Home

development of an advanced automated nondestructive evaluation

1. Abort Scan 1 Scan 1 gt Figure 6 9 Code section of Autoscan VI The VI is far too large to show in its entirety but this section shows the structure used to perform a point by point scan Scanner 1 Loop Priority 3 Read instruments on open channel control line meal Data File Path Scanner 1 Read Mode _a 17 Instrument Runtime Settings Auto Scan Scanner 1 7 Run Line Scanner Master Slave Co ftinstrumenth M S Inst Offsets Grid Settings Configuration Cluster Line Number k Camera Settings Inst Slave 7 7 While ideally not necessary iis prevents the two from ting out of sync frLine Number k 2 Dooooooo a Doo Scanner 2 Loop Priority 3 Read instruments on open channel control line scanni Data File Path 2 3 Scanner 2 Read Mode CSN Instrument Runtime Settings 2 E5 Scanner Master Slave 2 AMS Inst Offsets Grid Settings 2 h Configuration Cluster Line Number k 2 L32 gt frLine Number k Scanner 1 HA Scan Complete E t A Connect to Instrument gt ftScanner 1 Read Mode DoooooooooN oo iile ideally not necessary gt ftLine Number k 2 Hoop ooooo Oooo ooo Scanner 2 Ds Scan Complete F A Connect to Instrument 2
2. Line Count Grid n E Line Orientation Figure 6 26 Code snippet for covermeter and laser data realization VI A two dimensional display or C Scan image of covermeter data can be visualized as a top down view of the scan area with lightly colored pixels representing areas of shallow cover and darker pixels representing areas of deeper cover Due to the output 136 format of the covermeter used in this research when no reinforcement is detected a value of zero is reported leading to pure white areas on the C Scan display in areas of infinite cover where one would expect black pixels This effect is shown in Figure 6 27 where the data display is indexed in point number X and line number Y format Bj 4 points Figure 6 27 Example of a 2D intensity plot covermeter data visualization in the ANTARES Processing Suite The three dimensional plot display shown in Figure 6 28 allows the user to visualize the overall characteristics of the scanned specimen in any desired view orientation While this is useful for developing a sense of the overall configuration of the 137 reinforcement within the structurethe two dimensional intensity plot allows for precise measurements by dragging a special cursor which indicates the X Y and Cover values of any point on
3. 57 00 Ss ee SA ee DP EIR Wet sieved concrete place at pour roughly 6x6x10 pris 79 0 UNLESS OTHERWISE SPECIFIED NAME DATE DIMENSIONS ARE IN INCHES DRAWN JDN 06 20 13 i J primer CHECKED TITLE MO MACE DECIMAL E T ENGAPPR ANTARES Voids and Flaws THREE PLACE DECIMAL MEG APPR Block INTERPRET GEOMETRIC QA PROPRIETARY AND CONFIDENTIAL TOLERANCING PER COMMENTS EEE pinata All dims nominal SIZE DWG NO REV same c Site A V amp F_block IT me waen eS E a applicable sre PROHIBITED APPLICATION DO NOT SCALE DRAWING SCALE 1 20 WEIGHT SHEET 1 OF 4 5 4 3 2 1 212 6 00 7 10 00 z e o O g Notes Prisms to be constructed by compacting concrete of same mix design with sand removed into 6inch _ beam mold and allowing to set before block casting UNLESS OTHERWISE SPECIFIED NAME DATE DIMENSIONS ARE IN INCHES DRAWN JDN 06 20 13 Detail la Ib TOLERANCES 4 FRACTIONAL CHECKED TITLE ANGULAR MACH BEND P e TWOPLACEDECMAL ENG APPR ISC Prisms 3 and 6 inch THREE PLACE DECIMAL MFG APPR INTERPRET GEOMETRIC Q A oneee a couar AE EO SIZE DWG NO REV lt INSERT COMPANY NAME HERE gt ANY MERONEN PANT ORAS AWIKA py o A V amp F_block ua aoe APPLICATION DO NOT SCALE DRAWING SCALE 1 20 WEIGHT SHEET 2 OF 4 5 4 3 2 1 213 3 00 p 10 00 _ Notes Form triangular wedges from 1 16in thick steel sheet
4. 20 Cracking due to stress plastic shrinkage cracking curling and alkali silica KEE Pere re Ore nen TE erent ne TIAA nar nee erate eer E E A ter erent 24 Examples of SUR defoClS seriinin inania en ea a 24 National Instruments Smart Camera mounted behind the wheel of the Heavy Vehicle Simulator at the FDOT State Materials Office eeeeeeeeeeeees 25 Single point distance measurement using a laser distance SensoOr 006 28 A common covermeter with digital display and handheld instrument head 30 Example of the effect of orientation on covermeter measurement 005 32 Finite element simulation of an impact on a plate like structure showing the three types of induced stress wave ssssssssssreeesesrrrrrrnnrtresortrttrnnntnreeerrrren nnen 33 Modern impact echo system with hardened steel ball impactors 0 35 Test setup for measuring P Wave velocity using impact echo ASTM C 1383 36 Test setup for measuring concrete plate thickness using impact echo ASTM Fe a a pathaedaoiadapacinedp emia easieadiieteweaebiapeanels 36 Time domain and frequency domain data collected from an impact echo test of a 250mm thick concret atari tele err ere 37 Operating principle of a piezoelectric pulse wave ultrasound transducer 39 Test setup for standard UPV test in direct transmission mode cceeeeeeees 40 Operating principle of a shear wave ultrasound tr
5. Figure 7 9 Ultrasound scan of deep cover 153mm side of block showing indistinguishable wave reflections from all four ducts 159 Post grouting NDE scans Duct 1 the leftmost PT duct when facing the deep cover side of the Asymmetric Internal PT Block was filled with PT grout 51 days after fabrication of the block 56 days after placement of grout in the middle two ducts The grout was mixed using a high shear paddle mixer and pumped directly into the duct from the bottom up using a handheld transfer pump so as to prevent entrapped air This step was taken to produce an ideal duct one that was grouted undisturbed in situ with a tightly controlled mixing and pumping process Once the grout hardened it was used to compare instrument readings with the other ducts both to attempt to better define detection limits on poor grouting and to evaluate the success of the method by which the inclined void ducts were fabricated By this time ANTARES SVW was in its beta version and more advanced data visualizations could be realized As shown in Figure 7 10 scans of the block using ultrasound produced mixed results Careful inspection of the tomograph of the ducts revealed that in situ grouting duct 1 produced a better interface for transmission of the ultrasonic shear wave resulting in fewer reflections beyond the duct structure The more complete reflections of the pre grouted ducts 2 and 3 and ungrouted duct 4 ducts resulted in
6. ler Program Version 1 0b Ultrasound Filtering Display Control Original fa ic Wave Velocity 1600 m s Minimum Depth 50 mm m ica Maximum Depth 230 mm E 20000 Hz fone Figure 6 34 Ultrasound processing pane with trimming and filtering options enabled 145 The final step in the ultrasound processing VI is to apply a data transform to aid in visualization of the data plots The selection of a data transform depends on the operator s preference but three options are available 1 Original Trimmed filtered data is presented as calculated 2 Absolute Waveform amplitude intensity is presented as absolute values This eliminates some confusion in areas where the wave reflection oscillates rapidly and produces alternating light and dark layers on the intensity plots 3 Hilbert Applies a Hilbert transform to each waveform eliminating superfluous negative frequency content and emphasizing edge detection by creating a display that is more sensitive to change in relative intensities Ultra Scan Depth 54 mm Ultra Scan Depth 54 mm 50 gt j p 1 I g g g g t g g g p E I 1 1 g g g 1 j 1 1 p 1 ou 5 W 15 20 25 30 35 0 45 50 55 Of 65 70 7 5 10 15 2 2 30 35 4 45 50 55 6 65 70 Line Line Ultra Scan Depth 54mm 0 7 1 j t g g p g i p p p p uo oO s 1 15 20 25 30 5 40 45 50 55 60 Gu T Line F
7. One of the main features required from the ANTARES SVW software was to comply with State of Florida s Sunshine Law allowing public access to official government communications as well as FDOT s policy of open communication and data 129 sharing As a government agency FDOT was not only amicable but indeed legally bound in many cases to share its research and evaluation data with other agencies ANTARES was designed to be a research tool not only for FDOT but any agency that wished to collaborate in future research and development efforts Since ANTARES SVW was designed to accept and output several different types of files that may be read interpreted and modified by third parties it was required to produce a series of file I O subroutines that produced and interpreted data in a well defined and clearly formatted manner It was decided to build ANTARES SVW around three file types a master configuration file ANTCFG that would contain default settings for the scanners and instruments a grid file ANTGRD that could be used to save and load the grids upon which ANTARES would perform NDE scans and an open source text based universal data file ANTDAT that could be used with all the native instruments and would contain all the information to analyze an NDE scan of a given grid ANTARES Configuration File ANTCFG The ANTARES Configuration File ANTCFG is a binary file containing all vital settings necessary to operate th
8. Scanner to Control Loup T Scanner Port Cluster EZ geeeereeeereeerneeenneeenneeeneeeenneeeneeenneeenneeeneeeneeeneeeneeenneeenneeneeneeneeeneeneeeneenneeeneeeneenennennenneenenn a Scanner 1 Defa p a DAQm x Device P ActiveDev Figure 6 4 Command section of pneumatic instrument head command subroutine REI 2013 Scanner to Control fa Scanner 1 Defa vP o Scanner Port Cluster n DAQm x Device n ActiveDev DI Ports Sensor Head Status D 4 inet Digital Input Digital 1D Bool fone channel for each line gt enn W Figure 6 5 Pneumatic instrument head query subroutine with integral error handling The SubVIs responsible for interfacing with the NDE instruments and performing low level command and acquisition functions must be as unique as the instruments themselves Despite this incorporating new instruments into the ANTARES SVW system requires only writing of a new low level SubVI such as the EyYeCON subroutine shown in Figure 6 6 for that instrument and minor modification to the higher level handling VIs to allow use of the low level subroutine This simplicity is owed to the compartmentalized nature of LabVIEW and the architecture of the ANTARES SVW program 112 os aa sooo0o0o0naaoaaanaaaana aoaaa aaoo aaan aoaaa
9. instrument heads This mounting system included provisions for a new universal data and power connection scheme for current and future NDE instruments The next phase of the connectivity improvements was to establish a standardized and simplified methodology for communicating with the scanner systems and their attached NDE instruments This communication standard needed to be simple rugged and flexible enough to support current and future NDE instruments and was designed accordingly Finally concurrently with the previous two phases the entire system was readdressed for field survivability and modified using weather resistant electrical and data connections to ensure reliable operation outside a controlled laboratory environment 82 NDE Instrument Mounting System The original instrument mount of the Phase system was not only cumbersome but in fact posed hazards to the NDE instruments themselves as shown in Figure 4 1 Conversely the improved NDE instrument mounting system would not only have to be capable of quickly mounting and protecting the delicate instruments already in use with the automated NDE scanners it would have to be easily and inexpensively adaptable to new instruments and would have to take advantage of the very limited space available on the scanners instrument heads i inate Figure 4 1 Covermeter mounted on the Phase scanner system showing exposed and entangled data and power cables circled in
10. B Ultrasound transducer array Standard pneumatic head mounting bracket a Ultrasound operation unit 4 Standard instrument mounting palette LAN USB device server Figure 4 9 Germann Instruments EyeCON adapted to palette system No physical modifications were made to the instrument Photo by J Nelson 92 Figure 4 10 Germann Instruments EYeCON assembled on palette and ready for installation onto automated scanner Photo by J Nelson Though initially only two instruments were converted to the new LAN protocol new instruments were added to the system using a similar methodology Indeed the LAN protocol is capable of supporting almost any data communication type using servers in the case of digital communication busses and Ethernet DAQ devices in the case of analog devices 93 CHAPTER 5 INCORPORATION OF NEW INSTRUMENTS Approach With the new instrument mounting and communication protocol established with the NDE instruments from the Phase system and at the request of FDOT three additional instruments were incorporated with the automated NDE scanning system Ground penetrating radar GPR is a technique that has seen a sharp increase in interest from the structural concrete community in recent years and was the first and arguably most important new method to be addressed A GPR system was purchased for the Phase NDE system but never successfully automated due to its bulky supporting equipment lack of d
11. Independent loops on main program used to call Autoscan VI cceeeee 116 Emergency Stop VI with integrated reset dialog DOX ssseeeeeeeteeeeees 117 Initialization phase of main iNterfAaCe ccceeceeeeeeeeesteeeesteeeeesteeetsteeessteeeee 119 Human interface loop of ANTARES SV o NN EEIN IAEI INT FANTE EAA eae Clean up phase of the main Vl s sssssseseenssseersrsserersrsrnrnrsrsrernrrersrsrserersnereeee 120 The first four sequences of the Splash Screen Vl cccccescecessecessseeeessteeeeseee 121 Main loop of gen oe Screen eres the acai ANTCFG NR rcs a cess eae E ead EE AT sas 122 Grid Generation Wizard main loop with state machine architecture 123 ANTARES SVW splash screen with loading bar and feedback to operator 124 Configuration Settings Screen on Scanner Configuration tab 126 Configuration Settings Screen modifying a native instrument s default port and scanning operation settings stars acces resect uedsaaercracacmaeesentdnaees BAO Main interface of ANTARES SWW during a shear wave ultrasound scan 127 Grid Generation Wizard on one of the final ae to eaiilaaapail of a functional ET E E E E E E E aair adea Wee Grid configuration cluster as it appears in the Grid Generation Wizard 131 Excerpt of an ANTARES Data File sien AT showing generic file format and standard header inform h i 13 6 25 6 26 6 27 6 28
12. Ralls Accelerated Bridge Construction Designing for Contractors STRUCTURE no 9 p 5 2009 39 H Wiggenhauser D Streicher and M Friese Automated NDE of PT Concrete Structures Berlin Germany 221 40 E Schmidt Method and Apparatus for Testing Hardness of Specimens United States of America Patent 3 879 982 8 March 1974 41 S Mindess J F Young and D Darwin Concrete 2nd Ed Upper Saddle River NJ Prentice Hall 2003 42 J C Liu M L Sue and K Chang Huan Estimating the Strength of Concrete Using Surface Rebound Value and Design Parameters of Concrete Material Tamkang Journal of Science and Engineering vol 12 no 1 pp 1 7 2009 43 American Concrete Institute Structural Building Code ACI 318 11 and Commentary Section 8 5 Modulus of Elasticity Detroit American Concrete Institute 2011 44 International Atomic Energy Agency Guidebook on On Destructive Testing of Concrete Structures Vienna Austria IAEA 2002 45 American Society for Testing and Materials ASTM E1876 09 Standard Test Method for Dynamic Young s Modulus Shear Modulus and Poisson s Ratio by Impulse Excitation of Vibration West Conshohocken PA ASTM International 2009 46 American Society for Testing and Materials ASTM C805 13 Standard Test Method for Rebound Number of Hardened Concrete West Conshohocken PA ASTM International 2013 47 American Society for Testing and Materials ASTM C803 03 Standar
13. size of the instrument A small enclosure containing a DC power supply and voltage regulator was designed and constructed to provide the instrument with the requisite 12V to operate and the serial device server mounted on the palette with 5V as shown in Figure 5 5 A cable was constructed to connect the instrument to the serial device server on the palette and the components were assembled and stored in a watertight instrument case until needed 99 Laser Palette Component LAN serial device server 2 Power conversion and data connection enclosure 3 ODSL8 instrument on miniature pneumatic head instrument mount Figure 5 5 Laser profilometer system mounted on palette system and placed in watertight instrument case Photo by J Nelson lew k k U j PIR ME Figure 5 6 Laser profilometer mounted on scanner instrument head Photo by J Nelson 100 Surface Condition Surveyor The Surface Condition Surveyor SCS instrument was an idea borne out of observation of another FDOT SMO program The Accelerated Pavement Testing program had been using a camera system with LabVIEW vision processing support to detect cracks and pavement defects for years 4 and including a similar functionality with this automated NDE system for structural concrete was deemed prudent A National Instruments NI1742 Smart Camera was purchased for the project along with a ring light array that could be driven from the camera itself The
14. An all digital lossless universal communication protocol capable of cable lengths up to 100m TIA EIA 568 5 A without signal extenders indefinite cable length with signal extenders Multiple devices may be connected to a single inexpensive network hub Ruggedized weather resistant cables and connectors are available off the shelf Support for data sharing using publisher subscriber protocol PSP Easy integration with LabVIEW developed software Standard inexpensive readily available manufactured patch cables Compatibility with most computers 86 Direct compatibility with many available NDE instruments and DAQ devices Indirect compatibility with USB serial and analog voltage data types using off the shelf components Modification to Scanners In order to fully implement a LAN protocol the scanners themselves were first modified for an all digital interface Each scanner s X and Y movement is accomplished using serial command operated Cool Muscle stepper motors The stepper motors have onboard serial servers and can be daisy chain operated the Y axis stepper motor is connected directly to the X axis motor which receives serial commands for both motors The pneumatic cylinder which applies contact based NDE instruments to the concrete surface was driven by an electromechanical 24V relay which was operated remotely from the PXI s onboard analog output channels To integrate these systems with t
15. M H Ansley Structures Research Center in Tallahassee FI Photo courtesy of Google Earth Transportation of the System Moving an ANTARES scanner presented a significant logistical challenge The nearly 4x3m 180kg scanner was unbolted from its clamping and mounting framed as shown in Figure 8 4 laid down horizontally and manually carried onto a flatbed trailer as shown in Figure 8 3 The operation is feasible with four people but much safer with six 186 Figure 8 3 ANTARES Scanner 1 fastened to an open frame flatbed trailer outside of the FDOT State Materials Office in Gainesville Fl Photo by J Nelson Figure 8 4 Laboratory block clamping frame with ANTARES Scanner 1 removed for field transport Scanner 2 is still attached to the frame and could be operated in the laboratory independently of Scanner 1 Photo by J Nelson 187 Once the scanner was placed on wooden supports and strapped down to the open frame flatbed trailer numerous cardboard and plastic elements were used to protect the delicate components of the scanner from road debris The entire scanner was then wrapped in a waterproof cover for transport The field deployment began on Wednesday July 31 2013 at 6 00am and the 258km drive to the Ansley center took approximately 3 hours Ansley center personnel helped move the scanner to its first scan location and set up the required electricity compressed air and shade The scanner was laid directly
16. M Moore B Graybeal and G Washer A comparison of nondestructive evaluation methods for bridge deck assessment vol 36 2002 29 Parsons Transportation Group Grouting of Bridge Post Tensioning Tendons Training Manual Tallahassee Fl State of Florida Department of Transportation 2002 30 M Krause Localization of grouting faults in post tensioned concrete structures in Non Destructive Assessment of Concrete Structres Reliability and Limites of Single and Combined Techniques vol 1 Springer 2012 31 M Colombo and R Felicetti New NDT techniques for the assessment of fire damaged concrete structures vol 42 2007 32 J Heller The Day the Skyway Fell St Petersburg Times 7 May 2000 33 H Wiggenhauser D Streicher and M Friese Automated NDE of PT Concrete Structures BAM Federal Institute for Materials Research and Testing Berlin Germany 34 Leuze electronic Optical Distance Sensors ODSL 8 ODSL 30 ODS 96 35 J E Lucius and M H Powers GPR Data Processing Computer Software for the PC U S Geological Survey Denver CO 2002 36 D Clayton C Smith C C Ferraro J Nelson L Knazanovich K Hoegh S Chintakunta J Popovics H Choi and S Ham Evaluation of Ultrasonic Techniques on Concrete Structures Oak Ridge National Laboratory 2013 37 D Clayton and C Smith Summary of Large Concrete Samples Oak Ridge National Laboratory 2013 38 J McMinimee and M L
17. Simulator shown in Figure 2 3 This equipment is used in the Accelerated Pavement Testing program to detect cracks generated on a pavement surface during operation of the machine Using a similar camera and software arrangement on structural concrete one could detect small surface voids bugholes cracks honeycombing and surface deterioration due to sulfate attack alkali silica reaction and erosion Figure 2 3 National Instruments Smart Camera mounted behind the wheel of the Heavy Vehicle Simulator at the FDOT State Materials Office The software automatically detects and measures large cracks in the wake of the loading wheel Photo courtesy of FDOT State Materials Office 4 Image processing is often a difficult task to automate due to among other things the overall density of the data In the eye of a computer a digital image is nothing more than a two dimensional array of squares of varying colors While measuring the color and position of a particular pixel is not particularly difficult for a modern computer consider that modern high resolution cameras measure anywhere from 10 to 20 25 Megapixels up to 20 million data points per frame where each pixel can have up to 64 bits 264 1 845 x 101 of color level When a typical computer is assigned to process repeated frames stitched together in the case of a long composite image the demands on the computer s processing and memory architecture may rapidly become too
18. Smart Camera platform is essentially a high resolution black and white CCD camera coupled to an embedded real time controller an architecture which offered a number of advantages for the research team Built in compatibility with LabVIEW Real Time module and the ability to accept real time acquisition and data processing algorithms generated in LabVIEW Ability to be programmed reprogrammed and readapted to many different needs Direct compatibility with the LAN communication architecture using built in RJ45 ports Rugged construction ready for the field Small lightweight form factor After the instrument was acquired it was easily adapted to a standard aluminum instrument bracket described in Chapter 4 as shown in Figure 5 7 The camera system itself came with no installed software and was loaded with a custom designed real time application to allow plug and play operation on the scanning systems 101 Figure 5 7 SCS instrument adapted to standard aluminum mounting bracket and packed in watertight case Photo by J Nelson Developing the SCS into a complete and robust tool for detection and measurement of surface flaws was outside the scope of this research so for the purposes of this project it was decided to simply program the device to capture images at a set rate and push them over the LAN using a network published shared variable NPSV data architecture This made the program stored on the camera it
19. a and 1 8 thick polycarbonate sheet b Press into block at placement and remove after initial set UNLESS OTHERWISE SPECIFIED NAME DATE E DIMENSIONS ARE IN INCHES DRAWN JDN 06 20 13 Detail 2a 2b TOLERANCES TITLE FRACTIONAL CHECKED TWOPACEDECRIAL a gt ENG APPR Crack simulators 1 16 and THREE PLACE DECIMAL MFG APPR 1 8 in thicknesses INTERPRET GEOMETRIC QA PROPRIETARY AND CONFIDENTIAL TOLERANCING PER COMMENTS DRAWING IS THE SOLE PROPERTY OF ies SIZE DWG NO REV lt INSERT COMPANY NAME HERE gt ANY WITHOUT THE ola aati ee NEXT ASSY USED ON i A V amp Fb O C k mio oe APPLICATION DO NOT SCALE DRAWING SCALE 1 20 WEIGHT SHEET 3 OF 4 5 4 3 2 1 214 97 00 10 00 11 00 10 00 8 00 6 Grade 60 Rebar THE INFORMATION CONTAINED IN THIS DRAWING IS THE SOLE PROPERTY OF lt INSERT COMPANY NAME HERE gt ANY REPRODUCTION IN PART OR AS A WHOLE WITHOUT THE WRITTEN PERMISSION OF saabnc lt INSERT COMPANY NAME HERE gt IS PROHIBITED APPLICATION 5 4 ealed 1 2 ID flexible tubing Tie to rebar with non metallic ties UNLESS OTHERWISE SPECIFIED DIMENSIONS ARE IN INCHES ES TOLERANCES FRACTIONAL ANGULAR MACH TWO PLACE DECIMAL THREE PLACE DECIMA INTERPRET GEOMETRIC TOLERANCING PER MATERIAL ANGH BEND Le DO NOT SCALE DRAWING 3 215 NAME DATE DRAWN JDN 06 20 13 Detail 3 CHECKED TITLE ENG APPR Voided Rebar Detail MFG APPR QA COMMENT
20. a continuous series of secondary echoes clearly visible in the tomograph This difference is unfortunately not distinguishable enough for reliable internal PT duct inspection 160 daghi tates ma zi als eqn itty rare B B i pra iii dhas te I I I I I 0 200 400 600 800 1000 1200 1400 1600 1800 1900 X Axis mm Figure 7 10 Ultrasound XZ plane tomograph of Asymmetric Internal PT Block after in situ grouting of duct 1 A number of other scans were performed on the Asymmetric Internal PT Duct Block during the refinement of the ANTARES SVW software It was found that with the void spaces in Ducts 2 3 and 4 filled with water some interesting anomalies were witnessed in GPR scans where the water void directly faced the GPR antennae Essentially the water duct grout interfaces resulted in refraction of the radar wave and introduced a scattering effect only visible when all interfaces were presented to the antennae This scattering effect is visible in Figure 7 11 161 Figure 7 11 GPR trace of Ducts 1 4 showing radar scattering in Ducts 2 and 4 due to duct water grout interface refractory effects Void and Flaw Detection Block Design One of the most common applications for NDE methods in the construction industry is to evaluate the quality of consolidation of concrete within a completed structure Movement of rebar after initial set over under vibration mix segregation and development of bleed water
21. all major functions of the scanner system and could in fact be used to operate the entire system from start to finish without ever using another interface The layout and design of the interface will be discussed later in this chapter but the architecture of the interface is based in a three phase sequence where the first sequence initializes the program and loads configuration settings as shown in Figure 6 23 the second phase is a persistent loop allowing operator interaction and automated operation of the scanning systems and third phase closes any open data ports relinquishes memory and terminates the program cleanly 118 You may use ANTARES SVW to create configuration and grid i y Scanners are currently AHJA S gt 4 Bi Fs ae E This VI E offline or disconnected t VI 4 i t Vi x PFP RunTransparenth PFP RunTransparently parently p b FP Transparency Scanner 1f AScanner No Indicator PA Scanner Select AScanner Control Tab P Initial Blank YH r Incoming Data Display Tab Applied Research and Evaluation of Structures E fStop Loops ScannerVoll Wizzard Automated Non Destructive Testing for Program version 1 1 Welcome Panel dialog Release Date 06 27 2013 Developed by J Nelson Graduate Research Assistant University of Florid
22. and determine the size of reinforcing steel within a concrete surface An example of a common covermeter device is shown in Figure 2 6 Figure 2 6 Acommon covermeter with digital display and handheld instrument head Photo courtesy Proceq USA 7 Locating and sizing reinforcing bars rebar in concrete is important both for verification of construction practices as well as locating obstruction to drilling and cutting operations when portions of a structure are to be removed such as coring and slab removal operation Additionally covermeters are extensively used to verify the cover depth from surface of reinforcing steel a parameter with minimum values specified in concrete design guides which has a profound impact on the durability of a structure Reinforced concrete structures with insufficient cover are documented to experience rapid corrosion of the reinforcing steel resulting in premature failure of the structure Covermeters induce a time varying magnetic field in the concrete near the handheld instrument head by passing an alternating current source through an excitation coil Concrete having a high magnetic reluctance will resist the transmission 30 of this field and limit the magnetic flux registered by a second sensing coil When the instrument head is sufficiently close to ferromagnetic reinforcement the low magnetic reluctance of the nearby steel bar will yield a much higher flux registered by the instrument 8 On
