|<< Previous||Contents||Next >>|
Technical Manual for Design and Construction of Road Tunnels - Civil Elements
Chapter 7 - Soft Ground Tunneling
Chapters 6 through 10 present design recommendations and requirements for mined and bored road tunnels. Chapter 7 addresses analysis, design and construction issues specifically for tunneling (mostly shield tunneling) in soft ground including cohesive soils, cohesionless soils and silty sands. Chapter 10 addresses the design of various types of permanent lining applicable for soft ground tunnels.
Human kind has been excavating in soft ground for thousands of years. Archeological digs in Europe and elsewhere show that all kinds of tools were used by our ancestors to excavate soil (mostly for "caves" in which to live): bones, antlers, sticks, rocks and the like. However, there are tunnels in Europe that were built by the Romans, are over 2000 years old, and are still in service carrying water. As the population grows and we demand more and more transportation services, there can be no doubt that the requirement for tunnels will also grow. Through it all, the art of tunnel design and construction will also continue to develop, but it is doubtful that this art will ever develop into a science comparable to structural design. The structural engineer can specify both the configuration and the properties in great detail; the tunnel engineer must work with existing materials that cannot be specified and, in addition, are constantly changing, often dramatically.
Problematic soft ground conditions such as running sand and very soft clays are discussed in Chapter 8. Mining sequentially through soft ground based on the sequential excavation method (SEM) principles is discussed in Chapter 9. The data needed for analysis and design is discussed in Chapter 3. The results of the analysis and design presented herein are typically presented in the Geotechnical design memorandum (Chapter 4) and form the basis of the Geotechnical Baseline Report (Chapter 4).
7.2 Ground Behavior
7.2.1 Soft Ground Classification
Anticipated ground behavior in soft ground tunnels was first defined by Terzaghi (1950) by means of the Tunnelman's Ground Classification (Table 7-1). It can be also be discussed in terms of soil identification (by particle size) and by considering behavior above and below the water table as summarized in the following.
Cohesive Soils and Silty Sand Above Water Table Cohesive (clayey) soils behave as a ductile plastic material that moves into the tunnel in a theoretically uniform manner. Following Peck's (1969) lead for cohesive (clay) materials or materials with sufficient cohesion or cementation to sample and test for unconfined compression strength, an estimate of ground behavior in tunneling can be obtained from the equation:
Where Ncrit is the stability factor, Pz is the overburden pressure to the tunnel centerline, Pa is the equivalent uniform interior pressure applied to the face (as by breasting or compressed air), and Su is the undrained shear strength (defined for this purpose as one-half of the unconfined compressive strength).
|Classification||Behavior||Typical Soil Types|
|Firm||Heading can be advanced without initial support, and final lining can be constructed before ground starts to move.||Loess above water table; hard clay, marl, cemented sand and gravel when not highly overstressed.|
|Chunks or flakes of material begin to drop out of the arch or walls sometime after the ground has been exposed, due to loosening or to over-stress and "brittle" fracture (ground separates or breaks along distinct surfaces, opposed to squeezing ground). In fast raveling ground, the process starts within a few minutes, otherwise the ground is slow raveling.||Residual soils or sand with small amounts of binder may be fast raveling below the water tale, slow raveling above. Stiff fissured clays may be slow or fast raveling depending upon degree of overstress.|
|Squeezing||Ground squeezes or extrudes plastically into tunnel, without visible fracturing or loss of continuity, and without perceptible increase in water content. Ductile, plastic yield and flow due to overstress.||Ground with low frictional strength. Rate of squeeze depends on degree of overstress. Occurs at shallow to medium depth in clay of very soft to medium consistency. Stiff to hard clay under high cover may move in combination of raveling at excavation surface and squeezing at depth behind surface.|
|Granular materials without cohesion are unstable at a slope greater than their angle of repose (approx 30o -35o ). When exposed at steeper slopes they run like granulated sugar or dune sand until the slope flattens to the angle of repose.||Clean, dry granular materials. Apparent cohesion in moist sand, or weak cementation in any granular soil, may allow the material to stand for a brief period of raveling before it breaks down and runs. Such behavior is cohesive-running.|
|Flowing||A mixture of soil and water flows into the tunnel like a viscous fluid. The material can enter the tunnel from the invert as well as from the face, crown, and walls, and can flow for great distances, completely filling the tunnel in some cases.||Below the water table in silt, sand, or gravel without enough clay content to give significant cohesion and plasticity. May also occur in highly sensitive clay when such material is disturbed.|
|Swelling||Ground absorbs water, increases in volume, and expands slowly into the tunnel.||Highly preconsolidated clay with plasticity index in excess of about 30, generally containing significant percentages of montmorillonite.|
* Modified by Heuer (1974) from Terzaghi (1950)
Table 7-2 shows the anticipated behavior of tunneling in clayey soils (Modified from Peck 1969 and Phienwaja 1987.) Silty sand above water table may have some (apparent) cohesion but they typically behave in a brittle manner adjacent to the tunnel opening. Predicting their behavior by the above equation is more subjective but may be attempted as shown in Table 7-2.
