Journal of Structural Engineering

Masonry arch structures, and, more generally, vaulted structures, are traditionally assessed using a well-established approach, such as linear elasticity or limit analysis, whereby system behavior at the intermediate stage—which occurs when the material’s tensile strength has been exceeded but the collapse mechanism has not yet formed—is disregarded. A more accurate interpretation requires a thorough analysis that can take into account the intermediate cracking stage and uses a constitutive law providing a closer approximation to the actual behavior of the material. In this paper, an evolutionary fracturing process analysis for the stability assessment of masonry arches is presented. This method makes it possible to capture the damaging process that takes place when the conditions evaluated by means of linear elastic analysis no longer apply and before the conditions assessed through limit analysis set in. Furthermore, the way the thrust line is affected by the opening of cracks and the redistribution of internal stresses can be checked numerically. Finally, by applying this evolutionary method, a numerical calculation of the arch of the Mosca Bridge over the Dora River in Turin, Italy, is described, and the results are compared with those obtained by Castigliano’s iterative analysis.

Rigorous numerical analyses are conducted to study the effects of bidirectional lateral cyclic loading on the hysteretic response of stocky circular thin-walled steel columns. Four unidirectional and eight bidirectional lateral displacement controlled load paths are considered; different performance indices obtained from the lateral load paths under consideration are compared. The results obtained from these comprehensive numerical analyses indicate that bidirectional lateral cyclic loading has a very limited effect on the ultimate strength of stocky circular thin-walled steel columns, but significantly decreases the ductility when compared with unidirectional lateral loading.

It has been well established that stainless steel nails have superior corrosion performance compared to carbon steel or galvanized nails in treated wood; however, their mechanical fastening behavior is unknown. In this paper, the performance of stainless steel nails is examined with respect to two important properties used in wood connection design: withdrawal strength and nail bending yield strength. Different nail diameters, wood specific gravities, and nail manufacturers were examined. The current withdrawal design equations, developed from carbon steel nail data, overpredict the expected withdrawal strength when used for stainless steel nails, reducing the safety factor. As a result, a new equation was developed to predict the nail stainless steel withdrawal capacity. The data further indicate that nail bending yield strength values for stainless steel were similar to carbon steel data.

Existing approaches used to model the lateral load versus deformation responses of reinforced concrete walls typically assume uncoupled axial/flexural and shear responses. A novel analytical model for RC walls that captures interaction between these responses for reversed-cyclic loading conditions is described. The proposed modeling approach incorporates RC panel behavior into a two-dimensional fiber-based macroscopic model. The coupling of axial and shear responses is achieved at the macrofiber (panel) level, which further allows coupling of flexural and shear responses at the model element level. The behavior of RC panel elements under generalized, in-plane, reversed-cyclic loading conditions is described with a constitutive fixed-strut-angle panel model formulation. The sensitivity of model results to various modeling parameters is investigated and results of the sensitivity studies are presented, whereas detailed information on calibration and validation of the proposed modeling approach is presented in a companion paper.

This paper presents the experimental calibration and validation of the analytical wall model that incorporates interaction between shear and flexural responses under cyclic loading conditions described in the companion paper. The model is calibrated and validated against detailed experimental data obtained from tests on five moderately slender reinforced concrete wall specimens that experienced significant levels of shear-flexure interaction. Test measurements were processed to allow for detailed comparisons between the predicted and measured wall responses at various locations and response levels. Response comparisons reveal that the proposed analytical model captures the experimentally measured nonlinear shear deformations and their coupling with flexural deformations throughout the cyclic loading history. In addition, the analytical results successfully represent various experimentally measured responses, such as lateral-load versus wall-top-displacement relations, magnitudes and distributions of deformations associated with shear and flexure, and local responses including flexural rotations, vertical strains, and concrete crack orientations. Based on the comparisons presented, model capabilities are assessed and future model improvements are suggested.

