Journal of Structural Engineering

Forum papers are thought-provoking opinion pieces or essays founded in fact, sometimes containing speculation, on a civil engineering topic of general interest and relevance to the readership of the journal. The views expressed in this Forum article do not necessarily reflect the views of ASCE or the Editorial Board of the journal.

The yield-line method provides a powerful means of rapidly estimating the ultimate load that can be carried by a reinforced concrete slab. The method can reveal hidden reserves of strength in existing slabs and can lead to highly economic slabs when used in design. Originally conceived before the widespread availability of computers, the yield-line method subsequently proved difficult to computerize, limiting its appeal in recent years. However, it was recently demonstrated that the discontinuity layout optimization (DLO) procedure could be used to systematically automate the method, and various isotropically reinforced, uniformly loaded slab examples were used to demonstrate this. The main purpose of this paper is to demonstrate that the DLO procedure can also be applied to a wide range of more practical slab problems, for example involving orthotropic reinforcement, internal columns, and point, line, and patch loads. The efficacy of the procedure is demonstrated via application to a variety of example problems from the literature; for all problems considered solutions are presented that improve upon existing numerical solutions. Furthermore, in a number of cases, solutions derived using previously proposed automated yield-line analysis procedures are shown to be highly nonconservative.

Over the past two decades, there has been significant interest in research relating to concrete-filled tubes, and a corresponding penetration of this technology into practice. This paper aims to expound upon the effect of timber cores on the structural response of concrete-filled circular tubes under compression. A timber infill with different shapes and geometries surrounded by concrete and encased in a steel tube was employed. The effects of the combination of infill elements on the failure, axial capacity, ductility, and structural efficiency (weight versus capacity) are exhaustively set forth. For the specimens with the highest timber to concrete ratio, the capacity was enhanced by about two times the capacity of the hollow steel specimens. For these specimens a significant reduction in the total weight of the composite element was obtained relative to the fully concrete-filled specimens. These specimens showed the highest ductility among the other specimens. In addition, greater ratios of energy absorption to the mass were obtained for the specimens with different timber cores in comparison to the equivalent values for fully concrete-filled tubes, which is quite desirable in many practical scenarios. It is found that the use of timber as an inner core element in this new composite yields promising results in decreasing the weight and yet enhancing the capacity, ductility, and energy absorption, and can be a good alternative to double-skin concrete-filled steel tubes.

This study presents the experimental and analytical investigations carried out on high-strength concrete beam-column joints with Grade 700 longitudinal bars in beams and columns. Seven full-scale interior RC beam-column joints with various reinforcement detailing were designed according to the specific seismic provisions of current building codes. Test specimen variables included concrete compressive strength, reinforcement yield strength, and axial compression loading level. The test specimens were subjected to constant column axial loading and quasi-static lateral load reversals. The performance of each test assembly was examined in terms of cracking patterns, lateral loading capacity, reinforcement strain profiles, secant stiffness, energy dissipation capacity, joint shear strength, and bond performance. All specimens displayed ductile failure mode and it is concluded that the use of high-strength concrete and applied axial compression loading can improve specimen bond conditions. Parametric studies via finite-element analysis were performed to examine the influence of various parameters on the strength, bond condition, and energy dissipation capacity of the specimens. In addition, a backbone curve model for the shear behavior of the joints is proposed. An explanation for the observed cracking pattern and further analytical investigation on the joint shear capacity of the specimens is presented.

Porous safety fences are increasingly being incorporated into the deck furniture at or near the edges of many long-span bridges, and these tall fences (often 3–4 m high) may become blocked during winter storms. The accretion of ice or snow on these fences decreases the porosity and significantly impacts the air flow around the bridge. Due to the relatively low probability of concurring extreme wind and extreme snow/ice events, a methodology is developed and presented to assess the probability of failure due to flutter in these events using wind tunnel test results. Analysis of these test results also indicates that responses due to vortex-induced oscillation may amplify as the fences become increasingly blocked by snow or ice. Similar trends were observed for both single and parallel bridge configurations, suggesting the need for careful studies focused on these effects for bridges in climatic regions susceptible to snow and ice. The effects of snow and ice accretion on fences are also analyzed in the context of design wind loads; results suggest that the primary concern for these effects is the aerodynamic stability of the bridges rather than the increases in wind loading.

