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

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

24 Controlled Evaluation of Mechanical Wave Technologies: Portable Seismic Pavement Analyzer, Scanning Impact Echo, and Multiple Impact Surface Waves This chapter was prepared by Ray Brown and Haley Bell of the U.S. Army Corps of Engineers Engineering Research and Development Center (ERDC). Portable Seismic Pavement Analyzer Introduction The portable seismic pavement analyzer (PSPA), developed by Geomedia Research and Development, is a nondestruc- tive testing (NDT) device that measures Young’s modulus via ultrasonic surface waves, and completes the test within a few seconds. The PSPA is generally used to measure the in situ seismic modulus of pavements and to determine relative strength parameters for use in pavement evaluations. The device is operated from a laptop computer, which is connected to an electronics box by a cable that transmits power to the two receivers and the source. The source impacts the pave- ment surface and generates surface waves that are detected by the receivers. The measured signals are returned to the data- acquisition board in the computer. The velocity at which the surface waves propagate is determined, and the modulus is computed. For hot-mix asphalt (HMA) pavements, the PSPA reports the seismic modulus of the pavement temperature at the time of testing. An equation is used to standardize the measured modulus to a temperature of 77°F. Figure 4.1 shows a photograph of the PSPA. The PSPA was used to test the laboratory samples in the National Center for Asphalt Technology (NCAT) laboratory and to test the sections at the NCAT Pavement Test Track for Round 1 testing in October 2009. For Round 2 testing in Feb- ruary 2010, the PSPA was used only on the test track. Measure- ments during the Round 1 testing were made with a variety of hardware and software configurations, as shown in Table 4.1. Delamination of the laboratory slabs and the sections on the test track were detected by using modulus estimates. Analyzing modulus measurements alone appeared to be a fair to poor way of detecting HMA delamination. After the testing and data analysis of Round 1, the pavement structure condition of five of the 10 sections (Sections 1, 2, 3, 5, and 8) on the test track and the condition of the laboratory slabs were released to the vendor. Once this information was released, the vendor chose to reanalyze the data from Round 1 and complete Round 2 testing by using Configuration 1 (Table 4.2), which is the standard arrangement for thin HMA pavements, and some changes were made to the original software. All data presented in this report were tested and analyzed by using Configura- tion 1 and the updated software. Laboratory Testing Figure 4.2 shows the extreme waveforms for the four different conditions of the laboratory slabs. These results were deter- mined after the pavement structure of the laboratory slabs was known to the vendor. The four conditions included a shallow delamination, deep delamination, bonded, and delamination with reclaimed asphalt pavement (RAP) (simulating strip- ping). The red curve in Figure 4.2 is the source, the black curve is the near receiver (A2), and the green curve is the far receiver (A3). The waveforms on the left are the time domains, while the waveforms on the right are the frequency domains. Fig- ure 4.2 shows that the delaminated slabs have large amplitude receiver signals, with late low-frequency energy dissipations. The bonded slab has much lower amplitude receiver signals with a quick energy decay rate. The slab with RAP has wave- forms similar to the bonded slab. Further analysis was completed that identified the prob- ability of each test point to be bonded. Vector distances between the waveform characteristics of the bonded and delaminated slabs were defined. The vector distances were C h A P t e r 4

