Nondestructive Testing to Identify Delaminations Between HMA Layers: Phase III–Develop User Guidelines (2014)

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.

RIS Hi-Pave System Specification System Specification Recommended Laptop: Panasonic CF-19 Tough-Book Max. acquisition speed: 260 kmh (150 mph) Positioning: Survey wheel and/or GPS Number of control unit: DAD 1CH FW - DAD MCH FW Scan rate: (@ 512 samples/scan) 362 scans/sec or 724 scans/sec. Scan Interval: 10 scans/mt. Power Supply: SLA Battery 12VDC 12 AH Antennas Specification: Environmental: IP65 Antenna FootPrint: 51 × 22 cm Number of hardware channels: from 1 to 8 Antenna Central Frequencies: 2 GHz or 1 GHz Antenna Polarization: Horizontal (HH) Antenna type: Air launched K2 Fastwave Acquisition Software This software operates both with single GPR sections and with sets of homogeneously acquired single or multi-channel profile data; the radar maps can be viewed simultaneously and the sections run using automatic scrolling. It is possible to apply filters as well as to put marks directly on the radargram. The acquired radargrams are stored in raw files that can be opened with GRED HD software. Data can be exported in several formats. Radargram acquired with K2 Fastwave acquisition software. B35

GRED HD 3D Post-processing Software (Data Processing) GRED HD 3D is processing software designed to be an interface for the IDS RIS GPR family of products. The code can identify the various layers using both automatic and manual procedures. The radar Processing Software picture shows a radar map with several layers automatically identified on the radar map and reported in dedicated layers windows; the layers’ thicknesses are confirmed by the core sample data inserted in the GRED HD 3D windows. Data interpretation can be done also with GPR-Slice and Reflex software or alternatively IDS can be available to share proprietary information about the data format. Radar processing software. System Calibration The calibration is done automatically by the software. Having a template of the coring can be helpful to estimate with precision the propagation velocity that leads to real layer thickness. The software has a feature to introduce coring data and relate those with the layers in the radargrams in the post-processing phase. Data Export Data can be exported in several formats, including Excel, ASCII, CAD, and GIS with GIS coordinates. B36

GPS Compatibility The GPS device can be connected to the RS232 port of the data logger computer and GPS data saved together with radar profiles for knowing the position of each scan in absolute coordinates; the software can accept the GPS streaming from any device compatible with the NMEA standard. The GPS positions can be easily loaded on software like Google Earth. GPR track acquisition loaded on Goggle Earth map. B37

APPENDIX C1 SASW/IE User Guidelines C1

Use of Spectral Analysis of Surface Waves (SASW) and Impact Echo (IE) for Identifying Asphalt Pavement Delamination: User Guidelines Prepared by the SHRP R06D Research Team This document provides guidelines for using nondestructive testing (NDT) methods that utilize the spectral analysis of surface waves (SASW) and impact echo (IE) technologies to identify delamination in asphalt pavements. This guideline is applicable to all the SASW and IE devices for evaluating delamination in asphalt pavements. Users are advised to understand both SASW and IE because the test equipment may have the ability to measure both. Selection of which data has the highest level of accuracy and confidence will depend on the understanding of each technology and the field-testing conditions. 1 General Theory The SASW and IE methods are used for determining material properties and detecting defects based on the principles of elastic wave propagation. The methods are conducted by impacting the surface of a material to generate three primary elastic waves, including R-, P-, and S-waves and then measuring the waves propagating to some distance from the source impact. The SASW measures the changes in surface wave (i.e., R-wave) dispersion characteristics and elastic properties to determine the pavement material stiffness (modulus) and the potential for material defects. The IE method measures body-wave (i.e., P-wave) reflections in the material response to determine the thickness of the bound layer or the potential location of a defect in the bound layer. Surface waves propagate closer to the surface and have higher amplitudes than body waves. 1.1 Spectral Analysis of Surface Waves In the SASW test, the pavement is impacted with a short, high-frequency source creating a surface wave that propagates away from the source. Two receivers are spaced at different distances from the source to detect the arriving surface wave. The data from these two locations are used to calculate the wavelength versus velocity (dispersion curve) for the surface wave. Figure C1.1 illustrates the SASW testing and data analysis process. Since wavelength is related to depth of penetration, and since surface wave and shear wave velocities are very close, this dispersion curve is interpreted as a relationship between shear wave velocity and depth. A sharp drop in velocity at a particular depth is indicative of a discontinuity in the pavement structure which could be associated with delamination and stripping. Figures C1.2 (a) and C1.2 (b) show the dispersion curves for an intact pavement and a pavement with a known delaminated interface at a depth of 5 in. Figure C1.3 is a SASW data analysis screen display. The material modulus determined from the SASW measurements represents low-magnitude (less than 1 micro-strain), high-frequency (greater than 3,000 Hz) C2

