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
Tips for Slurry Wall Structural Design
By Camille H. Bechara, Boston 617-426-7330
In this abstracted version of the 1995 co-winner for PB Paper of the Year, Camille Bechara writes about general analysis and design of slurry walls. The article includes a case study and specific analysis and design suggestions.

Slurry walls offer several advantages over conventional retaining wall design, particularly in congested urban areas, in that they can:

  • Resist lateral earth and water pressures and earthquake forces
  • Behave as load-bearing elements
  • Serve as the temporary as well as the permanent structural wall for a tunnel or underground building, providing a finished concrete surface.

Fundamentals of Slurry Wall Design

Design of a slurry wall is greatly dependent on the method of analysis used and requires close communication between structural and geotechnical engineers. The three primary analysis methods are:

  • Equivalent Beam on Rigid Support (Rigid Method). The point of zero pressure is computed below the excavation cut (passive pressure set equal to active pressure), and acts as a fictitious support. As excavation proceeds, the wall spans as an elastic beam between rigid supports (provided by bracing members or floor slabs) with the lowest support at the point of zero pressure below the subgrade. This method may be conservative, but provides an easy structural solution. The computed wall deflections from the elastic analysis do not, however, correlate to predicted ground movements.
  • Beam on Elastic Foundation (Winkler Method). The passive side of the wall is modeled as a series of “springs” based on the modulus of subgrade reaction of the soil. The elastic properties of the soil are required in order to calculate the soil modulus. Analysis is carried out with general structural computer software to determine moments and shears. This method allows complete modeling of the wall based on soil springs, brace or floor springs (K-values based on stiffness and volume change). It seems to yield smaller forces than the Rigid Method and also gives a better indication of wall movement. Given the various unknowns inherent in soil properties, actual versus assumed loads, and other such factors, a cautious elastic approach to analysis should be exercised.
  • Finite Element Method (FEM). This method requires advanced computer software and modeling, and an in-depth knowledge of soil properties that are modeled in the program and construction sequences that are simulated. Its greatest benefit is better prediction of soil movement. Its use for structural design is limited and reliance on other methods is necessary.

I suggest using these three methods as follows:

  • The Rigid Method merits consideration as a first choice or as a preliminary tool due to its simplicity.
  • The Winkler Method offers an attractive solution in that it gives a more realistic representation of wall behavior. It could be used next as a check or for design. Then, engineering judgment should be used on a case by case basis taking into account the degree of conservatism desired.
  • FEM is a powerful tool that enables designers to analyze many construction stages and temporary excavation
    conditions. Its use is recommended as a supplement and where adjacent construction may be sensitive to wall movements and a better prediction of these is required.

Regardless of the method used, it is essential that close coordination and communication take place between geo-technical and structural engineers early in the design process, and a consensus and understanding be reached. This must be followed by coordination meetings with the contractor to minimize problems during construction.

Slurry Wall Penetration

The depth of wall penetration below subgrade or dredge level is governed typically by stability although water cutoff and bearing must also be considered. The forces pushing into the excavation must be balanced by resistance provided by the subgrade penetration and bracing. In most cases, the design depth is controlled by the loading case in which excavation is at its final stage, just before installation of the lowest slab or bracing member. For this calculation, it is typically assumed that the wall below this level does not contribute its own stiffness. Because the wall is free to rotate about its lower end, a safety factor with respect to passive resistance must be used in the design. A value of 1.5 is recommended typically.


Figure 1: Slurry Wall Penetration

The required penetration (d) is determined by equating to zero the sum of the moments on the active side and passive side about the brace point or floor level above (Figure 1). This condition yields a cubic equation in which d is easily solved by substituting trial values. The embedment is often arbitrarily increased by 20 percent.

Some engineers may want to take advantage of the additional bending resistance provided by the wall at the brace elevation and/or the moment capacity of the brace point at the connection to the wall. This is an engineering judgment that should be used on a case by case basis, depending on the degree of protection provided by other systems utilized, and several factors including overall degree of conservatism in selection of soil parameters, method of analysis, accuracy of loads and ground water pressure, existing adjacent foundations, etc. The same rationale should also be used in determining the magnitude and applicability of safety factors that must be incorporated in the design.

Another interesting calculation to perform is to consider the effect of base friction on the wall penetration, taking into account the coefficient of friction between concrete and the rock or bearing material. Here, too, an engineering judgment must be made as to whether or not to rely on this information.

