| description abstract | Tests were carried out on two continuous one-way slab specimens, representing, at a scale of 1:1.5, the floor slabs of a multistory building most affected by a postulated instant loss of a perimeter column owing to accidental loading. The bottom bars in the first specimen were discontinuous over the central support, and top reinforcement extended only over the supports. In the second specimen, reinforcing bars were continuous from one outer support to the other on both faces of the slab. Thanks to the membrane action due to the horizontal restraint by the rest of the floor system, both specimens—especially the one with continuous top bars—sustained the distributed load of the slab plus a line load along the sinking outer support of 10 kN/m without collapse in full scale, with settlement of that support as high as 0.5 m. The main features of the force-deformation behavior can be captured by a simple, hand-calculation model, taking into account geometric nonlinearities, including large deformations. Model predictions compare best with test results when tension stiffening is neglected; moreover, they are not very sensitive to the exact location where the slab first cracks or develops the critical plastic hinge. Test results were combined with those of the perimeter beams on which the slabs are supported to estimate the margin of bearing capacity of these beams over and above self weight, finishings, and quasipermanent loads on the slab. Membrane action in the slab seems to become a prime player in the global response to the loss of a perimeter column when the beam’s flexural mechanism and arch action are past their peak resistance. The slab’s membrane mechanism, together with the arch and catenary actions that develop at the edge beam in the transverse direction when the full structure is considered, are important for maintaining the capacity margin close to its peak value at larger beam deflections. This is better achieved if the slab is reinforced throughout its top surface. | |
