Parsons Brinckerhoff
Worldwide LocationsContact PBLegal NoticeSite MapHome
PB Websites search Advanced Search
MarketsServicesAbout UsPeople and CareersNews and EventsResearch LibraryProjects
PB Network Email This Page
Go To Other Issues | Contact PB Network | Print This Article 
<< Go To Previous Article | Table Of Contents | Go To Next Article >>
Slurry Walls
Fall 1996 • Issue No. 36 • Volume X • Number 3
Analysis and Design
Using SPTC Walls as Permanent Tunnel Walls
By Prabir K. Das, Boston CA/T 617-951-6163, Huang Ni, Boston CA/T 617-951-6525, Anthony L. Ricci, Massachusetts Highway Department and Paul Harrington, Fay Spofford Thorndike
Unlike many slurry walls, which are used as temporary support for excavation during construction, the SPTC walls on the CA/T project were designed also as the permanent wall system for the tunnels. This article focuses on their analysis and design.


Construction Site for the Central Artery/Tunnel

The I-93 portion of the Central Artery/Tunnel project includes approximately 7 km (23,000 linear feet) of slurry walls (Figure 1), most of which will be Soldier Pile Tremie Concrete (SPTC) walls. Steel wide-flange soldier piles, forming the primary support system for the wall, are installed at 1.2 m to 1.8 m (4 feet to 6 feet) spacings. In most locations, the concrete is designed to act as lagging spanning between the structural steel members. The concrete base slab is rigidly connected to the SPTC walls and the composite roof is pin connected to the walls to form the tunnel.

Groundwater Conditions

The groundwater level along the Central Artery is typically 2.4 to 3.1 meters (8 to 10 feet) below ground surface. Temporary or permanent lowering or raising of the groundwater table has been shown to affect buildings, streets and underground utilities adversely. Potential problems include deterioration of wood piles, ground subsidence and negative friction on piles.

To prevent the groundwater levels from fluctuating outside allowable ranges, a low permeability retaining wall system such as slurry walls is required for lateral support of the excavation. The walls are designed to extend into bedrock to reduce lateral movements during excavation and groundwater inflow, and to support vertical bearing loads. (See also “Slurry Wall Design Parameters” by Dave Druss.)


Figure 1: Location of tunnels using slurry walls

Figure 2: Typical subsurface soil and rock profile

Key Design Considerations

Several requirements that impacted the design of SPTC walls for the I-93 cut-and-cover tunnels fell into two categories-geotechnical and other.

Geotechnical. The geotechnical considerations (Figure 2) were that:

  • Permeability in the glacial till and bedrock is relatively high compared to the clay and glaciomarine soils.
  • Depressurization in the glacial till and bedrock outside the excavation had to be controlled to limit consolidation of the compressible clay on which numerous structures are supported.
  • The bedrock profile and quality rock are erratic, and localized areas of high groundwater infiltration could be expected.
  • Cobbles and boulders encountered in the glacial till could impact the installation of earth support systems.

Other. The following right-of-way, construction and alignment concerns had to be considered:

  • Many existing buildings abut the tunnel alignment. These include sensitive older structures, many of historic value, and also large, 40-story skyscrapers.
  • The presence of existing buildings leads to a tight lateral clearance in which to fit the tunnel. The design required stiff walls that needed to fit in a narrow corridor.
  • Underpinning of the existing Central Artery viaduct during the construction of new cut-and-cover tunnel required an excavation support wall system that would bear large vertical loads.

Having taken these key design constraints into consideration together with the cost, schedule, constructibility and environmental impact, we concluded that the optimal tunnel wall system for the Central Artery alignment was SPTC walls that would be included as part of the final tunnel structure. SPTC walls were chosen instead of conventional reinforced concrete slurry walls because SPTC walls are stiffer, more suitable for low headroom construction beneath the existing artery viaduct, and stronger in bending than a concrete wall of the same thickness.

For the CA/T downtown alignment, cross-lot bracing was designed to support the excavation. Tiebacks were not considered for the lateral support system due to right-of-way constraints and concerns that the tieback penetrations might result in leakage in the permanent tunnel wall.

Method of Analysis

It was necessary to consider the construction staging load along with the permanent load condition for the stress and deformation analysis because the SPTC walls are used both as part of support of excavation and as permanent final structure. Construction stages include sequential excavation and strutting with cross lot braces, installation of base slab, sequential removal of struts, installation of roof, and backfilling.

An interesting aspect of the SPTC wall design is that the maximum lateral deformation occurs at a stage when the total lateral load is relatively small compared with its maximum magnitude. This is true because the excavation induced lateral movement will reduce the lateral soil pressure from in-situ at rest value to some magnitude equal to or greater than the active soil pressure, depending on deflection of wall. In the final condition, the soil pressure will increase over time to at-rest pressure. This requires the structural design of the SPTC wall to first consider deformation control at the construction stage when the load is relatively small, and then the strength of the wall for the final permanent condition. For this purpose, the CA/T project adopted a design approach to use an allowable stress design method together with additional stringent deformation control requirements for design of the SPTC wall during temporary construction stage, and the load factor resistance design method for its final permanent condition.

