Visual experience is a powerful pedagogic tool. Extensive use of experimental studies prior to design and construction has made conceptualization of complex structures possible. Experiments on reduced-scale structures and specimens are also vital tools for teaching structural mechanics. As such, the Department of Civil Engineering at Virginia Tech has acquired a new apparatus for use as an educational demonstration tool in the area of structural mechanics. This work presents the results of a detailed study on the performance of this device as related to its accuracy and operation. To fulfill such objectives, two structural models (a continuous beam and a portal frame) were extensively tested under several loading and support configurations. The models were analysed using STRUDL as well as a computer program developed by the author. The comparison of the results (deformations) obtained in the two phases of the study have indicated that the apparatus is reasonably accurate to meet the requirements of a structural teaching model and adapting to a variety of structural models. / Master of Science / incomplete_metadata

This thesis investigates the performance of composite frame structures with smart partially-restrained (PR) concrete filled tube (CFT) column connections through simplified 2D and advanced 3D computational simulations. It also provides a design methodology for new types of innovative connections based on achieving a beam hinging mechanism. These types of connections intend to utilize the recentering properties of super-elastic SMA tension bars, the energy dissipation capacity of low-carbon steel bars, and the robustness of CFT columns. In the first part of this study, three different PR-CFT connection prototypes were designed based on a hierarchy of strength models for each connection component. Numerical simulations with refined three dimensional (3D) solid elements were conducted on full scale PR-CFT connection models in order to verify the strength models and evaluate the system performance under static loading. Based on system information obtained from these analyses, simplified connection models were formulated by replacing the individual connection components with spring elements and condensing their contributions. Connection behavior under cyclic loads was extrapolated and then compared with the monotonic behavior. In the second part of this study, the application of these connections to low-rise composite frames was illustrated by designing both 2D and 3D, 4 and 6 story buildings for the Los Angeles region. A total of 36 frames were studied. Pushover curves plotted as the normalized shear force versus inter story drift ratio (ISDR) showed significant transition points: elastic range or proportional limit, full yielding of the cross-section, strength hardening, ultimate strength, and strength degradation or stability limit. Based on the transition points in the monotonic pushover curves, three performance levels were defined: Design Point, Yield Point, and Ultimate Point. All frames were stable up to the yield point level. For all fames, after reaching the ultimate point, plastic rotation increased significantly and concentrated on the lower levels. These observations were quantified through the use of elastic strength ratios and inelastic curvature ductility ratios. The composite frames showed superior performance over traditional welded ones in terms of ductility and stability, and validated the premises of this research.

Curved and skewed I-girder bridges exhibit torsional displacements of the individual girders and of the overall bridge cross-section under dead loads. As a result, the girder webs can be plumb in only one configuration. If the structure is built such that the webs are plumb in the ideal no-load position, they generally cannot be plumb under the action of the structure's steel or total dead load; hence, twisting of the girders is unavoidable under dead loads. The deflected geometry resulting from these torsional displacements can impact the fit-up of the members, the erection requirements (crane positions and capacities, the number of temporary supports, tie down requirements, etc.), the bearing cost and type, and the overall strength of the structure. Furthermore, significant layover may be visually objectionable, particularly at piers and abutments. If the torsional deflections are large enough, then the cross-frames are typically detailed to compensate for them, either partially or fully. As specified in Article C6.7.2 of the AASHTO LRFD Specifications, different types of cross-frame detailing methods are used to achieve theoretically plumb webs under the no-load, steel dead load, or total dead load conditions. Each of the cross-frame detailing methods has ramifications on the behavior and constructability of a bridge. Currently, there is much confusion and divergence of opinion in the bridge industry regarding the stage at which steel I girder webs should be ideally plumb and the consequences of out-of-plumbness at other stages. Furthermore, concerns are often raised about potential fit-up problems during steel erection as well as the control of the final deck geometry (e.g., cross-slopes and joint alignment). These influences and ramifications of cross-frame detailing need to be investigated and explained so that resulting field problems leading to needless construction delays and legal claims can be avoided. This dissertation addresses the influence of cross-frame detailing on curved and/or skewed steel I girder bridges during steel erection and concrete deck placement by conducting comprehensive analytical studies. Procedures to determine the lack-of-fit forces due to dead load fit (DLF) detailing are developed to assess the impact of different types of cross-frame detailing. The studies include benchmarking of refined analytical models against selected full scale experimental tests and field measurements. These analytical models are then utilized to study a variety of practical combinations and permutations of bridge parameters pertaining to horizontal curvature and skew effects. This research develops and clarifies procedures and provides new knowledge with respect to the impact of cross-frame detailing methods on: 1) constructed bridge geometries, 2) cross-frame forces, 3) girder stresses, 4) system strengths, 5) potential uplift at bearings, and 6) fit-up during erection. These developments provide the basis for the development of refined guidelines for: 1) practices to alleviate fit-up difficulties during erection, 2) selection of cross-frame detailing methods as a function of I-girder bridge geometry characteristics, and 3) procedures to calculate the locked-in forces due to DLF cross-frame detailing.