25 scaled as a probability measurement point being taken on a bonded slab. Since the conditions of the laboratory slabs were known, the area with the shallow delamination was used for calibration. This area was deemed to be the worst case and was scaled to give a 5% probability of being bonded. Average vector distances near zero were given a probability of 100% (bonded). Figure 4.3 shows a summary of the probabilities of each test point on the two laboratory slabs to be bonded. The red colors indicate a strong departure from energy being carried in surface waves (delamination), and the green colors indicate bonded areas. The top part of Figure 4.3 shows Slab 1 while the bottom part shows Slab 2. Columns 1 through 3 of Slab 1 are fully bonded, and Columns 5 through 7 are debonded at a depth of 2 in. Column 4 is most likely a transition area from the fully bonded section to the delaminated section. Columns 8 through 10 of Slab 2 are debonded at a depth of 4 in., while stripping at a depth of 4 in. was simulated with RAP in Col- umns 13 through 15. Column 11 is most likely a transition area from the delaminated section to the stripped section. There is much variability in the shallow delaminated section of Slab 1 compared to the RAP section in Slab 2. The vendor stated that this variability was due to the plywood on the concrete slab showing more variable scattering characteristics, which were damped out in the RAP section. Figure 4.1. PSPA testing device. Table 4.1. PSPA Hardware Configurations Originally Used for Round 1 Testing Configuration Sensor Spacing (in.) Frequency Range (kHz) Source and A2 A2 and A3 1 4 4 4 to 40 2 6 8 0.5 to 5 3 4 6 4 to 40 4 3 to 10 walk-away 6 4 to 40 Table 4.2. Quantitative Results of PSPA Measurements for Identifying Delamination Section Number of Correct Measurements Total Measurements Percentage Correct October 2009 February 2010 October 2009 February 2010 1 10 9 10 100 90 2 17 18 20 85 90 3 20 20 20 100 100 4 5 5 12 42 42 5 7 9 20 35 45 6 8 11 12 67 92 7 18 17 20 90 85 8 10 9 12 83 75 9 3 7 10 30 70 10 10 17 20 50 85 Total Sections 108 122 156 69 78 Known Sections 64 65 82 78 79 Unknown Sections 44 57 74 59 77

26 Note: delam = delamination. Figure 4.2. Time domain and frequency domain power spectra of laboratory slabs.

27 Field Testing The conditions of Sections 1, 2, 3, 5, and 8 were known to the vendor during the February testing and for the data analysis presented here. Figures 4.4 and 4.5 show the time and frequency domains of four conditions [i.e., shallow delamination, deep HMA delamination over portland cement concrete (PCC), bonded HMA over PCC, and bonded HMA] in the known areas on the test track for the October and February testing, respectively. The waveforms for each pavement condition at the warmer and cooler temperatures compare well with each other and with the slabs measured in the laboratory. The time domain waveforms of the delaminated areas have larger amplitudes with slower energy decay rates, while the bonded areas have smaller amplitudes with faster energy decay rates. The same analysis completed in the laboratory showing the probability of a test area to be bonded was completed for the test points on the test track. For the test track data analysis, the shallow delaminated area of the laboratory slab and Section 3 on the test track were used for calibration. The average vector distances of the shallow delaminated portion of the laboratory slab were used as the worst case and assigned a 5% probability of being bonded. The average vector distances of the bonded Section 3 were used as the best case and assigned a 100% chance of being bonded. All other