Test Methods for Characterizing Aggregate Shape, Texture, and Angularity (2007)

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.

Disclaimer “The proposed test methods are recommendations of the NCHRP Project 4-30A staff at Texas Transportation Institute. These methods have not been approved by NCHRP or by any AASHTO Committee or formally accepted for the AASHTO specifications.” Proposed Test Methods A-1 A P P E N D I X A

Proposed Standard Method of Test for Shape, Angularity, and Texture of Aggregate Particles Using the Aggregate Imaging System (AIMS) 1. Scope 1.1 This method quantifies three-dimensional shape, angularity, and texture of coarse aggregate particles as well as angularity of fine aggregate particles. Testing and analyses are accomplished using the integrated Aggregate Imaging System (AIMS). 1.2 Analysis of Coarse Aggregates (Method A)—This method uses aggregates that are retained on a 4.75-mm (No. 4) sieve. 1.3 Analysis of Fine Aggregates (Method B)—This method uses aggregates that pass through a 4.75-mm (No. 4) sieve. 1.4 Aggregates scanned using this process should be washed to remove clay, dust, and other foreign materials and separated into the appropriate sizes before being analyzed. 1.5 This standard does not purport to address all of the safety problems, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use. 2. Referenced Documents 2.1 ASTM Standards: D 75 Practice for Sampling Aggregates C 136 Test Method for Sieve Analysis of Fine and Coarse Aggregates C 702 Practice for Reducing Samples of Aggregate to Testing Size E 11 Specification for Wire-Cloth Sieves for Testing Purposes 3. Terminology 3.1 Definitions: 3.1.1 Shape—describes the overall 3-dimensional shape of aggregate particles, e.g., round, elliptical, flat. The AIMS software sorts the three dimensions based on length and calculates the sphericity index as shown in Equation (1): where dL is the longest dimension, dI is the intermediate dimension, and ds is the shortest dimension. A sphericity value of one indicates that a particle has equal dimensions. 3.1.2 Angularity—is related to the sharpness of the corners of 2-dimensional images of aggregate particles. The angularity is analyzed using the gradient method. This method quantifies the change in the gradient on a particle boundary. The gradi- ent method starts by calculating the inclination of gradient vectors on particle boundary points from the x-axis (horizontal axis in an image). The average Sphericity d .d d A-1s I L 2 = ( )3 A-2 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity

change in the inclination of the gradient vectors is taken as an indication of angu- larity as follows: where the subscript i denotes the ith point on the boundary of a particle, and N is the total number of points on the boundary. 3.1.3 Texture—describes the relative smoothness or roughness of aggregate particles surfaces. The wavelet method is used to quantify texture. The wavelet analysis gives the texture details in the horizontal, vertical, and diagonal directions in three separate images. The texture index is taken at a given decomposition level as the arithmetic mean of the squared values of the wavelet coefficients for all three directions. The texture index is expressed mathematically as follows: where n denotes the level of decomposition and i takes a value 1, 2 or 3, for the three directions of texture, and j is the wavelet coefficient index. 4. Summary of Methods 4.1. Method A—Analysis of coarse aggregates includes 3-dimensional shape, angularity, and texture. The analysis starts by placing 56 aggregates particles on the aggregate tray at the specified locations. A 0.25X objective lens and camera acquire images of coarse aggre- gate particles. The maximum field of view achieved in the coarse aggregate module is 52.8 × 70.4 mm. The camera and video microscope assembly move incrementally in the x direction at a specified interval, acquiring an image of one particle at each increment. Once the x-axis range is complete, the aggregate tray moves in the y-direction for a spec- ified distance, and the x-axis motion and image acquisition process is repeated. This process continues until all 56 aggregates are scanned. Two separate scans are conducted using backlighting and top lighting, respectively. Backlighting is used to acquire two- dimensional images for the analysis of angularity, while top lighting is used for acquir- ing images for surface texture analysis. These two types of scans are necessary for complete analysis of coarse aggregates shape. 4.2. Method B—Analysis of fine aggregate angularity. The 0.5X objective lens is used for acquiring images. The analysis starts by uniformly spreading a few grams of fine aggre- gate particles on the aggregate tray. Backlighting is used to acquire all images in this analysis. The camera and video microscope assembly move automatically over the aggregate tray until the entire area is scanned. In each x-y scan, the z-location of the camera is stipulated to meet specified resolution criteria. Aggregates that are not within the size range for which the scan is conducted are removed from the image. 5. Significance and Use 5.1. Shape, angularity, and surface texture of aggregates have been shown to directly affect the engineering properties of highway construction materials such as hot mix asphalt concrete, Portland cement concrete, and unbound aggregate layers. Most methods currently in use for measuring these properties of aggregate particles are indirect Texture Index N D x yn i j j N i A-3= ( )( ) ( ) == ∑1 3 2 11 , , 3 ∑ Angularity N i ii (Gradient Method) = − − + = 1 3 1 3θ θ 1 3N − ∑ ( )A-2 Proposed Test Methods A-3

