RI 7831 Statistical Comparison of the Pulse and Resonance Methods for Determining Elastic Moduli

Elastic wave velocities and moduli were determined by both the pulse and resonance methods in a large number of specimens of St. Cloud Gray Granodiorite and Tennessee marble under the same moisture, temperature, and stress environment. Long cylinders were first tested by the resonance method to obtain longitudinal bar and torsional wave velocity. The cylinders were then sectioned in half and tested by the pulse method to obtain the transient pulse velocity. These velocities were later used to obtain various moduli for comparison. Statistical comparisons between the Young's, shear, and bulk moduli and Poisson's ratios obtained independently by the pulse and resonance methods indicate that the methods do not give equivalent results in nearly isotropic rock. The amount of difference varies for each modulus, with the least difference (less than 5 pct) occurring in the shear modulus and the greatest difference (as high as 34 pct) in the bulk modulus. Moduli determinations in aluminum show much closer agreement between the pulse and resonance methods than in rock. The discrepancy in pulse and resonance results probably is caused by frequency dispersion effects and by differences in error sensitivity in the various formulas used to calculate the moduli.