in the need to install more supports than would be needed if the opening could be made with fewer disturbances. To reduce the disturbance "Controlled Blasting" technique as discussed in Section 6.4.1.1 can be applied.

6.4.1.1 Controlled Blasting Principles

Explosives work by a rapid chemical reaction that results in a hot gas with much larger volume than that occupied by the explosive. This is possible because the explosive contains both the fuel and the oxidizer. When the explosive detonates, the rapidly expanding gas performs two functions: applying a sharp impulse to the borehole wall (which fractures the rock) and permeating the new fractures and existing discontinuities (which pries the fragments apart). To deliver this one-two punch effectively, the explosive is distributed through the rock mass, by drilling an array of boreholes that are then loaded with explosives and fired in an orderly sequence.

6.4.1.2 Relief

In order to effectively fragment the rock, there must be space for the newly created fragments to move into. If there is not, the rock is fractured but not fragmented, and this unstable mass will remain in place. Therefore the geometry of the array of boreholes must be designed to allow the fragments to move. This is optimum if there is more than one free face available. Creating an artificial "free face" is discussed in Section 6.4.1.5.

6.4.1.3 Delay Sequencing

To optimize the relief, internal free faces must be created during the blast sequence. To do this, millisecond delay detonators separate the firing times of the charges by very short lengths of time. Historically, because of scatter in the firing times of pyrotechnic detonators, "long" period delays between holes (on the order of hundreds of milliseconds) have been used in tunnel and underground mining, resulting in blasts that last several seconds. This is changing as more accurate electronic detonators are developed.

6.4.1.4 Tunnel Blast Specifics

As mentioned, tunnel blasting (like underground mine blasting) differs from surface blasting in that there is usually only one free face that provides relief. To blast some large tunnels, an upper heading is blasted first, and the rest of the rock is taken with a bench blast. Often, though, the whole face is drilled and blasted in one event. An array of blastholes is drilled out using drilling equipment that can drill several holes at once. The pattern of drill holes is determined before the blast, taking into account the rock type, the existing discontinuities in the rock (joints, fractures, bedding planes), and of course the desired final shape of the tunnel. Figure 6-5 shows a rather simplified example of a full-face tunnel round, with the various types of holes. The sequence of firings is Burn Cut (the holes in the neighborhood of the Open Cut Holes shown in the diagram), Production Holes (the holes in the "Blasthole Slash Area"), and the Smoothwall Holes (at the perimeter of the round).

Example of a Full-face Tunnel Blast

Figure 6-5 Example of a Full-face Tunnel Blast

6.4.1.5 Burn Cut

Because the start of each cut with a solid face has no relief, several extra holes are usually drilled and not loaded with explosives in the immediate neighborhood of the initiation point. These burn holes are generally larger than the explosively loaded holes, requiring an additional operation beyond the normal drilling. Many different geometries of burn holes are used to optimize the cut, depending on the rock type and joint patterns in a specific tunnel geology. These holes are fired first, with enough firing time to allow the creation of a free face for the following holes to expand into.

Production Holes The mass of the rock, following the initiation of the burn cut, are fired in a sequence so that the rock moves in a choreographed sequence, moving into the area opened up by the burn cut, and out into the open space in front of the blast.

Wiring up the charges in the right sequence can be a challenging task in the confined environment of a tunnel. Figure 6-6 shows the hook-up of a rather complex blast round, with the surface connectors shown in red, and the period (corresponding to a specific delay time) next to each blast hole.

Complex Round Hook-up

Figure 6-6 Complex Round Hook-up

The desired sequence will fire holes so that there is enough time for rock to move out of the way (create relief) but not so much time that the rock surrounding unfired blast holes will fracture (creating a cutoff).

Perimeter Control It is important to blast so that the final wall is stable and as close to the designed location as possible. The final holes are loaded more lightly, and called "perimeter holes" or "smoothwall holes", and fired with some extra delay so that there is sufficient time for rock to fracture cleanly and create little damage to the rock outside of the "neat" line (such damage is called overbreak). Typical blast charges for these smoothwall holes are shown in Figure 6-7 .

