Proceedings of the 17 th International Conference on Soil Mechanics and Geotechnical Engineering M. Hamza et al. (Eds.) © 2009 IOS Press. doi:10.3233/978-1-60750-031-5-413 413 Use of vertical reinforcement to reduce ground movement due to tunnelling Usage de renforcement vertical réduire le mouvement moulu dû à percement de tunnels A. Juneja and A. Hegde Dept. of Civil Engineering, Indian Institute of Technology Bombay, Mumbai 400076 India F.H. Lee Department of Civil Engineering, National University of Singapore, 117576 Singapore ABSTRACT This paper examines centrifuge results of tunnel excavation adjacent to vertical reinforcement in small-scale sand beds. The tunnel walls were represented by pressurised rubber liner. Results of 6 tests conducted at 50g using large diameter centrifuge at IITBombay are reported. Depth of the tunnel and spacing between the tunnel axis and reinforcement are the variables investigated. To simulate tunnel excavation, hydraulic pressure on walls of the rubber liner was decreased in steps whilst the centrifuge was in-flight. The magnitude and extent of settlement trough with increase in ground loss was captured using potentiometers and pore pressure transducers placed in vicinity of the tunnel axis. From the limited test observations, vertical rods of not more than 14-times the particle size is shown to affect the width of settlement trough. RÉSUMÉ Ce papier examine des résultats de la centrifugeuse d'excavation de tunnel adjacent à renforcement vertical dans les lits du sable peu importants. Les murs de tunnel ont été représentés par paquebot du caoutchouc pressurisé. Résultats de 6 épreuves menés à 50g utilisant grande centrifugeuse du diamètre à IITBombay sont rapportés. Profondeur du tunnel et espacer entre l'axe de tunnel et le renforcement est les variables enquêtées sur. Pour simuler l'excavation de tunnel, la pression hydraulique sur murs du paquebot du caoutchouc a été diminuée dans whilst des pas la centrifugeuse était dans-vol. La magnitude et ampleur de dépression du règlement avec augmentation dans perte moulue ont été capturées utiliser potentiomètres et transducteurs de la pression du pore placé dans voisinage de l'axe de tunnel. Des observations de l'épreuve limitées, tringles verticales de pas plus de 14 fois que la dimension de la particule est montrée pour affecter la largeur de dépression du règlement. Keywords : centrifuge modelling, sand tests, tunnel, vertical reinforcement 1 INTRODUCTION Preventing excessive ground movement due to tunnel excavation is an important engineering design problem. Many reinforcement techniques, such as, jet grouting, steel pipe umbrella and face bolting are widely used to reduce the inward soil movement during the tunnelling. The use of more than one of the above techniques is not uncommon for a particular site. Although much of the research has been directed towards the understanding of the mechanism and prediction of soil movement using the above methods (e.g. Haruyama et al. 2001, Jacobsz et al. 2005), many of these methods are applicable only to tunnelling in deep soil deposits. Surface settlement and consequently the effect on structures relatively close to the tunnel becomes significant if tunnelling is done in shallow ground. Under these circumstances, vertical reinforcement from the ground surface down to the tunnel level can be useful. This pre-tunnelling vertical support system is known to reduce the volume of the settlement trough at the surface and also increase the face stability. Kamata and Mashimo (2003) conducted centrifuge model studies using 80mm diameter tunnel at 25-and 30g in dry Toyoura sand. This corresponds to 2- and 2.4m diameter tunnel at prototype scale. Vertical reinforcement using 1mm diameter phosphor bronze bolts coated with sand were embedded from the surface up to the tunnel face. In all the tests, the reinforcements installed up to the tunnel invert were shown to significantly increase the face stability during the tunnel excavation. Bilotta et al. (2006) investigated the effect of reinforcement installed along the length of the tunnel. A 50mm diameter model tunnel (8m in prototype scale at 160g) was installed in about 160mm thick kaolin clay bed. Aluminium piles of 1.6mm diameter and 120mm long were used as reinforcement which corresponds to 256mm and 19.2m, respectively, at prototype scale. Excavation of the tunnel was simulated by reducing the air pressure inside the tunnel cavity. The results showed that pile spacing of about 3 times the reinforcement diameter significantly reduced the settlement trough at the surface. The width of the trough also altered at these pile spacing. The purpose of this paper is to examine the influence of vertical reinforcement on the settlement trough due to tunnelling in sands using centrifuge tests conducted at the Indian Institute of Technology Bombay (IITB) Geotechnical Centrifuge. All dimensions in the figures are in mm unless and otherwise stated. 2 EXPERIMENTAL SETUP A series of nine centrifuge model tests were conducted at 50g environment to investigate the effect of piles on ground deformation upon tunnel collapse in sand beds. The soil cover and relative density of the sand were the variables in the tests. The model tunnel was about 40mm in diameter which correspond to 2m in prototype scale. The purpose for choosing this diameter was to not model any specific tunnel but to observe the effect of tunnel depth on the extent of ground deformation. Model sand beds were prepared in centrifuge strongbox of internal dimensions 720mm x 450mm x 410mm which had its front wall made of transparent perspex sheet. This large sized box was taken primarily because of its