National Academies Press: OpenBook
« Previous: Chapter 3 - Research Approach
Page 5
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 5
Page 6
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 6
Page 7
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 7
Page 8
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 8
Page 9
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 9
Page 10
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 10
Page 11
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 11
Page 12
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 12
Page 13
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 13
Page 14
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 14
Page 15
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 15
Page 16
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 16
Page 17
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 17
Page 18
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 18
Page 19
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 19
Page 20
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 20
Page 21
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 21
Page 22
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 22
Page 23
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 23
Page 24
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 24
Page 25
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 25
Page 26
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 26
Page 27
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 27
Page 28
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 28
Page 29
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 29
Page 30
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 30
Page 31
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 31
Page 32
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 32
Page 33
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 33
Page 34
Suggested Citation:"Chapter 4 - Test Results and Analysis." National Academies of Sciences, Engineering, and Medicine. 2004. Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements. Washington, DC: The National Academies Press. doi: 10.17226/13777.
×
Page 34

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.

54.1 PART 1—MIX DESIGNS FOR SPECIMENS TO STUDY THE EFFECT OF t/NMAS ON DENSITY Of the 36 mix designs, 27 were Superpave-designed mixes and 9 were SMA mixes. The Superpave mixes were classified according to three gradations: above the restricted zone (ARZ), through the restricted zone (TRZ), and below the restricted zone (BRZ). The optimum asphalt content, the effective asphalt content (Pbe), voids in mineral aggregate (VMA), voids filled with asphalt (VFA), percent theoretical maxi- mum density at Ninitial (% Gmm at Nini), and ratio of dust to effective asphalt content (P0.075/Pbe) for the Superpave mixes are summarized in Table 3. Data for SMA mixes are shown in Table 4. The mix design information for both mix types is presented in Appendix A. Optimum asphalt binder con- tent was chosen to provide 4 percent air voids at the design number of gyrations. However, for the 19-mm NMAS lime- stone SMA mix, 4 percent air voids could be achieved with 5.7 percent asphalt content, which did not meet the mini- mum asphalt content requirement in accordance with the “Standard Practice for Designing SMA,” AASHTO PP44-01. Therefore, the minimum asphalt content of 6.0 percent was chosen, which resulted in 3.7 percent air voids at the design number of gyrations. Some designs did not meet the re- quirements of VMA, VFA, % Gmm at Nini, and/or dust/Pbe. Efforts were made to redesign the respective mixes by changing the gradation until the requirements were met or closely approximated. This is important in that the mixes used in this project were intended to duplicate mixes uti- lized in the field. No modification was made for the TRZ mixes that did not meet the requirements, as little could be done to modify these gradations and still pass through the restricted zone. 4.2 EVALUATION OF EFFECT OF t/NMAS ON DENSITY USING GYRATORY COMPACTOR Before the evaluation was done, two methods of measuring density, or bulk specific gravity, were compared: the AASHTO T166 (SSD) and the vacuum sealing (ASTM D6752-02a) methods. All samples were measured using both methods. Fig- ures 1 through 4 present these measurements for the three gradations of Superpave mixes and the SMA mixes. As shown in Figure 1, the air voids for ARZ mixes as mea- sured by the two methods are approximately equal at low air voids and deviate by approximately 0.5 percent at the high- est air void level. This figure indicates that for ARZ mixes, the two methods provide similar results. For the TRZ, BRZ, and SMA mixes, Figures 2 through 4 suggest that the bulk specific gravity measurements derived from the two methods moved farther apart as density decreased. The results also indicate that, as the gradation became coarser, the difference in the test results for the two test methods increased. This finding agrees with the research by Cooley et al. (11). The apparent reasons for the different results according to the two test methods is loss of water during density measure- ment when using the T-166 method and the effect of surface texture. The loss of water when blotting in the T-166 method causes a test error resulting in higher measured density. The surface texture can result in the vacuum seal device measur- ing a lower density than the actual density. Because the vac- uum seal device is more accurate in measuring the density of porous samples, it was used to determine density for this research project. The main objective of this part of the study was to deter- mine the minimum t/NMAS. To achieve this objective, rela- tionships of average air voids for the three aggregate types versus t/NMAS with respect to NMAS and gradation were evaluated; the results are illustrated in Figures 5 through 10. Originally it was intended to determine the t/NMAS at which the air voids began to level out and to pick that t/NMAS level as the minimum level recommended to achieve optimum compaction. However much of the data in Figures 5 through 10 indicate that the air voids continue to drop with increasing t/NMAS past typical t/NMAS values. These data therefore did not provide reasonable guidance for selecting a mini- mum t/NMAS. Hence an air void content of 7.0 percent was selected as the criteria to determine the minimum t/NMAS. This level of air voids was selected because compaction of most pavements in the field is targeted at 92.0 to 94.0 per- cent of theoretical maximum density. Because of the uncer- tainty in the relationship of average air voids to t/NMAS, as indicated by the data, it was determined to compact some laboratory samples with a vibratory compactor and also to compact some mixes in the field during reconstruction of the NCAT test track. These two efforts, which are discussed later in the report, should provide sufficient information to make reasonable conclusions concerning desired t/NMAS levels. One potential problem with the Superpave gyratory com- pactor is that it applies a constant strain to the mix during com- paction and the force required varies as necessary to provide the desired strain. This is not the approach that is observed in CHAPTER 4 TEST RESULTS AND ANALYSIS

the field where the stress is constant and the strain varies. Hence, the Superpave gyratory compactor likely does not pro- vide a reasonable answer because the compaction provided by this device is different from the field. The big problem with using this concept to establish a minimum t/NMAS is that the voids continue to increase significantly as the t/NMAS increases, making it impossible to select an optimum value. The optimum t/NMASs established using the Superpave gyratory compactor vary from less than 2.5 up to approxi- mately 8. This wide range of numbers did not allow specific criteria to be established. Hence, additional testing was per- formed using the laboratory vibratory compactor and field test section. 6 4.3 EVALUATION OF EFFECT OF t/NMAS ON DENSITY USING VIBRATORY COMPACTOR After obtaining the results for the Superpave gyratory com- pactor, it was concluded that more tests needed to be con- ducted to better simulate compaction in the field. The air voids determined from the vacuum seal device were utilized in the analysis. To further evaluate the relationship between density and lift thickness, a similar study was conducted, but on a smaller scale, using the vibratory compactor as the compaction mode. This was not part of the original proposed work, but it was believed that the vibratory compactor might provide compaction that has more typical of in-place compaction. Aggregate NMAS, Gradation Optimum Pbe, VMA VFA % Gmm P0.075/Pbe mm Asphalt, % % % % at Nini 9.5 ARZ 6.7 6.2 18.4 76 89.0 0.8 9.5 BRZ 5.3 4.9 15.7 73 86.7 1.0 9.5 TRZ 5.4 5.0 15.6 75 88.9 1.0 19.0 ARZ 4.7 4.3 14.1 72 89.5* 1.2 Granite 19.0 BRZ 4.4 3.9 13.3 68 86.0 1.0 19.0 TRZ 4.0 3.6 12.5* 68 88.8 1.4* 37.5 ARZ 4.2 4.0 13.7 69 89.8* 0.8 37.5 BRZ 3.3 3.0 11.3 64 86.8 1.0 37.5 TRZ 3.6 3.3 12.0 65 88.1 0.9 9.5 ARZ 6.7 6.5 18.3 78* 88.4 0.8 9.5 BRZ 6.2 5.6 16.7 75 86.5 0.8 9.5 TRZ 6.0 5.4 16.3 75 87.7 0.9 19.0 ARZ 4.9 4.4 14.0 72 88.5 1.1 Gravel 19.0 BRZ 4.5 3.9 12.9* 69 86.3 1.3* 19.0 TRZ 4.4 3.8 12.8* 69 88.0 1.3* 37.5 ARZ 4.4 3.9 13.0 70 89.7* 0.8 37.5 BRZ 3.6 3.2 11.7 63 85.5 1.0 37.5 TRZ 3.9 3.5 12.0 66 85.6 0.9 9.5 ARZ 6.0 5.7 17.4 76 87.8 0.7 9.5 BRZ 5.0 4.6 15.3 72* 85.5 0.9 9.5 TRZ 4.4 4.2 14.4 70* 86 1.2 19.0 ARZ 4.1 3.5 12.6* 66 88.3 1.4* Limestone 19.0 BRZ 4.7 4.4 14.3 71 85.5 0.7 19.0 TRZ 3.3 2.8 11.0* 62* 85.7 1.8* 37.5 ARZ 3.2 3.1 11.8 64 88.8 1.0 37.5 BRZ 2.7 2.6 10.6* 60* 86.0 1.2 37.5 TRZ 2.8 2.6 10.6* 61* 87.7 1.1 * Did not meet Superpave Design Requirements TABLE 3 Summary of mix design results for Superpave mixes Aggregate NMAS, Optimum Pbe, VMA, VFA, VCAmix,a VCAdrc,b mm Asphalt, % % % % % % 9.5 7.2 6.6 18.7 78 30.9 41.9 Granite 12.5 6.6 6.4 18.8 77 30.3 42.7 19.0 6.4 5.9 17.6 77 29.6 42.0 9.5 7.3 6.5 18.6 77 30.4 41.8 Gravel 12.5 6.8 6.1 17.7 77 31.1 42.1 19.0 6.7 6.2 17.8 76 29.3 42.0 9.5 6.2 5.8 17.4 76 30.7 38.4 Limestone 12.5 7.4 7.0 19.6 80 31.1 38.9 19.0 6.0 5.6 16.8c 77 29.8 40.3 aVCA = Voids in Compacted Aggregate bdrc = dry-rodded compacted cDid not meet SMA Design Requirements TABLE 4 Summary of mix design results for SMA mixes

