This test series includes a scale model of a two-bay prototype bridge structure with the sloping ground conditions that are assumed to exist at the site of the prototype bridge structure. Due to the sloping ground, the two bays are supported by three bents, but the clear height between the soil and bridge deck is different for each bent. The centrifuge test package also includes an independent two-pile bent corresponding to medium height bent in the prototype structure, another independent medium height bent which is fixed at the bridge deck and a pile cap at the ground surface level, and a single pile corresponding to pile in the tallest bent.
This test series complements quarter scale shaking table tests performed at University of Nevada at Reno. The Reno tests focus on modeling the structural aspects of the response of prototype two-bay bridge structure. The Reno shaking table tests involved accurately modeled quarter scale reinforce concrete columns with fixed supports on the shaking tables. The centrifuge tests use aluminum tubing to represent the columns as pile extensions, but the piles are supported in soil so that soil-structure interaction effects are accurately simulated. A significant collaborative effort was required to decide what free height there should be between the bridge deck and the soil surface in the centrifuge model to correspond to the free height between the bridge deck and the fixtures on the shaking table in Reno.
In this centrifuge test series, eleven 1.22 m outer diameter prototype concrete piles were modeled using aluminum tubing with an outer diameter of 19.05 mm (0.75 in.) and a wall thickness of 0.89 mm (0.035 in.). The soil profile was prepared with a 56 cm soil layer made with Nevada sand with relative density, Dr = 80%. FSB2, a flexible shear beam model container was used in this test.
The model was instrumented with accelerometers, linear potentiometers, and strain gage bridges to measure the translation, rotation and bending response of the foundations, columns, and bent cap. Additional accelerometers were embedded in the soil to measure soil response during dynamic loading. Vertical linear potentiometers were used to measure ground surface settlement during shaking so that any densification induced changes in soil properties can be estimated.
A total of 16 shaking events were applied to the model at a centrifuge acceleration of 52 g. The shaking was applied in the transverse direction to the bridge. The testing was done in four spins. In the first spin the model was tested with a small step wave primarily to verify that the instruments and data acquisition system function properly. During Spin 2 model was tested under low level of shaking with out pile head masses. The purpose of Spin 2 tests was to tune the shaker to produce a free field motion at a point of 50 mm below the ground surface (anticipated fixity point for the piles) corresponding to a target motion prescribed by the collaborators at the University of Washington. This target motion was calculated by site response analysis using the ground motion recorded during the 1994 Northridge Earthquake at the CDMG station 24389, Century City LACC North, 090. The motions were determined to agree very ill over a range of periods between 0.6s and 1.0s, which encompasses the natural period of the bridge bents. Before Spin 3, masses were attached to the piles and during Spin 3 a low level of shaking was applied. In the final spin (Spin 4), bents in the bridge model were connected with bridge deck and model was shaken with full intensity ground motion events. In Spin 3 and Spin 4 the first events were small step wave and the ground motion for subsequent events consisted of scaled versions of the modified base motion from Spin 2 tuning results.
Cite this work
Researchers should cite this work as follows: