LATERAL CAPACITY OF CORRODED PILE BENTS

This report presents the findings from two separate studies. The first contains results from the main study on the lateral capacity of corroded pile bents. The second relates to a supplemental investigation on the use of carbon fiber reinforced polymers for pile repair. A two-year experimental study was conducted to investigate the effect of corrosion on the lateral capacity of corroded pile bents. One-third scale models of corroded pile bents were fabricated and tested. Their capacity reduction was assessed in comparison to an identical uncorroded control. Surveys of characteristic corrosion damage in pile bents were carried out to establish the geometry of the pile bent and the location of the damage. Soil restraint was modeled so that it reproduced the axial load and bending moment distribution in the damaged portion of the piles from the point of inflection. Corrosion damage was simulated using a constant current system and stainless steel counter electrodes. The system developed was very successful in achieving targeted metal loss in strands and ties in a predictable time frame and in reproducing damage that was typically observed. Gravimetric testing was carried out to establish actual metal loss in the test specimens. Three different metal loss levels were examined - 10%, 30% and 50%. Lateral capacity was determined under sustained axial loads corresponding to 90 kips per pile in the prototype. A five pile bent was selected for the testing in which the piles were spaced at 6.5 times the pile size and all five piles were corroded. The center of the corroded region was located 0.15 x pile length from the underside of the pile cap. All model bents were heavily instrumented by load cells, shear cells, strain ages, LVDTs, a pressure transducer and string line transducers to assess the performance during testing. An MTS controller and hydraulic actuator was used to apply and measure the lateral load and five hydraulic jacks mounted on a trolley were used to simulate service loads during testing. All data was recorded using an acquisition system. The results indicated that ultimate capacity was not severely compromised by high levels of corrosion. Compared to the controls, the reduction in the lateral load in specimens corroded 10%, 30% and 50% were respectively 1%, 23% and 30%. Failure occurred in the pile cap in the controls and the 10% corroded bents, but in the corroded zones for the 30% and 50% metal loss levels. Analysis of the strain data indicated that the top and bottom ends of the piles, assumed fixed did not exhibit full fixity throughout the tests.