7y = 1.0576x + 0.0992 R2 = 0.9887 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 Air Voids, % - AASHTO T166 A ir Vo id s, % - Va cu um S ea l M et ho d y = 1.1074x + 0.3893 R2 = 0.978 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 Air Voids, % - AASHTO T166 A ir Vo id s, % - Va cu um S ea l M et ho d Figure 1. Relationship between air voids for ARZ mixes. Figure 2. Relationship between air voids for TRZ mixes. The vibratory compactor used compacted beam samples for the wheel-tracking device. Of the 36 mix designs analyzed for Part 1, 14 mixes were selected for further study. Two types of aggregates, granite and limestone were used. For Superpave designed mixes, two gradations were utilized (ARZ and BRZ) along with two NMASs (9.5 mm and 19.0 mm). The 37.5-mm NMAS mix was excluded from the study because the maximum thickness of the vibratory specimen that could be obtained was 75.0 mm, which would only be 2.0 t/NMAS. For the SMA mixes, three NMASs were selected (9.5 mm, 12.5 mm, and 19 mm). The t/NMAS ratios utilized were 2.0, 3.0, and 4.0. The compactive effort for each t/NMAS was varied over a range including 30 sec, 60 sec, and 90 sec of compaction. The range of com- pactive efforts was selected for two reasons: (1) there is no standard compactive effort for the vibratory compactor and

y = 1.6583x - 0.9272 R2 = 0.7185 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 Air Voids, % - AASHTO T166 A ir Vo id s, % - Va cu um S ea l M et ho d Figure 4. Relationship between air voids for SMA mixes. (2) the effects of compactive effort on density at different thicknesses could be evaluated. After compaction, the bulk specific gravity was measured and the data were analyzed to provide recommendations concerning the minimum t/NMAS. To determine the minimum t/NMAS, relationships between average air voids for the two types of aggregates and t/NMAS were plotted for each NMAS, compaction time, and grada- tion, as shown in Figures 11 through 17. In many cases there was very little difference between the densities for the dif- 8 ferent t/NMAS values. However, in a few cases there was a difference. Also, in many cases the best t/NMAS was 2.0, which is significantly lower than that observed on field proj- ects. Typically, it was assumed that coarse graded mixes would have a desired t/NMAS greater than fine-graded mixes. The results did not always follow that trend. It was judged that some fieldwork was necessary to validate the results with the Superpave gyratory compactor and with the vibratory compactor. y = 1.201x + 0.4379 R2 = 0.9264 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 Air Voids, % - AASHTO T166 A ir Vo id s, % - Va cu um S ea l M et ho d Figure 3. Relationship between air voids for BRZ Mixes.

9ARZ y = 22.729x-0.8484 R2 = 0.9771 BRZ y = 31.16x-0.8913 R2 = 0.9807 TRZ y = 32.71x-0.9062 R2 = 0.9998 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 t/NMAS Av er ag e A ir Vo id s, % ARZ BRZ TRZ ARZ y = 17.879x-1.0631 R2 = 0.9772 BRZ y = 26.047x-1.187 R2 = 0.9954 TRZ y = 24.593x-1.2151 R2 = 0.9963 0.0 2.0 4.0 6.0 8.0 10.0 12.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 t/NMAS Av er ag e A ir Vo id s, % ARZ BRZ TRZ ARZ y = 6.2838x-0.2991 R2 = 0.8546 BRZ y = 14.404x-0.841 R2 = 0.9874 TRZ y = 10.185x-0.7561 R2 = 0.9053 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 t/NMAS Av er ag e A ir Vo id s, % ARZ BRZ TRZ Figure 5. Relationships between air voids and t/NMAS for 9.5-mm Superpave mixes. Figure 6. Relationships between air voids and t/NMAS for 19.0-mm Superpave mixes. Figure 7. Relationships between air voids and t/NMAS for 37.5-mm Superpave mixes.

10 y = 36.496x-0.8298 R2 = 0.9988 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 t/NMAS A ve ra ge A ir Vo id s, % y = 29.982x-0.7232 R2 = 0.9945 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 t/NMAS Av er ag e Ai r V oi ds , % Figure 8. Relationships between air voids and t/NMAS for 9.5-mm SMA mixes. Figure 9. Relationships between air voids and t/NMAS for 12.5-mm SMA mixes. 4.4 EVALUATION OF EFFECT OF t/NMAS ON DENSITY FROM FIELD STUDY The field test sections consisted of 7 mixes that were to be placed on the test track. These mixes had to be verified before placing on the track; hence, these mixes could be placed and tested without significant costs. Some of the mixes did not meet volumetrics and other requirements, but they were judged sufficient for this part of the study because determin- ing the desired thickness range was a relative value based on t/NMAS. 4.4.1 Section 1 Section 1 was constructed on July 18, 2003, and con- sisted of a 2.0 to 5.0 t/NMAS overlay of an existing HMA layer. This construction was performed adjacent to the NCAT Test Track. The mix was a 9.5-mm NMAS fine-graded mix- ture. The length of the section was about 40 m, and the width was about 3.5 m. On some of the sections the placement began on the thick side and in some cases the placement began on the thin side. This technique was used so that there would be no bias due to the placement of the HMA. On this sec-

11 y = 27.14x-0.9206 R2 = 0.9974 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 t/NMAS Av er ag e A ir Vo id s, % 0 1 2 3 4 5 6 7 8 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 t/NMAS Av er ag e Ai r V oi ds , % 30 sec 60 sec 90 sec Figure 10. Relationships between air voids and t/NMAS for 19.0-mm SMA mixes. Figure 11. Relationships between air voids and t/NMAS for 9.5-mm ARZ mixes. tion the paving began with the thicker portion of the section and the thickness was slowly decreased as the paver moved down the test lane. The desired mat thickness was achieved by gradually adjusting the screed depth crank of the paver during the paving operation. The weather conditions during the paving were 84°F, overcast, with calm wind. The existing surface temperature prior to overlay was also 84°F. The roller utilized in this section was an 11-ton steel roller HYPAC C778B with a 78-in. wide drum that could operate in vibratory or static mode. The rubber tire roller available did not meet desired requirements for weight and tire pressure, and thus the data generated for the rubber tire roller compacted mixture were omitted from the analysis for this section. The breakdown rolling was performed with one pass in the static mode on the mat at a temperature of about 300°F. This was fol- lowed by three passes in the vibratory mode at low amplitude and high frequency (3800 vibrations per minute [vpm]) and one pass in the static mode. It was determined that this com- paction effort reached the peak density; hence, additional rolling was not performed.

