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.
43 a p p e N D I x F accuracy The portable seismic property analyzer (PSPA) performs the ultrasonic surface wave and impact echo tests simultane- ously. On the basis of the results obtained in SHRP 2 Renewal Project R06A, Nondestructive Testing to Identify Concrete Bridge Deck Deterioration, the measurement spacing should be equal to or less than the smallest delaminated area to be detected by either the ultrasonic surface waves (USW) or the impact echo (IE) method. To map the area of the delami- nated area accurately, the measurement spacing should be half the desired smallest dimension of the area that is of practical value. Figure F.1 represents the USW and IE results of the tradi- tional PSPA, along with the approximate horizontal distribu- tion of the defects from SHRP 2 Project R06A. According to an objective criterion defined by Azari et al. (2012), the accu- racy of the USW and IE methods in detecting the defects was estimated at about 83% and 85% of the points tested, respec- tively. The detectability of the combined USW and IE results in locating the defects improved slightly, at 86%. The new PSPA results are similar to the traditional PSPA results. The amplitude and dominant frequency spectra are shown in Figure F.2. The defective areas are indicated by high amplitude and low frequency. The USW method is about 15% accurate in approximating the depth of defects. It becomes less effective when the delami- nation is deeper than 6 in. The IE method is more effective in locating deep delaminations. The accuracy of the IE method in estimating the depth of delamination is about 10%. precision Precision was evaluated through statistical analyses of the three sets of data from the three runs of the USW and IE methods on the centerline of the specimen. The USW methodâs repeatability results from SHRP 2 Renewal Project R06A are shown in Figure F.3a. The upper and lower bounds were calcu- lated for each test by adding/subtracting one standard deviation µ of the three runs to/from the mean modulus of three runs. The coefficient of variation (cov), which was used as a measure of repeatability, was obtained by calculating µ of the three runs divided by their corresponding mean value µ (cov = s/µ). As shown in Figure F.3b, the average cov was about 12%. The repeatability of IE test results for estimating the thickness of the slabs was also evaluated. The main points contributing to the higher standard deviation are the severely deteriorated pointsâwhere slight spatial variation may cause differences in the values. Nazarian et al. (2006) have shown that for new construction, the average cov is less than 7%. The thickness is calculated on basis of the dominant fre- quency and compression wave velocity of each slab. The aver- age cov of thickness was about 6%. These values correspond well with the anticipated uncertainty of 5% to 10% reported in the literature for the IE method. As recommended by a number of researchers (Nazarian et al. 2006), the evaluative power of the thickness estimation with the IE method can be improved through a calibration process using one or two cores. Calibration procedures After initial calibration by the manufacturer, a rigorous cali- bration is not necessary unless the sensors are replaced. testing procedures Testing procedures are documented at http://www.geomedia .us/. To collect data with the PSPA, the user initiates the testing sequence through the computer. The high-frequency source is activated four to six times. The outputs of the two transducers from the last three impacts are saved and averaged (stacked). The other (prerecording) impacts are used to adjust the gains of the preamplifiers. The gains are set to optimize the dynamic range. Ultrasonic Surface Waves and Impact Echo Testing Criteria
44 Cost The PSPA costs about $25,000. The speed of data collection can also be considered in the cost category because of the cost of traffic control and losses associated with traffic interrup- tions. Although the PSPA collects data point-by-point, the PSPA is a relatively rapid testing device. The data collection speed of the PSPA is about 30 s/point. Limitations Although the USW and IE methods are shown to be success- ful in detecting internal defects, some apparent disadvantages should be considered. Both are localized testing methods, and testing a long tunnel may take a lot of resources and time. Although the IE method does have the ability to show the existence of a defect, the depth of defects that are shallow or extensive can be difficult to quantify. Inadequate contact will result in inaccurate and false measurements, especially for very rough concrete surfaces and oily and curved surfaces such as tunnel linings, which occasionally cause the device to slip during testing. The new PSPA has resolved some of these issues. Data Management The PSPA saves the raw data from each test point with appropriate meta-data indicating the time and information about the test parameters. The collected data can be reana- lyzed readily with new algorithms if necessary. At the start of a project, the user identifies the location where the data will be stored. Data analysis and Interpretation The data analysis is defined as the processing of the raw data collected by the PSPA and includes preprocessing, data anal- ysis and presentation, and data interpretation. In the pre- processing phase of the IE method, using a time window to remove the surface wave energy from the time records pro- vides a more robust and accurate thickness measurement (a) Average apparent modulus obtained by USW method (b) Dominant frequency obtained by IE method 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 A B C D E F G Longitudinal Axis Tr an sv er se A xi s 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 R esonance Freq u en cy (Hz) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 A B C D E F G Longitudinal Axis Tr an sv er se A xi s 2000 2500 3000 3500 4000 4500 5000 5500 Apparent M oduli (ksi) Figure F.1. USW and IE contour maps.
45 Figure F.2. New PSPA defect maps on the bridge deck. (b) Planar contour map of dominant frequency (a) Planar contour map of amplitude of waveforms Figure F.3. Precision of the USW method. (a) Average, upper, and lower bound of modulus for each run (b) Coefficient of variation of modulus 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 0 10 20 30 40 50 60 70 80 C oe ffi ci en t o f V ar ia tio n Longitudinal Distance, ft
46 than when the entire waveform is used. In the USW method, the surface wave energy should be reinforced by implement- ing proper filters to minimize the reflection and body wave energy. The USW and IE results are graphically displayed in color contour maps, namely, traditional with unlimited-color index, traditional with two-color index, and checkerboard (Azari et al. 2012). The traditional contouring uses a smoothing algo- rithm to ensure that the displayed contour lines change gradu- ally and incrementally from a minimum value to a maximum value. When the unlimited-color index approach is selected, a large number of shades of primary colors are used in the smoothing algorithm. The two-color index contours contain only two colors delineated by a threshold value. However, a smoothing algorithm is still used to depict the results. The checkerboard algorithm plots a rectangular array of cells. The value for each cell is determined by smoothing the results using the values of that cell and the four adjacent cells to define a surface rectangle. Recent studies have shown that represent- ing the data in a checkerboard format enhances the evaluative power of the results (Azari et al. 2012). Interpreting the results requires defining the modulus and frequency threshold to delineate between the intact and delami- nated areas. In the USW results, the target modulus was set at 0.86 to ensure that the delaminated areas were selected with a confidence level of about 95% (Nazarian et al. 2006). The test points with a modulus less than 0.86 appear in red, indicating that they are defective. The threshold in IE contour maps was based on the thickness of the slab and the depth and extent of delamination. The test points with dominant frequency less than thickness frequency are marked as red (defective). references Azari, H., D. Yuan, S. Nazarian, and N. Gucunski. 2012. Sonic Methods to Detect Delamination in Concrete Bridge Decks: Impact of Testing Configuration and Data Analysis Approach. In Transportation Research Record: Journal of the Transportation Research Board, No. 2292, TRB, National Research Council, Washington, D.C., pp. 113â124. Nazarian, S., D. Yuan, K. Smith, F. Ansari, and C. Gonzalez. 2006. Acceptance Criteria of Airfield Concrete Pavement Using Seismic and Maturity Concepts. Report IPRF-01-G-002-02-2. Skokie, Ill: Innovative Pavement Research Foundation, Airport Concrete Pave- ment Technology Program.
