Long-Term Deflections of Reinforced Concrete Beams
Abstract
The CSA A23.3 code provisions compute the long-term deflection of reinforced concrete flexural members by applying a multiplier to the short-term deflection, essentially ignoring the creep and shrinkage characteristics of concrete. The CAC Concrete Design Handbook presents more elaborate methods, but fails to account for the factors that influence the creep and shrinkage of concrete.
The four widely recognized models for computing creep and shrinkage strains yield predictions that differ by up to 30%. Studies by others to assess the accuracy of the models apply different of statistical analyses do different datasets and so yield contradicting outcomes, making it difficult to quantify the prediction error.
Mechanics-Based Methods for computing long-term deflections based on first principles are proposed and used to derive Alternative Simplified Methods. The accuracy of existing and proposed methods is assessed by quantifying test/predicted ratios. The CAC Handbook Method yields mean test/predicted ratios ranging from 0.97 (conservative) to 1.34 (unconservative). The mean test/predicted ratio for the Mechanics-Based Method and the Alternative Simplified Method range from 0.92 to 0.94, and from 0.97 to 1.05, respectively. The A23.3 Multiplier Method overestimates the deflection of lightly-reinforced members.