12 0 2 4 6 8 10 12 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 t/NMAS Av er ag e Ai r V oi ds , % 30 sec 60 sec 90 sec 0 1 2 3 4 5 6 7 8 9 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 t/NMAS Av er ag e Ai r V oi ds , % 30 sec 60 sec 90 sec Figure 12. Relationships between air voids and t/NMAS for 9.5-mm BRZ mixes. Figure 13. Relationships between air voids and t/NMAS for 19.0-mm ARZ mixes. A total of 16 cores were obtained from this section and the test results of the cores are presented in Figure 18. The results include the thickness of cores, t/NMAS, and the air voids determined from the vacuum seal device. A review of the data indicated that a polynomial function provided the best fit line. The best-fit line indicates that the air voids decreased as the t/NMAS increased to a point where additional thickness resulted in increased air voids. The rec- ommended thickness range was selected as the point(s) where the air voids increased by 0.5 percent (less than 0.5 percent were considered insignificant). This number is somewhat arbitrary, but it is realistic. Therefore, as shown in Figure 18, the recommended t/NMAS range for 9.5-mm fine-graded mix was 3.4 to 5.8. This does not mean that satisfactory com- paction cannot be obtained outside of these limits, but it does indicate that more compactive effort would be needed. So this recommended range should only be used as a guide and should not be a rigid requirement. The effect of t/NMAS on

13 0 2 4 6 8 10 12 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 t/NMAS Av er ag e Ai r V oi ds , % 30 sec 60 sec 90 sec 0 2 4 6 8 10 12 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 t/NMAS Av er ag e Ai r V o id s, % 30s 60s 90s Figure 14. Relationships between air voids and t/NMAS for 19.0-mm BRZ mixes. Figure 15. Relationships between air voids and t/NMAS for 9.5-mm SMA mixes. the measured density was determined from Figure 18. Data in the figure indicate that the lowest air voids (7.0 percent air voids) occurred at t/NMA 4.4. Table 5 shows the air voids at various t/NMAs as related to this minimum. 4.4.2 Section 2 Section 2 was constructed on August 7, 2003, and the t/NMAS for this overlay ranged from 2.0 to 5.0. The mixture was a 9.5-mm NMAS coarse-graded mixture. The length of the section was about 40 m, and the width was about 3.5 m. The paving started from the thick portion of the mat and pro- gressed toward the thinner portion. The weather conditions during the paving were 82°F, overcast, with calm wind. The existing surface temperature was 96°F. The roller utilized in this section was an 11-ton steel drum roller HYPAC C778B with a 78-in. wide drum that could operate in vibratory or static mode. The rubber tire roller was a 15-ton HYPAC C560B with a tire pressure of 90 psi. For the side of the mat utilizing only the steel drum roller, the initial rolling was performed with four passes in the vibratory mode at low amplitude and high frequency (3800 vpm) at a mix tem-

14 0.0 2.0 4.0 6.0 8.0 10.0 12.0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 t/NMAS Av er ag e A ir Vo id s, % 30s 60s 90s 0.0 2.0 4.0 6.0 8.0 10.0 12.0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 t/NMAS Av er ag e A ir Vo id s, % 30s 60s 90s R2 = 0 . 6 3 9 2 3 . 0 4 . 0 5 . 0 6 . 0 7 . 0 8 . 0 9 . 0 1 0 . 0 0 . 0 1 . 0 2 . 0 3 . 0 4 . 0 5 . 0 6 . 0 7 . 0 t/NMAS Air Vo ids , % Figure 16. Relationships between air voids and t/NMAS for 12.5-mm SMA mixes. Figure 17. Relationships between air voids and t/NMAS for 19.0-mm SMA mixes. Figure 18. Relationships of air voids and t/NMAS for 9.5-mm fine-graded mix.

15 tire roller. The effect of t/NMAS on the measured density was determined from Figure 19. Data in the figure indicate that the lowest in-place air voids (10 percent air voids for the steel wheel roller only and 10.5 percent air voids for the steel and rubber tire rollers) occurred at t/NMAS of 4.7 for the steel wheel roller and 3.8 for the rubber and steel wheel roller. Table 6 shows the air voids at various t/NMAs as related to this minimum. 4.4.3 Section 3 Section 3 was constructed on July 25, 2003, and consisted of a 2.0 to 5.0 t/NMAS overlay of an existing HMA layer. The mix was a 9.5-mm NMAS SMA. The length of the sec- tion was about 40 m, and the width was about 3.5 m. The paving started from the thick portion of the mat and pro- gressed to the thinner portion. The desired mat thickness was achieved by gradually adjusting the screed depth crank of the paver during the operation. The weather conditions during the paving were 95°F, partly cloudy, with calm wind. The existing surface temperature was 115°F. The roller utilized in this section was an 11-ton steel drum roller HYPAC C778B with a 78-in. wide drum that could operate in vibratory or static mode. The rubber tire roller was a 15-ton HYPAC C560B with a tire pressure of 90 psi. For the side of the mat utilizing only the steel drum roller, the ini- tial rolling was performed with one pass in the static mode followed by five passes in the vibratory mode operated in low amplitude and high frequency (3800 vpm) on the mat having a mix temperature of about 320°F. This was followed with two passes in the static mode for the finish rolling. For the side of the mat that used a rubber tire roller as an intermediate roller, the breakdown rolling was performed with one pass in the static mode and four passes in the vibratory mode oper- ated in low amplitude and high frequency (3800 vpm). This t/NMA Percentage points above lowest 4.4 (lowest air voids, 7.0 %) 2 2.5 3 1.0 0.0 4 0.1 5 0.1 TABLE 5 Relationship of air voids and t/NMAS for 9.5-mm fine-graded HMA compacted with steel roller Steel Roller R2 = 0.68 Steel/Rubber Tire Roller R2 = 0.5115 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 t/NMAS A ir V oi ds , % Figure 19. Relationships of air voids and t/NMAS for 9.5-mm coarse-graded mix. perature of about 300°F. This was followed with four passes in the static mode. For the side of the mat that used a rubber tire roller as an intermediate roller, the breakdown rolling was performed with four passes in the vibratory mode operated at low amplitude and high frequency (3800 vpm). This was fol- lowed with five passes of the rubber tire roller and one pass of the steel roller in the static mode. A total of 15 cores were obtained from the side that uti- lized only a steel drum roller and 16 cores from the side that used the rubber tire roller. The relationship of air voids measured from the vacuum seal device and t/NMAS was evaluated for each rolling pattern. The results are illustrated in Figure 19. A review of the data indicated that a polynomial function provided the best fit. As the thickness increased, the air voids decreased until a point where additional thickness resulted in increased air voids. The plots also suggest that the side uti- lizing only a steel drum compactor had better compaction. To determine the desired thickness, it was decided to use air voids 0.5 percent larger (a void level less than 0.5 percent dif- ferent was not considered significantly different) than the minimum air voids from the best-fit line. Therefore, as shown in Figure 19, the desired t/NMAS range for 9.5-mm coarse- graded mix was 3.5 to 5.9 for compaction with a steel wheel roller and 2.9 to 4.6 for compaction with the steel and rubber

16 Steel roller Steel and rubber tire rollers t/NMA Percentage points above lowest t/NMA Percentage points above lowest 4.7 (lowest air voids, 10.0 %) 3.8 (lowest air voids, 10.5 %) 2 2.5 2 2.0 3 1.0 3 0.5 0.0 0.0 4 0.5 4 0.0 5 0.0 5 1.0 TABLE 6 Relationship of air voids and t/NMAS for 9.5-mm coarse-graded HMA compacted with steel roller and with steel and rubber tire rollers Steel Roller R2 = 0.8335 Steel/Rubber Tire Roller R2 = 0.1864 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 t/NMAS A ir Vo id s, % Figure 20. Relationships of air voids and t/NMAS for 9.5-mm SMA mix. was followed with eight passes of the rubber tire roller and two passes of the steel wheel roller in the static mode. A total of 12 cores were obtained from the side that utilized only the steel drum roller and another 12 cores from the side that used the rubber tire roller. To determine the range of rec- ommended t/NMAS for this mix, the relationship of air voids from the vacuum seal device and t/NMAS was evaluated for each rolling pattern. The results are illustrated in Figure 20. The best-fit lines indicate that the air voids decreased as the thickness increased to a point where additional thickness resulted in increased air voids. The plots also suggest that the side utilizing only the steel drum compactor had higher den- sity. Rubber tire rollers are not used on SMA mixtures and these data confirm that there is no need to use the rubber tire roller. As shown in Figure 20, the recommended range for t/NMAS for the 9.5-mm SMA mix is 3.8 to 5.3 for the com- paction with a steel wheel roller and 2.6 to 5.1 for compaction with a steel and rubber tire roller. The effect of t/NMAS on the measured density was determined from Figure 20. Data in the figure indicate that the lowest in-place air voids (8.5 per- cent air voids for the steel wheel roller only and 10.3 percent air voids for the steel and rubber tire rollers) occurred at t/NMAS of 4.5 for the steel wheel roller and 3.8 for the rubber and steel wheel roller. Table 7 shows the air voids at various t/NMAs as related to this minimum. 4.4.4 Section 4 Section 4 was constructed on August 12, 2003, and con- sisted of a 2.0 to 5.0 t/NMAS overlay of an existing HMA layer. The mix was a 12.5-mm NMAS SMA. The length of the section was about 40 m, and the width was about 3.5 m. The paving started from the thinner portion and proceeded toward the thicker portion of the mat. The weather conditions during the paving were 80°F, overcast, with calm wind. The existing surface temperature was 85°F. The roller utilized in this section was an 11-ton steel drum roller HYPAC C778B with a 78-in. wide drum that could operate in vibratory and static modes. The rubber tire roller was a 15-ton HYPAC C560B with a tire pressure of 90 psi. For the side of the mat utilizing only the steel drum roller, the initial rolling was performed with four passes in the vibratory mode operated at low amplitude and high frequency (3800 vpm). The mat temperature was approximately 320°F. This was followed with three passes in the static mode includ- ing finish rolling. For the side of the mat that used a rubber tire roller as an intermediate roller, the initial rolling was per- formed with four passes in the vibratory mode operated at low amplitude and high frequency (3800 vpm). This was fol- lowed with four passes of the rubber tire roller and one pass of the steel roller in the static mode.

17 Steel roller Steel and rubber tire rollers t/NMA Percentage points above lowest t/NMA Percentage points above lowest 4.5 (lowest air voids, 8.5 %) 3.8 (lowest air voids, 10.3 %) 2 5.5 2 1.2 3 2.0 3 0.2 0.0 0.0 4 0.2 4 0.0 5 0.2 5 0.5 TABLE 7 Relationship of air voids and t/NMAS for 9.5-mm SMA mix compacted with steel roller and with steel and rubber tire rollers Steel Roller R2 = 0.87 Steel/Rubber Tire Roller R2 = 0.77 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 t/NMAS A ir Vo id s, % Figure 21. Relationships of air voids and t/NMAS for 12.5-mm SMA Mix. A total of 21 cores were obtained from the side that uti- lized only a steel drum roller and 21 cores from the side that used the rubber tire roller. To determine the recommended t/NMASs for this mix, the relationship of air voids from the vacuum seal device and t/NMAS was evaluated for each rolling pattern. The results are illustrated in Figure 21. The best-fit lines indicate that the air voids decreased as the thickness increased to a point where additional thickness resulted in increased air voids. The plots also suggest that the side utilizing only the steel drum compactor had higher den- sity. As shown in Figure 21, the suggested minimum t/NMAS for 12.5-mm SMA mix is 3.8 for compaction with steel wheel roller and 4.6 for compaction with steel and rubber tire roll- ers. For these mixes, the density increased as the t/NMAS increased even at the thicker portions. Also the curve did not fit the data as well as desired, so the data points were actually used to select the suggested t/NMAS number. Note in the plots that the data points continue downward with increasing t/NMAS to a point and then the air voids remain relatively constant as the t/NMAS increased. The effect of t/NMAS on the measured density was deter- mined from Figure 21. Data in the figure indicate that the low- est in-place air voids (4.7 percent air voids for the steel wheel roller only and 7.5 percent air voids for the steel and rubber tire rollers) occurred at t/NMAS of 4.5 for the steel wheel roller and 4.8 for the rubber and steel wheel rollers. Table 8 shows the air voids at various t/NMAs as related to this minimum. 4.4.5 Section 5 Section 5 was constructed on July 16, 2003, and consisted of a 2.0 to 5.0 t/NMAS overlay of an existing HMA. The mix consisted of a 19.0-mm NMAS fine-graded HMA. The length of the section was about 40 m, and the width was about 3.5 m. The paving started on the thin end of the section and pro- ceeded to the thicker portion. The desired mat thickness was achieved by gradually adjusting the screed depth crank of the paver during the operation. The weather conditions dur- ing the paving were 90°F, clear, with calm wind. The existing surface temperature was 96°F. The roller utilized in this section was an 11-ton steel roller HYPAC C778B with a 78-in. wide drum that operated in vibratory and static modes. The rubber tire roller used did not meet the tire pressure requirements and the results were omit- ted from the analysis for this section. The breakdown rolling was performed with four passes in the vibratory mode oper- ated in low amplitude and high frequency (3800 vpm). The

18 Steel roller Steel and rubber tire rollers t/NMA Percentage points above lowest t/NMA Percentage points above lowest 4.5 (lowest air voids, 4.7 %) 4.8 (lowest air voids, 7.5 %) 2 11.3 2 6.5 3 3.3 3 3.5 0.0 0.0 4 0.3 4 0.5 5 0.5 5 0.0 TABLE 8 Relationship of air voids and t/NMAS for 12.5-mm SMA mix compacted with steel roller and with steel and rubber tire rollers R2 = 0.77 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 t/NMAS A ir Vo id s, % Figure 22. Relationships of air voids and t/NMAS for 19.0-mm fine-graded mix. mat temperature was approximately 300°F. Three passes in the static mode and one pass for finish rolling followed this initial rolling. A total of 20 cores were obtained from this section. To determine the minimum t/NMAS for this mix, the relationship between air voids (from the vacuum seal device) and thickness was evaluated. The results are illustrated in Figure 22. The best-fit line indicated that the air voids decreased as the thickness increased to a point where additional thickness resulted in increased air voids. As shown in Figure 22, the rec- ommended t/NMAS range for the 19.0-mm fine-graded mix was 3.1 to 4.6. The effect of t/NMAS on the measured density was determined from the figure. Data in the figure indicate that the lowest in-place air voids (6.2 percent air voids) occurred at t/NMAS of 3.8. Table 9 shows the air voids at various t/NMAs as related to this minimum. 4.4.6 Section 6 Section 6 was constructed on August 6, 2003, and consisted of a range of 2.0 to 5.0 t/NMAS overlay of an existing HMA. The mix was a 19.0-mm NMAS coarse-graded HMA. The length of the section was about 40 m, and the width was about 3.5 m. The paving started from the thinner portion of the mat and proceeded to the thicker portion. The weather conditions during the paving were 79°F, cloudy, with calm wind. The existing surface temperature was 84°F. The roller utilized in this section was an 11-ton steel drum roller HYPAC C778B with a 78-in. wide drum that could operate in vibratory and static mode. The rubber tire roller was a 15-ton HYPAC C560B with a tire pressure of 90 psi. For the side of the mat utilizing only the steel drum roller, the initial rolling was performed with four passes in the vibratory mode operated at low amplitude and high frequency (3800 vpm). The mat temperature was approximately 300°F. This initial rolling was followed with six passes in the static mode. For the side of the mat that used a rubber tire roller as the intermedi- ate roller, the initial rolling was performed with four passes in the vibratory mode operated in low amplitude and high fre- quency (3800 vpm). This initial rolling was followed with four passes of the rubber tire roller and two passes with a steel wheel roller in the static mode. A total of 22 cores were obtained from the side that utilized only a steel drum roller and 16 cores from the side that used the rubber tire roller. To determine the minimum t/NMAS for this mix, the relationship between air voids from vacuum seal

19 HMA and utilized a modified asphalt. The length of the sec- tion was about 40 m, and the width was about 3.5 m. The paving started from the thicker portion of the mat and pro- ceeded to the thinner portion. The weather conditions dur- ing the paving were 90°F, clear, with calm wind. The existing surface temperature was 120°F. The roller utilized in this section was an 11-ton steel drum roller HYPAC C778B with a 78-in. wide drum that could operate in the vibratory and static modes. The rubber tire roller was a 15-ton HYPAC C560B with a tire pressure of 90 psi. For the side of the mat utilizing only the steel drum roller, the initial rolling was performed with four passes in the vibratory mode operated in low amplitude and high fre- quency (3800 vpm). The mat temperature was about 330°F. This was followed with another five passes in the vibra- tory mode operated at low amplitude and high frequency (3800 vpm). There was one additional pass with the steel wheel roller in the static mode to finish the mat. For the side of the mat that used a rubber tire roller as an intermediate roller, the initial rolling was performed with two passes in the vibratory mode operated at low amplitude and high fre- quency (3800 vpm). This was followed with ten passes with t/NMA Percentage points above lowest 3.8 (lowest air voids, 6.2 %) 2 3.1 3 0.6 0.0 4 0.0 5 1.3 TABLE 9 Relationship of air voids and t/NMAS for 19.0-mm fine-graded mix compacted with steel roller Steel Roller R2 = 0.1601 Steel/Rubber Tire Roller R2 = 0.4489 0 . 0 1 . 0 2 . 0 3 . 0 4 . 0 5 . 0 6 . 0 7 . 0 8 . 0 9 . 0 1 0 . 0 1 1 . 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 t/NMAS A ir Vo id s, % Figure 23. Relationships of air voids and t/NMAS for 19.0-mm coarse-graded mix. device and thickness was evaluated for each rolling pattern. The results are illustrated in Figure 23. The best-fit lines indi- cate that the air voids decreased as the thickness increased to a point where additional thickness resulted in increased air voids. The plots also suggest that the side utilizing the rubber tire roller had higher density. As shown in Figure 23, the rec- ommended minimum thickness for 19.0-mm coarse-graded mix was 3.0 for compaction with the steel and rubber tire rollers. There is too much scatter in the data to make a good selection of a recommended value for compaction with a steel wheel roller. The effect of t/NMAS on the measured density was deter- mined from Figure 23. Data in the figure indicate that the low- est in-place air voids (5.7 percent for the steel and rubber tire roller, the steel wheel roller alone was not used because it pro- duced too much scatter in the data) occurred at t/NMAS of 4.5. Table 10 shows the air voids at various t/NMAs as related to this minimum. 4.4.7 Section 7 Section 7 was constructed on August 14, 2003, and con- sisted of a range of 2.0 to 5.0 t/NMAS overlay of an existing HMA. The mix consisted of a 19.0-mm NMAS coarse-graded t/NMA Percentage points above lowest 4.5 (lowest air voids, 5.7 %) 2 1.8 3 0.6 0.0 4 0.1 5 0.1 *The steel wheel roller alone was not used because it produced too much scatter in the data Table 10 Relationship of air voids and t/NMAS for 19.0-mm coarse-graded mix compacted with steel and rubber tire roller*

