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.
41 Introduction This chapter describes the field performance of the HMA and WMA pavements in terms of rut depth based on the second-round field distress survey results, the analysis used to determine the significant determinants for rutting, and the method used to develop a predictive model for rut depth. Minimal rutting was found during the first-round distress survey. No moisture-related distress was identified in the field during either survey round. Rutting in the Field Because the first-round distress survey recorded minimal field rutting (less than 1â16 inch), the field rutting comparisons are based on the second-round field distress survey results, which include the in-service projects and new projects shown in Figure 5.1 (a). The rut depths were measured every 50 ft in three 200-ft sections for both HMA and WMA pavements, and the averaged rut depths of the HMA and WMA pave- ments were compared. The manual measurement of the rut depth had a precision of 1â16 inch, meaning that only a rut depth higher than 1â16 inch would be recorded. Also, 1â16 inch was used as a practical criterion to determine if there was any significant difference in the averaged rut depth values between the HMA and WMA pavements. Figure 5.1 (b) presents a summary of the HMA versus WMA comparisons in terms of averaged rut depth values. Among the 43 HMA-WMA pairs, the HMA pavements show rut depth values that are compa- rable to those of the WMA pavements, that is, the differences in rut depth between HMA and WMA are within 1â16 inch. Figure 5.2 plots the rut depth versus pavement age for the second-round distress survey. As shown, rut depth starts to build up as early as 3 years in some pavement projects. There is a general trend that the older the pavement, the higher the rut depth. For pavements 6 years old or older, the accumulated rut depths are more differentiable (more than 0.1 inches) for both HMA and WMA pavements. The WA SR 12 project developed quite significant rutting (0.1 to 0.2 inches) at 3 to 4 years old, which could be attributed to studded tire usage during winter time in Washington State. The TN 125 project with thin overlay thickness (1.25 inches) also developed relatively high rut depth early in its service life. Breaking down the rutting comparison into the different WMA technologies, as shown in Figure 5.3, the WMA pave- ments show rut depth values that are generally comparable with those of the HMA pavements. Rutting comparisons among the WMA pavements based on the limited second-round survey results are presented in Figure 5.4. As seen, the WMA pavements with different WMA technologies generally show comparable rutting. Paired Ranking Analysis for Rutting The material properties obtained from the first-round field cores were used to determine the potential significant deter- minants for rutting using paired ranking analysis. Figure 5.5 presents a summary of the number of HMA-WMA pairs that show a consistent trend between material properties and field rutting. Detailed statistics of the paired comparisons between HMA and WMA are presented in Appendix B. As shown, the RRI values (34 out of 41 pairs) and binder high tempera- ture PG (e.g., 70 for PG 70-22) (34 out of 41 pairs) provide the highest rankings. Relatively high RRI values and binder PGs correlate with low rut depth values in the field. Statistical Predictive Models for Rut Depth The PLS method was applied to develop a predictive rut depth model. The following factors were considered in the model development based on the findings from the literature: ⢠Mixture properties: RRI, dynamic modulus; in-place air voids; percentage passing the #4, #8, and #200 sieve; and NMAS. C H A P T E R 5 Rutting and Moisture Susceptibility
42 (a) Projects Having Measurable Rut Depth Note: The projects without measurable rut depths for the second-round distress survey are the MT I-15 and MN TH 169 projects. Field surveys were not performed for the CA HVS, TX FM973, or TX SH 71 projects. (b) Summary of Rut Depth Comparison 0 5 10 15 20 25 30 35 40 3 39 1 N o. o f P ai rs H>W H=W H<W Figure 5.1. Comparison of rut depth for HMA-WMA pairs based on the second-round distress survey results. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Se co nd -R ou nd R ut D ep th , i n. Pavement Age, years TN SR 125 WA SR 12 HMA WMA Figure 5.2. Rut depth versus pavement age for the second-round distress survey.
