Variations of Shear Wave Arrival Time in Unconfined Soil Specimens Measured with Bender Elements

Bender element test setups have gained much popularity in the measurement of shear wave velocity (vs) in soil specimens, with the purpose of estimating the small strain shear modulus (Go). However the determination of shear wave arrival time from bender element tests can be subjective with results varying over a wide range, depending on the method adopted to identify the arrival time. This paper describes a series of bender element tests conducted on a pair of unconfined specimens, 38 mm in diameter and 76 mm in height, where the average data of the two are reported. With shear waves triggered at frequencies between 1 and 20 kHz, identification of the arrival time in both the time and frequency domains were performed. The different methods presented varying degrees of problems and discrepancies, with no one method emerging as a consistent winner. The time domain methods were apparently preferable due to its simplicity, which is perhaps one of the key factors contributing to the growing popularity of bender elements. The frequency domain methods, on the other hand, involved complex manipulation of the original signals, which can be onerous and time-consuming. Based on the findings, it was concluded that the reliability of shear wave velocity measurement with bender elements can be increased and the errors kept to a minimum, if the same arrival time identification method is performed with consistent judgment in a particular test series.