Low-Cycle Fatigue Fracture Prediction of Grade 40 Reinforcing Steel Bars in Concrete Bridge Columns

Levi Arnold

doi:10.7273/000005141

The Washington Department of Transportation (WSDOT) has identified concrete bridge columns designed and constructed pre-1974 with Grade 40 steel reinforcement to be vulnerable to significant damage from large earthquakes, including a possible Cascadia Subduction Zone (CSZ) earthquake. This has prompted the development of a seismic retrofitting program by WSDOT to increase the seismic performance of these columns to preserve functionality and responsiveness of life-line routes throughout western Washington. However, the retrofitting strategy used of encasing the vulnerable columns with a steel jacket puts additional cyclic fatigue demand on the reinforcement steel. This research focused on generating fracture predictions of Grade 40 reinforcement for CSZ earthquakes to provide WSDOT with location-specific predictions of bar fracture to aid them in prioritization and planning of their retrofitting efforts. Previous literature has studied low-cycle fatigue failure predictions for constant and variable strain amplitude tests which has led to the development of the Koh-Stephens fatigue model. However, this model is very sensitive to the calibration parameters that are specific to the unique steel that was tested. This project conducted additional constant and variable amplitude reversed cyclic tests on Gr. 40 reinforcement bar to validate Koh-Stephens model parameters, Palmgren-Miner’s damage rule, and to estimate model variability. Then, a numerical model was created that accounts for variability and uncertainty in the calibration parameters for different heats and bar sizes of steel that could be encountered in the field using Monte-Carlo simulations. The numerical model was analyzed using longitudinal reinforcement strain histories from 60 ground motions, across 4 locations in western WA, for 8 prototype bridge columns at 5 bridge periods. The four locations analyzed represent the four different major site characteristics found in western WA of inland or coastal and basin or no-basin. Results of the fracture prediction model determined that the probability of fracture increased in locations of deep basins. It was also found that the affected period range of high probabilities of fracture was wider for locations with basins. Column design also affected the probability of fracture with columns having low reinforcement ratios and high axial load ratios resulting in greater probability of fracture.

Low-Cycle Fatigue Fracture Prediction of Grade 40 Reinforcing Steel Bars in Concrete Bridge Columns

Files and links (1)

Abstract

Metrics

Details