Two T-shaped reinforced concrete wall specimens were subjected to reversed cyclic loading quasi-statically to failure. Both represented half-scale wall assemblages of a 6-story prototype building. Modifications to the wall detailing were incorporated to study the effects of longitudinal reinforcement distribution and splicing, shear lag, increased amounts of shear reinforcement, and increased dimensions of the boundary elements beyond original code-based requirements. In addition, the minimum number of stories required to capture important aspects of multi-story wall behavior through physical experiments was investigated. Distributing the longitudinal reinforcement across the flange, rather than concentrating it within the boundary elements, was found to reduce crack widths, damage to the wall, and shear sliding across the wall panel. Concentrating large amounts of reinforcement in the flange tips tended to increase shear lag effects in the web-direction loading, but led to moderate increases in the in-plane strength and deformation capacity in the flange direction. Locating the lap splices at the second-story level avoided problems with localized damage observed in cases where lap splices are located at the wall-flange interface. Increasing the amount of shear reinforcement and dimensions of the boundary elements did not have a significant impact on behavior. A minimum of two stories was found to be necessary to characterize the behavior of this 6-story prototype structure; it was sufficient to capture the height over which plasticity occurred.

Experimental and numerical studies were conducted to investigate the vertical load-carrying behavior and seismic performance of a composite frame structure with an innovative composite transfer beam, which was proposed to overcome the disadvantages of traditional reinforced concrete or steel reinforced concrete transfer beams. In the experimental study, a vertical monotonic loading test and a lateral cyclic loading test were conducted. The test observations, load–displacement curves, and strain measurements were discussed to investigate the structural performance and failure mechanism. The experimental results demonstrated the excellent vertical load-carrying behavior and seismic performance of the composite transfer frame. The characteristics of the plastic hinge distribution were investigated based on extensive data of the steel strains measured to reveal the typical failure mechanism of the composite transfer frame. In the numerical simulation analysis, a multiscale modeling scheme was developed to make full use of the fiber beam–column elements and multilayer shell elements. Comparisons with the experimental results showed that the developed model predicted the overall structural behavior, the individual story and component behavior, and the failure mechanism with a reasonable level of accuracy. In addition, both the experimental tests and numerical analyses indicated that the shear deformation mode and the energy dissipation in the composite joint core were also significant mechanical characteristics of the composite transfer frame.

The use of duplex high-strength stainless steel (HSSS) Grade 2205 prestressing strand and austenitic stainless steel (SS) Grade 304 spiral wire reinforcement is proposed for reinforcing prestressed concrete piles in marine environments in order to provide a 100+ year service life. The study experimentally investigated 40.6-cm-square, 21.3-m-long piles, which were successfully driven to refusal without visible damage, extracted, and tested in flexure and shear. Flexural and shear strength of piles using duplex HSSS 2205 strands were greater than predicted by design specifications. Experimental prestress losses for piles using duplex HSSS 2205 strands were approximately 80% of the predicted losses by standard provisions and were not affected by pile driving and extraction. These findings indicate that duplex HSSS 2205 can be used for prestressing strands in combination with austenitic SS 304 for the transverse confinement and shear reinforcement for prestressed concrete piles, using the same design requirements and construction procedures used for conventional prestressing strand and wire reinforcement.

Incident and reflected peak overpressures and reflected impulses are often used to compute blast loads on structures, components of structures, and objects. Reflection coefficients are used to transform incident to reflected peak overpressures for varying angles of incidence. Values for the reflection coefficient are available in textbooks and technical manuals but vary by document, especially in the Mach reflection region. Results of computational fluid dynamics analyses are presented to resolve differences among these documents and compute reflected scaled impulses. Recommendations are made for design practice.

