Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
12 Laboratory Experiment and Results The modified Lottman protocol per AASHTO T 283 was used in NCHRP Project 9-49 to evaluate moisture suscepti- bility of WMA. In order to assess alternative moisture condi- tioning protocols and to investigate various specimen-drying methods prior to laboratory testing, the laboratory experiment presented in Figure 13 was completed. The selected mixture corresponded to a field project on State Route 196 in Wyoming. Four fractions of limestone aggregates and river sand were used to prepare the mixture. The inclusion of one percent lime as an anti-stripping agent was specified by the mix design, but it was not included in the laboratory experiment in order to promote moisture damage in the labo- ratory tests. The mix was coarse graded with a 12.5 mm nomi- nal maximum aggregate size. A PG 64-22 binder was used in the mixture with an optimum binder content per mix design of 5.0 percent by weight of the mixture. Additional moisture conditioning protocols evaluated in the experiment included MIST and hot water bath (HWB). Detailed moisture conditioning parameters for each protocol are summarized in Table 4; 1,000 and 2,000 MIST cycles at the equipment manufacturerâs default settings (i.e., 140°F [60°C] and 40 psi) were selected based on previous experience and relevant literature, and they resulted in less time-consuming protocols than the modified Lottman protocol (half a day versus three days). The HWB at 140°F (60°C) was included as a simplified modified Lottman protocol without vacuum saturation and freezing, but it required the same time span of three days. A set of laboratory fabricated specimens were subjected to the various moisture conditioning protocols prior to being characterized in the MR, IDT strength, and Asphalt Pave- ment Analyzer (APA) tests. The APA test was selected in the experiment over the HWTT test due to the fact that HWTT specimens are tested in a wet condition (i.e., under water) and thus, no moisture conditioning is needed prior to testing. Test parameters including MR stiffness, IDT strength, and APA rutting resistance parameter (RRP) were determined after each moisture conditioning protocol, and the corresponding ratios (MR ratio, TSR, and APA RRP ratio) were used to quan- tify the reduction in mixture stiffness, strength, and rutting resistance after moisture damage, respectively. In addition, four different specimen-drying methods for moisture condi- tioned specimens were evaluated in the MR and IDT strength tests after the modified Lottman protocol, including saturated- surface dry (SSD) per AASHTO T 166, 48-hour air dry at 77°F (25°C), CoreDry per AASHTO PP 75, and 24-hour oven dry at 104°F (40°C). Test results obtained in the experiment were used to determine if an equivalent level of moisture damage to the modified Lottman protocol was achieved by the MIST or HWB protocols, and to evaluate the effects of various specimen- drying methods on moisture susceptibility parameters. Moisture Conditioning Protocols Figures 14 through 16 present the MR, IDT strength, and APA results for mixtures after various moisture conditioning protocols, including the modified Lottman protocol per AASHTO T 283 consisting of vacuum saturation plus one freeze-thaw cycle, 1,000- and 2,000-cycle MIST at 140°F (60°C) and 40 psi, and three-day HWB at 140°F (60°C). All moisture conditioned specimens were tested at SSD conditions. In each figure, the bars represent the average value of MR stiffness at 77°F (25°C), wet IDT strength at 77°F (25°C), and APA RRP at 122°F (50°C); and the error bars represent one standard deviation from the average value of three replicates in the case of the MR and IDT strength tests or two replicates in the case of the APA test. In addition, the mixture property ratios (MR ratio, TSR, and APA RRP ratio) are shown in the text boxes above the bars. As illustrated in Figure 14, the dry control specimens had sig- nificantly higher MR stiffness than all the moisture conditioned specimens, indicating a significant reduction in mixture stiffness after moisture conditioning. In addition, equivalent mixture stiffness was achieved by the moisture conditioned specimens C H A P T E R 5
13 Moisture Conditioning Protocols Dry Control Modified Lottman Moisture Induced Stress Tester Hot Water Bath Saturated-Surface Dry (SSD) MR Test IDT Strength Test APA Test Specimen- Drying Methods Modified Lottman SSD Air Dry CoreDry Oven Dry MR Test IDT Strength Test Dry Figure 13. Laboratory experiment. Moisture Conditioning Protocols Parameters Total Testing Time Modified Lottman Vacuum Saturation (70 to 80% degree of saturation) + One Freeze (-18°C) / Thaw (60°C) Cycle 3 Days Moisture Induced Stress Tester (MIST) Temperature: 60°C Pressure: 40 psi Number of Cycles: 1,000 and 2,000 0.5 Day syaD 3 C°06 :erutarepmeT )BWH( htaB retaW toH Table 4. Moisture conditioning protocols and parameters. Figure 14. MR stiffness results for various moisture conditioning protocols. Figure 15. IDT strength results for various moisture conditioning protocols.
