Input parameters for FEM analyses and G-GAM curves

The most elementary way of using DMT in a FEM analysis is to assume the soil as linearly elastic, using as an operative Young’s modulus E by DMT. Hamza (1995) reports using E from DMT in Cairo Metro design (with E ≈ 0.8 M_{DMT}).

A more elaborate way is to use the parameters provided by DMT (e.g. M_{DMT}, E, OCR, Ko) to get an estimate of the parameters required by FEM.

Considering the DMT recognized ability to generally well predict settlements, DMT can be used to approximately verify the set of FEM parameters that the designer intends to use. To do this, using the parameters estimated by a DMT sounding, the designer estimates the settlements of a simple shallow foundation (e.g. a 5 m diameter tank applying a uniform load of 100kPa). Then he predicts settlement using the set of parameters he has intention to use in his FEM program. If the settlement predictions are similar, then the two predictions support each other. If the settlement differences are high, then additional attention to the parameters is necessary. In practice the DMT settlements prediction may serve as a FEM-confirmatory virtual load test.

If the G-gamma decay curves have been estimated, based on the small strain modulus and the working strain Modulus determined by the Seismic DMT (Amoroso, 2014), then estimates of intermediate moduli is possible based on the expected strain.

Input parameters for Plaxis (Hard Soil model “HS”). The basic input for this model is E_{50}^{ref}. In an intensive survey Schanz (Schanz 1997, cited in Plaxis bulletin 2003) found that in loose to very dense quartz sand E_{50}^{ref }is generally in the range E_{50}^{ref }= 15 to 75 MPa. Schanz also pointed out that E_{50}^{ref }is correlated to the constrained modulus. Experience in many quartz sands has shown that, in quartz sands, M_{DMT}too is, in general, in the same range (15-75 MPa). Hence, as a first approximation, one may adopt E_{50}^{ref}≈ M_{DMT.}

The Subgrade Reaction Method (soil modulus by springs) is widely used in current practice for design of multi-propped walls. Of course the critical step is the selection of an appropriate value of the coefficient of subgrade reaction K_{h}. In a 2004 paper Monaco et al. investigate the influence of various factors on K_{h}(soil and wall stiffness, excavation depth, prop spacing). The possible K_{h}-DMT relationship is investigated based on the results of FEM analyses carried out using the non linear Plaxis Hardening Soil model. A tentative correlation between K_{h}and M_{DMT}is proposed, having the form K_{h}= M_{DMT}/ B. The paper also gives broad indications for selecting the values of B – hence K_{h}.