measurements of the desired properties. This test method provides direct measurement of aggregate shape, angularity, and texture and thus provides consistent values that are comparatively more beneficial for use in software designed to predict performance of highway pavements and structures. 6. Apparatus 6.1. The AIMS is an integrated system composed of a camera, video microscope, aggregate tray, backlighting and top-lighting systems, and associated software. 7. Sampling 7.1. Obtain aggregate specimens in accordance with Practice D 75, and reduce the specimen to an adequate sample size in accordance with Practice C 702. 8. Preparation of Test Samples 8.1. Wash and oven dry the reduced sample at 110 ± 5°C (230 ± 9°F) to substantially con- stant mass. The coarse aggregate sample should contain at least 56 particles. The fine aggregate sample should be about 50 gm. 9. Procedure 9.1. Analysis of Coarse Aggregate Angularity, Texture, and Shape 9.1.1. The user must ensure that the objective lens used is 0.25X and that the micro- scope is placed in the coarse position on the dovetail slide. The objective lens can be replaced by removing the fiber-optic ring light by unscrewing the three screws on the ring. Then unscrew the ring light holder from the lower end of the microscope. The user will be required to install the lens (0.25X in this case), return back the ring holder, and fix the top lighting ring back. 9.1.2. Position the microscope on the dovetail slide by releasing the knob of the retaining pin on the left side and sliding the microscope assembly upward or downward until the “coarse” labels on the left-hand side of the two pieces line up. The user needs to ensure that the retaining pin is engaged to secure the microscope. Then tighten the thumbscrew on the right-hand side of the micro- scope assembly. 9.1.3. On the integrated computer desktop, double click on the “AIMS”icon. The pro- gram interface will display a window along with a real time image (Figure A-1-1). On the program interface window, there are several active buttons with labels that indicate the process they perform. 9.1.4. Start the analysis by clicking on the “Project Settings” button. The user must select a name for the project so the analysis results for the aggregate sample will be saved in a file name under the specified directory. This step will allow the user to specify type and size of aggregates to be analyzed. The user is required to click on the “Modify Parameters” button that is available in the “Analysis Parame- ters” window (Figure A-1-2). 9.1.5. At the “Project Parameters” window (Figure A-1-3), enter the drive and direc- tory path desired for the project. Then enter a project name for the aggregates to be analyzed. Then from the “Aggregate Range” drop-list, the user can select the type of aggregate to be evaluated. For Method A, the user must select “Coarse.” Then click “OK.” A-4 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity

Proposed Test Methods A-5 Figure A-1-1. Computer screen for setting up an AIMS test. Figure A-1-2. “Analysis Parameters” window for AIMS test setup.

9.1.6. Clicking the “OK” button on the “Project Parameters” window will display the “Coarse Aggregate Parameters” window (Figure A-1-4). From the “Analysis Type” drop-list, the user must select the type of analysis to be performed (i.e., Angularity, in this case), and click the “OK” button. A “Coarse Aggregate Pa- rameters” window will appear, and the user must select from the drop-list the aggregate size to be analyzed (Figure A-1-5) and click the “OK”button. The first program interface window will appear, showing the information previously entered for current project settings. 9.1.7. Turn on the light beneath the aggregate tray, and allow it to warm up for a min- imum of two minutes. 9.1.8. To calibrate the camera and microscope, click on the “Camera Setup” button. An “AIMS Camera Setup” window will appear, showing a real-time image (Figure A-1-6). Now, the user must focus the camera and microscope on the calibration point marked on the aggregate tray. This point will be used as a ref- erence point for the scan where (x, y, z) coordinates are set to (0, 0, and 0). The user must ensure that the target point is in the center of the image by moving the aggregate tray in x and y direction using the joystick on the controller box. This process is easier if the magnification is at the lowest level (M = 1.0). A mag- nification of 1.0 is achieved by rotating the dial on top of the controller box while the switch button on the front of the controller is at zoom position. The magnification (M-value) appears on the digital screen on the controller box; this value will change when rotating the dial. The minimum value is 1.0 and the maximum value is 16, where maximum magnification is achieved. 9.1.9. After centering the calibration point in the image window, the user must click on the “16X” button. Clicking this button will cause the microscope to zoom in and achieve maximum magnification. If the point is not clear or not viewable A-6 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Figure A-1-3. “Project Parameters” window for AIMS test setup.

Proposed Test Methods A-7 Figure A-1-4. “Coarse Aggregate Parameters” window for AIMS test setup. Figure A-1-5. “Coarse Aggregate Parameters” window for AIMS test setup.

in the image, move the switch at the front of the controller box to the “Focus” position. Then rotate the dial on top of the controller box to move the micro- scope up or down until the image becomes clear. If the calibration point does not appear in the image window, move the joystick in x and/or y direction until the calibration point appears in the center of the image (Figure A-1-8). Put the switch in the focus position, and use the dial to focus the image at the maximum magnification (M = 16). This approach is illustrated in Figures A-1-6, A-1-7, and A-1-8. 9.1.10. Once the calibration point is centered and well focused in the image, tap the “@” on the controller. This button will cause the microscope to perform auto- focusing and achieve the best image. Then, tap the “Zero” button on the con- troller box. Then tap “Home.” The “Zero” button will set the x, y, and z coordinates to 0, 0, and 0, respectively. The “Home” button will cause the cam- era and microscope to return to the start point after finishing the scan. Then, click the “Done”button on the “AIMS Camera Setup”window; this window will close, and the program interface window will appear again. 9.1.11. Image acquisition begins by clicking on the “Acquire Images”button on the com- puter screen. A new message window will appear giving the option for per- forming camera setup. If camera setup was not performed in the previous step, it can be done here; otherwise, select “No,” if already performed (Figure A-1-9). When omitting the camera setup option, a new message window appears with instructions (Figure A-1-10). 9.1.12. The term “camera origin,”on the screen, signifies camera setup may be performed at this time; however, that is normally performed in the previous step. If so, click cancel, and place aggregate particles on the tray at the indicated locations. Place- ment of aggregates can be performed at the beginning, but in that case, the user A-8 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Figure A-1-6. “AIMS Camera Setup” window.