23. avoid disturbing the internal wet sieved concrete volume After the formwork was struck off with a long screed as shown in Figure 7 17 the surface defects including the lubricated triangular plates and the pervious prism were embedded in the concrete surface at the desired locations using a guide board shown in Figure 7 18 Using a procedure similar to the first wet sieved simulated honeycombing area the surface visible area of wet sieved concrete volume was fabricated by removing a volume of concrete with a scoop sieving it on a vibratory table and replacing it as shown in Figure 7 19 168 Figure 7 16 Consolidating the Void and Flaw Detection Block with an internal vibrator Photo courtesy of FDOT State Materials Office Figure 7 17 Levelling the Void and Flaw Detection block with a wooden board screed Photo courtesy of FDOT State Materials Office 169 Figure 7 18 Installing the surface visible pervious concrete prism in the Void and Flaw Detection Block Photo courtesy of FDOT State Materials Office Figure 7 19 Installing the surface visible wet sieved concrete simulated honeycombing into the Void and Flaw Detection block Photo courtesy of FDOT State Materials Office 170 After initial set of the concrete the two moving rebar specimens were shifted using their respective installed cables The rebar piece which moved along the Z axis direction moved easily as concrete was displaced upwards out of the formw
24. be successfully used in the field the overly complicated mixed signal interface had to be simplified The research team addressed this need by proposing an all digital interface handled through manufactured off the shelf cables In order to make the connections more robust the research team proposed signal connections be systematically moved from cheap fragile connectors to washdown rated sealed connectors Additionally improved methods of mounting and protecting the instruments on the scanning system were to be designed and implemented Integration of Additional NDE Instruments While the Phase automated NDE system included an eddy current covermeter shear wave ultrasound and impact echo system fully integrated with the scanning hardware and software FDOT recognized the need to include several new instruments with the system to keep up with the rapidly expanding NDE instrument market The new instruments proposed by the research team were a ground penetrating radar GPR system a laser distance sensor used to profile the specimen surface and a high resolution camera system with embedded vision processing computer FDOT was already in possession of the GPR and laser profilometer devices however they were not successfully integrated with the scanning hardware and software in the earlier research phase 79 Redesign of Software The software package originally developed to operate the scanning systems worked extremely well for the first sta
25. board In this manner only one end of the 166 rebar was moved after concrete placement allowing for an increasingly severe trail of disturbed concrete behind the path of the rebar The middle rebar was installed with two sections of 12mm ID flexible rubber tubing each approximately 250mm long contoured and bonded to the rebar s surface in two locations in such a way that one presented its entire area to one surface of the block and the other was located transverse to any surface contact instrument s view Figure 7 15 Three rebar pieces including the two moving specimens in Void and Flaw Detection Block formwork Photo by J Nelson Concrete placement After the formwork was completed concrete was ordered from a local ready mix plant The mix was designed for structural strength but included enough retarding 167 admixture to allow enough working time to install the surface defects Concrete was carefully placed around the embedded pervious concrete Once level with the top of the embedded large pervious concrete prism a volume of concrete equal to the large prism was removed and wet sieved through a No 4 sieve and placed by hand immediately below the fabricated embedded pervious concrete prisms This area represented a more realistic but more geometrically amorphous area of internal honeycombing The formwork was filled and carefully consolidated with an internal vibrator as shown in Figure 7 16 with care taken to
26. boiling point of the free water within its microstructure may result in superficial spalling as the vapor pockets expand beyond the strength of the concrete surface Sustained material temperatures above 300 C will result in interlayer and some chemically combined water being driven off Microstructural cracks will also be observed above this temperature as differences in the thermal properties of the constituent materials results in crack initiation at the interfaces and a significant loss of strength and elastic modulus Decomposition of calcium silicate hydrates C S H begins at around 500 C and ends around 900 C resulting in an unrecoverable loss of strength in the paste 8 67 Frequently in the event of fire damage to a structure only small portions may need to be repaired or replaced due to the very low thermal conductivity of concrete NDE methods are frequently used to evaluate the extent of thermal damage to better reduce the cost associated with repairing a fire damaged concrete structure In the early 2000s Matteo Colombo and Roberto Felicetti of the Politecnico di Milano in Milano Italy engaged in a research program to evaluate the effectiveness of multiple NDE methods to rapidly evaluate the extent of thermal damage sustained by concrete structures To test the NDE instruments in depth analyses were conducted on laboratory specimens as well as two full scale structures subjected to fire damage Among their many contributions Co
27. camera can only perceive two dimensions Three dimensional 26 measurements are possible with two or more cameras that share a field of view from different viewing angles however this capability is outside of the scope of this research and was not investigated Surface measurement Precision measurements of the contours of a structural concrete surface are rarely performed as critical concrete surfaces are usually formed by precise reusable forms such as in a precast yard In the case of some cast in place structural pieces such as bridge decks however the shape and roughness of the finished surface can have an impact in the overall performance of the element in terms of the contour and friction characteristics of roadway Perhaps even more importantly large cast structures formed in place have a tendency to bulge in areas of insufficient formwork support which may completely change the dimensions and center of gravity of the structure Modern surface contour measurements over large areas are typically performed using laser distance sensors mounted to a known reference point Most reflection based laser sensors use a laser source mounted a known distance from an optical sensor to create a laser light path that can be modeled as a right triangle Using simple trigonometry a single point laser distance sensor can rapidly measure distance from the camera aperture to the concrete surface as demonstrated in Figure 2 4 27 Measurin
28. data processing offered the most significant contribution and indeed the most significant challenge to the successful development of the ANTARES Processing Suite The 2048 point waveforms generated by the EyYeCON instrument included significant noise and variation and were often difficult to interpret Extensive post processing is required to generate a clear data visualization from the raw ultrasound data Typical artifacts in a raw ultrasound waveform include False reflections near the surface of the specimen due to Rayleigh wave detection by adjacent transducers 144 False reflections beyond the specimen thickness due to wave echoes Wave reflections from off axis objects and compact geometries e High frequency electrical noise To remove these artifacts a selectable sequence of filtering steps are available to the processing software operator First to trim the false reflections before and after the desired range of depth for the specimen the operator provides a minimum and maximum depth for the scan area as well as a velocity for the shear wave used to convert the time domain ultrasound waveforms into depth domain in millimeters This data is trimmed and passed to a bandpass second order digital Butterworth filter with configurable cutoff frequencies ANTARES Processing Data File Information File Name Ultra_V amp F_B_06102013 ANTD Date Time 6 10 2013 11 26 AM Ultrasound GI EyeCON
29. eClosed gt ftScanner 2 Read Mode Camera Settings Inst fie Figure 6 10 Independent loops on main program used to call Autoscan VI 116 Another equally important task handling SubVI is the Emergency Stop VI shown in Figure 6 11 This VI is operated in a special priority loop on the main VI and bypasses all other operations It calls the low level CML VIs directly and commands all servo motors on all scanners to immediately decelerate at maximum deceleration to zero velocity and remain in place until overridden There was no provision for an emergency stop of the system in earlier software and unexpected scanner behavior that could potentially damage the sensitive NDE instruments the scanners themselves or potentially injure the operator was dealt with by interrupting power to scanners This method was also dangerous however since the scanners are mounted vertically and an interruption in power allows the instrument heads to fall under gravity with very little deceleration from the Y axis drive motor s back electromotive force EMF The Emergency Stop VI effectively locks the scanners in their current location until reset and aborts any active commands to the scanners The soft buttons are located in convenient areas on the front panel are large and easy to see and respond in less than ims when clicked gt 2013 JooooooOoUoUOoUooOUooUo ooo ooUoooUooUoUoooU ooo Uo ooUU Uo Uo ooo U oo
30. either a ANTDAT or ANTGRD file Fortunately due to efforts from the US Geological Survey USGS GPR manufacturers have agreed to make their binary file formats open source and USGS have posted these file formats for public use in previous reports 35 With this file format in hand a LabVIEW VI was developed to open a group of DZT files and convert them into a two dimensional array of unsigned 16 bit integers WASE 2013 f File Path o y D mr p m p mn 5 T ry Sm f size in bytes B Scan Out Prem z error out Gne p F Figure 6 31 GSSI binary data file DZT importing VI used to interpret GPR data for the ANTARES Processing Suite 141 Visualizing the imported GPR data was accomplished by using the grid settings imported from the ANTDAT or ANTGRD to simply scale the axes of the various plots to match the spacing between the GPR line scans With this methodology the two dimensional GPR data cube could be plotted directly to the intensity plot displays without introducing a mathematical scale to the stored data Two distinct intensity plot displays were designed for GPR data visualization a fly through C Scan display and a fly through Tomograph display The C Scan display shows GPR data in the XY plane with the ability to step through the thickness of the scan area using a manual slider control or an automatic fly through function that playe
31. in NDE Introduction A common trait amongst NDE research considered to this point is the need for dense repeatable data grids to produce reliable data Most NDE researchers would agree that there is simply no replacement for more data especially when trying to detect small defects in an area of heavy reinforcement The need for automation in 71 structural concrete NDE is obvious and several institutions have invested heavily in developing automated NDE systems capable of producing very dense scan areas This section will acknowledge the two most influential sources for this work though many other organizations are actively contributing to the development of automated concrete NDE Federal Institute for Materials Research and Testing BAM Berlin Germany One of the first organizations to act on the need for automation in concrete NDE the Federal Institute for Materials Research and Testing BAM in Berlin Germany has spent years developing a multitude of different NDE scanning systems Under the leadership of Herbert Wiggenhauser BAM Division VIII 2 has developed vertical and horizontal two dimensional scanning systems with pneumatic instrument application devices With sophisticated data acquisition revolutionary data processing and data fusion techniques and the ability to use shear wave ultrasound impact echo GPR and a multitude of other instruments BAM remains at the cutting edge of automated NDE research 33 Figur
32. in concrete from the reaction 51 of monosulfates C3A CSH gt _1g within the hydrated cement paste and sulfate S within the environment according to the following chemical reaction C3A C H12 18 2CH 2S 10 16 H at C3A 3CSH3 gt The ettringite crystals possessing a larger volume than their precursors cause expansion within concrete and result in highly destructive microstructural cracking The effect is mitigated using low water cement ratio concrete to reduce permeability and cement which resists the formation of ettringite by containing lower quantities of tricalcium aluminate CsA 25 Figure 2 20 SEM Microstructure images of a concrete sample subject to negligible top left moderate top right and severe bottom sulfate attack 25 52 Microstructural cracking causes a reduction in stress wave velocity since the discontinuities within the material inhibit the transmission of the wave This effect is well established and techniques such as UPV have been used successfully with concrete cylinders submerged in a sulfate solution Boyd et al sought to apply the same method to more realistic situations of sulfate attack involving large concrete blocks partially submerged in a sulfate solution In this manner the region of the block not submerged in the solution but subject to its effects due to absorption could be investigated End View Figure 2 21 Test block and testing locations for Boyd s exp
33. is not necessary for comprehension of this work Surface Methods Visual inspection Perhaps the most versatile and frequently employed method of non destructively evaluating the quality of concrete in place is to simply direct a trained eye at the exposed surface 2 For many deterioration mechanisms of concrete the degradation of the material occurs on the surface and works its way inward These include very common material wear mechanisms such as sulfate attack surface wear and erosion due to scour freeze thaw etc Inspection methods for such modes of deterioration are well established and standardized by certification agencies around the world and are demonstrated in Figures 2 1 and 2 2 23 Figure 2 1 Clockwise from top left cracking due to stress plastic shrinkage cracking curling and alkali silica reaction Photos courtesy of American Concrete Institute 8 Figure 2 2 Examples of surface defects Clockwise from top left bugholes blistering honeycombing cold joint line Photos courtesy of American Concrete Institute 3 While a certified concrete inspector can quickly and accurately judge the condition of a concrete surface automating the process using a high resolution digital camera and sophisticated image processing algorithms offers an interesting potential to 24 increase the precision and repeatability of the process The FDOT State Materials Office has applied such a technique to their Heavy Vehicle
34. oaaao ooo ooo ooo DoooDoooooODDUo oo oOUU Doo oaoa aoaaa aoaaa aaa Initialize and Read Default _v Call Mode g a ee CheckBulkBuffer error out InitUSB SendServicel Buffer eri i oF 2 oss t init 2 fet Returned Bytes 5 Eve ja pus offset er or A Scan Raw DH fE ERE fi32 05 2 Eele a LS Me N N M MeMeM Me MMeM M MeM MMeM MoM M M M M MeMoM MeM M MeM MeM M N MeM MeMeM M M MeM MeM M N N MeMeMeM 8 N NeN MeM M K He1 Dooooo ooo ooo ooo DODO Figure 6 6 Code snippet of Ultrasound A Scan SubVI one of four low level instrument I O subroutines used with ANTARES SWW Hardware handling subVIs The hardware handling class of SubVIs contains only two SubVIs the Scanner Movement VI and the Native Instrument A Scan VI The former manages all scanner movement including instrument protection by preventing movement outside of accessible scan area raising the instrument off the surface before moving where applicable and managing speed and acceleration This code section is shown in Figure 6 7 and is available in detail in Appendix A The A Scan VI is responsible for polling whichever instrument is connected to the scanner for data and converting that data into a form suitable for display to the operator in the real time data visualization window and saving into the ANTDAT data f
35. of water and corrosive agents which attack any and all exposed strands resulting in a catastrophic loss in tension Left unchecked this failure will inevitably lead to loss of structural capacity 29 Martin Krause and other personnel at the German Federal Institute for Materials Research and Testing BAM have researched non destructive methods to investigate grout quality within post tensioning ducts Through Krause s research a comprehensive comparison of radiographic impact echo ultrasonic echo and radar techniques for evaluating voided internal post tensioning grout is presented Among Krause s important recommendations are the use of GPR for evaluating internal post tensioning ducts where the ducts are made of plastic and not steel While steel ducts are common in older structures current practice in the State of Florida allows only plastic internal ducts 65 Additionally important data processing and interpretation methods are presented for acoustic techniques Using Synthetic Aperture Focusing Technique SAFT and carefully designed data displays the full field image developed from ultrasonic echo scanning clearly shows defected areas within the post tensioning ducts and lends credibility to the technique as a whole 30 e 0 T j T aal i Ee a a a a al 0 2 i Wree 0 4 1 l 05 1 15 2 25 3 35 Figure 2 29 Ultrasonic echo results from Krause s research using a similar instrument and scanning device to
36. of a specimen collimated stress waves are driven into the material to either be received on the other side of the material in direct transmission mode as shown in Figure 2 14 or received on the same side a short distance away in pulse echo mode The velocity of these pulse waves is well described by the equation listed in the preceding section and the fast repeated manner by which the pulses are created and received lends itself to making many measurements averaged over a short time span to yield much more accurate data than is obtained from a single pulse Additionally the generated waves are far more collimated than those originating from a point impact and the wave fronts do diverge greatly from the intended path reducing the effect of compact specimen geometry on wave reflection UPV methods as applied to concrete generally operate at much lower frequencies then their metals testing counterparts as standard high frequency oscillating transducers fail to overcome the scattering effect caused by the many paste and aggregate interfaces and pores present in concrete to create repeatable stress waves 38 The operating principle of the transducer is however identical a piezoelectric crystal within a housing is used to drive a hard contact surface into the concrete surface The air gap between the UPV transducer and the concrete surface must be filled with a thin layer of acoustic couplant in the form of a gel or grease in order to ensure comple
37. represent number position as shown in Figure 5 3 The byte stream was read over serial interface using NI VISA low level Vis and parsed according to the schematic provided by Leuze USA as shown in Figure 5 4 Measurement value 14 Bit i Low Byte Bit 0 0 2 High Byte Bit 0 1 7 i 7 aaa E LSE WSE Figure 5 3 Byte stream representation of ODSL8 distance measurement 34 98 A al 2013 VISA resource name Flush the serial buffer to read instantaneous data Read in two bytes 2 Convert the hexadecimal string into two 8 element boolean arrays 3 Use the first element in the array corresponding to Byte Index 0 to determine whether Byte Index 0 is the High or Low byte The first element in the array corresponds to the LSB Concatenate the last seven elements in each boolean array in the appropriate order 4 Convert the 14 element boolean array to a 14 bit number adapt to double precision and shift the decimal point appropriately Apply logic to replace the default 0 reading with Inf when the measurement is out of range 1 29 13 J Nelson University of Florida Figure 5 4 LabVIEW VI which reads and parses the 2 byte serial stream from the Leuze ODSL8 laser distance sensor Adapting the laser profilometer to the scanner system using the instrument carrier palette system was straightforward due to the low power requirements and small
38. research where he met Christopher Ferraro After three years of working together at the SMO and at the Civil Engineering Laboratories at the University of Florida Jordan became Dr Ferraro s first graduate student He plans on continuing to work with his friend and employer after graduation as a consultant in civil engineering instrumentation controls data acquisition and nondestructive evaluation 223
39. scan operation and not dependent upon time changing material properties such as elastic modulus was performed first Laser scans on the hand finished surface with surface visible defects A side produced faithful recreations of the surface defects and the marred surfaces around them as shown in Figure 7 21 Laser scans captured the inner contour of a simulated crack for a few data points as shown in Figure 7 22 and 7 23 during a particularly dense scan Figure 7 21 Unfiltered 3D trace of defected block surface Note the scattering above the honeycombed area 172 ww 32ue3sIg I 90 100 110 122 Figure 7 22 Laser C Scan image of defected block surface with cracks and honeycombing visible Figure 7 23 3D laser trace of block surface viewed from defected end in YZ plane Note the scattering above the honeycombed area circled upper left and trace of the inside of a simulated crack circled lower right 173 A laser profilometer scan of the form side B side of the block revealed a very smooth flat surface as expected however it also revealed a 22mm change from the depth of the upper left to the lower right corner This difference is likely due to the block being held slightly off level within the clamping frame the scanner mounts on the frame being slightly off plane with respect to the specimen or both Figure 7 24 Laser C Scan of B side of block revealing a smooth but not flat with respect
40. that used in this thesis Plan view showing defected areas top B Scan middle and phase displayed B Scan bottom 30 Ultrasonic echo is the principal method used in evaluation of post tensioning ducts in this thesis with GPR used a supplementary method The research performed by Krause and the rest of the concrete NDE scientists at BAM is considered the foundation for the work performed herein 66 Damage Assessment Introduction In the case of catastrophic damage to an in service structure by act of man or act of God engineers are called upon to rapidly and accurately assess the condition of the structure and dictate repair or replacement procedures Historically the tools available to these individuals have been highly limited Practices exist for conservative visual inspection of structures damaged by natural disaster impact and thermal events however a common application of NDE technology and a major area of NDE research is to quantitatively evaluate the extent of damage to an in place structure to better recommend appropriate repair procedures This application of an automated NDE system for use in the State of Florida cannot be ignored Nondestructive evaluation of fire damage While an inherently fire resistant material concrete exposed to sufficiently high temperature for long periods of time exhibits a number of unfavorable property changes If concrete is well hydrated and relatively impermeable temperatures higher than the
41. the data visualization presented in Figure 7 7 dark blue pixels indicate deeper reflections as the stress wave passed through the duct and reflected somewhere within it either off the tendon itself or the back surface of the duct Pink reflections indicate the wave reflected off the duct surface indicating poor grout bonding or a void within the duct 400 00 200 00 1600 00 2000 1200 00 X Axis Figure 7 7 View of ultrasound data of dry PT ducts in the XY plane All units are in mm After analyzing this initial ultrasound data the deeper reflections exhibited in areas of proper grout bonding compared to the shallow reflections of the ungrouted regions of the duct indicated that the shear wave ultrasound method was a promising method for detection of air voids in internal PT ducts Visualizing the data in this manner however would require a significant amount of data points to make accurate inspection possible as an A Scan waveform at any one point on a PT duct could potentially reflect off the top surface of the duct due to the minor density change or due 157 to a delamination of insignificant size Proper inspection of these ducts depended on developing a large data visualization in order to analyze the shape and scale of areas of shallow reflection in order to discern a true defect in the grouted duct system from shallower than expected reflections due to other effects This observation further reinforced the impor
42. the new dimensions for this block meant that the reusable formwork constructed for fabrication of the previous NDE evaluation blocks could not be used it was decided to also attempt a new method of construction Since it was hypothesized that delaminations were induced in the ducts of the first internal PT duct block when it was stood up and moved from its horizontal mold this block would be fabricated vertically in place within the scanner frame as shown in Figure 7 4 Figure 7 4 Formwork under construction within the scanner frames Also shown is the small block mold used to make sister blocks for wave velocity measurement Photo by J Nelson After fabrication the new block was stripped of its formwork and the excess length of PT duct and strand was cut off Later PVC screw caps were installed on the ends of the ducts using epoxy mortar and a layer of concrete was placed to smooth the top of the block and seal the PVC caps The removable caps would allow future users of 154 the evaluation block to place or remove air water or soft grout in the duct void spaces as desired to investigate several different forms of PT duct grout defects Pt w LAS Figure 7 5 Completed block immediately after formwork removal Photo by J Nelson Data Collection The Asymmetric Internal PT Duct Block was fabricated concurrently with the development of ANTARES SVW so initial NDE scans were as much an evaluation of the instrumen
43. the original evaluation blocks The block would be approximately 50mm thicker than previous blocks to allow the PT ducts to be offset from the middle plane of the block This design allowed approximately 63 5mm 2 5 of cover over the ducts on one face of the block and a deeper 152 4mm 6 0 of cover on the other face The shallower cover represents the minimum cover specified by FDOT design guidelines for durability while the deeper cover is typical of that found in internal PT duct placement in hammerhead piers To maintain acceptable weight limits for movement with a forklift the block was also shortened in the Y dimension as shown in Figure 7 3 3X Bundle 7 wire strand Centered each duct Size as available Type B Duct tapered pr a Type A Duct no flaw in situ grout flaw grout before concrete placement UNLESS OTHERWISE SPECIFIED NAME DATE DIMENSIONS ARE IN INCHES DRAWN JDN 5 14 12 TOLERANCES TITLE FRACTIONAL CHECKED ANGULAR MACH BEND TWOPLACEDEGMIAL a gt ENG APPR ANTARES Asymmetric PT Block THREE PLACE DECIMAL MFG APPR INTERPRET GEOMETRIC QA PROPRIETARY AND CONFIDENTIAL TOLERANCING PER COMMENTS i SIZE DWG NO ae o F A assy APPLICATION DO NOT SCALE DRAWING SCALE 1 32 WEIGHT SHEET 1 OF 1 5 4 3 1 Figure 7 3 Design for Asymmetric Internal PT Duct Block All dimensions are in Inches An enlarged view is available in Appendix B 153 Construction Since
44. through Y axis guideway connection on box on Y axis drive Impact Echo External Analog Direct to PXI 8 5 m cable Transformer Ultrasound Internal Battery USB Direct to PXI 1 0 m cable Figure 3 2 Instrument and scanner connection scheme of the Phase FDOT automated NDE system NDE instruments were physically attached to the moving instrument head using an improvised mount consisting of elastic bands and cable ties with their own unique data cables painstakingly routed through the cable guideways on the scanner drives 76 The result was a clumsy time consuming mounting process that left the instrument open and susceptible to damage from impact water spray etc At best a serial based instrument such as the Proceq Covermeter took 20 30 minutes to connect to the scanner undergo cable routing and connect to a remote serial port In the case of the Germann Instruments EyeCON shear wave ultrasound device and the Olson Instruments impact echo device the instrument had to be connected directly to the PXI chassis using their own proprietary cables leading to situations where short cables mandated close placement of the PXI to the scanner and created the potential for dangerous entanglement of cables during operation The operating software itself was also identified by FDOT as insufficient for the next phase of automated NDE development While the software performed well for research purposes it was far too complex for effective
