Home

Design and Testing of a Laboratory Ultrasonic Data

image

Contents

1. DNE M ar M come 26 FIGURE 3 6 ULTRASONIC SWITCHBOX CONTROLLER HARDWARE 1 2 14411 24 FIGURE 3 7 PXI 1006 CHASSIS WITH PXI 5102 DIGITAL OSCILLOSCOPES eene 28 FIGURE PALS DIGITAL OSCILLOSCOPE 29 99 9 922 2 _ _ _ ____ 30 FIGURE 3 10 HARDWARE FLOWCHART sccscoscscoscscscscescscescscesesccescscescscesescscscesescesescssescssescssescscesasees 32 FIGURE 3 11 LABVIEW PROGRAM 2 0 00 0000000000 00 33 3 5 FIGURE 4 1 SCALED IMAGE OF THE PLEXIGLAS PLATE WITH A HOLE PLATE 2 36 FIGURE 4 2 PLEXIGLAS PLATE TOMOGRAPHIC SETUP 36 FIGURE 4 3 SENSORS MOUNTED TO PLEXIGLAS A SOLID PLATE 1 AND B PLATE 2 WITH A HOLE USING _
2. 42 42 2 RIVe ERO 50 CHAPTER 5 SUMMARY AND CONCLUSIONS 2 cccccccscsccccccccccccccccccccccccccccsscccccsccsccssccsccecceees 63 IUE M MC DELE 66 _ lt 3 2 2 22 2222222 _ _ _ 70 M 71 2 BEREA SANDSTONE TOMOGRAMSG 0 ccccsccscoscoccaccecsecnscaccascecoscaccacceccscscanceccsccscaccancecosceecantececsscaecs 71 FIVE OAKS LIMESTONE TOMOGRAMS ccccscoscescescecescescsceccecescescestscescescesessescescessectssascescesescaseeseeses 73 APPENDIX loe 78 EL MAIN APPLICA TION MLM gt 79 D SETUP PROGRAM ve ON AR tox EIU eU me ON 81 B 3 SIMULTANEOUS ACQUISITION 000 0 00 88 B 4 PAIRED ACQUISITION PROGRAM 90 Bese ARRIVAL IME CORRELATION PROGRAM X
3. 15 244 APPLICATIONS OF TOMOGRAP 17 CHAPTER 3 ULTRASONIC ACQUISITION SYSTEM HARDWARE DESIGN 20 SEGEN _ 2 2 lt lt __ _ _ _ ___ _ _ _ 21 S SENSOR _____ lt 22__ ____ _____ A 22 22322232 2 2 22 lt 222 lt _ _ _ M 24 6 6 6 6 6 6 6 6 _ 25 3 5 DATA AXCOQUISITION INSTRUMBENTATION 28 IAB VIEW PROCESSING dente se caidas bed edi eee 31 3 7 LABVIEW PROGRAM FOR ULTRASONIC DATA ACQUISITION cccccsescsscccssccesccesccescceusccesccescesesensces 3 CHAPTER 4 ULTRASONIC DATA ACQUISITION SYSTEM CALIBRATION AND TESTING 22 99 9 2 2 0 _ __ _ 35 32222 233_ _____ _ _ __ 35 APPLICATION TESTING ooo Notre d dete eo OSES QUEUE 41 00 60
4. 92 OR eo uiv EN IM DH IM 95 vi LIST OF FIGURES CHAPTER T INTRODUCTION 1 CHAPTER 2 LITERATURE REVIEW sicesscsisccsscsasccsssssdsssssessscsvcactsnnsasssssensssscocvsesuasesscsucsessosvsesssossenssesse 4 FIGURE 2 1 ILLUSTRATION OF SNELL S LAW ACOUSTIC WAVE REFRACTION AND REFLECTION AT A INTERFACE WITH DIFFERENT ACOUSTIC VELOCITIES cccscceccecceccccscecceccsceccsceccecescescescscescesceseecescascuseess 7 FIGURE 2 2 SIGNAL ALIASING IN THE TIME DOMAIN 0 0000000 002000000 11 FIGURE 2 3 LABVIEW VIRTUAL INSTRUMENT FRONT PANEL ccccececcececcsceccsceccsceccsceceecececcscesescesesces 13 FIGURE 2 4 LABVIEW VIRTUAL INSTRUMENT BLOCK 00000000000000000000 14 CHAPTER 3 ULTRASONIC DATA ACQUISITION SYSTEM HARDWARE DESIGN 20 FIGURE 3 1 ULTRASONIC DATA ACQUISITION SYSTEM HARDWARE 2 2 2 21 FIGURE 3 2 BEESWAX COUPLANT 222220222020 ep pa ERR 22 FIGURE 3 3 PANAMETRICS 5077PR ULTRASONIC SQUARE 0 0000 24 FIGURE 3 4 ESG ULTRASONIC SWITCHBOX 5 0 00000000000000 26 DATAJACODISITION CAR D al
5. 6 9 0 9 2 6 V 37 FIGURE 44 PLATE TIME VS DISTANCE PLOT oranie ii RUE 37 FIGURE PEATE VS DISTANCE PLOT 38 FIGURE 4 6 VIRGINIA TECH ROCK MECHANICS LABORATORY ULTRASONIC VELOCITY VIEWER 38 FIGURE 4 7A TOMOGRAM OF THE SOLID PLEXIGLAS CALIBRATION PLATE 0 0400000 40 FIGURE 4 7 TOMOGRAM OF THE PLEXIGLAS CALIBRATION PLATE WITH 41 FIGURE 4 8 BEREA SANDSTONE TOMOGRAPHIC SETUP csscecoececcecececcecccecccscecescecescecesceceseecesescecesees 43 FIGURE 4 9 BEREA SANDSTONE TEST SAMPLE PRIOR AND B POST eene 44 FIGURE 4 10A BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 1 0 MPA 44 FIGURE 4 10B BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 2 17 24 MPA LOAD 45 FIGURE 4 10C BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 3 24 82 45 FIGURE 4 10 BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 4 34 47 MPA LOAD 45 FIGURE 4 10 BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 5 46 54 MPA LOAD 46 FIGURE 4 11A BEREA SANDSTONE TOMOGRAM AT INITIAL STATE 0 MPA een 48 FIGURE 4 11B BEREA SANDSTONE TOMOGRAM A
6. 5 0 383 0 707 0 924 0 924 0 707 SW NC 55 0 383 0 383 10 0 0 707 0 707 0 11 0 0 383 0 924 0 924 13 0 0 383 0 140 0 707 0 707 15 0 0 924 0 383 FIGURE 2 3 LABVIEW VIRTUAL INSTRUMENT FRONT PANEL The block diagram is where the front panel objects are wired together similar to an electrical schematic An example of a VI block diagram 15 shown in Figure 2 4 13 Wesley Johnson Literature Review Get SR vi Block Diagram File Edit Operate Tools Browse Window Help Draw lines From source locations to receiver locations Picture II P os 5 FIGURE 2 4 LABVIEW VIRTUAL INSTRUMENT BLOCK DIAGRAM The wires connect to executable icons called nodes The wires represent the flow of data from one function block to another 4 LabVIEW was created to interact with data acquisition cards and instrument controllers Using LabVIEW a custom measurement instrument can be made for a data acquisition system To acquire data a virtual measurement channel 15 created to represent the input for data acquisition Once a signal 15 digitized it can then be processed within LabVIEW using digital signal processing Instrument control can also be conducted within LabVIEW A custom instrument control panel can be created in a VI fora specific application 14 Wesley Johnson Literature Review 2 4 TOMOGRAPHY Tomography
7. 4 72 FIGURE 2 BEREA SANDSTONE TOMOGRAM AT LOAD 4 46 73 FIGURE A 3A FIVE OAKS LIMESTONE TOMOGRAM PRIOR TO LOADING 0 MPA UNITS FT SEC 73 FIGURE A 3B FIVE OAKS LIMESTONE TOMOGRAM 17 24 MPA UNITS FT SEC 74 FIGURE A 3C FIVE OAKS LIMESTONE TOMOGRAM AT 34 47 MPA 1 8 5 222222 74 FIGURE A 3D FIVE OAKS LIMESTONE TOMOGRAM AT 51 71 MPA UNITS 75 FIGURE A 3E FIVE OAKS LIMESTONE TOMOGRAM 68 95 MPA UNITS 75 FIGURE A 3F FIVE OAKS LIMESTONE TOMOGRAM AT 86 18 MPA UNITS 76 FIGURE A 3G FIVE OAKS LIMESTONE TOMOGRAM 103 42 MPA UNITS 76 FIGURE A 3H FIVE OAKS LIMESTONE TOMOGRAM POST FAILURE 0 MPA UNITS FT SEC TI PP BINDERS ____ _ 2 223223 lt 3 lt _ _3__ ___3_ ____ _ _ 78 FIGURE 8 1 UDAQ LABVIEW PROGRAM ARCHITECTURE 80 FIGURE 2 MAIN APPLICATION CONTROL FRONT 81 FIGURE B ORT SR PRON T PANE De inio e E e 82 FIGURE 8 4 SENSOR CALCULATION UTILITY FRONT PANEL 83 FIGURE 5 GET SR BLOCK DIAGRAM DRAW 85 FIGURE B 6 GET SR BLOCK DIAGRAM PICTURE 2 72 24
8. 72 Wesley Johnson Appendix A GeoTomCG Output Velocity 20000 16250 12500 6750 5000 FIGURE A 2E BEREA SANDSTONE TOMOGRAM AT LOAD 4 46 MPA A 3 FIVE OAKS LIMESTONE TOMOGRAMS slice 1 of 2 2 0 Velocity 0 5101 40000 0 4464_ 30500 I 3826 _ 0 3188 21000 0 2551 _ 0 1913 7 11500 0 1275 3 3 6e 002 Y oC 2000 0 o 1 bob I I d pP j I 0 1609 0 3216 FIGURE A 3A FIVE OAKS LIMESTONE TOMOGRAM PRIOR TO LOADING 0 MPA UNITS FT SEC 73 Wesley Johnson Appendix A GeoTomCG Output slice 1 of 2 2 D 1 1 1 1 1 1 1 1 1 Yelocity 0 5101 40000 0 4464_ 8 30500 0 3825 0 3188 7 21000 0 2551 0 1913 dj 11500 0 1275 3 3 be D 2 _ 2000 I to II 0 0 1609 0 3218 FIGURE A 3B FIVE OAKS LIMESTONE TOMOGRAM 17 24 MPA UNITS FT SEC slice 1 of 2 D Yelocity 0 5101 40000 0 4454 30500 0 3826 _ 0 3188 21000 I 2551 _ 0 1913 11500 0 12 5 002 _ Y oC 2000 0 o L f E l gd l Lm 0 1609 0 3218 FIGURE A 3C FIVE OAKS LIMESTONE TOMOGRAM 34 47 MPA UNITS FT SEC 74 Wesley Johnson Appendix A GeoTomCG Output slice 1 of 2 2 0 Velocity 0 5101 40000 0 4454 E 30500 0 3026 0 3188 a 21000 0 2551 _ 0 1913 11500 0 1275 1 3 be D _
9. Controlled source tomography for mining and engineering applications Seismic Tomography Theory and Practice London Chapman and Hall 1993 22 Semandeni Calder P N High frequency microseismic monitoring applied to the determination of stress levels in hard rock mines Rockbursts and Seismicity in Mines Fairhurst ed 1990 23 D robert X O Abraham GPR and seismic imaging in a gypsum quarry Journal of Applied Geophysics 45 pp 157 169 2000 24 Hanson D R T L Vandergrift M J DeMarco and K Hanna Advanced techniques in site characterization and mining hazard detection for the underground coal industry International Journal of Coal Geology 50 pp 275 301 2002 25 Friedel M J D F Scott M J Jackson T J Williams and S M Killen 3D tomographic imaging of mechanical conditions in a deep US gold mine Mechanics of Jointed and Faulted Rock Rossmanith ed pp 689 695 1995 67 Wesley Johnson References 26 Scott D F T J Williams D K Denton and M J Friedel Seismic tomography as a tool for measuring stress in mines Mining Engineering pp 77 80 January 1999 27 Itakura K K Sato and A Ogasawara Monitoring of AE Clustering Activity Prior to Main Faulting of Stressed Rock by Acoustic Tomography Technique Fifth Conference on Acoustic Emission Microseismic Activity in Geologic Structures and Materials Penn State University June 11 13 1991 Clausthal Zellerfeld Germany
10. Conversion DSP in LabVIEW Rock Sample Couplant Source Signal Received Signal FIGURE 3 1 ULTRASONIC DATA ACQUISITION SYSTEM HARDWARE FLOWCHART All of the devices within the system should be capable of being controlled by the user at one location This chapter outlines the selection and design process for the ultrasonic tomographic data acquisition system 3 1 SENSOR SELECTION The most sensitive components of the ultrasonic data acquisition system are the sensors Piezoelectric transducers PZT are often used for ultrasonic testing to translate vibration into an electrical signal Damping is used to tune PZT sensor to a specific resonant frequency where the sensor would be most sensitive and would be excited more by a wave that had a matching resonant frequency The advantage of a damped PZT sensor 15 that 1 can effectively filter out unwanted frequencies Undamped sensors can be excited by a wider range of frequencies The energy received by an undamped sensor is not as great as a damped sensor but the wider bandwidth allows for more data to be collected For this study undamped sensors were selected due to their wide bandwidth 21 Wesley Johnson Hardware Design The sensors are the point of contact between the acquisition system and the rock specimen Improper frequency matching of the sensor to the testing material will result in a reducti
11. into National Instruments 5102 Digital Oscilloscopes These analog to digital converters collect the analog signal at 20 MS second The digitized waveforms are then passed back into the personal computer for post processing Acquired LabVIEW programs were created to acquire and process ultrasonic sighWRve mnims the UDAQ system A flowchart of the structure of the control program 1s shown in Figure 3 11 32 Wesley Johnson Hardware Design a FIGURE 3 11 LABVIEW PROGRAM FLOWCHART The hierarchy of the program can be seen in Figure 3 11 When the program 15 first called a main application control menu is opened In this main panel there are four options The first 1s to set the source and receiver locations by calling a setup program The number of sources and receivers ultimately determines the pixel size of the final tomogram The pixel size is related to the number of raypaths going between source and receiver locations The optimum pixel size would be the same length as the wavelength of the ultrasonic waveform The program has a simple calculation to determine the number of sensors based on the estimated wavelength of the ultrasonic waveform The source receiver location program will then output the X Y Z location of each sensor in two tables one for receiver locations and one for source locations The saved location data
12. sec 1 E 05 0 00 Distance cm FIGURE 4 10B BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 2 17 24 MPA LOAD Sandstone 24 82 MPa 5 05 3 0589E 06x 1 7882E 07 4 05 R 0 91099 eo m N A e 1 05 0 00 Distance cm FIGURE 4 10C BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 3 24 82 MPA LOAD Sandstone 34 47 MPa 5 05 3 1227 06 7 9981 09 4 05 R 0 89986 3 E 05 2 E 05 Time sec 1 E 05 0 00 0 2 4 6 8 10 12 Distance cm FIGURE 4 10D BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 4 34 47 MPA LOAD 45 Wesley Johnson Calibration and Testing Sandstone 46 54 MPa 4 E 05 y 3 0105E 06x 9 7676E 10 3 C 05 R 0 87393 LU e 2E 05 E 1 05 0 00 Distance cm FIGURE 4 10E BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 5 46 54 MPA LOAD The sample failed at a load of 65 16 MPa 9450 psi The slope of the plots in Figure 4 10 15 important to note The slope tends to increases with increasing stress The scatter in the plot also increases with increasing stress The increasing slope indicates that the average velocity across the sample is decreasing as the stress increases Table 4 3 shows a summary of average velocity at different stress levels TABLE 4 3 SANDSTONE VELOCITY SUMMARY Stress Level Average Vel
13. signals The setup command uses the physical dimensions of a cylindrical sample to calculate the wavelength and the sensor locations around the sample The ultrasonic raypaths the straight line estimation of where the ultrasonic signal travels can be drawn between source and receiver locations showing the raypath coverage across the sample This drawing will show the user areas of low raypath coverage and by using different combinations of sources and receivers the raypath coverage can be optimized Once the sensor locations are determined the ultrasonic waveforms are collected simultaneously or by matching pairs of sources and receivers The acquired waveforms are processed by a correlation program which determines time of arrival for each waveform tomogram is made by inputting the sensor location information with the arrival times into 80 Wesley Johnson Appendix B LabVIEW User s Manual GeoTom A utility was created to format the data for direct input into GeoTom A screen capture of the front panel of the main control program is shown in Figure 2 main panel v1 File Edit Operate Tools Window Help Acquire Acquire FIGURE B 2 MAIN APPLICATION CONTROL FRONT PANEL B 2 SETUP PROGRAM The setup application for the ultrasonic data acquisition system determines the coordinates of the transmitting and receiving sensors around the sample A VI was developed to calculate these locations as
14. 00E 05 1 00E 05 0 00E 00 Distance cm FIGURE 4 13D FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 3 51 71 MPA 54 Wesley Johnson Calibration and Testing KLB 68 95 MPa 7 00E 05 0 000002x 0 000000 eee R 0 590846 5 00 05 4 00 05 3 00 05 2 00 05 1 00 05 0 00E 00 0 5 10 15 20 Distance cm FIGURE 4 13E FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 4 68 95 MPA KLB 86 18 MPa 7 00E 05 y 0 000002x 0 000000 6 00E 05 R 0 574178 5 00E 05 4 00E 05 Time sec 3 00E 05 2 00E 05 1 00E 05 0 00E 00 Distance cm FIGURE 4 13F FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 5 86 18 MPA 55 Wesley Johnson Calibration and Testing KLB 103 42 8 00E 05 0 000003x 0 000000 R 0 473871 7 00 05 6 00 05 5 00 05 4 00 05 Time sec 3 00E 05 2 00E 05 1 00E 05 0 00E 00 Distance cm FIGURE 4 13G FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 6 103 42 MPA KLB 0 Mpa Post Failure 7 00E 05 6 00E 05 y 0 000002x 0 000000 800609058 R 0 276317 4 00E 05 Time sec 3 00E 05 2 00E 05 1 00E 05 0 00E 00 Distance cm FIGURE 4 13H FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT POST FAILURE 0 MPA The trend to note in Figure 4 13 is that the scatter in the plots increases a
15. 1 54 500 54 500 Chassis MXI 3 PXI Chassis National Instruments 1 3 000 3 000 Controller 5077PR Ultrasonic Panametrics 1 2 000 2 000 Square Wave Pulser Ultrasonic Oo Engineering Seismology Pulser 1 2 000 2 000 Switchbox 80 Acoustic l un Physical Acoustics Corp 36 150 5 400 Emission Sensors This system 1s inexpensive when compared to other ultrasonic testing systems that are commercially available presently ESG Laboratory System 80 000 The cost specifications were obtained from vendor quotes at the time of ordering There are other ultrasonic tests that could be conducted using this system Preliminary testing with this system shows that it 15 capable of acquiring acoustic emission data from a rock sample under load The acoustic emission data could be used to construct tomograms using the acoustic emission event as the source for tomographic reconstruction Other ultrasonic experiments include and are not limited to void detection pore pressure imaging and three dimensional imaging 64 Wesley Johnson Conclusions The development of this system now provides a means for stress studies within rock and other materials that have acoustoelastic properties The experiment in sandstone shows the capability of this system for stress redistribution study By obtaining tomographic slices at different stress states the stress within the sample can be correlated to the increase or decr
