Loading Systems, Sample Preparation, and Liquefaction

Abstract

The liquefaction behavior of a clean, fine, uniform sand designated F‐70 Ottawa banding sand is examined. Specifically, an investigation into the effects of loading systems and sample preparation methods on the stress‐strain behavior is made. A comparison between the results obtained with a dead‐load apparatus and a hydraulic loading system on samples prepared by dry pluviation, moist tamping, and moist vibration indicates differences in the soil's response during undrained shear. The former loading system generates and maintains over a greater strain range lower residual deviator stresses and peak pore‐water pressures during steady‐state flow. Furthermore, the steady‐state velocities of the samples tested in the dead‐load apparatus are two‐three times higher than the velocities of those loaded hydraulically. These differences can be attributed to the dead‐load system's ability to provide a constant driving force fast enough to achieve a sample's maximum rate of strain once liquefaction commences. The relative position of the steady‐state curves is unaffected by the type of loading system. However, the method of sample preparation dictates their position. Samples prepared by dry pluviation exhibit residual effective minor principal stresses that are approximately one‐quarter to one‐half the value of samples prepared by either moist method.