45. young or highly saturated concretes 20 Related NDE Research Materials Oriented NDE Elastic Property Estimation Introduction One of the most common applications of concrete NDE and indeed one of the earliest requirements of the technology is to estimate useful measurements of material properties normally acquired from more destructive or labor intensive tests From a concrete mix design standpoint two of the most useful and desirable material properties to measure from hardened concrete are the compressive strength fc and the static modulus of elasticity E These two measurements are usually made from destructive tests performed on concrete cylinders or cores according to ASTM C 39 Standard Test Method for Compressive Strength of Concrete and ASTM C 469 Standard Test Method for Static Modulus of Elasticity and Poisson s Ratio of Concrete A common situation 46 faced by concrete engineers and testing professionals is a need to estimate the compressive strength or elastic modulus of a quantity of concrete already in place and in use where companion specimens are not available for traditional testing For example if a section of Interstate highway is temporarily closed and heavy traffic needs to be detoured over older infrastructure that may or may not meet the strength required to support heavy trucks depending on the age of the structure compressive strength and modulus data may not be available The options available to engineerin
46. 6 29 6 30 6 31 Example of a 2D laser profilometer data visualization of a pea experimental BG Ce e e E S E e 3D render of laser profilometer data a rough bridge deck surface 140 GSSI binary data file DZT im re VI used to interpret GPR data for the ANTARES Processing Suite Lies alah kaaa ails tiene baiana wee 141 Fly through C Scan GPR d naa tie a ado of dense rebar in a test UT i oneee balaias BOM an tae Instruments and signal connection scheme for the new ANTARES system 149 Grouting of middle ducts of Asymmetric Internal PT Duct Block 152 3D render of Asymmetric Internal PT Duct Block deSign seeseeees 152 a 153 154 156 7 7 View of ultrasound data of dry PT ducts in the XY plane cceeeseeeeee 157 7 8 View of the Asymmetric Internal PT Duct Block shear wave ultrasound data in the XZ plan 158 7 10 7 11 7 14 166 167 7 16 Consolidating the Void and Flaw Detection Block with an internal vibrator 169 7 17 Levelling the Void and Flaw Detection block with a wooden board screed 169 7 18 Installing the surface visible pervious concrete prism in the Void and Flaw Detection Be sa cea eNO tener nO Ree ED Renn ae mere aaa O 7 20 ANTARES V s rface defects visible sedina daadaa aa aa a a aaa a aad VE 7 21 Unfiltered 3D trace of defected block surface ou eeeceeeeeee
47. AS A WHOLE m maen drawings where OC WITHOUT THE WRITTEN PERMISSION OF NEXT ASSY USD ON licabl applicable lt GNSERT COMPANY NAME HERE gt IS PROHIBITED APPLICATION DO NOT SCALE DRAWING SCALE 1 20 WEIGHT SHEET 1 OF 4 Figure 7 13 Drawing for Void and Flaw Detection Block Dimensions are in Inches Complete CAD drawings are available in Appendix B 164 Construction Defects and formwork The first phase of fabricating the Void and Flaw Detection Block involved fabricating the prisms of pervious concrete used to simulate internal and surface level honeycombing a common symptom of poor consolidation practices The prismatic artificial shape was chosen since its location and volume could be well defined and documented before NDE analyses were performed Two prisms measuring 152 x 152 x 254 mm 6 x 6 x 10 in and two prisms measuring 76 x 152 x 254 mm 3 x 6 x 10 in were fabricated by compacting a pervious concrete mixture into a standard concrete beam specimen mold The pervious concrete mixture was fabricated in the laboratory following the mix design submitted to the local concrete plant for delivery however the fine aggregate and a portion of the cement was removed from the mix to reduce the overall paste quantity and approximate a poorly consolidate region of the delivered mix Once the pervious prisms set they were demolded and saw cut to length using a wet diamond saw The prisms to be installed within the finished block were
48. ATE MATERIALS OFFFICE O Figure 5 9 SCS placard design with origin and direction markers The placard is roughly the size of a business card 104 Figure 5 10 SCS instrument mounted to Scanner 1 with ring light illuminating an orienting and calibration placard Photo by J Nelson et ANTARFS SCS1 169 254 100 33 Felul LPR majom oluji 4 lpm ve i e e ol OL Fs i Press Cirl A to run program Figure 5 11 Threshold image of SCS placard origin marker demonstrating ability to digitize orientation and origin marker and calibration grid 105 CHAPTER 6 DEVELOPMENT OF THE ANTARES SYSTEM ANTARES and the Concept of Accessible Automated NDE Introduction From the beginning of this research effort one primary goal of this project was to deliver a system to be used by skilled technicians with limited training rather than engineers and research and development teams The system needed to be dependable robust and cohesive with an approachable well designed software interface fault tolerant programming and intuitive controls Until this point this manuscript has discussed hardware improvements made in service of the goal of creating a more usable system However the bulk of the effort invested in the development of this automated system was spent in LabVIEW software development creating a simple and robust software package for system control dat
49. D FILE THIS SHEET IS THE ELECTRON ABC test element drawings courtesy FDOT Structures Design Office 5 P 5 Rebar Refer to 3 0 Panel EX 1 for spacing Grout Shear Pockets 2 0 y 2 0 2 0 2 4 2 8 Z s Bars 5E Spacing See Keyway Detail on Sheet Ex 2 It W Min Refer to Haunch Detail Bars 4C Refer a FD to EX 1 for spacing TI 1 2 5E Bars i in Beams I amp 5 2 SE each hear pocket amp z 2 Beams 2 3 amp 4 nding Compound Len per Spec 926 2 Bars 5J T amp B Grout Shear Pockets Concrete Block Concrete Block Type A Epoxy Bonding Compound per Spec 926 ki 4D See Pian Beams 1 amp 5 36 Beams 2 Section A A See m 2 Bars 5J Bar 5E Detail Bars 5E Spacing 218 Bars 4C 12 O C T Bars 4C 6 0C B g ES hear Pocket T amp B 1 0 Joint CLOSURE JOINT DETAIL 2 Bars 56 T amp B Bars 4C 12 OC T Bars 4C 6 0 C B i E 1 9 Lap Typ E eee iat a Sa NZ a el Shear Pocket Bars SH N l Bars 5H Reinforcement at Shear Pockets Interior Beams STATE OF FLORIDA DEPARTMENT OF TRANSPORTATION a COUNTY NANET TA y ISAGI TOI AN Ce prejerte Test 1 Precast Full Depth Deck Shear Pocket TateardENEEFUCT UT PRN Details PRECAST SUPERSTRUCTURE Prerast_srfogen il super Oe O7 086 ILE S
50. DEVELOPMENT OF AN ADVANCED AUTOMATED NONDESTRUCTIVE EVALUATION SYSTEM FOR INVESTIGATION OF CONCRETE STRUCTURES AND VALIDATION OF CONCRETE NDE INSTRUMENTS By JORDAN D NELSON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2013 2013 Jordan D Nelson To Marc Ansley who showed a high school kid what it meant to be an engineer and to my family who has always shown me the love and support needed most of all ACKNOWLEDGMENTS would like to thank each member of my supervisory committee for their support and guidance throughout the development of this work particularly Dr Christopher Ferraro His extensive experience in the field and invaluable support on both a professional and personal level have made this project far more than words on paper My esteem of him as both a mentor and a friend cannot be overstated Drs Peter Ifju and Bhavani Sankar have been invaluable in successfully helping me succeed as a cross discipline engineering student Their guidance has allowed me to pursue a mechanical engineering degree while working on a civil engineering oriented research project successfully planting me in a niche in which wish to continue my engineering career Dr Daniel Algernon has been a tremendous resource to this research effort and his work in the earlier phases of this project laid the foundat
51. EIOS minnein a aa eaten 23 ETEA a g A E A O T 23 Su rlace measureMeNiiesreshissina a a A 27 Eddy Current MatOd S aessa aa teeaeea ee eae 29 PER MeMO S aoe aieaeiiaaiaeeauaaeeean 33 ae a AE E AT AT T E O A O ETA 33 Impact echo method cccceeeeeeeeeesesseceeeeeeeeteeeeeenaeceeeeeeeeseeesessaeeeees 34 Ultrasonic pulse velocity FIO sei crea tester trsdtosrntererttnadmdimnimeriecninadineaiaadiay 38 Shear wave ultrasound method cccccceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeess 40 Ground Feneraino Radar aannre ere Pere tere niere rinerrtore fears Ferien vere ttert tere nen tere twrr ere erer rere 43 Related NDE Research Materials Oriented NDE ccceeeeeeeeeeeeeeeeeeeeeeeees 46 Elastic Property ESTIMA roeestoncsascsacsscaienncineesnaidaaciencdeaennnsbensdeaiecnmnedencdasaaande 46 ISIE sorai 46 Elastic properties of concrete using ultrasonic pulse velocity 0 47 Early age elastic properties of concrete using shear wave ultrasound 48 Durability and Degradation of Concrete sssssssssseeereeseesrrernrrrnsserrrrrrnnnneeseee 49 RR CIG a 49 Permeability of bridge decks using ultrasonic pulse velocity 00 50 Nondestructive evaluation of sulfate attaCk cceeeceseseceeeeeeeeeeeeeeeeeeees 51 Nondestructive evaluation of alkali silica reaction ccceeeeeeeeeeeeees 55 Related NDE Research Structural NDE siisisinisiniiinisinenini
52. IGNED AND SEALED UNDER RULE 61615 23003 FAL TRONIC F SHEET IS THE ELEC AD OF THIS OFFICIAL REC THE 1 2 3 4 5 6 7 8 9 LIST OF REFERENCES B Mitchell Concrete Spalling Pioneer Masonry Restoration Co Inc C C Ferraro A J Boyd and G C Consolazio Evaluation of damage to bridge piers using pulse velocity tomography vol 38 2012 ACI Committee 201 ACI 201 1 R08 Guide for Conducting a Visual Inspection of Concrete in Service Farmington Hills MI American Concrete Institute 2008 FDOT State Materials Office Instrumentation of Florida s Accelerated Pavement Testing Facility Gainesville FI 2011 Laser Displacement sensor LAP LASER T J Solution Co Ltd Online Available http www tjsolution com Accessed 9 June 2013 Laser XVP Fugro Roadware Fugro Roadware 2013 Online Available http www roadware com Accessed 9 June 2013 Cover Meter Profometer 5 Covermeter Proceg USA Inc Online Available http www proceq com Accessed 6 June 2013 P K Mehta and P Monteiro Concrete Microstructure Properties and Materials New York McGraw Hill 1993 D Algernon D Hiltunen and C Ferraro Validation of Nondestructive Testing Equipment for Concrete University of Florida Department of Civil and Coastal Engineering 2010 10 N J Carino The Impact Echo Method An Overview Proceedings of 2001 Structures Congress and Ex
53. NTARES SVW The program begins by creating default folders for saved grid files ANTGRD configuration files ANTCFG and data files ANTDAT on the computer s hard drive in the current user s document library if they do not exist already Next the program searches for the last used system configuration file If the file does not exist the program launches the Configuration Settings Screen so that the operator may provide the vital system settings necessary to operate the ANTARES system The remaining steps of the program perform a check and initialization on each scanner s serial port and cCDAQ and returns the success or failure of each check as a visual Cue to the operator as well as a group of settings provided to the main interface that will prevent the operator from using hardware that is either offline or in an error state The first steps of the Splash Screen VI are shown in Figure 6 15 Create default folders if they don t exist 1 0 1 7 2 Load Configuration file Default File Paths E a Smi 1 Look for last configuration file O O rnnnoonoooona vee ue cece Data Files ia a rd ser Documents Ca gt fLoading Bar False gt n EE 7 al Configurations RHF Figure 6 15 The f
54. S SIZE DWG NO REV A V amp F_block SCALE 1 20 WEIGHT SHEET 4 OF 4 ABC test element drawings courtesy FDOT Structures Design Office Beam 70K oo 70 3 G 15 4c 8 Shear Packet 6 6 SK _torerete Blocks Typ 26 Spacing 7 0 3 SC BIRT 15 4C 6 8 Bars 4D each 4C Gar see Partial Reint Layout on Sheet EX 4 Bars 48 T Gtwe SH Bars ars 4B T Btwn Each 5J Bars Btwn Each 5 Gars Gars 46 T Stwo 5H Bars 4 Bars 5G T amp 8 Bars 4C oo DE ScAIPTIEW 216 STATE OF FLORIDA DEPARTMENT OF TRANSPORTATION 5 Bars TAS See Deck Plan for Spacing LONAN SSN SLEITE Precast Full Depth Deck Plan amp Elevation 0 000 End 7 Ele 0 030 End 2 Elev PRECAST SUPERSTRU MELAS APEEB Sune JCTURE ADEE OWT 3 FAL 0 23 01 RULE 61615 ROWIC FILE SIGNED AND SEALED t RD OF CO OFFICIAL AE ABC test element drawings courtesy FDOT Structures Design Office 10 5 Shear Pocket Typs Leveling Bolt Typ Lifting Leveling amp Shear Pocket Layout Washer Polyethylene 4 Sq Washer a s o zoso soso Pomerne 3165 j Sta mut 5 r 3040 4000 Poiyethyiene 4125 Buildup heights are variable amp assume a 2 cross slope Precast Panel Keyway Detail End 2 lt 1 Btm of Slab Grout haunches with High Performance Normal Setting Grout X Min Refer to Haunch Detail Sars 5J Re
55. Scott showed very close correlation between suspect areas detected with impact echo and the contours developed by taking cores By contrast GPR is fast and effective and characterizing large areas of a bridge deck very rapidly but must be carefully analyzed and is prone to missing smaller defects 28 Mr Scott s study was especially useful in the course of this thesis as one of the main applications for an automated NDE scanning system for FDOT would be evaluation of aging bridge decks The study suggests a combination of GPR to rapidly detect areas of significant damage where applicable followed by a slower methodical application of acoustic echo for precise detection of air voids A major conclusion of the study that acoustic methods are extremely effective in evaluating aging bridge decks subject to delamination but highly limited by their speed of application is addressed within this thesis Nondestructive evaluation of post tensioning systems Recently demand has been placed on the NDE community in the United States to develop advanced methods for evaluating post tensioning ducts in segmentally constructed structures Post tensioning is used to assemble pre cast reinforced concrete segments into a functioning structure and can consist of high strength threaded rods or long steel multi wire strands passed through adjacent segments and tensioned Between the two methods multi wire strands are usually selected for long 63 structur
56. Tallahassee Florida as shown in Figure 8 2 This specimen was fabricated at the request of the FDOT Structures Design Office for evaluation of construction practices and feasibility analysis A major concern of the officials witnessing the placement was the integrity of the grout interface between the bridge deck and girders Since this interface is levelled before grouting operations by extension of its relationship with the bridge deck roadway surface it is difficult to guarantee that all entrapped air is displaced and vented by the pressure grouting operation Air pockets and bleed water pockets in this grout interface present a readily available water infiltration path that can lead to premature corrosion and damage and could be catastrophic in areas prone to freeze thaw damage This test element and its associated concerns presented a unique opportunity for the research team of the ANTARES program to collaborate with the FDOT Structures Design Office to generate mutually beneficial data Logistically the test element was conveniently located supported by local FDOT personnel contained a multitude of interesting features to investigate and was relatively easy to deploy the scanner system onto as the bridge element was actually fabricated below ground level with the bridge roadway surface level with the Ansley center s existing outdoor patio 185 Location of ABC Test Element Figure 8 2 Location of the test element behind the FDOT
57. a This program supercedes ScannerVoll Wizzard ore VI Figure 6 12 Initialization phase of main interface This subroutine operates the splash screen loads configuration settings saved on the hard drive and sets the default state of the interface The second phase of the main interface can be thought of as the central program or operating system of the ANTARES system and is shown in Figure 6 13 It consists of four persistent loops 1 Priority Loop responsible for controlling emergency stop and reset conditions as well as providing the operator with feedback on the scanners position and activities 2 Human Interface Loop responsible for accepting commands from the operator and altering the behavior of the system accordingly 3 Scanner 1 Operation Loop responsible for controlling the operational state of scanner 1 idle open instrument read and automated scanning 4 Scanner 2 Operation Loop responsible for controlling the operational state of scanner 2 idle open instrument read and automated scanning 119 Humint Loop Priority 2 Setting changes popup windows manual scanner movement start pause abort scans open instrument connections El li Home Load Manual Home Load Auto Home Load Manual 2 Home Load Auto 2 Value Change I Type ODO oe oo oen o 0 heen Handle Homing Troe p Home Load Manual JDOooOUooODooOoo oUooooooooo
58. a acquisition and data processing Approach To begin the software development effort an overarching goal of the software architecture was established In the interest of software unity the research team decided to begin software development as an entirely new architecture with no code reused from the Phase operating software The automated NDE system each of the two scanners and all supporting hardware the laboratory evaluation blocks and all instruments integrated with the scanner systems were incorporated under a single program name that would also be applied to the software designed to support the system Automated Nondestructive Testing for Applied Research and Evaluation of Structures ANTARES 106 ANTARES Figure 6 1 FDOT ANTARES logo designed by the research team The software package was separated into two distinct programs that compartmentalized the programming effort allowed for future improvements to be programmed in in without impacting system capabilities and minimized the demand on the operating computer One program was designed for operation of the scanners and instruments data acquisition and saving and limited display of incoming data while the other was designed to process the data saved by the operating software into rich detailed two and three dimensional displays suitable for report generation and structural evaluation ANTARES SVW Operating Software Introduction As an homage to Dr Algernon a
59. a normal stress event the shear wave or S Wave corresponds to shear stress and the surface Rayleigh or R Wave corresponds to the disturbance propagating along an impacted free surface 10 HEF Reflected HHE 6 P wave Figure 2 8 Finite element simulation of an impact on a plate like structure showing the three types of induced stress wave 11 33 For an infinite isotropic perfectly elastic solid the wave velocities for each of the three wave types Cp Cs and Cr for P S and R wave respectively is described in terms of the modulus of elasticity E shear modulus G Poisson s ratio v and density p ae E 1 v p 7 p 1 v 2v Equation 2 1 a f TS D Equation 2 2 C 0 87 1 12v RRE c ee Equation 2 3 as follows 12 Some acoustic methods simply use a measurement of the time of flight of a stress wave in order to determine the wave velocity Wave velocity can be used to estimate the elastic properties of the material using the above equations or determine if the material has been damaged or has significantly degraded Other methods will register a complete waveform of reflected waves to better determine if flaws or defects exist in the material s thickness The two techniques and the principles required to employ them are discussed in the section for the method to which they pertain Impact echo method The impact echo method is based on monitoring the minute movements of the surface of a concrete speci
60. abled lon line length I Disabled Line Count A Test Grid Line Count similar to E Disabled and Grayed Out 7 pDisabled E point Count 4 point count 3 a save and Exit Disabled 3 A PDisabled 3 Wizard Step Co pDisabled 2a Ln Delta E a E Calculate how big across the grid is supposed to be check value Disabled E Lines Run in X Default gt 2 S gt f Grid Total Width is A Grid Settings aa bz False v Since you chose your lines to go in the X Direction have Dialog frScanner Position Inst adjusted the previous point v Figure 6 17 Grid Generation Wizard main loop with state machine architecture The Grid Generation Wizard VI software architecture is based around a simple state machine a case based structure in which the operator s actions change an enumerator control which alters the case state of the software This architecture allows the operator to step through the grid generation process in order go back one or more steps and perform each step in different ways as desired The software also validates the grid setting parameters at each step of the operation so physically impossible grid configurations or grids that may run the instrument into the scanner frame cannot be generated based on invalid user inputs 123 Interface Design Introduction Interface design is usually se
61. ae deunenenese eit Siena Pere AT sree Measure to nearest 1 mm the actual distance between centers of transducers Figure 2 10 Test setup for measuring P Wave velocity using impact echo ASTM C 1383 14 In a depth measurement or flaw detection analysis a single transducer located near the impact point is used to record the P Wave as it reflects from the back of the specimen to the impacted surface repeatedly for several reflections This waveform is often processed using fast Fourier transform FFT or other suitable technique to take the data from time domain to frequency domain Data Acquisition and Analysis System Ma lt 0AT Figure 2 11 Test setup for measuring concrete plate thickness using impact echo ASTM C 1383 14 36 Once the collected data is processed into frequency domain a single dominant peak is exhibited in the power spectrum 0 04 H H H H 8 67 kHz Af 0 122 kHz 7 i mej 2 a g gt S s iva 0 500 1000 1500 2000 2500 3000 o 5 10 15 20 25 Time us Frequency kHz Figure 2 12 Time domain left and frequency domain right data collected from an impact echo test of a 250mm thick concrete plate Note the single dominant peak in the power spectrum 14 The frequency of this dominant peak corresponds to the frequency of the P Wave as it bounces from the impacted surface to the point of original reflection The distance from the reflected poi
62. al deficiencies were noted by FDOT that were to be addressed in this phase of automated NDE research Figure 3 1 CAD drawing of as built scanner systems mounted to block clamping frame The two electric scanning gantry frames each measuring approximately 3 65 by 2 75 m were operated by a National Instruments PXI chassis with embedded computer through a combination of serial and analog voltage interfaces This configuration required a separate serial cable for each of the two stepping motors for each scanner and a custom built four conductor cable for each scanner s pneumatic cylinder used to 75 apply instruments to the concrete specimen surface These interfaces were unreliable fragile difficult to manage and much more expensive than a single digital interface Furthermore since so many data interfaces were needed the scanning systems were unable to be operated at the same time when switching to a different scanner the operator had to physically disconnect the first scanner s data cables and connect the second The PXI itself was also a liability as it is a large expensive fragile piece of equipment requiring special care and attention to function correctly Signal Connections mam Serial RS232 mmm Analog Voltage mm USB Permanent Temporary 7 7 I Po I 7 Impact Echo Ultrasound Instrument Power Supply Connection Cable Routing Covermeter Internal Battery Serial Partial route
63. anSGUCe l sseeeeeeeeee 41 Modern shear wave ultrasound instrument showing 24 transducer array and schematic of wave reflections eee ee eee eon DIRE On Oe nO nPIRRD TO nDNPND DEP PAD OR PESEESON PED T 42 Modern handheld GPR unit with built in color line display ceeeeeeeeeee 45 Schematic of Voigt s experiment illustrating the usefulness of the shear wave technique in determining viscoelastic properties of cement paste 006 49 Yaman s test apparatus and results showing a large UPV difference between teld and laboratory CON GS ccs cee ss aaceeaecetece se daa aes eeoeetaeesnaea aa aia 51 2 20 Wave velocity re S fro C 2 25 Concrete slab in Berlin used by Beutel and other researchersS 60 2 26 Results from Van Buren Road Bridge NDE study overlaid with defect map 62 2 27 Typical tendon Ge ecient BA 2 28 Layers of corrosion protection in a typical post tensioning tendon 65 Ultrasonic echo results from Krause s resea and scanning device to that used in this th ch using a similar instrument 2 30 UPV results of a fire damaged concrete panel cscenonnesonrenonenenonerenonenenens OD 2 31 Collapse of the southbound main span of the Sunshine Skyway Bridge 70 Impact block and UPV tomograph after barge impact cesses ZT 3 1 CAD drawing of as built scanner systems mounted to block clampi
64. and drawings of the ABC test element provided by FDOT 209 PART NUMBER QTY 1 lin T Slot Alum Elem 8in T Slot Alum Elem COMPANT NAME ANT REPROOUCTION IN PART OR AS A WHOLE WEMOUT THE WHITTEN oF NEKT assy udon GEE COMPANT NAME HERE 6 APPUCATION 5 4 210 ANTARES Instrument Carrier Assy SZE DWG NO REV A InstCarAssy SCALE 1 5 WEIGHT SHEET 1 OF 1 H 80 0 y 3X Bundle 7 wire strand Centered each duct Size as available 13 0 3 PLACES 13 50 2 PLACES Type B Duct tapered flaw grout before concrete placement PROPRIETARY AND CONFIDENTIAL THE INFORMATION CONTAINED IN THIS DRAWING IS THE SOLE PROPERTY OF lt NSERT COMPANY NAME HERE gt ANY REPRODUCTION IN PART OR AS A WHOLE WITHOUT THE WRITTEN PERMISSION OF NEXT ASSY USED ON INSERT COMPANY NAME HERE gt IS PROHIBITED APPLICATION 5 4 UNLESS OTHERWISE SPECIFIED DIMENSIONS ARE IN INCHES TOLERANCES FRACTIONAL ANGULAR MACHt BEND t TWO PLACE t THREE PLACE DECIMAL INTERPRET GEOMETRIC TOLERANCING PER MATERIAL FINGH DO NOT SCALE DRAWING 3 211 Type A Duct no flaw in situ grout NAME DATE DRAWN JDN 5 14 12 CHECKED TITLE ENG APPR ANTARES Asymmetric PT Block MFG APPR QA COMMENTS SIZE DWG NO REV SCALE 1 32 WEIGHT SHEET 1 OF 1 2 1 6 Rebar moved at one end after init set OVER TYP 20C 10 3 __ 110 _ _ 100 ___ 80
65. as applied to concrete structures is to verify the structure was built per design This verification often involves simply 56 checking the geometries of the finished structure to ensure formwork was installed accurately and reinforced sufficiently to prevent bulging In many cases however portions of the structure are inaccessible and a reliable measurement cannot be made This includes situations such as concrete pavements foundations and headwalls NDE methods are frequently called upon to reliably perform these thickness measurements Perhaps even more frequently NDE methods are used to verify the placement of reinforcement within the concrete structure in a method known by the generic term as locating The applicability and effectiveness of NDE methods used in the body of this research for these purposes will be investigated Applications of ground penetrating radar for pavement thickness measurements Since ground penetrating radar has recently become popular for structural concrete applications most research into thickness measurements using GPR originates from a desire to rapidly measure pavement thickness Pavement thickness measurements do however present an interesting problem as the desire is usually to measure a single layer of the composite pavement and the individual layers frequently do not have very disparate dielectric constants Given these challenges and the extreme speed and accuracy demands of the pavement industry suc
66. ata output and tight Federal Communications Commission FCC restrictions on operation The research team sought to solve the issues with integrating GPR into the system by creating a new and elegant approach to instrument assimilation that would become an additional method for the automated NDE system to rapidly work with new NDE instruments in the future A Leuze ODSL8 laser distance sensor was also purchased for the first stage system as a surface profilometer but not successfully integrated due to its somewhat unusual method of signal communication Since little additional cost was required to integrate this device it was decided to solve the signal parsing issue and assimilate the device Finally the research team recommended inclusion of a Power PC driven National Instruments NI1742 Smart Camera due to its potential for surface investigation It was decided to integrate the camera and program it to do basic image acquisition and transfer to support future research endeavors 94 Ground Penetrating Radar and Parallel Instrument Operation GPR is one of the most frequently used and often misused NDE techniques for structural concrete GPR has a reputation for being fast accurate and intuitive to deploy and operate The technique is so fast in fact that there has been little call to automate its application to structural concrete Here the technique was included simply as a compliment to the other instruments incorporated in th
67. aterial researchers performed ultrasonic pulse velocity tests on the impact blocks before and after barge impact Using a dense grid of UPV readings Ferraro constructed a three dimensional tomograph of the impact blocks illustrating the change in ultrasonic pulse velocity correlating to the areas of damage experienced by the block 2 70 impact Block Pre4tmpact Impact Block Post4mpact lt amp amp amp ar a4 oF D mB or se a 13 18 x X Y Z Dimenwons in Meters a Ware Velocity Proa 1 2 3 4 5 Figure 2 32 Impact block left and UPV tomograph right after barge impact While the severely damaged area of the block was clear the tomograph gave indications of sustained damage far from the impact on the block as well as the pier upon which it was attached Photo courtesy of C Ferraro 2 Perhaps the most important information gleaned from Ferraro s research is the requirement of highly dense measurement grids when using ultrasonic tomography to generate a clear snapshot of the damage sustained by a structure While most inspectors would have thought it prudent to take a few UPV readings on and around the impact block to verify the level of damage to the structure a full tomographic characterization of the structure better defines the extent of damage and is far more valuable to assessing the need of repair or replacement in the event of catastrophic impact Related NDE Research Automation
68. aw data generated by the scanner and attached 133 NDE instruments into two and three dimensional plots suitable for analysis The ANTARES Processing Suite would be a simple and accessible platform for processing of NDE data generated by ANTARES to be expanded and modified later as new development efforts were sponsored Approach Compared to ANTARES SVW the ANTARES Processing Suite would be relatively straightforward to develop Since no hardware communication and control was involved the software could be run in a single event driven structure Compared to a software loop that continuously executes such as those in ANTARES SVW an event driven loop relinquishes a computer s processor threads when idle greatly reducing hardware demand With the basic architecture designed the framework of the ANTARES Processing Suite was populated using a series of tab controlled panels for each instrument type to be processed A simple file opening interface was generated to read in ANTDAT files for all the native instruments and groups of DZT line data files for the GPR system Each instrument type received customized interfaces with processing and visualization features pertinent to the data to be displayed 134 ANTARES Processing Data File Information Automated Sas for Applied Research and Evaluation of Structures File Name eet Program version 1 0 b DERG Release Date 06 27 2013 Instrument Developed by J Nelson Graduate Resea
69. bedded prisms would be placed within the block A similar procedure would be performed on the end with the surface visible defects The overall design for this block is shown in Figure 7 12 and 7 13 163 fo Dominion Al 76x152x254mm Pervious con embedded A2 76x152x254mm Pervious con embedded A3 152x152x254mm Pervious con embedded A4 Approx 152x152x254mm wet sieved con mix embedded B1 X axis disturbed rebar e B2 Rebar w simulated delaminations c B3 Z axis disturbed rebar EL 1 5 mm thick simulated crack e c2 3 mm thick simulated crack c3 152x152x254mm Pervious con surface c3 visible c4 Approx 152x152x254mm wet sieved con mix surface visible Figure 7 12 3D render of Void and Flaw Detection Block showing relative location of intentional defects 6 Rebar moved ot one end after init set 4 0 COVER TYP 2 0 COVER TYP ol OE g Wet sieved concrete place at pour roughly 6x6x10 pris UNLESS OTHERWISE SPECIFIED DATE NAME DIMENSIONS ARE IN INCHES DRAWN JDN 06 20 13 TOLERANCES TITLE FRACTIONAL CHECKED IWOPLACEDECMAL io ENGAPPR ANTARES Voids and Flaws PLACE t THREE PLACE DECIMAL MFG APPR Block INTERPRET GEOMETRIC QA PROPRIETARY AND CONFIDENTIAL TOLERANCING PER Couns THE INFORMATION CONTAINED IN THIS TER s i THE INFORMATION CONTANED IN M mama All dims nominal SIZE DWG NO REV INSERT COMPANY NAME HERE gt ANY See detail REPRODUCTION IN PART OR
70. ble for the product they produce 19 Figure 1 1 Spalling of concrete due to insufficient cover left and poor consolidation right Both deficiencies could have been detected using NDE methods shortly after construction Photos courtesy of Pioneer Masonry Restoration Company 1 The Florida Department of Transportation FDOT has recently decided to invest in advanced NDE technology research to attempt to overcome the traditional limitations in NDE for structural quality control and quality assessment The subsequent chapters of this work chronicle the development of such a system an automated structural scanning system capable of interfacing with many common NDE instruments to produce spatially dense repeatable easily interpreted data from multiple methods suitable for laboratory system verification and field deployment for structural evaluation Organization of Thesis This thesis can be thought of as a narrative following the development of a complex electromechanical system and supporting battery of software hardware and ancillary projects suitable to verify the proficiency of the system in accomplishing its intended function It would be incomplete however to dispense such a narrative without providing the necessary technical background explaining the operating principles of the NDE instruments themselves and their potential applications as well as briefly 20 discussing the background research into advanced NDE methods
71. canner was to be an experimental field deployment to prove the system was ready and viable for use on in situ structures in the often unpredictable and harsh environment found in the State of Florida This field deployment would be conducted at a location of the research team s choosing with FDOT support and approval The site of the first deployment was chosen to be the FDOT M H Ansley Structures Research Center in Tallahassee Florida Among many other active research projects the Ansley center possessed a full scale construction of an experimental bridge that would be a convenient and useful location for both parties for the first ANTARES field deployment Since this deployment was a trial of the system and was intended to provide FDOT with an indication of whether or not a more permanent field deployment logistics support system is warranted for the ANTARES system support for the initial deployment was very limited The research team needed to invest a serious amount of time in planning and logistics on how to physically transport and deploy the large heavy scanner frame using readily available equipment Once at the Ansley center the system was deployed and activated and a series of scans were performed Since the field deployment was limited to a single day only the most useful NDE instruments were used and the scanned areas were sparse in comparison with the laboratory evaluation block scans 183 Despite the aforementioned challen
72. cent method in the world of concrete NDE Geological surveyors have been employing GPR for decades to detect sinkholes utilities and other anomalies hidden below ground but these applications utilize large low frequency antennae that would not resolve any fine near surface details as required in structural concrete NDE As smaller higher frequency GPR antennae have emerged into the market concrete inspectors have embraced the technology and the technique is quickly becoming one of the most popular NDE methods for structural concrete GPR is based on the principle of reflection of electromagnetic waves off of interfaces within the structure As with any wave reflection technique an approximate 43 impulse of energy is generated upon the structure and allowed to propagate through it Reflected waves are generated on boundaries where step changes in the velocity with which the wave may propagate occur and recorded upon their return using a receiving transducer The velocity of an electromagnetic wave in any material is described by the material s dielectric permittivity and magnetic susceptibility u Using Maxwell s equations the velocity takes the following form E Permittivity ratio Eo 2s2 1 o Permittivity of Free Space 8 85x107 1 Nm v Teo u Equation 2 5 ErEoHrho ur Magnetic susceptibility ratio C 0 N Uy Mag susceptibility of Free Space 4nx10 7 zz In free space t
73. city Compared to the point by point scan mode used with most other instruments this operating scheme results in highly dense far more rapid data collection although the data points may be unevenly spaced and some noise in the data due to vibration in the moving scanner is generated Additionally the laser profilometer is prone to scattering when incident on a sharp edge or sudden elevation change resulting in an erroneous reported value To overcome these limitations the laser profilometer data is plotted using the true scanner coordinates at the time of instrument read as given in the ANTDAT file and some filtering usually 4 points using the provided moving average filter is required 140 GPR processing Almost every major manufacturer of GPR equipment has at some point produced their own data visualization software These software packages are usually quite powerful and very well refined Despite the availability of this software it was decided to include some limited GPR processing ability in the ANTARES Processing Suite GPR Processing posed a unique challenge in that the software was required to work with the GSSI standard binary file format DZT due to the parallel operation nature of the SIR3000 GPR unit The GPR processing section of the ANTARES Processing Suite would need to open a group of DZT files where each file corresponds to a single line in a scanned grid and map these lines to a grid settings cluster imported from
74. computer for the first time the Splash Screen launches the Configuration Settings Screen automatically Configuration Settings Screen The Configuration Settings Screen allows the operator to import export and modify the configuration settings cluster for the ANTARES SVW software The interface with this data cluster is presented as a series of tabs with various numeric and drop down menu controls Default settings for the scanners themselves including ports calibration constants and default speed and acceleration settings port settings for each native instrument and default scan modes for each instrument are generated altered loaded and saved from this panel One of the most important settings generated in this screen is the COM port and cDAQ assignments for each scanner When these values are changed the Configuration Settings Screen automatically checks each port and verifies the selection is correct indicating success or failure with a red or green LED style indicator as shown in Figure 6 19 Similarly instrument default settings are changed as shown in Figure 6 20 125 ANTARES SVW FDOT State Materials Offic ANTARES a mQ em e S ea ead ee aes ET L cDAQS181 ANTARES ANTARES SVW FDOT State Poe a O ANTARES ANTARES SVW is Ultras E rotaya Ef Down Pontey Figure 6 20 Configuration Settings Screen modifying a native instrument s default port an
75. condary to function in software designed by an engineering research team however to create an intuitive interface for maximum usability significant time and effort was invested in creating a simple intuitive cohesive and attractive interface for the ANTARES SVW software This section discusses the design and use of the main software interfaces of ANTARES SVW The interfaces will be discussed in the order in which the operator is likely to encounter them in a user s manual style rather than in order of development use or complexity Splash Screen As discussed earlier in this chapter the Splash Screen VI is an 18 step sequence of events necessary to initialize the ANTARES hardware and software Each of these steps is completely automatic and requires no operator input For this reason it was decided to encase these long but necessary steps in a familiar software structure The VI was designed as a loading screen as shown in Figure 6 18 ANTARES SYW Figure 6 18 ANTARES SVW splash screen with loading bar and feedback to operator 124 The loading screen displays the progress of the initialization in a progress bar and provides the operator with text feedback about the procedure including any communication errors and startup issues One of the most important functions of this screen is to successfully load the configuration settings cluster If this load is unsuccessful as usually occurs when ANTARES SVW is launched on a new
76. d like a video through the plot s thickness as shown in Figure 6 32 The Tomograph display presented in Figure 6 33 allows visibility in the XZ or the YZ plane with the ability to manually step or automatically fly through the dimension not displayed Depth and distance information is provided from the GSSI instrument using their own calculations for GPR flight time and an integrated position encoder and the data array imported from the DZT file is already in a scaled format suitable for display 142 ANTARES Processing Data File Information File Name GPR_Rebar_SideA_X_08192013 Fly Speed Date Time 8 19 2013 10 49 AM Normal Instrument GPR GSSI SIR 3000 Program Ver Program Version 1 0b E ope se 1 1 1 1 0 50 200 250 100 150 Figure 6 32 Fly through C Scan GPR display showing a layer of dense rebar in a test specimen Depth of the display is controlled using the vertical slider on the left side of the display 143 ANTARES Processing Data File Information File Name GPR Rebar SideA Y 08192013 Instrument GPR GSS1SR3000 Program Ver Program Version 1 0b 0 ma r es T 1637 47 8125 36939 Figure 6 33 Tomograph display of GPR data showing a dense rebar mat in the YZ plane Two layers of rebar running transverse to the display are visible as a series of white hyperbolic traces on the pink background Shear wave ultrasound processing Shear wave ultrasound
77. d Test Method for Penetration Resistance of Hardened Concrete West Conshohocken PA ASTM International 2003 48 American Society for Testing and Materials ASTM C469 10 Standard Test Method for Static Modulus of Elasticity and Poisson s Ratio of Concrete in Compression West Conshohocken PA ASTM International 2010 49 Pioneer Masonry Restoration Comapny Inc Online Available www pioneermasonry com resources html Accessed 1 June 2013 222 BIOGRAPHICAL SKETCH Jordan Nelson was born in Tallahassee Florida to parents Clifton Jr and Susan Nelson He graduated 10 in his class from Lawton M Chiles High School there in May 2007 and began pursuing his Bachelor of Science degree in mechanical engineering at the University of Florida in August After receiving his degree in December 2011 he was accepted to the graduate school and began working on his Master of Engineering degree with his major in mechanical engineering but his research primarily civil engineering oriented in July 2012 A long time employee of the Florida Department of Transportation Jordan worked at the FDOT Structures Research Center in Tallahassee before and after his freshman year of college under Marc Ansley who fostered his early interest in structural testing fabrication and mechatronics Beginning his second year at the University of Florida on Mr Ansley s recommendation he began work at the FDOT State Materials Office the sponsoring agency of this
78. d scanning operation settings 126 Main Interface The main interface of ANTARES SVW is designed to provide the operator with complete control over the operation of the ANTARES scanner systems as well as display feedback about the position and instrument head status from scanners themselves and incoming data from the NDE instrument in use during a scan The interface is designed to be compact robust and easy to navigate with clear structure and order as shown in Figure 6 21 ANTARES SVW FDOT State Materials Office File Edit Operate Tools Window Help x Exit Program ANTAR a ES Time per Data Point Ea V IN 2 609 sec a Time per Line 172 219 sec Time Remaining 4hr 33min ETG 8 21 2013 9 39PM End of Line Hold na Pause Scan CD x Abort Scan fay 4 Down Point By Point Auto Scan Mode i independent Haw r ke a A 4 A Le Saye J We e Grid Wizard J f a lt Instrument Test Run Load Grid Origin fen 5 Lines Run in Y i 200 mm L int Line Y 200 mm Points ines Sy Usescs x 1 iia ER EL 1 Eo aE ST Da 400 500 600 700 800 900 1000 1100 1200 y a bi BEGIN I 8 C Doc Ultra_Rebar_SideB_08212013 ANTDAT peer Line Point mm Figure 6 21 Main interface of ANTARES SVW during a shear wave ultrasound scan The main interface provides settings and controls necessary to setup t
79. dix A Noteworthy subroutines and software functions will be discussed however the complexity of the program as a whole prohibits an in depth discussion of the code in its entirety In order to keep the front panel the interface visible to the software operator as clean and simple as possible the main panel and the limited number of modal popup windows used in the interface were designed around a common theme and operated by event driven controls that were clearly labeled and identified Controls that perform the same function across multiple windows are identified consistently and located in the same areas All of this effort was designed to make operating the ANTARES system highly intuitive The most important function of ANTARES SVW is to produce useful NDE data One improvement ANTARES SVW made over ScannerVoll Wizzard is the use of a common file format for the data output of all instruments The ANTDAT file type was 108 designed specifically for the ANTARES system and is a universal ASCII file format that can be opened using almost any text based processing program and easily interpreted by other operators meaning that other agencies and personnel wishing to utilize data generated by ANTARES SVW can do so using their own processing software if desired Software Tiers Introduction ANTARES SVW is a relatively complex LabVIEW application comprised of over 660 subroutines contained in 135 SubVIs written during the course of this res
80. e 2 33 Horizontal left and vertical right scanning systems developed at BAM for use on concrete structures Photo courtesy of BAM 33 72 The research and development performed at BAM and the cooperation between their organization the Florida Department of Transportation and the University of Florida have led directly to the developments contained in this Thesis Indeed the basis of the scanning system used in this work comes from designs generated at BAM Florida Department of Transportation State Materials Office FDOT SMO Gainesville Florida Following in the groundbreaking research by BAM the Florida Department of Transportation State Materials Office FDOT SMO in Gainesville Florida requested a research program in the early 2000s to validate NDE equipment used throughout the State The result was an automated scanning system similar to that used in BAM designed and developed by one of their scientists Daniel Algernon In the course of his post doctoral research at the University of Florida Dr Algernon and his team developed a pair of precision automated scanning systems with pneumatic instrument application devices The scanning systems were operated remotely using programs developed by Algernon in National Instruments LabVIEW The system ultimately operated and collected data from a covermeter impact echo and shear wave ultrasound device and verified the instruments using four laboratory specimen blocks constructed wit
81. e ANTARES system It is written or loaded manually in the Configuration Settings Screen loaded automatically in the Splash Screen and written automatically when altered within operation of the Main Interface Since the file can be moved and imported the ANTARES SVW software can be installed and configured on a new PC rapidly and the settings migrated over easily allowing almost any computer to be quickly adapted to the ANTARES system The file is produced by bundling all the settings established on the Configuration Settings Screen into a single data cluster and flattening this data cluster into a big 130 endian format binary string This binary string is written to a file and interpreted by loading the file and unflattening the string into a configuration settings data cluster ANTARES Grid File ANTGRD The ANTARES Grid File ANTGRD is a binary file formed by flatting a grid settings data cluster The grid settings data cluster contains the coordinates of the grid origin the orientation of the grid lines and the point and line count and spacing ANTARES Grid Files may be written or loaded from the Main Interface in the grid setup section Grid in Progress Grid Origin Lines Figure 6 23 Grid configuration cluster as it appears in the Grid Generation Wizard This data cluster can be written to and read directly from the ANTGRD file The grid settings data cluster is unique in that it is a complete container of cont
82. e automated NDE scanning system and due to the fact that FDOT was already in possession of a GPR system intended to be adapted to the scanning systems The GPR system to be adapted was a Geophysical Survey Systems Inc GSSI SIR3000 self contained operation and display unit with a 2 6GHz remote antennae The antennae came with a detachable rolling cart with integral position encoder as shown in Figure 5 1 Figure 5 1 GSSI SIR3000 system with 2 6GHz antennae to be integrated with automated scanning system Photo by J Nelson 95 The physical size of the GPR system was the first obstacle to its integration with the automated scanning system The SIR3000 unit itself was too large and heavy to ride on the scanners instrument heads and still required management of a thick 6 5m long multiconductor cable and a 15cm x 15cm x 5cm signal conditioner used to interface with the antennae itself It was decided to allow the signal conditioner to ride loose in the new instrument carrier and leave the SIR3000 stationary away from the scanner Routing the custom built cable through the scanner guideways before every scan was infeasible so the cable would lie near the cable and be carefully fed and retrieved during scanning The antennae and rolling cart were integrated using a specially designed mount that allowed an extension adapter supplied by GSSI to mate the cart to the pneumatic instrument head A rolling limit switch and self retrac
83. e automated scanning software generated an idea to make the system even more flexible and useful Instead of the software controlling both the scanner systems and the GPR input output operations it was decided to allow the software to operate the scanner without ever interfacing with the instrument to follow a user defined grid in a feed forward control scheme and output only coordinates and timestamps as its data file output The data files saved to the SIR3000 unit could then be mapped to the data points generated by the scanner operating software This methodology was dubbed parallel operation the support of which could allow new and previously incompatible instruments to be rapidly adapted to the automated scanning system The software support and potential capabilities of this methodology are discussed in Chapter 6 97 Laser Profilometer Incorporation of the laser profilometer purchased for the Phase system but never fully integrated was deemed a side project of this research effort if time permitted The instrument was revisited shortly before the reprogramming effort was started and determined to be fairly straightforward to integrate and quite useful The device itself a Leuze ODSL8 laser distance sensor was small in form robust and accurate at measuring distances down to 0 1mm The only major difficulty was parsing its data output a 14 bit measurement carried over a 2 byte stream where the high and low preceding bit
84. e of the main disadvantages of eddy current methods is that the method is insensitive to non magnetic features within the concrete surface Water pockets air voids organic contaminants and non ferromagnetic reinforcement do not exhibit the induced eddy current phenomenon required for covermeter operation This disadvantage can actually become a major advantage to an inspector when this technique is combined with another method that is sensitive to for example density changes as in acoustic methods where steel air water etc are all resolved within the structure but are difficult to tell apart from one another If the data from a covermeter and an ultrasound array were overlaid it would be very apparent to the operator which features were ferromagnetic e g steel rebar and which were some other nonmagnetic material It is important to note that covermeters are extremely effective at resolving cylindrical ferromagnetic rods of infinite length in shallow cover situations where the excitation and sensing coils are both centered over the centerline of the rod In situations where steel features have a non circular cross section short lengths are in deep cover dependent upon the sensor head used or are oriented perpendicular to the instrument head of acommon cover meter begins to lose accuracy Previous research has quantified this effect very well with the particular covermeter used in this research as illustrated in Figure 2 7 9 31 C
85. e3sig Figure 8 7 Laser profile of the ABC test element surface at location 1 Note the circular artifact at line 120 point 15 resulting from a grout port through the bridge deck 190 Figure 8 8 Surface of bridge deck at location 1 as profiled by the laser scan Photo by J Nelson After analyzing the ultrasound data some interesting characteristics of the element were noted The very rough roadway surface had indeed impacted the ultrasound data collection and resulted in a C Scan image with significantly more noise than a typical image produced from a laboratory specimen This coupled with an already coarse grid made data interpretation somewhat difficult Additionally the back surface of the bridge deck did not produce the kind of strong reflection usually seen in ultrasound C Scan images when performing the shadowing technique on laboratory specimens The back wall reflection was weak in the majority of its area with regions of more intense reflection It was determined that this was due to the fact that the test element was buried in the ground and therefore resting on compacted earth The shear stress wave was partially reflecting off the bottom edge of the slab but some of its energy was transmitted into the earth 191 To effectively analyze the data at location 1 the depth of the ultrasound C Scan display was set to 155mm where the interface between the girder and the bridge deck was visible In Hilbert transform space
86. earch effort This program represents over 300 operator hours of programming time closely following a programming hierarchy shown in Figure 6 2 Program Architecture Main Program Data File Human Data Interface terface 1 X i Manual Real Time Emergency Scanner Autoscan 1 Task Handling functions Visualization Movement Handler VI Ara ame Hardware Handling mamei ponm j 1 Low Level 1 O Subroutines Fests PERE gz Figure 6 2 Basic tiered architecture of the ANTARES SVW LabVIEW application Low level I O subroutines The most basic SubVIs included in ANTARES SVW the low level I O subroutines are designed to perform basic command and acquisition functions with the scanner hardware and NDE instruments They include command and feedback 109 subroutines for the scanner servo motors and encoders instrument head pneumatic cylinder relay and NDE instruments serial USB and NPSV interfaces The scanner servo motor subroutines operate the servo motors and receive feedback from their onboard encoders by communicating over serial interface using Cool Muscle Language CML The two VIs in this class CML Command and CML Query parse a plain text command given by an enumerator control into Cool Muscle Language and write it to the serial port in the case of CML Command or perform a read and parse operation returned as an enumerator indicator in the case of CML Query These VI
87. eck and the grout layer falling into two pieces as soon as it was removed The grout portions of these two cores with unbounded regions of the grout surface visible are shown in Figure 8 13 A core was also taken from a non suspect area on one of the girders and was found to be well bonded and intact as shown in Figure 8 14 Due to time constraints cores were not taken from ABC Location 2 however the indications of a delaminated area in this regions were far more certain With these results it is clear that the ANTARES system performed admirably on its first field trial with despite less than ideal operating conditions 195 Figure 8 13 Drilled core specimens from ABC Element Location 1 showing boundaries of regions with grout delamination Photos courtesy of FDOT M H Ansley Structures Research Center Figure 8 14 Drilled core of Location 1 in an area predicted by ANTARES to have a proper grout interface Photos courtesy of FDOT M H Ansley Structures Research Center 196 CHAPTER 9 CONCLUDING REMARKS The results of this research and development effort for the sponsoring agency can be thought of as simply a more advanced tool for quality control of concrete structures investigation of infrastructure components in situ and a platform for evaluating new NDE techniques as they come available However it is clear that this project has accomplished much more than that ANTARES is an expansive and comprehensive analysis syste
88. econdary processing at a depth of 90mm It was found that moving the depth control while focusing on regions of the block better allowed the user to discern whether pixel color changes were random or followed an actual defect suggesting that perhaps video displays of the ANTARES data visualizations would be more effective at communicating the processed data than still images alone Continuing through the depth of the block at a depth of 120mm the traces of the three rebar elements were easy to discern as shown in Figure 7 29 The trace of the Z axis moving rebar B3 indicated a significant region of disturbed concrete left behind after the bar s movement Though the X axis moving rebar B1 was not successfully moved the intended 25mm the C Scan image indicated some damage to the surrounding concrete near the moved end The simulated entrapped air located on the middle rebar B2 was somewhat difficult to discern after performing an absolute value 179 transform the tubing located on the side of the rebar from the perspective of the ultrasound transducers became very apparent though it was difficult to see in other view types The use of a Hilbert transform visualization shown in Figure 7 30 allowed a very clear image of the other section of tubing located on the middle rebar despite the defect being behind the rebar and masked from the ultrasound instrument s view Figure 7 30 Ultrasound C Scan image of Void and Flaw Detection Bl