16. 15 4 2002 66 Wesley Johnson References 14 Blaha P Selected Examples of Seismic Tomography Usage in Mining and Geotechnics Institute of Gephysics Polish Academy of Science M 24 340 pp 215 225 2002 15 Friedel M J M J Jackson E M Williams M S Olsen and E Westman Tomographic Imaging of Coal Pillar Conditions Observations and Implications International Journal of Rock Mechanics and Mining Sciences and Geomechanical Abstracts Vol 33 No 3 pp 279 290 1996 16 Cheng J L Li S Yu Y Song X Wen Assessing changes in the mechanical condition of rock masses using P wave computerized tomography International Journal of Rock Mechanics and Mining Sciences 38 pp 1065 1070 2001 17 Malinsky K Ultrasonic investigations in a coal mine Ultrasonics 34 pp 421 423 1996 18 Plocek J Equipment for ultrasonic investigation in mines Ultrasonics 34 pp 425 430 1996 19 Falls S D and R P Young Acoustic emission and ultrasonic velocity methods used to characterize the excavation disturbance associated with deep tunnels in hard rock Tectonophysics 289 pp 1 15 1998 20 Seebold I B Lehmann A Arribas C Ruiz T J Shepherd and K L Ashworth Development of tomographic systems for mining mineral exploration and environmental purposes Trasnacions of the Institution of Mining and Metallurgy Section Applied Earth Science 108 pp 105 118 1999 21 Goulty N R
17. 5102 Digital Oscilloscopes Figure 3 8 were chosen as the ADC for this system because of their high sample rate and relatively lower cost The specifications for this card can be seen Table 3 3 28 Wesley Johnson The cards have a vertical resolution of 8 bits which correlates to 256 quantization steps If the minimum vertical range of 50 mV is used then the smallest quantization step 15 0 4 mV The horizontal resolution of the cards is 20MS s which corresponds to 0 5 us The vertical and horizontal resolutions are acceptable to capture the arrival time of the NATIONAL INSTRUMENTS 5102 M2 Channel 20 5 1 Digital FIGURE 3 8 PXI 5102 DIGITAL OSCILLOSCOPE TABLE 3 3 PXI 5102 SPECIFICATIONS 5102 Specifications Bandwidth 15 MHz No of channels 2 simultaneously sampled Maximum 1 GS s repetitive 20 5 5 sample rate single shot Vertical ranges 50 mV to 5 V DC accuracy 2 5 of full scale at all gains Input coupling IDC AC Input impedance 1MO 1 in parallel with 25 pF 10 pF 29 Hardware Design Wesley Johnson Hardware Design waveform provided the waveform does not attenuate to a value less than 0 4 mV There exists a tradeoff between vertical and horizontal resolution for A D converters At higher acquisition speeds the vertical resolution 15 less than at lower acquisition speeds To improve the vertical resolution the vertical range can be reduced which r
18. Appendix B LabVIEW User s Manual Another event is used to handle saving the coordinate data the tables to a file This event is called when the Output Coordinates button is pressed on the front panel The event call is shown in Figure B 7 Select Receiver Coordinate File r open or create Y rec path Biz 2 Append R Coord elect Source Coordinate File source path rE SoCoord txt 123 8015 Append S Coord FIGURE B 7 GET SR BLOCK DIAGRAM OUTPUT EVENT When the event 15 called it takes the coordinates stored the tables and converts the data type to a double precision integer The numerical arrays for the source and receiver coordinates are then passed into a loop that erases any null data The data are then passed to a user specified file This file can be appended with more data points if the append buttons are pressed on the front panel When the Draw Raypaths button is pressed an event occurs that draws lines from the user specified source locations to the receiver locations This event is shown in Figure 8 8 87 Wesley Johnson Appendix B LabVIEW User s Manual 5 Draw Raypaths Value Change 7 Draw lines from source locations to receiver locations E FIGURE 8 GET SR BLOCK DIAGRAM DRAW RAYPATHS EVENT This event uses the location data to update the picture with raypaths between source and receiver locations
19. Because rock 1s not a uniform material differences in the internal velocity of the rock exist due to the presence of fissures discontinuities changing mineralogy and other geologic factors Ultrasonic tomography allows for a cross sectional image of ultrasonic velocities to be created Tomography is derived from the Greek word tomos meaning section Tomography uses the principals of the Radon Transform to invert ultrasonic data creating an image of the velocity within a cross section of rock Ultrasonic data 15 acquired transmitting an ultrasonic pulse through a rock and then measuring the velocity along the raypath of the ultrasonic wave The captured velocity data can then be processed using an inversion algorithm to produce a tomogram From the velocity tomogram an inference of the stress state within the rock can be inferred By comparing different velocity Wesley Johnson Chapter 1 Introduction tomograms obtained at different stress states patterns of stress movement can be identified and studied This imaging method has been used mining and tunneling operations to monitor and map the movement of stress within a rock mass By moving ultrasonic tomography into the laboratory stress redistribution studies can be conducted in a controlled environment through tomographic image collection and interpretation Tomography is an essential tool for stress redistribution studies in active mine workings The mapping of stress around und
20. Johnson Appendix B LabVIEW User s Manual APPENDIX B ULTRASONIC DATA ACQUISITION SYSTEM LABVIEW PROGRAM MANUAL FOR TOMOGRAPHIC DATA COLLECTION 78 Wesley Johnson Appendix B LabVIEW User s Manual USER S MANUAL FOR ULTRASONIC DATA ACQUISITION USING LABVIEW Laboratory Virtual Instrumentation Engineering Workbench or LabVIEW 15 a graphical programming language developed for testing and research engineers This intuitive programming language was developed by National Instruments to help engineers collect data from Data Acquisition DAQ equipment without having to learn programming syntax Within LabVIEW Virtual Instruments VIs are developed to control instruments read and process data and present data Within a VI there are two windows to be familiar with the front panel and the block diagram The front panel is the testing interface used when data is being collected analyzed or displayed The block diagram 15 the graphical user interface where the front panel items such as buttons graphs and user inputs are wired together to conduct an operation The block diagram is programmed graphically where function nodes are wired together by connecting terminals The wires pass data from one node to another so that the path of computation can be followed by tracing a wire similar to following current flow in an electrical schematic Groups of functions can be combined within a VI to form a subVI This subVI is an
21. OAKS LIMESTONE TOMOGRAM 103 42 MPA UNITS FT SEC 0 5 045 0 4 0 35 03 0 25 0 2 0 15 0 1 0 05 0 0 0 05 0 1 0 15 0 2 0 25 0 3 0 35 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM POST FAILURE 0 MPA UNITS FT SEC 40000 38000 36000 34000 32000 30000 28000 26000 24000 22000 20000 18000 16000 14000 12000 10000 8000 6000 61 Wesley Johnson Calibration and Testing The presence of the stress be seen in Figure 4 15 From the initial state Figure 4 15a a void or low velocity zone can be seen at the bottom of the tomogram In addition to this structure there also seems to be a thin low velocity zone running from the upper left to the lower right The low velocity zones observed in initial condition are known from pre and post failure inspections of the sample to be pre existing fractures within the sample The indentation load warm color at the top of the figure is clearly seen in Figures 4 15e 4 15f The location of the indentation does not appear to be in position shown in Figure 4 12 The reason for the discrepancy is from eccentric loading of the indentation platen The loading of one side of the platen more than the other results in the stress condition observed The uneven loading was due to non parallel load faces The location of the stress appears to be in the correct location however the large wavelength through the limestone reduces the precision in locating the str
22. Selector Labe _ _ _ ai Coords Horizontal E Coords Vertical Receiver Locations FIGURE B 6 GET SR BLOCK DIAGRAM PICTURE EVENT The picture event uses property nodes to collect information from the previous initialize event Property nodes contain values that are specific to the front panel object that they refer to A property node for an array would contain properties such as the values the array whether the control 15 visible or not visible the position of the control on the front panel etc The coordinates from the coordinate array calculated the initialize event are subtracted from the point at which the mouse button was depressed If this value is less than the radius of the sensor then data passes into the true case in a case structure if it is greater than the radius then the data passes to the false case If the case 15 true this indicates that a sensor was selected in the picture The function inside of the true case changes the color of the sensor from the default of blue to red to indicate the change from receiver to source or vise versa If the case is false then nothing is done After comparisons have been made of distance of the mouse relative to the radius of each sensor the coordinate array 15 compiled and updated This data is also sent to a subVI to fill in the receiver and source tables with the corresponding coordinates 86 Wesley Johnson
23. Trans Tech Publications pp 1 17 1995 28 Scott T E Ma J C Roegiers and 7 Reches Dynamic stress mapping utilizing ultrasonic tomography Rock Mechanics Nelson and Laubach eds Balkema Rotterdam pp 427 434 1994 29 Couvreur J F A Vervoort M S King E Lousberg and J F Thimus Successive cracking steps of a limestone highlighted by ultrasonic wave propagation Geophysical Prospecting 49 pp 71 78 2001 30 Falls S D R P Young S R Carlson and T Chow Ultrasonic tomography and acoustic emission in hydraulically fractured Lac du Bonnet grey granite Journal of Geophysical Research Vol 97 No B5 pp 6867 6884 1992 31 Villaescusa E M Seto and G Baird Stress measurements from oriented core International Journal of Rock Mechanics and Mining Sciences 39 pp 603 615 2002 32 Analog Devices Analog Digital Conversion Handbook N p Prentice Hall 1994 33 Goodman Introduction to Rock Mechanics 2nd ed New York Wiley 1989 34 Lama R D and V S Vutukuri Handbook on Mechanical Properties of Rocks Vol II Clausthal Germany Trans Tech Publications 1978 35 Karfakis M Lecture Notes Rock Mechanics 2002 36 UT Material Properties Tables 1996 NDT ED org 3 Nov 2004 lt http www ndt ed org gt 37 Bloomfield Philip E Wei Jung Lo and Peter A Lewin Experimental Study of the Acoustical Properties of Polymers Utilized to Construct PVDF Ultrasonic Transdu
24. a level that can be input into the PXI Chassis 8 The output of the voltage divider circuit 7 15 input into channel 0 of device in the PXI Chassis 8 This input is the trigger input for the UDAQ system The last step in wiring this system is to connect the acoustic emission sensors to the respective source relay terminal or acquisition terminal via a BNC Microdot connecting cable 98
25. average value of transmission frequency through the limestone was found to be 250 kHz These properties result in an ultrasonic wavelength of 2 66 cm This wavelength is longer than the radius of the uniaxial compression samples that were prepared for index testing The smallest structure that was expected to be resolved was 2 66 cm in length Knowing this restriction for tomographic reconstruction a limestone block was prepared for tomographic testing The dimensions of the block the placement and size of the load and the sensor locations are shown in Figure 4 12 51 Wesley Johnson Calibration and Testing Isometic View m Ys FEUI TS TIEVS UMEN LL PE Eire TC ET rr SLAP E n ep zi RE 4 Front wiew gt Source Location Receiver Location Ultrasonic Raypath Limestone Block Tomographic Setup Units Centimeters FIGURE 4 12 FIVE OAKS LIMESTONE TOMOGRAPHIC SETUP The indentation load was used to apply a concentrated stress state on the limestone sample This stress state would create a large stress contrast that should be able to be acquired by the UDAQ system Ultrasonic data were acquired at the points shown in Table 4 5 TABLE 4 5 LOAD CONDITIONS FOR TOMOGRAPHIC DATA COLLECTION Wesley Johnson Calibration and Testing At each load point 288 ultrasonic
26. derived from the Greek word tomos meaning section is an application of nondestructive testing to view the interior of a body without penetrating its surface by physical means In tomography radiation is either passed through a material or reflected inside a material along a straight line The radiation carries information about the physical properties in the transmitted medium 6 This process is repeated with different locations of transmitting and receiving radiation to achieve the desired resolution This method of imaging is used widely in the medical field using X Rays to create a cross sectional image The data 15 then inverted using a transform Tomographic inversion techniques are derived from the Radon Transform The transform is named after Radon 1917 who first derived an inverse transform The Radon Transform 1s described in depth by Nolet 5 The focus of this study is on geophysical applications of tomography This area of study was pioneered by Dines and Lytle in 1979 9 Geophysical tomography can utilize different radiation sources such as X Ray electrical resistivity reflection seismic and ultrasonic The radiated energy is passed through a medium and measured slightly different ways producing transforms that display different physical information in the cross section In X Ray tomography the cross sectional image is of the boundaries within a medium such as cracks in rock In electrical resistivity tomography the tomograp
27. embedded program within the parent VI The subVI can be called in different ways it can be called to open and ask for user input or it can run in the background SubVIs are useful for repetitive actions within a VI they also clean up the block diagram making it easier to read the block diagram This appendix is intended to be used as a guide to better understand the programming of the control applications developed for the ultrasonic data acquisition system The programs discussed are used for ultrasonic acquisition for the purpose of creating tomograms B 1 MAIN APPLICATION CONTROL The tomographic data acquisition system programs written in LabVIEW are used to acquire process and format data for tomographic inversion The tomographic 79 Wesley Johnson Appendix B LabVIEW User s Manual inversion is done using a third party program GeoTomCG The different applications for this system were tied together under one main control Each program can be run independently but are designed to be run together The program architecture is shown in Figure 1 Main Application _ Get Sensor Control Setup Locations Acquire Simultaneously Acquire 16 Acquire Pairs Process Correlate Arrival Format Geotom Input Times FIGURE B 1 UDAQ LABVIEW PROGRAM ARCHITECTURE The options within the main control are to setup acquire and to process ultrasonic