the rubber tire roller and two passes of the steel wheel roller in the static mode. A total of 23 cores were obtained from the side that utilized only the steel drum roller and 26 cores from the side that used the rubber tire roller. To determine the minimum t/NMAS for this mix, the relationship of air voids from the vacuum seal device and t/NMAS was evaluated for each rolling pattern. The results are illustrated in Figure 24. The best-fit lines indicate that the air voids decreased as the thickness increased to a point where additional thickness resulted in increased air voids. The plots also suggested that the side utilizing only the steel drum compactor had higher density. As shown in Figure 24, the minimum t/NMAS range for 19.0-mm coarse-graded with modified asphalt mix was 3.4 to 4.8. The effect of t/NMAS on the measured density was determined from Figure 24. Data in the figure indicate that the lowest in-place air voids (5.6 percent air voids for the steel wheel roller only and 7.4 percent air voids for the steel and rubber tire rollers) occurred at t/NMAS of 4.2 for the steel wheel roller and 5.3 for the rubber and steel wheel roller. Table 11 shows the air voids at various t/NMAs as related to this minimum. 4.4.8 Summary In summary, the data for the seven sections appear to be reasonable and to match past experience. A summary of the results compared to the t/NMAS for lowest voids is provided in Table 12. These results indicate that the t/NMAS should be somewhere between 3 and 5 for best results. Based on the lim- ited data, a t/NMAS of 3 is probably reasonable for fine-graded mixes, because there is less than 1 percentage point change in density when the t/NMAS is reduced from optimum to 3.0. 20 The t/NMAS should be set at 4.0 for coarse-graded mixes due to the significant increase in voids when reducing the t/NMAS from optimum down to 3.0. 4.5 EVALUATION OF THE EFFECT OF TEMPERATURE ON THE RELATIONSHIP BETWEEN DENSITY AND t/NMAS Three locations were selected for temperature measure- ments for each section in the field experiment; one near the beginning of the section, one near the middle, and one near the end of the section. To determine the effect of mix temperature on the density, the temperature at 20 minutes after placement of the mix at each location was selected because this provides a reasonable compaction time. Because the mixes in this study used two different types of asphalt binder, PG 67-22 and PG 76-22, the temperatures at 20 min- utes were normalized by subtracting the high temperature grade of the asphalt type from the temperatures at 20 min- utes. Table 13 presents the t/NMAS, the average tempera- ture readings at 20 minutes, the asphalt high temperature grade, and the difference between mix temperature and high temperature grade. The differences in temperature were plot- ted against the t/NMAS together with the core densities for each section, as shown in Figures 25 through 31. The relationship between density and t/NMAS for all sections is shown in Figure 32. The best-fit line has an R2 of 0.26 and indicates that the density increased as the thick- ness increased to a point where additional thickness resulted in a decrease in density. The effect of the layer thickness and cooling time on mix temperature is provided in Figure 33. The data were obtained from the thermocouples installed in the pavement. This plot indicates that, during hot weather, Steel/Rubber Tire Roller R2 = 0.81 Steel Roller R2 = 0.65 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 t/NMAS A ir Vo id s, % Figure 24. Relationships of air voids and t/NMAS for 19.0-mm coarse-graded mix with modified asphalt.

21 Steel roller Steel and rubber tire rollers t/NMA Percentage points above lowest t/NMA Percentage points above lowest 4.2 (lowest air voids, 5.6 %) 5.3 (lowest air voids, 7.4 %) 2 4.9 2 6.1 3 1.3 3 3.4 0.0 0.0 4 0.0 4 0.8 5 0.8 5 0.0 TABLE 11 Relationship of air voids and t/NMAS for 19.0-mm coarse-graded mix with modified asphalt compacted with steel roller and with steel and rubber tire rollers Description of Mix Increase in Air Voids for t/NMAS=2 Increase in Air Voids for t/NMAS=3 Increase in Air Voids for t/NMAS=4 Increase in Air Voids for t/NMAS=5 Section 1-9.5mm Fine Graded— Steel Roller 2.5% 1.0% 0.1% 0.1% Section 2-9.5mm Coarse Graded- Steel Roller 2.5% 1.0% 0.5% 0.0% Section 2-9.5mm Coarse Graded- Steel and Rubber Roller 2.0% 0.5% 0.0% 1.0% Section 3-9.5mm SMA(mod AC) Steel Roller 5.5% 2.0% 0.2% 0.2% Section 3-9.5mm SMA(Mod AC) Steel & Rubber Roller 1.2% 0.2% 0.0% 0.5% Section 4- 12.5mm SMA (mod AC) Steel Roller 11.3% 3.3% 0.3% 0.5% Section 4- 12.5mm SMA (mod AC) Steel & Rubber Roller 6.5% 3.5% 0.5% 0.0% Section 5-19mm Fine Graded Steel Roller 3.1% 0.6% 0.0% 1.3% Section 6-19mm Coarse Graded Steel and Rubber Roller 1.8% 0.6% 0.1% 0.1% Section 7-19mm Coarse Graded (mod AC) Steel Roller 4.9% 1.3% 0.0% 0.8% Section 7-19mm Coarse Graded (mod AC) Steel & Rubber Roller 6.1% 3.4% 0.8% 0.0% TABLE 12 Effect of t/NMAS on compactibility of HMA compaction time for a layer thickness of 1.5 in. is approxi- mately twice that for a 1-in. layer. This clearly shows that one of the problems in obtaining density is layer thickness regardless of the t/NMAS. If the amount of compaction time is reduced by 50 percent, it may be very difficult to compact the mixture to an adequate density. To place the same amount of compactive effort on an HMA mixture prior to cooling to some defined temperature will take twice as many rollers at a 1-in. thickness as that required for a 1.5-in. surface. It is likely to be significantly more difficult to compact a 1-in. layer than to compact a 1.5-in. layer simply because of the cooling rate.

22 gyrations were increased up to 300 gyrations. This shows the difficulty of compacting mixes at thinner lifts in the gyratory mold. Permeability testing was only performed on specimens that met the desired air voids. The results were very limited, but, did show that generally the coarser mixes (larger maxi- mum aggregate size or higher percentage of coarse aggregate) had higher permeabilities. 4.7 EVALUATION OF EFFECT OF t/NMAS ON PERMEABILITY USING VIBRATORY COMPACTOR All specimens compacted at t/NMAS of 2.0, 3.0, and 4.0 did achieve the target air void content, which was 7 ± 1.0 percent. Figure 34 shows the relationship between average permeabil- ity for the two aggregate types and t/NMAS. In general, the permeability decreased as t/NMAS increased. Most of the mixes had permeability values fewer than 50 × 10−5 cm/sec. However, at t/NMAS equal to 2.0, the 9.5-mm and 12.5-mm NMAS SMA mixes had average permeability values of 173 × 10−5 cm/sec and 196 × 10−5 cm/sec, respectively. These values for the SMA exceed the recommended maximum permeability value of 125 × 10−5 cm/sec. It appears from these data that a specification requirement of 7 percent air voids would be acceptable for all of the mixes if the t/NMAS is 3 or greater. The likely reason that the thinner samples have high permeability is that the voids are more likely to be inter- connected all the way through the samples when the samples are thinner. Hence when mixes are placed thin, in this case Section/Mix Temp. at Asphalt Difference 20 min., °C Grade, PG 1 2.5 60 67 -7 9.5mmFG 3.6 82 67 15 5.1 95 67 28 2 2.1 64 67 -3 9.5mmCG 2.4 72 67 5 5.1 105 67 38 3 2.2 65 76 -11 9.5mmSMA 3.7 100 76 24 5.2 112 76 36 4 2.2 72 76 -4 12.5mmSMA 3.1 118 76 42 3.8 120 76 44 5 2.6 124 67 57 19mmFG 3.0 122 67 55 5.2 130 67 63 6 2.1 82 67 15 19mmCG 3.2 120 67 53 5.1 118 67 51 7 2.7 86 76 10 19mmCG 3.8 120 76 44 5.2 142 76 66 TABLE 13 t/NMAS, temperature in C at 20 min., asphalt high temperature grade, and difference in temperature %Lab Density Steel Roller R2 = 0.6392 Difference in Temperature R2 = 0.9821 93.0 93.5 94.0 94.5 95.0 95.5 96.0 96.5 97.0 97.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 t/NMAS % La b De ns ity -20 -10 0 10 20 30 40 D iff er en ce in Te m pe ra tu re , o C Steel Wheel Roller Difference in Temperature Figure 25. Relationships between density, t/NMAS, and temperature for Section 1. 4.6 EVALUATION OF EFFECT OF t/NMAS ON PERMEABILITY USING GYRATORY COMPACTOR Specimens were compacted to 7.0 ± 1.0 percent air void content at t/NMAS of 2.0, 3.0, and 4.0. For most mixes, spec- imens could not achieve the target air voids even when the

