Nondestructive Testing to Identify Delaminations Between HMA Layers, Volume 3 - Controlled Evaluation Reports (2013)

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.

1Controlled Laboratory and Field Evaluations: Construction Report This chapter was prepared by Dr. Nam Tran and Dr. Michael Heitzman from the National Center for Asphalt Technology (NCAT). It summarizes the planning and construction of the test slabs for controlled laboratory evaluations and the con- struction of the sections at the NCAT Pavement Test Track. The chapter concludes with a list of lessons learned as the construction progressed. Task 6 was coordinated by Dr. Tran, and the data analysis was performed by the specialized research groups. To accomplish this task, three subtasks were conducted. Each subtask defined a specific set of activities, but the chronological sequence of the subtasks was divided by the desired temperature and mois- ture condition of the test pavement. This sequence of testing required each equipment vendor to bring their nondestructive testing (NDT) systems to the NCAT twice. For efficiency of the research and the NDT equipment vendors’ time, the mea- surements for Subtasks 6A and 6B under a prescribed climate condition were obtained at the same time. • Subtask 6A—Conduct Controlled Laboratory Evaluation. • Subtask 6B—Conduct Controlled Field Evaluation. • Subtask 6C—Refine the NDT Equipment or Software. To conduct the controlled laboratory and field evaluations under Subtasks 6A and 6B, two asphalt slabs and ten 25-ft asphalt pavement sections were constructed at the NCAT Pavement Test Track. The following sections describe how the test slabs and pavement sections were built. Construction of Test Slabs Figure 1.1 illustrates the design of the delamination con- ditions of the two test slabs. Two types of delamination— lack of bond and stripping—were simulated at two depths. Three interface treatments were used to achieve bonded and debonded conditions at the interfaces: (a) optimum amount of tack coat to the receiving surface for achieving full bond, (b) baghouse fines from the hot-mix asphalt (HMA) plant to the receiving surface to achieve no bond, and (c) placement of a separate 1-in.-thick, uncompacted coarse fractionated RAP (reclaimed asphalt pavement) to the receiving surface to represent a stripping condition. To facilitate construction and transportation, each of the two slabs was supported by four 4-ft by 8-ft laminated sheets of plywood. The plywood base was put on the ground, and hot asphalt tack was sprayed on the plywood surface (Fig- ure 1.2) to ensure a good bond between the asphalt slabs and the plywood base. Then two layers of HMA with a total thickness of 4 to 6 in. were paved over the plywood sheets and carefully compacted. Before the third asphalt layer was paved, the locations of two slabs were surveyed and marked on the pavement. Baghouse fines and RAP materials were placed on two 4-ft by 4-ft squares as shown in Figure 1.3 to simulate debonding and stripping interfaces at a 4-in. depth for Slab B. For Slab A, asphalt sur- face was tacked to make a good bond interface at a 4-in. depth. Then the third HMA layer that was approximately 2 in. thick was paved and carefully compacted. After the third HMA layer was completed, the locations of two slabs were again surveyed and marked on the pavement. To create a debonded interface at a 2-in. depth for Slab A, a 4-ft by 4-ft square covered with baghouse fines was placed as shown in Figure 1.4. Hot asphalt tack was sprayed around the square to make good bond interfaces at a 2-in. depth for Slab B and the other half of Slab A. Then the last (surface) HMA layer was paved to a thickness of approximately 2 in. and carefully compacted. After the construction was completed, two slabs were cut out of the pavement section (Figure 1.5). The locations of stripped and debonded interfaces were examined. Figure 1.6 shows Slab A, which has one half of the slab with all good bond interfaces, and the other half of the slab with a debonded interface at a depth of 2 in. Slab B, as shown in Figure 1.7, has an approxi- mately 1-in.-thick stripped layer at a depth of approximately C h a p T e r 1

2Full bond No bond at 2” depth 8” 4 ft 8 ft 2” HMA Stripping at 4” depth No bond at 4” depth 8” 4 ft 8 ft 4” 4” Slab A Slab B Figure 1.1. Design of two HMA slabs for controlled laboratory evaluation. Figure 1.2. Two plywood sheets for supporting test slabs. Figure 1.3. Two squares covered with baghouse fines and RAP materials. Figure 1.4. One square covered with baghouse fines. Figure 1.5. Slab A (foreground) and Slab B (background).