28. or the voxel size of a 3D tomogram based on the acoustic wavelength through the material The largest pixel size or voxel size depending on the number of dimensions selected 15 equal to the wavelength The front panel of this program 15 shown in Figure 8 4 X FIGURE B 4 SENSOR CALCULATION UTILITY FRONT PANEL 83 Wesley Johnson Appendix B LabVIEW User s Manual The output of the utility shows the number of pixels voxels and the number of sources and receivers needed to have that number of pixels voxels in the final tomogram The number of sources and receivers are always equal but could be changed to adjust the raypath coverage as long as the total number of sensors remains the same The setup VI uses events to control the flow of data through the program Events are handled LabVIEW through an event structure This structure is programmed to execute when a specific event occurs while the program is running For instance if a Boolean button is pressed on the front panel it will change the state of that button from false to true The change in the Boolean value can be programmed into an event structure so when that button 1s depressed an action such as the execution of a loop takes place the block diagram The VI 15 contained in a while loop that waits for events to happen this loop terminates and the VI will close when Continue is pressed on the front panel of the VI The button activates the
29. stop event in the block diagram and the value from the continue control button 15 passed to the stop function of the while loop The event call when the Initialize Drawing button 15 pressed 15 shown in Figure 8 2 94 Wesley Johnson Appendix B LabVIEW User s Manual gt Draw Value Change B Source Time CtiRef Oldval Picture Pi 73 Am 71 DrawAreaSize JEnE EE N FIGURE 5 GET SR BLOCK DIAGRAM DRAW EVENT This event takes the user input values for Z Level Number of Sensors R Sensor R Sample and Angle and uses them to draw a picture of the sample This VI will only create geometry and sensor locations for cylindrical samples The geometry is calculated inside a for loop Equation 1 shows how to calculate a point COS BEN MK Angle R Sample eq 1 NumberofSensors The calculated points are used as a center point to draw the sensors on the picture A cluster array Coord Array of the x and y coordinates and the color default 15 blue 15 output for each sensor by the for loop to be used in other events When a sensor 15 selected in the picture an event structure Figure B 6 determines the point where the mouse button was depressed and whether or not the user selected a sensor 85 Wesley Johnson Appendix B LabVIEW User s Manual 3 Picture Mouse Down L
30. waveforms were collected one waveform for each source and receiver pair From these waveforms the arrival time was correlated and checked To improve the picking accuracy of the program two correlations were done at each load and the arrival times that were not found both files were removed as outliers The arrival times were plotted against distance to observe the linearity of the picks These plots are shown in Figures 4 13a 4 13h KLB 0 MPa 4 00E 05 y 0 000002x 0 000000 3 00E 05 0 743193 e o 2 00 05 E 1 00 05 0 00 00 0 5 10 15 20 Distance FIGURE 4 13A FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT INITIAL STATE 0 MPA KLB 17 24 MPa 5 00E 05 4 50E 05 0 000002 0 000000 4 00 05 R 0 788785 3 50E 05 3 00E 05 2 50E 05 2 00E 05 1 50E 05 1 00E 05 5 00 06 0 00 00 Time sec 0 5 10 15 20 Distance cm FIGURE 4 13B FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 1 17 24 MPA 53 Wesley Johnson Calibration and Testing KLB 34 47 MPa 5 00 05 y 0 000002x 0 000000 R 0 846283 e 3 00E 05 4 00E 05 2 00E 05 Time sec 1 00E 05 0 00E 00 Distance cm FIGURE 4 13 FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 2 34 47 MPA KLB 51 71 MPa 7 00 05 6 00 05 y 0 000002 0 000000 R 0 572896 5 00 05 4 00 05 3 00 05 Time sec 2
31. well as to visualize the raypaths between ultrasonic sources and receivers as an estimate of the resolution within a tomogram The user inputs sample and sensor geometry and the VI draws a cross section of the sample If the number of sensors is not known a utility 1s included to calculate the number of sensors to use based on the wavelength within the material After entering the required input the user selects the source locations and can output the sensor coordinates view the raypaths or reinitialize the drawing to start over The front panel of this VI 15 shown in Figure B 3 81 Wesley Johnson Appendix B LabVIEW User s Manual S o o p L _ 9 00 0 924 0 000 10 0 0 707 0 707 0 000 0 924 0 000 11 0 0 383 12 0 0 000 1 000 0 000 Ma e FIGURE 3 GET SR FRONT PANEL This VI is called by its parent VI Main Panel vi To draw the sample and sensors the radius of the sample is input into R sample and radius of the sensor 15 input into R sensor The control Z Level refers to the vertical location of the plane that is being modeled The control Angle refers to a constant angle that should be added to offset the sensor array angularly Lastly the number of sensors is input into Number of Sensors Once these inputs are satisfied the model can be drawn by clicking the button Initialize Drawing This will draw a picture of the sample surrounded by the numbe
32. 00 0 05 9000 8000 7000 04 6000 5000 0 15 0 15 0 1 0 05 0 0 05 0 1 0 15 FIGURE 4 11B BEREA SANDSTONE TOMOGRAM AT LOAD 1 17 24 MPA 48 Wesley Johnson Calibration and Testing 0 15 20000 0 1 19000 18000 17000 0 05 16000 15000 14000 13000 12000 11000 10000 0 05 9000 8000 7000 0 4 6000 5000 0 15 0 15 0 1 0 05 0 0 05 0 1 0 15 FIGURE 4 11C BEREA SANDSTONE TOMOGRAM AT LOAD 2 24 82 MPA 0 15 20000 0 1 19000 18000 17000 0 05 16000 15000 14000 13000 12000 11000 10000 0 05 9000 8000 7000 0 4 6000 5000 0 15 0 15 0 1 0 05 0 0 05 0 1 0 15 FIGURE 4 1 10 BEREA SANDSTONE TOMOGRAM LOAD 3 34 MPA e 49 Wesley Johnson Calibration and Testing 015 11 arn n n 1h FIGURE 4 1 1E BEREA SANDSTONE TOMOGRAM AT LOAD 4 46 MPA 4 2 2 Five Oaks Limestone Experiment Limestone samples were obtained from the Kimballton Mine Ripplemeade VA which is located in the Five Oaks Limestone seam This rock was chosen for testing because it 1s a high quality limestone with few imperfections It was also readily available at the time of this study The index properties for this rock were not known prior to testing The index tests preformed on the limestone were done to classify the rock prior to ultrasonic tomographic imaging Limestone blocks were collected from pillars near the working section Blocks with rough dimensions of 37 x 2 x 1 were pried from the rib An arr
33. 2000 0 0 0 1609 0 3218 FIGURE A 3D FIVE OAKS LIMESTONE TOMOGRAM 51 71 MPA UNITS FT SEC slice 1 of 2 2 0 Yelocity 0 5101 40000 0 4464_ 30500 0 3826 0 3198 E 21000 I 2551 0 1913 11500 0 1275 3 3 be 002 Y oC 2000 0 I I I I I I I I D 0 1609 0 3218 FIGURE A 3E FIVE OAKS LIMESTONE TOMOGRAM 68 95 MPA UNITS FT SEC 75 Wesley Johnson Appendix A GeoTomCG Output slice 1 of 2 2 0 l Yelocity 0 5101 40000 0 4464_ 30500 1 3825 0 3188 7 21000 0 2551 _ 0 1913 7 11500 0 1275 3 3 b6e 002 _ Yo 2000 0 I I I I I I I I 0 1508 0 3218 FIGURE FIVE OAKS LIMESTONE TOMOGRAM 86 18 MPA UNITS FT SEC slice 1 of 2 2 0 Yelocity 0 5101 50000 0 4464_ 38100 0 3826 _ 0 3188 26200 0 2551 _ 0 1913 14300 0 1275 3 3 be D02 _ Y 2399 5 0 0 0 1609 0 3218 X FIGURE A 3G FIVE OAKS LIMESTONE TOMOGRAM AT 103 42 MPA UNITS FT SEC 76 Wesley Johnson Appendix A GeoTomCG Output slice 1 of 2 2 0 l 1 1 l Yelocity 0 5101 50000 0 4454 38301 0 3826 5_ 0 3188 26603 0 2551 0 1313 14904 0 1275 I1 47 6 9002 _ y 3205 4 0 0 0 1609 0 3218 A FIGURE A 3H FIVE OAKS LIMESTONE TOMOGRAM POST FAILURE 0 MPA UNITS FT SEC 77 Wesley
34. 4 4 1 0 48 00 000000000005 000000043 nnns 86 FIGURE B 7 GET SR BLOCK DIAGRAM OUTPUT 87 FIGURE B 8 GET SR BLOCK DIAGRAM DRAW RAYPATHS 88 FIGURE B 9 SIMULTANEOUS ACQUISITION 2 4040 00000 e een 89 FIGURE B 10 SIMULTANEOUS ACQUISITION BLOCK DIAGRAM 90 FIGURE B 11 PAIRED ACQUISITION FRONT 00 0 0 91 FIGURE FRONT PANE C 4 92 FIGURE B 13 GEOTOM FORMAT FRONT PANED 93 5 95 FIGURE ULTRASONIC DATA ACQUISITION SYSTEM HARDWARE WIRING FLOWCHART 96 1X LIST OF TABLES CHAPTER I INTRODUCTION 1 CHAPTER 2 LITERATURE REVIEW 4 CHAPTER 3 ULTRASONIC ACQUISITION SYSTEM HARDWARE DESIGN 20 TABLE 3 1 PHYSICAL ACOUSTICS MICRO 80 AE SENSOR PROPERTIES ccceccscsceccecceccscescecceccscacceseecs 23 TABLE 3 2 PANAMETRICS 5077PR ULTRASONIC SQUARE WAVE PULSER SPECIFICATIONS 25 125883 3 310250 ICI IC TIONS ett tui oat be une emere Pols ne Pos
35. 5 0 1 0 05 0 0 0 05 0 1 0 15 0 2 0 25 03 0 35 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM 34 47 MPA UNITS FT SEC 0 5 045 40000 38000 36000 34000 32000 0 35 30000 28000 gt 26000 24000 22000 0 25 20000 18000 02 16000 14000 12000 ae 10000 8000 0 1 6000 0 05 0 0 005 01 O15 02 025 03 0 35 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM 51 71 MPA UNITS FT SEC 40000 38000 36000 34000 32000 30000 28000 26000 24000 22000 20000 18000 16000 14000 12000 10000 8000 6000 59 Wesley Johnson Calibration and Testing 0 5 0 45 40000 38000 36000 34000 32000 0 35 30000 28000 T 26000 24000 22000 0 25 20000 18000 0 2 16000 14000 12000 10000 8000 0 1 6000 0 05 0 0 005 01 O15 02 025 03 0 35 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM 68 95 MPA UNITS FT SEC 0 5 045 40000 38000 36000 34000 32000 0 35 30000 28000 T 26000 24000 22000 0 25 20000 18000 0 2 16000 14000 12000 8 5 10000 8000 0 1 6000 0 05 0 0 005 01 O15 02 025 03 0 35 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM 86 18 MPA UNITS FT SEC 60 Wesley Johnson Calibration and Testing 0 5 0 45 40000 38000 P 36000 34000 32000 0 35 30000 28000 26000 24000 22000 0 25 20000 18000 02 16000 14000 12000 10000 8000 0 1 6000 0 05 0 0 005 01 015 02 025 03 035 FIGURE 4 15G FIVE
36. ATE 40 Wesley Johnson Calibration and Testing 12000 11200 11000 10500 10000 9500 9000 8500 9000 7500 TD 6500 6000 S500 5000 0 05 0 1 0 15 0 2 0 25 0 3 FIGURE 4 7B TOMOGRAM OF THE PLEXIGLAS CALIBRATION PLATE WITH HOLE The velocity scale shown in Figure 4 7 15 in feet per second The background velocity for both plates is the same at around 9000 ft sec 2743 m s The background velocity from this experiment matches the Plexiglas acoustic velocity value of 2700 m s 9058 ft sec reported by ndt ed org 36 The most significant finding in this experiment is the presence of the hole in the tomogram for Plate 2 Figure 4 7b The hole is not only clearly shown but it is accurately placed in the tomogram In addition to the presence of the hole artifacts are seen around the perimeter of the tomogram The cause of these artifacts is most likely due to the beeswax couplant The artifacts correlate closely to source and receiver locations By comparing the source locations Figure 4 2 to the red artifacts Figure 4 7 around the perimeter of the tomograms a close match is made The same trend can be seen when the receiver locations Figure 4 2 are compared to the blue artifacts Figure 4 7 4 2 APPLICATION TESTING Testing in rock was conducted to verify the application of the UDAQ system for stress redistribution tomography The application testing was conducted in Berea Sandstone and in Five Oaks Limestone An i
37. Design and Testing of a Laboratory Ultrasonic Data Acquisition System for Tomography By Wesley Byron Johnson Thesis submitted to the faculty of the Virginia Polytechnic Institute and State University in partial fulfillment of the requirements for the degree of Master of Science In Mining and Minerals Engineering Committee Members Dr Erik Westman Chair Dr Thomas Novak Dr Mario Karfakis December 2 2004 Blacksburg VA Keywords Tomography Ultrasonic Rock Mechanics Data Acquisition DESIGN TESTING OF A LABORATORY ULTRASONIC DATA ACQUISITION SYSTEM FOR TOMOGRAPHY Wesley Byron Johnson ABSTRACT Geophysical tomography allows for the measurement of stress induced density changes inside of a rock mass or sample by non invasive means Tomography is a non destructive testing method by which sensors are placed around a sample and energy 15 introduced into the sample at one sensor while the other sensors receive the energy This process 1s repeated around the sample to obtain the desired resolution The received information is converted by a mathematical transform to obtain a tomogram This tomogram shows a pixelated distribution of the density within the sample Each pixel represents an average value at that point The project discussed in this paper takes the principle of ultrasonic tomography and applies it to geomechanics A new instrumentation system was designed to allow rapid data collection through varyin
38. ER CIRCUIT B VOLTAGE DIVIDER CIRCUIT SCHEMATIC The equation used to calculate the resistances and the final voltage 15 shown Equation 3 1 30 Wesley Johnson Hardware Design 2 3 1 Where V Output voltage Ri 2 Resistance Vin Input voltage The input voltage value was 400 V originating from the Panametrics ultrasonic pulser The resistance for was calculated as 100 for R2 as resulting an output voltage to the digital oscilloscope of 3 96 V Using this signal as a trigger for the acquisition the arrival time was able to be referenced from the transmitted pulse of the pulser instead of a logic signal that might need to have timing corrections 3 6 LABVIEW PROCESSING After the waveforms had been digitized by the PXI 5102 A D cards digital signal processing DSP of the waveforms was done in LabVIEW The processing that needed to be conducted was the automated picking of the arrival times This was done by direct correlation To correlate the correct arrival time a LabVIEW program was written to compare a reference waveform to the actual waveform The reference waveform was compared to a piece of the total waveform and the correlation between the two was calculated At the point of highest correlation the arrival time was picked This method worked well for waveforms with a high signal to noise ratio SNR however all of the picks were double checked manually to ensure accurate picki
39. GURE B 10 SIMULTANEOUS ACQUISITION BLOCK DIAGRAM The acquisition call shows the master device initialization and setup in addition to the subsequent slave device calls The master device sets up the timing and triggering for the acquisition Once the program senses a trigger even it waits for the last slave device to get the trigger In the sequence structure the start acquisition call goes to each device in turn starting with the last device to receive the trigger The synchronization of the acquisition is handled by these programs automatically so no timing adjustments need to be made Once the waveforms have been acquired they are placed in to an array and saved to the file The program then activates the next source channel and continues acquisition until it runs out of source locations to activate B 4 PAIRED ACQUISITION PROGRAM There are circumstances where simultaneous acquisition might not be possible or not appropriate For these circumstances a program was made to acquire data based on source and receiver locations Each receiver is paired with a source turn and 90 Wesley Johnson Appendix B LabVIEW User s Manual ultrasonic waveform between the two sensors is acquired Figure B 11 shows the Front Panel of this VI File Edit Operate Tools Window Help pe ID X Coord Y Coord Z Coor P ee FIGURE B 11 PAIRED ACQUISITION FRONT PANEL The operation of this program is similar to the previ