89. eeeeeeeeeeeeteeeseeeeseeee 172 7 22 Laser C Scan image of defected block surface with cracks and 7 24 15 7 25 7 26 XZ Plane GPR tomograph of the Void and Flaw Detection block Gee IO 7 27 GPR C Scan image of Void and Flaw Detection block ceseseseeeeeees 177 7 28 Initial ultrasound data processing for Void and Flaw Detection Block 178 7 29 Ultrasound C Scan image with no secondary processing at a depth of 90mm 179 16 me flatbed trailer outside of the 184 187 191 192 A 2 A 3 A 4 A 8 A 9 A 10 A 11 STH C ie ila at a of 120mm TENN dense reinforcement on Drilled core of Location 1 in an area ae h i ANTARES to have a pips eaae e E E N E aaa loaa 196 Covermeter Single Point Acquisition V iicicsticccscdisssccndsdesscadiacareadeeeadadiceciucieicas 199 Laser Single Point Acquisition Vl hcccaccesssscisevescssvaessdnsecosacnsncenninbecearectesseidvaentons MUU SCS Startup VI Sequence 0 scccssesccesceseeeeceseeseeeseeeeeeaeeeeeseessetseeesetseteenees LOT SCS Startup VI Sequence 1 main l00p s secsseceseseeesseeeseeeeeeseeeesseeees 202 CML Query Vl PEN IEE E R A E AT A AN A EA N T E Pneumatic Head Command Vl ccicccccntcssusostacsnrasccsnnaninactasasaeddnaniuadsecnsesrinensneun LOG P eumatic Head Query Vl cd ciieieeit inden AG ANTDAT File Creation Vl sisscssocereaesrsvsvavseseseaesvaezenvseseavavavdeavnesesesvevzvseavaea
90. elson _ a WR n a Figure 4 4 Instrument installation process Note the panel the instrument is mounted on forms the back panel of the polycarbonate enclosure Photos by J Nelson 85 Establishing a Communication Standard Given the overly complicated and fragile system in which the PXI chassis was connected to the scanner systems and NDE instruments discussed in the opening section of this chapter it was decided that the entire communications protocol would have to be redesigned Furthermore one of the most useful instruments integrated with the scanner systems the shear wave ultrasound device was limited by its USB protocol interface to a meager distance of 1 5m from the PXI For field deployment using the Phase system in a difficult to access position on a bridge pier for example there would be no feasible arrangement for the PXI to be sufficiently close to the scanner s operating location to execute a successful ultrasound scan In order to extend the operating distance between the operating computer and the scanner system eliminate signal degradation due to transmitting analog voltages over long cable lengths and increase the maximum distance from the instrument to the operating computer it was decided to convert both of the scanner systems entirely to a local area network LAN based communication standard A LAN protocol offered a number of advantages over other digital or analog communication schemes
91. en concrete surfaces during use Covermeter Palette Component Covermeter sensing head 2 Standard pneumatic head mounting bracket 3 Covermeter Readout Box inverted for palette visibility Standard instrument mounting palette 5 LAN serial device server 6 5 VDC power supply 12 VDC power supply Figure 4 7 Proceq Profometer 5 covermeter modified for use with the scanner systems Photo by J Nelson Figure 4 8 Covermeter assembled on palette and packed in watertight instrument case Photo by J Nelson 91 The Germann Instruments EyeCON shear wave ultrasound device had posed problems in the past when attempting to interface over a Silex USB device server The device communicates its data as a 2048 element array of 8 bit integers over what is known as bulk transfer protocol a USB transfer mode that until recently has not been supported by USB device servers Fortunately shortly before the beginning of this project Silex launched a new USB device server that supported the transfer mode and allowed the EyYeCON to communicate over a network eliminating its previously short cable restriction The EyeCON device retained its original rechargeable internal battery which was capable of operating the device over multiple area scans before depleting and was mounted to a polycarbonate panel with its USB device server using removable hook and loop ties as shown in Figure 4 9 and 4 10 Ultrasound Palette Component
92. encss ZOT CANTO T PGT o Y aaa SOM 17 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering DEVELOPMENT OF AN ADVANCED AUTOMATED NONDESTRUCTIVE EVALUATION SYSTEM FOR INVESTIGATION OF CONCRETE STRUCTURES AND VALIDATION OF CONCRETE NDE INSTRUMENTS By Jordan D Nelson December 2013 Chair Peter Ifju Cochair Christopher Ferraro Major Mechanical Engineering Nondestructive evaluation NDE of concrete has historically been a niche discipline within civil engineering and a process carried out by an individual operator in the field to evaluate the extent of known or suspected flaws within a concrete structure NDE techniques such as Ground Penetrating Radar Ultrasound and Eddy Current methods depend heavily on the expertise and diligence of the operator to maintain accuracy and are prone to producing sparse subjective data In recent years the Florida Department of Transportation FDOT has called for NDE techniques to be applied in an automated manner to produce denser more precise and more defensible data suitable for structural evaluation and quality control in a variety of applications This body of work chronicles the development of such a system for use by FDOT a precision automated scanning system coupled with state of the art NDE instruments standardized laboratory evaluation specimens and purpose built operation and data proc
93. ents moving the scanner servo motors and reading and writing ANTARES Data Files Most images are code snippets and if preserved digitally as a PNG file can be dragged onto a compatible LabVIEW block diagram to generate the code 198 r Initialize and Read Defa v i Covermeter COM Settings Timeout s Figure A 1 Covermeter Single Point Acquisition VI 199 om eee ese ee O 2m i E A et Tue vb VISA resource name 1 Flush the serial buffer to read instantaneous data Read in two bytes 2 Convert the hexadecimal string into two 8 element boolean arrays 3 Use the first element in the array corresponding to Byte Index 0 to determine whether Byte Index 0 is the High or Low byte The first element in the array corresponds to the LSB Concatenate the last seven elements in each boolean array in the appropriate order 4 Convert the 14 element boolean array to a 14 bit number adapt to double precision and shift the decimal point appropriately Apply logic to replace the default 0 reading with Inf when the measurement is out of range 1 29 13 J Nelson University of Florida Figure A 2 Laser Single Point Acquisition VI 200 ooo0oooo0ooo0oo o la Initialize and Read Defaul PBL CheckBulkBuffer Figure A 3 Ultrasound Single Point Acquisition VI Startup sequence strobed light is visual feedback for opera
94. eriments Note the water line and the relative location of top middle and bottom test levels Photo courtesy of C Ferraro 25 Concrete blocks partially submerged in a sulfate solution were tested in parallel with identical panels partially submerged in lime water as a control The blocks were tested in a grid using surface impact echo and direct UPV techniques as well as other non destructive methods The results for the NDE methods were compared with destructive data from compressive strength static elastic modulus and pressure tension tests collected from cores of the blocks 53 Wave Velocity vs Age Block 18 0 65 W CM Ca OH lt or om ii E T eu cee Qe 3200 3000 i 10 20 30 40 50 Age Weeks Wave Velocity vs Age Block 20 0 65 W CM SO 5000 4800 4600 4400 4200 i a Ta a o oO _ ji oo O M Pil B sgn PRS 3400 6 Cn mome oO 8 3200 4 3000 0 10 20 30 40 50 Age Weeks O UPV Top O UPV Middle t UPV Bottom O UPV immersion o iIE Top 0 IE Middie t IE Bottom 0 IE Immersion Figure 2 22 Wave velocity results from lime water immersed top and sulfate solution immersed bottom concrete blocks Note the decay in wave velocity for the sulfate exposed block 25 54 Boyd s experiment contained several interesting results of note to the research contained in this thesis First and most obv
95. es such as segmental bridges due to their relatively easy installation procedure a manufacturing procedure that allows very long development lengths and comparatively low cost Multiple strands are combined to form a single tendon contained either internally in ducts embedded in the concrete structure or externally in ducts spanning the interior space of a segment from one diaphragm to the next Depending on the design of the segments multiple strands may be used for each tendon and multiple tendons may be used in each segment Se ee Pier Figure 2 27 Typical tendon locations Internal tendon ducts in a hammerhead pier left and external tendon ducts in a box girder right 29 After installation and tensioning of the tendons the ducts are filled and pressurized with specialized grout designed to promote load transfer between the tendon and the structure and most importantly protect the mild steel strands from corrosion The result is a multi layer protection for the tendon that is usually highly effective in preventing corrosion of the post tensioning tendon 64 Level 1 Exterior Surface Figure 2 28 Layers of corrosion protection in a typical post tensioning tendon 29 Recently a number of premature failures of post tensioning tendons has been noted across the United States Many of these failures were traced back to improper grouting Void spaces left in the grout within a post tensioning duct allow infiltration
96. essing software that together comprise the FDOT Automated Non Destructive Testing for Applied Research and Evaluation of Structures ANTARES system CHAPTER 1 INTRODUCTION Research Motivation Non destructive evaluation NDE of concrete structures is not a recent application of engineering science but historically its application has been limited due to a number of difficulties including modeling the behavior of an inhomogeneous time dependent material such as concrete Additionally NDE techniques are most frequently performed manually by trained technicians whose method of application and interpretation will invariably affect the quality of the reported data For these reasons NDE methods such as eddy current methods ultrasound and ground penetrating radar GPR are typically called upon by governing agencies only rarely and only to identify the extent of known defects or defects suspected to be present Concrete NDE techniques have shown the potential to reliably identify many forms of structural deficiency including poor material quality poor consolidation insufficient cover thickness of protective concrete layer over reinforcing steel and subsurface cracking and delamination due to damage such as that shown in Figure 1 1 The application of such methods for quality control purposes however requires spatially dense exhaustive data collection performed in an irrefutable manner suitable for holding construction agencies responsi
97. f a concrete structure in great detail with an easily interpreted data display Figure 2 17 Modern handheld GPR unit with built in color line display Photo courtesy of Geophysical Survey Systems Inc 19 Despite the popularity and flexibility of modern GPR technology the method still has some limitations First and foremost as with any wave reflection method GPR is sensitive to the masking effect In the case of GPR this is most commonly seen as a 45 case where an air or water pocket is masked by rebar the GPR reflection display is consistent with a normal rebar reflection and the defect is undetected Additionally a GPR antenna is sensitive to the manner in which it is coupled to the concrete surface Since the electromagnetic wave must first cross an air barrier before reaching the concrete surface the impulse wave is often attenuated before ever reaching the surface and all defects within the structure are effectively masked This effect is exacerbated in so called air coupled systems commonly used in pavement survey vehicles in which a relatively large gap is placed between the antennae and the surface and higher energy radar waves are used to compensate Finally free moisture in concrete serves to make the previous two limitations even worse by increasing its dielectric permittivity leading to a situation where the surface of the concrete itself becomes an effective mask for all targets in the structure in the case of
98. fer to EX 1 Stm of Slab for spacing Concrete Block Concrete Black we Pe pen See Ex 1 Extruded Polystyrene Form for Polyethylene Form width to be 2 in k weld pipe amp nut to washer r i ne of Shear I 2 Top of Slab cket Beyond 7 CO Maximum Negative Moment Leveling A Maximum Positive Moment Leveling Maximum Negative Moment Lifting Maximum Positive Moment Lifting Shear Pocket Spacing Notes 1 Rigging for lifting shal be assembled such that each lifting point shall not be subjected to any shear The panel weight is approximately equally distrfouted lifting device releasing rigging verify that each leveling bolt is in cont with the beam 3 When installing bolts add lubricant as necessary to achleve free turning of bolt by hand Remove bolt after haunches have been grouted amp cured then grout dlockout 4C Bars Refer to line of Shear EX 1 for spacing Pocket Beyond ay Slab Grout haunches 6x6x Stee Plate Ea Leveling Bolt d bolt or add bolt cap to create rounded edge Bolt ta be sufficiently greased before insertion to prevent bonding ta haunch Grout and to facilitate Leveling Bolt Detail poze sen Precast Full Depth Deck Details PENR STATE OF FLORIDA DEPARTMENT OF TRANSPORTATION Tivancra POET To ana Vibe TM2OI2 PE aISD AM 217 ENTOJ PRECAST SUPERSTRUCTURE WOR MELAS NAPED T Sune SOES ONT NDER RULE 61615 23 003 FAL IGNEO AND SEALE
99. g The flexibility of vision based data acquisition meant that the SCS could in the future be used to perform any number of surface evaluation tasks when used to scan structural concrete The research team recognized one application in particular that could be extremely valuable for field scanning operations orienting the scanner ona structure Historically attempting to repeat a nondestructive evaluation of a section of a structure either manually or using an automated scanner after some time had passed was extremely difficult as intimate knowledge of the previous grid layout and 103 origin were required The SCS instrument offered the potential to locate distinguishing marks on a surface and calculate an origin relative to the scanner coordinate system and an angle of rotation The research team designed etched aluminum placards with an origin marker and a space for an identification label specifically for this purpose Since the SCS requires a calibration at run time to calculate the size of surface defects essentially a conversion of pixels to millimeters that changes based on the distance from the camera aperture to the surface a calibration grid was also included on the SCS placard design shown in Figure 5 9 Future research should include development of vision processing algorithms to fully utilize this significant capability FDOT ANTARES Automated NDT Scan Area DO NOT REMOVE THIS PLACARD WITHOUT APPROVAL FROM THE ST
100. g Range Off g g Figure 2 4 Single point distance measurement using a laser distance sensor 5 A three dimensional contour can be created from a two dimensional scanning system with a single point laser measurement system To cover larger distances and make measurements at highway speed pavement surface measurements typically achieve a three dimensional contour using an array of laser sensors or a laser line projection system mounted to a survey vehicle Fugro Company s Automatic Road Analyzer ARAN system shown in Figure 2 5 uses such a system in conjunction with several other instruments to evaluate pavement surfaces at high speed and is an example of an entire class of survey vehicles built with technology such as this 28 Figure 2 5 Rear mounted laser surface contour system on Automatic Road Analyzer ARAN system Photo courtesy of Fugro Roadware 6 Eddy Current Methods Eddy current methods are a popular subset of a larger class of magnetism based NDE methods used in reinforced concrete The technique is based upon the tendency of long ferromagnetic rods to emanate a magnetic field when small electric currents are induced due to a nearby moving magnetic field The effect is described by Faraday s laws for mutual inductance The eddy current method is commercially available in a simplified device referred to as a covermeter a relatively inexpensive and accessible instrument typically used to 29 accurately locate
101. g authorities in this situation are limited and essentially boil down to two main options temporarily remove the structure from service in order to obtain core specimens to test traditionally or estimate the desired properties quickly and unobtrusively using a non destructive evaluation method Elastic properties of concrete using ultrasonic pulse velocity Acoustic methods are popular for estimating elastic properties since the speed of a stress wave in an isotropic elastic material is described analytically by its elastic properties as shown in the Introduction of the Acoustic Methods section A cornerstone of research to this end was completed in 1990 by Sandor Popovics 21 Carefully fabricated concrete specimens were subjected by Popovics and his assistants to UPV testing from several different transducer frequencies The elastic moduli and densities of concrete were calculated from the UPV results and used to estimate compressive strength The tendency of the measurements to be affected by material inhomogeneity directional preference and internal stress was addressed in detail The results were clear while an accurate determination of the speed of a P Wave or S Wave in concrete can be used to estimate the elastic or shear modulus with a high degree of reliability estimating concrete strength is not as straightforward While there is a correlation 47 between ultrasonic pulse velocity and compressive strength it varies between concrete mi
102. ge research it was designed to fulfill FDOT requested a more comprehensive robust and user friendly software package for the second phase of their automated NDE research The research team proposed a new operating software package designed from the ground up in National Instruments LabVIEW to be as user friendly and robust as possible while still allowing expansion in the future Additionally the software was to fulfill the need of real time data visualization while using the scanner as an additional source of feedback for the operator An additional software package for processing data into two and three dimensional displays was also to be developed as part of this research effort Construction of Additional NDE Evaluation Blocks The NDE Evaluation Blocks produced for FDOT by Algernon et al are an invaluable tool to compare NDE systems and methods when applied to controlled well documented structural features and defects FDOT wanted to expand upon this capability by producing two additional blocks The first block would address evaluation of internal polypropylene post tensioning ducts where potential defects include air voids water inclusions and soft grout in the grouted tendon ducts The second block would be used to evaluate NDE instrument effectiveness in locating cracks honeycombing and delamination of reinforcing steel Field Deployment After all the aforementioned improvement to the automated NDE system were implemented succes
103. ges once the scanner system was returned safely the data was processed and analyzed and results were sent to the personnel at the Ansley center The personnel there performed a series of coring operations on the bridge deck to test the validity of the analysis performed using the ANTARES system Deployment Site In one method of Accelerated Bridge Construction ABC a bridge is constructed using precast deck elements lowered onto precast concrete girders already in place as shown in Figure 8 1 The deck elements are levelled using a series of leveling bolts passed through the deck and in contact with the girders upon which they rest High strength grout is pumped under pressure to fill the interface between each girder and the deck elements it supports A similar grouting operation bonds the joints between adjacent deck elements together and fills the spaces left when the levelling bolts are removed This type of bridge construction offers a number of advantages over cast in place bridge decks specifically in reduced time to open the bridge to traffic z NAS N Al es E o ee _ gt x sam x m sama a cig n iir m Ta Figure 8 1 Prefabricated bridge deck panels are lowered onto concrete girders during an Accelerated Bridge Construction ABC operation Photo courtesy of STRUCTURE Magazine 38 184 A test element of such a bridge design was fabricated at the FDOT M H Ansley Structures Research Center in
104. h applications of GPR are worthwhile to consider as an automated GPR system to measure structural concrete element thickness is developed While Al Qadi and Lahouar focused their research primarily on hot mix asphalt HMA pavements their methods for interpreting GPR data to determine pavement thickness are certainly valid for concrete In the course of their research Al Qadi and Lahouar showed that their GPR measurements for the thickness of an HMA layer taken using a high speed air coupled GPR system closely followed the actual layer thickness 57 Their method for showing this involved scanning a special test road in Virginia which has multiple copper plates placed under the HMA layer on top of the base layer The copper plates show up exceptionally well on GPR and can be taken as a true reflection when attempting to distinguish reflections between pavement layers 20 Amplitude 12000 10000 8000 6000 4000 2000 0 2000 4000 6000 8000 ERS Surface Reflection i f Spurious Reflection OGDL 21 A Reflection Multiple Reflections from copper plates Figure 2 24 Data taken in Virginia by Al Qadi and Lahouar Note the clear reflections from the embedded copper plates 20 58 A number of important details are to be noted from this GPR research First the processing techniques are readily applied to the GPR processing algorithms to be developed herein Secondly the idea of using a metal p
105. h known features and defects The result was a first stage NDE validation facility for use by the State of Florida 9 73 Oe CH ym Out Gore ih fede cu g Figure 2 34 Scanning system left and software right used by Algernon This system formed the starting point of the research discussed in this thesis Photo courtesy of D Algernon 9 The work by Algernon et al lead directly into the research discussed in this thesis The first stage facility referred to subsequently as the Phase system developed by Algernon was an excellent starting point for the FDOT SMO as they ventured into automated NDE research The SMO required improvements to hardware and communication systems of the scanner systems themselves the incorporation of new instruments a redesign of the operating software a suite of improved data processing software and further research using new laboratory specimens to effectively create a system their personnel could use in the laboratory and field to rapidly take dense reliable easily interpreted NDE data These are the goals the research contained herein sought to achieve 74 CHAPTER 3 DEVELOPMENT STRATEGY Project Starting Point The scanning system developed in the course of the postdoctoral research by Algernon was left with the Florida Department of Transportation State Materials Office and inherited in this research effort While the system was very effective for its original purpose sever
106. he scanners and prepare for an automated scan on the left side of the screen This 127 includes a bank of manual scanner movement controls instrument settings grid settings and a control that toggles the settings between the two scanners if connected successfully Perhaps most importantly no matter how the operator configures the software the left side of the main interface always has an emergency stop soft button clearly displayed and enabled any time it is possible for the scanner to be in motion This applies even when an automatic scan is active as the Autoscan VI is designed to relinquish control to the Main Interface VI in emergency situations as described earlier in this chapter The right side of the screen is devoted to operator feedback and includes a three dimensional display of the scanner s position in space as well as its coordinates and instrument head up down status During an automated scan incoming instrument data is presented in real time and an information cluster containing the elapsed time of the scan as well as an estimated time to completion ETC is presented for the operator s reference Controls to pause or abort the automated scan are also provided for the operator s convenience Grid Generation Wizard The Grid Generation Wizard VI is accessible from the main program interface as long as the Autoscan VI is not active and an automated scan is not in progress It is a more convenient and intuitive way to ge
107. he new LAN protocol a Silex serial device network server was employed to operate the X and Y axis drives and a National Instruments NI 9181 Ethernet cCDAQ carrier with a NI 9401 8 channel Digital I O TTL cDAQ device was used to remotely operate the pneumatic relay installed on each scanner Since the low power DAQ device was incapable of driving the high power levels required to drive the electromechanical relay a simple transistor relay circuit was designed and constructed for each scanner control system These components as well as a network hub to interface them all were installed in washdown rated fiberglass enclosures and mounted to each scanner in place of the existing electrical junction box as illustrated in Figure 4 5 87 Control Box Component a ranior arven zev rany arc E 5 5VDC power supply 6 24 VDC power supply LAN serial device server Figure 4 5 Scanner control box components in fiberglass enclosure Photo by J Nelson To connect NDE instruments to the LAN for each scanner system a compact watertight ABS enclosure was mounted on each scanner s moving instrument head Each enclosure was outfitted with one National Electrical Manufacturers Association NEMA 5 15R electrical outlet to power each NDE instrument as well as a watertight RJ45 LAN panel connector to establish connectivity between the NDE instrument and the scanner s network hub 88 Figure 4 6 Electrical and data connection panel m
108. he permittivity and magnetic susceptibility ratios are exactly equal to unity Which yields the following calculation for the speed of light in a vacuum 1 m 2 998x10 Equation 2 6 y Eo Ho 2 For nonmagnetic materials the magnetic susceptibility ratio can be assumed V H C equal to unity Equation 2 1 then reduces to the following relation 1 C v _ Equation 2 7 V r oHo This relation effectively shows that the GPR method is sensitive to abrupt changes in dielectric permittivity 18 An extremely large change in dielectric permittivity such as a radar wave propagating through concrete incident upon a water pocket would cause a 44 correspondingly large portion of the wave to be reflected back towards the source This reflection is what is recorded and measured by a GPR data acquisition system Knowing the speed of the wave in concrete one could measure the time of flight of the reflected wave from the source multiply by the velocity and yield the distance the wave traveled twice the distance from the source to the water pocket Modern GPR equipment performs this single point measurement thousands of times per second to yield highly dense accurate data and generate a complete depth plot of the structure Connected to a rolling cart with built in position encoder such as the device shown in Figure 2 17 this equipment has the capability of producing long detailed line scans B Scans o
109. he standard practice in the State of Florida Currently all construction agencies using internal post tensioning duct technology in Florida and much of the US are required to use polypropylene or high density polyethylene HDPE ducts Galvanized steel PT ducts are still regularly used for external tendons but are rapidly being phased out in favor of cheaper easier to handle HDPE ducts Design of the new PT duct specimen included four polypropylene internal ducts with a bundle of three seven wire strand elements installed approximately on the center axis of each duct These ducts would be installed in the concrete mold and held in place with epoxy mortar Two of the ducts would be inclined on a work surface and partially filled with grout before installation into the concrete mold to create a void space within the duct of linearly increasing cross sectional area as shown in Figure 7 1 The other two ducts would be left empty with one to be filled completely with grout after concrete placement and the other to remain empty 151 PRRRHTHSHSET 7 Daana aapa Figure 7 1 Grouting of middle ducts of Asymmetric Internal PT Duct Block The ducts were filled on an inclined support to produce a varying void space Photo by J Nelson Figure 7 2 3D render of Asymmetric Internal PT Duct Block design Note the orientation of the two middle ducts 152 It was decided to fabricate the new PT duct block with different overall dimensions than
110. ignificantly easier for the operator An operator can manually complete the grid settings data cluster on the main interface and successfully run an automated scan without using the Grid Generation Wizard however using the wizard allows the operator to visualize the final size and density of the grid to be scanned The Grid Generation Wizard works by allowing the operator to manually move the scanner with the desired NDE instrument attached to the starting point of the grid the origin select the orientation of the lines to be scanned X or Y move the scanner 122 to the last point of the first line move the scanner to the last point of the last line and provide the desired number of points per line and desired lines in the bracketed scan area The wizard is intuitive and easy to use and provides detailed feedback to the operator over a text box Weis Next Value Change zH na Next and Back control message controls enable disable Calculate only on Next otherwise cases are identical YA Wizard Step EOL N Value Check vp 7 Sarp in other TE ISS ee f Disabled and Grayed Out aoe gt Disabled and Grayed Out F Disabled 4 seurik di E TET nner Seek ae 4 bDisabled m ye scs 4 Point Count if one or the ELE a aq aL Back Disabled other controls are changed Enabled vH Disabled l cesar eres N a jLine Orientation Sr Next Dis