3 4 in. in one half and a debonded interface at a depth of 4 in. in the other half. Finally, the two slabs were boarded and trans- ported to the NCAT main laboratory for testing (Figure 1.8). Extreme care was used to lift and transport the slabs without creating tensile stress cracks. Construction of pavement Test Sections at the NCaT pavement Test Track Ten controlled asphalt pavement test sections were built in the inside lane at the NCAT Pavement Test Track for the controlled field evaluations under Subtask 6B. There were no bond and good bond (control) at the interfaces between dense-graded asphalt layers. The research team ensured the good bond by using a tack coat and bad bond by using bond Figure 1.6. Slab A: (a) fully bonded and (b) debonded at depth of 2 in. (a) (b) Figure 1.7. Slab B: (a) stripped and (b) debonded at depth of 4 in. (a) (b) Debond @ 4” Depth Debond @ 2” Depth Full Bond Stripping @ 4”Depth Figure 1.8. Two test slabs in the NCAT laboratory.

4As previously described, the 10 pavement test sections were built in the inside lane adjacent to Section 5 between Stations 0+15 and 2+65. The old pavement section built between those stations was constructed in 2000 with a 24-in.-thick HMA layer on top of a 6-in.-thick aggregate base. Because there were deep cracks between Stations 50 and 75, repairs were done before construction of the delamination test sections. The asphalt and a portion of the aggregate base layers were milled at the begin- ning of the old pavement section (Figure 1.20). For the second half of the experimental section, the old asphalt layer was milled approximately 6 in. thick to accommodate the delamination test sections. After the milling and backfilling work was completed, a 6-in.-thick concrete slab was constructed from Station 0+15 through Station 0+65, as shown in Figure 1.21. Layers of HMA were paved from Station 0+65 through Station 2+65 to Note: Light gray = baghouse dust; tan = paper; delamination depth = ~ 5 in. (between PCC and HMA); O = locations where point-load methods were conducted; X = verification core; and S = standpipe. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 S 3 4 O O O O X O 5 6 O O O O O 7 8 9 O O O O O 10 11 O O O O O 12 Figure 1.10. Section 1: HMA over PCC (Stations 015 to 040). Figure 1.9. Layout of controlled field test sections. PCC = Portland cement concrete. Section 1 Section 2 Section 3 Section 4 Section 5 Section 6 Section 7 Section 8 Section 9 Section 10 Top 2-inch lift Full bond Full bond Full bond Partial no bond No bond Partial stripping Full bond Full bond Full bond Full bond Bottom 3-inch lift No bond Full bond Full bond Full bond Full bond Full bond Full bond Partial stripping Partial no bond No bond Existing surface PCC PCC HMA HMA HMA HMA HMA HMA HMA HMA Section 1 – no bond between 5-inch HMA overlay and PCC pavement Section 2 – full bond between 5-inch HMA overlay and PCC pavement (control section) Section 3 – full bond between 5-inch HMA overlay and HMA pavement (control section 1 of 2) Section 4 – partial bond between 2-inch HMA overlay surface lift and 3-inch HMA overlay leveling lift Section 5 – no bond between 2-inch HMA overlay surface lift and 3-inch HMA overlay leveling lift Section 6 – simulated stripping in the wheel path between 2-inch HMA surface lift and 3-inch HMA leveling lift Section 7 – full bond between 5-inch HMA overlay and HMA pavement (control section 2 of 2) Section 8 – simulated stripping in the wheel path between 3-inch HMA overlay leveling lift and HMA pavement Section 9 – partial bond between 3-inch HMA overlay leveling lift and HMA pavement Section 10 – no bond between 3-inch HMA overlay leveling lift and HMA pavement (text continues on page 9) breakers, including baghouse fines and two layers of heavy kraft paper. A 1-in.-thick, uncompacted coarse-fractionated RAP material was used to simulate a stripping condition. The design for the controlled field test sections is illus- trated in Figure 1.9. The test sections were designed to simulate 10 different bonded and debonded conditions that represent a majority of situations encountered in the top 5 in. of HMA pavements. Both full lane width and partial lane debonding conditions were constructed for evaluating the NDT methods. The partial lane debonding condition included the wheelpath and two 3-ft by 3-ft squared areas. Each test section was 12 ft wide (full paving width) and 25 ft long. To achieve compaction, the full lane width debonded areas were only 10 ft wide. The outer 1 ft was fully bonded to confine the experimental debonded areas for compaction. The detailed designs for these test sections are presented in Figures 1.10 through 1.19.