40. HIC SETUP 36 Wesley Johnson Calibration and Testing To mount the sensors beeswax was used The wax was warmed and placed around the perimeter of the plate The sensors were then placed into the wax Figure 4 3 I amp i b FIGURE 4 3 SENSORS MOUNTED TO PLEXIGLAS A SOLID PLATE 1 AND B PLATE 2 WITH A HOLE USING BEESWAX After acquiring the all of the waveforms the arrival times were picked using the Correlation program in LabVIEW The arrival time was then plotted with distance to produce a linear plot of time vs distance Figures 4 4 and 4 5 Plate 1 Time vs Dist 5 E05 y 3 7415E 06x 6 2344E 07 R 9 9120E 01 3 3 E 05 2 E 05 1 E 05 0 00 2 6 8 10 12 Distance cm FIGURE 4 4 PLATE TIME VS DISTANCE PLOT 37 Wesley Johnson Calibration and Testing Plate 2 Time vs Dist 6 E 05 y 4 0664E 06x 1 4729E 11 9 5148E 01 5 05 4 05 3 E 05 2 E 05 Time sec 1 E 05 0 00 Distance cm FIGURE 4 5 PLATE 2 TIME VS DISTANCE PLOT The calibration of the system was done by comparing the average velocity of all raypaths through the solid plate to average velocity measurements obtained from the ultrasonic P wave measurement device from the Virginia Tech Rock Mechanics Laboratory and to an average of reported velocity measurements from the literature 36 39 The apparatus used
41. T LOAD 1 17 24 0 0400000 48 FIGURE 4 11C BEREA SANDSTONE TOMOGRAM AT LOAD 2 24 82 0 49 FIGURE 4 11D BEREA SANDSTONE TOMOGRAM AT 3 3 49 FIGURE 4 11 BEREA SANDSTONE TOMOGRAM AT LOAD 4 46 50 FIGURE 4 12 FIVE OAKS LIMESTONE TOMOGRAPHIC SETUP 0 52 FIGURE 4 13A FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT INITIAL STATE 0 MPA 53 FIGURE 4 13B FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 1 17 24 53 FIGURE 4 13 FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 2 34 47 54 FIGURE 4 13D FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 3 51 71 MPA 54 FIGURE 4 13E FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 4 68 95 55 FIGURE 4 13F FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 5 86 18 55 FIGURE 4 13G FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT LOAD 6 103 42 MPA 56 FIGURE 4 13H FIVE OAKS LIMESTONE TIME VS DISTANCE PLOT AT POST FAILURE 0 56 FIGURE 4 14 FIVE OAKS LIMESTONE TEST BLOCK A PRIOR AND B POST 57 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM PRIOR TO LOADING 0 MPA UNITS FT SEC 58 FIGURE 4 15B FIVE OAKS LIMESTONE TOMOGRAM 17 24 MPA UNITS F
42. T SEC 58 FIGURE 4 15C FIVE OAKS LIMESTONE TOMOGRAM AT 34 47 MPA UNITS FT SEC 59 FIGURE 4 15D FIVE OAKS LIMESTONE TOMOGRAM AT 51 71 MPA UNITS 59 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM 68 95 MPA UNITS 60 FIGURE 4 15F FIVE OAKS LIMESTONE TOMOGRAM AT 86 18 8 5 60 FIGURE 4 156 FIVE OAKS LIMESTONE TOMOGRAM 103 42 1 5 5 61 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM POST FAILURE 0 MPA UNITS FT SEC 61 CHAPTER 5 SUMMARY AND CONCLUSIONS ccsccsccsssssssssssccccccssssssssccccccsssssssssccsscsssssssscocses 63 66 5 dul UN DI b 70 FIGURE A 1A TOMOGRAM OF THE SOLID PLEXIGLAS CALIBRATION PLATE 71 FIGURE 1 TOMOGRAM OF THE PLEXIGLAS CALIBRATION PLATE WITH HOLE 71 FIGURE 2 BEREA SANDSTONE TOMOGRAM AT INITIAL STATE 0 MPA een 71 FIGURE A 2B BEREA SANDSTONE TOMOGRAM AT LOAD 1 17 24 72 FIGURE 2 BEREA SANDSTONE TOMOGRAM AT LOAD 2 24 82 72 FIGURE 2 BEREA SANDSTONE TOMOGRAM AT LOAD 3 34
43. TSI 2 Y 112 gore R un E 4 aye mg z 51 15 j 414 o gt gt gt gt gt CLOCK uuum e 0 6 6 e gt 4 o O reset device INISCOPET INI SCOPE 77 i RIS CO PE nnn nnn nnn nnne ART scoppi b SORE 2 gt mf SCOPE gt CLOCK urs M eMPIGTRIG9 gt ubi FETCH I NISCOFE scor PHI inaf scoppi 1 4 gt L VERT gt gt gt Eunice CLOCK gt gt gt gt gt gt gt gt gt gt gt gt gt gt 2 gt gt gt gt gt gt 222 gt gt 061 S ai FETCH I NISCOFE _ gt VERT gt C CLOCK TRG be NISCOFE o 5 2 gt gt CLOCK jur 016 Tije fja FETCH o NISCOFE A m VERT gt 2 CLOCK o ee eeneie xXDIGTRIieu gt gt FETCH EE NISCOPE Acquire t NS e VERT L gt 4 CLOCK orra xD TRI FETCH ca i Ill Type INI SCOPE z A gt PUSS LFEI6H FI
44. The event calls to_pixel vi to convert the rectangular coordinates from inches to pixels so that they can be drawn in the picture B 3 SIMULTANEOUS ACQUISITION PROGRAM Once the setup has been conducted acquisition of the ultrasonic waveforms can be done One option for this is to simultaneously acquire ultrasonic waveforms from 16 channels on the PXI chassis This VI could be easily modified to handle more or less acquisition channels This program was written to acquire ultrasonic waveforms from 8 PXI 5102 digital oscilloscopes The oscilloscopes were setup in a master slave configuration total of 9 oscilloscopes are used with this program first oscilloscope reads a trigger signal from the square wave pulse The triggering pulse then triggers simultaneous acquisition on the other 8 oscilloscopes The front panel of this VI is shown in Figure B 9 68 Wesley Johnson Appendix B LabVIEW User s Manual a U Usa7 xc T Leo Acquire A 9 05 1 une jc ILII 1 FIGURE 9 SIMULTANEOUS ACQUISITION FRONT PANEL The user inputs the scope properties in the upper left portion of the Front Panel The inputs include the vertical range sample rate record length and the trigger level The user then inputs the number of source channels being acquired and a file path to save acquired ultrasonic waveforms Once these inputs have been satisfied the acquire button can b
45. Wesley Johnson Literature Review it is important to match the impedance of the transducer to that of the rock to maximize the energy transmitted it is also important to know the mineralology of the sample being tested to know if refraction will be an issue These concepts were discussed above At the reception point other factors such as sensor sensitivity reflections sensor directionality and the sound field become important Sensor sensitivity and directionality are properties of the sensor The sensor sensitivity 1s the sensor s ability to detect a signal Sensitivity is commonly associated with the ability to detect a weak signal A sensor can be damped to improve sensitivity By damping the sensor is tuned to resonate at a distinct frequency The sensor then becomes very sensitive to this frequency but not to others An undamped sensor 15 not sensitive but will have a wide frequency reception bandwidth Directionality refers to the angular reception and transmission range of a sensor for example a point source would have an angular transmission of 360 whereas a planar source would have a directionality of 0 The sound field consists of a near field and far field These fields are similar to the focal length of a lens where within the near field the image 15 unfocused but beyond that region the image becomes clear The near sound field 15 the area in front of the transducer where a signal 15 hard to measure due to multiple wav
46. a rock 15 mechanically loaded the acoustic velocity through the rock will increase proportionally with stress This is known as the acousto elastic effect A tomographic survey which is conducted at a stress state within the acousto elastic region will show a cross sectional picture of the relative internal stress To observe this the velocity within the material is first found by collecting an ultrasonic waveform by transmitting an ultrasonic wave through the material and the arrival time of the wave 15 found There are different methods of automatically finding the arrival time of the waveform In seismology automatic P wave arrival picking 1s very important and many studies have been done to make this process more robust The main difference between laboratory applications of arrival time picking and seismic applications 1s the difference in the signal to noise ratio SNR The SNR for laboratory testing normally 15 very high this 1s not the case for seismic testing Therefore the algorithms used for arrival time picking are complex for seismic data to account for the poor SNR The methods of fuzzy logic and wavelet transforms are too complex for the application in this study and will not be discussed The direct correlation method of arrival time picking is used in the laboratory and in the field This method uses statistics to determine the arrival time of the waveform It is a simple method in which a portion of a waveform 15 compared to a refe
47. a wide range of frequency matching to different rock types The small diameter 10 mm allows for a reduction in energy loss through the couplant from mounting the flat sensor face to the curved rock surface 3 3 ULTRASONIC PULSER The ultrasonic pulse that passes through the sensors and into the rock originates at a pulsing device The pulsing device selected for this data acquisition system was a Panametrics 5077PR This is an ultrasonic square wave pulser receiver unit The pulser can operate two modes pulse echo or thru transmission The thru transmission mode was used for this system and the receiving capability of the unit was not utilized A picture of this unit is shown in Figure 3 3 FIGURE 3 3 PANAMETRICS 5077PR ULTRASONIC SQUARE WAVE PULSER The pulser features adjustable output voltage output frequency and repetition rate Table 3 2 shows a summary of these features 24 Wesley Johnson Hardware Design TABLE 3 2 PANAMETRICS 5077PR ULTRASONIC SQUARE WAVE PULSER SPECIFICATIONS Panametrics 5077PR Specifications Pulse Type Main Bang Negative Square Wave adjustible duration amp amplitude Maximum Pulse Amplitude 400V no external load Pulse Width 10 preset fixed widths for the following transducer frequencies ranges 15 20MHz 10MHz 7 5MHz 5 0 6 0MHz 3 5 4 0MHz 1 2 0 5MHz 0 25MHz and 0 1MHz Each width can be fine tuned by at least 25 Repetition Rate 100 200 500 1000 2000 5000Hz f
48. ample The drawing in Figure 4 8 shows the location of the load platen the sensor spacing and the ultrasonic raypaths between each source and receiver pair Top View Isometric View Front views Ex Receiving Sensor Raypath Source Sensor Berea Sandstone Sample Tomographic Setup Units Centimeters FIGURE 4 8 BEREA SANDSTONE TOMOGRAPHIC SETUP 43 Wesley Johnson Calibration and Testing The sensors were attached using beeswax Figure 4 9 shows the sample before and after failure b FIGURE 4 9 BEREA SANDSTONE TEST SAMPLE A PRIOR AND B POST FAILURE Tomographic data were collected as the sample was loaded The loading was paused to collect data at 0 17 24 24 82 34 47 46 54 MPa using the Acq 16 in LabVIEW The arrival times for each load were then picked using the Correlation vi LabVIEW program Arrival times were plotted against distance to observe how well the arrival times were picked Figure 4 10 shows plots of arrival time vs distance for all different load conditions Sandstone No Load 4 E 05 y 2 8123E 06x 2 6767E 12 ES R 0 95435 Ls gt 2 05 E 4 E 05 0 E 00 0 2 4 6 8 10 12 Distance cm FIGURE 4 10A BEREA SANDSTONE TIME VS DISTANCE PLOT AT LOAD 1 0 MPA LOAD 44 Wesley Johnson Calibration and Testing Sandstone 17 24 MPa 5 05 y 2 9663E 06x 5 0622E 11 R 0 91134 4 E 05 3 E 05 2 E 05 Time
49. ant to have at least one intersection for each pixel so each pixel represents an average velocity 34 Wesley Johnson Calibration and Testing CHAPTER 4 ULTRASONIC DATA ACQUISITION SYSTEM CALIBRATION AND TESTING Ultrasonic data acquisition was conducted on three different materials to test the system for velocity calibration void identification and stress identification Velocity calibration and void identification was done by testing Polymethylmethacrylate PMMA or Plexiglas material of known velocity in the ultrasonic data acquisition system UDAQ Stress measurements were done using Five Oaks Limestone and Berea Sandstone Index properties for the limestone were obtained experimentally in this study the index properties for the sandstone had been previously tested This chapter describes the process of preparing and testing samples for index property values as well as the procedure for using the UDAQ system The reliability of velocity measurements from the UDAQ system was tested to calibrate and validate the system for velocity measurements The calibration of the velocity measurement in the UDAQ system was done using Plexiglas plates and initial application tests were conducted to image a void within a Plexiglas plate Final application testing was to image an indentation load in rock Two indentation tests were conducted on two different materials Five Oaks Limestone and Berea Sandstone The index properties for the rock we
50. ased by 2 so the distance traveled over the angular change from 0 to 27 1s the wavelength 1 The onset of the wave across a material marks the arrival time of the wave The time it takes to travel across the sample At is divided from the distance traveled Ax to determine the speed of the wave This relationship is also equal to the product of the wavelength and the frequency of the wave This is shown in Equation 2 2 Wesley Johnson Literature Review 2 2 where V sound velocity wavelength f frequency Ax change in position At change in time Ultrasonic waves depend on the elastic properties of a material Equation 2 3 shows how density and the bulk modulus are related to the speed of sound through a material 2 3 p where C elastic constant p density 2 4 V 2 2 46 3 1 2 2 4 2 p 2 4 where Bulk Modulus E Elastic Modulus v Poison s Ratio Ap change pressure P wave Velocity V S wave Velocity pu 2 5 where o stress Wesley Johnson Literature Review strain Elastic constants such as the bulk modulus K and the Young s Modulus E are found experimentally These index values can be used to characterize a rock sample as Equation 2 3 shows to find the theoretical material velocity The equations for these moduli are shown Equations 2 4 a c and 2 5 5 The theoretical value o
51. cal waves have a frequency range above 20 kHz These waves can only exist 1n a material medium such as air water rock 1 These Ultrasonic waves travel two modes longitudinal and transverse longitudinal ultrasonic wave 1s also known as a pressure wave or P wave This type of wave 15 the fastest moving ultrasonic wave The wave propagates along a straight line due to particles on the molecular level oscillating parallel to the direction of propagation A transverse wave 15 also known as a shear wave or S wave This wave propagates perpendicular to the direction of travel due to a shear oscillation on the molecular level The S wave 15 slower than the P wave due to the mode of oscillation To completely describe an ultrasonic wave a function can be formed that describes the shape of the wave This function is described Equation 2 1 1 h x t 2 1 The variable y represents transverse displacement of an element in the function at the time and in the position of x The amplitude of y is the magnitude of the displacement of the wave The phase of the function refers to the angular motion with respect to the position x The phase changes linearly with time 1 Repetitions of the pattern or shape of the function can be measured to determine the wavelength of the wave wavelength 15 the linear distance the wave travels in one cycle of wave s function For example a sine function begins to repeat itself when its angle 1s incre
52. cers and the Acousto Electric Properties of PVDF and P VDF TrFE Films IEEE transactions on ultrasonics ferroelectrics and frequency control 2000 38 Lee Yung Chun and Shi Hoa Kuo A New Point Contact Surface Acoustic Wave 68 Wesley Johnson References Transducer for Measurement of Acoustoelastic Effect of Polymethylmethacrylate IEEE transactions on ultrasonics ferroelectrics and frequency control Vol 51 No 1 p 114 120 2004 39 Onda Corporation References 2002 Onda Corporation 3 Nov 2004 lt http www ondacorp com gt 40 Stein J ed The Random House College Dictionary 1980 Random House Inc 69 Wesley Johnson Appendix A GeoTomCG Output APPENDIX A GEOTOMCG TOMOGRAPHIC OUTPUT 70 Wesley Johnson Appendix A GeoTomCG Output A PLEXIGLAS TOMOGRAMS Velocity 12000 10250 8500 6 50 5000 Velocity 12000 10250 8500 5750 5000 FIGURE 1 TOMOGRAM OF THE PLEXIGLAS CALIBRATION PLATE WITH A HOLE A 2 BEREA SANDSTONE TOMOGRAMS Velocity 20000 15250 12500 8750 5000 FIGURE 2 BEREA SANDSTONE TOMOGRAM INITIAL STATE 0 MPA 71 Wesley Johnson Appendix A GeoTomCG Output Velocity 20000 15250 12500 8750 5000 Velocity 20000 16250 12500 8750 5000 Velocity 20000 15250 12500 8750 5000 FIGURE 20 BEREA SANDSTONE TOMOGRAM AT LOAD 3 34 MPA
