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.
5This chapter describes the results of Phase 2 subcontract research conducted by Olson Engineering, Inc., at the National Center for Asphalt Technology (NCAT), Auburn University, in Alabama, and at test sites in Gainesville, Florida, and Pitts- burg, Kansas. The research examined nondestructive testing (NDT) and evaluation of the pavement at these sites with stress wave methods for debonded hot-mix asphalt (HMA) layers. The field portion of this investigation was performed in accordance with generally accepted testing procedures. Research team members were Patrick K. Miller, Yajai Tinkey, and Larry D. Olson. Introduction Phase 2 of the SHRP 2 R06D research project for stress wave detection of delaminated (debonded) asphalt pavement lay- ers focused on the use of spectral analysis of surface waves and impact echo (SASW and IE) tests to determine the con- dition of asphalt pavements. Olson Engineering has added four transducer wheels to the Bridge Deck Scanner (BDS) to accelerate the testing. The new system with six transducer wheels was used again on the NCAT Pavement Test Track site to verify the accuracy of the test results from the new sys- tem. Then the BDS with six transducer wheels was used to determine the conditions of asphalt pavements in Florida and Kansas. IE and SASW Test Setup The same test setup was used in all three test sites. The two transducer wheels within the same pair were spaced 6 in. Only one solenoid was used to impact the asphalt so that the transducer wheel nearest to the solenoid impactor acquired the IE data and both transducer wheels acquired the data for SASW tests. Distances between each sensor wheel pair were varied between 2 and 3 ft. IE and SASW Results from the NCAT Test Track Test results from the NCAT test track with known, con- structed delaminations are presented in the section on NCAT pavement. Summaries of findings from the NCAT test results are given below: 1. Results from the IE tests were able to identify debond- ing conditions with a depth of 5 in. However, the IE tests were not able to detect shallow delaminations (2-in.-deep delaminations). This result is most likely caused by the inability of the current system to excite the needed high frequencies and the fundamental difficulty to excite high frequencies in asphalt at above freezing temperatures because of asphaltâs temperature-dependent elastic moduli. This outcome is in contrast to the BDS perfor- mance for IE tests on concrete pavements and decks that have higher elastic stiffness/moduli and are not temperature-dependent to any significant degree. 2. The SASW tests were successfully used to identify debond- ing conditions on the NCAT asphalt test track pavement. A sharp drop in the surface wave velocities appeared to correlate well with the depth of debonding, as shown in the section on test results from the SASW tests. Both shallow debonding (2 in.) and deeper debonding (5 in.) were detected by using the SASW tests, as sufficiently high frequencies were excited for SASW tests at the nominally 50°F and higher test temperatures. SASW Test Results from the Florida Pavement On the basis of test results from the NCAT test track, only SASW test results were processed from the Florida pave- ment. In general, the conditions of the 2,000-ft section of the tested Florida pavement were similar to those identified C h a p t e r 2 Olson Engineering, Inc., Results
6with SASW tests of known delaminations at the NCAT track. The majority of test locations showed a sharp drop of sur- face wave velocity at wavelengths corresponding to depths between 0.2 and 0.4 ft. This drop was indicative of either possible debonding at these depths or an existence of a thin layer of lower stiffness/moduli material at these depths. In addition, the conditions of the pavement at the centerline of the lane seemed to be in better condition than did those of the wheelpaths. Finally, the general conditions of the pave- ment on the right wheelpath (0 to -5 ft) appeared to be in the worst condition, with slower velocity in comparison to the left wheelpath (0 to +5 ft). SASW Test Results from the Kansas Pavement On the basis of test results from the NCAT test track, only SASW test results were processed from the Kansas pavement, as well. Overall, the majority of the data from the Kansas test site had relatively high surface wave velocities near the pave- ment surface (<0.3 ft deep) and decreasing velocities with depth indicating weaker underlying materials. The teamâs interpretation of the data overall is that the road likely had a relatively new asphalt overlay of good condition with an older underlying pavement composed of weaker material with widely varying velocity/moduli conditions. Research by oth- ers on asphalt pavements has shown that decreases in surface wave velocity correlate well with aging/reduction of moduli in asphalt as pavements deteriorate with traffic loading and time. Summary of Stress Wave Findings from Scanning IE and SASW Testing IE scanning with the BDS system was able to detect 5-in.-deep delaminations, but not the 2-in.