Effects of confining pressure and degree of saturation on wave velocities of soils

Measuring wave velocities of soils is a common methodology of determining the small-strain stiffness of soils. Both confining pressure and degree of saturation affect the magnitude of the wave velocities and hence the stiffness of the soil. However, confining pressure and degree of saturation affect P- and S-wave velocities of the soil differently. At full saturation, increases in effective confining pressure (σ₃') increase S-wave velocity, but P-wave velocity remains fairly constant. As the soil desaturates, the effects of net confining pressure (σ₃ - u_a) and matric suction (u_a - u_w) were observed to be different from the effect of effective confining pressure. This difference in response was investigated, and the results are reported here. Degree of saturation has been commonly associated with the Skempton's pore-water pressure parameter B under an isotropic state of stress. However, some soils have a B-value less than 0.9 even at full saturation. In such cases, the P-wave velocity may be a better indication of full saturation because P-wave velocity increases rapidly from 90 to 100% degree of saturation. The objectives of this study were to investigate (1) the effect of matric suction on P- and S-wave velocities of soil and (2) the relationship between degree of saturation, B-value, and P-wave velocity. Experiments were conducted on sand using a triaxial cell modified to include bender elements and a high-air-entry disk for the testing of unsaturated soils. The P- and S-wave velocities and B-values at various net confining pressures and matric suctions were measured. The relationship between P-wave velocity and degree of saturation was found to be dependent on the saturation path (drying or wetting). The intensities of the high-frequency components (more than 20 kHz) greatly increased in the P-wave signals when the specimen approached full saturation. This observation shows that P-wave velocity and the intensity of its frequency components could be a better indicator of full saturation compared to Skempton's pore-water pressure parameter B. The P-wave velocity decreased rapidly when the matric suction exceeded the air-entry value. Test results show that P-wave velocity was not affected by changes in effective confining pressure or net confining pressure. The changes in S-wave velocity resulting from an increase in matric suction were not proportional to the changes resulting from an equal increase in net confining pressure. Changes in S-wave velocity resulting from changes in matric suction were generally less consistent and varied from soil to soil.