The objective of this study is to identify the significant parameters that influence the lateral load–drift response of steel posttensioned (PT) connections. In particular, the effects of variations in beam section size are included in the study. First, three-dimensional finite-element models are developed to simulate the monotonic lateral load behavior and limit states of PT connections with top-and-seat angles. Full models are developed to capture out-of-plane movement and beam local buckling behavior of PT connections. The analytical response is validated by using two sets of previous experimental studies. By comparing the test and analytical results, damage identification measures are defined for different limit states, such as angle fracture and strand yielding. The established damage identification measures are then used to examine the monotonic lateral load behavior of PT connections. Using a statistical design-of-experiment approach, the relative significance of 16 factors is assessed. These factors include geometry-related and material-related sources of uncertainty. Results demonstrate that beam depth, column height, and posttensioning strand force have large effects (with contributions greater than 15%) on at least one of the response variables (including initial stiffness, the onset of decompression, residual stiffness, and load capacity). Additionally, there is statistical evidence that the beam depth and the beam flange thickness and width influence all the response characteristics.

Since the devastating 2011 tornado season, there has been renewed interest in understanding tornado wind loads and developing methodologies to reduce the risk of tornado damage. This study focuses on the assessment of the system-level performance of five different residential light-frame wood building archetypes subjected to tornado winds. The system-level performance considered three critical components along the vertical load path within each building, namely, the roof sheathing, roof-to-wall connection, and wall-to-foundation anchorage. Although there has been significant work on each of these components independently, combined uplift and lateral forces have received less attention. Therefore, a laboratory testing program for wood shear walls, reported herein, was performed to determine the wall-failure mechanisms and capacities when subjected to combined shear and uplift forces. Each building archetype was designed at varying locations across the United States based on the current residential building code and/or observed practice. Fragilities were developed, based on the assumption that failure of any component along the vertical load path results in failure of the system. Because the forces in each connection along the vertical load path arise from the same tornado wind load, the connection failures are not statistically independent, and Monte Carlo simulation was applied to perform the system reliability analysis.

This paper describes a new method for fitting probability distributions for modeling annual maximum ground snow loads for use in structural design. These probability models are intended for use in reliability assessments to determine reliability-targeted ground snow loads, as described in the companion paper. The proposed method emphasizes the upper tail of the distributions, because the upper tail is most critical for the reliability assessment and for determination of design loads. A combination of site-specific and region of influence approaches results in annual maximum snow load distributions whose magnitudes are consistent with historical data observed at the site of interest but with upper tail shapes that are informed by historical snow records at a number of similar sites. Clusters of sites with similar snow accumulation patterns are utilized to improve the definition of the snow load distribution at sites in the cluster.

This paper describes the development of reliability-targeted ground snow load maps for use in building (roof) design. The proposed procedures aim to ensure that structures designed achieve a target safety index, taken to be 3.0 as defined in ASCE 7 (ASCE 2010). When applied to the U.S. state of Colorado, the reliability-targeted mapping procedure shows that to achieve this target reliability index, design ground snow loads may need to be larger or smaller than the 50-year return period load that ASCE 7 currently stipulates for design, depending on the site and winter climate conditions at that site. Sites with larger coefficients of variation in the annual maximum snow load (in Colorado, lower-altitude sites) generally need design loads larger than the 50-year values. The paper also advances snow reliability assessment procedures through the proposal of a new model for the ground-to-roof conversion factor needed for quantifying roof snow loads as a function of ground snow loads. Spatial smoothing and altitude-dependent mapping procedures are described.

