Combining Traditional and Non-Traditional NDT ...

Paper No. 03-4214

Duplication for publication or sale is strictly prohibited without prior written permission of the Transportation Research Board

Title: Combining Traditional and Non-Traditional NDT Techniques to Evaluate Virginia’s Interstate 81

Authors: Imad L. Al-Qadi, Trenton M. Clark, David T. Lee, Samer Lahouar, Amara Loulizi

Transportation Research Board 82nd Annual Meeting January 12-16, 2003 Washington, D.C.

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

Combining Traditional and Non-Traditional NDT Techniques to Evaluate Virginia’s Interstate 81 Imad L. Al-Qadi Charles E. Via, Jr. Professor of Civil and Environmental Engineering, Leader of the Roadway Infrastructure Group Virginia Tech Transportation Institute 200 Patton Hall Virginia Tech Blacksburg, VA 24061-0105 Tel: 540 231-5262, Fax: 540 231-7532 e-mail: [email protected] Trenton M. Clark Pavement Design and Evaluation Program Engineer Virginia Department of Transportation 1401 E. Broad Street Richmond, VA 23219 Tel: 804 328-3129, Fax: 804 328-3136 e-mail: [email protected] David T. Lee Salem District Materials Engineer Virginia Department of Transportation 731 Harrison Avenue Salem, VA 24153 Tel: 540-387-5383, Fax: 540 387-5503 e-mail: [email protected] Samer Lahouar Graduate Research Assistant Virginia Tech Transportation Institute 3500 Transportation Research Plaza Virginia Tech Blacksburg, VA 24061-0536 Tel: 540 231-1588, Fax: 540 231-1555 e-mail: [email protected]

Amara Loulizi Research Scientist Virginia Tech Transportation Institute 3500 Transportation Research Plaza Virginia Tech Blacksburg, VA 24061-0536 Tel: 540 231-1504, Fax: 540 231-1555 e-mail: [email protected]



Al-Qadi, Clark, Lee, Lahouar, and Loulizi

1

Combining Traditional and Non-Traditional NDT Techniques to Evaluate Virginia’s Interstate 81

ABSTRACT With increasing traffic volumes and truck weights on Virginia’s Interstate 81, the entire corridor will be rehabilitated and expanded to meet future demands. Initial construction estimates included the complete reconstruction of I-81. This paper presents the methods employed by the Virginia Department of Transportation (VDOT) to structurally evaluate approximately 108 lane km (68 lane mi) of pavement using Falling Weight Deflectometer (FWD) testing, Ground Penetrating Radar (GPR), and pavement coring in order to quantify reconstruction and rehabilitation limits. The recommended rehabilitation design based on this study is also presented. While pavement coring and FWD testing are not new evaluation techniques for VDOT, the use of GPR to measure continuous changes in pavement layer thickness and to identify areas of moisture in the pavement structure is. The FWD data and results were used to divide the flexible pavement portion of the project into eleven sections and to assess the structural capacity of the pavement. Material retrieved from pavement cores were used to verify layer thickness and condition. The results of the GPR testing increased the accuracy of the FWD analysis by providing accurate and continuous thickness data, located additional potential problem areas, and assisted in the pavement rehabilitation design process. By combining the FWD, core, and GPR results, VDOT was able to determine which segments of the project require reconstruction and which segments require major rehabilitation; thus, saving millions of future construction dollars.

Submission date: 11/15/02 Word count: 4420 + 9 (tables and figures) = 6670




2

INTRODUCTION As pavements are subjected to traffic over time, they tend to weaken because of structural fatigue, which is a phenomenon resulting from repetitive deflections beneath moving wheel loads, unlike the concept of bending fatigue in metal structures. It is possible to assess the amount of service life remaining in a pavement system if information about its structural response to predetermined loads is available and if traffic-loading conditions are known. A study was performed on a portion of Virginia Interstate 81 (I-81) to assess its structural conditions. The objectives of this study were to: (1) determine the in-place pavement structural capacity in terms of effective Structural Number (SNeff), (2) identify weak areas where reconstruction would be necessary, and (3) determine the amount of rehabilitation (material removal, replacement, and additional structure) required to meet future traffic loading. The ultimate goal was to reduce VDOT’s construction costs while not jeopardizing the structural and functional integrity of I-81. The structural assessment was accomplished by conducting a comprehensive loadtesting program at approximately 700 locations using a Falling Weight Deflectometer (FWD) system and analyzing its results, along with results of the subsurface investigation obtained by means of a Ground Penetrating Radar (GPR) system, laboratory tests, and visual surveys. This paper presents the techniques used to estimate the structural capacity of the studied section of I-81. It includes a description of the project, GPR data analysis method used to determine the layer thickness, FWD data collection and analysis methods to assess the structural capacity, and results and their incorporation into the final pavement recommendations. DESCRIPTION OF I-81 AND SCOPE OF STUDY Interstate 81 in the State of Virginia is a typical four lane divided highway built as part of the Eisenhower Interstate Highway System. It begins in far southwestern Virginia at the Virginia/Tennessee border and continues for approximately 523km (325mi) to the Virginia/West Virginia border. The majority of I-81 was constructed during the 1960’s and early 1970’s with the last section being completed around 1986. The Virginia Department of Transportation (VDOT) is in the very early stages of preparing to upgrade I-81 throughout its length. This upgrade is being primarily driven by the significant increase in truck traffic since the deregulation of the trucking industry and vehicle traffic. I-81 was originally designed to handle approximately 25,000 vehicles per day (vpd) with around 15% truck traffic. Today many sections carry well over 50,000 vpd with as much as 45% truck traffic and are predicted to handle approximately 100,000 vpd by 2030. This study covers a specific section of I-81 built between 1963 and 1965 within the counties of Roanoke and Botetourt in the Southwestern portion of Virginia. The evaluated section is split into six design segments totaling 27.2km (17mi) of centerline roadway and presently carries up to 66,500 vpd with approximately 40% truck traffic. As part of the design process and largely due to the significant increase in truck traffic along the corridor, VDOT’s Salem District has embarked on an extensive evaluation process in order to determine the condition of the existing pavement throughout the section. The pavement structure throughout the section is predominately flexible with only 6km (3.8mi) centerline roadway being composite (within these design segments). The original pavement design for the flexible sections consisted of a 150mm (6in) aggregate subbase material, a 190mm (7.5in) hot-mix asphalt (HMA) base, a 30mm (1.25in) HMA intermediate course, and a 20mm (0.75in) HMA wearing surface. The original design for the composite section initially consisted of a 150mm (6in) aggregate subbase, a 50mm (2in) limestone screenings-leveling course (fines passing the 6.3mm size sieve), and 228mm (9in) of wire mesh reinforced jointed concrete pavement. In 1995, a 90mm (3.5in) HMA concrete overlay was placed. This pavement was evaluated structurally using the FWD and GPR data. Pavement cores were obtained to verify VDOT’s construction history information and to evaluate the results of the GPR data. GROUND PENETRATING RADAR TESTING The use of GPR has developed considerably over the past thirty years to assess civil structures such as pavements and bridges. Ground penetrating radar was used to locate dowels (1) and detect voids (2) in rigid pavements. It was also used in flexible pavements to detect moisture in the HMA layer (3) and to locate moisture in the base layer that may lead to structural damage (4). Currently, the majority of GPR applications in pavements are focused on determining layer thicknesses. This data, which can be used by modern pavement management systems (PMS), is essential for planning project rehabilitation (5). Furthermore, it can be used as input for other nondestructive testing techniques such as FWD.




3

The GPR system used in this study is an impulse GPR, which is based on sending an electromagnetic (EM) pulse through the antenna to the ground and then recording the reflected signals from the layer interfaces where there is a contrast in the dielectric properties. Measure of the time difference between the reflected signals (usually known as two-way travel time or time of flight, TOF) can be used in conjunction with the dielectric properties of the traversed medium to determine its thickness. The dielectric properties of the layers can also be estimated from the reflected signal and, therefore, used to locate the presence of any defects in the subsurface such as moisture. For VDOT, the main goal of the GPR survey performed during this study was to evaluate the feasibility of GPR to determine the HMA layer thickness of the existing pavement system. The HMA thickness determination process included both the flexible pavement section and the overlay over the reinforced concrete section. A secondary objective of the GPR survey was to assess the performance of GPR in identifying moisture areas in the pavement system, which usually leads to stripping of the HMA. In fact, identification of these areas and their depth in the pavement structure greatly affects pavement rehabilitation techniques. An impulse GPR system with an air-coupled antenna and a ground-coupled antenna was used to accomplish VDOT’s goals. As discussed below, data collected by both systems were used in a modified common midpoint analysis technique to accurately estimate the HMA layer thickness. The air-coupled system is composed of a pair of separate horn antennas (one serves as a transmitter and the other as a receiver) each having a 1 GHz frequency bandwidth. The ground-coupled system is composed of a single antenna operating as transmitter and receiver and having a 900 MHz bandwidth. As depicted in Figure 1a, both antenna sets were mounted behind a van during the survey. The survey speed was approximately 16km/h; imposed by the use of the ground-coupled antenna, and the GPR scan rate was set to 40 scans per second, which resulted in approximately one scan every 110mm (4.3in). Moreover, a Global Positioning System (GPS) was used to locate the collected data spatially. To achieve an accurate positioning, the GPS speed data (with an accuracy of 0.1m/s or 0.4km/h), combined with the acquisition time and the GPR scan rate, was used instead of the absolute position recorded by the unit. During this survey, GPR data was collected continuously between milepost 137 and milepost 154 in each lane (traveling lane and passing lane) of the northbound and southbound directions. Ground penetrating radar data was collected using both antenna systems simultaneously allow using the common midpoint technique for HMA thickness estimation. Additionally, static measurements were taken near the core locations extracted by VDOT. This data was used during the analysis to verify the accuracy of the GPR results. It should be noted that to avoid interference between the two antenna systems used during the survey, the ground-coupled antenna wasn’t placed directly underneath the air-coupled antenna. Instead, it was placed at a fixed distance in front of it. FALLING WEIGHT DEFLECTOMETER TESTING Pavement surface deflection measurements provide valuable information on the structural condition of pavement systems (6). In fact, stronger pavements, having good quality materials and thick layers, deflect less under a given wheel load than weaker pavements with thin or deteriorated layers. Therefore, in response to the need for reliable tools to evaluate pavement structures, deflection devices for use in nondestructive testing have been developed, whereby tests can be rapidly conducted at any point along a pavement section. The falling weight deflectometer (FWD) is one of those nondestructive devices that has recently gained widespread acceptance (since the introduction of the static Benkelman beam at the WASHO Road Test in the early 1950's) for the structural evaluation of pavements based on their deflection responses (7). An in-depth summary of historical developments of nondestructive testing, backcalculation, and theoretical considerations, as well as associated technologies, is presented by Lytton in (8). The VDOT routinely uses the Dynatest Model 8000 FWD, shown in Figure 1b, to evaluate pavement structures for thickness design work and as an analysis tool for failure investigations. This FWD unit, which is trailer mounted and towed behind a van with an on-board processing computer, is the most widely used in the United States. The impulse force is created by dropping different weights, namely 50kg (110lbs), 100kg (220lbs), 200kg (440lbs), or 300kg (660lbs) from different heights ranging from 20mm (0.8in) to 380mm (15in). By varying the drop heights and drop weights, a peak force range of 6.7kN (1,500lbs) to 107kN (24,000lbs) can be developed. The load is transmitted to the pavement through a 300mm (11.8in) diameter loading plate and measured using a load cell. The resulting pavement deflections at different points on the surface are also captured using a set of velocity transducers. In keeping with standard VDOT procedures for evaluating flexible and composite pavement structures,




