The 14
th
World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China
EXPERIMENTAL INVESTIGATION OF PROGRESSIVE COLLAPSE OF
STEEL FRAMES UNDER MULTI-HAZARD EXTREME LOADING
Antonis Tsitos
1
, Gilberto Mosqueda
2
, André Filiatrault
3
and Andrei M. Reinhorn
4
1
Ph.D. Candidate, Dept. of Civil, Structural & Environmental Engineering, University at Buffalo-The State University of
New York, Buffalo, NY, USA - Email: atsitos@eng.buffalo.edu
2
Assistant Professor,
3,4
Professor, Dept. of Civil, Structural & Environmental Engineering, University at Buffalo-The State
University of New York, Buffalo, NY, USA
ABSTRACT :
Experiments were performed to evaluate the progressive collapse resistance of steel buildings considering
multi-hazard extreme loading. Two 1/3 scale three-story, two-bay steel frames, a special moment resisting frame
and a post-tensioned energy dissipating frame designed and previously tested for seismic performance on a shaking
table, were adapted for quasi static collapse testing. The experiments simulated the structural response after the
sudden failure of a column. The objective of the tests was to evaluate the effectiveness of earthquake resistant design
details in enhancing the resistance to progressive collapse. In these tests, an effort was made to document the load
resisting mechanism, the sequence of damage in the frames and correlate observed damage with changes in the
resisting strength. The experimental results demonstrate significant vertical displacement capacity for both frame
systems and the capacity to redistribute loads after the failure of a single column. However, the steel moment frame
appears to have significantly more ductile response by maintaining its yield strength up to the point of first
connection fracture. The vertical load carrying capacity of the special moment frame appears to be dependent on the
rotation capacity of plastic hinges before buckling and fractures occur. The vertical load carrying capacity of the
post-tensioned energy dissipating frame depends heavily on the performance and ultimate strength of the tendons
used for connecting the beams to the columns. The PTED frame lost a significant amount of strength after failure of
one of the tendon strands.
KEYWORDS:
Experimental study, progressive collapse, steel frames,
post-tensioned frames, multi-hazard loading.
1. INTRODUCTION
Since the collapse of the Ronan Point Towers in London 1968 and more recently spurred by the September 11
2001 events at the World Trade Center in New York City, the structural engineering community has focused
significant efforts towards better understanding of the phenomena of progressive collapse in building structures.
The ultimate goal is to establish rational and reliable methods for the assessment and the enhancement of
structural resistance to extreme accidental events. Although design methodologies (such as the alternative path
method and the tie force method) and analysis procedures to enhance resistance to progressive collapse are
proposed in guideline documents issued by the U.S. General Services Administration (2003) and the Department
of Defense (2005), there is a scarcity of experimental data to support the numerical modeling of building
structures under extreme loads, particularly to the point of failure.
Several numerical studies have been published investigating the progressive collapse of buildings (Kaewkulchai
and Williamson, 2004; Bazant and Verdure, 2007) and the adequacy of commercially available structural analysis
software to perform collapse analyses (Marjanishvili and Agnew, 2006). The dynamics of impact of collapsing
failed elements has also been investigated by Kaewkulchai and Williamson (2006). In a different approach to
capture the dynamics of impact and other phenomena involved in a progressive collapse, Sivaselvan and Reinhorn
(2006) proposed a method using a Mixed Lagrangian Formulation (MLF). In most cases, it is attempted to predict
the global response of the building using simplifying assumptions without explicitly considering detailed but
important phenomena, such as axial-flexure-shear force interaction in the beams, joint failure, local buckling and
panel zone deformation (for steel structures). To address this need, Khandelwal et al. (2008) and Bao et al. (2008)
proposed macro-models of beam-column subassemblies that capture the key local and global response