Pullout of Steel Grids in Dense Sand: Experiments and Design Insights

Abstract

Retaining walls mechanically stabilized with reinforcement frequently have proven their resilience while maintaining an advantage over conventional (gravity and cantilever) walls in terms of cost effectiveness and environmental impact. To achieve optimized design beyond the inherently conservative code specifications, reliable pullout models are needed. The paper enriches the existing literature with results from a series of laboratory pullout tests involving inextensible steel grid reinforcement (in the form of bar mat) embedded in dense sand. Comparative tests were used to determine the role in key response mechanisms of parameters such as the reinforcement width and the spacing of longitudinal and transverse elements. The dominant role of soil dilatancy was identified and its effect on the response of different grid configurations was interpreted by recourse to simple conceptual physical models. It was found that interference between zones of restrained dilatancy may be substantial enough to promote the use of coarser (and hence more economical) mesh configurations, especially at shallow depths. For the pullout factor in highly dilative soils, a closed-form expression is proposed which describes reasonably well the experimental results, and adequately captures the interrelated phenomena occuring in two directions, transverse and longitudinal.