14 using the modified Lottman protocol, the 1,000-cycle MIST, and the three-day HWB, while a significantly lower mix- ture stiffness value was observed for the 2,000-cycle MIST. A statistical analysis including analysis of variance (ANOVA) and Studentâs t-test (for each pair) was performed with a 5% significant level (i.e., a = 0.05) to further discriminate the MR stiffness results considering their variability, and the detailed results are presented in Appendix B. The statistical analysis results in terms of connecting letters report shown in Table 5 further confirmed that the 2,000-cycle MIST had a significantly lower MR stiffness, while no significant dif- ference was shown for the other three moisture condition- ing protocols. The same conclusion was also obtained by comparing the MR ratio results for various moisture condi- tioning protocols versus the d2s acceptable range of 10.0% (Epps Martin et al., 2014). A similar trend is shown in Figure 15, where a higher IDT strength value was observed for the dry control specimen as compared to the moisture conditioned specimens. For the com- parisons among various moisture conditioning protocols, the specimens conditioned using the 1,000-cycle MIST and three-day HWB protocols exhibited the highest IDT strength, followed by the modified Lottman protocol and then the 2,000-cycle MIST protocol. According to the statistical analysis results presented in Table 5, moisture conditioned specimens using the modified Lottman, 1,000-cycle MIST, and three-day HWB protocols had statistically equivalent wet IDT strength, which was higher than that of the specimens conditioned with the 2,000-cycle MIST protocol. The same conclusion was also obtained by comparing the TSR results for various moisture conditioning protocols versus the d2s acceptable range of 9.3% (Azari, 2010). The effect on mixture rutting resistance in terms of APA RRP results from various moisture conditioning protocols is illustrated in Figure 16. The RRP value represents the visco- plastic strain increment of the mixture at a critical number of load cycles (i.e., 10,000); and therefore, mixtures with lower RRP values are expected to have better rutting resistance than those with higher RRP values (Yin et al., 2014). The dry con- trol specimens had a lower RRP value as compared to the moisture conditioned specimens, indicating better rutting resistance in the APA test. Among the various moisture con- ditioning protocols, the modified Lottman protocol had the highest RRP value, followed by the 2,000-cycle and 1,000-cycle MIST protocols and then the three-day HWB protocol. To better discriminate various moisture conditioning protocols, the same statistical analysis introduced previously was per- formed to consider the variability of the APA RRP results. According to the statistical analysis results shown in Table 5, the effect of all moisture conditioning protocols on rutting resistance was significant and different from each other. The 2,000-cycle MIST protocol yielded the smallest difference as compared to the modified Lottman protocol, even though a statistically significant difference was observed. Table 5 summarizes the statistical analysis results in terms of connecting letters report for various moisture conditioning protocols investigated in the study; the more detailed results are presented in Appendix B. According to the MR stiffness and IDT strength results, the 2,000-cycle MIST protocol produced the most severe moisture damage, while no significant differ- ence was shown for the other three protocols. However, a dif- ferent trend was observed for the APA RRP results; the most severe moisture damage was created by the modified Lottman Figure 16. APA RRP results for various moisture conditioning protocols. Moisture Conditioning Protocols MR Stiffness IDT Strength APA RRP Dry Control A A A Modified Lottman B B D 1,000-cycle MIST B B B-C 2,000-cycle MIST C C B Three-day HWB B B C-D Table 5. Statistical analysis results for various moisture conditioning protocols.