111. igure 6 35 Ultrasound C Scan of a dense rebar mat at a depth of 54mm with Original top left Absolute top right and Hilbert bottom middle data transforms 146 After processing a number of displays for shear wave ultrasound data are available to the operator of the ANTARES Processing Suite A three dimensional scatter plot on the filter settings tab visualizes the depth of the maximum intensity of the reflected stress wave If filtering and trimming are done correctly this single data point per coordinate should correspond with either the back surface of the scanned specimen or a source of significant reflection within the structure Figure 6 36 3D ultrasound scatter plot of a densely reinforced concrete structure Frequently the scatter plot display is far too cluttered with spurious high reflections to be of sense to the operator For instances such as these the Fly Through C Scan and Fly Through Tomograph displays identical in function to their counterparts in the GPR processing section are much more useful in visualizing ultrasound data 147 So 1 Direction gt Fly 4 Tomo View Type gt Length Width Depth Current View Plane Y 1160 0 mm Manual Control A 58 8 ss 8 8 Depth mm Figure 6 37 Tomograph display of reinforced specimen with rebar mat top white and light blue and back side of specimen bottom light blue visible System Summary The fi
112. ile This VI is responsible for calling the correct low level hardware I O subroutine handling any errors associated with the read operation and outputting the instrument data in a standardized format A section of this code is shown in Figure 6 8 113 bsa 2013 TooOoOD oO ooo OD OOD OD OOD OD ODDO OOD OUD D UDO Oo UO o OOD o3 pnn JODO Do ODO DoD ooo ooo ooo po Oo oO DO oo po DO DO ooo TODD Oooo ooo Oooo Ao ooo Movement Command C Scanner to Control Za X Step Position Y Step Position Home Default v Constant timeout for homing seekir 2000o nonn Doo oo e MeN MsMsM MeM MoMsM MeN MeMo MMM MeM MsMeM MeM MoMeN MeM M MeN MsM N MeM MeMsM MeHoM MsN MeM M MeN MeM M MeN MeM M MeN MsMeM MeN MeMsM MeM MsMeNeMsM N MeN MsM N MeNMoM N MeN K MeN sK Figure 6 7 Portion of Scanner Movement VI with low level SubVIs visible as square icons with a blue top border 2013 a Covermeter Proceg Profometer 5 Default Ypf i B Read Only j Instrument Instrument 2 Configuration Cluster Camera Settings n k Figure 6 8 Code snippet of Native Instrument A Scan SubVI Task handling subVIs The Task Handling SubVIs are a novel way to handle plain text commands to the ANTARES system such as move to location automatically scan the provided grid with the chosen instrument tell me where the scanne
113. in the State of Florida that led to the research discussed herein The chapters devoted to the development of the scanning system will follow a roughly chronological order with some liberties taken for organizational flow Chapter 2 focuses on the requisite background knowledge to fully comprehend the development of the NDE scanning system Included is a history of NDE research performed by FDOT the sponsoring agency of this research Additionally a detailed description of each NDE method employed by the scanning system including data acquisition and interpretation practices based on material modeling is provided Chapter 3 describes the development strategy and goals of the research which comprises the framework of this thesis as planned before initiation of the research effort Chapter 4 discusses initial improvements to the inherited automated scanning system and their role in producing a more effective more versatile and more accessible system suitable for laboratory and field use Chapter 5 is devoted to discussion of new NDE instruments to be included in the system their relative strengths and weaknesses and the methods by which they were employed Chapter 6 chronicles the development of the software packages used to operate the scanning system and associated NDE instruments and process the data produced by the system into a form suitable for interpretation by the operator and owning agency alike This effort represented by far the larges
114. ion of entrapped air and water within structural concrete A series of ultrasound scans were conducted using ANTARES collecting on average around 6000 waveform data points per scan during an approximately 5 hour scan 177 Immediately upon processing the ultrasound ANTDAT file it was observed that ultrasound had reliably detected all of the most severe defects as shown in Figure 7 27 These defects were visible on the filter settings display which only displays the depth of maximum reflection and is therefore usually unreliable at displaying all objects of interest within a specimen g lt Ultrasound Filtering Display Control Hilbert Wave Velocity t 2200 m s Minimum Depth 50 mm Maximum Depth zomm l Bandpass Filter High 46000 Hz Low 10 Hz Ji 200 00 1 1 1 1 1 200 00 600 1000 00 1400 00 1800 00 2200 00 De ba Te E Figure 7 28 Initial ultrasound data processing for Void and Flaw Detection Block Focusing on the C Scan depth display stepping through the thickness of the specimen first revealed the three embedded pervious concrete prisms and the embedded wet sieved concrete volume as shown in Figure 7 28 Though their geometries were difficult to define and their outlines were faint these features could be identified reliably in the ANTARES Processing Suite 178 Figure 7 29 Ultrasound C Scan image with no s
115. ion upon which this thesis is built am very grateful for his continued involvement My sincerest gratitude goes to all the dedicated personnel at the FDOT State Materials Office particularly Messrs Mike Bergin Richard DeLorenzo and Pat Carlton whose support of this project has truly allowed it to flourish Acknowledgment is due also to Mr Sam Fallaha and the staff of the FDOT M H Ansley Structures Research Center whose assistance allowed the first field deployment of the Department s newest NDE asset to be a resounding success Last but certainly not least thanks are due to my dear friend and love Ms Rebecca Gabay for her support and understanding during my research and writing efforts TABLE OF CONTENTS page ACKNOWLEDGMENT is ssesctoncchonednsloaudentdenmhonidirasdancdonehenblonnduadlaneboitdiondudslouudentlenebsdadenal 4 LST OF TABLE c Moro memnerevenererooemtom teerte mentors mirnnereurne ints Marre enters i Tint Tse nun eo TenDr oO Te Te 9 LUST OF PLUME cas aoa da N AANRAKEN RAENT A ANERE ATATA 10 aera lt 8 A A AEEA A E A A A A O eae 18 CHAPTER 1 INTRODUC TKO N ocene ene er Werte Ruane ty irene yen ene a heer hve Wome sy enei erm urren 19 Fie ae KONV AMOT ae iatieeielatiaueielaliauaieliatinueiaiaiaudtiali 19 IGG TNES Sangu o EnA OAE 20 2 BAOF OROUND esseiden aa oleae acetal meld eal altel dlenelians 23 NDE Methods Considered in this RES rch cccccceceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeees 23 el M
116. ious there was a notable difference in the wave velocity between the top middle and bottom testing locations on the concrete block even in the control blocks This can be attributed to compaction and segregation during casting of the blocks as well as differential curing This result is a major reason why the majority of laboratory blocks used in this thesis were cast horizontally Secondly a rise in wave velocity as the concrete fully hydrates is noted remaining constant in the case of the lime water saturated panels The blocks exposed to sulfate solution however showed a decrease in wave velocity at later ages with a more dramatic decrease farther down the block This result suggests the sulfate attack mechanism verified by SEM images and other data 25 Nondestructive evaluation of alkali silica reaction In order to evaluate the effectiveness of several non destructive test methods to detect the presence of alkali silica reaction ASR in concrete Maryam Sargolzahi et al engaged in a series of laboratory trials applying NDE methods on cylinders fabricated from ASR reactive an nonreactive materials 26 Prisms were also cast from the batches to be tested in accordance with Canadian standard CSA A23 2 14A a measurement of expansion used to quantify the presence of ASR In addition to a mechanical test of the specimen compressive strength and elastic modulus Sargolzahi performed the following non destructive tests in an attempt to quantif
117. irst four sequences of the Splash Screen VI The Configuration Settings Screen shown in Figure 6 16 has the vital task of managing the persistent configuration settings data cluster in ANTARES SVW This data cluster contains the conversion factor for the servo motors from steps to millimeters the default velocity and acceleration limits the scan area limits the port settings for all native hardware including NDE instruments servo motors and each 121 scanner s CDAQ and the default operation mode for each NDE instrument This program is capable of altering the values listed above verifying them writing them to a file and importing them from a file z sar 5 Export Value Change yee o pe 0 0 1 DoODoooDoOoooOo ooo ooo ooo oo 7 5 Scanner Port Cluster E i ANTARES Pept Scanner Global Settings ANTARES Config Covermeter Settings 1 ANTCFG I gt Tam ANTARES Configuration File r Ultrasound Settings E E Save configuration file l selected path gt GPR Settings i SCS Settings gt Laser Settings L opooooooocooocooooooooooooocooooocoooocoooooooooooocooonoooooooo0oo0g Figure 6 16 Main loop of Configuration Settings Screen displaying the export ANTCFG file case Finally the Grid Generation Wizard is a supplemental VI to the Main Interface VI which makes creating a grid to be automatically scanned using the ANTARES scanner s
118. is a logical future application to be promoted Nondestructive evaluation of impact damage to concrete bridge piers The removal of a functionally obsolete bridge in Apalachicola Florida provided researchers with a unique opportunity to study the behavior of a concrete bridge pier when impacted by a barge or other large vessel Such a mode of damage had not been previously studied but was the cause of the tragic collapse of the Sunshine Skyway Bridge spanning Tampa Bay Florida in May of 1980 The freighter M S Summit Venture veered off course in poor weather and struck Pier 2S sending the main southbound span into the Bay and 35 people to their death A photo of this tragic accident is shown in Figure 2 31 69 a 4 VAI X AWE PEME MAAA AA ENIT gt Figure 2 31 Collapse of the southbound main span of the Sunshine Skyway Bridge Thirty six motorists fell into Tampa Bay only one survived Photo courtesy of St Petersburg Times 32 In the Barge Impact Study a partially laden barge was driven at low speed into heavily reinforced load instrumented impact blocks mounted to the targeted piers The structural research authorities involved in the study monitored the deformation of the blocks using various instruments and high speed cameras to refine models intended for use in developing more impact resistant bridge piers in the future In parallel with the structural data gathering teams Christopher Ferraro and other m
119. l established confidence in all the techniques used in the study and proved the reliability of appropriate acoustic and radar techniques to measure concrete thickness and locate rebar and post tensioning ducts 27 Beutel s comparative research established several sound practices and a significant amount of confidence in the NDE instruments used in this research His exhaustive study of a multitude of different methods including multiple instruments of each individual instrument type was invaluable for the work discussed in this thesis 60 Detection of Flaws Introduction Flaws within a concrete structure are both difficult to observe and have the potential to severely compromise the integrity of a structure A flaw within concrete from the perspective of a fabricator or inspector can include any number of things cracks air voids due to poor consolidation water pockets due to bleed contaminants such as construction debris waste organics etc NDE methods are usually called upon to evaluate the extent of a known flaw such as measuring the depth of a crack ina bridge pier Most of the research discussed in this section will be motivated as such however once automation is considered it is necessary to address NDE methods in a quality control capacity detecting unknown flaws within a structure before they become apparent Comparison of NDE methods for concrete bridge deck evaluation In order to provide a comprehensive comparison
120. late to create a known reflection using GPR could be readily applied to measuring the thickness of concrete elements In situations where the back side of an element is accessible a metallic plate could be adhered to the concrete surface and quickly found using GPR on the other side for an accurate thickness measurement Comparison of multiple NDE methods for slab thickness measurement Validation of NDE methods for their various applications is an important function of many research entities To this end Ralf Beutel and his research group used a specially design concrete specimen in Berlin Germany consisting of realistic but well documented intentional defects to evaluate the abilities of several different impact echo ultrasound and GPR methods 27 Their primary intent was to certify the instruments for quality control purposes in detecting slab thicknesses locating reinforcement and detecting delaminations contaminants and other flaws Of particular importance to the research contained in this thesis was the comparison of acoustic and radar techniques for reinforcement locating and geometry measurement purposes 59 Area 2 detail mm X 00 460 400 50 MO 260 MO 10 I0 s 4960 mm Area 1 Figure 2 25 Concrete slab in Berlin used by Beutel and other researchers The block consists of known defects reinforcement locations and dimensions including sudden changes in geometry 27 The research conducted by Beute
121. lombo and Felicetti established correlations between uniformly heated specimens and several NDE parameters including UPV reading and digital image colorimetry Indirect UPV was found to be extremely sensitive thought time consuming to apply and difficult to apply to anything but flat surfaces Loss in ultrasonic pulse velocity within a structure is also directly correlated to a loss in elastic modulus therefore no calibration data for the particular concrete mix is necessary when this technique is applied appropriately to evaluate fire damage Digital image colorimetry also showed great promise for rapid evaluation of concrete with well characterized constituent materials 31 68 after fire enor 3780 m s 27 before fire Vn 4530 m s depth mm travel time T Vasym mm panel 2 0 200 400 probes distance X mm 0 005 0 0 005 colour variation x y Figure 2 30 UPV results of a fire damaged concrete panel left Color variation in a section of a small core with a fire damaged surface right 31 For the purposes of the work contained in this thesis the research performed by Colombo and Felicetti established a new application for the automated system Ultrasonic pulse velocity may be readily derived from data taken using a shear wave ultrasonic echo array and the included vision system can be rapidly adapted to perform colorimetry calculations While neither application is within the scope of this project it
122. m and the beginnings of an entire research program that could potentially advance the industry of concrete NDE and indeed the construction industry as a whole It is a tool accessible to researcher and technician alike with unique capabilities and unique applications It is a powerful flexible and expandable software platform owned and operated by non profit agency The ANTARES system has proved itself able to resolve many forms of structural defect and do so in a repeatable and easy to understand manner The software produces useful open format data and easy to understand data displays and is available for use by any agency that wishes to collaborate The system is certainly not without its limitations but it is sincerely hoped that future research efforts and collaboration with others in the industry will work for mutual benefit and reduce those limitations systematically It is recommended that the potential of the ANTARES system for structural quality control applications be investigated and expanded upon including statistical analyses of the system s ability to reliably detect defects before a structural element enters service These capabilities could be further expanded upon by the inclusion of new NDE techniques including higher energy techniques where applicable 197 APPENDIX A SELECTED PROGRAMMING EXCERPTS This Appendix contains a few pieces of code developed in LabVIEW that perform critical functions such as reading instrum
123. men using a sensitive accelerometer following a short duration point impact The method was borne out of a technique called hammer sounding in which an operator would tap on the surface using a hammer and listen for a low frequency rattling sound indicative of near surface voids in the concrete specimen While the operating principle of the modern impact echo method is essentially the same 34 as hammer sounding computer based data acquisition is used to replace the trained human ear to interpret the frequency pitch of the returned acoustic wave and more consistent impacts are generated using specialized manual or electromechanical impactors Figure 2 9 Modern impact echo system with hardened steel ball impactors Photo courtesy of NDT James Instruments 13 In practice the impact echo method is used by first creating a short duration point impact on the concrete surface using a falling or spring loaded spherical ball made of hardened steel or an electric solenoid striker In a P Wave velocity measurement two transducers mounted a known distance apart on a spacer record the surface wave as it passes under their contact point allowing a simple distance by time calculation for wave velocity The P Wave is distinguished from the slower traveling R Wave by using the shortest recorded time delay between wave peaks usually performed in software automatically 14 35 Data Acquisition and Analysis System Spacer device
124. ment used data and time of file creation and grid settings in the header information The instrument data section contains a row for each data point with the date and time the data point was 132 taken the scanner coordinates the instrument s coordinates Scanner coordinates plus the instrument s offsets and A Scan single spatial point data The A Scan data column changes for each instrument as shown in the following table Table 6 1 A Scan data output formats for ANTDAT file format Instrument A Scan Data Type Covermeter Cover in mm 0 digits of precision Distance in mm 1 digit of Laser Profilometer precision Ultrasound 2048 point 32 bit integer waveform SCS Binary image string GPR Parallel Instrument Blank The universal nature of the ANTARES Data File allows the file format to be expanded easily to new instrument types and new data formats and allows for collaborative processing efforts on data output by the system Since a data file is generated even in the case of parallel instrument operation if an NDE scan is performed in parallel mode the data can be mapped to the grid given in the data file by using either the time stamps for each coordinate or by the coordinates themselves if the instrument has its own position encoder as was the case with the GPR system ANTARES Processing Software Introduction The capstone piece of software to be developed for the ANTARES system was the software suite for processing the r
125. mm I I I I 1 I I I 1 I I 1 I i I I 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Line Figure 8 10 Ultrasound C Scan of location 2 showing severe delamination circled left and stronger backwall reflection of adjacent bridge deck due to lack of contact with foundation right A GPR scan was also performed at location 2 to reinforce the findings of the ultrasound scans The bridge deck was densely reinforced so the C Scan image presented in Figure 8 11 contained large areas of steel reflections that occluded any 193 potential reflections from delaminations within the deck A closer look at the tomograph view shown in Figure 8 12 however revealed a strong reflection from the area of suspected severe delamination further suggesting an area of voided grout Figure 8 11 GPR C Scan image at a depth of 120mm showing dense reinforcement on either side of the keyway joint 194 Figure 8 12 GPR B Scan image of antennae path over the area of suspected severe delamination The large faint reflection over the girder indicates a probable air void Following the NDE analysis the personnel at the Ansley center took drilled core specimens from the two suspect locations marked on the test element at Location 1 by the research team with marking paint in order to confirm the findings from ANTARES In the two locations from the first scan area the core came out of the barrel bit with a clear space between the concrete bridge d
126. mpact Echo Method section The technique is also far more sensitive than pulse wave to fluid pockets within a solid since liquids and gases cannot transmit shear stress and as a result much stronger wave reflections are observed This makes the technique ideal for locating honeycombing bleed water pockets and similar defects 42 A shear wave ultrasound array typically produces a single waveform showing a composite of the reflected shear waves as received by the instrument head This waveform can be thought of as a single point representation of the thickness of the concrete specimen and is referred to as an A Scan When waveforms are collected iteratively over a line an array of waveforms can be assembled to generate a two dimensional intensity plot This two dimensional representation of a line of data through the thickness of the specimen is referred to as a B Scan Finally if repeated B Scans are assembled together to form a three dimensional intensity array the operator can view the specimen top down and investigate each received data layer of the specimen or generate a three dimensional render of the interior of the specimen This complicated data assembly is referred to as a C Scan Production of these data renders as well as all signal processing used in creating useful quality data from shear wave ultrasonic array is discussed in Chapter 6 Ground Penetrating Radar Ground Penetrating Radar GPR is a somewhat nas
127. much to handle Fortunately for the purposes of concrete surface inspection a grayscale camera of relatively low resolution is usually sufficient and allows a survey of a much larger area before requiring extremely high computer performance Conceptually processing the images generated by the camera system for these purposes is relatively simple Using a repeatable light source and tuning the camera at the time of testing an operator can establish a minimum light level for a good concrete surface Recessions in the surface due to cracks and voids will appear in the data processing algorithms as darker pixels which can be identified and counted if they fall below the established light threshold Variation in ambient light reflective areas and surface markings left intentionally may cause a false flaw detection in a simple algorithm such as this so more advanced detection techniques also establish minimum sizes and establish boundaries on shapes using edge detection algorithms when defining a defect Performing planar measurements with a single camera system is fairly simple Using a pre printed grid where easily distinguished lines are set a known distance apart a software package can be assigned a calibration factor where a designated number of pixels in its two dimensional field of view corresponds to a known distance This calibration loses accuracy when applied surfaces that are not planar with respect to the camera s lens as a single
128. n run independently and provides for pass through transmission instrument types to be used when run dependently The Autoscan VI makes multiple calls to the Scanner Movement VI A Scan VI and various other subroutines at different instances dependent upon grid settings and scan type as shown in Figure 6 9 The VI must be reentrant to allow calls for both scanners when the scanners are operated simultaneously and requires significant computational overhead to do so This requires ANTARES SVW to be run ona moderate performance computer 115 Scanner Timeout Line Start Ero Abort E x T02 7 a False 7 2 Sea s This section starts at Line K Point 0 E Scan Mode Point Count bd LJ Hi 3 i soa 6 l Instrument E t Instrument Runtime Settings l i LU Pneumatic Mode fl Always Up Down Poi Scan Mode Line Number k 0 4 l Grid Settings FF l M S Inst Offsets EE pipe jj jeene eee eeg fA Commigeration Chister Scanner Global Settings Camera Settings Inst EE mH Point by point IFOR N pts ca j l gt M S in position gt Pneumatic gt M S pneumatic mr gt Leiter l gt Read gt Loiter gt Move Data File Path Independent Default vP Scanner 1 Default gt
129. nalized ANTARES system is a fully integrated expandable system for automated NDE of concrete structures capable of operating five different instruments in a repeatable manner over dense two dimensional grids on a concrete surface The remainder of this thesis will discuss efforts to prove the viability of the new system and expand its capabilities both in the laboratory and in the field 148 To other scanner optional Signal Connections mam Ethernet Permanent Temporary Instrument Power Supply Connection Cable Routing Covermeter Palette mounted Serial Palette mounted server Ethernet to Transformer instrument head Laser Palette mounted Serial Palette mounted server Ethernet to Profilometer Transformer instrument head Ultrasound Internal Battery USB Palette mounted server Ethernet to instrument head SCS External Ethernet Direct to instrument head transformer Removable N A N A Parallel Instrument Battery Pack Figure 6 38 Instruments and signal connection scheme for the new ANTARES system 149 CHAPTER 7 LABORATORY EVALUATION BLOCK CONSTRUCTION Introduction The ANTARES Laboratory Evaluation Blocks represent a tremendous asset to FDOT and any collaborating agencies wishing to test new NDE instruments and techniques to investigate various structural characteristics and defects The blocks are fabricated with tightly controlled concrete mixes and precisely located features for detection by NDE ins
130. nd his original operating software package ScannerVoll Wizzard voll meaning full or complete in German the ANTARES scanner and instrument operating software was named ANTARES ScannerVoll Wizzard or ANTARES SVW for short ANTARES SVW was designed from the beginning to have simplified modular software architecture and an intuitive single pane interface One of the most frequent complaints about the Phase operating software was that it depended far too much on popup windows and multiple screens to the point where it was difficult to discern which 107 window was active and which was merely open as a subroutine of the active window Additionally concerns about safety were raised when unexpected scanner behavior was witnessed and no ability to interrupt an automated scan was included in the Phase interface ANTARES SVW was designed to address these two major concerns as well as include significant design changes to increase overall usability of the system Approach To make the ANTARES SVW software easy to troubleshoot and modify the program was designed around a tiered architecture starting from low level operating Virtual Instruments VIs LabVIEW code sections with graphically programmed subroutines and simplified user interfaces a nd moving upwards in complexity until ultimately reaching the master interface Vis The software architecture will be discussed here in detail and supported by code excerpts found in Appen
131. ndirect measurements are valid and statistically similar to direct measurements and it is possible that this method may one day be approved by standardized testing organizations 15 Transmitting Transducer Figure 2 14 Test setup for standard UPV test in direct transmission mode 16 Shear wave ultrasound method Shear wave ultrasound is a relatively new technique designed to overcome many of the limitations carried by standard pulse wave ultrasound when applied to concrete The mechanism by which stress waves are induced on the concrete surface is fundamentally different from ultrasonic pulse shear wave ultrasound transducers use a 40 piezoelectric oscillator whose principal axis is oriented parallel with the concrete surface to slide the transducer contact point along the surfaces instead of drive it into the surface In the case of both transducers the piezoelectric crystal can be used to either drive the transducer and generate stress waves by driving the crystal with an alternating electric current or receive the stress wave by measuring the current generated by the crystal as the transducer receives a stress wave In this way a sophisticated control system can take advantage of a relatively small number of transducers arranged in an array to both generate and receive stress waves at many different points using the same transducers Piezoelectric Crystal Soft Polymer Contact Figure 2 15 Operating principle
132. nenisenineninnininnnineneinininins 56 Geometry and Locating Applications xccsccsensssscsaressatseesiestarrsidieassteneecses 56 WUE seorieirerinioe a 56 Applications of ground penetrating radar for pavement thickness MESEM casana a A NER 57 Comparison of multiple NDE methods for slab thickness measurement 59 Detection of en ee eee eee te EAEI IESE eee eet eee en eee eee 61 ppc a e RENERE E ene A 73 Data Collection ae E E E peste A E diaries ee ac Initial NDE scans wee 56 Post grouti Void and hye De 8 FIELD DEPLOYMENT AND SYSTEM EVALUATION oaeee 183 e E a A E 183 CIS SIE aino 184 Transportation of the System restos sitet nics casscdeenedtdinndetedon done diandetederedesliandenent 186 DS A RESU eee ascertain gies ieee ie cel aia ens 189 9 CONCLUDING REMARKS murine idni inka kin riean nina niia kianat 197 APPENDIX A SELECTED PROGRAMMING CAUCE Vi scccgeccce cess eeren eee 198 B SELECTED CAD PAG sete sssnat reciente ceueddantecnedensenaidatcenedeneneidanuesasiod 209 UST OF REFERENCE Shoan tra R erat pagened axaeeE 219 BIJGRAPHICAL SKE TOH rypa iior nia AA A ere NT B Tere E rere 223 LIST OF TABLES Table page 6 1 A Scan data output formats for ANTDAT file format c eeeeeeeeeeeeeeeeeee 133 Figure 2 1 22 2 3 2 4 2 6 a7 2 8 2 12 aija 2 14 2 15 2 16 2 17 2 18 LIST OF FIGURES page Spalling of concrete due to insufficient cover and poor consolidation
