Resulte of an experimental-analytical investigation on a reinforced concrete box girder bridge are presenredo The experimental phase consisted of tests on a - scale, 3 P? direct model of an intermediate span of a b* inuous box girder bridge. Experimental data from the tests under symmetrical and unsymmetrical loads were analysed using some of the available analy- sis methods. These results are combined with other available information to formulate some recommendations for design of box girder structures.

The Nelson Street Bridge in downtown Atlanta, Georgia is a ten-span continuous concerete arch bridge. Designed for carrying electric street cars, it was built in 1906. The bridge is present-ly carrying local street traffic over active tracks of the Southern Railway System and automobile parking areas. Strength evaluation procedures contrasted the latest design speci-fications and modern analysis techniques with the original design approach in every aspect. A slab-grid model was used for analyzing lateral wheel load distribution on a multiple arch and slab system. The distribution factors were then compared with those recommended by current AASHTO specifications for concrete slabs on concrete T-Beams. Arch geometry was accurately simulated, and moving axle loads were generated in a computer program. Force envelopes were produced for arch sections spaced from one foot to six feet along the arch, using critical loading positions. Sections were then checked for combined compression and bending. This approach is particularly effective when the spacing of arches is unequal and span lengths vary. Concrete cores were taken from the bridge and tested for compres-sive strength and chemical properties. Test results were satisfac-tory when compared with recommendations in the AASHTO Manual for Bridge Rating and turn-of-the-century books on concrete design. Sound arch sections from original contract plans, and existing sec-tions with concrete and steel losses determined by field inspection were used for evaluating section capacities.

Precast panel deck bridges have been in use for many years. Although some cracking is inherent in this structural system, recent concern has been expressed because of the greater degree of cracking exhibited on the surface of some decks compared to conventional cast-in-place bridge decks. Full scale structural testing was initiated to estimate the strength and evaluate the performance of the deck panel system. In this program two deck panel bridges, each with different panel support details, and a conventional cast-in-place deck bridge were tested. Results showed that the deck panel system did not act as a continuous slab over the girders as is usually assumed for design. The conventional cast-in-place deck bridge did develop continuity. The effect is to increase the maximum positive moments in the slab but not to a degree as to render the bridges unsafe.

Because prior AASHTO design codes permitted a much greater shear stress in concrete than allowed by current codes, the shear strength of existing bridges may be questioned. This paper offers an approach to shear strength assessment of bridge beams. The condition of the existing structure, theoretical strength based on recent research, and the unique characteristics of older concrete bridges are considered. Field investigation and rating methods are discussed, and a case study is presented.

A series of static live load tests were performed on two concrete T-beam bridges to evaluate stress induced in tension reinforcement and transverse load distribution factors. The purpose of these tests was to evaluate the usefulness of service load testing in evaluating the load carrying capacity of concrete bridges. The two structures were similar in dimension and reinforcing but had concrete of markedly different quality as determined from deck cores and sonic pulse velocity measurements through the deck and beam stems. The load test results showed no difference in bridge behavior attributable to concrete condition. To obtain data on failure capacity, testing was performed on two single and one double T segment taken from the deteriorated struc-ture. Rebar stress, centerline displacement, and end rotation were measured during the loading program which culminated in com-pression failure after rebar yield. The measured failure moments exceeded the nominal flexural strength as given by ordinary ulti-mate strength design methods. The consequences of these findings on the load rating process are discussed and a strategy for rating older reinforced concrete T-beam bridges is proposed.