23 Difference in Temperature. R2 = 0.998 %Lab Density Steel Roller R2 = 0.6796 %Lab Density Steel/Rubber Tire Roller R2 = 0.5115 88.0 89.0 90.0 91.0 92.0 93.0 94.0 95.0 96.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 t/NMAS % La b De ns ity -20.0 -10.0 0.0 10.0 20.0 30.0 40.0 50.0 D iff er en ce in Te m pe ra tu re ,o C Figure 26. Relationships between density, t/NMAS, and temperature for Section 2. Difference in Temperature R2 = 0.976 %Lab Density Steel Roller R2 = 0.8335 %Lab Density Steel/Rubber Tire Roller R2 = 0.1864 90.0 91.0 92.0 93.0 94.0 95.0 96.0 97.0 98.0 99.0 100.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 t/NMAS % La b De n si ty -30 -20 -10 0 10 20 30 40 50 D iff er en ce in T em pe ra tu re ,o C Steel Wheel Roller Rubber Tire Roller Difference in Temperature Figure 27. Relationships between density, t/NMAS, and temperature for Section 3.

24 Difference in Temperature R2 = 0.8884 %Lab Density Steel Roller R2 = 0.8711 %Lab Density Steel/Rubber Tire Roller R2 = 0.7651 84.0 85.0 86.0 87.0 88.0 89.0 90.0 91.0 92.0 93.0 94.0 95.0 96.0 97.0 98.0 99.0 100.0 101.0 102.0 103.0 104.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 t/NMAS % La b De ns ity -40 -20 0 20 40 60 80 100 D iff er en ce in T em pe ra tu re , o C Steel Wheel Roller Rubber Tire Roller Difference in Temperatur e Figure 28. Relationships between density, t/NMAS, and temperature for Section 4. %Lab Density Steel Roller R2 = 0.7736 Different in Temperature @ 20 Min. R2 = 0.8191 95.0 96.0 97.0 98.0 99.0 100.0 101.0 102.0 103.0 104.0 105.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 t/NMAS % La b D en si ty 0 10 20 30 40 50 60 70 D iff er en ce in T em pe ra tu re ,o C Figure 29. Relationships between density, t/NMAS, and temperature for Section 5.

25 Difference in Temperature R2 = 0.6816 Steel/Rubber Tire Roller R2 = 0.4489 93 94 95 96 97 98 99 100 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 t/NMAS % La b De ns ity 0 10 20 30 40 50 60 70 D iff er en ce in T em pe ra tu re , ° C Steel/Rubber Tire Roller Difference in Temperature Difference in Temperature R2 = 0.9904 %Lab Density Steel Roller R2 = 0.6529 %Lab Density Steel/Rubber Tire Roller R2 = 0.8092 87.0 88.0 89.0 90.0 91.0 92.0 93.0 94.0 95.0 96.0 97.0 98.0 99.0 100.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 t/NMAS % La b De ns ity -40 -30 -20 -10 0 10 20 30 40 50 60 70 D iff er en ce in Te m pe ra tu re ,° C Steel Roller Steel/Rubber Tire Roller Difference in Temperature Figure 30. Relationships between density, t/NMAS, and temperature for Section 6. Figure 31. Relationships between density, t/NMAS, and temperature for Section 7.

26 y = -0.8216x2 + 7.2531x + 80.848 R2 = 0.2562 86.0 88.0 90.0 92.0 94.0 96.0 98.0 100.0 102.0 104.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 t/NMAS % La b De ns ity Figure 32. Relationships between density and t/NMAS for all sections. 0 20 40 60 80 100 120 140 160 180 0 10 20 30 40 50 60 70 Time, min. Te m pe ra tu re o f M ix , ° C 25 mm (1.0") 32 mm (1.25") 38 mm (1.5") 44 mm (1.75") 51 mm (2.0") 64 mm (2.5") 89 mm (3.5") Figure 33. The effect of layer t/NMAS and cooling time on mix temperature.

27 0 50 100 150 200 250 2:1 3:1 4:1 2:1 3:1 4:1 2:1 3:1 4:1 2:1 3:1 4:1 2:1 3:1 4:1 2:1 3:1 4:1 2:1 3:1 4:1 9.5 ARZ 9.5 BRZ 9.5 SMA 12.5 SMA 19.0 ARZ 19.0 BRZ 19.0 SMA t/NMAS @ each mix A ve ra ge P er m ea bi lit y x E- 05 c m /s ec Recommended maximum permeability, 125 x 10-5 Figure 34. Relationships between permeability and t/NMAS. Section Number Mix Type In-Place Air Voids (percent) Field Permeability (cm/s x 10-5) Lab Permeability (cm/s x 10-5) 1 9.5mm FG 6.6 to 8.8 1 to 28 1 to 35 2 9.5mm CG 9.0 to 12.6 14 to 632 107 to 1070 3 9.5mm SMA 7.7 to 12.6 110 to 651 29 to 168 4 12.5mm SMA 4.1 to 17.9 3 to 1778 0.1 to 5850 5 19.0mm FG 5.7 to 9.5 38 to 161 1 to 77 6 19.0mm CG 5.3 to 9.8 10 to 1760 1 to 141 7 19.0mm CG 4.8 to 15.2 72 to 3030 0 to 1203 TABLE 14 Comparison of laboratory and field permeabilities less than a 3:1 t/NMAS, the air voids have to be lower to ensure that the mixes are impervious. 4.8 EVALUATION OF EFFECT OF t/NMAS ON PERMEABILITY FROM FIELD STUDY Permeability tests were conducted on the seven HMA sec- tions that were evaluated in the field. These tests were con- ducted in-place with the field permeameter and in the labora- tory with the lab permeability test. Cores were taken from the in-place pavement for measurement of density and for mea- surement of lab permeability. The field permeability values were determined adjacent to the location where the cores were taken for density and for lab permeability. The results of these tests for the 7 sections are provided in Table 14. In summary, the coarse-graded mixes had permeability values that exceeded the recommended value when the air voids exceeded about 8 percent. The fine graded mixes never exceeded the recommended value even up 9 to 10 percent air voids. 4.9 PART 2—EVALUATION OF RELATIONSHIP OF LABORATORY PERMEABILITY, DENSITY AND LIFT THICKNESS OF FIELD COMPACTED CORES The average thickness, the average air void content by the vacuum seal device method, and the average laboratory permeability values were determined for each of the cores obtained from the work under NCHRP Project 9-9 (1). Figures 35 through 37 present the plots of in-place air voids versus permeability for each NMAS mix. The relationship between in-place air voids and permeability for 9.5-mm NMAS is illus- trated in Figure 35. The R2 values for both coarse-graded and fine-graded mixes were relatively high (0.70 and 0.86, respectively) and both relationships are significant (p-value = 0.000). At 8 percent air voids, the pavement is expected to have a permeability of 60 × 10−5 cm/sec for coarse-graded mix and 10 × 10−5 cm/sec for fine-graded mix. Because there are only a couple of data points for fine-graded mix above approximately 10 percent air voids, this model should not be used to predict permeability at these higher void levels. At

28 Coarse-graded y = 2.279e0.4225x R2 = 0.6942 Fine-graded y = 0.0309e0.7457x R2 = 0.8602 0 500 1000 1500 2000 2500 3000 3500 4000 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 24.0 In-place Air Voids (Corelok), % La b Pe rm ea bi lity , E -5 c m /s ec Coarse-Graded Fine-Graded Figure 35. Plot of permeability versus in-place air voids for 9.5-mm NMAS mixes. 0 100 200 300 400 500 600 700 800 900 1000 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 24.0 In-place Air Voids (Corelok), % La b Pe rm ea bi lity , E -5 c m /s ec 12.5 mm Coarse-Graded 12.5 mm Fine-Graded Figure 36. Plot of permeability versus in-place air voids for 12.5-mm NMAS Mixes. lower void levels the coarse-graded mixes are more permeable than fine-graded mixes. The relationships for the coarse-graded and fine-graded 12.5-mm NMAS mixes are shown in Figure 36. For these projects there was no significant difference between fine and coarse graded mixes. The relationships between in-place air voids and permeability for both gradation types were reason- able and significant with an R2 of 0.61 for coarse-graded mixes (p-value = 0.000) and 0.58 for fine-graded mixes (p-value = 0.000). As shown by the best-fitted lines, the permeability values for both gradation types were basically the same at a given air void content. The permeability at 8.0 percent air voids for coarse-graded and fine-graded mixes was approximately 30 × 10−5 cm/sec. Figure 37 illustrates the relationship between in-place air voids and permeability for fine-graded 19.0-mm NMAS mixes. The R2 value for this figure is 0.59 and the relationship is significant (p-value = 0.000). Based on the trend line, per- meability is very low at air void contents less than 8 percent. At air void contents above 8 percent, the permeability begins to increase rapidly with a small increase in in-place air void content. At 8 percent air voids, the fine-graded 19.0-mm NMAS mix has a permeability value of 16 × 10−5 cm/sec. 4.10 CONTROLLED LABORATORY EXPERIMENT TO EVALUATE METHODS OF MEASURING THE BULK SPECIFIC GRAVITY OF COMPACTED HMA 4.10.1 Introduction and Problem Statement A major concern of the HMA industry is the proper mea- surement of bulk specific gravity (Gmb) for compacted samples. This issue has become a bigger problem with the increased

