Search

This paper presents the results from an experimental pushover test on a one-third-scale model of a special inverted-V- braced steel frame with zipper struts. Zipper elements are vertical elements added at the intersections of the braces above the first floor and designed to carry upward the unbalanced loads resulting from buckling of the braces. As far as the writers know, this is the first test on such a configuration that has been suggested as an alternate to conventional braced frames in the American Institute of Steel Construction seismic specification for many years. The model was pushed to a target roof drift ratio of +3.58%, which was achieved without strength degradation and followed a trilinear backbone curve. The load was then reversed until the bottom story brace in tension fractured at a drift of -0.78%. After this fracture, the frame still carried about 37% of the maximum base shear. The zipper elements demonstrated their ability to activate buckling in all stories except the top one, redistributing the loads in the structure and minimizing strength losses. Two-dimensional and one three-dimensional frame models were analyzed and found to reproduce satisfactorily the experimental results up to the beginning of tearing in the tension brace. DOI: 10.1061/(ASCE)0733-9445(2008)134:10(1619) CE Database subject headings: Steel frames; Seismic design; Bracing; Struts.

The 2005 AISC specification provides a new method for stability design of steel frame structures termed the direct analysis method (DM). This paper addresses the consideration of partial base fixity using the DM. A refined model of the column base moment-rotation response is developed using an adaptation of the component method (CM) of Eurocode 3 plus a representation of the foundation stiffness. The paper also summarizes and applies an extension of the DM for frames containing web-tapered members. An example clear-span gable frame is checked using several models of its nominally simple four-bolt base detail: ideally pinned, elastic based on G = l0 from the AISC sidesway-uninhibited alignment chart, and elastic-perfectly plastic using the CM. It is demonstrated that the member strength unity checks and the frame deflections obtained by modeling of the column bases as elastic based on G = l0 are accurate to slightly optimistic compared to the refined base model. It is found that the contribution of the significant initial stiffness of typical bases to the overall frame response is limited by the connection strength. DOI: 10.1061/(ASCE)0733-9445(2008)134:9(1508) CE Database subject headings: Stability; Steel frames; Connections; Columns; Nonlinear response.