values. These material modulus values are higher than modulus values computed from current dynamic modulus tests. Figure C1.1. SASW testing and data analysis process. Source First Sensor Second Sensor Time Depth Shear Wave Velocity (a) Raw Waveform (b) Processed Waveform Station and Offset Depth Depth Slice (c) Processed Shear Wave Profiles at Successive Locations (d) Velocity Map at Depth Slice for an Array of Locations Station O ffs et Velocity gray scale C3

(a) (b) Figure C1.2. Dispersion curve from an intact pavement (a) and dispersion curve from a pavement with delamination at a depth of 5 in. (b). C4

Figure C1.3. SASW data analysis screen. 1.2 Impact Echo In the IE test, an impact source is used to transmit a high-frequency mechanical (sound) wave into the pavement and a receiver is utilized to measure the P-wave reverberation (resonant echo) between the top and bottom surfaces. The impact source and receiver are placed adjacent to each other on the pavement surface. The amplitude of the reverberation detected by the receiver is converted into the frequency domain as amplitude versus frequency. For a homogeneous pavement layer, there is a resonant or dominant frequency directly proportional to the thickness of the pavement layer. This resonant frequency is referred to as the thickness resonance. The frequency data are typically converted to thickness using the following equation with an assumed P-wave velocity that is modified by a beta (β) factor of 0.954 for a Poisson ratio 0.2: T = βV/2f (1) where T = the thickness, β = a beta factor, V = the P-wave velocity in the pavement, and f = the frequency. For a uniform pavement with no delamination, the calculated thickness resonance will be relatively uniform. However, when there is a shallower delamination, the reverberation will be C5

disrupted and both higher resonant impact echo and lower flexural frequency modes of vibration will occur. This change in frequency will lead to variation in calculated thickness values at delamination locations. For a series of IE tests conducted over an area, the calculated thickness can be plotted (see Figure C1.4), and areas where it changes (i.e., not expected in the pavement structure) are interpreted as delaminated. The change is indicated by a higher-amplitude, low- frequency response due to flexural vibrations for shallow delaminations, or a somewhat deeper but less than full pavement-thickness echo. Figure C1.4, Graph 3 on the left, shows thickness values (horizontal axis) plotted against distance (vertical axis) using the measured resonant frequency in Graph 2 and Equation 1. Note that the calculation requires an assumed velocity, V. Figure C1.4. IE testing and data analysis. (1)Sensor response (2)Measured amplitude & frequency De la m in at io n at 5 in ch es So un d 12 -in ch pa ve m en t (3)Computed Layer Thickness C6

2 Equipment Specifications Currently, most commercially available portable devices for conducting SASW and IE testing on pavements are point-test devices, and some of them can conduct both SASW and IE simultaneously. To improve the testing efficiency, prototype testing equipment with rolling wheels has been developed. The device can be pulled behind a vehicle or a push-cart. This equipment allows the SASW and IE tests to be performed continuously at a walking speed. This equipment has the potential to provide full lane-width continuous measurements for project-level investigations. Table C1.1 shows the specification for the continuous SASW/IE testing equipment with rolling wheels. The continuous testing equipment method can travel at a modest 1 mph using three testing units in an array to cover a half lane width, completing approximately 264 tests per minute over 528 ft2 (1 test per 2 ft2). The prototype equipment also demonstrated a test frequency of 1 test per 1 ft2. The slow rolling continuous test method does not provide for replicate testing at one location. Instead, the density of tests establishes the precision of the computed results. The commercially available point-test SASW/IE testing equipment can be implemented using the testing plan provided in this guide; however, testing will be more labor intensive. For purposes of this guide document, single-point-test devices with replicate measurements can, at best, cover 12 test locations per minute. To cover the same 528 ft2 for the same test density would take a minimum 22 minutes. A complete array of test results are required to develop a velocity map from SASW testing or a calculated thickness map from IE testing. The maps can be used to identify the potential delamination locations in the evaluated asphalt pavement. C7