This article presents an experimental study on the seismic behavior of arch-pier systems reinforced with tie-rods. In particular, the role of the tie-rod stiffness is stressed, with the aim of proposing innovative tie-rods characterized by lower stiffness than traditional ones. The experimental campaign was performed by adopting the inclined plane static test on a 1:10 scaled model. It is demonstrated that prestress applied to tie-rods barely influences the seismic capacity of the system. Moreover, it is shown that larger tie-rod deformability induces larger displacement capacity at collapse. Finally, a comparison between experimental results and simple analytical models based on rigid block systems shows the fundamental role of overall system deformability on the seismic response.

The structural performance and design of cold-formed lean duplex stainless steel members in combined compression and bending are investigated. A wide range of square and rectangular hollow sections has been performed by finite-element analysis. A finite-element model has been developed and verified against the available test data for lean duplex stainless steel members subjected to axial compression and minor axis bending. An extensive parametric study was conducted using the verified finite-element model, and 150 finite-element analysis results were obtained. A total of 233 data, including the numerical results obtained in this study as well as the experimental and numerical data from the literature, are compared with design predictions from the current American, Australian/New Zealand, and European specifications for stainless steel structures. Reliability analysis was carried out to assess the reliability of these design rules for the lean duplex stainless steel beam-column members. It is shown that the American Specification is capable of providing accurate prediction for the compression and bending capacities, whereas the Australian/New Zealand Standard (AS/NZS) and European Code provide quite conservative predictions. However, the calculation procedure in the American Specification involves tedious iterative processes. Therefore, two modified design methods are proposed for lean duplex stainless steel beam-columns in this study. The two modified design methods in the AS/NZS and direct strength method generally provide better predictions for the cold-formed lean duplex stainless steel beam-columns compared to the current design rules.

The use of steel concentrically braced frames (CBFs) has substantially increased recently. However, investigations and observations from past earthquakes suggest that some existing CBFs, particularly the older ones that are still in service but were designed without ductile detailing and compliance with state-of-the-art seismic provisions, may exhibit unfavorable performance during strong earthquake events. This research numerically examines the adequacy of a seismic rehabilitation technology for deficient multistory CBFs. The technology consists of one or multiple rocking cores (RCs) added to an existing CBF to redistribute the seismic forces applied to the frame and reduce the system seismic response. For demonstration purposes, one three-story and one six-story CBF buildings are selected and rehabilitated using the RC technology. Finite-element (FE) models of the considered systems, which explicitly take into account the effect of gusset plates, member yielding, brace buckling, brace rupture, and the $\mathit{P}\text{-}\mathrm{\Delta }$ effect, are developed and validated. Through extensive nonlinear response history analyses using two suites of ground motions, seismic performance of the rehabilitated systems are evaluated. The results suggest that the RC technology is effective in enhancing seismic performance of low-rise- and midrise-deficient CBFs. Furthermore, the rehabilitated systems can benefit from the use of hysteretic energy dissipating links between RCs and existing CBF frames.

When an elastomeric bearing is subjected to simultaneous vertical compressive load and lateral displacement, the shear force can pass through a maximum beyond which the bearing exhibits negative tangential horizontal stiffness and a condition of unstable equilibrium. This behavior has been experimentally demonstrated and has important implications on the stability and earthquake response of elastomeric seismic isolation bearings. Yet, analytical bearing models used for numerical earthquake simulation assume a positive second-slope stiffness irrespective of vertical load and/or bearing lateral displacement and therefore are unable to simulate the experimentally observed bearing behavior. Semiempirical bearing models have been developed and some of these models have been shown to simulate the influence of vertical load and lateral displacement on the shear force response with reasonable accuracy, however these models rely on a number of experimentally calibrated parameters, making them impractical for the purpose of design. An alternative approach to modeling the behavior of elastomeric bearings is explored in this study, which uses a series of vertical springs and a simple bilinear constitutive relationship to represent the rotational behavior of elastomeric bearings. The mechanistic bearing model, utilizing vertical springs, is shown to be capable of simulating the experimentally observed behavior with reasonable accuracy without relying on experimentally calibrated parameters.

