Avigdor Rutenberg

Seismic base isolation of asymmetric shear buildings

Abstract Effects of alternative designs of bilinear base isolation systems on the seismic response of ten, five and two-storey shear buildings, with one axis of symmetry consisting of three parallel frames, are investigated. Analyses with the El Centro and Taft records predict appreciably lower shear forces and torques compared with the unisolated case, whereas lateral roof displacements are likely to increase for the lower structures. Optimum levels of yield force, and secondary to primary slope ratio of the isolation system hysteretic force displacement relationship, are discussed. Locating the centres of rigidity and the yield forces of the bearings at the mass centre of the superstructure appears to be the best strategy for reducing twist. Other alternatives, however, may reduce base motion and inter-storey shear even more. It is shown that increased shear forces and very large displacements may result when the 1977 Rumanian earthquake record is applied, suggesting that a decision regarding the use of base isolation depends on the earthquake characteristics at the site.

ON ECCENTRIC SEISMIC POUNDING OF SYMMETRIC BUILDINGS

Results of a parameter study on eccentric pounding of two symmetric single storey systems under seismic excitation are presented. Linear behaviour is assumed, and pounding effect is considered using the restitution coefficient approach. The effect of impact eccentricity is studied on two sets of symmetric models symmetrically and asymmetrically aligned with respect to each other for several gap widths, period dependent gaps and three values of the torsional-to-lateral frequency ratio. Two time histories are used for input. On the whole it was found that impact eccentricity amplifies the response relative to symmetric impact but the effect is not proportional to first impact eccentricity. Also, increasing gap width is likely to be effective when the separation is sufficiently wide practically to eliminate contact. Larger torsional rigidity tends to lower response amplification. SRSS code-type gaps appear to be adequate, or even excessive, when the design spectrum is compatible with the expected earthquake record at the site.

A direct P-delta analysis using standard plane frame computer programs

Abstract A simple technique is proposed that enables a direct second order analysis to be performed by means of first order plane frame computer programs. The geometric-stiffness matrix is modelled as a fictitious column with negative stiffness. The effect of additional moments due to eccentricity of axial force about the deflected shape is considered. The approach is also applicable to three dimensional problems and can also be used with nonlinear computer programs to evaluate the effect of sway on elastic-plastic frames.

Spatial response spectra and site amplification effects

A simplified analysis of local site amplification effects on the seismic response of multi-support structures is presented. The site effects are modeled by considering reflections and transmission of vertically propagating shear waves from bedrock to the surface through a soil layer. A random vibration-based response spectrum of a simple oscillator on two supports, one of which is founded on rock outcrop and the other on soil layer, is formulated in order to study the influence of non-uniform excitations on multi-support structures. Joint inertial and pseudo-static effects in the overall response are studied in detail. The resulting response spectra are formulated as displacement and force spatial seismic coefficients, convenient for sensitivity analyses. Numerical analyses of the response spectra as affected by various soil parameters are included.

Inelastic seismic response of code designed single storey asymmetric structures

Abstract A parametric study of the inelastic earthquake response of single storey asymmetric structures is presented. The aim is to compare the behaviour of structures designed by the static seismic provisions of the ATC3 Code with those of the 1985 National Building Code of Canada (NBCC). Three different configurations of the resisting system were analysed for four earthquake time histories. A bilinear force-displacement relation was assumed. The study shows that the maximum ductility demand of structures designed by the NBCC is lower than those designed following ATC3. This is due to the higher total strength and the more efficient strength distribution among the resisting elements in NBCC designed structures.

Dynamic torsional coupling in asymmetric building structures

Abstract An approximate method is proposed for the dynamic analysis of torsionally coupled tall building structures by utilizing the properties of their uncoupled counterparts. An exact solution is first given for the particular case in which the lateral and torsional stiffness matrices are uncoupled by same transformation. The method is then applied to a wider class of structures where this condition is only approximately satisfied by reducing the dynamic coupling problem to an approximate two-degrees-of-freedom system. Simple formulae and graphical representations of dynamic magnification of static eccentricity are given. Two numerical examples illustrate the use of the proposed method, checking on its accuracy and comparing its results with seismic code provisions.

A simple approach to the seismic design of friction damped braced medium-rise frames

This paper is concerned with the seismic design of steel frames using friction damped slotted bolted connections (SBCs) in the diagonal braces as a means for passive control. A dynamic model that uses bilinear hysteretic behavior for the damper describes the response of a single degree of freedom (SDOF) steel frame and constitutes the basis for the design. A design procedure that attains the stiffness of the individual braces and their elongation at the threshold of activation is then applied to 10-story and 3-story steel frames. This design procedure is a two-phase iterative scheme of the analysis/redesign type, the analysis of which is modal. Finally, a full nonlinear dynamic analysis that uses a generalized version for MDOFs of the bilinear hysteretic SDOF model assesses the design.

Seismic code provisions for asymmetric structures: a re-evaluation

Abstract Results of a parametric study on earthquake time history response of asymmetric single storey structures with one axis of symmetry modelled as two degree-of-freedom systems are compared with static seismic code provisions. A large number of such systems with a range of mass centre to rigidity centre eccentricities, a range of uncoupled lateral natural frequencies and a number of torsional to lateral frequency ratios Ω 0 were subjected to several earthquake acceleration records. Five percent damping was assumed in the two coupled modes. Maximum displacements at several locations along the roof deck were computed, normalized with respect to the symmetric case, averaged, and compared with code oriented static methods. The study shows that the static approach does not give reliable estimates for the response of frames in asymmetric buildings, even when the amplification factors provided by earthquake codes are incorporated into the formulation. In particular, code provisions often underestimate the response of frames located on the side of the rigidity centre away from the mass centre for small to moderate eccentricities when Ω 0 2 = 0.5, 1.0 ; whereas for systems with higher torsional rigidities (Ω 0 2 ⩾ 2.0) , the static approach appears to yield reasonable results. For members located on the opposite side of the roof, code provisions, including recent revisions, appear to overestimate the response with decreasing frequency ratio.

On the sensitivity of bridge seismic response with local soil amplification

The paper presents a numerical sensitivity study of the local site effects on structural response. Following a recently developed model of spatial coherency and a concept of a simple site coefficient the local site effects are modelled as filtrations of excitation processes with a frequency shift. An analysis of a bridge response with supports founded on different soils is carried out. The joint effects of dynamic response and pseudostatic motion are considered. Two types of response are analysed: longitudinal and transverse. The differences between dynamic displacements and force responses are pointed out.

Lateral load response of belted tall building structures

Abstract This paper presents the results of an investigation on drift reduction in uniform and non-uniform belted structures with rigid outriggers under several lateral load distributions which are likely to be encountered in practice. Design aids in the form of graphical presentations of the somewhat complex solutions are provided to assist the practicing engineer in the preliminary design stages.