Home

Impact-Echo User`s Manual

1. 1 Nt 48 kHz Help Cancel Accept Figure A 1 The Parameters for Data Acquisition screen The parameters that can be selected or changed on this screen and their role in the analog to digital signal conversion process are the following Number of Samples When a test is performed the system always records 2048 data points voltages separated by the time interval specified under Time Between Samples Because these voltages are proportional to displacement a plot of voltage vs time is called a displacement time curve or waveform The Number of Samples entered on this screen is the number used by the computer to plot the waveform and to calculate the amplitude spectrum Specifying the number of samples as a power of 2 i e 512 1024 2048 makes the most efficient use of the mathematical algorithm called fast Fourier transform or FFT used to calculate the amplitude spectrum from the waveform The default value for this parameter is 1024 a value suitable for most testing The effect of using different numbers of samples to calculate the 101 spectrum can be explored by changing the horizontal scale for the waveform when a record is displayed Time Between Samples An important aspect of impact echo testing is the reliance on short sampling periods to limit the response to P wave reflections from the region directly beneath the impact point The sampling interval At must be sufficiently short to resolve the highest frequ
2. This situation is frequently encountered on concrete bridge decks for example where widespread cracking called delamination occurs at shallow depths due to corrosion in the reinforcing steel The resulting signal includes a large amplitude low frequency component due to the flexural vibration Flexural vibrations are similar to the vibrations in a drum and because the resulting surface displacements are far larger than those caused by P wave arrivals they dominate the sgnal The higher frequency component due to multiple P wave reflections across the thin layer is weak by comparison and sometimes difficult to detect This phenomenon is discussed in Chapter 10 of Sansalone and Streett pp 99 114 A schematic representation of the effects of flexural vibrations in thin layers is shown in Figure V 4 Flexural vibrations shown schematically in a have low frequencies typically 2 6 kHz and very large amplitudes compared to surface displacements caused by the arrival of reflected P waves shown schematically in c Figures b and d show the corresponding contributions to the spectrum flexural vibrations produce a high amplitude low frequency signal that dominates the waveform and spectrum while the peak resulting from Pwave reflections has a higher frequency and lower amplitude and is sometimes too small to be seen There are two methods for amplifying this high frequency peak 1 using a smaller impactor and 2 digital filtermg
3. Figure X 5 Examples of normal R waves Separated and irregular R waves For high energy impacts the R wave can become separated into several segments by steep vertical lines as a result of the transducer being bounced off the surface and momentarily losing contact In other cases an irregular surface or crushing under the point of impact can result in an irregular shape in place of the usual rounded bottom of the R wave well Examples of separated R wave are shown in Figures X6 a and b and an irregular R wave is shown in c Record 15 in file c ImpactDemo TestFile dat exhibits a separated R wave a b Separated Irregular fy N Figure X 6 Examples of abnormal R waves a and b separated R wave and c irregular R wave The presence of a separated or irregular R wave does not mean that the signal is invalid but suggests that the spectrum should be examined carefully A separated R wave for example often introduces high frequency components into the waveform that are not related to the important P wave reflections 95 Removing or cutting the R wave The problem of a separated R wave is solved by removing it from the waveform prior to transforming the signal into the frequency domain 1 To see an example of the distortion in the spectrum caused by a separated R wave start the ImpactDemo program and open the data file c ImpactDemo TestFile dat Open record 41 The waveform and spectrum are similar to Fig
4. Help Name Prev Record Replace Record J Label Peak Main Menu mau SEE in File 46 Z Show Record List Compare 9 To Destination File Restore Low Freg Figure VIII 1 Record 25 with the transducer resonance filtered out and the R wave clipped at 0 2 volts 2 Click Next Record Keyboard press Alt X to open record 26 The main peak in this case lies at 23 4 kHz corresponding to P wave reflections from a depth of 81mm or about 3 2 inches these numbers appear in the gray boxes at the upper right of the spectrum graph and in the plate cross section at the right of this graph Analysis and Interpretation Wave reflections from a depth of 8lmm indicate that the two layers of concrete are debonded Note that in addition to the transducer resonance peak at 1 0 kHz there is a small peak at about 2 kHz probably due to flexural vibrations in the thin top 71 layer of concrete To see the effects of removing these low frequency components place the spectrum cursor at about 10 kHz and click Filter Low Freg The effect of cutting or removing the R wave can be seen by placing the active waveform cursor at about 228 us and clicking the Cut R Wave button The result should appear as in Figure VIII 2 showing a single dominant frequency at 22 9 kHz corresponding to a depth of 82mm or 3 2 inches x Examine Test Data File c magodemo TestFile dal Str Name Type Plate Date Time __ 11713794 12 00 00