15 protocol, followed by the 2,000-cycle and 1,000-cycle MIST protocols, and then the three-day HWB protocol. Based on these results, the 1,000-cycle MIST protocol at 140°F (60°C) and 40 psi and three-day HWB protocol at 140°F (60°C) are proposed as two alternatives to the modified Lottman protocol per AASHTO T 283 that could be used as part of the moisture susceptibility guidelines in Figure 1. Specimen-Drying Methods Figures 17 and 18 present the MR and IDT strength test results for moisture conditioned specimens with various spec- imen-drying methods, including SSD, 48-hour air dry at 77°F (25°C), CoreDry, and 24-hour oven dry at 104°F (40°C). In each figure, the bars represent the average value of MR stiffness at 77°F (25°C) and wet IDT strength at 77°F (25°C) after the modified Lottman protocol per AASHTO T 283, and the error bars represent one standard deviation from the average value of three replicates in the case of the MR and IDT strength tests or two replicates in the case of the APA test. In addition, the MR ratio and TSR results for various specimen-drying meth- ods are shown in the text boxes above the bars. The same sta- tistical analysis introduced previously was performed for MR stiffness and IDT strength results for various specimen-drying methods. Table 6 summarizes the analysis results in terms of connecting letter reports, and the more detailed results are presented in Appendix B. As illustrated in Figure 17, an equivalent MR stiffness value was observed for SSD specimens and CoreDry specimens, which was slightly higher than those of air dry specimens and 24-hour oven dry specimens. As previously mentioned, the testing of MR specimens in the SSD condition could preclude an accurate measure of mixture property due to the fact that the water occu- pying the permeable pores of the specimens artificially increases mixture load-carrying capacity due to pore pressure and incom- pressibility of water (Azari and Mohseni, 2013; Laukkanen et al., 2012). This might be the primary reason for higher MR stiffness values observed for SSD specimens versus the air dry and oven dry specimens shown in Figure 17. According to the statistical analysis results shown in Table 6, no significant difference was observed among the four different specimen-drying methods. However, a slightly different trend was obtained by comparing the MR ratio results versus the d2s acceptable range of 10.0% (Epps Martin et al., 2014), where the air dry specimens had a slightly lower MR stiffness than the SSD and CoreDry specimens (with a 11.4% difference in MR ratio results). The mixture strength results shown in Figure 18 illustrated that the SSD and air dry specimens had equivalent wet IDT Figure 17. MR stiffness results for various specimen-drying methods. Figure 18. IDT strength results for various specimen-drying methods. Specimen-Drying Methods MR Stiffness IDT Strength SSD A A-B Air Dry A A-B CoreDry A A Oven Dry A B Table 6. Statistical analysis results for various specimen-drying methods.
16 strength, while slightly higher and lower strength values were observed for the CoreDry and 24-hour oven dry specimens, respectively. The difference between the SSD and air dry spec- imens in wet IDT strength was significantly reduced as com- pared to the difference in MR stiffness (Figure 17). This was possibly due to the fact that the IDT strength test is destruc- tive and the specimens are loaded monotonically (instead of repeatedly), and thus, the water present in the specimens does not offer the load-carrying capacity benefit as it apparently did in the MR test. According to the statistical analysis results shown in Table 6, no significant difference was observed for the four specimen-drying methods, with only one exception for CoreDry versus oven dry methods. A similar conclusion was obtained by comparing the TSR results versus the d2s acceptable range of 9.3% (Azari, 2010), where the CoreDry specimens had higher IDT strength than the air dry and oven dry specimens (with 9.7% and 14.5% differences in TSR results, respectively). According to the results shown in Figures 17 and 18, the SSD and CoreDry specimens had higher MR stiffness and IDT strength values than the air dry and oven dry specimens, although the difference was insignificant according to the statis- tical analysis results shown in Table 6. Considering that the water occupying the permeable pores of the specimens was likely to preclude an accurate measurement of MR stiffness and IDT strength, the SSD method was excluded from use in the revised flow chart. Instead, the other three specimen-drying methods of CoreDry, 48-hour air dry at 77°F (25°C), and 24-hour oven dry at 140°F (60°C) were recommended, with the CoreDry method preferred due to the shorter time requirement.
