Factors in the Variability of Measured Surface Settlements over EPB Driven Tunnels in Soft Clay

INTRODUCTION Soft, near normally consolidated clays generally provide ideal conditions for tunneling using Earth Pressure Balance (EPB) shields. It is easy to turn the clay into a low permeability, viscous spoil that can be pressurized to maintain the required face pres-sure. Despite this, there are records of relatively large, and often quite variable, surface settlements over EPB driven tunnels in soft clay. To understand this variation it is necessary to identify the various factors that contribute to settlement. Tunnels in soft clay are ideal for study these effects in the field, as an unsupported void will close almost immediately, due to the low shear strength of the soil. In the short term, deformation in a soft clay will occur at constant volume, so short term surface settlements can be related to movements at tunnel level. In this paper, data will be used from EPB driven tunnels in Singapore marine clay. In all of the cases discussed the lining consisted of precast concreted segmental rings erected in the tail of the shield. The marine clay in Singapore has been studied, among others, by Tanaka et al.(2001). The clay typically has a plasticity index of 40 to 60, and the undrained shear strength (SU ) can be derived from: SU = 0.22 svo' where svo' is the current vertical effective stress. Figure 1 shows a summary of undrained strength/depth plots used for design on two sections of Singapore’s North East Line. The stability of a tunnel in clay can be assessed by comparing the stability number (N) of the tunnel with the stability number at total collapse (NTC). The stability number can be calculated from the equation: N = (?Z + q – Pt)/SU where: ? = the average unit weight of the ground above the springline of the tunnel Z = the depth to the springline of the tunnel from the ground surface q = the surcharge Pt = the support pressure within the tunnel The stability number at total collapse depends on the size, geometry and depth of the tunnel, and typically ranges between 5 and 9. Charts for assessing the number at total collapse are provided in Kimura and Mair (1981), based on the testing of model tunnels in a geotechnical centrifuge. For tunnels in the Singapore marine clay, the stability number of an unsupported tunnel would be over nine if the tunnel ois more than 8m deep. As a result, a support pressure