This paper presents the results from strand end-slip measurements and load tests of members that were fabricated at six different precast concrete plants over the past 2½ years. All of the work reported herein is based on specimens that were produced using standard concrete mixtures and placement techniques. As such, the data presented are believed to be representative of current industry practice.
This study revealed that the occurrence of the so-called top-bar effect (top-strand effect) for pretensioned strands is primarily a function of the amount of concrete above the strand rather than the amount of concrete below it. Accordingly, the results of this investigation indicate that the current design assumptions for bond in pretensioned members are unconservative for members with strands near the top as-cast) surface. This phenomenon can result in extremely large transfer lengths for strands located within a few inches of the top surface, including those in thin members. In addition, the top-bar effect typically becomes more pronounced in members as concrete fluidity increases. However, these same findings also revealed that the current design assumptions for bond were generally accurate when strands were located deeper in the members. This was true for members made with either flowable concrete or self-consolidating concrete.
