Master of Science / Department of Architectural Engineering and Construction Science / Kimberly W. Kramer / Dual Lateral Force Resisting Systems are currently required by code to include a Moment Resisting Frame capable of resisting at least 25% of the lateral loads. This thesis evaluates the seismic performance of a specific type of dual system: a Special Moment Resisting Frame-Special Concentric Brace Frame System (SMRF-SCBF) under three different force distributions. The three distributions were 80% - 20%, 75% - 25%, and 70% - 30% with the lesser force being allotted to the Special Moment Resisting Frame (SMRF) portion of the system. In order to evaluate the system, a parametric study was performed. The parametric study consisted of three SMRF-SCBF systems designed with different seismic force distributions. The aim of this study was to determine accuracy of the three different seismic force distributions. The accuracy was measured by comparing individual system models’ data and combined system models’ data. The data used for comparison included joint deflections (both horizontal and vertical), induced moments at moment connections, brace axial loads, column shears, and column base reactions. Two-dimensional models using the structural software RISA 3D were used to assist in designing the independent Seismic Force Resisting Systems. The designs of the frames were not finely tuned (smallest member size for strength), but were designed for drift (horizontal deflection) requirements and constructability issues. Connection designs were outside the scope of the study, except for constructability considerations – the SMRF and the SCBF did not have a common column; the frames were a bay apart connected with a link beam. The results indicated that a seismic force distribution of 75% to the SCBF and 25% to the SMRF most accurately predicts that frame’s behavior. A force distribution of 80% to the SCBF and 20% to the SMRF resulted in moderately accurate results as well. A vast opportunity for further research into this area of study exists. Alterations to the design process, consideration of wind loads, or additional force distributions are all recommended changes for further research into this topic.

Three end-plate moment connection configurations are prequalified for special moment frames for seismic applications in AISC 358-10. The eight bolt extended stiffened connection is the strongest of the three configurations, but it can only develop approximately 30 percent of currently available hot-rolled beam sections. The strength of this configuration is limited by bolt strength. There is a need for a stronger end-plate moment connection, hence the reason for the development and validation of a twelve bolt configuration. Equations were developed for the design procedure using various analytical methods, which included yield line analysis and an effective tee stub model. An experimental program was conducted, which consisted of the full-scale cyclic testing of four end-plate moment connections. The intention of the testing was to develop and validate the design procedure, and prequalify a new twelve bolt configuration. A displacement-controlled loading protocol was applied according to AISC 341-10. The experimental results showed that the model for thick end-plate behavior is conservative by 6.7%, the model for end-plate yielding is conservative by 8.8%, and the model for bolt tension rupture with prying conservatively predicts by 18.5%. The specimens that were designed to form a plastic hinge in the beam fractured in a brittle manner. The deep beam specimen fractured in the first 2% story drift cycle, and the shallow beam specimen fractured in the second 3% story drift cycle. The fracture of the prequalification specimens was determined to have been caused by stiffeners of high yield stress relative to the beam yield stress. / Master of Science / End-plate moment connections are a common way to create a rigid joint between beams and columns. Before using a moment connection in a steel building to resist horizontal earthquake loads, each connection configuration must be tested at full-scale and meet performance criteria prescribed in the applicable building code (in this case, the Seismic Provisions for Structural Steel Buildings published by the American Institute of Steel Construction). Three end-plate moment connection configurations have been previously “prequalified” for high seismic regions, which means that sufficient previous testing has shown adequate performance. The eight bolt end-plate moment connection is the strongest of the three configurations, but it can only develop approximately 30 percent of currently available hot-rolled steel beam sections. The strength of this configuration is limited by bolt strength. There is a need for stronger end-plate moment connections, which motivated the development and validation of a twelve bolt configuration in this thesis. Equations were developed for the design of the twelve-bolt end-plate moment connection including equations to predict when the bolts would fracture and when the end-plate would yield. An experimental program was conducted, which consisted of the full-scale cyclic testing of four end-plate moment connections. The intention of the testing was to validate the design procedure and demonstrate that the connection could withstand significant inelastic rotation. The connection assembly was cycled back and forth according to a displacement protocol prescribed in the Seismic Provisions for Steel Buildings. The experimental results showed that the equations were able to predict bolt rupture within 6.7% of the applied moment at fracture, the equation for end-plate yielding was conservative by 8.8%, and the equation for bolt fracture with prying action was conservative by 18.5%. The specimens that were intended to show the connection could withstand significant inelasticity fractured in an unexpected brittle manner. The deep beam iv specimen fractured in the first 2% story drift cycle, and the shallow beam specimen fractured in the second 3% story drift cycle, neither of which reach the target of 4% story drift. The fractures were determined to have been caused by stiffeners that had too high a yield stress relative to the beam yield stress.