Table C1.1. Equipment Specification for Continuous SASW and IE Testing System Component SASW Specification IE Specification System typea Array of multiple sets of impact sources and pairs of sensors. The sets are lined up transverse to the direction of travel. Array of multiple sets of impact sources and sensors. The sets are lined up transverse to the direction of travel. Sensor frequency responseb Up to 50,000 Hz Up to 50,000 Hz Impact source input frequencyb Up to 50,000 Hz Up to 50,000 Hz Lateral spacing between sensorsc 2 feet between center of sensor pairs (maximum) 2 feet between sensors (maximum) Lateral coverage per passd 6 ft (half lane width) 12 ft (full lane width) Longitudinal data collection ratec 1 test per foot (minimum) 1 test per foot (minimum) Travel speed during data collectione 1 to 2 mph 1 to 2 mph Travel speed during mobilization Posted speed limit Posted speed limit Real-time displayf Single sensor pair waveforms in time domain at reduced display rate Waveform and resonant frequency at each senor System monitoring and control Within or outside the survey vehicle Within or outside the survey vehicle Data collection rate Based on speed and sensor spacing on the sensor array Based on speed and sensor spacing on the sensor array Spatial reference Vehicle DMI, external distance wheel, or GPS Vehicle DMI, external distance wheel, or GPS a The primary difference between SASW and IE hardware is the configuration of the impact source and the number and location of receiver sensors. SASW typically has two sensors spaced away from the impact source. IE has one sensor spaced relatively close to the impact source. b The NDT system hardware should include variable frequency sources and sensors so the testing can be effectively performed under diverse climate and material conditions. c The spacing between the units in the array must consider the desired level of measurement density and signal interference from adjoining units. Signal interference must be avoided by controlling the lateral and longitudinal spacing between units and the test sequence. The proposed equipment specification, at the maximum unit spacing and longitudinal collection rate, would generate data for every 2 ft2. Higher measurement densities of every 1 ft2 can be achieved with available prototype equipment. d The lane width covered by each pass is a function of the number of testing units in the array. Full lane width during a single pass can be achieved with a sufficient number of measurement units. The purpose of the array is to collect equally spaced parallel lines of data simultaneously so that coherent areas of delamination can be identified and mapped. e The measurement is sent and received in milliseconds, but the entire process includes lowering and seating the source and sensor(s), initiating the source impact, collecting the sensor response, lifting the source and sensor(s) and storing the data. Speed of data collection is influenced by this sequence of tasks. f Real-time data display should be used to monitor the consistency of the measured data. It will provide a preliminary level of pavement uniformity, but more importantly it will show if the NDT system array is operating properly. C8

3 Proposed Data Output and Display Requirements The field operation and playback software for SASW should be capable of the following displays: • Direct sensor time-amplitude waveform; and • Surface wave velocity–wavelength dispersion curve. Examples of this display are shown in Figure C1.1, Figure C1.2, and Figure C1.3. The field operation and playback software for IE should be capable of these displays: • Direct sensor time-amplitude waveform; • Converted amplitude-frequency curves; and • Longitudinal thickness profile for a given transverse offset. An example of this display is shown in Figure C1.4. Output Format • SASW output should be a volume of data with velocity as a function of x (longitudinal distance), y (transverse offset), and z (depth). • IE output should be an area of data with thickness as a function of x (longitudinal distance) and y (transverse offset). 4 Equipment Calibration and Verification Equipment calibration is critical. Calibration is defined as comparing the response of the test component against a known standard. Bench calibration of each component should follow the manufacturer’s recommendations. It is suggested that calibration of the system, or a component, be conducted whenever a verification test shows a problem. Verification of all the components of the system is recommended prior to the start of testing. Verification is defined as observing the response of the test against a known sample. The user should have a procedure to perform verification where the equipment is normally stored and in the field just prior to testing. A pavement slab constructed in a parking lot with known materials to exact dimensions can provide repeatable verification tests. Testing should only proceed after conducting all the verification steps. The manufacturer should have recommended verification procedures. A distance measurement instrument (DMI) or global positioning system (GPS) unit is used to record the location of each measurement. It is common to use a DMI to trigger each NDT test. The DMI will either be attached to the wheel of the survey vehicle or to a separate wheel attached to the test equipment. The measurement device should be regularly calibrated as prescribed by the manufacturer and verified by the user prior to the start of testing. The DMI is calibrated by running the system over a pre-measured distance (usually ranging from 1,000 feet C9

