Substructured Pseudodynamic (PsD) testing is used to study the transverse seismic response of a bridge with box-girder deck monolithically connected
to circular piers, but free to move laterally at the abutments. The tests are of two types, each one with two variants: with the pier footing rocking on elastic soil and
uplifting from it, or fixed to rigid ground. The first type concerns a two-span deck connected monolithically to a central pier and supported at each end on the abutment
through a pair of elastomeric bearings, which are stiff and strong in compression but very soft and weak in tension. The second one concerns the interior
piers of multi-span bridges having all spans equal and all piers similar. The pier was physically tested at 1:2 scale; everything else was numerically simulated
online with Opensees. The deck, the soil and the bearings were modeled as elastic, but not allowing tension to develop at the soil-footing interface, and the bearings
two orders of magnitude more flexible in tension than in compression. In the first type of tests, at a peak ground acceleration (PGA) of 0.15g, the footing uplifted,
the bearings developed tension, but the pier stayed elastic. In the second type of tests, with a PGA of 0.4g, uplifting of the footing did not prevent plastic hinging at
the pier base, but reduced significantly the pier damage. A cyclic quasi-static test in the end shows the deformation capacity margins of the pier.