-deep delaminations at the NCAT test track. SASW testing was able to detect delamina- tions at depths of 2 in. and deeper. It was further found that there could be additional benefits from SASW scanning of asphalt in terms not only of debonding but also by providing data on the moduli of the pavement materials where it was not delaminated, as well. To the best of the teamâs knowledge, the application of slow rolling at 1 to 2 mph, contacting IE and SASW scanning for asphalt pavements, was done for the first time in this study. Development of Bridge Deck Scanner hardware The initial IE and SASW testing performed as Phase 1 of the SHRP 2 research project at the NCAT Pavement Test Track in Opelika, Alabama, in October 2009 used a two-wheel rolling measurement system. The two-wheel pair system was developed under a National Cooperative Highway Research Program grant (NCHRP IDEAS Project No. 32âVehicle Mounted Bridge Deck Scanner [BDS] by Tinkey, Olson, and Miller) awarded to the research team members. The NCHRP research involved performing scanning IE tests, surface waves tests, and other tests for mapping out reinforcing corrosion- induced bridge deck delaminations and general condition assessment evaluation of concrete bridge decks. The BDS sys- tem had already evolved through multiple prototype versions into a fairly mature two-wheeled device. The basic BDS two-wheel system essentially consists of two transducer wheels made of high-density polyethylene (HDP) with six embedded Olson Instruments IE sensor heads around the circumference of the wheel at 6-in. intervals to allow for tests every 6 in. The sensor heads were paired with a small steel solenoid impactor. The wheels were coupled together with a rubber isolated axle that provided alignment of both displace- ment transducers for SASW testing or misalignment (offset 30°) of the two wheels in the case of IE testing to make impact at both wheels for IE tests only. The system had onboard sig- nal conditioning and was controlled by an Olson Instruments Freedom Data PC. The Freedom Data PC performed all trig- ger and timing functions as well as the data acquisition and the data analysis and presentation. Figures 2.1 and 2.2 show photographs of the two-wheel pair IE/SASW scanner in use during Phase 1 testing at the test track. After two separate rounds of testing at the NCAT test track, extensive analysis and comparison of the collected data to the known in-situ condition, and constructive feedback from the investigative team, a series of modifications of the IE/SASW scanning system were implemented before Phase 2 testing. The first testing round was in October 2009 on âwarmâ asphalt and the second testing round was in March 2010 on âcoldâ asphalt. The objective of these modifications was to improve Synchronized transducer wheels Automated Impactors Cable to data acquisition PC Vehicle ball hitch mount Figure 2.1. Photograph of the IE/SASW system used during Phase 1 testing.
7 the ability of the system to detect near-surface delaminations (~2 in. deep) as well as to increase the number of test wheels to improve the test area coverage of a single pass and thus to reduce the required test time of a single lane. In the initial Phase 1 study, a variety of NDT test methods was evaluated, and it was found that the IE test method pro- vided the most consistent results in determining debonded asphalt conditions. However, the IE test method was not able to detect shallow debonding at a depth of approximately 2 in. It was suggested by the research team that the SASW test method may provide a better indication of debonding condi- tions, particularly shallow debonding. It was also discussed that the transducer spacing of 1 ft, while excellent for IE test- ing, was not ideal for SASW testing. The possibility of the solenoid impactor being too close to the first transducer and creating so-called near-field wave effects was also discussed with the research team. Therefore, a series of experiments was conducted in the teamâs laboratory on both concrete and asphalt specimens, in which the distance between the two transducers was varied, as well as the distance between the impactor and the first transducer. The results of these experi- ments found that a 6-in. spacing between transducers would most likely be optimal at locating debonding conditions at depths ranging from ~1 to 10 in. This depth fully covers typi- cal asphalt pavement thicknesses. It was concluded that the original impactor to first transducer distance of 1.75 in. was not creating near field effects at a 6-in. transducer spacing but most likely was complicating the analysis at the previous transducer spacing of 12 in. Therefore, it was determined, on the basis of these experiments, to retain the impactor/first transducer spacing of 1.75 in. and to reduce the transducer/ transducer spacing to 6 in. This test configuration was used on the NCAT test track in February 2011 to verify the ability of the scanning SASW system to locate shallow debonding condition, and excellent results were achieved, as discussed in the section on NCAT pavement. On a related note regarding SASW testing, the alignment of a single pair of transducer wheels to allow for SASW test- ing is imperative. However, based on testing and experiments it was necessary to isolate the wheels from one another using rubber couplings to avoid excessive vibrations noise. Experi- ments were done with multiple different axle configurations incorporating rubber isolation, but there was trouble with wheel slippage at times, as had been the case at the Florida test site. The recent axle configuration used rubber isolated hubs attached to the wheel with steel shear pins and an aluminum axle, also with shear pins (see Figure 2.3). This configuration was used at the Kansas test site and experienced no slippage. Figure 2.2. Scanning IE/SASW testing performed during Phase 1 on the NCAT test track. Aluminum axle locked with metal pins Axle hub ring, rubber mounted with 6 steel shear studs Figure 2.3. Current IE/SASW scanning system, showing new axle configuration and 6-in. transducer spacing shown inline for SASW testing (offset for alternating IE testing).