Proposed Test Methods A-9 Figure A-1-7. Calibration point centered at an intermediate magnification. Figure A-1-8. Calibration point centered, focused, and at maximum magnification.

must ensure that the calibration mark is exposed so the camera setup can be performed. If calibration has been performed, one can place aggregates on every marking including the calibration mark. Placement of the aggregates begins by placing a translucent sheet (Mylar film) between the aggregate tray and the light- ing table, which has an alignment grid indicating the position for 56 particles A-10 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Figure A-1-9. Window providing second opportunity for camera setup. Figure A-1-10. Window providing options for AIMS test setup.

(Figure A-1-11). The Mylar sheet is prepared such that the spacing between the center of the particles is approximately 50 mm in the x-direction and 40 mm in the y-direction. To ensure that the aggregates are properly aligned, the two mark- ings on the right side of the glass aggregate tray should align with the corre- sponding markings on the Mylar grid sheet (Figure A-1-12). Remove grid sheet after all the particles are positioned. Figure A-1-13 shows the coarse aggregates properly positioned on the glass tray. 9.1.13. After all instructions have been followed, click “OK,” and AIMS will start scan- ning. Upon scanning all aggregate particles on the aggregate tray, the camera will return to the starting point. Figure A-1-14 shows an example of an image from the scanning process. Proposed Test Methods A-11 Figure A-1-11. Aggregate tray with Mylar grid sheet showing proper positions of aggregate particles. Figure A-1-12. Close-up view of Mylar sheet over light table. (Note: objects in photo appear misaligned due to parallax error. Look straight down on light table to achieve the optimum alignment.)

9.1.14. For analysis of coarse aggregate texture, the same steps are followed as in the angularity measurement, except in Step 9.1.6 for analysis type, select “Texture.” 9.1.15. Click on the “Acquire Images”button, and a message window appears, as shown in Figure A-1-15. Follow the instructions and turn off the bottom lighting, and turn on top lighting. 9.1.16. Once the “OK” button is pressed, the system starts scanning the aggregates and acquiring grayscale images for each particle. The system will automatically focus on the top of each aggregate particle and adjust the top lighting. The camera and A-12 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Don’t place aggregates here until camera is set up. Figure A-1-13. Coarse aggregates properly positioned on glass tray. Figure A-1-14. Example of a 2-dimensional image of an aggregate particle.

the microscope will return to the starting point when the scan is completed. Figure A-1-16 shows an example of the scanning process. The system records the vertical location of the microscope while it is in focus at the top of each particle. The difference between the location of the microscope while it is in focus at the top of a particle and its initial location when it is focused on the lighting table (see step 9.1.10) is recorded in a text file as the depth of the particle. 9.1.17 Once the images are collected, they are saved under the directory path specified in Step 9.1.5. Click on the “Process Images” button to process the images using the analysis software. In the new window that appears, specify the project name or the path of the directory in which the images are to be saved. If the analysis was conducted for different aggregate sizes under the same project name, the user has the option to run the analysis for one single size or for all sizes avail- able in that directory (Figure A-1-17). 9.1.18 Select “OK,” and a new window will appear showing that the analysis process is being performed (Figure A-1-18). As soon as the analysis is completed, the win- dow will close. 9.1.19. Click on the “Analyze Data”button to analyze and obtain the desired data.The new window displayed will allow the user to select the analysis for a single particle size or for all sizes in the directory (Figure A-1-19). Then another window appears (Figure A-1-20), allowing the user to select from a drop-list the directory that con- tains the processed images and the type of analysis desired (Figure A-1-21). Select the analysis type and the directory, and click on the “Analyze” button. The results will be plotted in cumulative distribution formats (Figures A-1-22 and A-1-23). This process can be repeated sequentially for different analysis types. 9.1.20. Results for each analysis type are saved in an Excel spreadsheet in a folder named “Analysis” that has been created in the directory where the images are saved. Proposed Test Methods A-13 Figure A-1-15. Message screen when entering the texture measurement mode.

A-14 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Figure A-1-17. AIMS permits processing of all images or only those of a given size. Figure A-1-16. An example of the texture scanning process.

Proposed Test Methods A-15 Figure A-1-18. Window showing that angularity and texture analyses are being performed. Figure A-1-19. Window for selecting one or all aggregate sizes for analysis.