to one mile, but may be shorter for smaller manual systems). The measured distance expressed by the DMI is compared to the known distance, and the DMI calibration factor is adjusted so that the two will agree. It is advisable to repeat the calibration after the calibration factor has been adjusted to confirm that the calibration has been carried out correctly. The measured distance should be within a foot (0.1% accuracy) of the pre-measured 1,000 feet distance after the repeat calibration run. Most commercial dashboard DMI units measure in increments less than 1 foot. DMI units for dedicated NDT sensor systems have much finer resolution (typically 0.1 ft). For longer surveys, over 1.0 mile, user marks should be entered into the data at ground control points (mile markers, intersections) for location referencing. 5 Climate and Pavement Conditions for Testing In general, the SASW and IE testing works better on stiff materials that have high moduli. Testing asphalt pavements at colder temperatures is preferable. SASW testing has been successfully conducted at asphalt surface temperatures of up to 100°F. IE generally requires comparatively cool asphalt for testing, as resonant echo amplitudes decrease with increasing asphalt mixture temperature. Both SASW and IE analysis rely, to some degree, on knowing material properties. IE uses assumed P-wave velocity to calculate thickness; SASW calculates modulus, but needs some seed values to perform the calculation. Better estimates of the material properties can be applied by initially testing a location of sound pavement with a known thickness. The user should recognize that asphalt material modulus is a temperature-depth- dependent gradient. Pavement surface temperature should be monitored and recorded at the time of testing. Testing should not be affected by the moisture present on the pavement surface. Frozen pavements with high moisture contents and frozen base/subgrade conditions should be avoided. Check with the manufacturer to establish if the equipment is weatherproof. To get good signal measurements, the SASW and IE techniques require good contact between the tip of each transducer and the pavement surface. Both SASW and IE test methods have been successfully conducted on older asphalt pavement surfaces with moderate raveling (but no significant loose material on the pavement surface). For cracked asphalt surfaces, the SASW system will generate faulty data when cracks intersect the energy wave passing from the impact point to the transducers. While a point-test system may be carefully placed between the severely failed areas on a pavement surface, it is impractical to adjust the rolling-wheel system, so it will generate irregular data. The rolling-wheel system will work best on dense, low-texture asphalt pavement surfaces. 6 Testing Modes and Required Settings The testing modes described in this section are prepared for a continuous testing equipment system. The discussion is also applicable to other SASW and IE equipment. The user should be familiar with the capabilities of the equipment they are operating. C10

The testing mode selected depends on the desired level of pavement condition detail. The permitted speed of data acquisition (travel speed of the equipment during testing) will establish the range of testing modes available. Currently, continuous SASW and IE test equipment is limited to speeds no greater than 5 mph. The slow testing speed limits the use of the equipment to project-level pavement condition evaluation and typically will require a lane closure. The lane-width coverage depends on the number of measurement units in the array and the desired level of evaluation detail. For IE testing, each single wheel test unit could be spaced at 1- to 2-ft increments. Six units spaced at 2-ft increments would provide full lane-width coverage with measurements at 1, 3, 5, 7, 9, and 11 ft transverse across the lane. If a higher test density is needed, the six units could be spaced every 1 ft to cover half the lane. The measurement units are staggered from wheel to wheel to minimize signal noise from adjacent units. For SASW testing, each test unit requires a set of two receiver sensors, plus the signal generator. The spacing between the receiver sensors should be adjusted for the thickness of the pavement. The optimum spacing depends on the thickness, condition, and stiffness of the pavement. Testing a thin pavement under cold, stiff conditions will obtain the best signal response with a short spacing. Testing a thicker pavement under warm conditions requires a longer spacing between the receiver sensors. The lateral distance between the receiver sensors is generally 6 to 12 in. Since shallower delamination in the pavement is often a greater concern, the spacing between the sensors is typically 6 in. The distance between the units is typically 2 ft. The SASW test measures the surface waves travelling across the pavement, so it is important that the testing is staggered. For project-level testing, multiple forms of spatial reference should be employed. In addition to the DMI or GPS system to log the test locations, the data collection software should permit the user to annotate roadway features into the data as the testing proceeds. Mile markers, intersections, bridges, and roadside traffic control signs are good physical references that should be in the data. The user should prepare a testing plan before starting field testing. The plan should include: • Proper assembly and verification of the test equipment; • Start and end of test section; • Condition of pavement surface; • Type of test to be used; • Frequency of measurement (i.e., distances between test points, number of replicates); • Test speed (i.e., distance covered during continuous testing) or the number of tests conducted per minute for point-test devices; • Type of output data that will be monitored during the test; and • Amount of time required to complete the field testing. This includes time to return to the beginning of the test for multiple passes when full lane-width testing is not possible. C11