8The major upgrade implemented to the IE/SASW scan- ning system was the expansion from a single pair of two wheels to three separate wheel pairs for a total of six wheels. The decision to expand to six measurement wheels was made when the testing plan was to perform IE testing on 2-ft spac- ings to cover a full 12-ft lane width in a single pass, leaving 1 ft of space on either side. However, as described above, through additional analysis, recommendations by the research team, and additional laboratory testing, the optimal test configura- tion for asphalt pavement delamination detection was deter- mined to be IE/SASW with a 6-in. displacement transducer spacing, as shown in Figure 2.4. With a spacing of 2 ft between each pair of wheels, it would require five or six pairs of wheels to provide full coverage of a single 12-ft lane. Therefore, when testing was conducted at the Florida and Kansas test sites, two passes were performed to cover the full lane width. Additional expansion of the system could be completed in the future to eliminate the need for two passes. The additional four measurement wheels (two pairs) were constructed to be nearly identical to the original pair of trans- ducer wheels. This construction included six IE/SASW trans- ducers per wheel evenly spaced around the circumference at 6-in. intervals, six impactor solenoids, and electronic circuitry for signal conditioning, solenoid firing, and data acquisition triggering. As the four new wheels were built, a few slight modi- fications were made to improve the design: a lip of the HDP wheel was added to secure the urethane tire (which previously had slowly rolled off the transducer wheel on long scans) and a variable attenuator circuit was added to the signal conditioning to allow the signal to be reduced in voltage amplitude to avoid clipping during data acquisition. Another major modification was the addition of trigger timing circuitry to synchronize the three independent transducer wheel pairs. This timing circuitry eliminated the possibility of multiple wheel pairs (or individual transducer wheels in the case of IE testing) firing at the same time, which would cause confusion in the propagated waves and data collection. The timing circuitry also forced the wheel pairs to fire sequentially (from left to right) to allow the data to be well organized in the software. The timing circuitry had two separate options, which allowed the scanning system to be run as three sets of SASW testing wheel pairs or as six individual IE testing wheels. The last major change necessary was the addi- tion of an independent distance wheel to record the distance from the start of the scan to each data collection point with any of the transducer wheels or pairs. The distance wheel output 128 pulses per revolution, resulting in a distance measurement resolution of approximately 0.1 in. Several other miscellaneous upgrades were also completed to improve the ease of use, ease of setup, and simplicity. For instance, the assembly of the wheel-pair carriage was simpli- fied and redesigned to make it easier and quicker to assemble. The towing assembly mounted to the truck was also slightly altered, adding more pin (rather than bolted) connections to ease field assembly (see Figure 2.5). The attachment of the wheel pairs to the towing assembly was changed to a clamp- ing attachment that could slide on the towing assembly bar, creating endless configuration possibilities. Lift bails were added to the towing assembly to allow the transducer wheels to be picked off the asphalt surface to back the system up or reposition the truck with greater ease (see Figure 2.6). The Source: Photograph from the Pittsburg, Kansas, test site. Olson Instruments Freedom Data PC Acquisition Computer Solenoid Impactor and Displacement Transducer Wheel Pairs Distance wheel Figure 2.4. Current IE/SASW scanning system, showing the SASW test setup of 6-in. transducer spacings with distance of 2 ft between pairs.
9 wiring between the transducer wheels and the data acquisi- tion computer was also simplified. The system is still run with the Olson Instruments Free- dom Data PC, which runs the acquisition software and pro- vides all data acquisition. The solenoids are powered either with a 12-VDC battery or with a cigarette lighter car plug. Ie and SaSW Method and Data acquisition with Bridge Deck Scanner The six-wheel BDS system was designed to perform either 1. IE testing on all 6 transducer wheels; or 2. IE testing on the transducer wheel near the impactor sole- noid and SASW testing from adjacent pairs of transducer wheels. The first type of testing (IE) is best applied to condition assessment of concrete bridge or garage deck slabs without an asphalt overlay. The later type of testing (IE and SASW) works well with asphalt material, either full depth or as an overlay. This project used the second test setup to perform primar- ily SASW testing; thus, only SASW will be discussed. The IE testing was also used at the NCAT test track and is discussed more in the section on NCAT pavement. To perform SASW testing, sensor elements in both transducer wheels (within a pair) were aligned and locked with a pin to prevent the slip- page. The sensor elements were offset approximately 2 in. between each adjacent pair of transducer wheels. Figure 2.7 shows a typical test setup, including the transducer locations of all wheels for the SASW setup. In this case, the system was set so that only the solenoids of the left transducer wheel (of the pair) are used for generating impacts. The solenoids of the right transducer wheel (of the Redesigned assembly connections Redesigned circuit box Figure 2.5. Current IE/SASW scanning system, showing simplified assembly connections and a smaller circuitry box. Wheel pair lift bail Sliding attachment to assembly bar Figure 2.6. Current IE/SASW scanning system, showing the sliding connection system and lift bails.