133. nerate a grid for an automated scanning and its mechanics are discussed in detail in the previous section The wizard s interface itself is a rearranged version of the manual controls and scanner location display in the main interface as shown in Figure 6 22 At the top of the panel a large text box provides the user with contextual instructions for using the wizard as well as buttons to proceed forward or backward in the grid generation process The text box allows the Grid Wizard 128 to communicate with the operator in descriptive phrases to aid in navigating the wizard while the soft panel itself is designed to be as simple as possible and to prevent the user from generating poor or unexpected scan grids X ANTARES SYW FDOT State Materials Office File Edit Operate Tools Window Help x Exit Program Time per Data Point 0 000 sec ener Line iec SW Finally all I need to know is how many lines to run Tell me in the grid panel by completing either the line spacing or line count I ll fill in the other for you Once you are done click Next Pause Scan opr ars b Grid in Progress tsn Grid Origin x 150 mm Use SCS Y 1520 mm Lines Run in Y Points Count gt 133 fine o 186 Grid Test Run Y 0mm e Q use ses a _ o BEGIN A SCAN Figure 6 22 Grid Generation Wizard on one of the final steps to completion of a functional grid File I O Introduction
134. ng frame 75 11 4 4 Instrument installation process Seen eee e eee eee eee eee eee eens See eee eee eee eee eee eee eee ee ee eee 85 Scanner control box components in fiberglass ENClOSULE eeeeeeeeeeeees 88 4 6 Electrical and data connection panel mounted to instrument head on Scanner ai 4 8 91 4 9 Germann Instruments EyeCON adapted to palette system ceeee 92 4 10 5 1 5 2 GPR antennae mount with limit switch for FCC compliance ses e1111111 96 5 3 Byte stream representation of ODSL8 distance measurement 98 5 4 LabVIEW VI which reads and parses the 2 fe serial Leuze ODSL8 laser distance sensor Startup subroutine of SCS run time application 0 cece 103 5 9 SCS placard design with origin and direction Markers eee eeeeeeeeeeeeees 104 5 10 5 11 6 1 FDOT ANTARES logo designed by the research team iiinis 107 12 6 10 6 11 6 12 6 13 6 14 6 15 6 16 6 17 6 18 6 19 6 20 6 21 6 22 Basic tiered architecture of the ANTARES SVW LabVIEW application 109 Parsing structure of CML Command SUV hisiminiuunimnniimimnniaun 111 Command section of pneumatic instrument head command subroutine 112 Pneumatic instrument head query subroutine with integral error handling 112 Code sechon GT AUTOSCAN VT cnssascinnsinsaininiarainininrainaniacaleitianalantianalantianaininennnias DEO
135. nt to the impacted surface can be calculated from the pulse wave velocity Cp and the calculated dominant frequency fas follows 10 T Equation 2 4 SAR 2f In situations where an air void water pocket area of poor consolidation delamination or other discontinuity exists between the transducer and the back wall of the concrete plate two distinct situations may occur 1 In the case of a small aspect ratio defect the stress wave will travel around the defect and the calculated thickness of the concrete plate will be larger than expected 2 In the case of a large planar defect the stress wave will reflect off the defect and the calculated thickness of the concrete plate will be smaller than expected 37 Flaw detection using impact echo is unreliable when defects are very small non planar with respect to the transducer or hidden behind another interface such as another anomaly or reinforcing steel in an effect known as masking It also unreliable to use the impact echo method in compact specimen geometries where wave reflections from the sides of the specimen may cause modes of vibration other than the desired plate like vibration mode to become dominant 10 Ultrasonic pulse velocity method Ultrasonic pulse velocity UPV methods applied to concrete structures were borne out of related techniques applied to metals and other homogeneous materials By coupling a high frequency piezoelectric oscillator to the surface
136. o ooo U noo ooo UU Doo ouoUo Uo ooooUoooooo Configuration Cluster EERE tana i gt Scanner 1 gt Motor Status 7 i Scanner 1 7 jantares Motor Talkback 1 eri Reset E Stop gt eee a Scanner 2 I e r Y E rand fad fad Ed Motor Talkback 2 Reset i i SCANNERSIN AMAR PP Ra 7 EMERGENCY ia STOP yj i DoOoOooooo Oooo Oooo Oooo OU Oooo oo coo oooooooooooooooooooooo 0o0oooooooooooo0ooooooooo0o0og Figure 6 11 Emergency Stop VI with integrated reset dialog box This VI effectively freezes both ANTARES scanners in place and halts execution of ANTARES SVW until reset 117 Human data interface Vis These top level VIs ultimately form the link from human to software or software to data output While their programming structure is not particularly complex the ways in which these VIs utilize lower tier subroutines are critical to the overall function of ANTARES SWW Essentially the human interface Vis can be thought of as event driven wrappers for the lower tier Vis and the data output VIs can be thought of as a converter of data arrays stored in RAM to formatted ASCII text files The human interface VIs are as follows Main Interface Splash Screen Configuration Settings Screen Grid Generation Wizard The Main Interface VI controls
137. ock at a depth of 120mm showing damage from depth moving rebar 1 lack of rebar reflection due to earlier reflection from simulated entrapped air 2 and minor concrete disturbance due to length moving rebar 3 180 Figure 7 31 Hilbert transform C Scan ultrasound image of Void and Flaw Detection Block at 120mm depth Finally as the depth of the C scan display reached the back surface of the block 250mm a continuous plot of similar color would be expected for a monolithic concrete block In the case of the Void and Flaw Detection Block however the enclosed defects meant that the shear stress waves transmitted through the block s thickness at those areas had already reflected back to the source and so never made it to the back surface of the specimen A trace of some kind was visible for every defect in the block as shown in Figure 7 31 although these traces did not necessarily adequately reconstruct the defect s geometry and no information on the depth of the defect can be inferred This effect is known as shadowing and can be used to rapidly investigate a specimen of continuous known thickness such as a slab or bridge deck 181 a Figure 7 32 Ultrasound C Scan shadow image of the back surface of the Void and Flaw Detection Block 182 CHAPTER 8 FIELD DEPLOYMENT AND SYSTEM EVALUATION Introduction One of the major milestones in developing the ANTARES system from the Phase FDOT automated NDE s
138. of a shear wave ultrasound transducer Note the piezoelectric crystal is mounted transverse to the transducer and particle motion follows this direction The operating mechanism of shear wave ultrasound allows the transducer to be designed in such a way where a stress wave can be reliably generated and transmitted through a soft polymer tipped transducer without the use of acoustic couplant This simple advantage over standard UPV testing is not only more convenient when the 41 technique is applied in manual testing but allows the technique to be much more easily automated as an ancillary system to manage application of couplant and verification that sufficient couplant was used for a reliable measurement is not required This is the primary reason why this technique was chosen for use in this body of research and UPV was excluded Figure 2 16 Modern shear wave ultrasound instrument showing 24 transducer array left and schematic of wave reflections right Photo courtesy of Germann Instruments 17 A return signal using the shear wave ultrasound technique depends on successful transmission and reception of S Waves which leads to a few properties of interest for the concrete inspector S Waves travel significantly slower than P Waves and with less energy so they tend to reflect off of small defects and delaminations easily While this means the technique is very sensitive it also exacerbates the masking effect discussed in the I
139. of several popular NDE techniques for use on concrete bridge decks Scott et al engaged in a study of four NDE methods used on the Van Buren Road Bridge over Quantico Creek Virginia Scott and other researchers employed chain dragging impact echo and two forms of GPR to evaluate the condition of an aging bridge deck and compare the test results with cores taken from the structure 28 The data was overlaid with a map of the known defect areas and used to compare the effectiveness of the NDE methods 61 Care Cows Cored Cor 10 Figure 2 26 Results from Van Buren Road Bridge NDE study overlaid with defect map a Chain dragging with suspect areas shaded b Impact echo with dark spot suspect locations and c GPR scans with section AA from proprietary HERMES system and section BB from GSSI system similar to the one used in this thesis 28 The major cause of distress in the Van Buren Road Bridge is delamination of the concrete layer from the rebar mat nearest the surface of the bridge deck This is a common mode of failure for concrete bridge decks and is usually fairly easy to detect 62 using acoustic methods especially when a scanning system is used to increase data point density The most important information to glean from this particular study is the reliability of impact echo and indeed any acoustic method in detecting planar air voids within a thin plate like structure of reinforced concrete Indeed the map provided by
140. on the roadway surface as shown in Figure 8 5 and the first scan began at 10 00am Figure 8 5 ANTARES system in operation at the Ansley center performing a laser profilometer scan of the first location on the ABC test element Photo by J Nelson Scans were conducted until 5 00pm until an encroaching storm forced an end to the field deployment The system was placed back on the flatbed trailer and transported back to Gainesville 188 Deployment Results Scans using the ANTARES system covered two areas on the ABC test element Each location contained a joint running in the Y direction and a girder running in the X direction as shown in Figure 8 6 the intersection of which was clearly visible on the surface and in the NDE data Figure 8 6 ANTARES scan locations on the ABC test element The first scan location received a laser profilometer scan and an ultrasound scan The second received a GPR scan and an ultrasound scan All four scans were 189 performed on a much coarser grid than would have otherwise been used due to time constraints The laser profilometer scan showed a very coarse roadway along with residual features from the grouting operations as shown in Figure 8 7 This scan emphasized its value as a supporting instrument indicating regions where suspect areas determined from contact instruments such as ultrasound may just be artifacts of the surface itself and not an actual structural defect ww aau
141. oncrete block Cover mm i 1 5 2 2 5 Distance inches Figure 2 7 Example of the effect of orientation on covermeter measurement These line scans were performed on the same specimen with the instrument head parallel blue and perpendicular red to the rebar 9 Despite the obvious limitations of eddy current methods and the covermeter in particular the method was included in this body of work primarily to complement the other included methods and to include a method that is well known well understood and commonly used manually in the field 32 Acoustic Methods Introduction Acoustic methods is a moniker encompassing an entire range of methods based on the measurement of stress wave propagation in a material For the purposes of this work we will consider only ultrasonic methods in depth with a few ancillary methods that help explain them The research described herein utilizes shear wave ultrasound as its acoustic method so this technique will be discussed in detail To better understand the mechanism behind shear wave ultrasound method the earlier impact echo and ultrasonic pulse velocity methods are discussed All acoustic methods are based on the detection of one or more of three distinct stress wave types which propagate through a material when a stress event is induced on the surface or within the continuum of a material as shown in Figure 2 8 The pulse wave or P Wave corresponds to
142. oooM _Scanner Select CO a Home Load Auto Bt PEASE Scanner 1 7 e Scanner 1 gt F Home gt fe Set Current as Home Configuration Cluster EH DoOoDooooOooU ooo UU ooo Uo ooo Scanner Enabled E ii ee i Bre F 5 x J eaea True gt I Home Load Manual 2 i OOOO oooooOon ooo ooooooooo oe oe alt alt ie Scanner 2 7 gt Scanner 2 Fj gt Home gt gt Set Current as Home Home Load Auto 2 Ensure the scanner s path ooooo000oooonoooooooonnaA 4 n x ia E Stop Loops TEN Figure 6 13 Human interface loop of ANTARES SWW The final phase of the Main Interface VI closes any existing data port sessions relinquishing those resources for use by other programs The code section also operates a clean program termination ending the program safely and relinquishing the RAM the program has reserved This code section is shown in Figure 6 14 Figure 6 14 Clean up phase of the main VI NI VISA sessions are closed and the program is terminated in a safe clean manner The Splash Screen VI is a piece of interfacial code that performs the vital tasks of initializing communication with the scanner hardware and informing ANTARES SVW of 120 any hardware faults missing communication settings or disconnected scanners The program itself is an 18 step startup sequence that presets the scanner hardware for use by A
143. or T T Scanner Status Scanner Port Cluster ke T Scanner 1 Defa 7 zl a DAQm x Device A m Enum strict RE wD A a b ActiveDev Value i r Scanner TSi a DiPorts anner 1 Status ae Timeout iat Bl ne channel for each ine figer oopa Digital ny F ane NChan 1Samp T Up Default VPA iE 4 Command a DAQm x Device Sensor Head Status lone channel for each line Figure A 10 Pneumatic Head Query VI 206 098 Instrument Lo nan a RES Figure A 11 ANTDAT File Creation VI 207 Data File Path i B Header abc BH ANTDAT p EE g Coe a a Foon OHO AAA OBO H OOP eee ob 1 0 1 pPRee ooo ono neon oe eee Build instrument specific columns la ANTARES Laser Profilometer Data File Path Instrument Lot 3 Br m pey Data Cluster L eS Scanner Status Laser Offset X f Laser Offset Y Configuration Cluster t sic eh li ie ea eh Se egal ach end ea ERE Figure A 12 ANTDAT File Append VI 208 APPENDIX B SELECTED CAD DRAWINGS This Appendix contains selected drawings for designed components mentioned in this report This includes the instrument carrier design laboratory blocks
144. ork and allowed to resettle below the block Its free end was moved the desired 25mm per design The laterally moving rebar specimen had a more complicated cable system guided through a lubricated tube into the formwork itself This system failed to move the rebar as desired and it was confirmed after measurement that this rebar piece moved only 8mm rather than the desired 25mm The formwork was removed after 24 hours and the new block was cured under wet burlap with a timer operated soaker hose to maintain saturation for 7 days Ina carefully orchestrated forklift operation the block was flipped vertically using the bottom formwork as a lever and protective shield and wooden support boards to allow a forklift to lift the block from the bottom edge It was then installed in the scanner frame as shown in Figure 7 20 to begin NDE scans using ANTARES Figure 7 20 ANTARES Void and Flaw Detection Block installed in scanner frame with surface defects visible Photo by J Nelson 171 Data Collection For the first time an NDE laboratory specimen was evaluated entirely by automated techniques using ANTARES and the newly completed ANTARES SVW software package Where computing power and lack of instrument interference permitted both scanners operated continuously and simultaneously on both sides of the Void and Flaw Detection Block quickly producing a mountain of data to be processed Laser profilometer The laser profilometer scan being a rapid
145. ounted to instrument head on Scanner 1 Photo by J Nelson With these modifications to the scanner complete additional benefits were realized First both scanner systems could be operated at the same time since they no longer required multiple dedicated serial resources to operate Secondly because the two scanners could be connected to each other using independent onboard network hubs a single Ethernet cable was all that was required to operate both scanners with their own independent instruments from a single computer Finally the implementation of the LAN protocol and the all digital interface had effectively eliminated the need for 89 the PXI chassis to operate the system Almost any inexpensive computer could be used in lieu of the more expensive and heavy PXI chassis in watertight case Modification to Instruments Once the scanner hardware was fully integrated to the new LAN protocol the instruments themselves had to be modified to communicate over the new all digital data communication system The three instruments in use with the first stage scanner systems the covermeter shear wave ultrasound and impact echo devices communicated over serial USB and analog voltage formats respectively Through prior discussion with FDOT and based on the recommendations of the research team it was decided that the impact echo device could be eliminated from the automated NDE system for this phase of development In the course of the pre
146. pockets can all lead to entrapped voids within a concrete structure and compromise its structural integrity For these reasons it was decided that an additional NDE evaluation block needed to be fabricated containing simulated honeycombing delamination and entrapped air around rebar and simulated cracking The Void and Flaw Detection Block the sixth and newest ANTARES Laboratory NDE Evaluation Block was designed to fit in the standard reusable horizontal block 162 mold The block would contain three pieces of 6 rebar one would have plastic tubing affixed to simulate entrapped air one would be moved horizontally in the mold X axis after initial set of the concrete and one would be moved vertically in the mold Z axis after initial set The three rebar lengths would be located in the middle third of the block while on both ends forced honeycombing and cracking would be put in place On one end section of the block three prisms of pervious concrete would be suspended along the central plane of the block The other end would have an additional prism of pervious concrete placed at surface level during concrete placement along with two simulated angled cracks On each end a small region of the block was reserved for a more realistic but amorphous in shape form of simulated honeycombing On the end with the embedded features at the time of concrete placement a measure of the concrete mix wet sieved to approximately the same volume as the em
147. position American Society of Civil Engineers 2001 11 M Sansalone N J Carino and N N Hsu A Finite Element Study of Transient Wave Propagation in Plates Journal of Research of the National Bureau of Standards pp 267 268 1987 12 J Krautkramer and H Krautkramer Ultrasonic Testing of Materials Journal of Sound and Vibration pp 157 158 1969 13 Vu Con System R ImpactEcho System For Field Evaluation of Concrete Structures NDT James Online Available http www ndtjames com Accessed 10 June 2013 219 14 American Society for Testing and Materials ASTM C1383 04 Standard Test Method for Measuring the P Wave Speed and the Thickness of Concrete Plates Using the Impact Echo Method West Conshohocken PA 2004 15 1 O Yaman G Inci N Yesiller and H M Aktan Ultrasonic Pulse Velocity in Concrete Using Direct and Indirect Transmission AC Materials Journal pp 450 457 2001 16 American Society for Testing and Materials ASTM C597 09 Standard Test Method for Pulse Velocity Through Concrete West Conshohocken PA ASTM International 2009 17 Germann Instruments Inc EyeCon Evanston Il 18 J Hugenschmidt and J P Balayssac Ground Penetrating Radar in State of the Art Report of the RILEM Technical Committee 207 INR 2012 19 Ground Penetrating Radar Equipment StructureScan Minis Geophysical Survey Systems Inc Online Available http www geophysical com str
148. que in determining viscoelastic properties of cement paste 22 While the research discussed in this thesis is concerned primarily with later age structures the effect of early age material properties on the behavior of the shear wave ultrasound technique cannot be ignored as the automated system will inevitably be applied in the future to early age structures as a quality control measure Durability and Degradation of Concrete Introduction General degradation of concrete whether due to natural environmental weathering action chemical reaction or traumatic damage has a profound impact on the performance of reinforced concrete structures pavements and indeed any practical application of concrete as a primary material Degradation of concrete in steel reinforced concrete structures is especially harmful as increased concrete permeability leads to an increase in chloride ion infiltration paths Once an increase in permeability either locally or globally in the structure is sufficiently severe corrosion of the reinforcing steel within the structure will initiate and inevitably lead to a structural failure if left unmitigated 49 Durability is a term which describes the ability of concrete to withstand environmental degradation and prevent infiltration of corrosive agents to the reinforcing steel contained within a structure While the definition of a durable concrete product and a less durable concrete product is somewhat subjec
149. r is or emergency stop They represent the highest programming tier that does not require direct human control The number of SubVIls in this class is far too great for detailed individual discussion so only a handful of important VIs are discussed in this chapter 114 The Autoscan VI is arguably the most important and complex SubVI in the ANTARES SVW application This VI is responsible for accepting user defined scanner grids instrument settings and data file paths and performing an entire automated scan Since the user may decide to pause or interrupt an automated scan at any time the Autoscan VI must be able to receive commands from the main interface during execution and alter its own function Since the Autoscan VI is a SubVI which must fully execute before relinquishing control to the calling VI per LabVIEW execution methodology it was decided to force the Autoscan VI to relinquish control to the Main VI after each step a single point in a point by point scan or an entire line in a line scan to allow the main interface to update and alter behavior if necessary ANTARES SVW introduced the capability to run an NDE scan using a single scanner both scanners independently and simultaneously or both scanners dependently following a master slave protocol in which the slave scanner matches the mirror image coordinate of the scanner chosen as the master before data is collected This protocol allows more rapid data collection whe
150. rch Assistant University of Florida Program Ver This program processes ANTDAT data files produced by ANTARES SVW and DZT files produced by the GSSI SIR 3000 GPR Currently this software supports basic processing for Covermeter Ultrasound GPR and Laser Profilometer Future releases will include support for ANTARES SCS and Data Combination packages To process ANTDAT files simply open the file and this software will import the grid and data information and begin the appropriate processing mode To process DZT files open either the ANTGRD or ANTDAT file produced at GPR runtime and open the bank of DZT files corresponding to the GPR scan By default this program assumes each DZT file is a line scan and files are opened in the order the lines were run For this reason negative line or point displacements may result in an opposite movement sense in data visualization Please refer to the instructional manuals for further information and troubleshooting Figure 6 25 Welcome panel of ANTARES Processing Suite While designing a software package to output the various processed data sets into formatted report files proved to be impractical each plot display on the software panel could be exported as a simplified image or data set by right clicking on the plot itself These images could then be used to generate a report in the program of the user s choice Covermeter processing Processing of covermeter data proved to be relatively
151. red and the susceptibility of the instrument to collision during operation Photo by J Nelson A small T slot extruded aluminum frame polycarbonate enclosure was designed to fit inside the space between the back side of the connector flange on the moving instrument head and the Y axis linear drive as shown in Figure 4 2 and expounded 83 upon in Appendix B 1 The NDE instrument and any supporting components to be used were mounted onto a polycarbonate panel which when inserted into the aluminum frame formed the rear panel of the instrument enclosure The exposed end of the enclosure was then closed with an aluminum panel fastened by aluminum thumb screws This methodology allowed maximum available space for instruments as well as a small level of protection against water and debris infiltration Additionally since the dimension of the back panel of the instrument carrier was standardized between the two scanners a new instrument may be incorporated with the scanner simply by fastening it and any supporting hardware to an inexpensive piece of cut polycarbonate sheet Similarly the aluminum right angle bracket which fastens the instrument sensing head to the pneumatic cylinder is a standardized part which can be fabricated at very low cost Figure 4 2 CAD drawing excerpt and 3D render of new instrument carrier design 84 Figure 4 3 Completed instrument carrier as mounted to the Y axis linear drive on Scanner 1 Photo by J N
152. rols for all the numerical and enumerator values necessary to visualize run and recreate an automated NDE grid This makes the cluster ideal for direct flattening to a binary string in the same manner as the configuration settings cluster ANTARES Data File ANTDAT The ANTARES Data File is a simplification of the multiple file types and formats utilized by the first stage system software The Phase software wrote a unique text file 131 for each instrument and separate file for each data point or line of data depending on the instrument This made the files difficult to interpret and subject to inadvertent separation and loss of individual files ANTDAT format files contain an entire automated NDE scan and share a common header and data range format They are text based tab delimited files that can be read from any number of available third party programs as well as the ANTARES Processing Software ANTARES SVW Data File Program Version 1 0b ANTARES Laser Instrument Profilometer Date Time of creation 5 30 2013 10 16 Grid Settings Origin X 106 Origin Y 255 Line Orientation X Point Spacing 25 Line Spacing 10 Point Count 80 Line Count 74 Instrument Data Time S Coord X S Coord Y I Coord X I Coord Y A Scan 5 30 2013 10 16 106 255 168 315 86 2 Figure 6 24 Excerpt of an ANTARES Data File ANTDAT showing generic file format and standard header information ANTARES Data Files contain the program version instru
153. s a result the later age durability of a concrete placement 50 0 0040 0 0035 0 0030 0 0025 0 0020 vy UPV amp 0 0015 Field Deck S26 0 0010 0 0005 0 0000 12000 13000 14000 15000 16000 17000 UPV ft s Figure 2 19 Yaman s test apparatus and results showing a large UPV difference between field and laboratory concrete This difference is attributed to Paste Quality Loss PQL 24 Yaman employed an array of UPV transducers to generate a statistical model of the pulse velocity of a bridge deck placed in the field When compared to the pulse velocity of standard specimens taken in the laboratory there was a significant difference in the average UPV reading that correlated closely with the PQL value calculated using the standard method This correlation was also shown valid in areas of known defects such as poor compaction In addition to establishing the basis of UPV as what could be a valid permeability test method for concrete bridge decks Yaman showed the superiority of the method versus the Figg s apparatus in terms of speed The UPV based measurement took roughly half as long as a measurement using the Figg s apparatus a speed increase that could likely be improved if purpose built UPV systems were fabricated 24 Nondestructive evaluation of sulfate attack Sulfate attack in concrete is a mechanism by which environmental conditions promote the formation of ettringite C A 3CSH3 crystals