Structural Modeling of SPTC Wall

The structural model of the SPTC wall during the construction phase is a nonlinear and stress-history-dependent soil structural interaction analysis. Prior to tunnel excavation, the SPTC wall is in a state of equilibrium with the surrounding soil. The first stage excavation on one side of the SPTC wall creates different ground elevations on two sides of the wall and mobilizes soil pressure on the retained side to push laterally against the wall. The soil below the bottom of excavation acts as support to maintain the stability of the wall.

A strut is placed at elevation about 0.6 meter (2 feet) above the bottom of excavation and braced against SPTC walls on both sides of excavation. In order to minimize the ground movement, the strut is pre-compressed between the two SPTC walls. The strut acts as a prestressed spring support to the wall with spring constant equal to the axial stiffness of the strut. Preloading of the strut pushes the SPTC wall against the retained soil and generates additional soil reaction on the wall.

For each subsequent excavation stage, the excavation support structure experiences following changes sequentially:

  • Excavation removes the support of subgrade soil to the SPTC wall and lowers the wall’s soil support down to the bottom of the excavation stage.
  • Excavation releases the subgrade soil reaction of the previous stage and causes stress redistribution between the SPTC wall and soil below the bottom of the excavation.
  • Excavation results in additional lateral soil pressure on the SPTC wall.
  • Installation of a support strut adds a support, which is modeled as a linear spring in the analysis.
  • Precompression of the strut may generate additional soil reactions on the SPTC wall by pushing the wall back into the soil.

The final stresses in the SPTC wall are accumulated in increments from one excavation stage to the next. As excavation proceeds, the load on the SPTC wall increases while the structural configuration of the wall/support system as well as its boundary conditions change. The final stresses in the wall are not linearly related to the displacements of the structure, preventing use of the principle of superposition and requiring an incremental nonlinear analysis.

Our structural model assumes for each excavation stage that all loads mobilized by excavation will be applied to the SPTC wall at once. The excavation and installation of struts to support the SPTC wall is a staged process, so lateral wall movement occurs prior to the installation of strut. To model this staged deformation mechanism, the deflection of the SPTC wall at strut level prior to its installation is accounted by introducing a fictitious concentrated load at end of strut to short the strut by same amount. The effects of construction staging are cumulative. As excavation proceeds, the wall tends to move inward. Moments are generated in the wall, which become “locked in” to the structure in the permanent condition.

The wall bending stresses and strut reactions from the incremental analysis are examined at each phase of the excavation to check the capacity of the soldier pile and struts under construction conditions using allowable stress design method. In addition, the final “locked-in” bending stresses are applied with appropriate design factors and combined with results from a final net tunnel model using the additional long-term loadings not included in the incremental analysis. These results are then compared to a final tunnel model using all the loading cases to obtain the maximum design stresses along the wall.


Figure 3: Invert slab to SPTC wall connection

The thickness of the SPTC slurry wall varies between three to four feet thick. The soldier piles are spaced between 1.2 to 1.8 meters (4 to 6 feet). The structural steel soldier piles are designed as primary load carrying members and the concrete between the soldier piles is considered as lagging between the piles. The flexural stress in the concrete is very small, so no steel bar reinforcement is needed. In some locations where the surcharge loading from adjacent building foundations is high, however, reinforcement needed to be provided. In most cases, W36 x 393 wide flange sections with 3.45 x 105 kPa (50 ksi) yield strength are used as soldier piles. The minimum 28 days strength of the tremie concrete is 2.76 x 104 kPa (4000 psi) with a slump between 180 to 255 mm (7 to 10 inches).

The concrete base slab, with thickness that varies between 2.4 to 4.5 meters (8 to 15 feet), is rigidly connected with the steel soldier piles as shown in Figure 3. The roof structures are simply supported by the soldier piles. The roof structures are composite structural steel plate girders with a concrete roof top slab. Waterproofing material is provided under the base slab and on the top of the roof slab.

Conclusion

The downtown Central Artery tunnels used SPTC slurry walls that were designed to serve as support of excavation, water cutoffs, underpinning support for the existing artery viaduct, part of the final tunnel structure and to handle a variety of challenging soil and construction conditions.


PB, as part of the Bechtel/ PB joint venture, is serving as management consultant of the Massachusetts Highway Department’s CA/T project.

[This article was abstracted from a paper entitled “Using Soldier Pile Tremie Concrete Slurry Walls as Permanent Tunnel Walls,” which was published in the proceedings of the 1996 Transportation Research Board conference.]

Other CA/T articles in this issue are by Brenner and Gelfer ("Constructing Slurry Walls Under Low Head Room,", "Should Slurry Walls Be Part Of the Final Structure?"), and Druss ("Slurry Wall Design Parameters").

<< Go To Previous Article | Table Of Contents | Go To Next Article >>
Go To Other Issues | Contact PB Network | Print This Article 
Markets  |  Services  |  About Us  |  People + Careers  |  News + Events  |  Research Library  |  Projects
Worldwide Locations  |  Contact PB  |  Legal Notice  |  Site Map  |  Home
© Parsons Brinckerhoff