5. Measure Wave Speed Di Depth Waller Examine Test Data Eile Utilities Help Set Data Acquisition Parameters Begin Impact Echo Test Program Parameters Exit Figure II 8 The Main Menu screen with all command buttons enabled The next step in preparation for impact echo testing is to describe the structure by entering the structure name characteristic dimensions and P wave speeds One or more of these quantities may have to be estimated In this example impact echo tests are to be performed to determine the thickness of concrete plate The thickness is unknown but is thought to vary from about 300mm to 600mm A starting estimate of 400mm will be used The P wave speed has been measured independently using two transducers held a fixed distance apart on the surface the method is described later and found to be 3996 m s Describe the Structure 17 Click the Describe Structure command button on the Main Menu screen The Choose Structure Type option box appears Figure II 9 A plate structure about 400mm thick is to be tested Choose Structure Type Plate Plate With Overlay Circular Column Rect Beam or Column e Cc C Sq Beam or Column e o Hollow Cylinder Figure II 9 The Choose Structure Type option box 18 Select the Plate option and click OK The Plate screen appears Figure II 10 Plate Thickness Structure Name Test Plate Thickness mm P Wave Speed 3996 mis Help Can
6. The voltage setting V appears at the lower left of the waveform graph when the results of a test are displayed on the computer screen If the vertical scale on the waveform graph is set to the same value as the voltage and the waveform remains within the graph when a test is performed the output of the transducer is within the selected voltage range and V need not be changed If the waveform is cut off at the top and bottom of the graph the transducer output exceeds the selected voltage setting and V should be increased A voltage setting of 2 0 volts is satisfactory for most testing of plates beams columns and other common structures Higher voltage settings can be used if the signals are very strong and lower voltage settings can be used if the signals are especially weak For wave speed and crack depth measurements where the highest resolution is needed a default value of 0 1 volts is used Values of 0 2 or 0 1 volts are recommended for these tests Trigger Voltage This parameter instructs the data acquisition system to trigger or save a set of digitized data points when the analog voltage from the transducer unit reaches this value and the slope of the voltage time curve is negative the voltage is decreasing The significance of the Trigger Voltage parameter is illustrated in Figure A 3 which shows the leading part of a typical waveform 103 Arrival of stress wave at transduc Beginning of record x e Voltage T
7. These are discussed below a 4 Amplitude Amplitude Frequency kHz Frequency KHz Figure V 4 The principal components of the response produced by impact on the surface of a concrete slab containing a shallow delamination a flexural mode and c thickness mode The contributions to the spectrum are shown in b and d 7 Open record 12 in the file e ImpactDemo TestFile dat This shows the results of a test on the same 240mm thick plate examined in the preceding paragraphs 57 Recosnizing Flexural Vibrations In the waveform there is a low frequency component with a large amplitude that decays very slowly To confirm the slow decay click the 3 H Scale text box at the lower right of the waveform graph enter 2048 in place of 1024 and press Enter The result is a plot of the waveform over the entire range of 2048 recorded data points about 4 milliseconds The decay in the magnitude of flexural vibrations over this period is small To emphasize this point click Prev Record Keyboard press Alt V and compare the waveform in record 11 to that in record 12 To return to record 12 click Next Record Keyboard press Alt X These two records have comparable vertical scales so the amplitudes can be directly compared Notice how rapidly the amplitude of the P wave reflections in record 11 decays compared to that of the flexural vibrations in record 12 The presence in the waveform of a high amplitude low
8. With a seven pound notebook computer the system weighs about 13 pounds 6 kg and can operate on its own internal batteries for 2 3 hours or longer if extra notebook computer batteries are available A 12 volt DC source such as a car or truck battery can be used to power the system and or recharge the batteries Storage and Retrieval of Test Data When impact echo tests are performed with the field test system test results can be saved to a file in the computer for later examination and analysis and for printing copies of test results The results of a single test are stored as a unique record in a random access file The file can be on the hard disk or on a floppy disk or another storage device connected the computer 10 11 Section II Setup and Preparation for Testing Getting Started With ImpactDemo To set up a real impact echo test system cables are connected between the data acquisition system the transducer unit and the computer The instrument can be powered by internal batteries or by an external source 12v DC or 220 110v AC The data acquisition system and transducer unit are not needed when using the ImpactDemo software Their actions are simulated by the software If you have purchased an instrument with a computer ImpactDemo is already installed Installing ImpactDemo Software 1 Insert Disk 1 in the appropriate drive typically the a drive and select Start and Run in the lower left corner of the Windows scr