A Case Study: North Station Parking Garage

Boston’s North Station Underground Garage illustrates some of the current methodology used in the design of slurry walls. The construction of this 6-level underground structure was at the site of the new Shawmut (now Fleet) Center sports arena. It included 0.9-m (3-foot) thick reinforced concrete slurry walls 18 to 24 m (60 to 80 feet) deep, embedded in bedrock. The wall panels vary in length from 2.6 to 7.6 m (8.5 to 25 feet). The “top-down” method of construction was selected because of the large footprint area of the garage, the scheduled arena construction above, and the extremely tight physical constraints of the site, which was surrounded by the existing Boston Garden, the underground Orange Line subway tunnel, the Federal Administration Building (GSA), and the Central Artery Ramp structure.

First, the slurry walls were built around the site. These walls acted as horizontal water cut-off and also became the permanent wall of the structure. Interior steel columns were installed from grade in a slurry trench technique (filled with tremie concrete at the base) to form supporting caissons. The ground floor level was then built around the columns, which support it. The completed slab braced the slurry wall while the garage was mined below it. The ground floor slab braced the excavation, eliminating the need for conventional internal bracing or tiebacks. This step was repeated for the lower levels.

When designing the slurry wall, we analyzed it for all stages of construction because it was to be both a temporary and permanent excavation support element.

  • During Construction. Active lateral pressures were assumed and a triangular pressure distribution was used. The passive pressure was computed at various excavation levels and the point of “zero pressure” was used to compute the reaction below excavation level (Rigid Method). This was checked later using the Winkler Method. A finite element analysis was also performed, and the wall analyzed based on predicted slurry wall deformations. The calculated bending moments were lower than the ones obtained from the other analyses.
  • End of Construction. Prior to casting the lowest base slab (six levels down), active lateral pressures were assumed, but based on a rectangular distribution. The pressures were used to determine slurry wall penetration and stability, which was obtained by equating to zero the active and passive moments about the level above.
  • Long Term (Permanent Loading). At rest, lateral pressures (triangular distribution) were used for the long-term design of the slurry wall

Figure 2: Reinforced Slurry Wall Section

Figure 3: A section of the stirrup reinforcing detail.

The slurry wall was designed for soil, groundwater and surcharge pressures in all three stages, and for seismic pressures in the long-term stage. A safety factor of 1.5 was applied to passive soil pressures. To obtain a comprehensive bending moment envelope along the height of the wall, the support reactions provided by the floor slabs at each stage were considered “rigid supports” (zero deformation) as well as “flexible springs,” taking into account the elastic shortening of the slab and volume change deformations resulting from creep, shrinkage, and temperature variations.

During construction, an extensive monitoring program was performed that included observation wells, inclinometers, and settlement measurements. This program was necessary to verify that wall movements were similar to those assumed during design.

Theory Versus Practice

What does all this mean to structural design? Engineers often fall back to the ACI 318 concrete building code and design a slurry wall for flexuq`re and shear that are the same as for walls cast above ground. There is a major difference, however, in that slurry walls are constructed somewhat “in the blind” without the convenience inherent with exposed construction as far as placing and curing concrete, controlling form dimensions, and placing reinforcement. For these reasons, a moreconservative approach to design and detailing is warranted, including making allowances for anticipated construction tolerances and keeping in mind that the finished product is not as shown by the neat straight lines on the contract drawings (Figure 2).

Some specifics of flexural and shear design are shown in the accompanying box. Crack control and wall/soil deflection/ displacement must also be considered, however, in a typical design.

Flexural Design

(57kb)

If you do not have Adobe Reader please download it from here.

Download Adobe Acrobat Reader

Conclusion

In summary, flexural and shear design for slurry walls must take into account not only the structural integrity and code requirements, but also:

  • The detailing of the contract documents
  • Construction practices in the area
  • The increased tolerances associated with this type of construction.

The recommendations presented in this article are intended to facilitate structural design, bring awareness to the differences between design and construction, and provide a more durable concrete wall that will exhibit fewer cracks and leaks and help reduce the long-term maintenance that may be required.

As structural engineers, we must acknowledge that the ultimate performance of the slurry wall is more significantly affected by construction techniques and site-specific subsurface conditions than it is by the structural details and code interpretations.


For copies of the original, complete paper, please call Camille Bechara at 617-426-7330.

<< 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