29 use of coarse gradations. Bulk specific gravity measurements are the basis for volumetric calculations used during HMA mix design, field control, and construction acceptance. Dur- ing mix design, volumetric properties such as air voids, voids in mineral aggregates, voids filled with asphalt, and percent theoretical maximum density at a certain number of gyrations are used to evaluate the acceptability of mixes. All of these properties are based upon Gmb. In most states, acceptance of HMA construction by the owner is typically based upon percent compaction (density based upon Gmb and theoretical maximum density). Whether nondestructive (e.g., nuclear gauges) or destructive (e.g., cores) tests are used as the basis of acceptance, Gmb measurements are equally important. When nondestructive devices are uti- lized, each device first has to be calibrated to the Gmb of cores. If the Gmb measurements of the cores are inaccurate in this calibration step, then the nondestructive device will provide inaccurate data. Additionally, pay factors for construction, whether reductions or bonuses, are generally based upon percent compaction. Thus, errors in Gmb measurements can potentially affect both the agency and producer. For many years, the measurement of Gmb for compacted HMA has been accomplished by the water displacement method using saturated-surface dry (SSD) samples. This method consists of first weighing a dry sample in air, then obtaining a submerged mass after the sample has been placed in a water bath for a specified time interval. Upon removal from the water bath, the SSD mass is determined after patting the sample dry using a damp towel. Procedures for this test method are outlined in AASHTO T166 (ASTM D2726). The SSD method has proven to be adequate for conven- tionally designed mixes, such as those designed according to the Marshall and Hveem methods, that generally utilized fine- graded aggregates. Historically, mixes were designed to have gradations passing close to or above the Superpave defined maximum density line (i.e., fine-graded). However, since the adoption of the Superpave mix design system and the increased use of SMA, mixes are being designed with coarser gradations than in the past. The potential problem in measuring the Gmb of mixes like coarse-graded Superpave and SMA using the SSD method comes from the internal air void structure within these mix types. These types of mixes tend to have larger internal air voids than the finer conventional mixes, at similar overall air void contents. Mixes with coarser gradations have a much higher percentage of large aggregate particles. At a certain overall air void volume, which is mix specific, the large internal air voids of the coarse mixes can become inter- connected. During Gmb testing with the SSD method, water can quickly infiltrate into the sample through these intercon- nected voids. However, after removing the sample from the water bath to obtain the saturated-surface dry condition the water can also drain from the sample quickly. This drain- ing of the water from the sample is what causes errors when using the SSD method. Because of the potential errors noted with the saturated surface-dry test method of determining the bulk specific grav- ity of compacted HMA, the primary objective of this task was to compare AASHTO T166 with other methods of measur- ing bulk specific gravity to determine under what conditions AASHTO T166 is accurate. The plan for this part of the study was to evaluate two sep- arate sample types: laboratory compacted and field compacted. Laboratory compacted mixtures having various aggregate types, nominal maximum aggregate sizes, gradation shapes, and air void levels were prepared. Each of the prepared samples was tested to determine bulk specific gravity by four different test methods: water displacement, vacuum-sealing, gamma ray, and dimensional. y = 0.0437e0.7362x R2 = 0.5923 0 20 40 60 80 100 120 140 0.0 2.0 4.0 6.0 8.0 10.0 12.0 In-place Air Voids (Corelok), % La b Pe rm ea bi lity , E -5 c m /s ec Figure 37. Plot of permeability versus in-place air voids for 9.5-mm NMAS mixes.

For the field compacted samples, cores obtained during the field validation portion of this study were subjected to the same four bulk specific gravity test methods. Because cores have a different surface texture than laboratory compacted samples, it was necessary to evaluate them also. Testing also conducted on core samples included laboratory permeabil- ity tests and effective air void content using the vacuum- sealing device. 4.10.2 Field Compacted Samples Each of the cores obtained during the Task 5 field valida- tion were tested to determine bulk specific gravity using the same four tests as the laboratory experiment: water displace- ment, vacuum sealing, gamma ray, and dimensional analysis. Because of the differences in surface texture between labora- tory compacted samples (surface texture around entire sam- ple) and field compacted samples (surface texture only on top of sample because of core bit and sawing), the experiment was also extended to core samples. Because of the differences in resulting air voids for the four methods of measuring bulk specific gravity, a Duncan’s multiple range test (DMRT) was conducted to determine which methods, if any, provided similar results. This analy- sis method provides a ranking comparison between the dif- ferent methods. The range of sample means for a given set of data (method) can be compared to a critical valued based on the percentiles of the sampling distribution. The critical value is based on the number of means being compared (four, rep- resenting the different methods) and number of degrees of freedom at a given level of significance (0.05 for this analy- sis). Results of the DMRT analysis for the Superpave mixes are illustrated in Figure 38. 30 Statistically, results of the DMRT comparisons show that all methods produced statistically different air void contents. However, vacuum-sealing and gamma ray bulk specific grav- ity methods provided similar results given a difference of 0.24 percent air voids. On average, the dimensional method resulted in the highest air void contents, followed by the vacuum-sealing and gamma ray methods, respectively. Air void contents determined from AASHTO T166 resulted in the lowest air void contents. None of the alternative meth- ods provided similar results to AASHTO T166. The results for SMA mixtures are provided in Figure 39. As with the Superpave mixes, the vacuum-sealing and gamma ray methods resulted in similar air void contents. The dimen- sional method again resulted in the highest air voids and the AASHTO T166 method resulted in the lowest air voids. Analysis of both the Superpave and SMA data indicated that the four methods of measuring bulk specific gravity signif- icantly affected resulting air voids. For both mix types, the vacuum-sealing and gamma ray methods provided similar air voids; however, the dimensional method provided significantly higher air voids and AASHTO T166 provided significantly lower air void contents. Theoretically, the dimensional method should provide the highest measured air void content, as this method includes both the internal air voids and the surface texture of the sam- ple. Therefore, the results in Figures 38 and 39 pass the test of reasonableness for the vacuum-sealing, gamma ray, and AASHTO T166 methods as all three provided air void content lower than the dimensional method. Because it was assumed that the T-166 method would be accurate at low water absorption levels, it was decided to test the mixes with low absorption, less than 0.5 percent, to see which mixes provided results similar to the T-166 method. The results are provided in Figure 40. This figure shows that the vacuum-sealing and AASHTO T166 methods provided 9.39 7.50 7.26 6.22 0 1 2 3 4 5 6 7 8 9 10 11 Dimensional Vacuum-Sealing Gamma Ray AASHTO T166 Test Method A ve ra ge A ir Vo id C on te nt , % A B C D Letters represent results of Duncan's Multiple Range Test for air voids resulting from the bulk specific gravity methods. Methods with the same letter ranking are not significantly different. Figure 38. Average air voids and DMRT results for Superpave mixes.

