The best test for a method is its comparison with field behavior and the results presented are in most cases reasonably accurate. However, some deviations can be explained only be the shortcomings of the model's main assumption. In one instance, for example, the settlements according to the complete analysis deviate further from the measured values than those of the total stress analysis. The assumption of one-dimensional confined compression being the strain condition prevailing within the centrally located core of a dam with stiff shells is valid only for the axis of a symmetrical core and may lead to eroneous results in the case of a nonsymmetrical core, such as Mica Dam. Furthermore, the assumption that total stresses do not change during consolidation under constant load is not valid when construction is not continuous, as was the case with Mica Dam. The discrepancy between measured and computed values in settlements may also be explained by the underestimation of load transfer and arching. The reuslts of a finite element analysis of a model, assuming both linear and nonlinear elasticity and two-dimensional consolidation under conditions of plane strain, show significant variations in total stresses during consolidation. It is felt that comparisons between measured and computed horizontal movements, especially at the interface, could have been useful, as would have been simulation of the development of those minor principal stresses which control the potential for hydraulic fracturing. Another problem is that the governing partial differential equation used to analyze the pore pressure disruption during dam construction is only applicable to saturated soils, while dams generally consist of unsaturated materials. Some assumptions arising from studies on the consolidation of unsaturated soils can be used to predict the consolidation behavior of dams during construction, namely, that the air-water mixture in the pores of a clay compacted wet of optimum water content can be as a homogenized compressible fluid, that the consolidation process of a clay with a lower degree of saturation can be treated as the transient flow of two immisable pore fluids through a compressible medium, and that Darcy's law is applicable to each of the separate fluids. It is felt that more realistic predictions of consolidation behavior would result from the use of an elastic-plactic work hardening model to simulate the effective stress-strain relationship of unsaturated soils with an appropriate effective stress equation, taking into account the variation of soil permeability with degrees of saturation and void ratio.

  • Availability:
  • Supplemental Notes:
    • This paper is a discussion of ASCE Proceeding Paper 13161 by Zdenek Eisenstein and Steven T.C. Law.
  • Corporate Authors:

    American Society of Civil Engineers

    345 East 47th Street
    New York, NY  United States  10017-2398
  • Discussers:
    • Cavounidis, S
  • Publication Date: 1978-7

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00179323
  • Record Type: Publication
  • Report/Paper Numbers: ASCE 13902 Proceeding
  • Files: TRIS
  • Created Date: Sep 14 1978 12:00AM