Journal of Structural Engineering

This paper summarizes studies in which the effects of masonry and lightweight concrete infills on R/C moment existing frame buildings were studied experimentally and analytically. The experimental investigation consisted of a series of quasi‐static cyclic and monotonic load tests on 1/3‐scale models of the lower 3‐1/2 stories of an 11 story‐three bay reinforced concrete frame infilled in the outer two bays. Different panel material and reinforcement combinations were tested. For reasons of economy, ease of construction, favorable mechanical properties, and efficiency of different types of masonry infill, it was concluded that the most promising panel configuration consisted of solid brick laid in mortar reinforced with two mats of welded wire fabric, one bonded to each side of the wall in a layer of cement stucco (mortar). The implications of these experimentally obtained results are analyzed by investigating how the infills affect the dynamic response of R/C moment resisting frame buildings, as well as considering the effect of these implications on design of new buildings, and retrofitting of existing buildings located in regions with differing seismic risk.

This paper reports the strength and behavior of various diameter J‐shaped bolts embedded in concrete masonry. The bolts were tested monotonically, in direct tension, direct shear, and in combined shear and tension. The results of the monotonic tests show that for the combined shear and tension and for tension tests, strengths of the embedments increased with bolt size up to the 3/4‐in. (19 mm) diameter. For larger diameter bolts, the masonry began to fail resulting in approximately equal strengths for increasing bolt diameter. The strength of bolts in shear was found to increase with bolt size as long as the bolt was pretightened sufficiently.

Concrete bearing stress test results on large embedded inserts subjected to shear failure loads are presented. Tests were performed on eight individual inserts that were cast into a prototype concrete slab reinforced similarly to the concrete floor slab in the Gas Centrifuge Enrichment Plant (GCEP) at Piketon, Ohio. Analyses were made to examine the rigidity of the insert as a beam on an elastic foundation. These analyses, which agreed with field measurements, indicate that an insert rotates as a rigid body. With this knowledge and assuming that the ACI rectangular stress block exists in front of the insert, the maximum bearing stress is computed and compared to $f_c^{\prime }\text{.}$ The results indicate that the average bearing stress on the concrete at the failure load is $3.78f_c^{\prime }\text{.}$

Present methods for approximating the elastic lateral buckling load of a continuous beam are unnecessarily conservative when the beam is loaded by concentrated moments. This is because the significant warping restraints which may develop at points of concentrated moment are neglected. In this paper, an improved approximate method is developed which allows for these warping restraints. The method uses an effective length concept to allow for the warping and minor axis flexural interactions between adjacent segments at brace points and supports. The warping restraint effective length factors are estimated from the concentrated moments and the anti‐symmetry of the bending moment distribution about the brace point, and initial estimates are made of the buckling loads of the individual segments. These estimates are then improved by allowing approximately for the flexural interaction between adjacent segments. The accuracy of the new method is demonstrated for a wide variety of cases.

The Queensboro Bridge in New York City is presently undergoing major rehabilitation to its road surfaces and wind bracing system. A three dimensional computer model of this bridge was developed since significant cost savings were feasible if accurate analyses of the structure could be provided during the 15‐yr rehabilitation period. The size of the model was reduced considerably by simplifying ornate portal frames and neglecting the contribution of the floor systems (upper and lower) to the torsional rigidity of the entire structure. The accuracy of the model was verified by comparison to previously computed geometry and member stresses. The first use of the three dimensional model resulted in providing the contractor maximum flexibility in conducting his work and thus should reduce his costs.