9. a compression wave at transducer 1 causes an upward surface displacement and a positive voltage time t while the diffracted wave that first reaches transducer 2 is a tension wave which causes a downward displacement and a sudden voltage drop time t The elapsed time between t and t the wave speed and the known distances H H and H are used to calculate the depth D In practice the operator measures the distances H Hs performs the impact and positions the cursors on the graph to mark t and t The software calculates the crack depth and displays it on the computer screen Setup for Crack Depth Measurement The following action steps describe the setup and procedure for determining crack depth and explain the interpretation of test results using data from real impact echo tests 1 Start the ImpactDemo program see p 14 and create and open a new data file c ImpactDemo CrakDpth dat on the c drive in the directory ImpactDemo see pp 16 18 or press Help for information on creating and opening a new file When control is 67 returned to the Main Menu screen click Measure Depth of Surface Opening Crack Keyboard press Alt C to open the Setup for Crack Depth screen Figure VI 4 wm Setup for Crack Depth Wave Speed Distance H1 Distance H2 Distance H3 Help Figure VI 4 The Setup for Crack Depth screen 2 Enter Test Slab as the structure name 3800 as the wave speed and
10. below Place the spectrum cursor on the highest peak 6 3 kHz and click J Label Peak Keyboard press Alt J The letter k is placed at the top of the peak and the frequency 6 3 kHz in this case appears next to it Up to ten peaks can be labeled in this fashion and the labels will appear on printed copies of the record The labels are not saved and will not reappear when this record is next accessed The Filter Low Fre q button is used to activate a digital filter that mathematically filters removes frequencies below the frequency marked by the current position of the spectrum cursor This is especially useful for removing an artificial signal associated with the natural resonance of the transducer unit which is about 1 kHz Open record 19 click on the box in the lower left hbeled Number Keyboard press Alt b This record 35 shows a large peak at 1 5 kHz in the spectrum This results from the slow oscillation of the waveform around the zero Place the spectrum cursor at about 4 kHz and click Filter Low Freq Keyboard press Alt q Observe the change in the waveform which now appears as a decaying sine wave dominated by a single frequency The caption on the button has changed to Restore Low Fre q Click on this button several more times and observe the changes in the waveform with the removal and restoration of the low frequency components This command must be used with care to avoid removing important fr
11. low even at 20 kHz This explains in part why the peak near 20 kHz in record 12 is small and suggests that if there were more energy in that part of the spectrum it would be larger A smaller impactor which produces stress waves with a broader range of frequencies might be effective in clarifying the structure in the spectrum around 20 kHz Click Restore W to restore the full waveform record 12 Place the waveform cursors on two adjacent large peaks in the waveform at 738 and 948 us for example The approximate period of the low frequency component of the waveform is 210 us which corresponds to a frequency of about 4 8 kHz the approximate frequency of the large peak in the spectrum Place the cursors on adjacent secondary peaks superimposed on the low frequency signal at 338 and 390 us for example Figure V 7 The horizontal scale has been set at 512 w Examine Test Data Str Name TestPlste Type Plte Date Time Thickness mm Wave Speed m s 3744 6 Thickness 7 8 Frequency kHz 2 5 v 1 Clip Level 25 me 487 u wat Contact Time Change Curs Cut R Wave Cut Waveform 2V Scale 1 25 3H Scale 512 60 Figure V 7 The waveform in record 12 with cursors on two adjacent peaks of the high frequency component The period is 52 us shown in the box at the upper right of the graph corresponding to a frequency of about 19 KHz This is a good indication that there is a frequency component near 20
12. 75 for each of the distances H Hs and click Begin Test to open the Measure Crack Depth screen Figure VI 5 w Measure Crack Depth ile c imagodemo CrakD pth dat Structure Name Date Time 8 17 93 12 00 00 AM Cursor to Cursor EI us Distance H1 Duar H2 Distanca H3 Seins Voltage Setting ELO v Sampling Interval 7277 u s Calculate Grack Depth 1 Upper 2 Lower 3 Horiz Vert Scale Vert Scale Cursors Scale Last Record Saved a Fr Es U o Records In File 0 Figure VI 5 The Measure Crack Depth screen ready for a test to be performed 3 This screen contains a single graph on which the two waveforms are plotted when a test is performed The three distances H H and the wave speed appear in boxes at the upper right of the screen After a test has been performed these numbers can be changed by selecting one of the boxes inserting a new number and pressing Enter Because a new file has been created and opened the graph is blank The green border around the Trigger Armed X button is flashing indicating that the system is ready to receive the signals from a test At this point the operator positions the transducers and the impact point as shown in figure VI 4 and performs the test 4 To see the screen as it would appear after a test for crack depth has been performed click Main Menu to return to the Main Menu screen Click Open Test Data File and open the existing file c ImpactDemo TestFile dat