4

pavement basin testing was performed to generate information about the response of subsurface pavement layers to the applied test loads. The FWD sequence of operations used in this project was as follows. The FWD unit was moved to the selected test points in the travel lane. The loading plate and transducers were then lowered hydraulically to the pavement surface. Next, six drops [2 at 53kN (12,000lbs), 2 at 40kN (9,000lbs) and 2 at 71kN (16,000lbs)] were conducted, and the deflections resulting from the 40kN (9,000lbs) and 71kN (16,000lbs) drops were recorded for further analysis. The system automatically recorded and stored deflections measured by the nine velocity transducers, which were located at radial distances of 0, 203, 305, 457, 610, 914, 1219, 1524, and 1829mm (0, 8, 12, 18, 24, 36, 48, 60, and 72in) from the center of the load plate. In addition to the load and deflection data, pavement surface temperatures were measured automatically with the FWD at each test location. DATA ANALYSIS AND RESULTS Ground Penetrating Radar Data The collected GPR data was analyzed in two steps: the first step was to manually examine all the collected data using a LineScan view in order to separate the different homogeneous pavement structures such as bridges, rehabilitated areas, and concrete sections, in addition to the suspicious areas. Those homogenous sections were later analyzed during the second step with data analysis software developed by Virginia Tech, to determine the HMA layer thickness and its dielectric constant along all the surveyed sections. Figures 2 and 3 represent a colored LineScan view of the raw GPR data. In a LineScan view, the x-axis represents the scan number (proportional to the spatial position) and the y-axis represents depth (shown here in ns, since raw data is presented, but could be converted to mm after determining the dielectric constant of each layer). For each scan, the intensity of the reflection is mapped to a color according to the scale on the right rectangle of the figure, with zero reflection corresponding to black. A non-linear mapping scale was used in order to intensify the low reflections. These figures show samples of some important features that were detected visually from the collected raw GPR data. Figure 2a depicts a typical cross-section of the flexible pavement where the HMA, base, and subgrade layers are clearly identified. The presence of a reflection within the HMA layer in this figure is a good indicator of moisture (and therefore most probably stripping) presence. Figure 2b shows a typical rigid pavement overlaid with HMA, with the reinforcement shown as small hyperbolas, which is the GPR signature of any metal cylinder embedded in the pavement. Figure 3a illustrates a full-depth HMA repair within the concrete section, and Figure 3b shows a flexible pavement section with very irregular layer thicknesses. The data analysis algorithm works on a scan-by-scan basis to determine the average dielectric constant of the HMA layer and therefore its thickness, at a spatial resolution of 110mm corresponding to the acquisition resolution. First, the analysis starts by selecting a scan obtained by the ground-coupled antenna and the corresponding scan obtained by the air-coupled antenna, taking into account the number of scans corresponding to the distance between the two antenna systems. Next, using a matched filter detector (9), the system locates the reflections of interest from both scans, namely the surface reflections and the HMA/base reflections. Based on the reflection locations found, the two-way travel times in the HMA layer are computed. Then using a common midpoint (CMP) algorithm (10), the dielectric constant of HMA and therefore the speed of EM waves in that layer are calculated. Consequently, using the two-way travel time and the velocity of the EM waves determined by the previous steps, the HMA layer thickness is estimated at the location corresponding to the processed scan. Finally, using GPS data, the scan numbers are converted to spatial locations and the results are saved for later display. A detailed description of the data analysis procedure is presented in (11). The results obtained from the entire surveyed portion show that the HMA layer thickness varies from 250mm to 400mm in the flexible pavement with some areas of accumulated moisture confirmed by the relatively high dielectric constant. The HMA overlay thickness over the concrete section varies between 90mm and 110mm with the presence of many irregular full depth repaired areas with different lengths and spacing. The length of the repaired areas varies from less than 10m to around 575m with a spacing as low as 8m in some cases. The repaired sections are usually very irregular in thickness and have different levels of moisture content. Generally, the traveling lane in each direction appears to be in a more severe condition than the passing lane. Virginia Department of Transportation’s knowledge of the repairs performed to the composite pavement section confirms these results. To verify the accuracy of the GPR data analysis software, the stationary measurements collected near the core locations were used to estimate the HMA thickness. The results for the northbound and southbound traveling lanes showed that the error ranges from 1 to 15% with a mean around 6.8%. Figure 4 illustrates the correlation