This paper describes an investigation of buckling and post‐buckling behavior of steel beam‐columns. Beam‐columns having a H‐shaped cross section were studied for their deformability as well as ultimate strength. A constant axial thrust and monotonically increasing end moments were applied to the beam‐columns. Both experimental and analytical investigations were made in the study. Variables in the experimental study were cross‐sectional shape, length, and intensity of axial force. Numerical analysis demonstrated its capability to properly simulate the behavior of the tested beam‐columns. On the basis of parametric study using the numerical analysis, important findings can be drawn. First, lateral torsional deflection is more distinguished for beam‐columns with less axial force. Second, beam‐columns having a small slenderness ratio do not lead to instability by lateral torsional buckling, indicating a stable plastic range. Third, deformability drastically decreases once the slenderness ratio exceeds a certain level. Available formulas which estimate the ultimate strength of beam‐columns, in addition, are found to be effective.

The mechanical performance of built‐up roofing membranes, fully bonded to an underlying deck or substrate was modeled. Both linear and nonlinear stress‐strain behaviors were considered in the model development. The model is compared to previously developed models. It was found that the equality of the complementary strain energy of the fabric or felt layer with the strain energy of the bonding adhesive or bitumen layer governs both the conditions under which membrane integrity is lost and the mode of failure. Failure can occur either by membrane splitting or adhesive disbonding. The testing criteria developed are applied to a sample case.

The post‐buckling behavior of a simple column guyed in a plane by symmetric guy wires is considered. The column is built‐in at the ground, and the structure is loaded by a single vertical force W. Consideration is given to both the nonlinear behavior of a post‐buckled column and to the inherently nonlinear behavior of extensible guy wires subjected to their own weight. Numerical results are presented for the column alone and for the system consisting of a particular column with particular guy wires with varying degrees of pretension. The buckling behavior is found to depend strongly on the amount of pretension in the guy wires.

Temperature distribution in composite box girder bridges is predicted from the diurnal variation of the solar radiation to which the structure is exposed. Parametric studies are conducted in order to find the extreme temperature differences developed. Selected cases of temperature distributions are used as input for analyses to find temperature‐induced stresses in a two span continuous structure.

The equivalent frame method rules in ACI Building Code Requirements (ACI 318‐77) are only applicable to single story gravity loaded frames. New rules are proposed for the effective stiffness of the equivalent beam in beam‐slab buildings, and for the effective width of slab in flat plate floors. The proposed rules, derived from parametric studies using finite element techniques, are suitable for gravity and laterally loaded buildings, are simpler to use than the current procedures and eliminate the artificial device of the equivalent column. A laterally loaded beam and slab building is analyzed using both methods and the results compared. Also, a model flat plate building is analyzed and the predicted sway deflections, by the proposed rules, are in close agreement with the measured deflections.

Some aspects of analysis and design of protective structures are discussed. It is demonstrated that parametric analyses can be used to obtain an approximate model for reanalysis. The latter describes the major phenomena involved in the problem; i.e., those that most influence the design variables. Examples are used to describe the approximations, and an optimization approach is suggested. The suggested design process is based on a multilevel scheme involving a few separate design spaces, each of which contains a class of variables of the same type. The variables are classified as cross‐sectional, structures dimensions, and relative location (distances) variables. Using equivalent charge, ideal load histories and response functions, and the multilevel scheme methods of optimization for each level are chosen. Examples are used to test the suggested approach, which is shown to supply an efficient design tool. Also included is a discussion of possible further use of the approach for different problems.

The growing use of continuous bridge decks, as a structural type, has demanded more sophisticated analytical methods to investigate static behavior under live loads. These methods are based, essentially for the beamand‐slab decks, on the analysis of computer aided equivalent grillages. Solutions in close form are not available since harmonic analysis is sound only for simply supported bridge decks. But, if a differential equation system describing deck behavior is studied by modal analysis techniques it is possible, in two limiting cases, to give a solution in close form made up of a set of uncoupled differential equations for a set of independent beams on elastic foundation. Then, it is easy to write compatibility equations for continuous decks. These two cases occur when longitudinal beams have either null or infinite torsional rigidity. Nevertheless they are of great interest in practice since the true cases can generally be, with enough accuracy, assumed as one or the other limiting case. A numerical example shows a continuous bridge deck analyzed by hand calculations.