13. Frequency kHz v 1 Clip Level 25 vel 25 5 Contact Time Change Curs Cut R Wave Cut Waveform 2V Scale 0 82 0 82 3H Seale ar Resolution Lasa Depth 315 mm 124 in 0 10 20 30 40 50 60 kHz 10070 50 40 30 20 15 10 Depth 4V 5cale 1 5 H Scale 60 This Record Other Records Save This Record Printing Number 7 Plate thickness Next Record Save as New Rerd Prepare to Print Help Name peo Prey Record Replace Record J Label Peak Main Menu Records in File 45 Z Show Record List Compare 1 To Destination File Filter Low Freg Astart A ES 6 ty maco micr GyLocal Disk C _ 68 rutorialz000 t Examine Te Guo asi pm Figure III 1 A test record on the Examine Test Data screen 5 In standard Windows protocol activating a command or selecting an object from the keyboard requires that the Alt key be pressed together with the underlined character in the caption 27 Fixed Labels labels in white lettering on a gray or blue background in the figures in this manual black lettering not in a box that describe adjacent text boxes or information labels Example in the upper right of the screen the Fixed Label Thickness mm identifies the contents of the adjacent text box Information Labels gray boxes with a narrow black border containing a variable associated with the test record displayed on the screen These boxes are not interactive
14. Impact Echo Test to enter the Begin Testing screen for the Plate With Overlay structure type Figure VIII 4 Begin Testing File c imagodemo TestFile da Str Name Type Plate Overla Date Time 9 8 97 4 29 05 PM Ds Overlay Thickness mm Overlay Wave 3000 Speed m s 6 Overlay Thick 11 1 Frequency kHz 7 Plate 180 Thickness mm 8 Plate Wave 1 Clip Level E Speed m s Contact Time Change Curs LutBeWaye CutWavim 2V Scale 0 7 3H Scale 2048 rea u Resolution J Depth 7 mm EJ n BE AV Scale 1 5 H Scalel 60 De By Cursor Last Record Saved Number Plate thickness Save as New Berd Help Trigger Armed X cared Records in File 45 Z Show Record List Compare Filter Low Freg Figure VIII 4 The Begin Testing screen for the Plate Overlay case Note that this screen differs from the screen for simple plates in that there are six key parameters in the upper right of the screen thickness and wave speed for both the concrete plate and the asphalt overlay an overlay thickness frequency and a composite thickness frequency The system is now ready to receive the results of a test on a concrete asphalt structure with the estimated thickness and wave speed values shown in these boxes 6 To view the results of tests on this concrete asphalt structure return to the Main Menu screen and open the file c ImpactDemo TestFile dat Records 31 35 in this file are from t
15. KHz and that the subdued peak in that vicinity of the spectrum represents the thickness frequency of the thin layer above the flaw Two methods are available to further clarify this point 1 using a smaller impactor and 2 removing low frequencies by digital filtering Using a Smaller Impactor to Amplify Hish Frequency Components 12 Open record 13 in the file e ImpactDemo TestFile The results on the screen are from an impact echo test at the same point as the previous record but with a smaller impactor There is now a significant peak at 19 KHz which confirms the suspicion that this is the thickness frequency of the thin layer The indicated thickness of this layer is 98mm a thickness likely to exhibit flexural vibrations Careful measurement of the contact time see step 9 above shows that it is about 14 us This indicates that a sphere of approximately 3mm diameter was used in this test producing a maximum useful frequency of 89 kHz Choosing the Right Impactor In routine testing it is usually best to start with a large impactor 10mm diameter or larger and use a smaller impactor if it is necessary to amplify or bring up features that are associated with frequencies of about 20 kHz and higher A 3mm diameter impactor the smallest practical size is 1 8 inch or about 1 5mm in diameter produces stress waves with useful frequencies up to almost 90 kHz and wavelengths as small as 0 04 m At the high end of this frequency range the