5

between the thicknesses measured from the cores and the thicknesses computed from the GPR data. It is clear from this figure that the majority of the data points are within the 6.8% error margin. It should be noted that thickness errors greater than one standard deviation from the mean could be attributed to the position where the data was collected, since in some sections the stationary GPR scanning was not performed exactly over the core locations, because the cores were taken prior to the GPR survey. In addition, the thickness error could be attributed to direct thickness measurements on cores. In fact, it was reported that thicknesses measured from the same core could vary with 2.7% on average (12). Falling Weight Deflectometer Data Based on a preliminary analysis of the FWD data using the cumulative sums approach outlined in the AASHTO Design Guide (13) and changes in overall pavement structure, the project was divided into eleven sections – six northbound and five southbound as shown in Table 1. Virginia Department of Transportation determined that these sections may have homogeneous structures based on the deflection responses. Results of the deflection testing activities were used in conjunction with subgrade soil classifications to backcalculate and validate the subgrade resilient modulus results, and therefore determine the structural capacity of the existing pavements. Since the strength of flexible pavements depends on the temperature of HMA materials in the structure, deflections were corrected to the equivalent of 20ºC (68ºF) in accordance with AASHTO guidelines (13). The next step involved calculating the effective structural number (SNeff) for each location (or group of structurally similar contiguous segments within a site). The SNeff is a function of subgrade resilient modulus, as well as the condition and total thickness of all pavement layers above the subgrade. The calculations for the effective structural number and subgrade resilient modulus were based on equations found in the AASHTO guide Appendix L (13). A VDOT developed deflection analysis program named TAG was used for these calculations. The structural capacity for the pavement was evaluated using deflection and load data collected with the FWD and pavement structure information provided by the GPR analysis for the homogeneous sections. A summary of the SN results found for the 11 sections are provided in Table 2, whereas the subgrade resilient modulus results are provided in Table 3. To provide a comparison between a sound pavement structure and the existing pavement structure, the sound or original structural number (SN0) was estimated using the layer and material information provided by the GPR analysis and coring. Figures 5a and 5b illustrate the estimated SN0 and SNeff for the northbound and southbound traveling lanes of I-81 through the analysis area. A significant difference in the SN0 and SNeff for the length of the analysis area is very clear particularly in the northbound data. It should be noted that the area from milepost 142 to milepost 143 was reconstructed in 1996-97, which explains the relatively high SN in that area. PAVEMENT REHABILITATION RECOMMENDATIONS Based on the data collected for this study and the preliminary traffic forecasts, pavement rehabilitation recommendations were developed for Roanoke and Botetourt Counties. Roanoke County Section (Milepost 137.0-147.4) Virginia Department of Transportation concluded from the results of the FWD structural analysis that the total pavement structure between Milepost 137.0 and 140.6 has deteriorated to a point beyond its useful life (less than 30% of remaining structural life) as an integral pavement element and therefore it should be completely reconstructed. The deteriorated condition of the HMA layers was evident in the pavement cores. Laboratory tests did not indicate that HMA was prone to striping; therefore, it was hypothesized that HMA materials had fatigued due to repeated truck loading. For many years VDOT did not mill deteriorated HMA, but instead overlaid it, that would be the reason why the deteriorated HMA exists in the middle of the pavement structure. By combining the results of the structural analysis, the evaluation of the pavement cores recovered, and the GPR data analysis, it was determined that the upper half of the pavement structure between Milepost 140.6 and 147.4 has deteriorated to a point beyond its useful life as an integral pavement element and should be removed. However, the lower half of this pavement section appears to have a significant portion of its useful life remaining. Rehabilitation of this pavement section by removing and replacing the upper layers appears feasible. Additional structure will be required to meet future traffic demands.




