6.7 Tunneling

Deformation due to tunneling

The first centrifuge model test on 2-D tunneling was carried out at University of Cambridge, followed by the test on 3-D behaviour of tunnel heading. One of the main interest of these tests were settlement profile at ground surface above the tunnel (Fig. 2).

Kimura, T. and Mair, R.J. (1981) : Centrifugal testing of model tunnels in soft clay, Proc. of 10th ICSMFE, Vol. 1, pp. 319-322.

Influence of shield tunnelling on pile foundation

The influence of shield tunnelling on pile foundations has been studied by a series of centrifuge tests. As most of the buildings and houses in the western part of the Netherlands are founded on long, end-bearing piles with the tip placed in dense sand, shield tunnelling in the neighbourhood of such foundations may lead to problems (Fig. 3). As shield tunnelling always leads to some volume losses, the stress situation and density of the sand around the tunnel is changed. This mostly leads to a reduced bearing capacity of the piles and possible settlement of piles. Therefore, at Delft Geotechnics centrifuge tests have been performed with instrumented piles, placed in dense sand overlain by soft clay (Photo 1). Near the sand/clay interface a tunnel was placed, of which the diameter could be reduced, thereby modelling volume losses. Three tests have been performed with various depths of the tunnel with respect to the sand/clay interface, but fixed depths of the pile tips and distances from the tunnel. The results are confidential, but they have lead to an improved quantitative understandings of the behaviour of foundations pile adjacent to shield tunnels.

Bezuijen, A. and van der Schrier, J. (1994) : The influence of a bored tunnel on pile foundations, Proc. of CENTRIFUGE 94, pp. 681-686.

Face stability of a slurry shield tunnel boring machine

The face of a slurry shield tunnel boring machine is stabilized by a bentonite slurry, which is kept under pressure. On behalf of the Centre for Underground Construction, the Netherlands, a prediction for the minimum required and maximum allowed slurry pressures has been made for the construction of Second Heinenoord tunnel by means of centrifuge tests. If the support pressures are too low (active failure), instatability of the face occurs and large settlements may be expected. If the pressures are too high (passive failure), loss of bentonite and large soil disturbance will occur. For this test series a soil model, consisting of three sand layers with varying densities has been made with the dynamic fluidisation compaction technique. For the tests, conducted at 70 G, a mini tunnel face has been constructed with a diameter corresponding with the prototype diameter of 8.55 m. At the front a soft membrane was placed (Photo 2). Behind the membrane a piston was present, which stabilized the face during spinning up. The chamber was filled with a support fluid. After spinning up, the pressures in the support fluid were increased until the piston became free. Subsequently, in the first test the support pressures were lowered to simulate active failure and in the second test increased to simulate passive failure. During the test face deformations and ground surface deformations were recorded. After the active test, the sand was made capillary and face deformations were made visible with the aid of coloured layers, which has been placed in the sand during preparation (Photo 3). The tests have been back-calculated with finite element analysis. The test results have been used to validate the finite element model.