Alessandro Dazio

Shear Deformations of Slender Reinforced Concrete Walls under Seismic Loading

Note: Discussion Reference EPFL-ARTICLE-176186 Record created on 2012-04-13, modified on 2016-08-09

Quasi-Static Cyclic Tests of Two U-Shaped Reinforced Concrete Walls

U-shaped or channel-shaped walls are frequently used as lateral strength providing members in reinforced concrete (RC) buildings since their form does not only provide strength and stiffness in any horizontal direction but is also well suited to accommodate elevator shafts or staircases. Despite this popularity, experimental results on the seismic behavior of U-shaped walls are scarce. For this reason a research program with the objective to provide additional experimental evidence for such walls under seismic loading was developed. It included quasi-static cyclic testing of two U-shaped walls at the structural engineering laboratories of the ETH Zurich. The walls were built at half-scale and designed for high ductility. The main difference between the two walls was their wall thickness. The project was chiefly focusing on the bending behavior in different directions and therefore the walls were subjected to a bi-directional loading regime. This article discusses the design of the test units, the test setup and the test predictions. Finally the main results are summarized in terms of failure mechanisms and force-displacement hystereses.

Quasi-Static Cyclic Tests on Masonry Spandrels

This paper presents the results of an experimental campaign on masonry spandrels. Within this campaign, four masonry spandrels were subjected to quasi-static cyclic loading. Two different spandrel configurations were tested. The first configuration comprised a masonry spandrel with a timber lintel, and the second configuration, a masonry spandrel on a shallow masonry arch. For each configuration, two specimens were tested. The first was tested with a constant axial load in the spandrel, while for the second specimen, the axial load in the spandrel depended on the axial elongation of the spandrel. This paper summarizes the properties of the four test units, the test setup, and the most important results from the experiments, documenting the failure mechanisms that developed and the force-deformation hysteresis of the spandrel elements. The paper also presents a mechanical model for estimating the peak strength of masonry spandrels.

Inelastic Wide-Column Models for U-Shaped Reinforced Concrete Walls

Although core structures are often used in reinforced concrete buildings as members providing lateral strength and stiffness, experimental and numerical studies on their inelastic behavior are scarce. In an experimental program recently completed at the ETH Zurich, two U-shaped walls were subjected to a bi-directional quasi-static cyclic loading regime. In this article, inelastic wide-column models for these two test units are developed. The wide-column analogy was chosen because it combines the merits of representing the U-shaped wall as a three-dimensional structure with inelastic properties while still being relatively simple and easy to set up when compared to shell or solid element models. It is therefore a tool which is not only available to researchers but also to design engineers. The article commences with the analysis of wide-column models that have been built according to recommendations found in the literature. Since these recommendations had been derived from analyzes of elastic systems, they are then revisited in a sensitivity study in which the effects of different modeling assumptions on the inelastic behavior of wide-column models are investigated. Finally, comparing the numerical results with the experimental evidence from the tests, the article concludes with practical recommendations for setting up wide-column models of U-shaped walls subjected to large inelastic deformations.

Simulating Maximum and Residual Displacements of RC Structures: I. Accuracy

Estimation of likely global and local response measures plays an important role in seismic performance assessment. The capabilities and limitations of beam-column element modeling strategies in predicting the dynamic nonlinear flexural response of RC models are investigated in this study. For this purpose, 12 shake table tests are numerically reproduced. Correlations of the predicted deformations with the measured ones are evaluated. The results show that maximum displacements can be estimated with sufficient accuracy if the adopted hysteresis model takes into account stiffness degradation. However, accurate estimation of the residual displacements is found to be difficult to achieve. The results suggest that the assumed small-cycle behavior has a strong influence on the estimated residual displacements. Fiber-section models are found to provide relatively more accurate estimates of the residual displacements than modified Takeda hysteretic and bilinear models. A companion paper, Part II: Sensitivity, presents the sensitivity of the simulated displacements to a set of the model parameters and idealizations.