Typical Blast Charges

Figure 6-7 Typical Blast Charges

Though fired after the Production Holes have been detonated, the Smoothwall Holes are often fired on the same delay period, creating a "zipper" effect of the holes generating a smooth fracture on the perimeter.

Environmental Effects - Vibration and Airblast Not all of the energy from blasting goes into fragmenting rocks - some of it is unavoidably wasted as vibration that propagates away from the blast area. This vibration can be cause for concern both for the stability of the tunnel itself, as well as neighboring underground and surface structures.

Airblast is an air pressure wave that propagates away from the blast site, due to movement of the rock face and also possible venting of explosive gases from the boreholes. This is not so much a problem in tunneling, where personnel are evacuated from the blast area before a blast, but still must be taken into account.

Both of these issues are covered in more detail in Chapter 15, Instrumentation.

6.4.1.6 Blasting - Art vs. Science

As mentioned earlier, explosives have been used for a long time to excavate rock. With the passage of time, engineers have studied the scientific relationships between the properties of explosives, the controllable variables such as the geometry of a blast and the timing, and uncontrollable variables such as variations in rock type and existing jointing and fracturing. Many relationships can show the most appropriate configuration of the blastholes, timing, and explosive type. However, as can be seen from the Figure 6-8 of actual drilling for a tunnel blast, the ideal is difficult to achieve.

Holes are marked out with spray paint on an irregular surface, and drilled in a dirty, often wet environment. The roof is supported with rock bolts (shown by the red squares in Figure 6-8 ) and meshes. Lighting is limited. Overall, this makes for a very challenging work environment.

Experience, or the "art" of blasting, comes into play in implementing the desired blast design. Choice of an experienced and capable blasting contractor, as well as a blast consultant to advise the contractor, is important to obtain the desired results.

Drilling for a Tunnel Blast

Figure 6-8 Drilling for a Tunnel Blast

6.4.2 Tunnel Boring Machines (TBM)

While progress and mechanization continued to be applied to drill and blast excavation well into the 1960's, the actual advance rates were still quite low, usually measured in feet per day. Mechanized tunneling machines or tunnel boring machines had been envisioned for over a century but they had never proven successful. That began to change in the 1960's when attempts were made to apply oil field drilling technology. Some progress was made, but it was slow because the physics were wrong - the machines attempted to remove the rock by grinding it rather than by excavating it. All of that changed in the later 1960's with the introduction of the disk cutter, see Figure 6-9. The disk cutter causes the rock to fail in shear, forming slabs (chips) of rock that are measured in tens of cubic inches rather than small fractions of a cubic inch. Much of the credit for this development, which now allows tunnels to advance at 10's or even 100's of feet per day, belongs to The Robbins Co.

Chipping Process between Two Disc Cutters

Figure 6-9 Chipping Process between Two Disc Cutters (After Herrenknecht, 2003)

Today, tunnel boring machines (TBM) excavate rock mass in a form of rotating and crushing by applying enormous pressure on the face with large thrust forces while rotating and chipping with a number of disc cutters mounted on the machine face (cutterhead) as shown in Figure 6-10. Design of disc cutters RPM, geometry, spacing, thrust level, etc. are beyond the scope of this manual.

Rock Tunnel Boring Machine Face with Disc Cutters for Hard Rock, Australia.

Figure 6-10 Rock Tunnel Boring Machine Face with Disc Cutters for Hard Rock, Australia.

6.4.2.1 Machine Types and Systems

Tunnel Boring Machines (TBMs) nowadays are full-face, rotational (with cutter heads) excavation machines that can be generally classified into two general categories: Gripper and Segment as shown in Figure 6-11. Based on Figure 6-11, there are three general types of TBMs suitable for rock tunneling including Open Gripper/Main Beam, Closed Gripper/Shield, and Closed Segment Shield, as shown within the dashed box on the Figure.

The open gripper/beam type of TBMs are best suited for stable to friable rock with occasional fractured zones and controllable groundwater inflows. As shown in Appendix D, three common types of TBMs belong to this category including Main Beam (Figure 6-12), Kelly Drive, and Open Gripper (without a beam or Kelly).