test points were assigned probabilities based on those two representative conditions. Figure 4.6 shows the probabilities of the test points on the test track to be bonded based on the average vector distances. The probabilities are given for the data collected in October and February along with the pavement temperature at the time of testing. The vendor stated that Section 2 had a more abnor- mal and sensitive power distribution because the PCC layer may have provided additional reflected energy to confound the simple analysis method. Section 2 consisted of bonded HMA over PCC. Each test point on the test track was quantified to determine the accuracy of the PSPA for identifying delamination. For this analysis, any probability of 0.6 or more was considered bonded, while any probability of 0.5 or less was considered delaminated. Table 4.2 shows these quantitative results of the PSPA data presented in Figure 4.6. For each section, only the test point locations that were most likely to represent the bonding condition that it was designed to represent were selected to be evaluated. If there were a strong possibility of a location not being representative of the design condition due to being adjacent to another section, then that point was eliminated from the analysis. The analysis was completed for all test sections, including sections known by the vendor as well as those sections not known by the vendor. Table 4.2 shows that the PSPA was able to accurately detect approximately 69% and 78% of delamination in the warm and cool temperatures, respectively. The high percentages of the section known to the vendor (Sections 1, 2, 3, 5, and 8) were not surprising. For the sections not known to the vendor, the PSPA was able to identify approximately 59% and 77% of the delamination on the test track in October and February, respectively. The PSPA did a better job of identifying the delamination in the cooler pavement temperatures. Table 4.3 shows the probability of good bond on the basis of PSPA results compared to the actual percentage of bonded points. The probability of bonding determined from the PSPA test compares very well to the percentage of points actu- ally having good bond. For example, all the samples predicted with the PSPA test to have between 0.51 and 0.75 probability of being bonded actually had 67% of these points that were bonded. The actual percentage of bonded points compared well to the predicted probability for other ranges as well. The good comparison between the predicted probability of good bond and the actual percentage of bonded points occurred within sections known to the vendor as well as in sections in which bond conditions were not known by the vendor. This device measures individual points and does not allow for complete coverage at highway speeds. However, the potential for this test to locate delamination is good, and it is possible that a procedure can be developed later that allows for decreased test time and more complete coverage. Summary Overall, the PSPA was able to identify the bonded sections accurately. The nondestructive device had a difficult time Note: STA = station. STA 1 2 3 4 5 6 7 1.0 0.89 0.50 1.00 0.64 0.03 0.04 0.03 1.5 1.00 0.57 0.95 0.78 0.02 0.04 0.04 2.0 0.89 0.62 1.00 0.50 0.02 0.03 0.05 2.5 0.92 0.35 0.63 0.63 0.03 0.06 0.07 3.0 0.60 0.69 0.80 0.46 0.04 0.04 0.06 STA 8 9 10 11 12 13 14 1.0 0.04 0.16 0.17 0.17 0.63 1.00 0.83 1.5 0.12 0.12 0.16 0.12 0.79 1.00 1.00 2.0 0.12 0.14 0.19 0.11 0.98 1.00 1.00 2.5 0.13 0.13 0.10 0.14 0.81 0.93 0.71 3.0 0.13 0.16 0.15 0.16 0.95 1.00 0.62 Figure 4.3. Probabilities of laboratory slabs to be bonded. (text continues on page 31)