9.2. The “AIMS Analysis Workbook” is another program that can be used to obtain statistics from the data analysis, provide percent of particles in each shape property category, and plot analysis results on Excel charts. The program is self-guided and very easy to use. 9.3. Method B—Fine Aggregate Angularity Analysis Procedure: Analysis of fine aggregates is similar to that for coarse aggregate, except the fine aggregates are uniformly spread on the aggregate table. 9.3.1 Analysis of fine aggregates starts by uniformly spreading a few grams of fine aggregate particles on the aggregate tray such that individual particles are not A-16 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Figure A-1-20. Window for selecting directory type of analysis desired. Figure A-1-21. Drop-list for selecting type of analysis.

touching each other. The 0.5X objective lens is used for acquiring images of fine aggregates. The maximum field of view achieved in the fine aggregate module is 26.4 mm × 35.2 mm. Backlighting is used to acquire all images in this analy- sis. The camera and video microscope assembly move incrementally in the x direction at a specified interval and acquire an image at each increment. Once the x-axis range is complete, the aggregate tray moves in the y-direction for a specified distance and the x-axis motion is repeated. This stepwise process Proposed Test Methods A-17 Figure A-1-22. Example of cumulative distribution for surface texture index. Figure A-1-23. Example of cumulative distribution for gradient angularity index.

continues until the entire area is scanned. In each x-y scan, the z-location of the camera is stipulated to meet some specified resolution criteria. Aggregates that are not within the size range for which the scan is conducted are consequently removed from the image automatically by the system. 9.3.2 Ensure the objective lens is 0.5X and the microscope is placed in the fine position on the dovetail slide.The objective lens is interchanged by removing the fiber-optic ring light and the ring light holder from the lower end of the microscope.Then the required objective lens can be installed on the microscope. The microscope can be easily positioned on the dovetail slide by releasing the knob of the retaining pin on the left side and sliding the microscope assembly upward or downward until the “Fine” labels on the left-hand side of the two pieces are aligned. Ensure that the retaining pin is engaged so the microscope cannot fall. Then tighten the thumb- screw on the right-hand side of the microscope assembly. 9.3.3 Specify the drive and directory path for the project. Enter a project name for the aggregates to be analyzed. From the “Project Parameters”drop-list, select “Fine,” then select “OK” (Figure A-1-24). 9.3.4 Selecting “OK” on “Project Parameters” will display the “Fine Aggregate Para- meters” window (Figure A-1-25). From the drop-list, select the desired aggre- gate size, then select “OK,” and the first program interface window will display showing the new entered information for the current project. 9.3.5 Turn on the bottom light and allow it to warm up for minimum of two minutes. 9.3.6 Adjust the camera settings following the same procedures used for coarse aggre- gates (Subsection 9.1.8). 9.3.7 Initiate image acquisition selecting “Acquire Images.” A new message window will appear giving the option to perform camera setup. If not accomplished in the previous step, camera setup can be performed here; otherwise, select “No.” When omitting the camera setup option, a new message window displays with instructions that must be followed (Figure A-1-26). A-18 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Figure A-1-24. Aggregate range drop-list for fine aggregate.

Proposed Test Methods A-19 Figure A-1-25. Drop-list for fine aggregate parameters. Figure A-1-26. Specific instructions for fine aggregate analysis. 9.3.8 Spread fine aggregate uniformly on aggregate tray (translucent Mylar align- ment grid is not used in this segment), and click the “OK” button. The AIMS system will scan the entire tray and return to the starting point. 9.3.9. Fine aggregate image processing is identical to that for coarse aggregates, except no texture images are acquired.

A-20 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Figure A-1-27. Gradient angularity screen for AIMS fine aggregate analysis. Figure A-1-28. Gradient angularity screen for showing analysis type drop-list. 9.3.10. Data analysis for fine aggregate is similar to that for coarse aggregate, except the number of analysis parameters (gradient angularity, radius angularity, and 2-D form) are fewer (Figures A-1-27 and A-1-28). 9.3.11. Results for each analysis type are automatically saved in an Excel spreadsheet in a folder named “Analysis” that is created in the directory selected in Step 9.3.3. 10. AIMS Analysis Workbook 10.1 The AIMS Analysis Workbook contains additional software that can be used to generate statistics from the analysis data, provide percent of particles in each shape category, and plot analysis results on Excel charts. The program is self-guided and very easy to use.