|Stability Factor, Ncrit||Soft Ground Tunnel Behavior|
|4-5||Creeping, usually slow enough to permit tunneling|
|6||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|
|1/2 - 1||Raveling|
Cohesionless Granular Soils including Silty Sand Below the Water Table From the tunneling perspective, dry or partially saturated sand and gravel above the groundwater table may possess some temporary apparent cohesion from negative pore pressure. When the material is below the water table, it lacks sufficient cohesion or cementation and the behavior is more subjective and can easily run or flow into the excavation. The behavior of sands and gravels in tunneling were summarized by Terzaghi (1977) and that summary still applies (Table 7-3). Note that the cleaner the sand, the more liable it is to run or flow when exposed in an unsupported vertical face during tunnel construction. Chapter 8 provides more discussion for running and flowing sands
For silty sands below the groundwater table, they can be problematic and flow if the uniformity coefficient Cu is not less than 3 and flowing to cohesive running if Cu is less than 6 (Terzaghi 1977).
7.2.2 Changes of Equilibrium during Construction
Excavation of a soft ground tunnel opening and the subsequent construction of supports change the stress conditions for the tunnel and the surrounding medium. These changes may be continuous or in stages. A comprehension of the deformations associated with these changes is necessary for an understanding of the behavior of tunnel supports.
"The state of the medium before the excavation of a tunnel cavity is one of equilibrium in a gravity field. The process of tunneling evokes new equilibrium conditions which will change during the various stages of tunneling and construction of supports until a final equilibrium is reached. In this final equilibrium, all changes in strain and stress around the tunnel opening cease and a new equilibrium condition is established.
|Designation||Degree of Compactness||Tunnel Behavior|
|Above Water Table||Below Water Table|
|Very Fine Clean Sand||Loose, N<10|
|Cohesive RunningFast Raveling||FlowingFlowing|
|Fine Sand With Clay Binder||Loose, N <10|
Dense, N >30
|Rapid RavelingFirm or Slowly Raveling||FlowingSlowly Raveling|
|Sand or Sandy Gravel with Clay Binder||Loose, N<10|
|Rapid RavelingFirm||Rapidly Raveling or FlowingFirm or Slow Raveling|
|Sandy Gravel and Medium to Coarse Sand||Running ground. Uniform (Cu <3) and loose (N<10) materials with round grains run much more freely than well graded (Cu >6) and dense (N>30) ones with angular grains.||Flowing conditions combined with extremely heavy discharge of water.|
A region of changing stresses, characterized by increased vertical pressure, travels ahead of the advancing face of the tunnel. Changes of equilibrium conditions are also felt at a considerable distance behind the face. The distribution of stresses has a three dimensional character at a point near the face, but approaches a two-dimensional state as the face advances. The rate at which the two-dimensional state is approached is influenced by the rate of advance of the face in relation to the time-dependent behavior of the medium.
The continuous or frequent changes in the conditions for stress equilibrium cannot take place without deformations in the medium. If supports are employed, these will deform as well. There is always an immediate deformation response to a change in equilibrium conditions, and commonly there is an additional, time-dependent response. In a waterbearing medium, the excavation of a tunnel changes the pore water pressures around the opening, and flow of water is induced. In fine grained materials with a low permeability, the establishment of hydrostatic or hydrodynamic equilibrium is not immediate. The associated time-dependent changes in effective intergranular pressures in the medium then lead to time-dependent deformations.
