Seismic design issues of the panel zone in special moment frames with deep-column sections

Abstract

Deep-column sections (deeper than W14 sections) are increasingly used in moment frames for numerous reasons. Our previous study indicated that use of the deep-column might have economical benefit while there is no significant difference in performance. However, concerns have been raised about the issues related to the panel zone in the deep-column moment frame. Numerous investigations, before and after the 1994 Northridge Earthquake, have shown that the panel zone plays an important role in seismic performance of beam-to-column connections in a moment frame. The current American Institute of Steel Construction (AISC) Seismic Provisions (ANSI/AISC 341-02) for panel zone were based on the studies using typical W14 column depth in the beam-to-column connection assemblage. A limited number of experimental and numerical studies on deep-column connections revealed some special features associated with the deep-column section, such as potentially higher slenderness ratio in the panel zone. This paper presents a numerical study on panel zone inelastic behavior. Beam-column sub-assemblages with a range of deep-columns from W27 to W33 were subjected to ever-increasing story drift, and yielding strength and the strength at the ductility of four (used as the design strength by ANSI/AISC 341- 02) of the panel zone were compared with conventional W14 columns, as well as with the design provision in ANSI/AISC 341-02. The study found that the current seismic design strength, primarily based on the tests of light W14 sections, overestimates the panel zone strength of the heavy W14 sections by 20% at yielding and at ductility of 4. However, such overestimation becomes smaller particularly at large inelastic panel zone deformation. It appears that the strain hardening in a deep-column panel zone is not as significant as in a W14 column. While recommending further experimental study to confirm the findings, the authors have not observed any adverse impact to use deep-columns in terms of panel zone behavior, and the current design strength for panel zone, although less conservative, has a better correlation with analytical results than that for W14 columns. © 2012 Elsevier B.V., All rights reserved.