Quasi-static monotonic and cyclic tests on composite spandrels

In modern unreinforced masonry (URM) walls, the vertical piers are connected at the story levels by reinforced concrete (RC) ring beams—also known as bond beams—or RC slabs. Particularly, in the outer walls, the spandrel element also includes a masonry spandrel on top of the RC beam or slab (“composite” spandrel). Numerical simulations have shown that spandrels significantly influence the global behavior of the URM building when subjected to seismic loading. Despite their importance, experimental data on the cyclic behavior of composite spandrels were lacking. This paper presents the results of an experimental campaign on five composite spandrels. Each test unit consisted of an RC beam, a masonry spandrel and the adjacent masonry piers required for applying realistic boundary conditions to the spandrel. The investigated parameters included the type of loading, the brick type and the reinforcement content of the RC beam.

Simulating Maximum and Residual Displacements of RC Structures: II. Sensitivity

The simulated response of a structure subjected to seismic excitation is sensitive to the idealizations made to model its response. This paper examines critical idealizations and assumptions that have a strong influence on the accuracy of the maximum and residual displacements predicted by response-history analysis. A set of shake table tests are numerically reproduced for this purpose. The investigated idealizations include the discretization scheme, the axial load, the steel hysteretic model, the viscous damping ratio, and the time-integration step size. The results indicate that the simulated residual displacements are significantly more sensitive to the model idealizations than the maximum displacements. It is found that the adopted discretization scheme and the utilized steel hysteresis model have very large influences on simulated residual displacements.

Structural Testing of New East Bay Skyway Piers

This paper discusses 2 proof-of-concept tests conducted at 1/4 scale on a typical hollow, rectangular, reinforced concrete pier designed to support the new East Bay Skyway Structures of the San Francisco-Oakland Bay Bridge. The test units were loaded according to quasi-static, fully-reversed, cyclic loading histories to simulate the force-displacement response of the prototype pier under seismic forces. Both tests withstood inelastic deformations greater than those expected in response to the 1500-year governing design earthquake for the Bridge. Based on results from both tests, structural details such as the architectural concrete and transverse reinforcement were improved to enhance constructibility and service-level performance. This paper discusses 3 different approaches to evaluating experimental plastic hinge lengths and demonstrates the ambiguity inherent in evaluating the force-displacement behavior of such piers according to strain-based seismic performance criteria.

Discussion of “Simulating Maximum and Residual Displacements of RC Structures: I. Accuracy”

(1) Tazarv and Saiidi state that a unique “best modeling approach” is not proposed in the paper. This is correct; proposing a “best modeling approach” is not intended in the papers. The papers are aimed at investigation of the capabilities and limitations of some common beam-column element modeling strategies. Consequently, development of an ultimate modeling strategy is beyond the scope of the investigation. The authors reviewed the paper once more but could not find any statements suggesting that an approach for accurate estimation of residual displacements is provided. Tazarv and Saiidi ask why the relationship between the small cycle hysteresis response and the bias in the simulated residual displacements is presented based on the results obtained using displacement-based elements and not using force-based elements. The importance of small cycle response is presented based on the measured and simulated response histories of column “A1” tested by Hachem et al. (2003) in Figures 7 and 8. For the test unit “A1,” the maximum and residual displacements obtained from the simulations made using displacement-based and force-based element formulations are marked using “•”symbol in Figures 4 and 5, respectively. In these figures, it can be seen that the maximum and residual displacement obtained using displacement-based and force-based formulations for test unit A1 are practically the same. Therefore, it can be deduced that a very similar plot would be obtained if force-based elements were used instead of displacement-based elements. In brief, the difference would be practically irrelevant for the considered test unit. (2) Tazarv and Saiidi argue that three references which were available at the time of submission of the paper were missed. They provide references to two reports (i.e., Jeong et al. 2008, Lee 2007) and a journal paper (Lee and Billington 2010). The article by Lee and Billington appeared on May 2010 issue of the relevant journal. The submission date of our paper was 7 February 2010 (which is indicated

Das Erdbeben in Italien vom 26. September 1997

Der Aufsatz vermittelt eine erste Analyse des Erdbebens in Umbrien und Marken. Es werden seismologische und geotechnische Aspekte behandelt und Schaeden an neueren sowie an historischen Gebaeuden, an Industrieanlagen und an Strassenbruecken beschrieben. (A*)