7 Test Output Data Formats The goal of field data collection is to produce a three-dimensional volume of measurements. The x distance is defined as the longitudinal direction of vehicle travel in the lane. The y distance is the transverse direction, including the spacing between measurement units. The z values represent depth as determined from the measured waveform received by the receiver sensors. SASW tests generate a data file that ties the signal response waves of the impactor and two receivers to the x (longitudinal distance) and y (transverse offset) test location. Once the data is processed, the output data file converts the response waves into a surface wave velocity as a function of z (depth). The final output data file is a 3-D array of material response measured as wave velocity. Each x, y, z coordinate has a computed wave velocity. IE tests generate a data file that ties the signal response wave frequency measured by the receiver sensor to the x (longitudinal distance) and y (transverse offset) test location. Once the data is processed, the output data file converts the response wave frequency into a measured pavement thickness. The final output data file is a three-dimensional array of the contour of the bottom of the pavement. When the test encounters a debonded area, the measured pavement thickness reflects the depth of the delamination. The format of this data volume will vary with each equipment manufacturer. The equipment manufacturer’s software will combine these files into the 3-D volume. The wave data analysis may be generated directly during data collection and further refined as a part of post- processing. SASW field operation and playback software should be capable of the following displays: • Direct time domain waveforms from each of the two receivers; • Dispersion curve for each wheel pair; and • Waterfall plot of dispersion curves collected versus distance covered for each wheel pair. IE field operation and playback software should be capable of the following displays: • Direct time domain waveforms from each source-receiver pair and • Running amplitude/thickness plot, or equivalent B-scan, for each sensor wheel. 8 Test Output Data Quality Control Checks At the end of testing, the signal data collected from the transducers should be checked for quality according to the manufacturer’s data quality control check procedure. This quality control check should be performed while the equipment is at the pavement site and the traffic control is still in place. If problems are encountered or there are data of questionable quality, the data collection should be repeated. This process is partially automated but may require considerable user interaction. The quality control process should include: C12

• Correlating the field notes with the data files to ensure that all noted files actually exist; • Checking all of the recorded data files to confirm that their size is consistent with the amount of data collected; • Scrolling through each data file to ensure that the data looks reasonable and that there are no problems with the data: and • Checking the recorded pavement length of each file (recorded distance) and confirming that it agrees with the intended length. It is also a good practice to check for the roadway features recorded in the data file to make sure the distance measurement data recorded and lined up with the signal data correctly. After reviewing the NDT results and other pavement condition survey information, it is recommended that field cores be cut at the locations where potential delamination is identified to verify the delamination condition. The equipment manufacturer’s analysis software should have one or more features to identify data that would be viewed as outliers, based on expected data ranges. This is similar to scrolling through the data as a reasonableness test. This feature is particularly important for continuous tests systems that do not have replicate values and could encounter poor test conditions, like cracks and potholes. The software should highlight all suspect data and allow the engineer to determine the credibility of the values. 9 Data Analysis Each SASW data set collected during field measurement associates two receiver sensor surface wave traces with a single x-y pavement coordinate location. Differences in the time-history of each wave response are analyzed and converted into surface wave velocity versus wavelength, which is referred to as the dispersion curve. The wavelength component is related to pavement depth. The resulting surface wave velocity versus depth curves will be smooth when the test is measuring a sound pavement. See Figure C1.2. When the test encounters material delamination, there is a break and significant drop in the wave velocity. See Figure C1.3. Once the surface wave velocities versus depth are computed, the 3-D array of velocities can be studied for changes in the wave-velocity pattern. Visual analysis is used to help identify the location and depth of the delaminated area that is associated with a sharp drop in the SASW surface wave velocity. The processed surface wave velocity data is divided into increments of depth (depth slices), and the velocity at each depth is displayed as grayscale or color-coded 2-D maps. Figure C1.5 shows examples of depth slices at different depths. Higher surface wave velocity (dark shade) is an indicator of better pavement condition. Anomalies can be seen as light spots where the velocities are lower. The light-colored areas in the slice at 0.4 ft (4.8 in.) depth represent the constructed delaminated interface at a depth of approximately 5 in. The lower wave-velocity measurement will continue to reflect in the depth slices below the location of the delamination, even though the material below the delamination is sound. C13