28 Note: AC = asphalt concrete. Figure 4.4. Samples of time domain and frequency domain power spectra on test track in October 2009.

29 Note: AC = asphalt concrete. Figure 4.5. Samples of time domain and frequency domain power spectra on test track in February 2010.

30 Oct. 2009 17.5 22.5 27.5 32.5 37.5 42.5 47.5 52.5 57.5 62.5 67.5 72.5 77.5 82.5 87.5 92.5 97.5 102.5 107.5 112.5 1 0.1 0.2 0.2 0.2 0.1 0.4 0.9 0.6 0.5 1.0 1.0 0.8 1.0 0.8 0.9 0.6 0.5 0.4 2 0.2 0.3 0.2 0.1 0.1 0.9 0.8 0.9 0.9 1.0 1.0 0.8 1.0 0.7 1.0 0.9 0.9 0.9 1.0 0.8 3 0.2 0.2 0.2 0.3 0.5 1.0 1.0 0.7 1.0 0.8 1.0 1.0 0.9 1.0 1.0 1.0 0.4 0.8 0.2 1.0 4 0.4 0.2 0.1 0.1 0.1 0.8 1.0 0.9 0.9 0.9 1.0 0.9 0.7 1.0 0.7 0.9 1.0 0.9 0.8 0.9 Temp. (oF) 63 64 65 64 66 69 70 70 71 70 70 71 71 70 71 72 74 81 81 84 Feb. 2010 1 0.1 0.2 0.2 0.5 0.2 0.7 0.3 0.3 0.7 0.6 0.9 1.0 1.0 0.9 1.0 1.0 1.0 0.5 0.8 0.3 2 0.3 0.2 0.3 0.2 0.3 0.7 0.6 0.6 1.0 1.0 1.0 1.0 1.0 1.0 0.8 0.9 1.0 1.0 1.0 1.0 3 0.2 0.2 0.2 0.3 0.7 0.6 0.6 0.7 0.7 1.0 1.0 0.8 0.8 1.0 1.0 0.8 0.7 1.0 0.2 1.0 4 0.6 0.1 0.1 0.2 0.7 1.0 0.9 0.6 1.0 1.0 0.9 1.0 1.0 0.9 0.9 1.0 1.0 1.0 1.0 Temp. (oF) 41 42 41 41 40 42 44 47 49 51 52 52 53 53 54 55 56 57 59 59 Oct. 2009 117.5 122.5 127.5 132.5 137.5 142.5 147.5 152.5 157.5 162.5 167.5 172.5 177.5 182.5 187.5 192.5 197.5 202.5 207.5 212.5 1 0.0 0.6 0.3 0.9 0.8 0.8 0.3 0.8 0.7 0.6 0.8 0.7 0.8 0.7 0.8 0.5 0.5 0.6 0.5 0.3 2 0.1 0.3 0.8 0.8 0.7 0.9 0.6 1.0 0.8 0.8 0.7 0.9 0.5 0.7 0.5 0.6 0.6 0.7 0.7 0.2 3 1.0 0.4 0.1 1.0 0.3 0.8 0.6 0.8 0.8 0.8 0.9 1.0 1.0 0.7 0.9 0.8 0.7 0.7 0.8 0.4 4 0.7 1.0 0.7 0.7 0.6 0.2 0.1 0.8 1.0 0.9 0.6 0.7 0.9 0.6 0.7 0.6 0.7 0.7 0.4 0.6 Temp. (oF) 85 85 86 86 86 87 86 86 87 84 84 83 82 82 84 85 84 84 82 52 Feb. 2010 1 0.1 0.3 0.2 0.5 1.0 1.0 0.1 0.6 1.0 1.0 1.0 1.0 0.3 0.5 1.0 1.0 0.2 0.4 0.5 0.3 2 0.1 0.2 1.0 1.0 1.0 0.7 0.9 0.8 1.0 1.0 1.0 1.0 1.0 1.0 0.7 0.3 0.4 1.0 1.0 0.3 3 1.0 0.4 0.1 1.0 0.7 1.0 1.0 0.8 0.9 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 0.5 0.6 0.8 4 1.0 1.0 0.8 0.5 0.8 0.2 0.1 0.1 0.2 0.6 1.0 1.0 0.9 1.0 0.5 0.5 1.0 0.6 0.3 1.0 Temp. (oF) 58 53 51 51 52 53 53 51 51 51 51 51 51 51 51 51 51 52 53 53 Oct. 2009 217.5 222.5 227.5 232.5 237.5 242.5 247.5 252.5 257.5 262.5 1 0.3 0.7 0.6 0.5 0.9 0.5 0.3 0.4 0.8 0.2 2 0.8 0.9 1.0 0.6 0.9 0.2 0.3 0.6 0.6 0.3 3 1.0 0.7 0.7 0.5 0.6 0.7 0.6 0.5 1.0 0.6 4 0.6 0.6 0.5 0.6 0.8 0.7 0.8 0.7 0.5 0.4 Temp. (oF) 52 52 52 52 52 53 54 55 58 58 Feb. 2010 1 0.4 0.4 0.6 0.4 0.5 0.3 0.4 0.3 0.5 0.2 2 0.5 0.5 0.9 0.3 0.3 0.1 0.4 0.1 0.3 0.2 3 0.6 0.2 0.4 0.4 0.3 0.3 0.5 0.4 0.5 0.7 4 0.6 0.4 0.2 0.6 0.3 0.4 0.6 0.8 0.5 0.3 Temp. (oF) 53 54 54 55 57 58 58 58 57 58 Section 9 Section 10 Section 1 Section 2 Section 3 Section 4 Section 5 Section 6 Section 7 Section 8 Figure 4.6. Probabilities of test track test points to be bonded.

