Experimental Study of Continuous-Beam Lateral Torsional–Buckling Resistance with a Noncomposite Concrete Deck

Abstract

Some of Louisiana’s bridges built in the 1950s and 1960s used two-girder or truss systems, in which transverse floor beams are carried by main longitudinal members and continuous (spliced) beams supporting the deck are supported by the floor beams. The main longitudinal members are either two edge (fascia) girders or trusses. Continuous-beam bottom flanges are in compression in the negative moment region, which could result in lateral torsional buckling. When the continuous beams are load rated, Cb is calculated in accordance with a standard specification that does not account for potentially beneficial bracing effects from a noncomposite concrete deck and could therefore underestimate actual flexural strength. As a result, the rating may require a restrictive bridge posting or closure. This issue affects bridges that are key parts of Louisiana’s highway system, i.e., longer, multispan crossings with high average daily traffic (ADT) counts. The calculated load rating could require expensive, and possibly unnecessary, bridge rehabilitation or replacement with significant traffic disruption. This research reassessed the methodology behind load rating these continuous beams, with efforts focusing on deriving more realistic values for Cb. Its main objective was to evaluate the capacity of bridges built with continuous beams and noncomposite decks and to develop a new approach for rating those beams by more accurately representing the moment gradients using Cb. An experimental study was conducted to evaluate lateral torsional–buckling resistance of a reduced-scale, two-span grillage system that included three lines of continuous beams supporting a noncomposite concrete deck. A significantly higher moment gradient factor as compared to the existing codes and standards was recommended by accounting for the bracing effect provided by a noncomposite deck.