The use of fiber-reinforced polymers (FRP) has become widely accepted engineering practice for strengthening reinforced concrete members. It is well established that lateral confinement of concrete with FRP composites can significantly enhance its strength and ductility. As the confinement pressure generated by FRP on the confined concrete depends on the lateral expansion of concrete, the mechanism of concrete expansion inside the FRP shell is of significant interest. A review of the existing stress-strain models of FRP-confined concrete revealed the need for a model that accurately predicts the dilation characteristic of confined concrete as it provides the essential link between the response of the concrete core and the passive confinement mechanism of the FRP shell. It is also understood that knowledge established from the research area of actively confined concrete can be employed in the development of a model applicable for both FRP-confined and actively confined concretes. Based on a large number of experimental test results of both FRP-confined and actively confined concretes, a generic model is proposed to describe the lateral strain-to-axial strain relationship of confined concrete. The instrumentation arrangements of the tested specimens have allowed for the lateral strain-axial strain relationships of confined concrete to be captured throughout the tests. The trend of the lateral strain-to-axial strain relationship of confined concrete is shown to be a function of the confining pressure, type of confining material and concrete strength. Assessment of models with the experimental databases showed that the predictions of the proposed model are well above existing models and in good agreement with the test results of both FRP-confined and actively confined concretes.

A cable-spring truss system for supporting ethylene tetrafluoroethylene (ETFE) cushions spanning between buildings to create atria is described. The truss system uses cables together with springs encased in telescoping tubes. The system is remarkably lightweight. The springs provide the system with flexibility to accommodate differential displacements of the supporting buildings under wind and seismic loading. Design equations are developed for the springs, tubes, and cables. A design example is investigated, and the flexibility is compared to that of an equivalent cable-strut support system.

A theoretical model is developed in this paper for investigating the nonlinear behavior of one-way high-strength reinforced concrete wall panels. The model accounts for geometric and material nonlinearities, including the strain softening of concrete in compression, cracking, tension stiffening, and the yielding of steel reinforcement. The governing differential equations of the model are solved by the nonlinear shooting method, along with the use of the arc-length continuation method. A smeared cracking model is adopted, and an iterative procedure is conducted at each load step to determine the unknown rigidities of the cracked section and the length of the cracked region. An experimental study that includes testing to failure of eight panels under eccentric axial compression is carried out. The effects of the reinforcement ratio and arrangement, load eccentricity, and slenderness ratio are examined in the experimental study. A good correlation between the theoretical model and the experimental results is obtained. The results reveal that buckling dominates the failure of the tested panels, which is greatly influenced by cracking, and it is sensitive to the initial load eccentricity and slenderness of the panel and almost insensitive to the reinforcement ratio and location. The theoretical and experimental results are compared with design code predictions, and the model is further verified through comparison with other test results in the literature.

Steel pins are an attractive and efficient alternative technology to screws when connecting cold-formed steel or other materials to cold-formed steel members. However, unlike screws, pins remain an essentially proprietary connection technology based on manufacturer claims of special pin shank design. As such, current cold-formed steel design standards and specifications do not include provisions for their direct use and designers have no simple, rational means for making informed design decisions beyond information provided by manufacturers. This paper reviews the current design model for screws and examines the basic tension pullout/withdrawal limit state of helical knurled steel pin connections for cold-formed steel thicknesses common in residential construction. Included in this paper is a determination of material properties using a nondestructive test method as an alternative to traditional methods for estimation of steel base metal thickness. A simple lower bound friction model is proposed for helical steel pins and the model is calibrated using data from 32 physical pin tests. The calibrated model is shown to provide an acceptable degree of reliability and safety consistent with that of screws in current cold-formed steel design standards and specifications.