10 pair) are disabled for the duration of the testing. The impact sequence starts with firing the single solenoid used for the left pair of transducer wheels, followed by the solenoid for the mid- dle pair of transducer wheels, and finally, the solenoid for the right pair of transducer wheels. The sequence is then repeated. Once one of the solenoids fires, data acquisition for all three pairs of transducer wheels (all six channels) is recorded simulta- neously, but only the data from the pair of wheels where the sole- noid is fired and impacts are analyzed. At each acquisition, two channels (one wheel pair) will actually have valid surface wave data. The other four channels (two pairs) acquire background noise. The postanalysis software scans through all the acquired data and pulls out only valid SASW data for further analysis. Features in the postanalysis software include auto- windowing, dynamic masking/dispersion curve display, and composite velocity calculations. Plotting of the test results is performed in Golden Software Surfer software. Short discus- sions of the IE and SASW methods are presented below to explain and illustrate the methods that are discussed in other references, including Nondestructive Test Methods for Evalu- ation of Concrete in Structures by American Concrete Insti- tutes Committee 228, Report ACI 228.2R-98. IE Method The IE method involves hitting the concrete surface with a small impactor (or impulse hammer) and identifying the reflected wave energy with a displacement (or accelerometer) receiver mounted on the surface near the impact point [ASTM Standard C-1383(07)]. After the impact, the resulting displacement or acceleration response of the receiver is recorded. Although the resonant echoes are usually not apparent in the time domain, they are more easily identified in the frequency domain (linear displacement spectrum). Consequently, the time domain test data are processed with a Fast Fourier Transform (FFT), which allows identification of frequency peaks (echoes). If the thick- ness of a slab is known, the compression wave velocity (Vp) can be determined by Equation 2.1: Vp d f= 2 2 1 β ( . ) where d = slab thickness and f = resonant frequency peak. Equation 2.1 is modified by a beta factor of ~0.96 for con- crete slabs and algebraically rewritten to calculate the echo depth thickness, d. An example result for one IE scan line from a new bridge with cracking/void concerns that was tested with the BDS system is shown in Figure 2.8. Thickness echo data shown on the left include an approach slab from 0 to 20 ft followed by 80 ft of the concrete bridge deck. The time domain signal is in the upper right, and its correspond- ing IE thickness echo of 8.05 in. in the linear displacement frequency spectra is in the lower right in Figure 2.8. Shallow delaminations of less than a few inches may not produce a distinct echo peak response in an IE test if the necessary high frequencies cannot be generated, as often occurs for warm to hot asphalt. However, such a delamina- tion has the potential to be identified by a low-frequency flexural response similar to that occurring in BDS-IE test- ing of concrete bridge decks. Shallow deck delaminations less than a depth of 3 to 4 in. are identified by an apparently much thicker echo depth due to the resonant flexing associ- ated with a concrete delamination (due to reinforcing cor- rosion and expansion) parallel to the deck surface. However, unless the asphalt is tested at freezing and lower temperatures and, thus, the asphalt is almost concrete-like in terms of its modulus, then this flexural response may not be measured. In this study, all asphalt temperatures were above freezing at the NCAT test track, and there was no indication of the hoped- for low-frequency flexural response that might have indi- cated the presence of the shallow, 2-in.-deep delaminations. In Phase 1, some apparent, possible flexural responses indica- tive of shallow asphalt delamination were measured. These results likely indicate either one of two conditions. The first condition is that the material could be lacking consolidation or could be of low density, which causes a decrease in the res- onant frequency of the slab, or could be lacking consolidation and low density. The second condition is that the area could have a nearâsurface, shallow delamination that has a strong low-frequency flexural resonance and the full-depth asphalt pavement slab resonance is not observed. If a delamination is present, the depth of the delamination cannot usually be determined through IE testing because the flexural resonance of the delamination would depend strongly on the delami- nation depth and area. In concrete, with special solenoid IE testing equipment, team members have been able to excite frequencies as high as 50 kHz and measure concrete thick- nesses as thin as 1.5 in. (0.125 ft), with an Olson Instruments Impact Echo-1 Super Thin test head. However, because of the viscous, temperature-dependent stiffness/moduli of asphalt, it is difficult to excite such high frequencies in asphalt, partic- ularly as temperatures increase. At low temperatures of freez- ing and below, asphalt is more concrete-like, and it is easier to excite high frequencies and measure echoes from thin asphalt pavements or shallow delamination depths. Because this is Figure 2.7. Typical SASW testing setup.
11 generally not the case, the flexural response of a delamination that corresponds to an apparently greater thickness of asphalt was generally found to be the best indication of delamina- tions less than a depth of 5 in. in Phase 1 studies, but this condition was not consistently apparent in the IE results. SASW Method The SASW method uses the dispersive characteristics of sur- face waves to determine the variation of the surface wave velocity (stiffness) of layered systems with depth. Shear wave velocity profiles can be determined from the experimental dispersion curves (surface wave velocity versus wavelength) obtained from SASW measurements through a process called âforward modeling.â Forward modeling is an iterative inver- sion process to match experimental and theoretical results. Materials that can be tested with the SASW method include concrete, asphalt, soil, rock, masonry, and wood. Applications of the SASW method include but are not limited to (1) deter- mination of pavement system profiles, including the surface layer, base, and subgrade materials; (2) determination of seis- mic velocity profiles needed for dynamic loading analysis; (3) determination of abutment depths of bridge substructure; and (4) condition assessment of structural concrete. SASW can also measure crack depths (for cracks perpendicular to the surface) in bridge decks. The SASW method uses the dis- persive characteristics of surface waves to evaluate concrete integrity with increasing wavelength (depth). Open, unfilled cracks will result in slower surface wave velocities. Weak, fire- damaged, and poor-quality concrete (less stiff/lower elastic moduli) also produce slower surface wave velocities. In the NCHRP IDEAS BDS research, SASW tests were conducted with two of the same microphones that were used as noncontacting transducers. The distance between the two microphones was 4 in. (10.1 cm), and a solenoid impactor was used as an impact source. The microphones were mounted 3 in. (7.6 cm) above the concrete slab, and the source was located between 8 and 18 in. (20.3 and 45.7 cm) away from the closest microphone. Good-quality SASW data (top trace in Figure 2.9) with a good coherence up to 10,000 Hz (second trace from top in Figure 2.9) was captured by both microphones. The SASW data shown in Figure 2.9 were exponentially windowed after the large amplitude sur- face wave pulse traveled by the data to eliminate other wave vibration modes. The surface wave velocity can be calculated from the phase plot (third trace from top in Figure 2.9) as a function of wavelength (velocity = frequency à wavelength). The surface wave velocity is calculated to be approximately Figure 2.8. Example IE scanning results from the BDS-IE system on a new bridge.
