Curling of New Concrete Pavement and Long-Term Performance

Abstract

Curling results from the temperature differential across the concrete slab thickness and may induce undue stresses in newly placed slab. This study deals with the finite element (FE) analysis of curling, curling stresses, field measurement of curling on a newly built jointed plain concrete pavement, and comparison of its long-term performance using both the Mechanistic-Empirical Pavement Design Guide (MEPDG) and HIPERPAV II software. The FE analysis was performed with a software program ANSYS. The test section was modeled as a three-layer system with 12 in. (300 mm) concrete slab, 4 in. (100 mm) treated drainable base, and 6 in. (150 mm) lime-treated subgrade. All layers were assumed to be linear elastic. Temperature data were collected at five different depth locations across the concrete slab with digital data loggers. Curling was measured on five different days with a simple setup. The effect of temperature nonlinearities across the slab thickness was also examined. The results show that both upward and downward curling increase as the temperature differential increases. The maximum stress resulting from the combined effect of curling and traffic loading due to positive temperature differential is higher than that due to the negative temperature differential of the same magnitude. Since temperature differential has a significant influence on curling, both curling and curling stresses can be mitigated at an early age with temperature control, namely via enhanced curing. Both MEPDG and HIPERPAV II (HIgh PERformance concrete PAVing) showed approximately the same performance for the PCC thickness ranging from 12 in. (300 mm) to 8.5 in. (215 mm) for this project. Performance prediction from HIPERPAV II is very sensitive to the change in PCC thickness less than 9 in. (230 mm), whereas MEPDG prediction is not as sensitive to the thickness change as with HIPERPAV II.