53. cy of the arrival time picking determines the accuracy of the tomograms The Front Panel of this program 15 shown in Figure B 12 correlation pick vi Front Panel File Edit Operate Tools Browse Window Help it Set Ref WFM Values _ 027 _ _ 0 00234 188 8 FIGURE B 12 ARRIVAL TIME FRONT PANEL This program uses direct correlation to assist in the arrival time picking procedure Direct correlation allows for a comparison of a reference arrival time to the actual waveforms The correlation function picks the point where the waveform best fits the 92 Wesley Johnson Appendix B LabVIEW User s Manual reference arrival time To use this program the user first needs to input the source location receiver location reference waveform and acquired waveform files This being done the reference waveform properties can be adjusted by clicking on the Set Ref WFM Values The default reference waveform properties should be acceptable for most applications Having entered in the file locations and the reference waveform values the arrival times can be correlated by clicking on the Correlate t a button This will call the direct correlation function to pick the arrival times on each waveform After the correlation 15 finished the user then needs to cycle through each waveform to check the fit To cycle through the waveforms click on the WFM Index numerical control Changing the value in this n
54. d mines but 15 mainly being studied on the laboratory scale An overview of other engineering applications and equipment design for ultrasonic investigations can be found the literature 17 18 21 and 26 Ultrasonic tomography has been conducted the laboratory to better understand the implications of this monitoring method By analyzing sample behavior in the laboratory many experiments with different loading conditions and sample characteristics can be conducted rapidly The design of a tomographic data acquisition system for the laboratory would be advantageous to allow these experiments to be conducted Laboratory tomographic systems have been developed in past experiments 6 29 These 18 Wesley Johnson Literature Review systems were not designed to accommodate new technologies such as faster acquisition computers and higher resolution A D converters Experiments that have been investigated involve the stress distribution from indentation 28 stress distribution and acoustic emission location hydraulically loaded samples 30 31 19 Wesley Johnson Hardware Design CHAPTER 3 ULTRASONIC DATA ACQUISITION SYSTEM HARDWARE DESIGN The accuracy and precision of a tomographic imaging system 15 determined by the quality of the acquisition system used The data acquisition system for this study was designed to be used a laboratory setting Samples for this system can have a cross sectional area betw
55. d mining The cause of these geologic hazards can be a result of a number of geologic factors including stress redistribution resulting from mining Monitoring of geologic hazards to predict failure within mine workings has been conducted with minimal success Microseismic and acoustic emission monitoring methods have been used to collect acoustic signals through the roof back or the floor thereby assessing the stress state in those structures The results from these mine wide monitoring systems have been inconclusive in determining the direct result of stress redistribution to the prediction of geologic failure The phenomenon of stress redistribution must be understood before field predictions be made Ultrasonic investigations used the petroleum industry show promise in imaging stress redistribution Ultrasonic investigations rock mechanics have been historically used to determine the index properties of rock By determining the pressure and shear wave velocities dynamic properties such as the modulus of elasticity can be computed New technology allows ultrasonic investigations to go beyond index properties and infer the stress state within the rock When a mechanical stress 15 applied to a material that behaves according to Hooke s Law 3 stress increases with strain the sound velocity increases with increasing stress This is known as the acoustoelastic effect This effect makes ultrasonic stress redistribution studies possible
56. e pressed to begin acquisition When this button is depressed the counter in the upper left will count up to the number of sources specified The acquired waveforms can be viewed by changing the value in the numerical control The waveform numbering begins at zero and counts up to the number of sources x 16 To stop the program press the stop button The block diagram programming for this VI 15 simple as well The same event structure programming used in the previous programming is used again here This programming allows for simple control of operations by the user and also makes the program easier to read The main function in this program 15 the acquisition call Figure B 10 shows the acquisition call to the block diagram 69 Wesley Johnson Appendix B LabVIEW User s Manual lt ICIXI gt 16 71 Block Diagram File Edit Operate Tools Browse Window Help cleo enr 4 Value Change v File Path of Sources 13231 N Hsm 8 po s fe Ze gt Master Device E NISCOFE NI SCOFET NIS COFE Nisco SCOPE m 2 7 PUMA config VERT ess P Adjust Samples ric a Tria Out e reset device trigger coupling trigger slope 20 0 clock sync pulse source R
57. e range 15 too wide the waveform will become pixilated and if the range is too narrow the waveform will be clipped It 15 important to know the properties of the wave being measured and the properties of the sensing device before assigning a digitizing range 10 Wesley Johnson Literature Review Analyzing an acquired waveform using digital signal processing DSP can remove unwanted frequencies which will improve the signal of interest LabVIEW has built in DSP functions that allow for a waveform to be filtered transformed and otherwise manipulated digitally to better resolve a noisy or incomplete waveform These methods are useful when post processing data but can create false data points by over smoothing or interpolating the waveform 2 2 1 SAMPLING The information contained within an analog waveform can be captured by sampling the waveform Continuous analog data are sampled at discrete intervals The sampling interval must be chosen carefully to insure an accurate representation of the original signal Faster sampling of a continuous signal will allow for a more accurate picture due to the abundance of sampling points but 1f fewer samples are taken a point is reached where information about the signal can be lost This point 15 defined by Nyquist s criteria 32 The Nyquist criteria requires that the sampling frequency fs be at least twice the signal bandwidth f or information about the signal will be lost The loss in in
58. ease of velocity the tomogram Future studies will explore phenomena of stress redistribution in different rock types using this data acquisition system along with discrete element models in two and three dimensions This new tool for ultrasonic imaging of rock provides a means to better understand the implications of tomographic imaging in geophysics The greatest benefit to the mining industry 1s for identifying geologic hazards within mining operations Accurate interpretations of tomographic data as well as the design of mine wide monitoring systems will allow for geologic hazards to be correctly identified Through further tests with this ultrasonic data acquisition system different stress conditions relating to mine conditions and geologic hazards can be identified the laboratory providing the mining industry a means to correctly interpret stress redistribution readings using current mine wide monitoring systems 65 Wesley Johnson References REFERENCES 1 Halliday D R Resnick and J Walker Fundamentals of Physics 6th ed New York John Wiley amp Sons Inc 2001 2 NDT Resource Center 1996 Iowa State University Sept 2004 lt http www ndt ed org gt 3 Lempriere B M Ultrasound and Elastic Waves Frequently Asked Questions New York Academic Press 2002 4 Bishop R Learning with LabVIEW 61 Upper Saddle River NJ Prentice Hall 2001 5 Nolet G Ed Seismic Tomography With A
59. economic cost that mines face due to geologic failures monitoring must be conducted The laboratory provides a controlled environment where it is possible to explore what ultrasonic tomographic data is implying about rock samples The study presented in this report describes a new laboratory ultrasonic data acquisition system for acquiring and processing ultrasonic tomograms This system uses National Instrument s LabVIEW programming to control the acquisition of ultrasonic waves from a rock sample under uniaxial compression Tomograms were collected Wesley Johnson Chapter 1 Introduction during uniaxial compression of Five Oaks Limestone and Berea Sandstone The limestone samples were collected from the Kimballton Limestone Quarry which 1s owned and operated by Chemical Lime Corporation The sandstone samples were obtained from Cleveland Quarries in Amherst Ohio The data acquisition system designed this study is adaptable to be used with different rock types as well as adapting to image different cross sectional planes This study details the development and tomographic testing of a data acquisition system as well as recommendations for other ultrasonic investigations that are capable with this system Wesley Johnson Literature Review CHAPTER 2 LITERATURE REVIEW 2 1 ULTRASONIC WAVE PROPAGATION Ultrasonic sound waves are classified as waves that have frequencies higher than what can be heard by the human ear These mechani
60. educes the quantization step increasing the resolution However amplitude data might be lost due to clipping of the signal Clipping occurs when the vertical range of the acquisition window is less than the actual vertical range of the analog signal being captured The top of the waveform appears to be cut off and the data outside of the digitizing range 1s lost Therefore the vertical range that is used for acquisition must be set so that the arrival time of the waveform can be acquired with minimal clipping of the waveform LabVIEW programming allows for the vertical range to be adjusted easily minimizing any loss of data due to clipping Precise triggering of the digital oscilloscopes reduces timing errors and improves the overall precision of the acquisition system To achieve this the output signal of the pulser was wired into one of the digital oscilloscopes using a voltage divider circuit The input voltage was dropped from 400V to 4V by using the circuit shown in Figure 3 9 O CPFSUIHUPI ODDE ANDAA GOT UNIVEROAL BREAUBUARL san Jaani WERNE KAHENE KEEA EV 1 Fee nur Bn els III vaada XERHE ERREP FEILE qana ee cote FEEFEE EBENE b FIGURE 3 9 4 BREADBOARD VOLTAGE DIVID
61. een 20 100 cm Sensors with smaller diameters should be used to improve the contact between the sensor and rock additionally the small diameter increases the number of sensors that can be attached to the sample The selection of the proper couplant is important to increase the signal to noise ratio SNR of the received signal A high SNR will increase the accuracy of arrival time picking The sensors must be capable of transmitting an ultrasonic pulse or receiving the pulse The ultrasonic pulser that is used with this system should have adjustable output frequency and output amplitude This will allow for more precise matching of the resonant frequency of the transmitted signal to that of the rock Once the signal 15 received the signal is digitized using an analog to digital A D converter The resolution of this device must be fine enough to resolve the first break of the received waveform to accurately calculate the velocity of the wave The resolution of the A D converter is determined by the number of quantization steps vertical resolution and also by the acquisition speed horizontal resolution The flowchart in Figure 3 1 shows the relationship of the instruments selected for the ultrasonic data acquisition system 20 Wesley Johnson Hardware Design Ultrasonic Pulse Source Pulsing Couplant Device Receiving AE Sensor Transmitting AE Sensor Analog to Digital
62. en 29 35 TABLE 4 1 ULTRASONIC DATA ACQUISITION SYSTEM VELOCITY CALIBRATION MEASUREMENTS 39 TABLE 4 2 BEREA SANDSTONE INDEX PROPERTIES cccesceccecceccscescesceccscescesceccscescescestecescescescssescesceseess 43 TABLE 43 SANDSTONE VELOCITY SUMMARY 46 TABLE 4 4 FIVE OAKS LIMESTONE INDEX 8 2 020000000000000000000000000 51 TABLE 4 5 LOAD CONDITIONS FOR TOMOGRAPHIC DATA COLLECTION ccccscesceccecceccecscesceccecesceseecs 52 CHAPTER 5 SUMMARY AND CONCLUSIONS ccccccccccccccccccccccccccccccccccccccccccccceccccccccccecces 63 ____ _ __ eee 64 2 23222 2 2 222 2 22 2 2222 _ _ ___ _ _ _ _ __ __ E 606 32 3 90 70 APPENDIX D ____ __ 2 222 222 23 lt _ _ 78 2 3222 22 2 lt 22 lt 2 _ _ __ _ _ _ _ 95 TABLE C 1 HARDWARE DESCRIPTIONS 0 0000000000000000 97 Wesley Johnson Chapter 1 Introduction CHAPTER 1 INTRODUCTION Roof falls and pillar failures are common workplace hazards in undergroun
63. erground openings helps identify hazards within the mine Tomography can also be used to identify geologic structures abandoned mine workings and other interfaces within a rock mass The data collected from these investigations are often very hard to interpret due to factors such as groundwater complex geology faults and fissures poor ultrasonic reception and background noise Some of these factors be minimized the laboratory reducing errors that influence tomographic results The Mine Safety and Health Administration MSHA Accident Injury Illness Database shows the past four years 2000 2003 11 8 of fatal incidences were a result of the failure of a geologic structure roof rib pillar etc surface or underground mine This statistic could be lowered by monitoring the stress states inside of these dangerous places Non fatal incidences resulting in man hours lost totaled 83 213 man hours in the past four years In underground coal mining the number of man hours lost totaled 70 963 hours Labor statistics show that the average tons per man hour in underground coal mines equals 4 05 tons With an estimated market value of 40 per ton over the last four years this equates to 11 5 million dollars in lost production due to geologic failure in underground coal mines This statistic does not take into account down time for production equipment directly resulting from geologic failures inside of a mine To reduce the human and
64. es from all parts of the transducer The distance of this area can be found using Equation 2 8 4 27 2 8 Where d transducer diameter wavelength The far sound field is the region outside of the near field where the transducer waves coalesce to produce a plane wave whose on axis intensity decreases inversely with distance 3 From Equation 2 8 it 1s seen that the diameter of the transducer influences the length of the near field The advantage of a small diameter transducer 1s that it can measure thin materials the disadvantage 15 that the intensity of the wave rapidly decreases with distance In contrast to a small diameter transducer a large diameter Wesley Johnson Literature Review transducer will have a larger near field but the intensity of the wave does not diminish as quickly Wesley Johnson Literature Review 2 2 DATA ACQUISITION Data acquisition is the process of collecting data a digital or analog format to describe a physical occurrence The components included in a data acquisition system consist of a sensing device a signal conditioning device and analog to digital conversion The resolution of a DAQ system 15 relative to the resolution of the sensing device and of the A D converter The sensing device used in most applications of ultrasonic testing is a piezoelectric transducer PZT These sensors consist of a piezoelectric crystal that is damped to produce a sensor that r