31 purpose of the October testing was to identify HMA delamina- tion at warmer pavement temperatures. Round 2 testing was conducted in March 2010 where the conditions of the labora- tory slabs and Sections 1, 2, 3, 5, and 8 on the test track were provided to the vendor after the analysis of Round 1 testing. The purpose of the March testing was to determine whether the vendors could do a better job in identifying delamination in cooler temperatures. The scanning IE device consists of two synchronized 1-in.-diameter transducer wheels that measure the IE and spectral analysis of surface wave (SASW) vibrations induced by the on-board automated impactor. The scanning IE records a measurement every 6 in. while being towed behind a truck and operating at speeds of 1 to 2 mph. For the labo- ratory testing, the device was rolled manually. The IE data were analyzed by determining the resonant frequency of the pavement. The MISW method collects data similar to the SASW but involves multichannel data-processing techniques to determine pavement modulus and thickness information. The amount of data collected with the MISW technique is therefore significantly greater than the amount collected with the SASW technology. An accelerometer is placed on the pavement surface, and several impactors are triggered at various distances from the accelerometer, mea- suring the surface wave responses. The GPR data were col- lected with 1-ft-wide scans by using a Geophysical Survey Systems, Inc. (GSSI), SIR-3000 computer and a 1,500-MHz antenna. The SIR test method is performed by using a com- puter, an instrumented 3-lb-impact hammer, and a 4.5-Hz geophone. This test is operated by impacting the HMA pavement and measures the impact force and the resultant vibration to determine the pavement structure’s relative stiffness. The SIR data were analyzed by calculating the average mobility, or frequency domain transfer function between the system input and output, between 200 and 800 Hz in the laboratory and 100 to 500 Hz on the test track. Round 1 Testing Summary After the testing and analysis of Round 1 data, the pavement structure condition of five of the 10 sections (Sections 1, 2, 3, 5, and 8) on the test track and the condition of the laboratory slabs were released to the vendor. Once this information was released, the vendor chose to reanalyze the laboratory and field data from Round 1. All four methods tested on the laboratory slabs seemed to perform well in identifying the bonded and delaminated areas. However, the scanning IE had a difficult time identifying the RAP section (stripping) in Slab 2. The GPR and MISW data showed that there was a difference between the bonded section and the RAP section. Table 4.4 shows a summary of the four test methods on the laboratory slabs and the in situ condition of the slabs. Table 4.3. Probability of Good Bond on the Basis of PSPA Testing Results All Sections Probability Number Bonded Number Unbonded % Bonded 0 to 0.25 0 22 0 0.26 to 0.50 9 28 24 0.51 to 0.75 16 8 67 0.76 to 1.00 56 17 77 Known Sections 0 to 0.25 0 13 0 0.26 to 0.50 4 9 31 0.51 to 0.75 13 2 87 0.76 to 1.00 30 11 73 Unknown Sections 0 to 0.25 0 9 0 0.26 to 0.50 5 19 21 0.51 to 0.75 3 6 33 0.76 to 1.00 26 6 81 detecting the 2-in.-deep delamination when the baghouse dust was used. The PSPA was able to detect the 5-in.-deep delaminated areas by using the baghouse dust fairly well, particularly with the cooler pavement temperatures. This test does not provide continuous measurement of the bond, but the test does a reasonably accurate job of identifying bond. With some improvements, the PSPA has potential to help state departments of transportation identify pavements with delamination issues. Scanning Impact echo and Multiple Impact Surface Waves Methods Introduction Olson Engineering, Inc., performed a suite of nondestruc- tive tests during the Round 1 HMA delamination detection testing in October 2009. The tests included scanning impact echo (IE), multiple impact of surface waves (MISW), ground- penetrating radar (GPR), and slab impulse response (SIR). Olson Engineering, Inc., also used a point-by-point IE testing device. The four test methods were conducted on the labora- tory slabs and on the test track in October 2009. The conditions of the laboratory slabs and the sections on the test track were unknown to the vendor during the October testing. The (continued from page 27)

32 The second analysis of the Round 1 test track data revealed that the SIR performed poorly in identifying delamination. The device showed potential but is not ready to be implemented for identifying delamination. The GPR also performed poorly in identifying delamination in the field; however, the GPR method did show promise for identifying layer thicknesses and material boundaries. The scanning IE method seemed to be good at identifying deeper delaminations (4 to 6 in.), while the MISW method showed potential for identifying delaminations at a variety of depths. Figure 4.7 shows the results of the scanning IE on the test track sections during the October 2009 testing. The results are displayed in thickness (in.) by using the color scale to the right in the figure. An IE compression wave velocity of 7,000 ft/s was used for all thickness calculations. The resonant frequency displayed in Figure 4.7 is directly related to the pavement structure’s thickness. The thicker areas shown in blue, purple, and green (delamination) rep- resent low-frequency flexural resonance of the delamina- tion. Orange and yellow colors are considered to be bonded areas. The red areas have high-frequency resonances, which indicate thin layers or thicker, deeper delaminations. The results show that the IE seemed to do a good job of detect- ing the 5-in.-deep delaminations simulated with paper and the bonded areas. The scanning IE did a fair-to-poor job of detecting the delaminations simulated with the baghouse Figure 4.7. Round 1 scanning IE test results on test track. Table 4.4. Laboratory Summary Results of Olson Engineering, Inc. Test Method Slab 1 Slab 2 Test Locations 1-1 to 3-3 Test Locations 5-1 to 7-3 Test Locations 8-1 to 10-3 Test Locations 12-1 to 14-3 Point-by-Point IE Bonded Delaminated Delaminated Bonded Scanning IE Bonded Delaminated Delaminated Bonded SIR Bonded Delaminated Partially delaminated Bonded MISW Bonded Delaminated at 2.76 in. Delaminated at 3.54 in. Slightly debonded layer at 1.6 in. GPR Bonded Layer boundary at 2 to 3.5 in. Layer boundary at 3 in. Layer boundary at 2 to 3 in. Actual Condition Bonded Delamination at 2 in. Delamination at 4 in. Stripping at 4 in.