An understanding of how corrosion affects fatigue behavior of reinforced concrete (RC) structures is essential for lifetime-oriented maintenance and design of existing bridges. This paper covers an experimental investigation on fatigue behavior of corroded RC beams. Corrosion in six RC beams was artificially accelerated by the impressed current method before fatigue loading was administered. All beams failed with fatigue fracturing of their tension steel bars. The fatigue test results indicated that the more severe the corrosion on steel bars was, the sooner the beams failed. The behavior of corroded steel bars determines the fatigue behavior of RC beams. This paper introduces a segmented linear model to simulate nonlinear behavior of corroded RC beams under fatigue loading, which was verified by experimental results. Of particular interest was the deterioration of corroded RC beams under fatigue loading, including the fatigue damage to concrete and corroded steel bars as well as the bond behavior between steel bars and concrete. The proposed model can simulate the development of strains in concrete and steel bars under fatigue loading, thereby permitting a fatigue life assessment. Further analysis investigated the effects of corrosion level, longitudinal variation of the cross-sectional areas, reinforcement ratio, load amplitude, and bond behavior as well. With the increasing degree of corrosion, load amplitude and longitudinal variation of the cross-sectional areas, or the decrease of the reinforcement ratio, the fatigue life of corroded RC beams decreased. The deterioration of bond behavior had little effect on fatigue life, but it led to a reduction in the flexural stiffness of beams.

In this study, two methods used to estimate surface terrain conditions ($z_0$) surrounding Automated Surface Observing System (ASOS) stations in tropical cyclone-prone regions were analyzed. The first method utilized ground-based photography from the National Oceanic and Atmospheric Administration Hurricane Research Division’s Tropical Cyclone Wind Exposure Documentation Project (WEDP). The second applied a modified effective $z_0$ method using ASOS wind data. Comparisons reveal that WEDP $z_0$ estimates are typically larger in magnitude and do not account for changes in upstream $z_0$. Variability in the $z_0$ estimates showed a distinctly skewed nature in the probability distributions, which may have some physical meaning. It was also shown that standardized maximum 1-min sustained wind speeds using both the WEDP and MM2010 median $z_0$ estimates for short fetch lengths can have differences as large as 15%. The MM2010 $z_0$ estimates were also compared with single weighted mean $z_0$ estimates to be incorporated in the next version of the U.S. wind loading standard and suggest that a single weighted mean $z_0$ value does not effectively capture the variability of the terrain with respect to wind direction.

Fluidic self-centering devices operate on principles similar to those of fluid viscous dampers, but with additional capabilities to provide recentering force (or preload) and stiffness. The functions of preload, stiffness, and damping are entirely controlled by the physical properties, volume, and initial pressure of fluid, and the shaping of the orifices. They are characterized by compactness in comparison to the output force and stroke. They are proposed as self-centering bracing systems for buildings. This paper presents: (1) a description of the behavior of these devices that is on the basis of principles of mechanics, (2) models of behavior of the devices, (3) results of testing of a number of these devices to demonstrate their behavior, and (4) validation of the analytical models. The test results include those of two devices tested over a period of 22 years that demonstrate the stability of the properties of the devices.

In recent years, the need for resilient structural systems has led to renewed interest in the use of rocking members instead of conventional ones for the design of earthquake-resistant structures. Due to the usually constrained nature of such members, in addition to the deformability along their height, the nonlinear stress distribution at their base resulting from the partial loading of the rocking interface needs to be taken into account for an accurate estimation of the rocking motion. In this paper, a new approach is proposed for the prediction of the response of elastic rocking bodies under static loading that is able to consider such effects. The nonlinear stress distribution near the contact area results in additional displacements compared to those of the technical theory of bending, and a method to estimate them is presented and incorporated into a macroelement formulation that can be used in the context of a finite-element program. Results for typical examples are presented and compared with those of corresponding commercial software models, showing excellent agreement, while the proposed macroelement requires extremely low runtimes compared to conventional finite-element codes.

Design models for timber structures with fire-resistance requirements usually take into account the loss in cross section due to charring and the temperature-dependent reduction in strength and stiffness of the residual cross section. Standards for glued timber elements like glued-laminated timber beams have assumed that structural adhesives do not influence the fire resistance, but further specifications of adequate adhesive requirements are missing. Investigations were therefore conducted to determine the influence of different adhesives on the load-carrying behavior of glued-laminated timber beams in fire. Large-scale fire-resistance tests in combination with extensive finite-element analyses showed that a modification of the zero-strength layer thickness of $d_0=7\mathrm{mm}$ currently used in a standard design code should be discussed for glued-laminated timber loaded in bending in a future release of this design code. An increase of the zero-strength layer $d_0$ considering the finger-joint strength depending on the approved structural adhesive is not needed.