65. esonates at a specific frequency The orientation of the crystal determines whether the sensor will detect compressional or shear wave types To create a sensor that is highly sensitive a small amount of dampening is used To create a sensor that 15 selective a larger amount of dampening 15 used to allow the sensor to resonate at a specific frequency The size of the sensor determines the length of the near field and the resonating frequency determines the minimum wavelength that can be resolved by the sensor To maximize the resolution of a sensor applied to geophysical monitoring the frequency of the rock type that 1s being measured must be known Most rock types will only transmit ultrasonic frequencies the range of 200 kHz to MHz Therefore a sensor with a wide bandwidth at a lower frequency range would be preferable for measuring ultrasonic waves rock The resolution of an analog to digital A D conversion device is determined by its quantization The quantization of an A D device 1s defines its digital resolution The resolution 15 determined by the number of quantization steps This is commonly reported by the bit resolution of the A D device 32 A bit consists of 2 steps where n is the number of bits The digital encoding of an A D device takes an analog value and assigns it to the nearest digital step within the digital range being measured The range of encoding 15 important to the final resolution of the digitized wave If th
66. ess zone The final tomograms Figures 4 15g h show the presence of very low velocities and very high velocities The reason for the high contrast in these final tomograms 15 due to the fracturing of the limestone In Figure 4 15g the low velocity zones follow the patterns observed in the initial tomogram The contrast 1s increased because loading has dilated the existing fractures In Figure 4 15h the high velocities are due to reflections of the ultrasonic waves at the fracture interface To enhance the resolution within the tomograms the wavelength should be reduced To reduce the wavelength in the limestone a higher frequency wave should be introduced Another option 15 to use same frequency but use a sample with a lower ultrasonic velocity so that the wavelength is smaller 62 Wesley Johnson Conclusions CHAPTER 5 SUMMARY AND CONCLUSIONS A tomographic data acquisition system has been presented in this study The system was developed in a LabVIEW programming environment The advantage of using this program 1s that all ultrasonic data acquisition can be done within LabVIEW as well as all signal processing LabVIEW is an intuitive and easy to learn programming language that is geared towards engineering applications The plug and play capabilities of LabVIEW allow for upgrades of the control computer sensors and acquisition cards without any new programs being written This allows for the data acquisition system to be upg
67. eswax to fill gaps formed by incompatible geometries can be seen in Figure 3 2 BEESWAX BEND LIMESTONE SAMPLE FIGURE 3 2 BEESWAX COUPLANT APPLICATION 27 Wesley Johnson Hardware Design When the beeswax 15 heated it becomes soft and pliable and when it cools it hardens The hardening of the wax ensures that the sensors are held in a rigid position not sliding when the sample is moved or loaded A disadvantage to using wax 15 that the sample and sensors must be kept warm or the wax becomes brittle pulling away from the sample under high loads A cynoacrylate adhesive was used to mount sensors to flat non reactive surfaces This adhesive becomes very rigid when dry and can be released by a debonding agent The sensors selected for this system are Physical Acoustics Corporation Micro 80 miniature acoustic emission sensors Properties for these sensors are listed in Table 3 1 Table 3 1 Physical Acoustics Micro 80 AE Sensor Properties Micro 80 Specifications DIMENSION DIAXHT 10x12 4x 5 mm Inch WEIGHT gm OPERATING TEMP C SHOCK LIMIT g CASE MATERIAL FACE MATERIAL CONNECTOR TYPE CONNECTION SIDE LOCATION PEAK SENSITIVITY dB 57 65 1V m s 1V ubar OPERATING FREQ 175 1000 RANGE kHz RESONANT 250 FREQUENCY kHz DIRECTIONALITY dB 325 1 5 SEAL TYPE EPOXY 23 Wesley Johnson Hardware Design The sensors have a wide bandwidth allowing for
68. f velocity may not be the same as the measured value due to the presence of discontinuities within the rock sample Ultrasonic waves can travel through any natural material as explained above The resonant frequency of a material 15 an important property to note when attempting to pass ultrasonic waves through the material In rock this value is usually nonlinear The resonant frequency of a rock can change with saturation or stress Matching the resonant frequency of the rock to the transmitted ultrasonic signal will maximize the amount of the wave s energy that is passed through the rock The resistance to the flow of sound wave through the rock is known as acoustic impedance Z The equation for acoustic impedance 1s shown in Equation 2 6 Z 2 6 where Z Acoustic Impedance material density V material velocity Frequency matching 15 a common way to maximize the signal strength and minimize acoustic impedance Refraction is another concern when passing an ultrasonic signal through a rock Refraction is the incomplete transmission of a wave through an interface When a wave interacts with two materials that are not the same 1 e mineralization faults or bedding planes inside of a rock the wave 15 split in two and 15 said to be refracted 1 2 Part of this refracted wave is reflected and the other part 1s transmitted at a different angle This angle of refraction 1s influenced by the angle of wave incidence Wes
69. formation will result in what 1s known as aliasing An example of aliasing 15 shown in Figure 2 2 Mhuistration of the Miygast Theorem signe ahasmng in Dame Gown gt E a 5 Frequency Time Alger Signal Nyquist Frequency FIGURE 2 2 SIGNAL ALIASING IN THE TIME DOMAIN 11 Wesley Johnson Literature Review Proper sampling rates must be used to prevent aliasing Sampling rates higher than the required rate determined by the Nyquist theorem improve the resolution of the first break of the waveform allowing for more accurate arrival time picking 12 Wesley Johnson Literature Review 2 3 LABVIEW PROGRAMMING Laboratory Virtual Instrumentation Engineer s Workbench LabVIEW 1s a graphical programming language developed by National Instruments This program was developed for instrument control and data collection Virtual Instruments VIs are made to collect data from measurement devices control or manipulate instrumentation and process data A VI consists of a front panel and a block diagram The front panel is a VI s control panel while the program is executing On this screen graphs knobs buttons and other objects can be manipulated by the user An example of a VI front panel is shown in Figure 2 3 File Edit Operate Tools Browse Window Help 3 in 3 d 1 A 17
70. g sample geometries and rock types with a low initial investment The system is composed of sensors an ultrasonic pulser a source switchbox and analog to digital converters it 1s tied together using a LabVIEW virtual instrument LabVIEW 1s a graphical development environment for creating test measurement and other control applications Using LabVIEW virtual instruments VIs are created to control or measure a process In this application LabVIEW was used to create a virtual instrument that was automated to collect the data required to construct a tomogram Experiments were conducted to calibrate and validate the system for ultrasonic velocity determination and stress redistribution tomography Calibration was conducted using polymethylmethacrylate PMMA or Plexiglas plates Uniaxial loads were placed on limestone and sandstone samples The stress induced density contrasts were then imaged using acquisition system The resolution and accuracy of system is described The acquisition system presented is a low cost solution to laboratory geophysical tomography The ultimate goal of the project is to further the ability to non invasively image relative stress redistribution in a rock mass thereby improving the engineer s ability to predict failure 111 ACKNOWLEDGEMENTS I would like to thank the Virginia Tech Mining and Minerals Engineering Department and the National Science Foundation for the opportunity to contribute to t
71. h shows the resistivity through the medium 8 Reflection tomography measures reflected ultrasonic waves as they bounce off of interfaces within a body This method can be used to locate geologic structures or old mine workings 24 Seismic tomography is very similar to ultrasonic tomography The difference between these two methods of tomographic imaging 1s the frequency ranges used Seismic tomography utilizes low frequency waves to measure structures within the earth The low frequencies correspond to long wavelengths which are capable of traveling long distances Ultrasonic tomography is used more for laboratory and mine wide monitoring systems Ultrasonic waves have small wavelengths which can resolve small structures however ultrasonic waves attenuate quickly and can therefore only be transmitted over short distances Ultrasonic tomography can be used to measure the 15 Wesley Johnson Literature Review slowness of the P or S wave across the sample as well as amplitude variations across the sample This study involves ultrasonic P wave velocity tomography This type of tomography produces a cross sectional image of velocity across a body Rocks behave according to Hooke s Law where stress and strain are related When the rock 1s under a load it reaches an equilibrium point where the rock will change in length strain to accommodate a change the load across the area of the rock stress This is known as elastic loading As
72. he third party tomogram program This program adjusts the fit of the data so that the general linear trend of the time versus distance plot passes through the origin An offset was created in the acquisition program to ensure that all ultrasonic data 15 collected before and after the trigger pulse The validation of the data acquisition system for tomographic data collection was achieved by using acoustic emission sensors on three different materials Plexiglas Berea Sandstone and Five Oaks Limestone The sensor locations and assignments source or receiver were entered into the LabVIEW program to determine X Y Z locations The acquisition was done using the simultaneous option to eliminate some of the timing error that might be introduced by multiplexing through the sensor array The arrival times were then picked using the direct correlation program and the times were adjusted to fit a linear trend passing through the origin Tomographs were created GeoTomCG This software was developed by GeoTom LLC for creating tomograms from geophysical data This program uses simultaneous iterative reconstruction technique SIRT to perform tomographic inversions To create a tomogram in GeoTomCG a model is generated in the program This model then undergoes an iterative inversion using SIRT The inverted model is then displayed showing a pixilated cross section of the model Each pixel represents a point velocity at raypath intersections It 1s import
73. his project am thankful for the educational and financial support that was extended to me for this project I would like to thank my advisor Dr Erik Westman for his guidance and support throughout this project and throughout my academic career I would also like to thank my other major advisors to this project Dr Mario Karfakis and Dr Thomas Novak I am thankful for my committee s patience and support in mentoring me on this project To my wife Laura I am grateful for her love and sacrifice to see me finish this project In addition I would like to thank my parents and family for their support and love 1V TABLE OF CONTENTS ETS TOR FIG 5 22202 RR 2 2 2 2 2 2323 3 22232 22 322 2 lt _ X CHAPTER 1 INTRODUCTION deed eve utei recie v ei vae po aive pee 1 CHAPTER 2 LITERATURE REVIEW ee eeeeeeeeece eee eeeee esee eese eee see sees eese esses eese 4 Dol ULTRASONIC WAVE PRO PAAT O N editus deren n d ara ud cuo E uv ode 4 2 qu d pq ds ON ICS EIN SORS 1 aim etus idet eia etsi Totg Li aA ec e Eis AL 7 AC OUTS EC 10 _ 3 32222 2_ __ 11 29 LCABVIEWPROGRA M MN O ees a 13
74. ice which utilized Visual Basic programming to control the relay function of the switchbox This card was not used because the control computer did not have an ISA slot to accommodate the factory control card A National Instruments PCI 6503 digital input output data acquisition card was used as a digital control for the relay box Figure 3 5b THMATIONAL CICOPYSTGHT 1584 TT RSPAS T m l ereeecaaeceaec eadteettae eeancecocececeaetatenncecc a b FIGURE 3 5 DIGITAL INPUT OUTPUT CONNECTOR BLOCK B NI PCI 6503 DIGITAL INPUT OUTPUT DATA ACQUISITION CARD Wires from the digital I O connector block Figure 3 5a were connected to a 25 pin dongle A dongle was made to plug into a 25 pin serial cable and was wired according to the pin out diagram shown below 26 Wesley Johnson CHI DQO CH2 DQI CH3 DQ2 CHA DQ3 5 DQ4 CH6 DQ5 CH7 DQ6 CH8 DQ7 N C 10 Vcc 9V 11 GND 12 N C 13 N C d WN Hardware Design CH9 DQO CH10 DQI CH11 DQ2 CH12 DQ3 CH13 DQ4 CH14 DQ5 CH15 DQ6 CH16 DQ7 N C Vcc 9 GND TTL pulse out Power for the box was supplied by an Agilent power supply running at a constant output voltage of 10V The complete switchbox controller is shown in Figure 3 6 FIGURE 3 6 ULTRASONIC SWITCHBOX CONTROLLER HARDWARE CONNECTIONS The setup shown in Figure 3 6 allows for the relay of the ESG ultrasonic switchbox
75. irly uniform due to the tight linear trend The plot in Figure 4 5 shows the linear fit for Plate 2 The trend in this plot 15 linear like Plate 1 but the fit is not as tight R 0 95 The variation in fit is most likely due to the presence of the hole in Plate 2 Using GeotomCG velocity tomograms were made The smallest structure that can be viewed in the tomograms 15 dependant on the wavelength of the ultrasonic wave that 15 received through the Plexiglas The calculation for wavelength 15 shown in Equation 4 1 4 1 Where 4 Wavelength V Material Velocity 39 Wesley Johnson Calibration and Testing f Frequency The ultrasonic frequency used for testing the Plexiglas was 1MHz and the ultrasonic velocity was found experimentally to be 2 820 m s Using equation 4 1 the resulting wavelength was 0 28 cm Tomograms were made from the ultrasonic data using GeoTomCG Shown Appendix A The resulting tomographic data was smoothed using the Nearest Neighbor interpolation method The smoothing did not add erroneous data points and made the visual interpretation of the tomograms easier The tomograms in Figure 4 7 show that the hole in Plate 2 was clearly imaged and the solid plate Plate 1 has uniform velocity across the tomogram 17000 11400 11000 10400 10000 9500 9000 8500 8000 000 6500 6000 S400 5000 0 15 0 1 0 15 02 0 25 0 3 FIGURE 4 7A TOMOGRAM OF THE SOLID PLEXIGLAS CALIBRATION PL