6

Botetourt County Section (Milepost 147.4-153.0) The analysis of the FWD data for this section of composite pavement was inconclusive with backcalculation results showing the elastic modulus of the reinforced Portland cement concrete exceeding 200,000MPa (30,000ksi). This has been VDOT’s experience with composite pavement structures. However, based on VDOT’s knowledge of the underlying jointed reinforced concrete pavement and its condition, the significant amount of repair found, and the moisture in the HMA overlay as indicated by the GPR data, this section would require reconstruction. In addition to the structural evaluation, visual and ride quality surveys for this section show an increase in pavement distress and decrease in ride quality as a result of the slabs moving. Pavement Designs Pavement designs for Roanoke and Botetourt Counties were developed using mechanistic and empirical based methodologies. For the majority of new and rehabilitation pavement designs, VDOT uses the procedures and guidelines outlined in the 1993 AASHTO Pavement Design Guidelines. However, based on the traffic projections for the I-81 corridor and realizing the limitations of the empirical AASHTO pavement design procedure, a mechanistic analysis was performed using the Asphalt Institute’s DAMA software, which utilized the methods detailed in MS-1 (14). For I-81, VDOT desired a long-life pavement. For the fatigue and deformation life criteria, the pavement structure must not fail prior to 30 years and preferably remain structurally sound for 50 years. The structure must provide a high level of serviceability with limited pavement maintenance in the future to minimize user delay. Using these criteria and inputs detailed in Table 4, the following pavement design was developed for the reconstruction sections: • • • •

38mm (1.5in) HMA Wearing Surface 50mm (2.0in) HMA Intermediate Course 300mm (12in) HMA Base Course 150mm (6in) Cement Treated Aggregate Sub-Base Course

For the rehabilitation sections, approximately 50% of the HMA material thickness (150–175 mm) will be removed and replaced with: • • •

38mm (1.5in) HMA Wearing Surface 50mm (2.0in) HMA Intermediate Course 150mm (6in) HMA Base Course

While these pavement designs exceed the criteria established for the pavement, the additional thickness is considered insurance to cover unknowns in future truck weights, tire types and pressures, and traffic volumes. CONCLUSIONS A 27km portion of Interstate 81 (I-81) was studied in both directions to determine future pavement rehabilitation needs. Virginia Department of Transportation incorporated traditional and non-traditional methods to evaluate the existing pavement. While the overall goal for the study was to quantify rehabilitation and reconstruction limits, VDOT’s objectives for incorporating Ground Penetrating Radar (GPR) into this study were to determine GPR’s accuracy in estimating the thickness of the HMA and other structural pavement layers, and identifying any defects such as moisture, which causes stripping and weakens the pavement system. By comparing the GPR results to a set of cores used to verify the accuracy of the method, a mean thickness error of 6.8% was achieved. Additionally, the GPR located moisture pockets in the pavement structure both known and unknown to VDOT. Overall, VDOT was able to define 11 flexible pavement segments in this study. Each segment was evaluated with FWD, GPR, and core results. The results were used to identify approximately 5.6 directional km (3.5mi) of flexible pavement where reconstruction is required. Additionally, 11.2 directional km (7.0mi) were determined to be deficient but could be rehabilitated with a 150–175mm removal and replacement. Initially, VDOT intended to reconstruct the entire 27.2km (17mi); however, through a thorough investigation of the pavement, the reconstruction limits were reduced and millions of future construction dollars were saved. ACKNOWLEDGMENTS The authors would like to thank Mr. William Hughes for his analysis of the FWD data, and Mr. Brian Prowell of NCAT (formerly with VTRC) for his condition analysis of the HMA pavement cores.




7

REFERENCES 1. 2.

3. 4. 5. 6. 7. 8.

9. 10. 11.

12. 13. 14.