The closed shield type of TBMs for most rock tunneling applications are suitable for friable to unstable rocks which cannot provide consistent support to the gripper pressure. The closed shield type of TBMs can either be advanced by pushing against segment, or gripper. Note that although these machines are classified as a closed type of machine, they are not pressurized at the face of the machine thus cannot handle high external groundwater pressure or water inflows. Shielded TBMs for rock tunneling include: Single Shield (Figure 6-13), Double Shield (Figure 6-14), and Gripper Shield.

The typical machine elements and backup system for each category are discussed in the following section. Pressurized-face Closed Shield TBMs are predominantly utilized in tunneling in soft ground and are discussed in Chapter 7. Appendix D presents descriptions for various types of TBMs.

Classification of Tunnel Excavation Machines

Figure 6-11 Classification of Tunnel Excavation Machines

6.4.2.2 Machine Main and Support Elements

A TBM is a complex system with a main body and other supporting elements to be made up of mechanisms for cutting, shoving, steering, gripping, shielding, exploratory drilling, ground control and support, lining erection, spoil (muck) removal, ventilation and power supply. As shown in Figures 6-12, 6-13 and 6-14, the main body of a typical rock TBM (either open or closed) includes some or all of the following components:

  • Cutterhead and Support
  • Gripper (Except Single Shield TBM)
  • Shield (Except Open TBM)
  • Thrust Cylinder
  • Conveyor
  • Rock Reinforcement Equipment

In addition, the main body of a TBM is supported with a trailing system for muck and material transportation, ventilation, power supply, etc. A fully equipped TBM can occupy over 1000 ft of tunnel.

Appendix D includes detailed descriptions for each of the above TBM types.

Typical Diagram for a Open Gripper Main Beam TBM.

Figure 6-12 Typical Diagram for a Open Gripper Main Beam TBM (Robbins).

Typical Diagram for Single Shield TBM

Figure 6-13 Typical Diagram for Single Shield TBM (Robbins)

Typical Diagram for Double Shield TBM

Figure 6-14 Typical Diagram for Double Shield TBM (Robbins)

6.4.2.3 Compatible Ground Support Elements

Most ground support elements discussed in Section 6.5 can be specified with the use of hard rock TBMs, especially if the TBMs are manufactured specifically for the project.

  • Rock reinforcement by roof bolting
  • Spiling/forepoling
  • Pre-injection
  • Steel ring beams with or without lagging (wire mesh, timber, etc.)
  • Invert segment
  • Shotcrete
  • Precast concrete segmental lining
  • Others

Details of the above support measures are discussed in Section 6.5 of this Chapter, and Chapter 10 Tunnel Lining.

6.4.2.4 TBM Penetration Rate

With a rock TBM, the penetration rate is affected by the following factors (from Robbins, 1990):

  • Total machine thrust
  • Cutter spacing
  • Cutter diameter and edge geometry
  • Cutterhead turning speed (revolutions per minute)
  • Cutterhead drive torque
  • Diameter of tunnel
  • Strength, hardness, and abrasivity of the rock
  • Jointing, weathering and other characteristics of the rock.

However, penetration rate (an instantaneous parameter) by itself does not assure a high average advance rate. The latter requires a good combination of penetration rate and actual cutting time. In turn, actual cutting time is affected by the following factors:

  • Learning (start-up) curves
  • Downtime for changing cutters
  • Downtime for other machine repairs/maintenance
  • Overly complex designs
  • Back-up (trailing) systems
  • Tunnel support requirements
  • Muck handling
  • Water handling
  • Probe hole drilling, grouting
  • Available time (total and shift)

The bottom line is that actual utilization typically runs in the range of 50% as shown by Figure 6-15.

TBM Utilization on Two Norwegian Tunnels

Figure 6-15 TBM Utilization on Two Norwegian Tunnels (After Robbins, 1990)

<< Previous Contents Next >>
Updated: 06/19/2013
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000