The analysis of IE data is more direct. Each set of IE data measured in the field is a single receiver sensor waveform tied to a single x-y pavement coordinate location. The P-wave data is analyzed and the resonant frequency of the wave is determined. The resonant frequency is then converted into the computed thickness of the pavement. The final 2-D display is the computed thickness of the pavement based on the x-y locations of the measurements. Figure C1.6 is an example of an IE pavement-thickness display. With a general knowledge of the constructed pavement thickness, IE results that show a thinner (or thicker) pavement thickness are areas with delamination or other significant change in material properties. The IE test analysis does not require the examination of depth slices to locate delamination as is done with GPR and SASW technologies. 10 Test Reporting The speed of SASW and IE testing (not over 5 mph) limits the use of these NDT technologies to project-level analysis. As such, common reporting formats used for network-level summary of pavement distress are not applicable to these technologies. This section discusses an alternative method of reporting SASW and IE data for project-level engineering review. Using color-coded depth slices for visual identification of areas with potential delamination can be labor intensive. There are tabular reporting methods to identify potential delamination based on percentage (or count) of test locations with velocities in specified ranges for each depth-slice increment. The example in Table C1.2 shows the data divided into 0.1 lane- mile increments. In the example, Lane-Sections 35.3 and 35.4 show a significant increase in low- velocity measurements at the 0.5 to 0.75 ft depth. These sections would be highlighted and reviewed in more detail. Summary data at 0.1 lane-mile increments gives the engineer a quick method to identify pavement lengths with areas of interest (potential delamination). For typical pavement rehabilitation projects that are 5 to 10 miles long, the 0.1 mile summary data generates 50 to 100 sets of data to examine. Each 0.1 lane-mile increment should generate over 1,000 velocity tests at each depth slice. Simple summary statistics for over 1,000 data points may not identify areas of delamination. The summary data must identify differences in the data by focusing on specific surface wave-velocity ranges. C14

Figure C1.5. Depth slices of the SASW measured surface wave velocity at incremental pavement depths. Surface Wave Velocity (ft/sec) Depth = 0.1 – 0.2 ft Depth = 0.2 – 0.3 ft Depth = 0.3 – 0.4 ft Depth = 0.4 – 0.5 ft Depth = 0.5 – 0.6 ft Depth = 0.6 – 0.7 ft C15

Figure C1.6. Example of impact echo pavement-thickness plot. Table C1.2. Example Reporting of SASW-Processed Surface Wave Velocity Results Lane Section Depth = <0.25 ft Depth = 0.25 to 0.50 ft Depth = 0.50 to 0.75 ft (MP) (direction) Velocity greater than 4,500 fps 4,000 to 4,500 Less than 4,000 Velocity greater than 4,500 fps 4,000 to 4,500 Less than 4,000 Velocity greater than 4,500 fps 4000 to 4,500 Less than 4,000 35.1 EB 90 8 2 85 12 3 75 20 5 35.2 EB 92 7 1 86 11 3 77 18 5 35.3 EB 90 7 3 85 13 2 40 40 30 35.4 EB 92 7 1 55 35 10 10 30 60 35.5 EB 91 8 1 86 13 1 76 20 4 35.6 EB 90 7 3 86 11 3 75 19 6 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Thickness Color Scale (in) C16

