ACI SP-106

Abstract

<strong>INTRODUCTION</strong> <br><strong>Background</strong> <br>Application of precast, prestressed girders to bridge construction started in the United States in the early 1950's. Use of pretensioned I-girders with cast-in-place concrete decks grew rapidly. Until the early 1960's, bridges built with pretensioned I-girders and cast-in-place concrete deck were designed as simply supported spans. However, longitudinal reinforcement placed in continuous deck slabs above the piers provided negative moment capacity. Therefore, these I-girders could be considered as partially continuous for negative moments at the piers. The degree of continuity depends on the time-dependent effects and the positive and negative moment connection details provided at the piers. <br>Application of precast, prestressed girders to bridge construction started in the United States in the early 1950's. Use of pretensioned I-girders with cast-in-place concrete decks grew rapidly. Until the early 1960's, bridges built with pretensioned I-girders and cast-in-place concrete deck were designed as simply supported spans. However, longitudinal reinforcement placed in continuous deck slabs above the piers provided negative moment capacity. Therefore, these I-girders could be considered as partially continuous for negative moments at the piers. The degree of continuity depends on the time-dependent effects and the positive and negative moment connection details provided at the piers. <br>In a pretensioned member, prestress will usually cause the member to camber. If the member is simply supported, the ends of the member will tend to rotate, as shown in Fig. l(a). When members are made continuous through the deck and pier diaphragms, the ends of the pretensioned girder are restrained from rotating. As a result, a positive restraint moment, as shown in Fig. l(b) may occur at the pier. Positive moment also occurs at the piers when alternate spans have live loads. Reinforcement for positive moment connection is designed for the summation of positive moment due to time-dependent effects and live load application. Construction of the positive moment connection detail is generally expensive and time consuming. <br>In 1961, the Portland Cement Association (PCA) conducted an experimental research program on this type of bridge (1). The research program studied the influences of creep in the precast girders and differential shrinkage between the precast girders and the cast-in-place deck slab on continuity' behavior after an extended period of time, As a result of these studies, procedures were developed for design of the positive moment connection and the negative moment reinforcement over supporting piers (2). <br>There are several uncertainties associated with the PCA procedures. Some of the uncertainty stems from the simplifying assumptions made in the PCA procedures. One assumption is that girder concrete and deck concrete have the same creep and shrinkage properties. This would not generally be the case, particularly if the sequence of construction results in significantly different ages between the girder, diaphragm, and deck concrete. Different concrete mixes and curing conditions for girder, diaphragm, and deck concretes also cause differences in creep and shrinkage properties. Also, for the PCA simplified analyses, the continuity connections are considered to have zero length and to be fully rigid. Full continuity is assumed in calculation of live-load positive and negative moments. The actual connections have finite lengths and rotational stiffnesses. The moment of inertia of the reinforced concrete section at the connection after cracking from either positive or negative moment will be significantly lower than the prestressed girder section. In addition, when positive restraint moment from time-dependent effects causes cracking in the diaphragm concrete, these cracks must close before the full section becomes effective for negative live load moment.