Time lags may also be associated with visco-elastic or visco-plastic phenomena such as creep in the medium itself or along joint planes in the medium. Whatever the cause of the time lags, their most important effect is that a final equilibrium for a set of boundary conditions often is not reached before new changes in boundary conditions occur.
Tunnel construction not only changes the equilibrium conditions but in many cases the medium itself. Blasting commonly reduces the strength of the rock around the opening; shoving by a closed or nearly closed shield disturbs and may remold the soil. Indeed, disturbing the material in the immediate vicinity of the opening is hardly avoidable. Where a tunnel is advanced without blasting in a medium which requires little or no immediate support, however, the disturbance may be minimal.
7.2.3 The Influence of the Support System on Equilibrium Conditions
Most tunnel openings are supported at some stage of construction. The behavior of a tunnel opening and a support system is dependent on the time and manner of the placement of the support and its deformational characteristics.
The reasons for providing support are manifold. Sometimes support is required for the immediate stability of the opening. It may be furnished even before excavation, for example by air pressure, forepoling or ground improvements. Under these circumstances the interaction between the medium and the supporting agent commences during or before excavation. When a shield is used for immediate support, a lining is erected inside the shield, and the annular void cleared by the shove of the shield is at least partly filled with pea-gravel and/or grout. The lining may be intended as a permanent support consisting, for example, of precast concrete segments. It may alternatively be a relatively flexible one in which a stiffer permanent lining will later be constructed. In this event, at least three different equilibrium conditions must be considered.
Where there is need for long-term but not immediate support, the support may be constructed at some distance behind the face. A partial relaxation with associated movements may then take place before the support interacts with the medium. Often a liner is erected and expanded into contact with the medium. The expansion induces a prestress in both the liner and the medium and influences subsequent deformations.
Even where instability or collapse of the opening is not imminent, support may still be required for various reasons, usually to control or limit deformations. Large deformations may lead to undesired settlements of the ground surface or to interference with other structures. Such deformation must be restrained at a suitably early stage. Deformations of a soil or rock mass commonly result in an undesirable reduction in strength and coherence of the medium. In a jointed or weak rock the material above the opening tends to loosen and may sooner or later exert considerable loads on the support. These loads are reduced if loosening is prevented by suitable support.
Although the initial stability may be satisfactory, conditions may be such that final equilibrium cannot be reached without support. This may occur in jointed rock mass subject to progressive loosening, in creeping or swelling materials, and in materials whose strength decreases with time. Except in such creeping materials as salts, these long term phenomena are associated with volume changes.
It is impossible and undesirable to avoid deformations in the soft ground altogether. Some movement is necessary to obtain a favorable distribution of loading between the medium and the support system. In each instance, the engineer must determine how much movement is beneficial to the behavior of the tunnel, and at what movements the effects will become detrimental. The engineer's conclusions regarding these matters determine whether and where restraints are to be applied to the tunnel walls. His conclusions also determine the character and magnitude of those restraints. In tunnels in hard rock the beneficial movements take place almost immediately, and subsequent movements are likely to lead to loosening and additional loading. Hence, in this case rapid construction of supports is usually desirable.
It is apparent that many factors determine whether and where a support system should be constructed for structural reasons alone. The final choice of whether and where supports are actually employed is influenced by additional factors such as the psychological well-being of the workers, or the economy that might be achieved by adopting a uniform construction procedure throughout the same tunnel even though the properties of the medium vary.
No matter what the reason for using restraints, the loads to which a support will be subjected depend on the stage of equilibrium prevailing at the time the support is introduced. Thus, if final equilibrium has been reached before support is provided, the support may not receive loads from the medium at all. On the other hand, when support is furnished before final equilibrium has been established, new boundary conditions are superimposed on the conditions existing at the time the support is constructed. The new final conditions depend on the time the support was provided and involve the interaction between the support and the medium. If a stiff support could be installed in the medium before excavation by an imaginary process that did not in any way disturb the remaining material, it would be subjected to stresses resembling those of the in-situ condition existing before the excavation. However, the at least temporary reduction of the radial stresses to atmospheric pressure (or to the air pressure in the tunnel), as well as many other activities, generally introduce such deformations into the medium that the stresses ultimately acting on the tunnel support bear little or no resemblance to the initial stresses in the medium.
