Published Online: 21 May 2013
Page Count: 8
Professor, Civil and Environmental Engineering, North Carolina A&T State Univ., Greensboro, NC
Professor, Civil Engineering, Univ. of Texas at El Paso, El Paso, TX
Almadhoun, A. Y.
Former Graduate Student, Univ. of Texas at El Paso, El Paso, TX
(Received 7 June 2012; accepted 20 March 2013)
The torsional resonant column is used to measure dynamic properties of soils. For this purpose, a sinusoidal torque is applied on a cylindrical specimen. The response of the specimen is used to derive the shear modulus and damping ratio of the specimen as functions of the induced shearing strain amplitude. The test interpretation assumes perfect coupling between the specimen and the end-platens. The main purpose of the present paper is to document the effects of specimen coupling on the interpreted properties of granular soils tested using several different coupling techniques. The granular specimens were prepared by raining granular soil in six layers into a mold. The soils consisted of angular and rounded sands and spherical glass beads. The coupling techniques included smooth end-plates, porous stone end-plates with and without embedded blades, and smooth end-plates coupled to the specimen with epoxy resin. The performance of the coupling was evaluated based on the relationship of the shear modulus and internal damping ratio with the shearing strain amplitude. The program included one relative density for each soil and one confining pressure. For each set of soil type and coupling technique, five repetitions were performed to illustrate test variability and provide statistical evidence of significant differences. The significance of the observed differences was based on the 95 % confidence interval of the mean response. The results showed significant differences among the interpreted soil properties depending on the coupling technique. The results from epoxy-coupled specimens were used as the reference and in general provided somewhat higher moduli and lower internal damping ratios. Particle size had small effects, but particle shape significantly affected the results, especially at high strain amplitudes. Existing suggestions in the literature of a threshold peak friction that would require special coupling techniques in order to test these soils were found to be excessively large.
Paper ID: GTJ20120104