Donohue & Associates, Inc. A Nondestructive Method for Determining the Thickness of Sound Concrete on Older Pavements, Final Report HR-250, Iowa Highway Research Board, Iowa, USA, 1983. Smith, S. S., and T. Scullion, Development of Ground Penetrating Radar Equipment for Detecting Pavement Condition for Preventive Maintenance, Final Report Strategic Highway Research Program SHRP-672, National Research Council, Washington, DC, USA, 177pp, 1993. Al-Qadi, I. L., Using Microwave Measurements to Detect Moisture in Hot-Mix Asphalt, Journal of Testing and Evaluation, JTEVA, ASTM, Vol. 20, No. 1, pp. 45-50, 1992. Rmeili, E. and T. Scullion, Detecting Stripping in Asphalt Concrete Layers Using Ground Penetrating Radar, Paper No. 97-0508, Submitted to Transportation Research Board, 1997. Maser, K. R., Condition Assessment of Transportation Infrastructure Using Ground Penetrating Radar, Journal of Infrastructure Systems, Vol. 2, no. 2, pp. 94-101, 1996. Hoffman, M. S., and M. R. Thompson. Comparative Study of Selected Nondestructive Testing Devices. In Transportation Research Record 852, TRB, National Research Council, Washington, DC, pp. 32-40, 1983. Graves, C. R., and V. P. Drnevich. Calculating Pavement Deflections with Velocity Transducers. In Transportation Research Record 1293, TRB, National Research Council, Washington, DC, pp. 12-23, 1991. Lytton, R. L., Backcalculation of Pavement Layer Properties, In Nondestructive Testing of Pavements and Backcalculation of Moduli, ASTM STP 1026, American Society for Testing and Materials, Philadelphia, PA, pp. 7-38, 1989. Therrien, C. W., Discrete Random Signals and Statistical Signal Processing, Prentice Hall, Englewood Cliffs, 1992. Schneider, W. A., The Common Depth Point Stack, Proceedings of the IEEE, Vol. 72, No. 10, pp. 1238-1254, 1984. Lahouar, S., I. L. Al-Qadi, A. Loulizi, C. M. Trenton, and D. T. Lee. Development of an Approach to Determine In-Situ Dielectric Constant of Pavements and Its Successful Implementation at Interstate 81, Paper No. 02-2596., Presented at The 81st Transportation Research Board Annual Meeting, Washington DC, January 13-17, 2002. Al-Qadi, I. L., S. Lahouar, and A. Loulizi, Successful Application of GPR for Quality Assurance/Quality Control of New Pavements, Submitted to Transportation Research Board Annual Meeting, 2003. AASHTO Guide for Design of Pavement Structures, 1993. Asphalt Pavement for Highways and Streets Manual Series No. 1, Asphalt Institute, 1981.




8

LIST OF TABLES AND FIGURES

TABLE 1 Project Sections Based on FWD Data TABLE 2 Structural Number Results for the Studied Sections in I-81 TABLE 3 Subgrade Resilient Modulus Results for the Studied Sections in I-81 TABLE 4 Pavement Design Inputs

FIGURE 1 (a) GPR Van Used during the Survey Showing Antennas Configuration, (b) VDOT’s FWD Unit. FIGURE 2 (a) GPR Scans Showing HMA/Base Interface, Base/Subgrade Interface and Moisture within HMA Layer, (b) GPR Scans Showing a Concrete Section. FIGURE 3 (a) GPR Scans Showing a Full Depth Repair in a Concrete Section, (b) GPR Scans Showing a Section with Irregular Layer Thicknesses. FIGURE 4 Correlations between Core Thickness and Computed GPR Thickness. FIGURE 5 Structural Number Results for (a) I-81 Northbound (b) I-81 Southbound.




