Analysis of fatigue prone details on a continuous steel girder bridge

The most important parameters in fatigue life evaluation of existing steel highway bridges are the detail type, the truck traffic volume, the number of stress cycles per truck, and the effective stress range. The uncertainties in predicting remaining fatigue life can be attributed to any one of these parameters. Field testing of a six lane long span continuous highway girder bridge under normal traffic was conducted to determine the number of stress cycles per truck passage for various fatigue stress category details in both main and secondary members; to evaluate the effect of cycle counting methods, stress-range frequency of occurrence histogram construction, and field test duration on the effective stress range and the number of cycles per truck; and based on field observations, develop a formula for the present truck volume. Analysis of field test results revealed that the provisions of the AASHTO Guide specifications with respect to these parameters do not reflect actual conditions on a bridge.

A refined remaining fatigue life estimation procedure, based on the widely accepted method of comparing the number of cycles to date with the number of cycles to failure, was developed. A comparison of remaining fatigue life estimates based on the AASHTO Guide procedure and on the refined procedure showed that the refined procedure gave a more accurate estimate of the actual remaining fatigue life than the AASHTO Guide procedure.

As part of this research an impact study was conducted. Crawl and full speed test data were analyzed to examine the influence of bridge approach, surface condition, and truck velocity on impact factors for fatigue prone details in both main and secondary bridge members. The measured impact factors for fatigue stress categories A, B, E, and E' details were compared to values obtained using the AASHTO Standard specification formula. For fatigue prone details, the test results showed that surface roughness, truck velocity, and bridge approach affect the impact factor. Measured impact factors varied with the location of the detail. Also, the study showed that the AASHTO design impact factors were unconservative for details located in the approach (end) span.