Procedures for the analysis and design of tunnel supports are necessarily simplified, but they should be based on the considerations of equilibrium and deformations briefly outlined above. In addition, a number of factors which are not directly related to the interaction between a support system and the medium are significant in the actual design of supports. Such factors, which are dealt with in the following section, sometimes even override considerations of structural interaction." (After Deere, 1969).
7.3 Excavation Methods
7.3.1 Shield Tunneling
Generally soft ground tunneling did not become viable until the introduction of the tunnel shield (accredited to Sir Marc Brunel), except for small hand-excavated openings in soft ground and somewhat larger ones in soft rock, tunneling. Brunel wrote: "The great desideratum (sic) therefore consists in finding efficacious means of opening the ground in such a manner that no more earth shall be misplaced than is to be filled by the shell or body of the tunnel and that the work shall be effected with certainty" (Copperthwaite, 1906). In other words, never open more than is needed, can be excavated rapidly, and quickly supported. Brunel patented a circular shield (Figure 7-1 ) in 1818 that was described by Copperthwaite (1906) as covering "every subsequent development in the construction and working of tunnel shields."
Figure 7-1 Patent Drawing for Brunel's Shield, 1818 (Cooperthwaite, 1906).
If we fast forward, we find that nearly all soft/ground tunnels driven in North America into the 1960's and the early 1970's were mostly under 10 feet (3 m) diameter and driven using the basic concepts of the Brunel tunnel shield; viz, compartmentalized, face breasting with timber and lots of hand labor.
In ground conditions that required a higher level of support than the basic Brunel shield, compressed air was commonly used (actually from the mid 1800's into the 1980's). When used correctly, compressed air provided the needed support and allowed many tunnels to be completed that would otherwise not have been possible. Because of the decompression required and all the associated equipment and procedures, not to mention the potential hazards to the workers, e.g., the bends or even death, compressed air has largely been eliminated as a tunneling adjunct.
Starting in the late 1960's and early 1970's, mechanization began to be introduced by incorporating excavating machines within the circular shields, hence the term digger shield (Figures 7-2 and 7-3).
Figure 7-2 Digger Shield with Hydraulically Operated Breasting Plates on Periphery of Top Heading of Shield used to Construct Transit Tunnel.
Figure 7-3 Cross-section of Digger Shield
However, digger shield machines too often met with poor results and were usually unsatisfactory for three reasons:
- Ground loss occurred ahead and above the shield when retracting the doors or poling plates. Typically, the orange-peel doors could not be retracted in tune with the forward progress of the shield. Also when retracting the doors, the miner does not have access to deal with running ground. Thus the machine encouraged unwanted ground movement, rather than controlling it.
- Maintaining the right soil "plug" in the invert was always a headache.
- Mounting the digger in the center created a "Catch 22": if the ground movement in the center became excessive, the only way to stop it was to cram the digger bucket into the face. However, that made it impossible to excavate and move the shield forward because to do so meant the bucket had to be moved, allowing the face to fail.
Shields with open faced wheeled excavators were another, early step in mechanization of soft-ground machines that have some things in common with their cousins the hard rock TBMs. Wheeled excavators were used with success in firm ground conditions, but not so well in running or fast raveling ground conditions. In some ground conditions this arrangement was marginally successful but in general it was not possible always to control the amount of ground allowed through the wheel to be equal to only that described by the cutting edge of the shield.
Figure 6-11 shows the types of tunnel boring machines suitable for soft ground conditions. The various conventional shield tunneling methods are summarized by Zosen (1984) as show in Table 7-4 . The following sections focus on the modern Earth Pressure Balance (EPB) and Slurry Face Machines (SFM).
7.3.2 Earth Pressure Balance and Slurry Face Shield Tunnel Boring Machines
As a turning point in global tunneling equipment development, soft ground tunnel shields equipped with wheeled excavators were exported to Japan . Further development of soft-ground tunneling machines was flat in the USA for many years, Japan, however, took a good idea, invested heavily in equipment development and within a decade or so exported vastly improved tunneling methods back to the USA in the form of pressurized-face tunneling machines.
Thus as soft ground tunneling in the USA was affixed with traditional shield tunneling, the Japanese, Europeans (read that Germans), the UK, and Canadians were developing two more "modern" machines - the earth pressure balance machine (EPB) and the slurry face machine (SFM) that are also summarized in Table 7-4 (also Figure 7-4 to Figure 7-7 ).
At first-hand, these machines are similar in that they both have:
- A revolving cutter wheel.
