Creep in prestressed lightweight aggregate concrete

Abstract

This study was made to develop a criterion for predicting creep in prestressed lightweight-aggregate concrete beams and to investigate the influence of magnitude of prestress on creep in lightweight-aggregate concrete. Prestressed lightweight-aggregate concrete beams 5 in. by 8 in. in cross section and 8 ft. long were loaded by prestress only. Seven-wire, 3/8 in. diameter high tensile strength steel strands were used for prestressing these beams. The designed initial compressive stress in the beams, with the exception of the unloaded control beam, was held constant at 3000 psi at the bottom fiber and was varied in the different beams from 0 to 1500 psi at the top fiber. Smaller specimens of the same concrete 3 in. by 3 in. in cross section and 16 in. long were axially loaded by springs at the same time that the prestressing strand was released. The compressive stress in these specimens varied from 0 to 3000 psi. Strain data were taken from inserts spaced 10 in. apart at mid-length of specimens from the time of prestressing. From these data, curves of unit shrinkage strain and unit creep strain versus time were developed for both the prestressed beams and the spring loaded specimens. Data from the spring loaded specimens were reduced to creep per unit of stress with respect to age of concrete, and that information was then used in a program to predict creep in the prestressed beams. Creep predicted at any definite age was then compared with the creep taken from the prestressed beams. The method used in the prediction divided time into small increments and the strains were computed for each increment by using the creep information obtained from the spring loaded specimen. Loss of prestress due to creep and shrinkage was computed for each time interval. The strains computed for each time increment were summed for the total time duration and the losses of prestress were also summed to determine the total effect. The result shows that the computed strain agrees very well with the observed strain for the laboratory specimens. The method provides the designer information from which the time dependent creep and shrinkage strains for a prestressed flexural member may be computed. From these computations he may obtain a reasonable estimation of the deflection of structure brought about by creep of the concrete.