5 Note: No delamination; O = locations where point-load methods were conducted; and X = verification core. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 3 4 O X O O O O 5 6 O O O O O 7 8 9 O O O O O 10 11 O O O O O 12 Figure 1.11. Section 2: HMA over PCC, control section (Stations 040 to 065). Note: No delamination; O = locations where point-load methods were conducted. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 3 O O O O O 4 5 O O O O O 6 7 8 O O O O O 9 10 O O O O O 11 12 Figure 1.12. Section 3: HMA pavement, control section (Stations 065 to 090). Note: Light gray = baghouse dust; delamination depth = ~ 2 in.; O = locations where point-load methods were conducted; and S = standpipe. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 S 3 O O O O O 4 5 O O O O O 6 7 8 O O O O O 9 10 O O O O O 11 12 Figure 1.13. Section 4: HMA pavement, wheelpath delamination (Stations 090 to 115).

6Note: Light gray = baghouse dust; delamination depth = ~ 2 in.; O = locations where point-load methods were conducted; X = verification core; and S = standpipe. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 S 3 O O O O O 4 X 5 O O O O O 6 7 8 O O O O O 9 10 O O O O O 11 12 Figure 1.14. Section 5: HMA pavement, full width delamination (Stations 115 to 140). Note: Dark gray = RAP; bottom of delamination = ~ 2 in.; RAP thickness = ~ 0.75 in.; O = locations where point-load methods were conducted; and S = standpipe. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 S 3 O O O O O 4 5 O O O O O 6 7 8 O O O O O 9 10 O O O O O 11 12 Figure 1.15. Section 6: HMA pavement, partial stripping (Stations 140 to 165). Note: No delamination; O = locations where point-load methods were conducted. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 3 O O O O O 4 5 O O O O O 6 7 8 O O O O O 9 10 O O O O O 11 12 Figure 1.16. Section 7: HMA pavement, control section (Stations 165 to 190).

7 Note: Dark gray = RAP; bottom of delamination = ~ 5 in.; RAP thickness = ~ 0.75 inches; O = locations where point-load methods were conducted; X = verification core; and S = standpipe. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 S 3 O O O O O 4 X 5 O O O O O 6 7 8 O O O O O 9 10 O O O O O 11 12 Figure 1.17. Section 8: HMA pavement, partial stripping (Stations 190 to 215). Note: Light gray = baghouse dust; tan = paper; delamination depth = ~ 5 in.; O = locations where point-load methods were conducted; X = verification core; and S = standpipe. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 3 O O O O O 4 X 5 O O O O O 6 7 8 O O O O O 9 S 10 O O O O O 11 12 Figure 1.18. Section 9: HMA pavement, wheelpath delamination (Stations 215 to 240). Note: Light gray = baghouse dust; tan = paper; delamination depth = ~ 5 in.; O = locations where point-load methods were conducted; and S = standpipe. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 3 O O O O O 4 5 O O O O O 6 7 S 8 O O O O O 9 10 O O O O O 11 12 Figure 1.19. Section 10: HMA pavement, full width delamination (Stations 240 to 265).