- An internal bulkhead that traps cut soil against the face (hence, they are called closed face), and that maintains the combined effective soil and water pressure and thereby stabilizes the face.
- No workers are at the face but rely on mechanization and computerization to control all functions, except segment erection (to date).
- Precast concrete segments erected in the shield tail, with the machine advanced by shoving off those segments.
|Open face, hang-dug shield||
|Slurry face Machine||
|Earth pressure balance (EPB) machine||
|Earth pressure balance (EPB) high-density slurry machine||
The actual functioning of the machines, however, has some distinct differences: in the EPB the pressure is transmitted to the face mechanically, through the soil grains, and is reduced by means of friction over the length of the screw conveyor. Control is obtained by matching the volume of soil displaced by forward motion of the shield with the volume of soil removed from the pressurized face by that screw conveyor and deposited (at ambient pressure) on the conveyor or muck car. Clearly the range of natural geologic conditions that will result in suitably plastic material to transfer the earth pressure to the face and, at the same time, suitably frictional to form the "sand plug" in the screw conveyor is rather limited - generally only combinations of fine sands and silts.
Figure 7-4 Earth Pressure Balance Tunnel Boring Machine (EPB) (Lovat).
Figure 7-5 Simplified Cross-section of Earth Pressure Balance Tunnel Boring Machine (EPB).
In contrast, the SFM transmits pressure to the face hydraulically through a viscous fluid-formed by the material cut and trapped at the face and mixed with slurry (basically bentonite and water). In this case the pressure transmitted can be controlled by means of pressure gages and control valves in a piping system. By this system a much more precise and more consistent pressure control is attained. The undesirable aspect of this system is the separation plant that has to be built and operated at the surface to separate the slurry from the soil cuttings for disposal and permit re-use of the slurry. Finding a site for the slurry separation that is satisfactory for the process and acceptable to the public can present interesting challenges.
Figure 7-6 Slurry Face Tunnel Boring Machine (SFM) (courtesy of Herrenknecht).
Figure 7-7 Simplified Cross-section of Slurry Face Tunnel Boring Machine (SFM)
During the last decade or so, great strides have been made in developing new families of conditioning agents that can be used in both types of closed face machines. These additives tend to blur the distinctions portrayed above and widen the range of applicability of both types of machines. Indeed, we predict that in another decade we will not be talking about the two type of machines but rather a new family of machines that will operate interchangeably and with equal efficiency as an open face wheel machine in stable ground or as a closed face machine (with conditioners) that will cut any type of soft ground. Herrenknecht, for one, is already moving ahead with development of this new breed of machine.
Throughout all of this development, the role of the miner at the tunnel face is steadily being diminished. With any closed face machine, the miner is not doing any excavating or breasting of the face. The miner is operating machines that, unfortunately, can not always do the job as advertised. (After Hansmire and Monsees, 2005)
7.3.3 Choosing between Earth Pressure Balance Machines and Slurry Tunneling Machines
The choice of the type of closed-face tunneling machine and its facilities is a critical decision on a soft-ground tunneling project. This decision will be guided by thorough assessment of the ground types and conditions to be encountered and by numerous other aspects.
Other aspects that will influence the choice include the particular experience of the project's contractor, the logistics and configuration of the works, and requirements of the contract as a means to ensure that the client's minimum specification is met. The initial choice is guided by reference to the grading envelope of the soils to be excavated. Since it is likely that the geology will fall into more than one envelope, the final choice may require a degree of compromise or development of a dual-mode open/closed-faced TBM system or a dual slurry/EPB system.
Review of Ground Types In many tunnel drives the conditions encountered along the route may vary significantly with a resulting need to specify a system capable of handling the full range of expected conditions. Closed-face tunneling machines can be designed and manufactured to cope with a range of ground conditions. Some machines are capable of handling many or all of this range of anticipated conditions with a limited degree or reconfiguration for efficient operation.
There have been several attempts to classify the naturally occurring range of soft-ground characteristics from the tunneler's perspective. This work was summarized most recently by Whittaker and Frith (1990) and the following categorization is based partly on their work. It consists of eight categories of physical ground behavior that may be observed within the soft-ground tunnel excavation range. The characteristics are summarized in Table 7-5 . Each of these may be associated with particular types of soils.
