Long-term deflection test and theoretical analysis on cracked prestressed concrete box beams

A long-term deflection test is performed for 470 days on two simply supported prestressed concrete box beams under sustained uniform load. The deflection growth rules of the box beams under both cracked and uncracked states are studied. The measured longterm deflection of the cracked beam is 18.4% larger than that of the uncracked beam, whereas the deflection growth coefficient of the cracked beam is 65.3% that of the uncracked beam. The larger long-term deflection of the cracked beam is mainly attributed to the huge value of the instantaneous deflection caused by the crack. Then, the cracks lead to the release of concrete stress, thereby reducing the creep effect. For the uncracked beam, the influences of concrete shrinkage, creep, and prestress loss on the long-term mechanical behavior of components are considered, and then deflection is predicted using the blended method based on the age-adjusted modulus method, the prestressed equivalent load method, and the calculation method for prestress loss. Results show that the theoretical and measured values are in good agreement. For the cracked beam, the crack influence coefficient is introduced to revise its effective elastic modulus by considering the influence of cracks on the long-term deflection of the cracked beam. Then, a formula for calculating the long-term deflection of cracked beams that considers stiffness reduction for instantaneous deflection correction is proposed. The formula exhibits high precision in calculating and analyzing the long-term deflection of the cracked beam. The deviation between the calculated and measured values is within 10%.