8621.0 621.5 622.0 622.5 623.0 623.5 624.0 624.5 625.0 0 25 50 75 100 125 150 175 200 225 250 El ev at io n (ft) Longitudinal Distance (ft) Backfilled with RAP Bottom of asphalt layer Bottom of Aggregate Base Layer Surface of old asphalt layer before milling Surface of RAP backfill Surface of old asphalt layer after milling Figure 1.20. Milling and backfilling profile (Stations 000 to 265). 621.0 621.5 622.0 622.5 623.0 623.5 624.0 624.5 625.0 0 25 50 75 100 125 150 175 200 225 250 El ev at io n (ft) Longitudinal Distance (ft) Backfilled with RAP 6" Concrete slab Leveling with HMA Bottom of asphalt layer Bottom of aggregate base layer Section 1 Section 2Section 3 Section 4 Section 5 Section 6 Section 7 Section 8 Section 9 Section 10 Bottom of delamination test sections (5" deep) Milled surface Interface between 2 layers of delamination test sections (2" deep) Surface of delamination test sections Surface of RAP backfill Figure 1.21. Pavement test section profile (Stations 000 to 265).

9 prepare the bottom surface for the delamination test sections. Figure 1.22 shows the concrete slab and the leveling layers of HMA being constructed between Stations 0+65 and 2+65. As previously illustrated in Figure 1.9, Sections 1, 8, 9, and 10 were designed to have delamination problems at 5 in. depth. As shown in Figure 1.23, two layers of brown paper were used as a bond breaker on the left, and baghouse fines material was used on the right in Section 1, as shown in Fig- ure 1.10. As previously shown in Figure 1.17, a partial strip- ping condition was designed at 5 in. depth. An approximately 1-in.-thick, uncompacted RAP layer was used to create the stripping condition in the field (Figure 1.24). Following the proposed designs for Sections 9 (Figure 1.18) and 10 (Figure 1.19), the research team used two layers of brown paper on the right and baghouse fines on the left as bond breakers in the field, as shown in Figure 1.25. No traffic was permitted on the newly paved experimental sections. A hot spray-applied asphalt tack was placed between lifts, and no tack was placed on delaminated areas. Despite the presence of the delaminated conditions, the placement of the 3-in. bottom lift for the delamination test sections went smoothly, except for two problems: the paver tires picked up the paper, and the screed ripped the top layer of paper at one location in Section 9 (Figure 1.26). The HMA layer was repaired (filled), but the double paper condition was lost. Figure 1.22. Concrete slab (Stations 015 to 065) and leveling HMA (Stations 065 to 265). Figure 1.23. Brown paper (left) and baghouse fines (right) used as bond breakers in Section 1 at depth of 5 in. Figure 1.24. Uncompacted RAP material approximately 1 in. thick placed in Section 8 at depth of 5 in. Section 8 Section 9 Section 10 Figure 1.25. Brown paper and baghouse fines used at depth of 5 in. in Sections 9 and 10, near Section 8 (Figure 1.24 above). (continued from page 4)