76. is then passed back into the main panel program so the location data can then be called by the acquisition programs The next two options are acquisition options The first acquisition option 15 to acquire pairs Choosing this option opens up a program that allows the user to acquire tomographic data from unbalanced source receiver combinations the number of sources the number of receivers This program matches each source with a receiver and then collects the waveform between the pair before moving to the next pair The other option of acquisition is for simultaneous acquisition 33 Wesley Johnson Hardware Design This option opens up a program that will simultaneously collect waveforms from 16 reception locations from a single source at one time This acquisition is a little faster and introduces less timing errors because 1 simultaneously acquires instead of individually acquiring the waveforms After the waveforms are acquired they are saved The last option in the main panel 15 to pick the arrival times When this option is selected a program opens that cross correlates the acquired waveforms with a reference waveform to determine the arrival time The correlated arrival times can then be cycled through to manually adjust any necessary picks A plot of time versus distance 15 plotted to help determine the fit of the correlation The final step of processing the data in LabVIEW is to format the data into the correct format for input into t
77. ley Johnson Literature Review sind sin8 2 7 nm where 0 angle of incidence 05 angle of refraction V velocity of material 1 V velocity of material 2 FIGURE 2 1 ILLUSTRATION OF SNELL S LAW ACOUSTIC WAVE REFRACTION AND REFLECTION AT A INTERFACE WITH DIFFERENT ACOUSTIC VELOCITIES The equation for Snell s Law 15 shown in Equation 2 7 and 15 illustrated Figure 2 1 1 2 1 1 ULTRASONIC SENSORS When selecting a transducer it 1s important to understand how an ultrasonic wave enters a material and how it travels through the material to the place of reception There are two methods of transmitting and receiving commonly used in ultrasonic applications These methods are the Pulse Echo and Through Transmission techniques The Pulse Echo technique uses one transducer that switches from active to passive to measure reflections within a material This method 1s commonly used in flaw detection to determine the depth to a flaw when the transmission length and material velocity are known The Through Transmission technique uses two transducers with one actively transmitting a wave and the other passively receiving the wave This method is used to determine material velocity through a cross section of the material Reflections can be measured using this technique to locate flaws When using either method it is important to know how a transducer will transmit and receive the wave When transmitting a wave
78. ndentation test was selected to test the 41 Wesley Johnson Calibration and Testing system for tomographic imaging of stress The purpose of the indentation test was to create a large stress contrast within a rock sample and capture the stress zone at different stress levels To be able to know the levels at which to acquire ultrasonic data index testing was conducted to find the average compressive strength ultrasonic velocity and elastic modulus for the rock being tested Two rock types sandstone and limestone were used to observe the ability of the UDAQ system to image the indentation stress and stress redistribution patterns for the rock samples Index properties for each rock type were found experimentally in the Virginia Tech Rock Mechanics Laboratory The properties tested were porosity density compressive strength and ultrasonic velocity Standard testing procedure was followed for these tests 33 34 Porosity was measured to determine the percentage of pore space within the rock A high porosity was favorable because it allowed for the coalescing processes to be potentially imaged Compressive strength was tested to determine the stress states at which tomographic imaging would be conducted Ultrasonic pressure and shear waves were passed through a sample to obtain the ultrasonic pressure p wave and shear s wave velocities This was done to be able to calculate the theoretical dynamic constants such as the Young s Modul
79. ndertaken due to cost constraints 24 If further geologic characterization 1s undertaken it usually requires boreholes to be drilled from the surface to intersect the ore deposit To reduce characterization costs ultrasonic tomography can be employed to detect hazards ahead of mining or within the mine workings Some hazards that can be detected are old mine workings structural changes in geology or high stress zones 24 To detect old mine workings reflection tomography can be employed Structural changes in geology are detected by measuring anisotropy 12 high resolution 3D tomography 13 or by noting velocity characteristic changes across a rock mass 14 20 Using these methods faults strata changes and some cases water content can be defined within a rock mass This information is very valuable in mine planning High stress detection is also an important application of ultrasonic tomography Field studies have been conducted successfully while excavating tunnels 19 excavating longwall panels 15 identifying burst prone faces 22 25 and measuring mining induced geological changes 16 Coupling tomography with other imaging techniques such as ground penetrating radar GPR can give an indication where surfaces are located in addition to knowing the stress at these interfaces 23 Another combination combines acoustic emission detection with ultrasonic tomography to monitor faulting or failure 27 This testing method has been use
80. ng The accuracy of the picks 15 very important due to the small scale that is used therefore great care was taken to ensure the picks were correct 3 7 LABVIEW PROGRAM FOR ULTRASONIC DATA ACQUISITION Tomographic data was acquired from ultrasonic sensors attached to the sample and was processed by a data acquisition system A simplified flowchart of the data acquisition system is shown in Figure 3 10 31 Wesley Johnson Hardware Design 9 19 19 101010 Koko 91010 ofall ol ol FIGURE 3 10 HARDWARE FLOWCHART The data acquisition system is controlled by a personal computer An ultrasonic pulse originates from a Panametrics 5077PR Ultrasonic Square Wave Pulser The pulser Control PC transmits an ultrasonic frequency that matches that of the material being measured at a repetition rate of 100 Hz This signal is relayed to one sensor on the sample through an ESG Ultrasonic Switchbox Pulser The switchbox relays are controlled digitally by a LabVIEW program The computer multiplexes the pulse through each of the predetermined source locations switching after the signal has been received by every receiving sensor Once the signal is relayed it enters a transmitting sensor on the rock sample inside of the load frame This signal is received by the reception sensors which are passively waiting for the ultrasonic signal The received data passes from the sensors
81. ocity psi MPa ft sec m s 5000 3 447 6750 3 881 The average velocity represents the average velocity along each raypath on which an arrival time was picked The tomograms in Figure 4 11 show the velocity levels across the sample for different stress levels The scale of the velocity for the tomograms 15 the same for all tomograms and the units are in ft sec The background velocity for the different stress levels seems to stay the same A transition in stress state can be seen at a load of 34 MPa At this level the stress distribution 15 more scattered than in the previous 46 Wesley Johnson Calibration and Testing two tomograms 17 and 25 MPa The previous tomograms show an anomalous presence which corresponds to the indentation load applied to the sample The tomogram at 34 MPa shows a wider scatter of stress and an increase in velocity the indentation area which could correspond to the pore spaces being completely closed and the rock beginning to be elastically loaded The final tomogram shows an additional increase in velocity directly under the indentation area The sample failed after the last tomogram and no other data were collected 47 Wesley Johnson Calibration and Testing 20 015 1 105 1 n an 11 I5 FIGURE 4 11 BEREA SANDSTONE TOMOGRAM INITIAL STATE 0 MPA 0 15 20000 0 1 19000 18000 17000 0 05 16000 15000 14000 13000 12000 11000 100
82. on of transmitted energy into the rock A perfect match between the resonant frequency of the sensor and the rock would result in the maximum energy transfer of the ultrasonic signal through the rock specimen The other consideration at the interface of the sensor and the rock involves acoustic impedance Acoustic impedance 15 the opposition to the flow of sound through a material 40 The effect of acoustic impedance can be minimized by using a couplant that has impedance close to the impedance of the rock The typical frequency range for ultrasonic testing in rock 15 100 kHz to IMHz The sensors used in this acquisition system were selected based on the criteria above 3 2 SENSOR MOUNTING A couplant is required to mount sensors to rock core samples because ultrasonic contact acoustic emission sensors were selected The sensors will not collect ultrasonic signals through the air so any air pockets between the sensor and the material must be eliminated by using a couplant Common couplants used for ultrasonic testing consist of viscous oils or epoxy adhesives These couplants are very effective on flat surfaces but on curved surfaces gaps between the sensor and the rock surface can exist Therefore beeswax was used to mount sensors to cylindrical samples This material 1s appropriate because it molds to the curved surface of the rock core and also fills any gap between the sensor and the sample due to the curved surface The effective use of be
83. or all transducers except internal and external trigger that maximum is limited to 2000Hz for 0 5MHz transducers 1000Hz for 0 25MHz transducers and 500Hz for 0 1 MHz transducers Pulse Rise and Fall Time 2015 max 10 90 Typically 10nS minimum pulse voltage and no load Sync Signal Output 3 0V into 50 ohms Pulse duration 0 5 10nS Capable of driving up to 26 standard TTL Loads 2 4V Sync Out precedes leading edge of Main Bang by 30 6015 External Trigger Positive edge 2 4V minimum Min pulse duration 50nS Input is AC coupled through 1000pF into 10K ohms Internal delay between External Trigger and leading edge of Main Bang is approx 215 Mode Pulse echo or Thru Transmission selectable Operating Temperature 0 to 50 The ultrasonic pulser 15 a versatile device that has adjustable pulse width and amplitude in addition to different triggering options The triggering features were not utilized for this system but are useful for other ultrasonic testing applications 3 4 ULTRASONIC SWITCHBOX To move the source pulse from one location to another an ultrasonic switchbox was used The ultrasonic switchbox pulser device was manufactured by Engineering Seismology Group ESG Solutions Figure 3 4 25 Wesley Johnson Hardware Design FIGURE 3 4 ESG ULTRASONIC SWITCHBOX PULSER This device has 16 relay channels with high voltage and trigger inputs An ISA control card was included with this dev
84. ous program The first step is to load in the source and receiver location files by clicking on the Load S R Locations to Tables button This activates a file dialog to open the files The user then sets the properties for the scope by clicking on the Scope Properties button This brings up a dialog where the scope parameters can be set The final step in acquisition is to click the acquire waveforms button This will start the acquisition of ultrasonic waveforms for every combination of source and receiver sensors The waveforms can be viewed by scrolling through the collected waveforms table in the upper right corner and then clicking the view waveform button To save the ultrasonic waveforms click on the output waveforms button This will open a save file dialog The program is stopped by clicking on the continue button 91 Wesley Johnson Appendix B LabVIEW User s Manual The block diagram for this VI is very similar to the previous acquisition program The program again uses event structure program to divide the functions in the program and make reading the block diagram easier 8 5 ARRIVAL TIME CORRELATION PROGRAM The final step in ultrasonic data acquisition for tomographic imaging 15 to pick the arrival times for the acquired waveforms The waveform file created 1n the acquisition step is used here to determine the ultrasonic arrival time This data processing step is the most important step in tomographic acquisition The accura
85. ow was painted on each block to indicate the direction of the roof so that samples cut from these blocks could be oriented with respect to the rib they were pulled from The blocks were pulled from pillars in the 12 East mains at a depth of 2200ft The mineralogy of the limestone deposit 15 99 7 pure calcium carbonate with some impurities Mg S102 When the blocks were collected in the mine the vertical direction of the blocks was marked in paint This orientation was important to note so that each block could have core samples cut in the same vertical orientation as the pillars This ensured that the 50 Wesley Johnson Calibration and Testing samples in the laboratory were tested in the same orientation as in the mine A summary of the index properties are shown Table 4 4 TABLE 4 4 FIVE OAKS LIMESTONE INDEX PROPERTIES Metric Units English Units P Velocity m s ft sec 6 65E 03 2 88E 02 2 18 04 9 43E 02 S Velocity m s ft sec 2 5E 03 1 31 02 9 01E 03 4 30 02 Effective Porosity 0 0 o 0 Tensile Strength MPa 9 79 00 1 42E 03 psi Compressive Strength 1 64 02 4 35E 01 2 38E 04 6 31 03 MPa psi Elastic Modulus MPa psi 3 51E 04 4 43 03 5 09E 06 6 42E 05 psi Not enough samples to calculate the standard deviation From Table 4 4 the P Wave velocity was found to be 6 650 m s From testing the frequency of a test waveform the
86. pplications in Global Seismology and Exploration Geophysics Boston MA D Reidel Publishing Co 1987 6 Terada M and T Yanagidani Application of Ultrasonic Computer Tomography to Rock Mechanics Ultrasonic Spectroscopy and its applications to material science 1986 pp 205 210 7 Trampert J and J J Leveque Simultaneous Iterative Reconstruction Technique Physical Interpretation Based on the Generalized Least Squares Solution Journal of Geophysical Research Vol 95 No B8 pp 12 553 12 559 1990 8 Daily W and E Owen Cross Borehole resistivity tomography Geophysics Vol 56 No 8 1228 1235 1990 9 Dines K A and R J Lytle Computerized Geophysical Tomography Proceedings of the IEEE Vol 67 No 7 pp 1065 1073 1979 10 Beard M D and M J S Lowe Non destructive testing of rock bolts using guided ultrasonic waves International Journal of Rock Mechanics and Mining Sciences 40 pp 527 536 2003 11 Molyneux J B and D R Schmitt First break timing Arrival onset by direct correlation Geophysics Vol 64 No 5 pp 1492 1501 1999 12 Song L H Liu S Chun Z Song and S Zhang Mapping an underground rock mass by anisotropic acoustical transmission tomography Ultrasonics 52 pp 1009 1012 1998 13 Marti D R Carbonell A Tryggvason J Escuder and A P rez Esta n Calibrating 3D tomograms of a granitic pluton Geophysical Research Letters 29 No 17 pp 15 1
87. r of sensors specified equally spaced around the sample The table showing Receiver Locations will then fill up with all of the sensor coordinates The Source Locations table will remain empty until a sensor 15 selected to be a source To select a sensor to become a source click on the desired sensor and the color of that sensor will change from blue to red indicating the change The coordinate for this sensor will then move from the Receiver Locations table to the Source Locations table To output these coordinates click on 62 Wesley Johnson Appendix B LabVIEW User s Manual Output Coordinate Files This action will save both tables to a text file specified by the user To add another plane to the coordinate files previously saved reinitialize the drawing with a new Z Level value Once this is done select the sensors that are to be sources Click on the Click to Append So File and Click to Append Rec File buttons to append both the source and receiver files that were previously saved If this is not done then the files will be rewritten with whatever data is in the tables for Source Locations and Receiver Locations To stop this VI and return to the Main Panel click on the Continue button To determine the number of sensors to use in creating a tomogram a utility has been incorporated into this VI The utility is called by clicking the Calculate Pixel Size button This calls a program that calculates the pixel size of a 2D tomogram
