Sands under Cyclic Loading

Silver, M. L., Seed, H. B. (1971a). Deformation characteristics of sands under cyclic loading. 

The effects of cyclic loading are extremely dependent upon the soil and nature of cyclic deformation as shown in the previous section. The potential for pore water pressure generation either in clays, silts, or sands governs their behavior under cyclic loading. Two analytical approaches have been developed to model this behavior. 

At a given relative density or (NUeo, sandy soils containing significant amount of fines (particle size <0.075 mm) are known to have higher liquefactitMi resistance than clean sands with no fines (Youd et al. 2001). Nonplastic silts, containing 100 % fines but with very low plasticity (such as plasticity index PI < 5 %), are considered to behave under cyclic loading in a similar manner as sands. The liquefaction resistance of nonplastic silts is evaluated as sands with upward correction on fines content. 

Figure 21.5 demonstrates the excess pore pressure ratio (tu) generations in medium dense sand specimens with different degrees of saturation under the shear strain record shown in Fig. 21.4c. The results reveal that during cyclic loading, excess pore pressures also generate in partially saturated sands and can remain high as the degree of saturation increases. However, as the degree of saturation reduces, maximum excess pore pressure (rumax) lower, while at the same time the number of cycles required to reach rumax (Nmax) gets higher.

CF tests may be carried out in an apparatus designed by the Continental Oil Company. The apparatus consists of a Monel tank with four samples subjected to cyclic bending. The first step consists of determining the displacement caused by the applied load. The exact stresses are determined by strain gauges. The electrolyte is deaerated with 3% sodium chloride. Polished or sand blasted samples are used, and the behavior of the alloy in CF may be studied at the free corrosion potential under different percentages of stress amplitude of the elastic limit. From potentiokinetic curves, I =J E) the protection or pitting potential applied during the stress test. Each test can have four samples and the difference between results (85) for similar tests does not exceed 15%. 

---