154. s are responsible for operating the servo controls in a limited feedback control scheme where the movement of the servo motors is controlled in the form of movement steps or pulses of the stepper motors and absolute values are stored locally on the operating computer The movement command is parsed as a text stack including desired motor state in pulses and the velocity and acceleration parameters for the movement step This text stack is then passed over the network to the scanner s serial port and interpreted by the onboard controller in the master servo motor The VI used to accomplish a CML motor command is shown in Figure 6 3 110 Cool Muscle Defaults S x in 100 s of Pulses Sec A x in 1000 s of Pulses Sec squared kpps2 P x in pulses X Input mm Dsi Y Input mm P514 Scanner Global Settings Default Default vp Command LO Figure 6 3 Parsing structure of CML Command SubVI Command and query functions for the pneumatic cylinder relay were also generated as low level subroutines These Vis operated the digital out ports of the 5V TTL cDAQ card to command the pneumatic instrument head to move up or down and read in the status of the instrument head from inductive proximity sensors on the digital in ports These subroutines are shown in Figure 6 4 and 6 5 and are available in full form in Appendix A 111 9 2013 Scanner to Control 2 amir Port Cluster 2
155. s expected The change in cover on the rebar that was moved upwards in the mold after initial set was very apparent in the C Scan image as shown in Figure 7 25 which would inform the investigator of a moving reinforcement issue despite the lack of an obvious indication of damaged concrete in the vicinity of the rebar In a tomograph data view the entrapped air at the surface of this moved rebar element is visible as a weak diffused surface reflection as shown in Figure 7 26 175 Figure 7 25 GPR C Scan image of rebar in Void and Flaw Detection block at a depth of 105mm Figure 7 26 XZ Plane GPR tomograph of the Void and Flaw Detection block By carefully adjusting the depth of the C Scan image and fine tuning the contrast of the display some evidence of the enclosed honeycombing also became visible It is 176 clear that while GPR is capable of detecting such flaws its data needs to be carefully scrutinized to do so successfully and it is still susceptible to missing smaller defects Indeed the smaller pervious concrete prisms and the surface visible cracks were not apparent in the data visualizations produced by the ANTARES Processing Suite Figure 7 27 GPR C Scan image of Void and Flaw Detection block Shear wave ultrasound Shear wave ultrasound was the most promising technique used to evaluate the Void and Flaw Detection Block The technique is sensitive to density changes and has shown great reliability in the detect
156. self as simple as possible and allowed an open source unprocessed image to be produced for later processing should future research efforts invest the time in creating vision processing algorithms The run time program was written using the LabVIEW Real Time module with support from the LabVIEW Vision Acquisition module as shown in Figure 5 8 The VIs that would be used to control the camera acquisition exposure settings lighting settings and operate the ring light array itself were designed to interface with future 102 control program development using the same NPSV methodology that allowed the image files to be transferred to the control computer over the network It was decided to transfer and store captured images as flattened data strings despite being far more memory demanding than a binary file so that the output data could be easily interpreted and processed using other software packages and used by other personnel if desired The completed Vis were published as a real time application rtexe and written to the nonvolatile memory of the Smart Camera so that the program would execute immediately on power up of the camera Figure 5 8 Startup subroutine of SCS run time application This section of code activates the Smart Camera and sets the default exposure time gain and lighting settings for the camera The ring light is illuminated briefly to inform the operator the device is workin
157. sfully FDOT requested a trial field deployment of the system to evaluate the feasibility of using the system to evaluate the condition of real world 80 structures in situ The field deployment was to be performed at the discretion of the research team and could be a vertical or horizontal application of a single or both scanning frames The research team delayed choosing a site and time for deployment until the scanning system was sufficiently complete and FDOT could recommend a site that would offer them a useful set of data for their own research as well as a convenient location to deploy the system 81 CHAPTER 4 IMPROVEMENT OF INSTRUMENT AND SCANNER CONNECTIVITY Approach At the beginning of this sponsored research project FDOT made it clear that this new automated NDE system was to be considered satisfactory only if it was able to be deployed in the field to evaluate structures in situ This application warranted a system that was able to withstand the rigors of field deployment while being simple and reliable to set up In the interest of simplifying the earliest software development efforts while also addressing this end goal it was decided to simplify and ruggedize the system interfaces as the first step in the development process The first phase of this effort was to design and implement an improved instrument mounting system capable of protecting the relatively fragile and valuable NDE instruments once attached to the scanners
158. straightforward since the data format is scalar in nature The ANTDAT file generated for a covermeter scan contains a table of coordinates for the instrument and the measured cover in millimeters 135 at that coordinate The instrument itself performs all necessary calculations and adjustments onboard and reports the measured cover exactly as a technician using the instrument in the field would use it A data realization VI shown in Figure 6 26 was developed to generate a two and three dimensional plot from the provided covermeter data This VI accepts the values from the instrument coordinates and A Scan columns of the ANTDAT file the loaded grid settings and a desired number of points for a moving average smoothing filter if desired The VI then generates various data matrices suitable for plot generation using the built in tools found in LabVIEW for three dimensional scatter plots and two dimensional intensity plots gt i 2013 r Moving Average Points sam fe RI I This methodology should work for line scan or point by point ral LJ 3D Scatter Out Inst Coords In L20 Plot Helper Matrices 7 C Scan Out Lines Run in X Defaul p 0 5 gt g
159. suspended from fiberglass rods designed to support the prisms weight during formwork construction as shown in Figure 7 14 The fiberglass rods were also stiffened vertically and laterally to hold the prisms down against buoyant force and laterally in place during concrete placement The fiberglass rods were chosen to be small enough and similar enough in density to not appear on ultrasound scans and were non metallic so as not to reflect radar waves 165 Figure 7 14 Pervious concrete prisms suspended along the middle plane of the formwork for the ANTARES Void and Flaw Detection Block Photo by J Nelson The remaining pervious concrete prism along with two triangular plates that would be used to form the surface visible simulated cracks on the other end of the block were to be inserted at the time of fabrication Each triangular plate was used to form a 254mm 10in long crack of linearly varying depth from 0 to 76 2mm 3in The plates were set into a small slot cut into a piece of lumber and epoxied in place to make installation and removal easier The three sections of 6 rebar for the middle section were suspended in place by recessing the last 50mm of their length on either end into a layer of polymer coated form board installed inside the main boards of the reusable formwork as shown in Figure 7 15 The rebar pieces to be moved after initial set of the concrete had one end installed into a slot rather than a hole in the inner form
160. t investment of time into the development of the system 21 Chapter 7 emphasizes the design construction and subsequent non destructive evaluation of a pair of concrete blocks intended to simulate a number of structural defects likely to be found in real world concrete structures This effort is a direct continuation of earlier research discussed in Chapter 2 but demonstrates the improved usability and data processing capabilities of the system Chapter 8 reviews a field trial of the system and addresses the viability of automated NDE as applied to reinforced concrete structures currently in use Chapter 9 closes with a summary of the results of this research as well as recommendations for continued research and development by FDOT and other agencies 22 CHAPTER 2 BACKGROUND NDE Methods Considered in this Research The list of NDE methods that have successfully been applied to concrete structures is incredibly long and is growing every day For this reason instead of presenting an exhaustive review of a large collection of methods this chapter investigates in depth only those few methods used specifically with the automated system and a few older methods that illustrate how the method of interest is utilized While one of the main goals of the development of the automated system was to create a system capable of expansion to include new NDE instruments a review of a myriad of potential instruments to be included in future research
161. tance of the ANTARES system as an investigation tool capable of collecting spatially dense data sets Ultrasound of the Asymmetric Internal PT Duct Block also revealed an unintentional defect the formwork had deflected significantly during concrete placement resulting in curved outer surfaces This defect was clearly visible in the tomograph view shown in Figure 7 8 I I I I I I I I I 1 I I U 150 200 250 300 350 400 450 500 550 600 650 700 750 800830 Line Figure 7 8 View of the Asymmetric Internal PT Duct Block shear wave ultrasound data in the XZ plane The back wall reflection bottom white shows the curve induced by formwork deflection 158 Early scans of the deep cover side of the block revealed a loss of sensitivity to grout defects compared to the shallow cover scams Though the shear wave ultrasound device could detect the presence and depth of the internal post tensioning ducts as shown in Figure 7 9 the difference in shear wave reflections from quality grouted duct areas compared to debonded and air voided duct areas was unclear Since the stress wave loses energy as it travels through the concrete medium it is likely that the density and consequently the shear wave velocity change from the concrete to the polypropylene medium at the duct surface resulted in a complete reflection of the stress wave back to the source with no influence from the condition of the grout within the duct Scan Depth 155 mm
162. te transfer of the stress wave The transducer is driven by an electric function generator in transmit mode and the impact of the transducer on the concrete surface generates a series of collimated longitudinal stress waves The stress waves are received using an identical transducer whose piezoelectric crystal is connected to a data acquisition system which records the voltage generated by the movement of the transducer The time of flight of the wave pulses is calculated and averaged and the ultrasonic pulse velocity is determined by dividing the measured transit distance by the calculated transit time Piezoelectric Crystal Hard Steel Contact Acoustic Couplant Figure 2 13 Operating principle of a piezoelectric pulse wave ultrasound transducer Note the longitudinal stress wave and corresponding particle motion The receiving transducer may be located on the opposite side of a planar specimen from the transmitting transducer in a technique known as direct transmission 39 or located on another surface in a technique known as indirect transmission Due to the difficulties in measuring an accurate transit distance and ensuring constant material properties along the wave path in indirect transmission mode ASTM C597 currently only allows direct transmission to be used to determine the ultrasonic pulse velocity of a concrete specimen More recent research and improved practices have shown that in certain circumstances i
163. the plot Figure 6 28 Example of a 3D scatter plot covermeter data visualization in the ANTARES Processing Suite Laser profilometer processing Functionally the processing scheme for the laser profilometer was no different than that of the covermeter The format of the data is identical a single numerical data point for each coordinate however the laser distance measurement affords another decimal of precision resolving to 0 1mm under ideal conditions A C Scan image of a laser profilometer scan of a rough bridge deck is shown in Figure 6 29 where the 138 difference between the highest light pixels and lowest dark pixels regions of the display is measured using the cursor to be 15 4 mm Figure 6 29 Example of a 2D laser profilometer data visualization of a rough experimental bridge deck When rendered in three dimensions laser profilometer data presents the operator with a movable colorized surface representative of the scanned area that can be used to quickly identify overall shape and texture of the specimen surface A three dimensional display of the rough bridge deck mentioned discussed earlier in this section is provided in Figure 6 30 139 Figure 6 30 3D render of laser profilometer data a rough bridge deck surface The laser profilometer data is usually collected in line scan mode meaning the instrument is read continuously at a high rate while the scanner traverses a grid line ata constant velo
164. ting cord were attached to the GPR antennae mount and designed to replace the remote control handle installed on the GSSI pole extension The mount is shown in Figure 5 2 GPR Mount Component 1 Extended standard pneumatic head mounting bracket W 2 Adjustable articulating adapter for AES GSSI pole mount z ie GSSI pole mount z T Connector for remote deadman S switch 5 Rolling cart with 2 6GHz antennae 6 Rolling adjustable limit switch Figure 5 2 GPR antennae mount with limit switch for FCC compliance Photo by J Nelson 96 FCC regulations require a deadman type switch to control activation of all GPR antennae due to their ability to interfere with nearby communication systems and potentially harm human tissue The limit switch was a convenient way to mimic the deadman switch and comply with FCC regulations while not modifying the as bought GPR system in any way The final and most important obstacle to assimilating the GSSI GPR system was retrieving and saving data from the SIR3000 unit While the SIR3000 had multiple interfaces built in for its embedded PC operating computer their functions were limited to only low level interfacing operations and manual data transfer It became clear early in the project that the system simply could not be integrated in the same manner as the previous instruments While initially a significant setback to research efforts the inability to interface the GPR system with th
165. tive and depends heavily on the environment in which the concrete is placed the characteristics of a durable concrete are low permeability good chemical and biological compatibility with the environment ability to resist freeze thaw degradation if applicable resistance to bleed during placement and good resistance to wear 23 While standardized tests exist for estimating permeability and wear resistance of concrete non destructive evaluation techniques are frequently requested to more rapidly detect significant loss in concrete durability due to weathering cracking alkali silica reaction ASR sulfate attack fire damage and other common sources of catastrophic concrete failure This subset of concrete NDE continues to be the subject of research efforts across the scientific community Permeability of bridge decks using ultrasonic pulse velocity In a Michigan DOT funded study I 0 Yaman and a research team attempted to better describe concrete permeability in bridge decks using the UPV method 24 Yaman compared UPV measurements of a bridge deck to UPV measurements of a standard ideally cured cylinder specimen in a manner similar to an established test metric known as Paste Quality Loss PQL PQL uses a porosity probe known as a Figg s apparatus to compare the permeability of field placed concrete with ideally fabricated concrete to determine to what degree the past is compromised This metric establishes the later age porosity and a
166. to the scanner concrete surface GPR Ground penetrating radar is sensitive to the dielectric constant of the material upon which it impinges In the case of concrete if there is a large amount of free water within the capillaries of the specimen an incident radar wave will be almost completely reflected off the surface This is a well known phenomenon and GPR manufacturers 174 recommend applying radar techniques to specimens that are at least moderately dry for investigation purposes For this reason GPR scans were delayed for 28 days until the specimen had chemically absorbed or expressed enough free water to be viable for inspection by GPR The block was scanned using ANTARES and GPR in a parallel operation manner on the form side B side of the block to attempt to detect the enclosed honeycombing and air voided steel GPR is not usually the NDE method of choice for detection of enclosed air voids due to reasons discussed in Chapter 2 and it takes an experienced operator to discern defects such as honeycombing due to their irregular interfaces and low GPR reflectivity Nevertheless it is a common practice in modern field structural NDE to use GPR to evaluate structural integrity in cases where poor consolidation is suspected For this reason a detailed GPR scan was conducted on the form side of the Void and Flaw Detection Block Initial data visualizations clearly showed the locations of the three rebar specimens within the block a
167. tor P Gain Lighting Model Figure A 4 SCS Startup VI Sequence 0 201 Camera Shutdown Figure A 5 SCS Startup VI Sequence 1 main loop 202 u AR Es al WAR L J Time Stamp Out f f L j D J w a m Figure A 6 SCS Image Stream VI 203 Cool Muscle Defaults Sxin 100 s of Pulses Sec Ax in 1000 s of Pulses Sec squared kpps2 P x in pulses Step Position bd Scanner Port Cluster Scanner 1 Default v fr Scanner to Control Liz 4 gt X Input mm 05 Y Input mm 05 Scanner Global Settings LE EEBEE Default Default v Custom Velocity Velocity Velocity Lo 4 Command Le Doouoo ooo Oooo oD ooo oD oo Doo oD oo Uo ooo Uo ooo ooo oo ooo UU ooo UU ooo oo error out Figure A 7 CML Command VI 204 2013 CT ee J Scanner to Query i i l Bla l l l Timeout ost i n J error out Raw Chatter Figure A 8 CML Query VI 205 7 gt 4 2013 2 Write value to global variable 0 Send command to pneumatic cylinder L Wait for confirmation Send message if error or timeout 5 Enum strict H a Scanner 1 Default 7 i E Vae A f Scanner to Control Laut x Sens
168. truments and supported extensively by laboratory tests for material properties Blocks may be loaded into a steel clamping frame which also serves as a mount for the two ANTARES scanner systems for simultaneous front and back access or can be scanned using manually operated instruments for comparison with automated techniques FDOT has already executed several collaborative research contracts with other agencies centered around data collection and comparison with the Laboratory Evaluation Blocks at the time of the writing of this thesis 86 37 As a result of this utility and corresponding interests it was decided that the research team would produce two additional Laboratory Evaluation Blocks in the course of this research effort The first block would be designed as an improved internal post tensioning duct evaluation block and the second would include some common construction defects including honeycombing deep cracking and delamination of reinforcing steel Asymmetric Internal PT Duct Block Design While the Phase system produced by Dr Algernon produced a specimen for evaluating internal post tensioning ducts it was noted in the conclusion of his research 150 that the specimen did not perform as expected and included unintentional defects that prevented its use as an effective evaluation specimen for NDE instruments 9 Furthermore the original PT duct block was constructed with galvanized steel internal ducts which are no longer t
169. ts and the construction of the block as an evaluation of the performance of the operating software itself A series of scans were performed emphasizing ultrasound above all other instruments due to its potential to detect defects in grouted ducts Some data was taken on older software or even manually before ANTARES SVW was ready 155 Initial NDE scans Before the ANTARES SVW software application was complete manual GPR scans were made of the block to attempt to discern the difference between air voided ducts and properly grouted ducts The data was qualitative in nature but did indicate that modern high frequency GPR was capable of detecting significant air voids in non metallic post tensioning ducts As shown in Figure 7 6 the GPR B Scan plot shows strong double reflections of the ungrouted ducts 1 and 4 at all three scan locations The partially grouted ducts 2 and 3 show a weaker reflection terminating in a reflection of only the steel strand within the duct as the grout fraction of the duct increases indicating more wave energy was transmitted through the duct interfaces Figure 7 6 Manually performed GPR scans illustrating the effect of air voids in plastic internal PT ducts on GPR reflections Additionally shear wave ultrasound scans were performed using the older operating program ScannerVoll Wizzard These data sets were then processed using 156 simplified disposable programs built in LabVIEW As shown in
170. two dark traces indicated clearly the intersection between the keyway joint and the girder Two regions were identified at the grout interface where a significant ultrasound reflection had occurred which should not occur if the grout layer was complete and properly bonded An additional suspect area was initially noted but after comparison with the laser profilometer data it was determined to be a reflection from a grout port and not likely an actual defect The C Scan image is presented in Figure 8 9 Filter Settings Fly Through C Scan Ultra Scan Depth 154 mm 556 us Line Figure 8 9 Ultrasound C Scan image at ABC Location 1 The two areas of suspected incomplete grouting are circled in red while a false defect corresponding to a grout port is indicated with a green arrow 192 Scans conducted at location 2 revealed similar results The ultrasound scan indicated one area of severe delamination on top of the girder and one area that was suspect but not severe Interestingly the adjacent bridge deck element across the keyway joint partially visible in the scan had a much stronger bottom surface reflection than any area previously investigated It is hypothesized that this section of the bridge deck was not in contact with the earth below the deck elements due to erosion or foundation settlement and therefore exhibited a stronger backwall reflection Filter Settings Fly Through C Scan Tomograph Ultra Scan Depth 147
171. ucturescanmini htm Accessed 10 June 2013 20 L Al Qadi and S Lahouar Measuring layer thicknesses with GPR Theory to practice vol 19 2005 21 S Popovics J L Rose and J S Popovics The Behavior of Ultrasonic Pulses in Concrete Cement and Concrete Research pp 259 270 1990 22 T Voigt The Application of Ultrasonic Shear Wave Reflection Method for Nondestructive Testing of Cement Based Materials at Early Ages An Experimental and Numerical Analysis 2005 23 ACI Committee 201 ACI 201 2R 08 Guide to Durable Concrete Farmington Hills MI American Concrete Institute 2008 24 O Yaman R Birgul H M Aktan N Hearn and J F Staton Test Method for Appraising Future Durability of New Concrete Bridge Decks vol 1798 2002 25 A J Boyd B Birgission C Ferraro and S Cumming Nondestructive Testing for Advanced Monitoring and Evaluation of Damage in Concrete Materials University of Florida Gainesville Fl 2005 26 M Sargolzahi S A Kodjo P Rivard and J Rhazi Effectiveness of Nondestructive Testing for the Evaluation of Alkali Silica Reaction in Concrete vol 24 no 8 2010 220 27 R Beutel H W Reinhardt C U Grosse A Glaubitt M Krause C Maierhofer D Algernon H Wiggenhauser and M Schickert Comparitive Performance Tests and Validation of NDT Methods for Concrete Testing vol 27 2008 28 M Scott A Rezaizadeh A Delahaza C G Santos
172. use by technicians and prone to poor behavior when used improperly The software package also included little room for expansion as new instruments were incorporated into the system 77 Figure 3 3 Phase scanner operating software based in multiple subroutines and windows leading to an overly complicated interface Design Goals Expanding the Capabilities of the FDOT NDE Validation Facility The FDOT NDE Validation Facility or Phase system established by the research performed by Algernon et al was an excellent first stage facility for the FDOT s foray into automated NDE A logical next step in development for the FDOT was to expand the capabilities of the Phase system by generally improving the hardware and software interfaces for better system survivability integrating additional NDE instruments rebuilding the system around a more cohesive and advanced software package designing and constructing new NDE validation blocks and proving the system as a whole is viable for deployment in the field on real world infrastructure 78 components This need for research as expressed by the FDOT was addressed by the research team in the course of this thesis Hardware Improvement FDOT expressed several deliverable goals in the design of this research effort in terms of ruggedizing and simplifying the ways in which the automated scanning system interfaces with the system operating computer First and foremost if the scanners were to
173. vious research by Algernon et al it was shown that for the sorts of structural defects and characteristics FDOT wished to investigate the impact echo method offered few advantages over shear wave ultrasound and was not worth the cost associated with converting it into an all digital device 9 Since both methods are based on stress waves they are sensitive to the same sorts of defects however the shear wave ultrasound instrument offered higher sensitivity better resolution faster operation and easier integration with the scanner system as its data output was already digitized The Proceq Profometer 5 eddy current covermeter originally operated on a 9VDC supply provided by 6 AA batteries and communicated its data using a serial byte stream over an RS232 interface It was found that the AA battery supply was insufficient to power the device for the duration of a single scan so the device was modified to accept a low profile plug in transformer that could be conveniently connected to the ABS enclosure on the scanners instrument heads A Silex serial device server was 90 connected to the covermeter s RS232 port and the instrument and server were mounted on a polycarbonate panel for convenient installation into the instrument carrier as shown in Figure 4 7 and 4 8 The rolling cart sensing head used by the covermeter instrument was mounted to the standard aluminum mounting bracket using a rubber buffer to allow the cart to adapt to unev
174. xes due to varying material properties and overall material behavior Attempting to estimate the strength of a particular concrete mix using an estimated elastic modulus and an empirical correlation without first establishing a calibration data set for the mix proved to be highly unreliable 21 Early age elastic properties of concrete using shear wave ultrasound A decade and a half after the groundbreaking research by Popovics Dr Ing Thomas Voigt of the University of Leipzig employed shear wave ultrasound reflection in a novel approach to evaluating the elastic properties of very early age cementitious specimens 22 In response to demand form the construction industry to better evaluate the quality of in place concrete days or even hours after placement Dr Voigt addressed the feasibility of using shear wave reflections to monitor the set of cementitious materials as their complex viscoelastic modulus becomes increasingly solid like Shear waves were passed through a Poly methy methacrylate PMMA buffer into an early age mortar specimen as shown in Figure 2 18 and the reflections were recorded to determine the energy lost The loss in wave energy is directly correlated to the viscous portion of the material s modulus and as such a lower loss in energy is readily correlated to the completeness of set of the cementitious material 22 48 Figure 2 18 Schematic of Voigt s experiment illustrating the usefulness of the shear wave techni
175. y the severity of ASR UPV Dynamic modulus of elasticity Nonlinear frequency shift and harmonic generation Petrographic examination 55 Of particular interest to this thesis is the investigation using UPV as wave velocity can be easily gleaned from the shear wave ultrasonic array in a manner analogous to the indirect UPV method While Sargolzahi noted that UPV did correlate well to ASR expansion by showing a reduction in wave velocity due to the increase in void fraction within the specimen the research showed that the technique is not particularly sensitive when compared with other methods 26 so x 30 Elastic matulus GPa Pulse ve beity vs 0 5 10 15 2 25 Age month Age moath Figure 2 23 Results of Sargolzahi s research using UPV to detect ASR Data in blue is from nonreactive specimens while orange is ASR reactive The static elastic modulus test left showed excellent sensitivity compared to UPV right 26 The results of Sargolzahi s research dissuaded the research team involved in this thesis from investigating ASR using the automated system at this time With the recent success of chemical tests for ASR using dye indicators a better approach may be to use a chemical method coupled with vision acquisition to automate early age detection of ASR in future research Related NDE Research Structural NDE Geometry and Locating Applications Introduction One of the most basic requirements of NDE

development of an advanced automated nondestructive evaluation

Contents

Download Pdf Manuals

Related Search

Related Contents