Search

Several experimental studies have shown that rocking shallow foundations have beneficial seismic performance features: recentering and energy dissipation with little damage. A new publicly available database, “FoRDy” (Foundation Rocking—Dynamic), summarizes the results of dynamic physical model tests of single-degree-of-freedom-like structures supported on rocking foundations. It contains data from five centrifuge and three 1- g shaking table test series that were conducted at experimental facilities in the United States, Greece, and Japan. The database includes 200 model “case histories” that span a wide range of model sizes, soil and structure properties, and seismic excitations. It is compiled as the first step toward building a comprehensive dynamic rocking foundation database, and it has the potential to grow in the future. To illustrate its usefulness, the data are used to show example correlations between the peak drift ratio demand and selected ground motion intensity measures. The results suggest that peak ground velocity (PGV), peak ground displacement (PGD), and the geometric mean of the linear spectral displacement over the period range of 0.2–3 times the initial natural period predict the peak drift ratio response reliably.

The standard ASCE 41-13 Seismic Evaluation and Retrofit of Existing Buildings includes new provisions for linear and non-linear modeling parameters and acceptance criteria for rocking shallow foundations. The new modeling parameters and acceptance criteria were largely based on model tests on rectangular rocking foundations with a limited range of footing length to width ratio (L/B). New model test results are presented, including a systematic variation of L/B and also non-rectangular (I-shaped) footings. This new data along with previously published results are presented to validate the trilinear modeling parameters and acceptance criteria of ASCE 41-13. This paper investigates the effects of footing shape on the residual settlement, residual uplift, rocking stiffness, and re-centering. Overall, the new data supports the provisions of ASCE 41-13; however, the acceptance limits for rocking rotation of I-shaped footings could be reduced to produce performance consistent with the acceptance limits for rectangular footings.

ASCE 41-13 supports three methods of modeling the soil-structure interaction for rocking footings as components of a foundation-building system: Method 1 uses uncoupled moment, shear, and axial springs; Method 2 uses a nonlinear gapping bed of springs; and Method 3 is used for structural footings that are flexible relative to the underlying soil. New component action tables in ASCE 41-13 provide modeling parameters and acceptance criteria for nonlinear and linear analysis of shallow foundation components. The values in the component action tables for nonlinear procedures were largely based upon analysis of foundation performance in model tests on rocking foundations. The primary measure to assess foundation performance is residual settlement or uplift. The acceptance criteria for linear analysis procedures ( m-factors) were derived from the allowable rotations for nonlinear procedures. A design example is presented in an online Appendix to illustrate differences between the current and previous versions of ASCE 41 and ASCE 31.

Physical modeling of buildings founded on layers of liquefiable sand has produced insightful data describing ground and building performance. A new publicly available NEEShub database, “FLIQ,” summarizes data from 9 large-scale centrifuge tests with 4 different lead experimentalists, various soil profiles, 49 stations of site and structure response, and more than 60 shaking events, that is, 405 events in total. Example correlations between ground-motion intensity measures and performance measures are presented to illustrate the usefulness of the database. The relationship between PGA at the surface and base is highly variable when liquefaction is observed; however, the relationship between the cumulative absolute velocity ( CAV) at the surface and base is relatively linear, even when liquefaction occurs. Additionally, the CAV was found to correlate well with settlement and can account for settlement history. Contrary to intuition, foundation settlement in liquefied soil is not directly correlated to bearing pressure; larger bearing pressures can decrease liquefaction occurrence and related settlements. The database has potential to grow through future community input.

A two-dimensional nonlinear dynamic finite element (FE) model was developed and calibrated against dynamic centrifuge tests to study the behavior of soil-pile-structure systems in liquefied and laterally spreading ground during earthquakes. The centrifuge models included a simple structure supported on pile group. The soil profiles consisted of a gently sloping clay crust over liquefiable sand over dense sand. The FE model used an effective stress pressure dependent plasticity model for liquefiable soil and a total stress pressure independent plasticity model for clay, beam column elements for piles and structure, and interface springs that couple with the soil mesh for soil-structure interaction. The FE model was evaluated against recorded data for eight cases with same set of baseline parameters. Comparisons between analyses and experiments showed that the FE model was able to approximate the soil and structural responses and reproduce the lateral loads and bending moments on the piles reasonably well. © 2013, Earthquake Engineering Research Institute.