88. raded as technology advances with minimal reprogramming The precision in locating anomalies within tomograms was dependant on the number of sensors the sensor locations and the algorithm used to calculate the tomogram Increasing the number of sensors used for tomographic acquisition increases the raypath coverage across the sample The locations of the sensors are another factor in the precision of the acquisition system Complete coverage around the body being imaged allows for undistorted representation 1n the tomogram The algorithm used 15 important because it determines how each pixel within a tomogram 15 calculated Errors within the algorithm will result in erroneous tomograms The accuracy of the data acquisition system 1s comparable to other commercial acquisition systems The main disadvantage of this system 15 the 8 bit resolution of the ADC cards This could be improved by investing higher resolution cards The error of measurements was determined experimentally to be 0 5 when compared to an ultrasonic P wave test unit in the Virginia Tech Rock Mechanics Laboratory and 3 8 when compared to ultrasonic test conducted in the literature The costs of the components in this system are summarized in Table 5 1 63 Wesley Johnson Conclusions TABLE 5 1 SYSTEM COST Device Cost Unit Total 5102 National Instruments 1 500 15 000 Oscilloscopes PXI 1006 National Instruments
89. re needed to design the tomographic test The procedure for capturing the index properties for the Five Oaks Limestone 1s discussed in detail below The same procedure was done to assess the index properties for the Berea Sandstone 4 1 CALIBRATION TESTING In polymer classification testing PMMA is used as a reference for calibration of ultrasonic transducers 37 38 The material exhibits an acoustoelastic effect when subject to a load 37 This material was used for calibration and initial application testing of the UDAQ system A test was conducted on two Plexiglas plates to calibrate the ultrasonic test system and to verify tomograms created from the ultrasonic data One plate was solid and a second plate had a hole drilled in the lower left quadrant of the 35 Wesley Johnson Calibration and Testing plate Both plates had the same dimensions in length width and height of 10 x 10 x 2 cm Figure 4 1 shows a scale picture of the plate with a hole plate 2 FIGURE 4 1 SCALED IMAGE OF THE PLEXIGLAS PLATE WITH A HOLE PLATE 2 A total of twelve sources and sixteen receivers were evenly placed around each plate The drawing in Figure 4 2 shows the source and receiver locations as well as the ultrasonic raypaths between source and receiver pairs view Isometric View EB Recewing Sensor Raypath Source Sensor Plexiglas Sarnple Tomographic Setup Unts centimeters FIGURE 4 2 PLEXIGLAS PLATE TOMOGRAP
90. rence waveform and the correlation between the two plots 15 compared A high correlation indicates a high probability that the sections of the waveforms are the same This method of arrival time picking is explored in depth by Molyneux and Schmitt 11 Once the arrival time is picked the velocity of the wave can be calculated The velocity between 16 Wesley Johnson Literature Review one source and receiver pair represents an average velocity along that raypath By conducting a tomographic survey where multiple raypaths are measured at different orientations across the sample many velocities along the different raypaths can be found These velocities are then inverted to produce a velocity profile along the plane being measured The algorithms used for ultrasonic tomography have improved since the first application of geophysical tomography by Dines and Lytle Two common algorithms used for tomographic inversion are the algebraic reconstructive technique ART and the simultaneous iterative reconstructive technique SIRT The ART method is used for linear models This method inverts all of the data at once to form a tomographic representation of the data This method is simpler than the SIRT method however if a large data set 15 used it requires large amounts of computer memory and processing time to conduct the transformation The SIRT method iteratively inverts rows of data that improve the initial inversion the iterations continue un
91. s the load increases This scatter is most likely due to fractures within the sample as 1 approaches failure The fracturing of the block from the indentation load can be clearly seen in Figure 4 14 56 Wesley Johnson Calibration and Testing a b FIGURE 4 14 FIVE OAKS LIMESTONE TEST BLOCK A PRIOR AND B POST FAILURE The sensors were attached to the limestone using a cynoacrylate adhesive The flat surfaces allowed for this application method of the sensors Tomograms were calculated from the arrival time data using GeoTomCG to get a clear picture of what was happening inside of the limestone as stress increased The tomograms for this sample are shown in Figures 4 15a 4 15h 57 Wesley Johnson Calibration and Testing 0 5 045 0 4 035 0 3 0 25 0 2 0 15 0 1 0 05 0 0 0 05 0 1 0 15 0 2 0 25 0 3 0 35 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM PRIOR TO LOADING 0 MPA UNITS FT SEC 0 5 0 45 40000 38000 36000 34000 32000 0 35 30000 28000 26000 24000 22000 0 25 20000 18000 02 16000 14000 12000 ae 10000 8000 0 1 6000 0 05 0 0 005 01 015 02 025 03 055 FIGURE 4 15 FIVE OAKS LIMESTONE TOMOGRAM 17 24 MPA UNITS FT SEC 40000 38000 36000 34000 32000 30000 28000 26000 24000 22000 20000 18000 16000 14000 12000 10000 8000 6000 58 Wesley Johnson Calibration and Testing 0 5 0 45 0 4 0 35 0 3 0 25 0 2 0 1
92. t be connected 1 3 4 5 correctly The personal computer 1 controls the acquisition of the ultrasonic signals from the PXI 1006 chassis 8 and also controls the ESG Ultrasonic Switchbox Pulser 6 A fiber optic cable connects the chassis to the computer The Ultrasonic Switchbox Pulser 6 is connected through a Digital Input Output Acquisition Card The card is connected via a cable to a screw terminal connector block 2 The wires from the connector block are connected to a 25 pin serial dongle The dongle 15 wired according 97 Wesley Johnson Appendix C Hardware Wiring Instructions to the channel specifications of the Switchbox 6 and the power requirements for the switchbox The power for the switchbox is provided by an Aglient Technologies Power Supply 3 The voltage requirement for the switchbox is 9V The wired dongle with the channel relay and power requirements 1s connected to a serial cable that connects to the Switchbox 6 The ultrasonic pulse 1s supplied by a Panametrics 5077PR Ultrasonic Square Wave Pulser 4 This device outputs a 400V pulse that 1s relayed to source locations via the Switchbox 6 A BNC T 5 connector is used to split the output signal from the 5077PR 4 One side of the T 5 is connected to the input terminal of the Switchbox 6 and the other is connected to the input of the Voltage Divider Circuit 7 The voltage divider circuit 7 drops the voltage from the 5077PR 4 to
93. til all data has been processed This method is useful for large data sets because the amount of computer memory required is significantly less than the amount required when using the ART method A comprehensive review of both methods is found in the literature 5 7 2 4 1 APPLICATIONS OF TOMOGRAPHY Tomography has been used widely for geophysical applications such as earthquake tomography three dimensional structural imaging geologic hazard identification and stress identification For large structures seismic tomography 15 commonly used because seismic waves can travel longer distances than an ultrasonic waves The theory behind seismic tomography is comparable to that of ultrasonic tomography Field studies using seismic tomography have been conducted in mines to characterize underground rock masses 12 and map underground structures 13 14 Tomography has also been used in mining to monitor for roof or pillar failure identify regions of relatively higher stress and strata identification The use of these mine wide monitoring systems can increase miner safety and also help maintain production by preventing potential roof failure I Wesley Johnson Literature Review Hazard detection is one way in which ultrasonic tomography can increase miner safety In active mining sections structural analysis can be done by observing the roof rib and floor surfaces Further geologic characterization and hazard recognition efforts are not u
94. to be controlled in a LabVIEW environment along with the rest of the system This 1s favorable because it allows for the user to manipulate one program instead of two or more 2 Wesley Johnson Hardware Design 3 5 DATA ACQUISITION INSTRUMENTATION After the sensors had been mounted to the sample using a couplant the ultrasonic pulse could pass through the rock to a reception location The receiving sensor collected the analog waveform and passed it into the analog to digital converters ADC National Instruments PXI 5102 Digital Oscilloscopes were used as the ADC for this system These cards were mounted into a National Instruments PXI 1006 PCI Extension Interface Chassis Figure 3 7 IONAL ENTS RUMEN FIGURE 3 7 PXI 1006 CHASSIS WITH PXI 5102 DIGITAL OSCILLOSCOPES This chassis has advanced timing and synchronization capabilities through the Real Time Signal Interface RTSI bus that connects all the instrument slots in the chassis chassis 15 connected to the control computer using National Instruments 3 technology The MXI 3 control consists of a PCI card that is installed on the control computer A similar control card is connected to the control slot in the PXI 1006 chassis A fiber optic cable connects the two control cards The use of fiber optics to transfer data allows for high data transfer rates without delaying acquisition on any of the instrument cards in the PXI chassis The PXI
95. to find the ultrasonic velocities 15 shown in Figure 4 6 FIGURE 4 6 VIRGINIA TECH ROCK MECHANICS LABORATORY ULTRASONIC VELOCITY VIEWER Using the Virginia Tech Rock Mechanics Laboratory Ultrasonic Velocity Viewer 10 velocity measurements were made on Plexiglas samples of varying length to calculate 38 Wesley Johnson Calibration and Testing an average ultrasonic velocity The transducer frequency used on the ultrasonic velocity viewer was IMHz The same test frequency was used in Plexiglas testing using the UDAQ system Table 4 1 shows the average velocity values the standard deviation and the comparable error of the measurements TABLE 4 1 ULTRASONIC DATA ACQUISITION SYSTEM VELOCITY CALIBRATION MEASUREMENTS Velocity Standard Error in sec m s in sec Rock Lab 84 Deviation m s m recta 9050 280 188 4620 o Table velocity represents an average from 36 39 From Table 4 1 it can be seen that the velocity measurements from ultrasonic tomographic data collection system in grey are accurate Knowing that the velocity measurements from the system are accurate some observations of the measurements shown in Figures 4 4 and 4 5 can be made The two plots of time vs distance are not the same Figure 4 4 shows the plot for Plate 1 This plot shows a tight linear trend with a high statistical correlation to the linear trend 0 99 It is expected that the tomogram for this data set will be fa
96. tructions 910 10 0 89 10 10 10 00000000 FIGURE ULTRASONIC DATA ACQUISITION SYSTEM HARDWARE WIRING FLOWCHART 96 Wesley Johnson Appendix C Hardware Wiring Instructions The wiring of the instrumentation components are shown in Figure C 1 this figure is used as a reference for the wiring descriptions given in this appendix The components are numbered 1 8 The description of each component 15 given Table TABLE 1 HARDWARE DESCRIPTIONS Computer Control Contains a PCI Control card for the PXI Chassis and a PCI Digital I O ACQ card for the Ultrasonic Switchbox Pulser Control Digital I O Connector Block with 25 pin serial dongle for connection to the Ultrasonic Switchbox Pulser a Aglient Technologies Power Supply 2 provides power for Ultrasonic Switchbox Pulser Panametrics 5077PR Ultrasonic Square Wave Pulser generates the ultrasonic source pulse BNC T Connector splits the output pulse from the 5077 ESG Ultrasonic Switchbox Pulser relays the ultrasonic pulse to different sensors 7 Voltage Divider Breadboard Circuit drops the high voltage pulse to a level that can be input into a PXI 5102 acquisition card PXI 1006 Chassis contains PXI 5102 digital oscilloscopes for acquiring ultrasonic signals To use the Ultrasonic Data Acquisition System all the components mus
97. umerical control will produce a plot of arrival times versus distance in the right hand graph window The linearity of this curve is dependant on the quality of picking Once all arrival times have been picked for each waveform the arrival times need to be formatted into the acceptable GeoTom format This is done by clicking on the Format to Geotom button The popup window from this call 1s shown in Figure B 13 geotomoutput vi Ele Edit Operate Tools Browse Window Help gt C Documents piesr FIL 0 003967 0 019990 FIGURE B 13 GEOTOM FORMAT FRONT PANEL 93 Wesley Johnson Appendix B LabVIEW User s Manual The input into Geotom assumes that the linear curve passes through the origin have this done the intercept must be corrected to pass through the origin Enter in the correction into the correction numerical control The play button must be pressed to make changes within this VI Once the curve is acceptable change the Save slide control towards the green indicator Click the play button and save the Geotom format 3dd To create a tomogram simply open the saved 3dd file in Geotom and follow the instructions in the Geotom manual to create a tomogram 94 Wesley Johnson Appendix C Hardware Wiring Instructions APPENDIX C HARDWARE WIRING INSTRUCTIONS 95 Wesley Johnson Appendix C Hardware Wiring Ins
98. us of Elasticity Poisson s Ratio and the Bulk Modulus using ultrasonic velocity and density values 4 2 1 Berea Sandstone Experiment Berea Sandstone 1s commonly used for laboratory testing because the variance between samples is smaller than that for other rock types The sandstone used in this experiment was obtained from Cleveland Quarries in Amhurst Ohio The index properties were tested in the Virginia Tech Rock Mechanics Laboratory by Dr Karfakis prior to this study 1993 and 1994 A summary of index properties for this Berea Sandstone 15 shown in Table 4 2 42 Wesley Johnson Calibration and Testing TABLE 4 2 BEREA SANDSTONE INDEX PROPERTIES Porosity n Density Specific Tensile Str Compressive Compressive Vel g cm J Ib f Gravity G MPa psi Str m s ft sec St St St St Dev Dev Dev Dev Dev 1 5 081 81 2 2 35 2 292 10 3 57 57 0 37 37 6 26 26 0 36 36 2 2 200 s wer w se or 96 A testing frequency of 300 kHz was used to image 10 16 x 2 54 cm 4 in diameter x height cylindrical sample The effective wavelength through the sample was estimated to be 0 73 cm 0 29 in The cylindrical platen with a diameter of 5 08 cm 2 in used the limestone experiment was used to apply an indentation load to the sandstone sample Thirty two sensors were equally spaced around the sandstone s

Download Pdf Manuals

image

Related Search

Related Contents

    SpectraLink PTB450 User's Manual  AFU 300/600 Instrucciones de Funcionamiento y  取扱説明書(PDF:316KB)  Casio EX-Z600 Digital Camera User Manual  « LES BD »  

Copyright © All rights reserved.
Failed to retrieve file