LAUSR

Abstract Due to poor understanding of the output signal, the reliability of the travel time estimated by bender element tests has been their most serious problem since their introduction to the soil testing. Recent establishment of a concept interpreting the bender element apparatus as a linear system enables us to calculate and probe the output signals more theoretically. In this study, first, the concept is reviewed to understand the linearity of the whole system, which is composed of some fundamental subsystems, reflecting the responses of the soil specimen, and the source/receiver bender elements including their peripherals. For modeling the response of the whole system, simplified models for the subsystems are employed and combined, with some updates to describe the interaction among the subsystems, by sharing some parameters throughout the models. With a set of hypothetical, but expected parameter values, the responses of each subsystem and the combined whole system are simulated in the time and frequency domains, and the resulting travel times are thoroughly examined. The validity of this model is demonstrated by comparing the system responses between the simulation and the experimental data previously conducted in the author’s laboratory. A potential that removing the response of the bender element subsystems from the whole response could dramatically improve the accuracy of travel time estimation in actual tests is concluded.