Selection Criteria Based on Particle Size Distribution and Plasticity An SFM is ideal in loose waterbearing granular soils that are easily separated at the separation plant. By contrast SFMs have problems dealing with clays and some silts.
If the amount of fines (particles smaller than 60 mm or able to pass through a 200 sieve) is greater than 20% then the use of an SFM becomes questionable although it is not ruled out. In this situation it will be the difficulty in separating excavated spoil from the slurry, rather than the operation of the TBM, that is likely to affect critically the contract program and the operating cost.
An EPBM will perform better where the ground is silty and has a high percentage of fines both of which will assist the formation of a plug in the screw conveyor and will control groundwater inflows. A fines content of below 10% may be unfavourable for application of EPBMs. For an EPBM the costs of dealing with poorly graded or no-fines soil will be in the greater use of conditioners and possibly, in extreme cases, the use of positive displacement devices, such as rotary feeders or piston dischargers, at the screw conveyor discharge point to maintain EPB pressures.
Higher plasticity index (PI) clays ('sticky clays') can lead to 'balling' problems and increased problems at the separation plant for SFMs. Similarly these materials can be problematic for EPBMs where special attention is required in selecting the most appropriate conditioning agents.
|Firm ground||Ground in which the tunnel can be advanced safely without providing direct support to the face during the normal excavation cycle and in which ground support or the lining can be installed before problematic ground movement occurs. Where this short-term stability may be attributable to the development of negative pore pressure in the fine grained soils, significant soil movements and/or ground loading of the tunnel lining may occur later. Examples may include stiff clays and some dewatered sands. A closed-face tunneling machine may not be needed in this ground type.|
|Raveling ground||Ground characterized by material that tends to deteriorate with time through a process of individual particles or blocks of ground falling from the excavation surface. Examples may include glacial tills, sands and gravels. In this ground a closed-face tunneling system may be required to provide immediate support to the ground.|
|Running or flowing ground||Ground characterized by material such as sands, silts and gravels in the presence of water, and some highly sensitive clays that tend to flow into an excavation. Above the water table running ground may occur in granular materials such as dry sands and gravels. Below the water table a fluidized mixture of soil and water may flow as a liquid. This is referred to as running or flowing ground. Such materials can sometimes pass rapidly through small openings and may completely fill a heading in a short period of time. In all running or flowing ground types there will be considerable potential for rapid over-excavation. Hence, a closed-face tunneling system will be required to support such ground safely unless some other method of stabilization is used.|
|Squeezing ground||Ground in which the excavation-induced stress relief leads to ductile, plastic yield of ground into the tunnel opening. The phenomenon usually is exhibited in soft clays and stiffer clays over a more extended period of time. A closed-face machine may be required to provide resistance to squeezing ground, although in some conditions there is also a risk of the TBM shield becoming trapped.|
|Swelling ground||Soil characterized by a tendency to increase in volume due to absorption of water. This behavior is most likely to occur either in highly over-consolidated clay or in clays containing minerals naturally prone to significant swelling. A closed-face machine may be useful in providing resistance to swelling ground although, as with squeezing ground, there is a risk of the shield becoming trapped.|
|Weak rock||Weak rock may be regarded effectively as a soft-ground environment for tunneling because systems used to excavate soft-ground types may also be applied to weak rock materials such as chalk. Weak rock will often tend to be self-supporting in the short term with the result that closed face tunneling systems may not be needed. However, groundwater may be significant issue. In these instances a closed-face machine is an effective method of protecting the works against high volumes of water ingress that could also be under high hydrostatic pressure.|
|Hard rock||A closed-face TBM may also be deployed in normally self-supporting hard rock conditions. The main reason would be to provide protection against groundwater pressures and prevent inundation of the heading.|
|Mixed ground conditions||Potentially, the most difficult of situations for a closed-face tunneling system is that of having to cope with a mixture of different ground types either along the tunnel from zone to zone or sometimes from meter to meter, or within the same tunnel face. Ideally the vertical alignment would be optimized to avoid, as far as possible, a mixed ground situation, however, in urban locations the alignment may be constrained by other considerations. For changes in ground types longitudinally, a closed-face machine may have to convert from a closed-face pressurized mode to an open non-pressurized mode when working in harder ground types to avoid over stressing the machine's mechanical functions. Such a change may require some modification of the machine and the reverse once again when the alignment enters a reach of soft, potentially unstable ground. In the case of mixed ground types across the same face, the tunneling machine will almost certainly have to operate in a compromise configuration. In such cases great care will be needed to ensure that this provides effective ground control. A common problem, for example, is a face with a hard material in the bottom and running ground at the top. In this situation the TBM will generally advance slowly while cutting the hard ground but may tend to draw in the less stable material at the top leading to over-excavation of the less stable material and subsequent subsidence or settlement at the surface. Different ground types at levels above the tunnel will also be of significance. For example, in the event that over-excavation occurs, the presence of running or flowing materials at horizons above the tunnel will increase the potential quantity of ground that may be over-excavated and again lead to subsidence or surface settlement. Another potential problem occurs when a more competent layer exists over potentially running ground in which case possible over-excavation would create voids above the tunnel and below the competent material, giving rise to potential longer-term instability problems.|
Permeability As a general guide the point of selection between the two types of machines is a ground permeability of 1x10-5 m/s, by using SFMs applicable to ground of higher permeability and EPBMs for ground of lower permeability. However, an EPBM can be used at a permeability of greater than 1x10-5 m/s by using an increased percentage of conditioning agent in the plenum. The choice will take into account the content of fines and the ground permeability.