Proposed Standard Method of Test for Aggregate Particle Shape Using Multiple Ratio Analysis 1. Scope 1.1. Multiple Ratio Analysis (MRA) quantifies and categorizes the overall shape of coarse aggregate particles. Testing and analyses are typically performed using a digital flat and elongated (F&E) measuring device, however, testing and analyses can be performed using a proportional caliper with multiple posts. 1.2. This standard does not purport to address all of the safety problems, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use. 2. Referenced Documents 2.1. AASHTO Standards: T 2 Sampling of Aggregates T 248 Reducing Samples of Aggregate to Testing Size T 27 Sieve Analysis of Fine and Coarse Aggregates 2.2. ASTM Standards: D 75 Practice for Sampling Aggregates D 4791 Test Method for Flat or Elongated Particles in Coarse Aggregate C 136 Test Method for Sieve Analysis of Fine and Coarse Aggregates C 702 Practice for Reducing Samples of Aggregate to Testing Size E 11 Specification for Wire-Cloth Sieves for Testing Purposes 3. Terminology 3.1. Definitions: 3.1.1. Shape—describes the maximum and minimum dimensions of coarse aggregate particles. 3.1.2. Coarse Aggregate—aggregates that are retained on a 4.75-mm or No. 4 sieve. 3.1.3. Multiple Ratio Analysis—a method for quantifying and categorizing the overall shape of coarse aggregate particles into five different F&E ratios, i.e., <2:1, 2:1 to 3:1, 3:1 to 4:1, 4:1 to 5:1, >5:1. 4. Summary of Method 4.1. MRA is an improved method for categorizing the various particle shapes found in a sample of aggregate. The ability to define the various particle shapes found within the coarse aggregate particles will lead to improved hot mix asphalt mix design procedures for performance optimized combined gradations based around particle shapes. MRA analysis may be used in the aggregate production process to optimize crusher per- formance and evaluate product consistency. This method features a new digital flat and elongated measuring device that easily and accurately determines the various coarse aggregate particle shapes found in an aggregate sample. Proposed Test Methods A-21

5. Significance and Use 5.1. Shape of aggregate particles has been shown to directly affect the engineering proper- ties of highway construction materials such as hot mix asphalt concrete, Portland cement concrete, and unbound aggregate layers. Most methods currently in use for measuring these properties of aggregate particles are indirect measurements of the desired properties. This test method provides direct measurement of aggregate particle shape and thus provides consistent values that are beneficial for use in software designed to predict performance of highway pavements and structures. 5.2. MRA is typically performed on an aggregate sample from a given stockpile. MRA gives an accurate representation of the particle shapes within an aggregate sample by evaluating the sample based on five different F&E ratios (<2:1, 2:1 to 3:1, 3:1 to 4:1, 4:1 to 5:1, >5:1) instead of one (as with ASTM D 4791). The caliper device measures the different F&E ratios found within a sample at the same time. Particles can easily be sorted into the various ratios with- out having to change the pivot point and re-measure the particles for each separate ratio. With the single ratio caliper device,a sample would need to be measured five separate times. 5.3. Important information can be obtained by examining the MRA of the various size frac- tions in the aggregate blend for the job mix formula of a paving mix. An example show- ing the percentages of various sizes of materials in each of five flat and elongated categories is shown in Figure A-2-1. 5.4. This procedure is useful for evaluating aggregates used in paving mixtures and for evaluating aggregates from different phases of the aggregate production process. 6. Apparatus 6.1. The MRA apparatus (Figures A-2-1 and A-2-2) is an integrated system composed of a device for measuring minimum and maximum aggregate dimensions and a computer with associated software to store data, categorize particle shapes, and produce tables and graphs to illustrate the findings. 7. Sampling 7.1. Obtain aggregate specimens in accordance with Practice D 75, and reduce the specimen to an adequate sample size in accordance with Practice C 702. 8. Preparation of Test Samples 8.1. A suitable reduced coarse aggregate sample should contain at least 100 particles. 8.2. Normally, coarse aggregate samples are suitable for testing without any processing. If aggregates are dirty, wash and air dry or oven dry the reduced sample at 110 ± 5°C (230 ± 9°F) to substantially constant mass. 9. Procedure 9.1. Validate the accuracy of the MRA caliper by sequentially placing three flat-faced steel blocks of known lengths (from about 2.54 mm to 25.4 mm or 0.1 to 2 inches) in the caliper, and ensure the measurements match the known lengths of the cylinders within ±0.0254 mm (±0.001 inch). (Note: These quantities and tolerances are merely a sugges- tion for this draft standard and are based on no analyses). If the readings are not accurate within the specified limits, the device must be calibrated following the manufacturer’s instructions. Validate caliper at the beginning of each day of testing. A-22 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity

Proposed Test Methods A-23 MultipleRatio Analysis Individual Size Fractions Grouped by F&E Ratios 0 10 20 30 40 50 60 70 80 90 100 % M ax im u m to M in im u m 3/8x4 1/2x3/8 3/4x1/2 1x3/4North Carolina #57 A 1:1 2:1 3:1 4:1 5:1 > Flat and Elongated Ratios 3:1 = 23.5% 5:1 = 4.1% (granite) Figure A-2-1. View of the original prototype multiple ratio analysis system. Figure A-2-2. View of a commercially-available multiple ratio analysis system.

