ABSTRACT
In the course of the past decade, a large number of small-scale tests have been carried out in the small geotechnical centrifuge of the University of Delft, on a variety of geotechnical problems. In the centrifuge testing technique, the soil samples are subjected to stress conditions similar to those experienced at full scale. This guarantees a more realistic behavior of the small models. In model tests complicated problems can be analyzed and new ideas can be tested easily. An advantage of model tests is that the behavior of a structure can be monitored much more accurately than in a full-scale test. A small size centrifuge allows several tests to be performed in quick succession. This is particularly effective for the study of failure mechanisms and the influence of changes in design. Furthermore, it is possible to prepare small soil samples with precision, so that the effect of slight differences in design can be made visible. This has facilitated the observation of several unexpected phenomena, which are not yet predicted by mathematical models. Aspects of the following subjects will be discussed in this paper: behavior of conical footings; bearing capacity of suction piles; widening of road embankments; cratering during gas blowout; slope stability and failure of sand embankments during water infiltration.
