Response of Nonductile RC Frame Systems and Components

John Wallace
Associate Professor
Department of Civil and Environmental Engineering
University of California, Los Angeles

A major component of the PEER Center research involves reducing risk in large urban earthquakes through development and application of performance-based design methodologies. A research program that includes nineteen integrated research tasks was established in Year 1 to begin work on achieving this goal. Among these tasks is a study of older reinforced concrete (RC) building frame systems and components, commonly referred to as "nonductile" RC frames. Reinforced concrete structures constructed before the mid-1970s have characteristics that can result in failure during earthquakes.

PEER research on older concrete buildings is focused, in the first year, on column behavior. The study includes five coordinated experimental and analytical projects involving research teams from southern, central, and northern California (UCLA; California Polytechnic State University, San Luis Obispo; and UC Berkeley), and the University of Washington. Additional research facets are to be incorporated in upcoming years.

The behavior of columns and beam-column connections are of particular interest due to poor performance of these components in laboratory studies (Fig. 1)

Fig. 1. Failure of a column specimen with details typical of older reinforced concrete columns

and in Northridge (fig. 2).

Fig. 2. Five-story RC Kaiser Permanente building, Northridge, California. Photo by Mark Aschheim. Northridge Collection, Pacific Earthquake Engineering Research Center, University of California, Berkeley.

Deficiencies that often characterize older reinforced concrete construction include (1) insufficient transverse reinforcement to confine the column core and to restrain buckling of longitudinal reinforcement; (2) inadequate lap splices located immediately above floor levels where inelastic actions may be concentrated; (3) insufficient shear strength to develop the column flexural capacity, or the potential degradation of column shear strength with increasing flexural ductility demand; (4) inadequate column strength to develop a strong-column, weak-beam mechanism; and (5) deficient joint dimensions and details.

The primary objectives of the PEER Center research task on older RC buildings are to characterize component strengths and deformation capacities, and to assess the impact of near-field ground motions on system and component behavior. An emphasis of Year 1 work is to assemble and evaluate the existing database of column test data to fill critical gaps in these data. Detailed evaluation of data and experimental studies of columns will serve as the basis for statistical information on parameters affecting column strength and deformation capacity. Directed by Professor Marc O. Eberhard at the University of Washington, this project will provide a vital link in developing a performance-based framework for older RC buildings and in planning future experimental efforts.

Two experimentally oriented projects initiated at UCLA and UC Berkeley are also addressing gaps in data on columns. Both studies include testing of large-scale columns subjected to reverse curvature bending, in combination with general analytical studies. Both projects include significant involvement from PEER's Business-and-Industry Partners. The study at UC Berkeley, under the supervision of Professor Jack P. Moehle, builds on previous work (Lynn et al. 1996) and is being conducted in cooperation with Assistant Professor Abe Lynn at California Polytechnic State University, San Luis Obispo. Rutherford and Chekene, Consulting Engineers, San Francisco, are assisting in the study. The study parameters include the impact of short-duration loading, biaxial loading, and variations in column details on the shear strength and deformation capacity of columns. Professors John W. Wallace and Joel P. Conte are directing work investigating short duration loading, axial load level, biaxial response, and variations in column details on strength, and capacity of columns with lap splices. Data from the UCLA and UC Berkeley studies will be incorporated into the University of Washington study.

Detailed analytical studies are being carried out in parallel with the experimentally focused work. At UCLA, Professor Conte is leading an effort to develop nonlinear finite element (damage) models that capture experimental responses observed at UC Berkeley and UCLA. The models are being developed to allow the work to be incorporated into the larger PEER Center analytical platform being developed in a parallel research task, Next Generation Analytical Platform for Nonlinear Dynamic Analysis.

Analytical studies of overall system responses are also being conducted in a research project led by Professor Gary C. Hart at UCLA. The system studies will establish component demands for a variety of building configurations and ground motions as well as assess how changes in individual parameters (i.e., component strengths or stiffness values) influence demands and the distribution of damage.

International collaboration has been initiated to broaden the research program on the response of nonductile frame systems and components. Current efforts include collaboration with researchers in Argentina, Japan, and Mexico. The coordinated research program builds on previous research and Center strengths in analytical and experimental research and prior research activities to develop performance-based tools for older RC buildings. Updates featuring abbreviated research findings will appear in future newsletters.

Lynn, A. et al. 1996. Seismic evaluation of existing reinforced concrete building columns. Earthquake Spectra 12 (4) (November): 715-39.