Currently two commonly used restrainer design methods are those mandated by the American Association of State Highway Transportation Officials (AASHTO) and the California Department of Transportation (CALTRANS). To evaluate these methods and perhaps develop a new method, the Washington State Department of Transportation (WSDOT) sponsored this study.
Using sample WSDOT designs and seismic retrofitting guidelines from WSDOT and CALTRANS, the researchers developed a model of a two-frame bridge with a single in-span hinge. The nonlinear response of the bridge was studied to determine the maximum opening experienced at the in-span hinge and the maximum relative abutment displacements.
To identify the parameters most important in restrainer design and in predicting the unrestrained maximum relative abutment displacements, the researchers varied eleven parameters. The parametric study identified the parameters that significantly influenced the maximum relative hinge displacement (MRHD) and the maximum relative abutment displacements (MRAD).
The AASHTO empirical seat width equation and the CALTRANS restrainer design methods were compared with the results of nonlinear time history analysis. The empirical seat width equation produced conservative results while the CALTRANS method produced inconsistent results, a large amount of scatter and some significantly unconservative values.
Using the results of the parametric study, the researchers developed a new restrainer design method that predicted the MRHD much more accurately than the CALTRANS method. The researchers also developed a method for estimating the unrestrained MRAD.