Many engineers are hesitant to specify rocking foundations for ordinary bridges because of the unsubstantiated notion that rocking bridges are more susceptible to instability than conventional fixed-base bridges. A parametric study using a finite element model including large deformation effects compares the performance and stability of stiff, flexible, tall, and short hinging-column and rocking-foundation systems. Eighty different ground motions, scaled using incremental dynamic analysis, were considered. Results show that, in a probabilistic sense, bridges with rocking foundations are more stable than bridges with hinging columns if their fundamental periods are the same and if base shear coefficients to initiate hinging or rocking mechanisms are the same. Maximum drifts are not much affected by changing between rocking and hinging mechanisms except near collapse, but residual drifts are smaller for rocking systems. The results also challenge the notion that rocking systems require a different design approach than hinging column systems.

This paper discusses the effects of using different input-motion-boundaryconditions on the sensitivity of numerical simulations results to errors in material properties of a specimen tested on a shaking table. In the flexibleactuator-prescribed-force-boundary-condition, input is specified by a force across an actuator element that connects the shaking table to a reaction mass. In the prescribed-displacement-boundary-condition, the measured shaking table motion in the experiment is prescribed in the simulation. The flexibleactuator-prescribed-force approach yielded smaller, almost constant sensitivity of simulation results to input properties. The prescribeddisplacement approach yielded larger and more variable sensitivities. The sensitivity of results depends on the how the boundary conditions are handled has further implications: the assessment of a comparison between a simulation and an experimental result should be performed with due consideration to the effect of the boundary conditions on the comparison, and numerically determined sensitivities may not be physically meaningful if the boundary condition is not accurately modeled. DOI: 10.1193/1.3383214

Practical guidelines for characterization of soil-structure interaction (SSI) effects for shallow foundations are typically based on representing foundation-soil interaction in terms of viscoelastic impedance functions that describe stiffness and damping characteristics. Relatively advanced tools can describe nonlinear soil-foundation behavior, including temporary gap formation, foundation settlement and sliding, and hysteretic energy dissipation. We review two tools that describe such effects for shallow foundations and that are implemented in the computational platform OpenSees: a beam-on-nonlinear-Winkler foundation (BNWF) model and a contact interface model (CIM). We review input parameters and recommend parameter selection protocols. Model performance with the recommended protocols is evaluated through model-to-model comparisons for a hypothetical shear wall building resting on clay and model-data comparisons for several centrifuge test specimens on sand. The models describe generally consistent moment-rotation behavior, although shear-sliding and settlement behaviors deviate depending on the degree of foundation uplift. Pronounced uplift couples the moment and shear responses, often resulting in significant shear sliding and settlements. Such effects can be mitigated through the lateral connection of foundation elements with tie beams.

Analytical design tools for evaluation of soil-pile-structure interaction during seismic events are evaluated and modified. Several implementations of the “Beam on Nonlinear Winkler Foundation” (BNWF) method were used to predict results of centrifuge model tests of single piles in a soft clay soil profile. This paper shows that calculations from these computer codes can be sensitive to the details of the arrangement of nonlinear springs and linear viscous dashpots. Placing the linear viscous dashpots (representing radiation damping in the far field) in series with the hysteretic component of the p-y elements (representing the nonlinear soil-pile response in the near field) is shown to be technically preferable to a parallel arrangement of the viscous and hysteretic damping components. Preliminary centrifuge data is reasonably modeled by the numerical calculations using this implementation of damping, but additional field or physical model data are needed to fully evaluate the reliability of BNWF procedures.

Five centrifuge models were tested to begin to quantify the effectiveness of liquefaction remediation techniques and to optimize the required volume of soil to be treated to reduce liquefaction consequences to a tolerable level. Detailed models of a bridge site including a river channel, flood plains, and bridge abutments were tested in the large centrifuge at the University of California, Davis. The models consisted of a deep, level, liquefiable sand layer overlain by a sloping, less-permeable clay layer with a free face at the river channel. Densification of a limited region of sand and in-ground water barriers were evaluated as possible remediation treatments. While spinning at a centrifugal acceleration of 30 g, the base of each model was shaken with realistic earthquake base motions. Base motions, pore pressures in the treated and untreated zones, profiles of ground deformations, settlements, and lateral sliding are compared. Settlements were controlled satisfactorily, but lateral sliding of the surface clay layer was not adequately controlled by the limited remedial treatments.

Soil-pile-superstructure interaction in liquefiable sand is evaluated using dynamic centrifuge model tests and pseudostatic p-y analyses. Select recordings from a recent centrifuge test are presented to illustrate typical behavior with and without liquefaction in an upper sand layer. Pseudostatic p-y analyses of single-pile systems in two recent centrifuge model tests show that the apparent reduction in p-y resistance due to liquefaction was strongly affected by changes in the relative density of the sand and drainage conditions.