Engineering research is undergoing dramatic changes with novel, large-scale research facilities being developed to help reduce the growing economic losses associated with natural disasters. The wall of wind (WOW), at Florida International University (FIU), is such an experimental facility, capable of simulating hurricane winds and wind-driven rain to better understand hurricane effects on buildings and other infrastructure. The WOW has been designated by the national science foundation (NSF) as one of the experimental facilities (EFs) under the natural hazards engineering research infrastructure (NHERI) program, which aims to allow researchers to enable innovations and help prevent natural hazards from becoming societal disasters. The WOW EF’s goal is to facilitate research to improve design practices for structural systems, building envelopes, and lifeline infrastructure. This paper presents an overview of the design and development of the WOW research facility and delineates its capabilities to assess and mitigate the impacts of hurricane wind, rain, and debris on civil infrastructure. Advantages and limitations of the facility are explained. To illustrate the EF’s research capabilities, details and results from three recent case studies related to large-scale and full-scale testing of building components and traffic infrastructure are described.

Reinforced masonry (RM) shear walls are key structural elements widely used to resist lateral loads in masonry buildings due to their capability to provide lateral strength, stiffness, and energy dissipation. This paper investigates the in-plane seismic performance of fully grouted RM shear walls dominated by shear failure. The experimental work involved assessing the response of five single-story RM shear walls when subjected to in-plane axial compressive stress, cyclic lateral excitations, and top moment. The studied parameters were the horizontal reinforcement ratio, level of axial compressive stress, and shear span to depth ratio. Most of the current provisions in masonry design codes for nominal in-plane shear strength, $V_n$, for RM shear walls consider a percentage of the yield capacity for the shear resistance provided by the horizontal reinforcement, $V_s$. Moreover, they limit the effect of the shear span to depth ratio, $M/V{d}_v$, to an upper value of 1.0. The test results concluded that the horizontal reinforcement could contribute 100% of its yield capacity toward $V_n$. In addition, increasing $M/V{d}_v$ from 1.25 to 1.875 resulted in a significant reduction in the shear strength by 25%, which means that limiting the effect of $M/V{d}_v$ to an upper value of 1.0 is overestimating $V_n$ of RM shear walls at high values of $M/V{d}_v$. However, this reduction is accompanied by higher displacement ductility. On the other hand, increasing the axial compressive stress resulted in a higher $V_n$ with more brittle failure.

In this paper, a finite-element-based computational method is proposed for time-dependent structural stability analysis of a concrete-filled steel tubular (CFST) arch with uncertain parameters. Specifically, the targeted uncertainty includes the mercurial effects of the creep and shrinkage of the concrete core, which inevitably affect the structural performance of the CFST arch. The structural stability of the composite arch is systematically investigated under the influence of uncertain creep and shrinkage in a time-dependent fashion. The proposed computational scheme efficiently establishes the quantitative long-term stability envelope for CFST arches against uncertain viscoelastic effects. In order to demonstrate the effectiveness and efficiency of the proposed time-dependent structural stability analysis for CFST arches, practically motivated numerical examples are thoroughly investigated throughout this work.

This paper investigates the fatigue behavior of reinforced concrete beams at low temperatures compared to similar beams tested in fatigue at room temperature. Four large-scale steel-reinforced beams ($200\times 400\times \mathrm{4,200}\mathrm{mm}$) were fabricated and tested. All of the beams had temperature differentials over their depth to simulate solar radiation and in-service temperature of the bridges. Two beams had stirrups while the other two had no shear reinforcement. One beam of each type was tested at room temperature and the other at low temperature. The beams were cyclically loaded to failure with a stress range representative of the ratio of live to dead loads found in most bridges. This study showed that the fatigue life of the reinforced concrete improved at low temperature and this improvement was much more pronounced when shear reinforcement was not present. In fact, low temperature changed the mode of failure of the reinforced concrete beams without stirrups from shear to flexural fatigue failure as a result of the higher shear strength and stiffness of the reinforced concrete, the lower number of cracks with smaller widths, and improved fatigue properties of steel at low temperature. The results showed that the fatigue life of the reinforced concrete beam without shear reinforcement at room temperature primarily depended on the concrete and concrete-to-steel bond while fatigue lives of the other beams depended on the stresses in the tensile reinforcement. The strains and stiffness degradation were much lower in the low-temperature beams compared to their room-temperature counterparts. The findings of this study showed that the contribution of concrete to the performance of the beam under either static or cyclic load becomes more pronounced at low temperatures.

