Journal of Structural Engineering

A multimass model is applied to simulate the impact‐load characteristics for an impact collision. Comparisons with verification tests indicate that the model is capable of predicting the resultant impact‐load characteristics and acceleration response. The multimass model is then linked to a dynamic finite‐element analysis for concrete‐slab structures through an interactive procedure. The interactive procedure is useful as the effects of effective mass and effective stiffness can be simulated accurately and upgraded at each time step. Finally, the ultimate behaviors and failure modes for reinforced‐concrete guardrails and fiber‐reinforced‐plastic (FRP) reinforced‐concrete slabs under impact collisions are studied. The kinetic energy transmitted by an impacting body as well as the energy absorbed by the concrete‐slab structures are found to be affected by the final failure modes in the structure.

A pointwise equilibrating polynomial (PEP) element is proposed for nonlinear analysis of frames in which each member can be modeled by one element in most cases. The formulation of the element is based on the imposition of compatibility conditions at end nodes as well as the satisfaction of equilibrium at midspan. The resulting expression for this new element is accurate in describing the force‐versus‐displacement relations at the element level, is simple, and it does not lead to a significant numerical truncating error in the computer analysis. Its implementation in a nonlinear analysis program is straightforward. The accuracy of the analysis results by the element was found to be considerably higher than its cubic counterpart in a number of well‐known examples. The limitation or inconvenience of the method of stability function such as the separation of the solution for tensile, compressive, and zero load cases is eliminated.

A new approach for the dimensioning of reinforced‐concrete structures, in particular wall and flat‐slab structures, so as to ensure structural safety against collapse is presented. The approach, which is based on the theory of plasticity, represents an extension of the finite‐element method from the analysis task to the dimensioning task: the reinforcement and the concrete are required to withstand not the stresses within each element (these do not even need to be evaluated), but the element nodal forces transmitted from the neighboring elements and acting on the element like external loads. Thus each finite element becomes an independent “dimensioning unit.” Alternatively, the new approach can be viewed as a generalization of the truss model approach, the finite‐element method being used not for finding a reliable stress distribution within the structure, but for building an equilibrium system of concentrated nodal forces to be used for dimensioning purposes. Only the basic principles of the new method are discussed in the paper. (A related software development project is under way and will be presented in future publications.)

The descending part of the hysteretic load‐deformation history of a structure with softening behavior discloses important structural properties such as strength, ductility, residual deformation, hysteresis loop shape, and energy‐dissipation capacity, especially under earthquake conditions. In the present paper, a finite‐element algorithm for hysteresis analysis of bonded and unbonded prestressed‐concrete frames subjected to arbitrary loading history including postultimate descending portion is put forward. One‐dimensional constitutive models for concrete, tendon, and reinforcing steel are proposed. The crack‐contact effect is included in the constitutive model for concrete. A bond‐slip relation between tendon and grout is suggested. A method for implementing the bond‐slip relation into the element in which tendons are embedded is also put forward. Being of self‐correcting incremental type, the algorithm incorporates both increment‐force and increment‐displacement procedures whose transition from one into the other is automatically performed according to the variations of the hysteresis curve. The results of laboratory tests of three prestressed‐concrete frames are reported. Comparison of computed results with experimental ones shows good agreement.

Bridge testing is rapidly becoming a major tool in evaluation. Despite improved structural analysis there are numerous cases in which field testing may be the best way to help ensure future safe performance. The reasons for testing include uncertainties in material and structural modeling and concerns for serviceability limit states. Switzerland has long experience in testing new bridges before allowing them to go into service. The general testing plan, the responses that are modeled, and an example are described. The aim in this paper is to extend these bridge‐testing methods to strength‐evaluation activity. The Swiss approach in testing offers a compromise between a diagnostic performance test and a proof‐loading check on capacity. The level of test loading is normally about 80–85% of the code‐specified serviceability or unfactored load, which is still significantly above the expected lifetime maximum traffic loading. The paper describes in detail a risk model appropriate to modeling the risks during the testing as well as the benefits in higher safety indices gained after the testing. An example is described in the paper.

This paper investigates the behavior and performance of open‐web truss‐moment framing system under cyclic loading. An open‐web truss‐moment frame of a four‐story building was designed as an ordinary moment resisting frame (OMRF) for this investigation. Three full‐scale, half‐span subassemblages from this building were tested under severe cyclic deformation to determine their hysteretic behavior. Based on the experimental results, a simple analytical model for inelastic cyclic behavior of open web truss frames was developed. This model was used to analytically evaluate the behavior and performance of the four‐story building subjected to several severe earthquake excitations. The system performs poorly because of buckling and fracture of the web members, which led to unstable hysteretic loops.

This paper presents a new design concept and procedure for seismic resistant special truss‐moment frames, which have been validated by full‐scale tests on full‐span truss‐column subassemblages. The new design concept results in increased economy and excellent inelastic response with stable hysteretic behavior of the special trusses in moment frames when subjected to severe earthquake motions. The new concept also has an added advantage that the structure can be retrofitted after a major earthquake by simply replacing the damaged web members in the special segments of floor trusses. The proposed system is an excellent and efficient seismic resistant framing system for building structures.

