Placement Rates for Highway Embankments with Vertical and Horizontal Drainage : Informational Report

Abstract

The objectives of this study were (a) to elucidate the practical and theoretical bases for using the controlled rate of construction technique to design a highway embankment underlain by soft ground, and (b) to synthesize presently available procedures in a comprehensive computer program in which special attention is given to the horizontal and vertical drainage, without sand drain installations. An existing program SAND which considers sand drains has been modified for this purpose. To facilitate the mathematical treatment, the overall problem was divided into four parts, which deal with the initial increase in excess pore water pressures caused by an increase of the vertical load on the surface of the compressible layer, the process whereby these pore water pressures are dissipated with time, the associated settlements, and the stability of the embankment-foundation system. With regard to the computation of the initial excess pore water pressure distribution, the following conclusions can be drawn: (1) the form of the stress equations requires the numerical integration of oscillating integrands, and convergence of the extended Simpson's rule or Filon's formulae with interval halving depends on the geometry of the problem. Poorest convergence was obtained in cases of heavily oscillating integrands when the ratio of the load width to the thickness of the compressible layer was large; (2) when the pore pressure coefficient B is held constant and equal to unity, the influence of the pore pressure coefficient A increases as the thickness of the compressible layer increases, and the average pore water pressures are larger and extend farther in the horizontal direction when A is larger; (3) as the compressible layer becomes thinner relative to the load width, closer agreement is obtained between the applied vertical load and the resulting average pore pressure distribution; and (4) the influence of shear stresses causes some concentration of average pore water pressures near the edges of the load. With regard to the computation of primary consolidation settlements, direct proportionality between the average degree of consolidation and the resulting settlement will occur only when constant coefficients of consolidation and a constant coefficient of compressibility are used. With regard to the stability analyses, the following conclusions can be drawn from a critical comparison of the charts in the report: (1) depending on the geometry of the embankment and the soil parameters of the embankment and the subsoil, the assumption of a circular slip surface will give reliable factors of safety only for sufficiently large subsoil thicknesses; (2) the stabilizing influence of flattening the embankment slope decreases as the thickness of the subsoil increases; (3) the slip circle resulting in a minimum factor of safety generally tends to penetrate the soft subsoil as deep as possible; and (4) the factor of safety is not proportional to the height of the embankment, but, given identical soil parameters, it depends on the ratio of the embankment height and the thickness of the compressible layer.