9.2. The MRA system uses five different colors to represent five different flat and elongated ratios (<2:1, 2:1 to 3:1, 3:1 to 4:1, 4:1 to 5:1, >5:1). Therefore, prepare five empty bowls with the same color indicators to receive the aggregates as they are categorized by the MRA system. Place the bowls near the caliper. 9.3. Select a single aggregate particle from the sample and place it in the caliper with the maximum dimension in the vertical orientation. Slowly lower caliper head until it con- tacts the particle. Press the foot pedal to record the maximum dimension of the parti- cle on the computer. 9.4. Place the same particle under the caliper with the minimum dimension in the vertical orientation. Slowly lower caliper head until it contacts the particle. Press the foot pedal to prompt the computer to record the minimum dimension of the particle and calcu- late the flat and elongated ratio category of the aggregate particle. 9.5. When the color code appears on the screen, place the aggregate particle in the appro- priate color-coded bowl. 9.6. Select another aggregate particle and repeat Steps 9.3 through 9.5. Repeat these steps with all aggregate particles in the sample. 9.7. Weigh and determine the mass of the aggregate particles in each bowl. Sum the masses of the aggregate particles in all five bowls. Determine the percentage of aggregate par- ticles in each flat and elongated category by dividing the mass of particles in each bowl by the total mass of all particles. 9.8. Alternatively, count the number of aggregate particles in each bowl. Sum the number of aggregate particles in all five bowls. Determine the percentage of aggregate particles in each flat and elongated category by dividing the number of particles in each bowl by the total count of all particles. A-24 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity

Proposed Standard Test Method for Volume, Flat and Elongated Ratio, Angularity, and Surface Texture of Coarse Aggregate Particles Using the University of Illinois Aggregate Image Analyzer (UIAIA) 1. Scope 1.1. This method is intended for simple three-dimensional (3-D) reconstruction of indi- vidual coarse aggregate particles for volume, flat and elongated ratio, angularity, and surface texture of coarse aggregate particles. Testing and analyses are performed using the integrated University of Illinois Aggregate Image Analyzer (UIAIA). 1.2. Analysis of Coarse Aggregates—This method uses aggregates that are retained on a 4.75-mm (No. 4) sieve. 1.3. Coarse aggregates scanned using this process should be washed to remove clay, dust, and other foreign and deleterious materials and separated into the appropriate sizes before being analyzed. 1.4. This standard does not purport to address all of the safety problems, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use. 2. Referenced Documents 2.1. ASTM Standards: D 75 Practice for Sampling Aggregates C 136 Test Method for Sieve Analysis of Fine and Coarse Aggregates C 702 Practice for Reducing Samples of Aggregate to Testing Size E 11 Specification for Wire-Cloth Sieves for Testing Purposes 3. Terminology 3.1. Definitions: 3.1.1. Volume—with three orthogonally positioned digital cameras, the UIAIA aggre- gate image analysis system has the ability to perform volume computation for an aggregate particle. An estimate of its weight can then be determined using the known bulk specific gravity. The imaging based volume computation is achieved by combining the information in the three 2-D binary images as shown in Figure A-3-1. The 3-D space is meshed into a 3-D array of pixel cuboids or voxels. It is then simply required to count the number of voxels cor- responding to the particle contained in the rectangular box in Figure A-3-1. Any voxel belonging to the particle has the corresponding three projection pixels in the x-y, y-z, and z-x planes. The number of voxels that satisfies this condition finally gives the volume of the particle in units of pixel length cube. The volume computation program used in the UIAIA scans over the entire 3-D space and examines if each voxel belongs to the particle. 3.1.2. Flat and Elongated Ratio (F&E Ratio)—describes the overall 3-dimensional shape of aggregate particles. The UIAIA software sorts the evaluated aggregate particles into three categories: F&E Ratio <3:1, 3:1 < F&E Ratio < 5:1 and F&E Ratio > 5:1, based on F&E Ratio calculated using Equation 1: Proposed Test Methods A-25

where Longest Dimension is the longest dimension measured by UIAIA software from the three orthogonally positioned camera images, i.e., the top, side and front images. Shortest Perpendicular Dimension is the shortest dimension from the three images, which is perpendicular to the Longest Dimension as shown in Figure A-3-2. 3.1.3. Angularity Index (AI)—is related to the corner sharpness of 2-D images of aggre- gate particles. The angularity index (AI) is defined based on tracing the changes in slope of the particle image outline obtained from each of the top, side and front images. The outline of each image is extracted and approximated by an n-sided (n is taken as 24 in the AI definition) polygon as shown in Figure A-3-3. The frequency distribution of the changes in the vertex angles is established in 10-degree class intervals. The number of occurrences in a certain interval and the magnitude are then related to the angularity of the particle profile. Accord- ingly, the AI procedure first determines an angularity index value for each 2-D image as shown in Equation A-1-2: where e is the starting angle value for each 10-degree class interval and P(e) is the probability of each angle change in the range e to (e+10). Then, a final AI is established for the particle according to Equation A-1-3, by taking a weighted average of its angularity determined for all three views. The AI has the same degree unit as an angle does. AI Angularity Area Area i i i i i = × (= = ∑ ∑ ( ) 1 3 1 3 A-1-3) Angularity A e P e e , ( ) A-1-2= × ( ) = ∑ 0 170 F E Ratio Longest Dimension Shortest Perpendicu & = lar Dimension A-1-1( ) A-26 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity x yz Y m ax Xmax Z ma x Figure A-3-1. The smallest rectangular box encompassing a particle. shortest dimension, perpendicular to the longest dimension longest dimension Figure A-3-2. The longest and shortest dimensions of a coarse aggregate particle.