33 Table 4.5. Round 1 MISW Results on Test Track Test Station (ft) Vs (ft/s) Delaminated Depth (in.) Predicted In Situ Condition Note Condition Note 1 2.5 5,413 4.3 Delaminated Small, low frequency peak; slow Vs convergence na Outside test section 2 2.5 5,741 3.5 Delaminated Huge, low frequency peak; poor Vs convergence na Outside test section 3 52.5 5,906 na Bonded Control section Bonded HMA over PCC Section 2 4 52.5 5,906 na Bonded Control section Bonded HMA over PCC Section 2 5 102.5 5,906 Mixed results Delaminated Multiple frequency peaks; inconsistent Vs Bonded Section 4 = Bonded 6 102.5 5,906 6.5 Delaminated Sound with debonded layer; extra frequency peak at high frequency Bonded Section 4 = Bonded 7 152.5 5,906 na Bonded Control Section Bonded Section 6 8 152.5 6,234 Mixed results No conclusion Frequency spectrum good; poor Vs convergence Bonded Section 6 9 202.5 6,070 1.2 Delaminated Huge, low frequency peak; poor Vs convergence Delaminated Section 8 = Delaminated at 5 in. with RAP 10 202.5 5,577 2 Delaminated Low frequency peak; poor Vs convergence; likely delamination Bonded Section 8 11 252.5 5,741 1.5 Delaminated Low frequency peak; poor Vs convergence; likely multiple layers and delamination Delaminated Section 10 = Delaminated at 5 in. with paper 12 252.5 6,070 2 Delaminated Frequency spectrum good; poor Vs convergence Delaminated Section 10 = Delaminated at 5 in. with paper 13 297.5 6,234 1.5 Delaminated Low frequency peak; poor Vs convergence; likely delamination na Outside test section 14 297.5 6,070 5.5 Delaminated Sound with debonded layer; extra frequency peak at high frequency na Outside test section Note: Vs = shear wave velocity. dust at 5 in. deep. The scanning IE was not able to identify the 2-in.-deep delaminations simulated with the baghouse dust, but the method did a fair job of identifying the sections with RAP at a depth of 2 in. Table 4.5 shows the results of the MISW on the test track during the October testing compared with the actual condi- tions on the test track. Tests 1, 2, 13, and 14 shown in that table were outside the test section, which was unknown to the vendor at the time of testing. The MISW accurately predicted bonding or delamination for six of the 10 tests conducted on the test track. The test method had the ability to determine the delamination depth; however, the measurements were not accurate. The findings of the second analysis resulted in the vendor using only the scanning IE and MISW for the Round 2 testing conducted in March 2010. Round 2 Testing Summary For Round 2 testing, the scanning IE and MISW were per- formed only on the test track. Figures 4.8 through 4.12 show detailed results of the scanning IE test in the form of reso- nant frequencies. The same thickness scale used in Figure 4.7 also applies to these results. The thicker areas shown in blue, purple, and green (delamination) represent low-frequency flexural resonance of the delamination. Orange and yellow colors are considered bonded areas. The red areas have high-

34 Section 4 Figure 4.9. Scanning IE thickness plots for Sections 3 and 4. Figure 4.10. Scanning IE thickness plots for Sections 5 and 6. Figure 4.11. Scanning IE thickness plots for Sections 7 and 8. Figure 4.8. Scanning IE thickness plots for Sections 1 and 2.