43 (a) Comparison Between Chemical and Organic (b) Comparison Between Chemical and Foaming 1 4 1 0 2 4 6 8 10 N o. o f P ai rs Chemical>Organic Chemical=Organic Chemical<Organic 0 6 1 0 2 4 6 8 10 N o. o f P ai rs Chemical>Foaming Chemical=Foaming Chemical<Foaming (c) Comparison Between Organic and Foaming 1 4 1 0 2 4 6 8 10 N o. o f P ai rs Organic>Foaming Organic=Foaming Organic<Foaming Figure 5.4. Rut depth comparisons among WMA technologies based on the second-round distress survey results. Figure 5.3. Breakdown comparison of rut depth between HMA and specific WMA technologies based on the second-round distress survey results. Note: The numbers in parentheses indicate the number of HMA-WMA pairs in the rut depth comparison. H>W (1) H=W (10) (a) HMA versus Sasobit H>W (1) H=W (13) H<W (1) H>W (1) H=W (16) (c) HMA versus Foaming(b) HMA versus Chemical H=W (6) (e) HMA versus Water-Containing Foaming H>W (1) H=W (10) (d) HMA versus Water-Based Foaming
44 ⢠Binder properties: Jnr0.1, Jnr3.2, Jnrdiff, R0.1, R3.2; high tempera- ture PG; and effective binder content. ⢠Pavement structure: HMA thickness and overlay thickness. ⢠Climate, traffic, and pavement age parameters: high- temperature hours; AADTT; and pavement age. Table 5.1 contains the range of each variable measured in the project. Equation (5.1) presents the model for predicting the field rut depth using the PLS method. Five parameters were selected: pavement age (+), RRI (-), AADTT (+), total HMA thickness (+), and overlay thickness (+). The sign in parentheses indicates the trend between the predictor variables and the responses. As indicated, a lower rut depth value is expected in a pave- ment with a higher RRI value, younger age (in months), lower AADTT value, and a thinner pavement thickness. The effect Note: RRI results for the TX FM 324 project are not available because the specimens were too thin. RRI High PG 0 5 10 15 20 25 30 35 40 N o. o f P ai rs NegativePositive Jnr0.1 Jnr3.2 Mix creep compliance (86°F) Hamburg Rutting Resistance Index R0.1 R3.2 Binder high temp PG Mix E* (86°F) Figure 5.5. Number of HMA-WMA pairs that have consistent rankings between material property ranking and rut depth ranking based on the second-round distress survey results. Variable Unit Range Rut depth inch 0.002-0.394 RRI NA 769.8-18718 Service life month 49-110 AADTT NA 5-57000 Overlay thickness mm 19-76 Total HMA thickness mm 51-363 In-place air voids % 1.8-9.1 Mixing temperature °C 107-168 Accumulated hour of air temperature>25°C hour 87-10944 Asphalt film thickness mm 10.5-33.3 Effective binder content % 7.5-13.8 High temperature performance grade (PG) °C 60.3-96.6 MSCR Jnr0.1 kPa-1 0.03-1.47 MSCR Jnr3.2 kPa-1 0.03-1.62 MSCR Jnrdiff % 1.12-47.6 MSCR R0.1 % 4.6-69.4 MSCR R3.2 % 2.8-68.3 Mixture dynamic modulus, 30°C, 0.1Hz MPa 186-2009 Percent passing the #4 sieve % 45-82.2 Percent passing the #8 sieve % 24.9-57.6 Percentage passing the #200 sieve % 2.8-11.9 NMAS mm 9.5-12.5 Table 5.1. Variables and their ranges used to develop the rut depth prediction model.