12 Figure 2.9. Example SASW data processing from noncontact microphones from NCHRP IDEAS BDS research.
13 7,000 ft/s uniformly from wavelengths of 0.2 to 0.4 ft (6.1 to 12.1 cm), as seen in the last trace in Figure 2.9, which agreed well with the results from contacting displacement trans- ducers. Coherence near 1.0 is high-quality data from two or more impacts. Thus, for the SASW tests conducted with the BDS system on asphalt pavement, the calculation of coher- ence was not possible since there was only one impact per test location during rolling. At this time, microphones for SASW testing do not pro- duce SASW results of as high quality as do the rolling, dis- placement transducers. Consequently, only the rolling sensor wheels were used in the R06D field testing program, which is reported. Data Interpretation of SaSW test results Sound pavement yields a high and constant surface wave veloc- ity shown as a flat and horizontal line in a dispersion curve throughout the depth of the pavement. A typical example is shown in Figure 2.10. It was found that the presence of a sharp drop in the surface wave velocity within the dispersion curve acted as a reliable indication of potential debonding within asphalt pavements. The location (wavelength) of the velocity drop related closely to the depth of the debonding. Figure 2.11 presents a dispersion curve (surface wave velocity versus wave- length) from a location from Section 1 from the NCAT site. Su rfa ce w av e ve lo cit y (ft/ se c) Figure 2.10. A dispersion curve from a sound location on NCAT pavement. Su rfa ce w av e ve lo cit y (ft/ s) Figure 2.11. A dispersion curve from a location with debonding at a depth of 5 in.
14 The actual as-built plan showed baghouse dust at a depth of 5 in. to simulate debonding within the HMA. Review of Fig- ure 2.12 shows a sharp drop in the surface wave velocity at a depth of 0.41 ft, or 4.9 in., which is indicative of the delami- nation depth. Figure 2.12 presents a dispersion curve from a location in Section 6 (HMA pavement with partial stripping) where the bottom of delamination is approximately 2 in. deep. Reviews of Figures 2.13 and 2.14 show 5-in.-deep delamina- tions for baghouse dust and thin paper, respectively. NCat pavement test track SaSW and Ie results Test Setup The BDS system with six wheels was connected onto a hitch of a vehicle, as shown in Figure 2.15. The IE and SASW scanning tests were performed in two runs. For the first run, the center- lines of the left, middle, and right transducer wheels were located at Grid Lines 4.25 ft, 6.25 ft, and 10.25 ft, respectively. For the second run, the centerlines of the left, middle, and right transducer wheels were located at Grid Lines 5.25 ft, 7.25 ft, and 9.25 ft, respectively. Figure 2.16 shows the test locations from both runs. Test Results from the SASW Scans This section presents test results from the SASW scanning of the NCAT pavement. The test results are presented in surface wave velocity plots at different depths from 0.1 ft to 0.7 ft from the surface. Surface waves velocities are presented in a gray- scale ranging from 3,000 (shown in light gray) to 5,000 ft/s (shown in black). The higher the surface wave velocity, the better the condition of the asphalt pavement. Anomalies can Su rfa ce w av e ve lo cit y (ft/ s) Figure 2.12. A dispersion curve from a location with debonding at a depth of 2 in. Note: Surface wave velocity decreases from ~5,200 ft/s (vertical scale) to ~4,300 ft/s at a wavelength of 0.4 ft (~5 in.) and indicates the presence of the delamination. Figure 2.13. SASW dispersion curve example of baghouse dust delamination built at a depth of 5 in.
