Optimisation of the Crack Pattern in Continuously Reinforced Concrete Pavements
Recent field investigations on several new Continuously Reinforced Concrete Pavements (CRCP) in Belgium indicate that its crack pattern is characterized by low mean crack spacing along with a high percentage of clusters of closely spaced cracks. Field surveys also indicate that it is difficult to significantly reduce the probability of a non-uniform crack pattern - such as closely spaced cracks, meandering, and Y-cracks - by only slightly adjusting the amount of longitudinal steel. Non-uniform crack patterns are inevitable and common in conventional CRCPs. Extensive distress data analyses of many CRCP sections in the United States have shown that the majority of punchouts develop at short spaced transverse cracks. Moreover, a non-uniform crack pattern and a high variability in transverse crack spacing were found to have a higher probability of punchout development. It is generally understood that the long-term performance of a CRCP is largely determined by it early age behaviour. Previous experiences have shown that the early entry method can eliminate the clusters of closely spaced cracks and a more regular crack pattern is achieved. In the present study a new early entry method, partial surface notch, is proposed to improve the crack pattern of CRCP. The primary objective of this study is to optimize the crack spacing pattern of CRCP through an active crack control method. To realize the research objective, predicting the pavement temperature at early age is a good starting point to understand the early age behaviour of CRCP. This study thus firstly provides a procedure to predict the early age temperature development of a concrete pavement based on concrete mixture composition, the thermal characteristics of the concrete and the underneath pavement layers, the environmental conditions, and the construction time and curing methods. Available heat flux models for the pavement surface are initially reviewed and adjustments to improve the accuracy of the predicted early age concrete pavement temperature are suggested. The proposed model enables to simulate the use of blended slag cement and the plastic sheet curing for the Belgium CRCP practice. This temperature model is verified with field measured data of two projects in Belgium, and the result is quite satisfactory. Lastly, an approach is proposed to generate reliable and real-time climatic inputs by using limited weather forecasting climate data for the temperature prediction model during the construction phase. This allows the contractor to optimize construction operations, especially the time of saw cutting. Because notches are made at early age to induce transverse cracks at the designated locations, in addition to the mechanical properties of tensile strength and elastic modulus, the evolution of the fracture energy has to be known to evaluate the cracking tendency of the notched concrete pavement. A deformation-controlled uniaxial tensile test on unnotched specimens is performed for the typical CRCP concrete mixture used in Belgium. Experimental results show that the applied unnotched parabolic shape concrete specimens, the used tension set-up with three hinges, and the applied test procedures succeed in obtaining the complete softening curves for the Belgium CRCP concrete mixtures ranging from 24 hours to 90 days. In order to correlate the concrete properties in field and laboratory conditions for accurately predicting the cracking in a concrete pavement, degree of hydration based descriptions of the early age concrete properties are given based on the experimental results of the tension tests. Using the proposed temperature prediction model and measured early age concrete properties, the concrete stress history at early age is calculated by the superposition principle (through a step-by step numerical method). The time dependent relaxation of the early age concrete, which was described as a function of the degree of hydration is considered as well. The zero stress temperature, peak pavement temperature, built-in temperature gradient, and the cracking time are determined by the estimated early age temperature and stress development. Extensive parametric simulations have shown that the early age concrete temperature and stress development are closely related with various environmental and construction conditions, such as time on the day of concrete placement, construction season, plastic sheet curing, concrete placement temperature etc. A fracture mechanics based procedure is developed to calculate the saw cut depth and saw cut timing for the active crack control method. The estimated final set time gives the lower limit for the saw cutting operation to avoid ravelling while the predicted cracking time indicates the upper limit of the saw cutting window before initiation of randomly occurring natural cracks. Theoretical analyses demonstrate that the applied saw cut depth and saw cut length is appropriate for the current CRCP conditions in Belgium. Extensive field investigations were conducted on two recently constructed CRCP sections in Belgium to evaluate the effect of longitudinal reinforcement percentage and active crack control methods on the crack pattern of CRCP. The crack pattern development, crack width, and crack width movement due to daily temperature variation, were regularly investigated. Field evidences have shown that the proposed active crack control method is very effective in inducing cracks. Moreover, the transverse cracks in the active crack control sections are much straighter and more regular spaced. The active crack control method significantly reduces the percentage of short spaced cracks and cluster cracks and thus reduces the risk of punchout development in the long-term of CRCP.