3.1.4. Surface Texture (ST)—describes the surface irregularities of aggregate particles. The surface texture of an aggregate particle is defined using an image analysis technique known as Erosion and Dilation. Erosion cycles followed by the same number of dilation cycles tend to smooth the surface of a particle by losing shape peaks and patching sharp dents on the boundary. The image area difference before and after erosion and dilation of the same number of cycles leads to the definition of the ST for one of the three particle projection images as shown in Equation A-1-4: where ST = Surface texture parameter for each 2-D image; A1 = Area (in pixels) of the 2-D projection of the particle in the image; A2 = Area (in pixels) of the particle after performing a sequence of “n” cycles of erosion followed by “n” cycles of dilation. Then, an ST index, denoted as STparticle, is established for the particle by taking a weighted average of each ST determined from all three views, which measures the overall surface irregularities of a particle. STparticle, is computed as according to Equation (A-1-5). The ST is a dimensionless quantity, as it measures the ratio of the areas before and after erosion and dilation. where i takes values from 1 to 3 for top, front, and side orthogonal views. STi is the surface texture parameter for each 2-D image, and Areai is the correspon- ding area of each 2-D image. 4. Summary of Method 4.1. Analysis of coarse aggregate particles includes determining volume, flat and elongated ratio, angularity, and surface texture. The UIAIA features a moving conveyor belt that carries the individual aggregate particle into the view of a sensor, which detects the particle and immediately triggers the cameras. Once triggered, the three synchronized ST ST Area Area particle i i i i i = ×( ) (= = ∑ ∑ 1 3 1 3 A-1-5) ST A A A = − × ( )1 2 1 100 A-1-4 Proposed Test Methods A-27 α α α αn = 24 n-1 n = 1 2 3 3 2 1 4 Figure A-3-3. An n-sided polygon approximating the outline of an aggregate particle.

cameras capture in one-tenth of a second the images of the front, top, and side views of the particle. The captured images are then processed for size and shape properties and indices using software developed specifically for this application. 5. Significance and Use 5.1. Volume, shape, angularity, and surface texture of coarse aggregates have been shown to directly affect the engineering properties of highway construction materials such as hot mix asphalt concrete, Portland cement concrete, and unbound aggregate layers. Most methods currently in use for measuring these properties of aggregate particles are indi- rect measurements of the desired properties. This test method provides objective and direct measurements of aggregate volume, shape (flatness and elongation), angularity, and texture to quantify these properties and provide repeatable results that are compar- atively more beneficial for use in performance prediction of highway pavements and structures. 6. Apparatus 6.1. The UIAIA is an integrated system with a fixture framework for mounting and posi- tioning the cameras, sensor, and other components. Three progressive scan CCD cam- eras are adopted to capture the images of moving particles, which are commonly used in motion control applications. The mechanical details of the UIAIA include a working conveyor belt operated using a variable speed AC motor, which provides smooth and steady operation at speeds as low as 3 inches/sec. Three fluorescent lights were used behind the cameras to provide adequate brightness. A black background was provided for all three views in order to provide a contrast and collect sharp images. 7. Sampling 7.1. Obtain aggregate specimens in accordance with Practice D 75, and reduce the specimen to an adequate sample size in accordance with Practice C 702. 8. Preparation of Test Samples 8.1. Wash and oven dry the reduced sample at 110 ± 5°C (230 ± 9°F) to substantially constant mass, typically 1 or 2 kilograms of aggregates depending on the average particle sizes. 9. Procedure 9.1. UIAIA System Setup—the UIAIA system is shown in Figure A-3-4 with an operator. Turn on the AC motor and warm up the system by keeping the AC motor running for ten minutes, so that the conveyor can move smoothly and steadily. Turn on the cameras and the sensor, and adjust the lens of the cameras until images of particles with sharp boundaries are obtained. Before image acquisition starts, take the images of a dummy calibration specimen to make sure the cameras and sensors work properly. 9.2. UIAIA System Calibration—calibration is a process by which measurements made in pixels from digitized images can be converted to equivalent engineering units through proportionality or equivalency factors. The calibration factors are determined from images of standard objects with known dimensions. Such calibration factors are of the form “X pixels = Y length units”. Once the calibration procedure is completed and calibration factors are established, the original configuration of the test setup including the camera focus, image resolution, light conditions, and so on, should not be altered. A-28 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity

Commercially manufactured white colored, precision spheres of diameter 0.5-inch, 0.625-inch, 0.75-inch, and 1-inch were used as standard specimens to establish calibra- tion factors in the UIAIA system. The relative sizes of the spheres are shown in Figure A-3-5. The sizes of the standard specimens chosen are representative of typical coarse aggregate particle sizes encountered in paving applications. Furthermore, the choice of a regular shaped object such as a sphere was made to expedite the calibration process by mak- ing it easier to detect and correct measurement irregularities between the different views. To establish calibration factors, images of the spheres need to be captured while the belt is moving. The diameters of the spheres are measured in pixel units from each of the three views. Calibration can then be accomplished by taking an average of the sphere diameter (in pixels) measured from the front, top, and side images for each trial and for each sphere size. The calibration factor for each size is obtained by comparing the diam- eters of the spheres in real dimensions in the form of “X pixels = Y length units”. To acquire images for the calibration process, please refer to Step 9.2 in this protocol. Image Acquisition—to start the image acquisition of the aggregate sample, first go to the software package of UIAIA, click on the LABVIEW file titled “triggered_capture” that has an extension of “.vi”. A window is opened as shown in Figure A-3-6. Operators can decide whether to display images, use trigger, or save gray scale and/or binary images during the image acquisition. Also, operators can specify the starting number of images and path of the saved images. The time delays for the three cameras indicate the time interval between the triggering of the sensor and the front camera, and the time interval between the front camera and the top camera and the side one. These three time delays have been calibrated, therefore should not be changed dramatically. Proposed Test Methods A-29 Figure A-3-4. Photo of the UIAIA system with operator passing aggregate particles. 0.5” 0.625” 0.75” 1.0” Figure A-3-5. Perfect spheres used for the calibration of UIAIA system.