APPENDIX C2 SASW/IE Vendor Features C17

Olson Engineering, Inc./ Pavement and Bridge Deck Scanner (PBDS) SASW-IE (prepared by Olson Engineering) 1. General overview of the system The system consists of a minimum of two transducer wheels and up to six transducer wheels. All transducer wheels are identical. An Olson Instruments computer based data acquisition system (Freedom Data PC) is used to control the transducer wheels. An external distance wheel attached to the moving mechanism is used to record of the location of the tests. The system can be mounted to either a vehicle hitch (if all three pairs of transducer wheels are used) as shown in Figure C2.1 or a scanning cart (if only a pair of transducer wheels are used) as shown in Figure C.2.2. The system performs well on dry and damp surface conditions. The system should not be used while raining or snowing. However, the system can be used immediately after the road is clear. Each set of transducer wheels of the BDS system can perform: a. Impact Echo (IE) tests on both wheels simultaneously in two scan lines (one scan per each wheel) by offsetting the transducer elements (by approximately 3 inches or 76.2 mm) and having the impactors from both wheels turned on. Note that the spacing between the two adjacent transducer wheels can be set between 6 inches (0.15 meter) and 2 feet (0.61 meter) depending on the scan resolution desired for the testing. b. Impact Echo and Spectral Analysis of Surface Waves (SASW) scanning simultaneously by aligning the transducer elements of both transducer wheels. The first transducer wheel (with the impactors on) can be used to perform the Impact Echo test. The second wheel (with the impactors off) can be used as the second transducer to acquire data for the SASW test analysis. Note that both Impact Echo and Spectral Analysis of Surface Waves tests can be performed simultaneously in a single scan. C18

2. PBDS Software a. Real-time Display and Processing The software acquires and displays the time domain waveform of each transducer wheel in real-time for both SASW and IE scanning. For the Impact Echo Scanning option, the software calculates the linear displacement frequency spectrum from the Fast Fourier Transform (FFT) of each time domain waveform, displays the displacement spectra, automatically identifies the dominant frequency and calculates the echo thickness associated with the dominant IE resonant frequency. All of these processes are performed real-time. For the spectral analysis of surface waves test data, the software calculates the phase information and dispersion curves (surface wave velocity versus wavelength) for each impact(s). Delamination of the asphalt can be automatically identified from the drop (decrease) of surface velocities in the dispersive curve at wavelengths corresponding to the asphalt lift depth(s). Figure C2.1. Pavement and bridge deck scanner (PBDS) towed behind a vehicle— 3 pairs of wheels at 2 feet (0.61 meter) apart. C19

Figure C2.2. PBDS system attached to a scanning cart on asphalt overlaid bridge deck—single pair of wheels at 6 inches apart for combined impact echo/surface waves scanning. b. Data Analysis The data analysis of the IE tests includes identifying the dominant frequency peak in the spectrum to determine the depth (thickness) of the asphalt (or delamination). The IE tests work best when the asphalt is colder (50°F or less) and therefore its modulus and surface hardness are more concrete-like so that the weaker compressional wave energy resonates (echoes) strong. For the SASW tests, the software automatically calculates the phase data and automatically masks or removes the glitches in the phase data. Based on the processed phase data, the software calculates the dispersive curve. Delamination of the asphalt can be automatically identified from the drop (decrease) of surface velocities in the dispersive curve at wavelengths corresponding to the asphalt lift depth(s). A screen shot of the processed/analyzed data from the NCAT test track over known delamination (debonding) of the asphalt lift is shown in Figure C2.3 below. The analysis is performed automatically by the software. C20

Figure C2.3. A screenshot of the processed/analyzed data. 3. Final Output Test results (thickness or delamination depth) from each scan line are put together. The final output is a graphical image with color or grayscale plots representing depths of delaminations. 4. Equipment Upgrades and Service More transducer pairs can be added on to the system (currently up to 3 pairs or 6 transducer wheels). 5. Future Developments It is planned to further develop the software so that upon completion of scanning a pavement or bridge deck, the data analysis is done and ready for review in a matter of 5 minutes or less. In addition, recent laboratory experiments indicate that the system may be capable of impacting the pavement/deck with the normal strong initial impact and then a second weaker impact at each test location. This will provide for the calculation of coherence (indicator of signal to noise ratio with a value of l.0 being excellent quality data and a value of 0 being very poor quality data) versus frequency (Hz) in SASW tests. The second impact may also improve the quality of the IE data by being able to compare single impacts versus the average of two impacts at each test location. A larger diameter distance wheel is also planned to more accurately measure distance. In addition, GPS may be incorporated to the Freedom Data PC for general test location purposes although it is expected the C21

distance wheel will be the most accurate measurement of distance. If desired, a video camera could also be incorporated into the PBDS system to provide operators with surface images of each scan. C22