A new damping system, the viscoelastic coupling damper (VCD), has been developed to enhance the wind and seismic performance of coupled shear wall high-rise buildings by adding high damping elements in place of reinforced concrete coupling beams. VCDs replace structural members, such as outriggers or coupling beams, and therefore do not occupy any usable architectural space. When they are properly configured in high-rise buildings, they provide supplemental viscous damping to all lateral modes of vibration, which mitigates building tenant vibration perception problems and reduces both the wind and earthquake response. Experimental results from tests on five small-scale viscoelastic (VE) damper specimens of 5- and 10-mm thicknesses are first presented, followed by the results from six full-scale VCDs representing two alternative configurations. The first was designed for areas where moderate seismic ductility is required, and the second was designed with built-in ductile structural fuses for areas where high seismic ductility is required. The VE material tests exhibited stable hysteretic behavior under the loading conditions that are expected in high-rise buildings under wind and earthquake loading. The full-scale tests validated the overall system performance within a realistic coupled wall configuration, and confirmed the performance of the wall anchorages and all connecting elements as well as the VE material behavior. The full-scale test results also demonstrated the targeted viscoelastic response during wind and low level earthquake loading and the targeted viscoelastic-plastic response for extreme earthquakes, where the response is a combination of the VE response and the nonlinear behavior of the structural fuses.

The earthquake on February 22, 2011, in Christchurch, New Zealand, resulted in the first documented field fractures of links in eccentrically braced frames (EBFs). A comprehensive forensic analysis of these fractures, which occurred in the parking garage of the Christchurch Hospital, is presented. The analysis is based on mechanical and spectrochemical testing and three-dimensional (3D) scanning of procured physical samples of the fractured links. The analysis features nonlinear time history simulations to characterize deformation demands and continuum finite-element simulations to determine the capacities based on a sophisticated fracture mechanics model. The exercise represents a multiscale end-to-end simulation of the system and provides insight regarding the observed fractures. The analysis reveals the inherent challenges in determining the proximate cause of the fractures because the fractures occurred because of a confluence of several interacting factors, primarily the intensity of shaking (several times the intensity that was expected during a design-level event) and the frame geometry, which severely amplified the imposed demands. In addition, the fractured links also suffered from an erection (fit-up) error, in which the link stiffener was not located (as specified) directly above the brace flange, producing a severe strain concentration. This flaw significantly reduced the deformation capacity; however, the simulation of hypothetical scenarios indicates that, even with this flaw, the links would (with high likelihood) have survived a design-level event. Strategies for mitigation include stricter tolerances for stiffener location, consideration of frame geometry to reduce amplification of rotation, and enhancement of the seismic hazard used for design. The limitations of the study are outlined.

Buckling of longitudinal bars is a common form of damage in reinforced concrete (RC) structures subjected to earthquakes. Previous research has illustrated the impact of cyclic loading on bar buckling which often occurs upon the reversal from a tensile loading cycle. This paper presents a finite-element method to predict reinforcement buckling under seismic loading that also captures the details of the buckling mechanism. This method combines a fiber-based model to simulate the reinforced concrete member itself and an independent finite-element model of the local plastic hinge region. The strain demands in the plastic hinge region are determined from the fiber-based model of the overall structure subjected to the ground motion. The strain history is then imposed on the finite element bar buckling model to predict the localized behavior. Comparisons between the model performance and experimental observations are shown to assess the accuracy of the proposed method.

Design provisions for cold-formed steel (CFS) light frame shear walls with wall height-to-width/length ratios (aspect ratios) greater than $2:1$, and less than $4:1$, requires a reduction in nominal strength based on the wall aspect ratio. This reduction is attributed to increased wall flexibility. The current strength reduction is not supported by existing data and it is not justified given building code lateral displacement limits and the availability of expressions to estimate wall displacement. Additionally, a significant underestimation of nominal or peak strength is not helpful where capacity-based design is adopted. In this paper, the results of an experimental program that included 11-mm oriented strand board (OSB) sheathed CFS shear walls with aspect ratios between $2:3$ and $8:1$ are presented, and a conservative, more consistent and less severe strength reduction expression is offered—an expression that facilitates more sustainable design and meets the intention of the current standards of engineering practice.