15 be seen as a light spot (zone) as the velocities are lower. Fig- ures 2.17 to 2.26 present the surface wave velocity profiles from all 10 sections of the NCAT Pavement Test Track. Reviews of Figure 2.17 (Section 1) show a drop of surface wave velocity between depths of 0.4 and 0.5 ft. Thus, the results show a likelihood of debonding between depths of 0.4 and 0.5 ft (4.8 and 6.0 in.). This interpretation of the SASW test results agrees well with the as-built condition. Reviews of Figures 2.18 and 2.19 (Sections 2 and 3) show a sound pavement with high velocities consistently from top to bottom of the pavement. This interpretation of the SASW test results agrees well with the as-built condition. Reviews of Figure 2.20 (Section 4) show lower velocities starting at depths of 0.3 to 0.4 ft from Grid Line 4 ft to 7 ft. This interpretation does not correlate well with the as-built condition. Reviews of Figure 2.21 (Section 5) show that data from approximately 25% of the test locations have a drop in velocity at depths of 0.2 to 0.3 ft. The majority of the test location shows a drop in velocity at depths of 0.3 to 0.4 ft. Reviews of Figure 2.22 (Section 6) show low velocities from Grid Lines 9 and 10 at the shallow depth (0.1 to 0.2 ft). This low velocity indicates debonding at depths between 0.1 and 0.2 ft. Reviews of Figure 2.23 (Section 7) show a sound pavement except at the lower-left corner (a few test locations), where the test results show a shallow delamination between 0.1 and 0.2 ft. Reviews of Figure 2.24 (Section 8) show delamination below Grid Line 4 ft at depths between 0.4 and 0.5 ft (approx- imately 30% of the data points) and between depths of 0.5 and 0.6 ft (approximately 60% of the data points). Reviews of Figure 2.25 (Section 9) show a sound pavement that does not agree with the as-built condition. Note: Surface wave velocity decreases from ~5,300 ft/s (vertical scale) to ~4,300 ft/s at a wavelength of 0.43 ft (~5 in.) and indicates the presence of the delamination. Figure 2.14. SASW dispersion curve example of thin paper delamination built at a depth of 5 in. Figure 2.15. Test setup for IE and SASW at NCAT test track. Centerline of the Transducer Figure 2.16. Test locations on any 25-ft sections of NCAT pavement test track. (text continues on page 26)
16 Note: Section 1 = 0 to 25 ft. Surface Wave Velocity (ft/sec) Depth = 0.6 â 0.7 ft Depth = 0.5 â 0.6 ft Depth = 0.4 â 0.5 ft Depth = 0.3 â 0.4 ft Depth = 0.2 â 0.3 ft Depth = 0.1 â 0.2 ft Figure 2.17. Profile plots of surface wave velocity.
17 Depth = 0.1 â 0.2 ft Depth = 0.2 â 0.3 ft Depth = 0.3 â 0.4 ft Depth = 0.4 â 0.5 ft Depth = 0.5 â 0.6 ft Depth = 0.6 â 0.7 ft Note: Section 2 = 25 to 50 ft. Surface Wave Velocity (ft/sec) Figure 2.18. Profile plots of surface wave velocity.
18 Depth = 0.1 â 0.2 ft Depth = 0.2 â 0.3 ft Depth = 0.3 â 0.4 ft Depth = 0.4 â 0.5 ft Depth = 0.5 â 0.6 ft Depth = 0.6 â 0.7 ft Note: Section 3 = 50 to 75 ft. Surface Wave Velocity (ft/sec) Figure 2.19. Profile plots of surface wave velocity.
19 Depth = 0.1 â 0.2 ft Depth = 0.2 â 0.3 ft Depth = 0.3 â 0.4 ft Depth = 0.4 â 0.5 ft Depth = 0.5 â 0.6 ft Depth = 0.6 â 0.7 ft Note: Section 4 = 75 to 100 ft. Surface Wave Velocity (ft/sec) Figure 2.20. Profile plots of surface wave velocity.
20 Depth = 0.1 â 0.2 ft Depth = 0.2 â 0.3 ft Depth = 0.3 â 0.4 ft Depth = 0.4 â 0.5 ft Depth = 0.5 â 0.6 ft Depth = 0.6 â 0.7 ft Note: Section 5 = 100 to 125 ft. Surface Wave Velocity (ft/sec) Figure 2.21. Profile plots of surface wave velocity.
21 Depth = 0.1 â 0.2 ft Depth = 0.2 â 0.3 ft Depth = 0.3 â 0.4 ft Depth = 0.4 â 0.5 ft Depth = 0.5 â 0.6 ft Depth = 0.6 â 0.7 ft Note: Section 6 = 125 to 150 ft. Surface Wave Velocity (ft/sec) Figure 2.22. Profile plots of surface wave velocity.
22 Depth = 0.1 â 0.2 ft Depth = 0.2 â 0.3 ft Depth = 0.3 â 0.4 ft Depth = 0.4 â 0.5 ft Depth = 0.5 â 0.6 ft Depth = 0.6 â 0.7 ft Note: Section 7 = 150 to 175 ft. Surface Wave Velocity (ft/sec) Figure 2.23. Profile plots of surface wave velocity.
23 Depth = 0.1 â 0.2 ft Depth = 0.2 â 0.3 ft Depth = 0.3 â 0.4 ft Depth = 0.4 â 0.5 ft Depth = 0.5 â 0.6 ft Depth = 0.6 â 0.7 ft Note: Section 8 = 175 to 200 ft. Surface Wave Velocity (ft/sec) Figure 2.24. Profile plots of surface wave velocity.