Hydrostatic Head High hydrostatic heads of groundwater pressure along the tunnel alignment add a significant concern to the choice of TBM. In situations where a high hydrostatic head is combined with high permeability or fissures it maybe be difficult to form an adequate plug in the screw conveyor of an EPBM. Under such conditions an SFM may be the more appropriate choice especially as the bentonite slurry will aid in sealing the face during interventions under compressed air.
Settlement Criteria Both types of machine are effective in controlling ground movement and surface settlement - providing they are operated correctly. While settlement control may not be overriding factor in the choice of TBM type, the costs associated with minimizing settlement should be considered. For example, large quantities of conditioning agent may be needed to reduce the risk of over-excavation and control settlement if using EPBM in loose granular soils. See Section 7.5.
Final Considerations Other aspects to consider when making the choice between the use of an SFM or an EPBM include the presence of gas, the presence of boulders, the torque and thrust required for each type of TBM and, lastly, the national experience with each method. These factors should be considered but would not necessarily dictate the choice.
The overriding decision must be made on which type of machine is best able to provide stability of the ground during excavation with all the correct operational controls in place and being used.
If both types of machine can provide optimum face stability, as is often the case, other factors, such as the diameter, length and alignment of the tunnel, the increased cutter wear associated with EPBM operation, the work site area and location, and spoil disposal regulations are taken into consideration.
The correct choice of machine operated without the correct management and operating controls is as bad as choosing the wrong type of machine for the project. (After British Tunneling Society, BTS, 2005)
7.3.4 Sequential Excavation Method (SEM)
In addition to shield tunneling methods discussed above, soft ground tunnels can be excavated sequentially by small drifts and openings following the principles of the Sequential Excavation Method (SEM), aka New Austrian Tunneling Method (NATM) first promulgated by Professor Rabcewicz (1965). The SEM has now been defined as "a method where the surrounding rock or soil formations of a tunnel or underground opening are integrated into an overall ring-like support structure and the following principles must be observed:
- The geotechnical behavior must be taken into account
- Adverse states of stresses and deformations must be avoided by applying the appropriate means of support in due time.
- The completion of the invert gives the above mentioned ring-like structure the static properties of a tube.
- The support means can/should be optimized according to the admissible deformations.
- General control, geotechnical measurements and constant checks on the optimization of the pre-established support means must be performed. (From ILF, 2004)
The underlying principle of SEM is actually the same as that stated by Sir Marc Brunel almost two centuries ago: "The great desideratum therefore consists in finding efficacious means of opening the ground in such a manner that no more earth shall be displaced than is to be filled by the shell or body of the tunnel and that the work shall be effected with certainty". (Copperthwaite, 1906) In other words, never open more than is needed, can be excavated rapidly, and quickly supported.
As applied to soft ground tunneling, SEM generally cannot compete with tunneling machines for long running tunnels but often is a viable method for:
- Short tunnels
- Large openings such as stations
- Unusual shapes or complex structures such as intersections
Refer to Chapter 9 for detail discussion regarding SEM/NATM.
|<< Previous||Contents||Next >>|