The image acquisition can be started for a washed and oven dried aggregate sample by clicking on the arrow icon in the Tools bar of the user interface. An operator is needed to drop the individual particles one by one onto the moving conveyor belt (see Figure A-3-4). During the image acquisition process, captured images can be monitored by both the audio signal and the acquired three images shown on the computer screen. All images captured are automatically saved in a temporary folder in the computer. 9.3. Calculation of Coarse Aggregate Size and Shape Properties—the size and shape indices from the three-camera based aggregate particle reconstruction, i.e., the volume, gradation, flat and elongated ratio, angularity, and surface texture are computed using the algorithms or the virtual instruments (VIs) processing the acquired images for each sample. Each program to calculate each of these quantities is an individual VI file in UIAIA system. 9.3.1. Coarse Aggregate Volume The imaging based volume computation is achieved by combining the informa- tion in the three 2-D binary images as shown in Figure A-3-1. The 3-D space is meshed into a 3-D array of pixel cuboids or voxels. It is then simply required to count the number of voxels corresponding to the particle contained in the rectan- gular box in Figure A-3-1. Any voxel belonging to the particle has the correspon- ding three projection pixels in the x-y, y-z, and z-x planes. The number of voxels that satisfies this condition finally gives the volume of the particle in units of pixel A-30 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Figure A-3-6. UIAIA user interface (virtual instrument [VI]) for image acquisition.

Figure A-3-7. Screenshot showing operation of volume computation user interface or VI. length cube. The volume computation program used in the UIAIA scans over the entire 3-D space and examines if each voxel belongs to the particle. The volume computation algorithm can be viewed during operation as the three front, top, and side images of an aggregate particle are switched one by one on the screen. A screenshot of this process is shown in Figure A-3-7. 9.3.2. Coarse Aggregate F&E Ratio Go to the software package of UIAIA, click on the LABVIEW file titled “fe_sieve_maxinter” with the extension of .vi. The window as shown in Figure A-3-8 will be opened. Set up the parameters as shown in Figure A-3-8 by enter- ing the drive and directory path desired for the project, and specifying a proj- ect name for the aggregates to be analyzed. The F&E Ratios are computed for each particle in the aggregate sample by clicking on the arrow icon in the Tools bar of the user interface. The F&E Ratios of all the particles are automatically saved in an Excel file, feratio.xse in the Results folder under C:\, which needs to be established beforehand. Three other Excel files will also be generated that measure the sieve dimension, the maximum dimension and the minimum dimension of the individual particles respectively. The sieve dimension will be used to plot the gradation of the evaluated aggregate sample. The maximum dimension and the minimum dimension report the length and width of the individual particles. 9.3.3. Coarse Aggregate Angularity Go to the software package of UIAIA, click on the LABVIEW file titled “angular- ity” with the extension of .vi. A window as shown in Figure A-3-9 will be opened. Set up the parameters as shown in Figure A-3-9 and the angularity index (AI) is calculated for each particle in the aggregate sample by clicking on the arrow icon Proposed Test Methods A-31

A-32 Test Methods for Characterizing Aggregate Shape, Texture, and Angularity Figure A-3-8. Screenshot of user interface or VI for coarse aggregate F&E Ratio. Figure A-3-9. Screenshot of user interface or VI for coarse aggregate angularity.

in the Tools bar of the user interface. Angularity of all the particles will be auto- matically saved in an Excel file, ang.xse in the Results folder under C:\. 9.3.4. Coarse Aggregate Surface Texture Go to the software package of UIAIA, click on the LABVIEW file titled “surftex” with the extension of .vi. The window as shown in Figure A-3-10 will be opened. Set up the parameters as shown in Figure A-3-10; the surface texture (ST) index is computed for each particle in the aggregate sample by clicking on the arrow icon in the Tools bar of the user interface. Surface texture of all the particles will be automatically saved in an Excel file, surftex.xls in the Results folder under C:\. When the size and shape properties and indices, i.e., maximum, minimum, interme- diate dimensions, gradation curve, flat and elongated ratio, angularity, and surface texture, are calculated for all the individual aggregate particles in a sample, the aver- age flat and elongated ratio, angularity, and surface texture can be calculated to eval- uate the size and shape property of the aggregate sample. 10. UIAIA Analysis Workbook 10.1. The UIAIA Analysis Workbook contains additional software that can be used to plot the gradation of the evaluated aggregate sample. The program is self-guided and easy to use. Proposed Test Methods A-33 Figure A-3-10. Screenshot of user interface or VI for coarse aggregate surface texture.