• Proceedings of the Korean Geotechical Society Conference
• /
• 2005.03a
• /
• pp.627-632
• /
• 2005

In this study, a series of large-scale oedometer tests was performed to investigate the compressibility of rock fill materials. The testing samples were prepared to have three different grain size distributions and for each distribution, exist in two different states(dried and saturated). The test results indicated that particle breakages occurred mainly for the particles larger than 4.75mm in size and increased with increasing grain sizes. Also, it was found that, for a dry sample as it became well-graged, its compressibility decreased and accordingly, its tangent constrained modulus increased. A comparion between the samples in dry and saturated states revealed that compressibility of the materials increases with increasing water content. The values of tangent constrained modulus calculated for the tested dry samples were larger by about 10 to 20%, on average, than those for the saturated samples.

• Geomechanics and Engineering
• /
• v.25 no.4
• /
• pp.341-345
• /
• 2021

1D sand compression response to ko-loading experiences volume contraction from low to high effective stress regimes. Previous study suggested compressibility model with physically correct asymptotic void ratios at low and high stress levels and examined only for both remolded clays and natural clays. This study extends the validity of Enhanced Terzaghi model for different sand types complied from 1D compression data. The model involved with four parameters can adequately fit 1D sand compression data for a wide stress range. The low stress obtained from fitting parameters helps to identify the initial fabric conditions. In addition, strong correlation between compressibility and the void ratio at low stress facilitates determination of self-consistent fitting parameters. The computed tangent constrained modulus can capture monotonic stiffening effect induced by an increase in effective stress. The magnitude of tangent stiffness during large strain test should not be associated with small strain stiffness values. The use of a single continuous function to capture 1D stress-strain sand response to ko-loading can improve numerical efficiency and systematically quantify the yield stress instead of ad hoc methods.