24 Depth = 0.1 â 0.2 ft Depth = 0.2 â 0.3 ft Depth = 0.3 â 0.4 ft Depth = 0.4 â 0.5 ft Depth = 0.5 â 0.6 ft Depth = 0.6 â 0.7 ft Note: Section 9 = 200 to 225 ft. Surface Wave Velocity (ft/sec) Figure 2.25. Profile plots of surface wave velocity.
25 Depth = 0.1 â 0.2 ft Depth = 0.2 â 0.3 ft Depth = 0.3 â 0.4 ft Depth = 0.4 â 0.5 ft Depth = 0.5 â 0.6 ft Depth = 0.6 â 0.7 ft Note: Section 10 = 225 to 250 ft. Surface Wave Velocity (ft/sec) Figure 2.26. Profile plots of surface wave velocity.
26 Reviews of Figure 2.26 (Section 10) show delamination below Grid Lines 6 and 7. The SASW scanning was performed past the end of the pavement section and the excess data are included in the plots. Test Results from the IE Tests Figure 2.27 shows the test results from the IE tests on the tested sections of the NCAT track. The plots are surface thick- ness tomograms separated into 50-ft sections and presented in a three-dimensional (3-D) thickness tomogram to graphi- cally show the general condition of the tested pavement at the NCAT site. The color thickness/echo depth scales are in inches and are presented in Figure 2.27. Gray represents sound pave- ments where shallow echoes (indicative of delaminations) are not present. Red represents debonding at depths between 4 and 6 in. Yellow represents debonding at depths between 2 and 4 in. Review of Figure 2.27 shows that the test results from the IE tests correctly identified debonding conditions at a depth (continued from page 15) of 5 in. (Section 1 = 0 to 25 ft and Section 8 = 175 to 200 ft). Less than 25% of the test results from the IE tests correctly identified the debonding conditions at a depth of 5 in. in Sec- tions 9 and 10 (200 to 250 ft). The IE test results were not able to identify the shallow debonding at 2 in. that was reported to be present in several sections as either a distinct echo peak corresponding to the depth of the debonding, or as a low- frequency flexural response indicative of delaminations. The asphalt surface temperature ranged from 61°F to 67°F during testing on February 27, 2011. This was warmer than during Phase 1 testing, when the asphalt surface temperature ranged from 54°F to 57°F during testing on March 19, 2010. Florida pavement results Test Setup The Florida test site was on I-75 near Gainesville, Florida. The BDS system was connected to the hitch of a towing vehicle, as shown in Figure 2.28. The six transducer wheels were coupled Note: Deeper delaminations are shown in red; shallow delaminations would be shown in yellow although none could be identified with the IE tests. Thickness 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 Figure 2.27. IE test results from the pavement at the NCAT site.
27 Note: The wheel pairs were spaced at 6 in.; separate wheel pairs were spaced at 2 ft on-center and two separate runs were performed to cover entire lane width. Figure 2.28. Test setup for Florida test site with six transducer wheels. Figure 2.29. Example of dispersion curve indicative of sound pavement. into three pairs of two wheels. Each pair of wheels was set 6 in. apart. Separate pairs of wheels were spaced at 2 to 3 ft on-center from one another. The IE and SASW scanning was performed in two runs. For the first run, the centerlines of the right, middle, and left transducer wheel pairs were located at 3, 1, and -1 ft (0 is the centerline of the tested lane), respec- tively. For the second run, the test centerline locations were located at -0.5, 3, and 5 ft, respectively. Test Results from SASW Scanning Figures 3.1 through 3.44 in Chapter 3 of this volume graphi- cally present the SASW results from the Florida pavement in 50-foot sections to allow detailed assessment. The discussion of the results that follows applies to overall trends within the data. The test results are presented in surface wave velocity plots at different depths from 0.1 ft to 0.7 ft from the surface. Surface wave velocities are presented with a color scale rang- ing from 3,500 to 1,000 ft/s. This color scale is lower than that used at the NCAT test site, which had smooth asphalt and was adjusted based on the specific site results. In general, the higher the surface wave velocity, the better condition of the asphalt pavement. Anomalies can be observed as light spots (or zones) where the surface wave velocity is notably lower. Overall, there were three categories of test results observed from the Florida test site. These categories include 1. A relatively constant and high value (â¥3,500 ft/s) of sur- face wave velocity from depths of 0.1 to 0.7 ft, as shown in Figure 2.29. This condition indicates sound pavement. Note that fewer than 2% of the tested locations were assessed with this condition. 2. A sharp drop (â¥500 ft/s drop) of surface wave velocity between 0.2 and 0.4 ft. The location of the drop was most likely the depth of debonding or a thin layer of low veloc- ity (low strength) material. Figure 2.30 presents a disper- sion curve from a location where debonding (or a layer of low-strength material) was most likely present at around 0.24 ft. Note that the majority of test results presented this type of condition. 3. A slow drop of surface wave velocity in the dispersion curve, as shown in Figure 2.31. This category indicates the condition of degradation of pavement at greater depths. Note that approximately 5% of the tested locations were identified with this type of condition. Although a significant amount of surface distress was observed on the surface of the test pavement, the majority of the data from the Florida test site had relatively high surface wave velocities (>3,500 ft/s) near the pavement surface (depths between 0.1 and 0.2 ft). This indicates that the visually observed distress only occurs on the shallow surface of the pavement.