10 After the bottom lift was placed and cooled down, the research team placed the delamination conditions at a depth of 2 in. Because of the problems with the paper as previously discussed, the paper was not used as a bond breaker at the 2-in. depth. Figure 1.27 shows an approximately 1-in.-thick, uncompacted RAP layer that was placed in the three locations in Section 6, as previously detailed in Figure 1.15. Baghouse fines material was used in Sections 4 and 5 as a bond breaker (Figure 1.28), and the dimensions of the delaminated areas in Sections 4 and 5 are shown in Figure 1.13 and Figure 1.14, respectively. Figure 1.29 shows the paver placing HMA on the delami- nated area of baghouse fines at a depth of 2 in. in Section 5. The placement of the 2-in.-thick HMA surface layer in Sections 4 and 5 went smoothly. As shown in Figure 1.30, the paver was moving to the delaminated areas and using uncompacted RAP materials in Section 6. The placement of the surface layer went Figure 1.26. Brown paper torn by paver at one location in Section 9. Section 4 Section 5 Section 6 Figure 1.27. Approximately 1-in.-thick, uncompacted RAP materials placed in Section 6 at depth of 2 in. Section 6 Section 5 Section 4 Figure 1.28. Baghouse fines used as bond breaker in Sections 4 and 5 at depth of 2 in. Figure 1.29. Paver placing 2-in. surface lift over baghouse fines in delaminated area in Section 5. well at the beginning of Section 6. However, the uncompacted RAP layer may have been too thick for the 2-in.-thick surface layer, because the screed behind the paver was pushing the RAP materials to the surface toward the end of Section 6, as shown in Figure 1.31. The problem was immediately repaired in the field. After the first round of laboratory and field evalua- tions under Task 6 in late October and November 2009, the research team extracted five cores to verify the interface conditions of the field test sections. Figure 1.32 shows two portions of Core 1 broke at a depth of approximately 5 in. from the pavement surface during coring. This confirmed that the baghouse fines placed on top of the concrete slab (at a depth of approximately 5 in. from the surface) caused the debonding problem at the interface, as was anticipated. Figure 1.33 shows Core 2 extracted from Section 2 (one

11 of the control sections), and this core showed no signs of delamination, as expected. Figure 1.34 shows Core 3 extracted from Section 5. It was anticipated that the interface at approximately 2 in. depth from the pavement surface would break during coring; however, this interface was intact even though a thin layer of baghouse fines could be seen at the interface. More cores will be extracted from this test section to evaluate the delamination condition further. Core 4 shown in Figure 1.35 was extracted from Section 8. The interface at a depth of approximately 5 in. broke during coring. This interface was delaminated with an approximately 1-in.-thick, uncompacted RAP layer. The last core (Figure 1.36) was cut from Section 9. The core broke at the 5-in. interface where two layers of brown paper were used as a bond breaker. More cores will be extracted after all the field evaluations under Task 6 are completed. Figure 1.30. Paver moving into delaminated areas and using RAP materials in Section 6. Figure 1.31. Paver pushing RAP materials to the surface in Section 6. Figure 1.32. Core 1 from Station 035 in Section 1. Figure 1.33. Core 2 from Station 045 in Section 2. Figure 1.34. Core 3 from Station 135 in Section 5.

12 Lessons Learned The following lessons were drawn from the experience with the construction of the test slabs for controlled laboratory evaluations at the NCAT Pavement Test Track: • Kraft paper was not strong enough to resist tensile forces generated by HMA paving screed as the screed passed over the paper. The top layer of paper tore and slid with the screed. A heavier or stronger type of paper should be used for the upper layer. • Delamination sections were built as overlay on a thick HMA pavement that resisted test load deflection. • After RAP material is placed, the material should be allowed to soften from solar heating and then compacted with one pass of a rubber tire. This process tightens the material in place and reduces the potential for the paver screed to move the material ahead. • Paper was picked up by paver tires. To avoid that problem, the paper should be covered with loose mix ahead of the paver. references Van Dam, T., K. Kirchner, M. Shahin, and E. Blackmon. 1987. Conse- quence of Layer Separation on Pavement Performance. Report DOT/ FAA/PM-86/48. U.S. Department of Transportation, Federal Aviation Administration. Ziari, H., and M. Khabiri. 2007. Interface Condition Interface on Pre- diction of Flexible Pavement Life. Journal of Civil Engineering and Management, Vol. 13, No. 1, pp. 71–76. Figure 1.35. Core 4 from Station 210 in Section 8. Figure 1.36. Core 5 from Station 220 in Section 9.