45 of pavement thickness on rut depth is counterintuitive. How- ever, based on both field measurements and prediction models, the relationship between pavement thickness and predicted rut depth can either be positive or negative (Von Quintus et al. 2012; ARA 2009). The effect of HMA thickness on rut depth (increase or decrease) is mixture dependent for individual proj- ects. For instance, Von Quintus et al. (2012) found that, for the Mississippi SPS-5 and Wisconsin SPS-5 field projects, the maxi- mum rut depth increased with increasing HMA layer thickness, while several other projects from LTPP test sections showed an increased rut depth with reduced HMA layer thickness (ARA 2009). Additional research should be conducted to investigate the effect of HMA thickness on rutting. Y X X X X X 0.489776 0.119943ln 0.062497ln 0.018386ln 0.043839ln 0.086717ln (5.1) 1 2 3 4 5 ( ) ( ) ( ) ( ) ( ) = â + â + + + where Y = field-measured rut depth, inch; X1 = pavement age, month; X2 = RRI; X3 = AADTT; X4 = total HMA thickness, mm; and X5 = overlay thickness, mm. Figure 5.6 (a) shows the relationship between field-measured rut depths and predicted rut depths. This model gives a R2 of 0.76, standard error of the estimate of 0.04, and Mallowâs Cp of 5.0, indicating fairly good prediction quality. Figure 5.6 (b) presents the validation of the rut depth model using the LOOCV method. As shown, most of the validated data are located fairly close to the line of equality. The R2 value of 0.67 and standard error of the estimate of 0.06 indicate reasonable validation results. Potential Significant Determinants for Rutting Table 5.2 contains the identified potential significant deter- minants for rutting based on two rounds of paired ranking analysis and the statistical models. In the table, â+â means a positive relationship, that is, a higher value will lead to higher rut depth; â-â means a negative relationship, that is, a higher value will lead to lower rut depth. As seen, RRI was identified as a potential significant deter- minant for rutting by both paired ranking analysis and mod- eling methods. The RRI parameter can be obtained using the Hamburg wheel tracking (HWT) test, a method that has been implemented by several DOTs as a mixture screening test during mix design. Binder high temperature PG was also found to correlate well with the field rutting results based on the paired ranking method. However, it is only a binder prop- erty, and as such it may be difficult to serve as a significant material determinant to differentiate mixtures that used the same binder but different aggregates and mix designs. (a) (b) 0.0 0.1 0.2 0.3 0.4 0.5 0.0 0.1 0.2 0.3 0.4 0.5 Pr ed ic te d Ru t D ep th , i n. Field Measured Rut Depth, in. HMA WMA Line of Equality R2=0.76 SEE=0.04 0.0 0.1 0.2 0.3 0.4 0.5 0.0 0.1 0.2 0.3 0.4 0.5 Va lid at ed R ut D ep th , i n. Field Measured Rut Depth, in. Validated Line of Equality R2=0.67 SEE=0.06 Figure 5.6. Rut depth statistical model development and validation: (a) relationship between predicted and field-measured rut depths and (b) validation of the rut depth model. 2nd Round Ranking Mix rutting resistance index, (â) Binder high PG, (â) Jnr0.1, Jnr3.2, (+) Mix E*, (â) Mix creep compliance, (+) Statistical Model Rutting resistance index, (â) Pavement age, (+) AADTT, (+) Total HMA thickness, (+) Overlay HMA thickness, (+) Table 5.2. Potential significant determinants for rutting.
46 Moisture Susceptibility Overall, no moisture-related distress, such as raveling, was found in the field during the first- or second-round distress surveys. Therefore, the moisture susceptibility of the mixes was analyzed based on the HWT test results only. It should be pointed out, though, that moisture damage is a long-term performance issue. The lack of moisture-related distress at the time of the distress surveys may not necessarily ensure the long-term moisture damage resistance of the HMA and WMA pavements. It is possible that moisture damage in the field had not progressed to the extent that moisture-related distress was evident. Figure 5.7 presents a summary of the num- bers of cycles to the stripping inflection points (SIPs); the projects with zero values shown in Figure 5.7 indicate that no SIP was observed. Out of the eight projects that showed SIPs within 20,000 passes, seven projects did not use anti- stripping agents. In other words, the only project that used an anti-stripping agent showed a potential for moisture dam- age. Such results suggest that anti-stripping agents may be generally useful. Future study is proposed to continue moni- toring these field projects for long-term moisture damage. Note: The project names are in color to signify the following: red indicates projects where no anti-stripping agents were applied, black indicates the projects where anti-stripping agents were applied, and green indicates that the anti-stripping agent information was not available. N o. o f P a ss es a t S IP Project Figure 5.7. Hamburg wheel tracking test results: number of cycles to stripping inflection point.
