Hardness Variation in Driven Rivets for Bridge Evaluation: Fid-Bau Portal

Hardness Variation in Driven Rivets for Bridge Evaluation

Hebdon, Matthew H. / Sherman, Ryan J. / Connor, Robert J.

Analysis and load rating of historic, riveted steel structures typically requires an assumption of rivet shear strength. The AASHTO Manual for Bridge Evaluation (MBE) provides rivet shear strengths based on the year of construction; however, the values provided are more conservative than legacy test data. Following the 2007 collapse of the I-35W bridge in Minneapolis, MN, riveted gusset plate connections, primarily on older structures, became a national concern. Many owners found using the MBE prescribed shear strength values resulted in some connections not satisfying load rating requirements. In such cases, some have elected to remove a number of rivets and perform destructive laboratory testing to obtain actual material properties. This alternative can be an expensive and time-consuming process. As an alternative, some have performed hardness testing on in-situ rivets to estimate the strength of the rivet using known hardness-to-ultimate strength correlations. The following study was performed to evaluate the variability of hardness throughout a rivet to establish if in-situ hardness measurement and correlation with the hardness of the shank is a reasonable method to estimate the strength of a rivet. Hardness testing is an economical and efficient process easily conducted in the field using handheld devices. However, questions have arisen regarding whether measurements made on the surface of the driven rivet are representative of the hardness of the shank, which was not mechanically worked as compared to the button head or head that was created during the driving process. For the current study, rivets were removed from seven structures of varying age and the hardness was measured. In addition, modern driven and undriven rivets were tested to compare historic results to modern materials. Hardness was measured on both the shop and driven rivet heads as well as at several locations throughout the shank of the rivet and compared to historical specification requirements. Variation in hardness throughout the rivet was analyzed to determine if the hardness measurement on the head could be correlated with the rivet shear strength. Measured rivet head hardness was found to have good correlation with hardness measurements throughout the rivet shank.