Affiliations: [a] L. Raymond and Associates, Irvine, CA, USA | [b] T.Y. Lin International, San Francisco, CA, USA | [c] Dominion Engineering, Inc., Reston, VA, USA | [d] California Department of Transportation, District 4, Oakland, CA, USA | [e] Alta Vista Solutions, Richmond, CA, USA
Correspondence:
[*]
Corresponding author. W. Crumly, L. Raymond and Associates, Irvine, CA, USA. Tel.: +714 394 3628; Fax: +949 474 9807; [email protected]
Abstract: After tensioning high-strength steel rods installed on the lower housing of Shear Key S1 and Shear Key S2 at Pier E2 of the San Francisco-Oakland Bay Bridge, 32 of 96 threaded rods failed. The rods were tensioned to 0.7 Fu and the failures were delayed failure, from 3 days to 14 days after tensioning. Failure analysis found that failure was due to Hydrogen Embrittlement. A six part test program performed to evaluate safety of additional rods in the field. This paper reports on Test V and Test VI – Small Threaded Specimen Testing. Small (Charpy-sized) threaded specimens were used to predict the threshold Load for the onset of hydrogen induced stress corrosion cracking of large A354BD Rods. The procedure used was to measure the hydrogen embrittlement threshold load for a small specimen and use a validated model to calculate the corresponding threshold stress intensity factor from the threshold load. With this threshold stress intensity factor, a model was used to calculate threshold load of the full size rod. Three independent methods were used to validate the accuracy of using small specimens to predict full-size failures. 1) Use finite element analysis to relate the local conditions at the thread root for a sub-size specimen to the thread root for a full-size threaded rod. 2) Correlate stress intensity factor between sub-size specimens to full-size threaded rods. 3) Mechanical testing of full-size rods in tension. It was found that the analysis methods agreed with each other, thus validating the use of data from sub-size specimens to predict the failure performance of full-size threaded rods. The data generated from Test V has the additional advantage of being able to quantitatively determine the effects of variations in the applied potential from compositional differences in the galvanizing and the variation in the hardness of the rods on the threshold stress.
Keywords: Hydrogen embrittlement, stress corrosion cracking, fracture mechanics, low carbon steel rods, small scale fracture mechanics specimen testing, validation of small scale specimen, determining full size performance from small scale testing, San Francisco-Oakland Bay Bridge, SFOBB