31 7.24 7.09 4.97 10.11 0 2 4 6 8 10 12 Dimensional Vacuum-Sealing Gamma Ray AASHTO T166 Test Method A ve ra ge A ir Vo id C on te nt , % A B B C Letters represent results of Duncan's Multiple Range Test for air voids resulting from the bulk specific gravity methods. Methods with the same letter ranking are not significantly different. Figure 39. Average air voids and DMRT results for SMA mixes. 4.9 4.3 4.0 3.8 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 Dimensional Gamma Ray Vacuum-Sealing AASHTO T166 Test Method A ir Vo id C on te nt , % Letters represent results of Duncan's Multiple Range Test for air voids resulting from the bulk specific gravity methods. Methods with the same letter ranking are not significantly different.A B C C Figure 40. Comparison of test methods, mixes with low water absorption level. similar results and that both were significantly different than the dimensional and gamma ray methods. The dimensional method provided the highest air void content, as expected. The AASHTO T166 method is accurate for low water absorp- tion mixes and at these low void levels provide similar den- sity values to that of the vacuum seal method. These results suggest that the vacuum-sealing method provides an accurate density for low voids, which indicates that it also provides an accurate density at higher void levels because the plastic seal will clearly prevent water from being absorbed into the mix- ture. Figures 38 and 39 suggest that the gamma ray method does an overall adequate job of estimating bulk specific grav- ity; however, Figure 40 suggests that it is not as accurate as AASHTO T166 or the vacuum-sealing methods. Refinements to the gamma ray method may make this method a viable option in the future. 4.10.3 Analysis of Field Compacted Samples Included within this portion of the study were the cores obtained during the Task 5 field validation experiment. Only the vacuum-sealing and AASHTO T166 test methods were analyzed, as they were shown most accurate during the labo- ratory phase of this experiment. Figure 41 illustrates the rela- tionship between air voids determined from the two methods for all field cores obtained from the 20 field projects during Task 5. This figure illustrates that when air void content is less

than about 5 percent, the two methods provided approximately similar results. Above 5 percent air voids, the vacuum-sealing method resulted in higher air void contents. As air voids increased, the two methods diverged and it is believed that the reason for this divergence is the loss of water during the SSD method. Hence, at low air voids, both methods should be close to correct; however, at higher air voids the vacuum-sealing method should be more correct. 4.11 FIELD VALIDATION OF RELATIONSHIPS BETWEEN PERMEABILITY, LIFT THICKNESS, AND IN-PLACE DENSITY The main objective of the field portion of NCHRP 9-27 (Task 5) was to provide a field validation of the relationships between permeability, lift thickness, and in-place density so the overall objectives of the study could be accomplished. In order to field verify the relationships between air voids, lift thickness, and permeability, 20 HMA construction projects were visited. Testing at these projects included tests on plant- produced mix and on the compacted pavement. Testing of the plant produced mix included compacting samples to both the design compactive effort and to a specified height. Test- ing on the compacted pavement included performing field permeability tests with the NCAT Field Permeameter. Selec- tion of the 20 projects was based upon the following factors: NMAS, gradation type (fine-graded, coarse-graded, and SMA), and the lift thickness to NMAS ratio (t/NMAS). Table 15 presents the 20 projects evaluated. Table 15 shows that both fine- and coarse-graded Superpave designed mixes were investigated for each of four NMAS, ranging from 9.5 to 25.0 mm NMAS. SMA mixes were inves- 32 tigated for 12.5 and 19.0 mm NMASs. The effect of lift thick- ness was evaluated within the 9.5, 12.5, and 19.0 mm NMASs. To determine if a general trend occurred between in-place air voids and t/NMAS, a regression was performed on the com- bined data. Figure 42 illustrates this general relationship. From this regression, a low R2 of 0.09 was found. The trendline sug- gested that as the ratio of lift thickness to NMAS increased, in-place air voids decreased. To determine if the relationship between in-place air voids and the t/NMAS ratio was significant, an ANOVA was conducted on the regression. For the combined data, the p-value was 0.014, which indicated that the overall rela- tionship was significant. Then the data were separated into the three mix types. When an ANOVA was conducted on the regressions for the mix types, it was found that the relation- ship was not significant for any of the mix types (p-values of 0.956, 0.994, and 0.107 for fine-graded, coarse-graded, and SMA, respectively). There is a lot of scatter in the data, but, as can be seen in Figure 42, every increase of 1 in the t/NMAS results in a decrease in voids of approximately 0.6 per- cent. This finding involves average numbers, and it must be realized that many other factors affect the density of these field projects. Another factor to consider for these projects is the specifi- cation requirements were approximately the same for all of these mixes. Hence, the contractor was trying to compact all mixes to a low void content. Even with the same target density the t/NMAS affected the results. For Figure 43, a best-fit line was produced on the com- bined data for the 12.5-mm NMAS mixes. A low correlation was also found for this regression (0.19), but the general trend suggested that in-place air voids decreased as the lift y = 1.2486x0.842 R2 = 0.8676 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 24.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 20.0 22.0 24.0 26.0 Air Voids (Vacuum-Sealing), % A ir Vo id s (A AS HT O T1 66 ), % Figure 41. Comparisons between AASHTO T166 and vacuum-sealing methods, field projects.

33 Project ID NMAS Fine or Coarse Gradation Average Lift Thickness (mm) Actual Lift Thickness/ NMAS Ratio AC Performance Grade Ndesign 1 9.5 Fine 48.7 5.1:1 70-22 65 2 19.0 Coarse 65.7 3.5:1 64-22 65 3 9.5 Coarse 32.3 3.4:1 64-22 65 4 12.5 Fine 68.6 5.5:1 * 75 5 9.5 Fine 41.0 4.3:1 70-22 100 6 12.5 Coarse 50.3 4.0:1 58-28 75 7 9.5 Fine 40.6 4.3:1 64-28 75 8 19.0 Coarse 58.9 3.1:1 64-22 100 9 19.0 Coarse 96.4 5.1:1 64-22 100 10 19.0 Coarse 70.9 3.7:1 64-34 100 11 19.0 Coarse 38.0 2.0:1 64-34 125 12 25.0 SMA 42.6 1.7:1 76-22 50 13 25.0 Fine 70.0 2.8:1 67-22 100 14 9.5 SMA 26.8 2.8:1 76-22 75 15 19.0 Coarse 50.4 2.7:1 76-22 100 16 12.5 Coarse 43.8 3.5:1 67-22 86 17 12.5 Fine 43.3 3.5:1 64-22 75 18 12.5 Coarse 44.5 3.6:1 67-22 75 19 9.5 Fine 41.5 4.4:1 67-22 75 20 12.5 Fine 34.5 2.8:1 67-22 80 * Designated RA295 by the agency TABLE 15 Field project summary information y = -0.0017x + 7.9818 y = -0.0488x + 9.1967 y = -3.0894x + 16.837 y = -0.5809x + 10.792 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Thickness-to-NMAS Ratio In -p la ce A ir V oi ds , % All Data Fine Coarse SMA All Data Fine CoarseSMA All Data: R2 = 0.09, p-value = 0.014 Fine: R2 = 0.00, p-value = 0.956 Coarse: R2 = 0.00, p-value = 0.994 SMA: R2 = 0.17, p-value = 0.107 Figure 42. Relationship between t/NMAS and in-place air voids—9.5 mm, all data. thickness increased. An ANOVA conducted for the com- bined regression indicated that the relationship was signifi- cant (p-value = 0.001). The data were then separated into the different mix types to see if the relationship was significant for each mix type. For the fine-graded mixes, the relationship was significant (p-value = 0.000). The coarse-graded mixes did not have a significant relationship between in-place air voids and t/NMAS (p-value = 0.932). These data indicate that an increase of 1 for the t/NMAS resulted in an average decrease in air voids of 0.5 percent. Figure 44 shows the relationship between lift thickness and in-place air voids for the combined data set for the 19.0-mm NMAS mixes, as well as for the individual mix types. For the combined data, the regression produced a low R2 value (0.09). An ANOVA performed on the regression determined that the relationship between t/NMAS and in-place air voids for the 19.0-mm NMAS mixes was significant (p-value of 0.000). The data indicate that an increase of 1 for the t/NMAS results in an average decrease of 1.0 in the air voids. In summary, even though there is a large amount of scatter in the data for the three NMAS mixes, the results suggest that the air voids dropped 0.5 to 1.0 percent for each increase of 1 in the t/NMAS. This shows the importance of making sure that the t/NMAS is sufficiently high.

y = -0.6277x + 11.1 y = 0.0412x + 7.2391 y = -0.5205x + 10.079 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Thickness-to-NMAS Ratio In - pl ac e A ir V oi ds , % All Data Fine Coarse Fine Coarse All Data All Data: R2 = 0.19, p-value = 0.000 Fine: R2 = 0.41, p-value = 0.000 Coarse: R2 = 0.00, p-value = 0.932 Figure 43. Relationship between t/NMAS and in-place Air Voids—12.5 mm NMAS. y = -1.0455x + 11.09 R2 = 0.0913 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Thickness-to-NMAS-Ratio In - pl ac e A ir V oi ds , % All Data (All Coarse) p-value = 0.020 Figure 44. Relationship between t/NMAS and in-place air voids—19.0 mm NMAS. 34

Next: Chapter 5 - Conclusions and Recommendations »
Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Report 531: Relationship of Air Voids, Lift Thickness, and Permeability in Hot-Mix Asphalt Pavements provides guidance for hot-mix asphalt pavement construction designed to achieve satisfactory levels of in-place air voids and permeability. This guidance was developed from the findings of a research project that examined the relationship of air voids content to permeability and hot-mix asphalt lift thickness. The full finding of the research were published as NCHRP Web Document 68.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!