Abstract: Two centrifuge model tests on foundation pit engineering with cantilever retaining wall in sand and silty clay ground are carried out respectively. Test procedures and some feasible solutions to the key problems in the tests are introduced, and results of the two tests are listed and compared in this paper. It can be summarized that the preparation of unsaturated soil ground is difficult because it is hard to control parameters of remolded soil. Therefore, stratified compaction method needs to be improved. In contrast, sand ground prepared by sand-rain method is easier to control parameters.Evolution of soil pressure, foundation deformation, and bending moment in the sand ground test is better than that in the silty clay ground test. Therefore, geotechnical centrifuge model tests in unsaturated soil ground can be replaced by the tests in sand ground appropriately. In foundation pit engineering with cantilever supporting structures, the induced surface settlement curve is an exponential pattern in sand ground but a linear pattern in silty clay ground according to test results, and the induced ground deformation range is larger in silty ground than that in sand ground. The induced bending moment of the retaining wall is larger in sand ground than that in silty ground, and the positions of the maximum bending moment for both grounds move down during the excavations. In sand ground, soil excavation causes decrease of the earth pressure behind the retaining wall, while in silty clay ground, there is an increase at the bottom of the wall. The complete test procedures and data processing methods presented in this study provide technical premises and guidance for further tests.

Key words: centrifuge model tests, foundation pit engineering, sand ground, silty clay ground, cantilever retaining structure