9

TABLE 1 Project Sections Based on FWD Data Section # 1 2 3 4 5 6 7 8 9 10 11


Direction North North North North North North South South South South South

From Milepost 137.00 138.58 141.87 142.41 145.34 145.46 136.96 138.61 140.77 141.31 142.03

To Milepost 138.58 141.87 142.41 145.34 145.46 148.22 138.61 140.77 141.31 142.03 148.21



10

TABLE 2 Structural Number Results for the Studied Sections in I-81 Section # 1 2 3 4 5 6 7 8 9 10 11

Avg. SNeff 4.40 4.13 7.30 5.55 5.77 5.29 5.00 4.91 4.56 5.50 5.49

Std. Dev. SNeff 0.28 0.34 0.57 0.37 0.32 0.40 0.34 0.39 0.37 0.38 0.38

CV SNeff(a) 6.4 8.2 7.8 6.7 5.5 7.5 6.8 7.9 8.2 6.9 7.0

SNo(b) 6.222 5.676 7.788 6.672 6.504 5.874 6.222 6.390 6.054 6.978 5.874

Cf(c) 0.71 0.73 0.94 0.83 0.89 0.90 0.80 0.77 0.75 0.79 0.93

Notes: a) CV SNeff: Coefficient of Variance for the effective structural number; higher the percentage, more variation in the section. b) SNo: Original structural number for the pavement c) Cf : Structural Condition Factor for the section (SNeff / SNo)




11

TABLE 3 Subgrade Resilient Modulus Results for the Studied Sections in I-81 Section # 1 2 3 4 5 6 7 8 9 10 11

Avg. Esg 32,300 32,000 48,300 34,000 41,300 34,400 37,400 34,100 32,800 43,100 37,800

Std. Dev. Esg 9,200 8,600 7,200 11,400 3,300 9,900 8,200 8,500 8,500 8,900 10,200

CV Esg (a) 28 27 15 33 8 29 22 25 26 21 27

Notes: a) CV Esg: Coefficient of Variance for the subgrade modulus; higher the percentage, more variation in the section




12

TABLE 4 Pavement Design Inputs Parameter Number of Dual Tire Repetitions per Month Tire Pressure Tire Load HMA Surface Course Poisson’s Ratio HMA Surface Course Elastic Layer Modulus HMA Intermediate Course Poisson’s Ratio HMA Intermediate Course Elastic Layer Modulus HMA Base Course Poisson’s Ratio HMA Base Course Elastic Layer Modulus Cement Treated Aggregate Sub-Base Course Poisson’s Ratio Cement Treated Aggregate Sub-Base Course Elastic Layer Modulus Subgrade Poisson’s Ratio Subgrade Elastic Layer Modulus


Value 1,500,000 670kPa (100psi) 2040kg (4500lbs) 0.35 3,800MPa (550,000psi) 0.35 3,450MPa (500,000psi) 0.35 3,100MPa (450,000psi) 0.20 4,100MPa (600,000psi) 0.45 51MPa (7,500psi)



13

(a)

(b)

FIGURE 1 (a) GPR Van Used during the Survey Showing Antennas Configuration, (b) VDOT’s FWD Unit.




14

Scan number

Moisture in HMA HMA

Time (ns)

Base

Subgrade

(a)

Scan number HMA

Concrete

Time (ns)

Reinforcement

(b)

FIGURE 2 (a) GPR Scans Showing HMA/Base Interface, Base/Subgrade Interface and Moisture within HMA Layer, (b) GPR Scans Showing a Concrete Section.




15

Time (ns)

Scan number

Concrete section

Full depth repair

Concrete section

(a)

Time (ns)

Scan number

(b)

FIGURE 3 (a) GPR Scans Showing a Full Depth Repair in a Concrete Section, (b) GPR Scans Showing a Section with Irregular Layer Thicknesses.




16

Equity line 400

+6.8% 375 350

-6.8% 325 300 275 250 250

275

300

325

350

375

400

Core Thickness (mm)

FIGURE 4 Correlations between Core Thickness and Computed GPR Thickness.




17

10 9

SNeff

8

SNo

7 6 5 4 3 2 1 0

Position (milepost)

(a)

10 9

SNeff

8

SNo

7 6 5 4 3 2 1 0

Position (milepost)

(b)

FIGURE 5 Structural Number Results for (a) I-81 Northbound (b) I-81 Southbound.

