Measurement of small-strain shear modulus G_max of dry and saturated sands by bender element and resonant column tests

The bender element (BE) and resonant column (RC) tests on two clean sands with various densities and mean effective stresses under dry and saturated conditions are performed. Based on the test results, a reliable method for determining the travel time in BE tests is evaluated. Also, the deviation of the shear wave velocity between BE and RC tests is explained. The studies on the effect of different excitation signals on shear-wave velocity indicate that the shear-wave velocity by the Start-Start method shows a slight increasing trend with an increase in the excitation frequency. The shear-wave velocity determined by the Start-Start method under the excitation frequency of 10 kHz is consistent with that from the RC tests. The shear-wave velocity values from the BE tests are about 6%~10% greater than those of the RC tests. The differences can be explained by the dispersion of shear wave under saturated conditions. This study suggests that the dispersion of shear wave should be considered in saturated clean sand because the excitation frequency of BE tests is greater than the characteristic one. However, the dispersion of shear wave should not be considered in the saturated natural sand or silt for the reason that the excitation frequency of BE tests is lower than the characteristic one. The results of this study provide effective methods for the BE tests on clean and natural sands under saturated conditions.