In an effort to mitigate major costs associated with isolation systems, especially in retrofit applications, isolation bearings may be placed on the tops of columns. In this configuration, a seismic gap and additional rigid diaphragm are no longer necessary, and excavation and foundation work may potentially be avoided in retrofit applications; however, columns under the isolation layer may not provide rigid boundary conditions, resulting in bearing end plates not remaining parallel. To investigate the effects of flexible columns in a column-top isolation system with elastomeric bearings, quasi-static cyclic testing of bearing-column subassemblies were conducted. Findings show that flexible end conditions can significantly reduce the lateral stiffness of elastomeric bearings. A simple analytical stiffness matrix is derived for bearings consisting of translational and rotational degrees of freedom at both end plates and is compared against experimental results. The simple model compares well for initial behavior but diverges from experimental results with larger end-plate rotations; however, when calibrated nonlinear material properties are included, the model performs well at capturing the behavior of elastomeric bearings when subjected to partially restrained boundary conditions.

Research was conducted to investigate how the inclination angle of the diagonal tension field action varies in steel plate shear walls (SPSWs) and to determine what optimum constant angle best matches the demands obtained from finite-element (FE) analysis. An FE model was first calibrated against experimental results that surveyed inclination angles across the web plate of an idealized SPSW as a function of drift and that showed significant differences in inclination angles at different locations across the web plate. Then, four real SPSWs with varying aspect ratios and numbers of stories were designed and modeled for FE analyses. The variations in angle in the web plate and along the boundary elements were documented as a function of drift and showed significant variations. Combined moment–axial force demand ratios in the SPSW boundary elements were calculated and compared for all real SPSWs to determine the preferable value of single angle that could be used in design. Overall, using 45° was found to be a reasonable compromise for both horizontal and vertical boundary element (HBE and VBE, respectively) design if a single constant angle is desired. Furthermore, the demand on the web plate is not sensitive to the variation of inclination angle. Consequently, the single angle of 45° is recommended for the design of the entire SPSW.

To investigate anchorage performance in a multidirectional arrangement of reinforcement bars, simulations are carried out using a three-dimensional discrete model, specifically the three-dimensional rigid-body-spring model (3D RBSM). In 3D RBSM, a material is partitioned into an assemblage of rigid bodies interconnected along their contact boundaries through discrete springs. In this study, for example, RC is meshed into rigid bodies with a size of 1–2 cm, where the opening, closing, and sliding of cracks can be modeled by the deformation or transmission of internal forces between two rigid bodies through the use of nonlinear springs along their contact boundaries. Simulations are conducted, one with the number of transverse reinforcement bars as a parameter and another varying the clear space between column reinforcement and embedded reinforcement in a congested joint. The results of these simulations are evaluated through comparison with experimental data; good agreement is observed with respect to anchorage capacity, crack pattern, and failure mode. The simulation results indicate the concrete strength is reduced if the reinforcement spacing between the column reinforcement and the embedded reinforcement is very close because of a nonhomogeneous behavior of concrete, where in the previous experiment large voids around reinforcing bars were observed. Simulations also suggest that the propagation of local cracks depends on the arrangement of reinforcing bars. This research shows how mesoscale analysis using the 3D RBSM can be a useful technique to consider the effect of a multidirectional arrangement of reinforcing steel that results in a complex stress and strain state due to the 3D bond transfer mechanism. It can be used for the design of a reinforcement arrangement in congested areas that is not addressed by the design code. As a result, a rational reinforcement arrangement can be designed on a case-by-case basis.