Previously derived stress‐strain relationships for compressed concrete confined by various quantities and arrangements of transverse reinforcement are used in cyclic moment‐curvature analyses of a range of reinforced concrete columns to derive design charts. The design charts permit the enhanced flexural strength of confined columns to be obtained. They also permit the quantities of transverse reinforcement required to achieve particular curvature‐ductility factors in the potential plastic‐hinge regions of reinforced concrete columns to be determined. The column section is considered to have reached its available ultimate curvature when either the moment resisted has reduced to 80% of the ideal flexural strength, or the strain energy absorbed in the transverse reinforcement has reached its strain energy absorption capacity, or when the logitudinal steel has reached its limiting tensile or compressive strain, whichever occurs first. Refined design equations to determine the quantities of transverse reinforcement required for specified ductility levels are derived on the basis of the design charts. The equations are an improvement on the current provisions of concrete design codes.

Simulated seismic load tests conducted on 11 reinforced concrete columns are described. The columns had a 400‐mm (15.7‐in.) square or octagonal cross section and contained various quantities of transverse reinforcement. The columns were subjected to either low, moderate, or high axial compressive loads and to reversible quasi‐state lateral loads that simulated severe earthquake loading. The tests were conducted to check an analytical approach for the determination of the available flexural ductility and the enhancement of flexural strength of columns with various levels of transverse confining reinforcement. The analytical approach for determining the quantities of transverse reinforcement required for confinement is shown to be conservative and to lead to equations more suitable for the design of transverse reinforcement for confinement than current code equations. Conclusions are also reached concerning the required length of the confined region in the potential plastic‐hinge zones of reinforced concrete columns.

Reliability indexes associated with typical short and slender columns are evaluated. Designs are made in accordance with ACI 318‐89 and Mexico City concrete design regulations. Loads associated with ACI 318‐89 designs are taken from ASCE 7‐88. The Monte Carlo simulation technique is used for the analysis. A general view of β indexes relative to different limit states (obtained previously by the first writer) is presented. The study is extended to β indexes relative to slender columns. It is shown that slender columns subjected to loads with small eccentricity ratios present the highest β‐reliability values, with respect to other limit states such as flexure, shear, and torsion. In all cases, RCDF‐87 β values are higher than those associated with ACI 318‐89. This is due to the following reasons: (1) The probabilities of exceedance of the maximum loads adopted for both codes are drastically different; (2) the codes define the concrete strength with different reduction factors; and (3) load and strength factors differ for both regulations.

The purpose of the present paper is to provide experimental data and analytical results for the tensile strength and ductility of welded wire fabric (WWF) so that engineers can confidently use WWF as structural reinforcement. The material for the tested specimens was provided by manufacturers in the United States, Germany, and Canada. The specimens included steel from rods (steel before the cold‐drawing process), plain wires, deformed wires, epoxy‐coated deformed wires, tempered and nontempered WWF, and nontempered wires without welds. The experiments conducted show that the WWF made in the United States can be considered a ductile steel based on the relevant ASTM specifications. In addition, the fatigue life of the epoxy‐coated WWF made in the United States exceeds those for reinforcement bars as tested by the Portland Cement Association. The results indicate that WWF is suitable as a structural reinforcement since it possesses sufficient ductility to meet ASTM specifications.

The purpose of the present paper is to provide experimental data and analytical results for the fatigue strength of welded wire fabric (WWF), so that engineers can confidently use WWF as structural reinforcement. The material for the tested specimens was provided by manufacturers in the United States, Germany, and Canada. The specimens included steel from rods (steel before the cold‐drawing process), plain wires, deformed wires, epoxy‐coated deformed wires, tempered and nontempered WWF, and nontempered wires without welds. The experiments conducted show that the WWF made in the United States can be considered a ductile steel based on the relevant ASTM specifications. In addition, the fatigue life of the epoxy‐coated WWF made in the United States exceeds those for reinforcement bars as tested by the Portland Cement Association. The results indicate that WWF meets the fatigue requirements of structural reinforcement for bridge decks.

The buckling strength of wood electric‐power transmission and distribution structures is examined using finite‐element analysis. The elastic and geometric stiffness matrices for a simple beam element are used to form a generalized eigen problem. The eigen problem is solved for the buckling strength of poles of constant diameter and various end fixities. There is good agreement with Euler's formula even for models consisting of relatively few finite elements. Critical buckling loads are also calculated for tapered poles with circular cross sections. The results agree closely with the buckling loads predicted by Gere and Carter's formulas. The buckling strength of a tapered pole is often approximated by using a pole of constant cross section with the cross section equal to that at one‐third the distance from the small end. This method is shown to be very conservative. The numerical analysis is extended to several practical examples including a single‐pole tangent structure, dead‐end structures guyed two ways and one way, multilevel guyed structures, and an H‐frame.

The theory is presented for a two‐layered one‐dimensional elastic member acting either as a beam or a beam‐column, including both connector and geometric nonlinearities. This member is capable of having slip at the interface of the two layers and displacements perpendicular to the longitudinal axis of the member that are on the order of magnitude of the depth of the member. The force‐slip relationship for the interlayer connectors is assumed to be nonlinear in accordance with accepted constitutive relationships for nailed joints. An energy approach is employed in the formulation from which the three coupled, nonlinear, ordinary differential equations governing the behavior of the member are derived; the associated boundary conditions are also derived through variational procedures. Tests of two‐layered wood beams and beam‐columns are conducted to verify the theory and very good agreement is achieved for deflections, slip between layers, and normal strains.