Numerical Modeling of the Casagrande Cup Test with Newmark's Method

Abstract

Numerical modeling of Casagrande cup test can be conducted by applying Newmark’s method in a finite element program. Casagrande is a test that applies acceleration as a load to minislope that is molded in a cup until the slope fails. Modeling of the test cannot be performed without scaling because finite element program must have its limit in mesh size. Thus, a series of scaling up and down needs to be carried out to match the parameter for obtaining a realistic result. The test is done in a series of water contents to gain information about the change in behavior of the slope, which is then validated by the result of a physical test, on which the input for the modeling data is gained. By doing so, the difference in real test and modeling result can be equally compared. This study aims to propose a method of modeling the Casagrande cup test, which includes the scaling procedure and the reduction in Young’s modulus to obtain the desired result. In conclusion, reduction in Young’s modulus of the base plate is necessary to ensure an accurate result of the model.Comparison of the result of numerical modeling and that of the real Casagrande test presents a small error, which is still acceptable because the Casagrande test has high repeatability between tests even with the same sample. A proper method of finding the exact value of other properties followed by proper scaling is important to obtain the desired soil behavior.