The effect of coarse aggregate on pavement performance has been attributed to the volume of aggregate used in pavement construction. The different patterns of crack development for limestone (LS) and siliceous river gravel (SRG) are a typical example of aggregate-induced variable performance in continuously reinforced concrete pavement (CRCP). An attempt was made to find a reasonable solution for pavements with SRG. As a way to solve the performance problem observed from the SRG pavement, a blended aggregates mixture was suggested. Laboratory and field tests were performed to check the feasibility of their application in pavements. From the laboratory test, a 50:50 blending ratio was suggested after considering the effect on tensile strength and thermal coefficient of expansion. Field test sections were also constructed to verify previous performance observations for the two aggregates and to provide performance data for new variables such as blended aggregates and special curing methods. Unexpectedly, the blended mixture did not improve the performance of SRG pavement; rather it experienced worse cracking than SRG alone. A controlled experiment with additional field test sections is needed to verify or disprove this finding. The only definitive finding was that selection of aggregate in the concrete pavement is a vital consideration for the design of the pavement. The CRCP8 analytical program reasonably predicted crack spacing for both SRG and LS pavements, predicting mean crack spacing of 0.99 m (3.25 ft) for SRG and 1.98 m (6.41 ft) for the limestone. These values are somewhat below the actual spacing observed at 100 days. Data collected after the first winter period will be required to calibrate the program.
