Click for next page ( 25

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 24
24 CHAPTER 3 Interpretation, Appraisal, and Application The primary objective of NCHRP Project 4-19(2) was to because the CA-3 (Uncrushed Gravel) mixture has no data at validate performance-related aggregate tests recommended that point. The matrix is shown in Table 20. The correlations by Kandhal and Parker (1). A discussion of the relationships between the rutting parameters and the coarse aggregate of these aggregate test methods to mixture performance and UVA and UVB are significant. Each of the test methods was aggregate properties follows. assigned a descriptive ranking as shown in Table 20. The rank- ing was determined by assigning a value from 0 to 9 depending on the correlation coefficient between the given test method and Coarse-Graded Mixtures the rutting parameters. For example, if the correlation coeffi- Kandhal and Parker recommended the coarse aggregate cient for a given test was higher than 0.80, but less than 0.90, the UVA (AASHTO TP 56) and the FOE21 as the first and second descriptive ranking was assigned as 8. The ranking for a given best coarse aggregate tests related to rutting, respectively. test method was then averaged for the rutting parameters used. They also found that FOE51 was an important variable, but it Thus the higher the ranking number, the better the aggregate was not recommended because of the narrow range in results test relates to HMA rutting performance in the APT. Table 20 and did not provide a clear measure of a coarse aggregate's shows that the coarse aggregate UVA and UVB have the highest performance. descriptive rankings, 9.0, while the FOE21 has the next closest The regression equations developed by Kandhal and at 8.0. The remaining aggregate tests have much lower rankings. Parker (1) suggested that as UVA increases, both rutting and Plots of the three permanent deformation parameters rate of rutting decrease. An increase in FOE21 causes a and the coarse aggregate UVA values are shown in Figures reduction of mixture stiffness and an increase in the rate of 16 through 18. These figures suggest that higher coarse rutting. In summary, Kandhal and Parker (1) suggested that aggregate UVA values result in greater resistance to perma- high coarse aggregate UVA and low FOE21 values are desir- nent deformation. The plots also indicate that continued able for improved pavement performance. The current increase in rut resistance becomes negligible for UVA values research also found a strong correlation between FOE21 and higher than approximately 50 percent. coarse aggregate UVA (r = 0.786, p-value = 0.064). The cor- Although the flat or elongated and flat and elongated values relation suggests that the FOE21 value significantly influ- at the 3:1 ratio (FOE31and F&E31) were expected to correlate ences the orientation and structure of coarse aggregate as positively with FOE21, the data shown in Table 20 indicated reflected by the UVA values. that neither of the 3:1 ratios correlates well to the rutting Because the two variables, coarse aggregate UVA and parameters. The relationships between rutting and FOE31 or FOE21, are highly correlated, it is not appropriate to use both F&E31 suggest that factors other than particle shape contribute as independent variables in a statistical model. Moreover, the to explaining rut resistance. Coarse aggregate UVA appears to conclusion drawn by Kandhal and Parker (1) about the effect successfully capture the effect of particle shape, surface charac- of FOE21 on pavement permanent deformation can become teristics, or mineralogy on rutting performance of the HMA unrealistic. In fact, a high percentage of FOE21 aggregate par- mixtures tested in this research. ticles are desirable for HMA pavements. To further investigate interaction of aggregate tests, multiple For the current study, a correlation matrix was developed for regression analyses were performed between rutting parame- eight of the coarse aggregate tests and three rutting parameters. ters and coarse aggregate test data. The response variables The total rut depth at 20,000 wheel passes was not included included total rut depth at 5,000 wheel passes and total rut rate

OCR for page 24
25 Table 20. Correlation matrix of coarse aggre- 5 gate properties and rutting parameter. 4 Rutting Parameter Coarse Total Rut Early Rut Rate, Aggregate mm/log(N) Property Depth @ Early Rut Later Rut Descriptive 3 5000 Passes Rate Rate Ranking -0.947 -0.963 -0.945 UVA 9.0 2 0.015 0.008 0.015 -0.983 -0.995 -0.983 UVB 9.0 0.003 0.0004 0.003 y = -0.40x + 21.52 -0.161 -0.242 -0.151 1 LALOSS 1.0 R2 = 0.93 0.796 0.695 0.809 0.364 0.291 0.372 MDEV 2.7 0 0.547 0.635 0.537 -0.826 -0.845 -0.819 40 45 50 55 FOE21 8.0 0.086 0.071 0.090 UVA, % -0.360 -0.436 -0.351 FOE31 3.3 0.552 0.463 0.563 Figure 17. Early rut rate/coarse -0.476 -0.534 -0.467 F&E31 4.3 aggregate UVA relationship. 0.418 0.354 0.428 -0.468 -0.550 -0.462 F&E51 4.3 0.426 0.337 0.434 The relationships between coarse aggregate UVA and APT traffic to reach 3.5- and 7.0-mm rut depths appear to be expo- for both early and late traffic. The results, shown in Table 21, nential and significant. The regression equation shown in Fig- indicate that the coarse aggregate UVA is a good indicator of ure 19 indicates the number of APT wheel passes to reach a HMA mixture rutting performance. When the results from 3.5-mm total rut depth increases by approximately 53 passes other aggregate properties such as FOE21 were combined with for an increase in coarse aggregate UVA from 45 to 50 percent. UVA, the R2 does not change significantly. This finding indi- Figure 20 shows the number of wheel passes increases by cates that the coarse aggregate UVA alone is a sufficient predic- approximately 3,715 to reach a 7.0-mm total rut depth for the tor of HMA mixture rutting performance. However, it should same 5 percent increase in coarse aggregate UVA. The expo- be recognized that these data were developed for coarse-graded nential relationship would seem logical up to a point. For mixtures using a few coarse aggregate types. example, for coarse aggregate UVA values between 35 and 40 To determine sensitivity of coarse aggregate UVA to traffic percent, the number of wheel passes to reach a given rut depth levels, the effect of coarse aggregate UVA on the number of increases very little. However, the increase is significant for APT wheel passes to reach total rut depths of 3.5 and 7.0 mm coarse aggregate UVA values between 45 and 50 percent. In was examined; results are shown in Figures 19 and 20, respec- practice, most crushed aggregates appear to have coarse tively. The 7.0-mm total rut depth was selected because it was aggregate UVA values less than approximately 50 percent. the minimum total rut depth exhibited at 20,000 wheel passes Figure 21 presents the APT rut depths as a function of the during the APT tests; the 3.5-mm total rut was arbitrarily number of wheel passes for each of the coarse aggregates used chosen as one-half of the 7.0-mm rut depth. in the study. The coarse aggregate UVA values are somewhat 35 15 y = -2.76x + 144.11 30 2 y = -1.54x + 78.59 R = 0.90 5000 Wheel Passes, mm R2 = 0.89 Total Rut Depth @ 25 Later Rut Rate, 10 mm/log(N) 20 15 10 5 5 0 0 40 42 44 46 48 50 52 40 45 50 55 UVA, % UVA, % Figure 16. Rut depth/coarse aaggregate Figure 18. Late rut rate/coarse UVA relationship. aggregate UVA relationship.