Forty-four different mixes used in Colorado by the Colorado DOT were sampled. The samples were split into 100 mm and 150 mm replicated specimen sizes and compacted in the Superpave gyratory compactor to Nmaximum. The bulk specific gravity of the specimens was measured at Nmaximum and back calculated to Ndesign.
The average bulk specific gravity at Ndesign of the 150 mm diameter specimens was 2.347 and for the 100 mm diameter specimens was 2.336. This was an average difference of 0.011. The standard deviation of the differences was 0.021. Using the t-test, this is a statistically significant difference. However, this difference represents about 0.4% air voids or approximately 0.2% asphalt content. The 150 mm specimens were more dense, or the 100 mm specimens would give slightly higher asphalt contents. Compared with typical laboratory repeatability, this difference is not huge, although it is significant.
The air voids of the specimens at Ndesign ranged from 1.2 to 6.2. The average was approximately 4.0%. The best fit line of the data was within 1% air voids of the line of equality.
Considering the advantage of 100 mm specimens for use in the Hveem stabilometer and ease of sampling size, 100 mm specimens are being considered for use in the Superpave mixture design process by the Colorado DOT.
Those with questions or comments on this study by the Colorado DOT should contact:
Mr. Tim Aschenbrener