28 In general, the conditions of all test locations on the 2,000-ft section of the tested Florida pavement are similar. The majority of test locations showed a sharp drop of surface velocity at depths between 0.2 and 0.4 ft. This drop is indica- tive of either possible debonding at these depths or the exis- tence of a thin layer of low-strength material at those depths. In addition, the condition of the pavement at the centerline of the lane seemed to be in better condition than that of the wheelpaths. Last, the pavement on the right wheelpath (0 to -5 ft) appeared to be in worse condition than that of the left wheelpath (0 to 5 ft). It should be noted that the Florida pave- ment was rough with exposed aggregate as compared to the smoother Kansas pavement and the smooth NCAT test track. Kansas pavement results Test Setup The Kansas test site was on US-400 near Pittsburg, Kansas. The BDS system was connected to the hitch of a towing vehicle, as shown in Figure 2.32. The six transducer wheels were coupled into three pairs of two. Each pair of wheels was set at a spac- ing of 6 in. Separate pairs of wheels were spaced at 2 ft on- center from one another. The IE and SASW scanning tests were performed in two runs. For the first run, the centerlines of the right, middle, and left transducer wheel pairs were located at 10.7, 8.7, and 6.7 ft from the centerline of the roadway, respectively. For the second run, the test center- line locations were at 5.7, 3.7, and 1.7 ft from the roadway centerline, respectively. Test Results from the SASW Scanning Figures 4.1 through 4.72 in Chapter 4 graphically present the SASW results in 50-foot sections to allow detailed assess- ment. This discussion of the results applies to overall trends within the data. The test results are presented in surface wave velocity plots at different depths from 0.1 ft to 0.7 ft from the surface. Surface wave velocities are presented with a color scale ranging from 3,000 to 1,000 ft/s. This color scale is lower Figure 2.30. Example of dispersion curve indicative of debonding condition. Figure 2.31. Example of dispersion curve indicative of degradation of pavement at greater depths.
29 than that used at the NCAT test site and was adjusted on the basis of specific site results. In general, the higher the surface wave velocity, the better the condition of the asphalt pave- ment. Anomalies can be observed as light spots (or zones) where the surface wave velocity is notably lower. Overall, the majority of the data from the Kansas test site had relatively high surface wave velocities (>3,000 ft/s) near the pavement surface (<0.3 ft deep) and decreasing veloci- ties with depth, thus indicating weaker underlying materi- als. Overall interpretation of the data is that the road has a relatively new asphalt overlay of good condition with an older underlying pavement composed of weaker material with widely variable condition. In some areas, the velocity decrease with depth was gradual and was interpreted as an expected result of newer materials overlaying older material. However, in other areas the velocity decrease was sharp and severe, indicating that the underlying materials were of poor condition. Sharp drops in the dispersion curve (such as those observed at the NCAT test track) were not typically observed in the data. Instead, the drop Note: The wheel pairs were spaced at 6 in.; separate wheel pairs were spaced at 2 feet on-center, and two separate runs were performed to cover the entire lane width. Figure 2.32. Test setup for Kansas test site with six transducer wheels. in the surface wave velocity was typically steady and indicated a more gradual change in material strength with depth. How- ever, the degree of change appears severe. In many areas, dis- persion curves at depths of 0.4 to 0.7 ft were less than half of those at the pavement surface. ⢠0 to 160 ft. In this section the roadway is actually a concrete bridge deck that had velocity far above the color scale maxi- mum of 3,000 ft/s. ⢠160 to 210 ft. The section immediately after the concrete bridge had relatively high velocities throughout the pave- ment section and exhibited less of a velocity decrease with depth. This may indicate stiffer subbase preparation for the bridge approach slab. ⢠210 to 350 ft. In relation to much of the Kansas test seg- ment, this section appeared to be in reasonably good con- dition with slightly lower velocities and velocity values decreasing with depth. ⢠350 to 1,000 ft. This section showed significant degrada- tion with depth from 0 to 5 ft from the centerline of the roadway. As noted above, the surface layer still had reason- ably sound velocities, but the lower layers had significantly lower velocities. ⢠1,000 to 2,200 ft. The degradation in this section was similar to that in the previous section but extended across the full lane width. Again, the surface layer had reasonable surface wave velocities, which then drastically decreased with depth. ⢠2,200 to 2,500 ft. This section had relatively high velocities throughout the cross section and was considered one of the best areas of the test segment. ⢠2,500 to 3,100 ft. This section showed typical degradation, increases with depth, and full lane width. ⢠3,100 to 3,285 ft. This section has relatively high velocities throughout the cross section and was considered one of the best areas of the test segment. ⢠3,285 to 3,495 ft. This section showed typical degradation, increases with depth, and full lane width. ⢠3,495 to 3,600 ft. This section had relatively high velocities throughout the cross section and was considered one of the best areas of the test segment.