35 Figure 4.12. Scanning IE thickness plots for Sections 9 and 10. Table 4.6. Quantitative Results of Scanning IE for Identifying Delamination Section IE Total Evaluated Percentage CorrectMarch 2010 1 10 10 100 2 20 20 100 3 20 20 100 4 5 12 42 5 5 20 25 6 5 12 42 7 20 20 100 8 12 12 100 9 6 10 60 10 13 20 65 Total 116 156 74 Known Sections 67 82 82 Unknown Sections 49 74 66 frequency resonances, which indicate thin layers or thicker, deeper delaminations. Again, the conditions of Sections 1, 2, 3, 5, and 8 were known to the vendor. The results show that the scanning IE identified the bonded areas well and did a fair job of identifying the delaminations simulated with paper at a depth of 5 in. and the stripping simulated with RAP at a depth of 2 in. Overall, the scanning IE did a poor job of identifying the delaminations simulated with the baghouse dust at depths of 2 and 5 in. It was important to try to quantify the results of the scan- ning IE to compare with the PSPA results. The same analysis procedure used for the PSPA was used with the scanning IE shown in Figures 4.8 through 4.12. However, only the results from the March testing were quantified for the scanning IE. Table 4.6 presents the results of the quantitative analysis. On the basis of the results, it appears that the IE method did a fair job of identifying the areas of delamination. Of course, the con ditions of Sections 1, 2, 3, 5, and 8 were known before the data were analyzed. It was important to provide some data to the vendors so they could calibrate their equipment on the basis of known conditions. Most emphasis for delamination identifi cation was placed on the predictions for Sections 4, 6, 7, 9, and 10, the sections unknown to the vendor. For those sections, the prediction was correct approximately 66% of the time. The amount of testing with the MISW was minimized because of the test time required with this equipment. The MISW was performed at 12 locations on the test track. The results of the 12 tests along with the actual conditions are presented in Table 4.7. The MISW accurately identified

36 Table 4.7. Round 2 MISW Results on Test Track Test Station (ft) Vs (ft/s) Delaminated Depth (in.) Predicted In Situ Condition Analysis Note Condition Note 1 27.5 6,234 5.5 Delaminated Low frequency peak Delaminated Section 1 = Delamination with baghouse dust at 5 in. 2 27.5 5,906 6 Delaminated Low frequency peak Delaminated Section 1 = Delamination with paper at 5 in. 3 52.5 6,660 na Bonded Stiff supporting layer; surface wave velocity pulls up at lower frequency Bonded HMA over PCC Section 2 4 52.5 6,398 na Bonded Stiff supporting layer; sur- face wave velocity pulls up at lower frequency Bonded HMA over PCC Section 2 5 102.5 6,070 na Bonded Layer Low velocity layer near 5.5-in. depth Bonded Section 4 6 102.5 6,693 1.5 Delaminated Some low frequency energy Bonded Section 4 = Bonded but near delaminated area 7 152.5 6,726 6.5 Bonded Layer Bonded Section 6 8 152.5 6,070 na Bonded No data beyond 10,000 Hz Bonded Section 6 = Bonded but near partial stripping area 9 202.5 6,726 6 Bonded Layer Bonded Section 8 = Bonded but near partial stripping area 10 202.5 6,857 10 Bonded Bonded Section 8 11 252.5 6,759 10 Bonded Delaminated Section 10 = Delamination with paper at 5 in. 12 252.5 6,693 10.5 Delaminated Low frequency peak Delaminated Section 10 = Delamination with baghouse dust at 5 in. Note: na = not applicable. the bonded or delaminated areas of 10 of the 12 tests per- formed. However, of the 12 tests conducted, half of the condi- tions were known to the vendor. Also, no tests were conducted on the areas that had 2-in.-deep delaminations. Therefore, it was difficult to accurately judge the test method’s potential for identifying HMA delamination. Summary After Round 1 testing, the vendor elected to remove the SIR and GPR from the program, since they did not provide a suitable answer and there were several other vendors using various forms of the GPR. The IE and MISW continued to be evaluated in Round 2. Both methods showed some promise for measuring delamination, but the MISW was time-consuming and not practical for rapid testing. The MISW limitations resulted in only a few data points taken. It was difficult to judge the potential for the MISW to iden- tify HMA delamination accurately on the basis of the few data points collected. The scanning IE was quicker and easier to use, and it showed some potential for measuring delamination.