16. Keyboard press Alt 6 enter the observed frequency of 7 8 kHz and press Enter to accept this value into the computer memory 9 Inthe Recalculate Parameters box that appears Figure IV 6 select P Wave Speed as the parameter to calculate and click Calculate This screen disappears and the correct P wave speed of 3744 m s appears in the text box at the upper right of the screen displaying record 5 The blue thickness frequency line on the spectrum now coincides with the observed peak at 7 8 kHz O Thickness mm Choose paramater to O P Wave Speed recalculate Z Frequency The formula is P Wave Speed 2 x Thickness x Freq Cancel Figure IV 6 The Recalculate Parameters box The known thickness of 240mm and the observed frequency of 7 8 kHz have been used to calculate the correct wave speed of 3744 m s This is the apparent P wave speed in a plate The true P wave speed is this value divided by 0 96 or 3900 m s see Sansalone and Streett pp 51 52 Measuring Plate Thickness When the P Wave Speed is Known After the wave speed has been measured impact echo tests can be used to determine the thickness of concrete plates This is illustrated using impact echo records 6 to 8 from the file c ImpactDemo TestFile dat These records are from tests to determine the thickness of a concrete tunnel wall The wave speed was measured independently using two transducers on the surface and found to be 3996 m s this
17. Type and Parameter to Change screen The type of structure for which a parameter is to be changed is selected by clicking one of the option buttons in the box at the upper left This will enable make active the appropriate parameter option buttons in the box at the right In this example a Plate structure has been chosen so the only parameters that can be changed are Structure Name Wave Speed and Thickness or Dimension If Plate Overlay is selected as the structure type the additional parameter buttons Overlay Wave Speed and Overlay Thickness will also be enabled Selecting one of the buttons in the box on the right will cause the text box below to be enabled and an appropriate instruction on what to enter will appear to the left of the box In this example the instruction is Enter new wave speed and the number 3860 has been entered in the text box If the Change Parameter button is activated all records of the selected record type plate will have the designated parameter changed to the new value on the screen In this case the wave speed for all records in the current file in the selected range that are for plate structures will be changed to 3860 m s Wave speeds for records in that range in the current file for other types of structures columns beams hollow cylinders etc will remain unchanged Copying Sets of Records from the Current File to Another File The Destination File The two command buttons Open Destinati
18. absent altogether If there is debonding at the interface there will be complete P wave reflection from that level and no response from the concrete layer below A complete analysis of P wave propagation in such a structure can be found in Chapter 16 of Sansalone and Streett pp 159 161 Because the two layers have different acoustic properties the structure cannot be treated as a simple plate It is treated as a special class in impact echo testing as explained in the following paragraphs 4 Return to the Main Menu screen click Open Test Data File and create and open a new file c ImpactDemo Overlay After returning to the Main Menu screen click Describe Structure and from the Choose Structure Type option box that appears select Plate With Overlay and click OK The Plate With Overlay screen appears as shown in Figure VII 3 w Plate With Overlay yy Overlay Thickness Plate Thickness Structure Name 180 Plate Thickness mm Plate P Wave Speed m s Overlay Thickness 135 mm Overlay P Wave Speed m s Help Cancel 79 Figure VIII 3 The Plate With Overlay screen Enter Overlay as the structure name 180mm as the plate concrete thickness 4500 m s as the plate P wave speed 135mm as the overlay thickness and 3000 m s as the overlay P wave speed When these values have been entered click Accept Click Accept again on the Parameters for Data Acquisition screen 5 At the Main Menu screen click Begin
19. appearance of a sine wave and the spectrum exhibits a strong thickness frequency peak at 7 8 kHz 98 10 Click the 3 H Scale text box at the lower right of the waveform enter 2048 in place of 1024 and press the Enter key The spectrum is re plotted using 2048 data points a record length of about 4 milliseconds The resulting spectrum although it still has a strong peak near 7 8 kHz is more complicated This is the result of the effects of wave reflections from the side boundaries of the structure and the excitation of additional modes of vibration in a bounded structure In this case the addition of 2 milliseconds of time to the record length only serves to complicate the spectrum and to make it more difficult to interpret Place the active waveform cursor at 2048 us and click Cut Waveform This removes the last two milliseconds from the waveform and recalculates the spectrum The result is an effective record length of 2 milliseconds and the spectrum is generally similar to that obtained in step 3 above Open record 44 This shows the results of a test at the same point as in record 43 but with a sampling interval of 4 Us instead of 2 us The record length is 1024 x 4 Hs or about 4 milliseconds and the spectrum is similar to that obtained by using the same record length in step 4 above Place the active waveform cursor at 2048 ms and click Cut Waveform The resulting spectrum based on 512 data points at 4 ms a record
20. by the data acquisition system This setting is not used by ImpactDemo It is used only during testing with Impact E software system see Instrument Manual Par 7 The Power Saving Interval Min is designed to conserve battery power in the data acquisition system After a signal is received by the system it will automatically go into the Sleep Mode reducing power consumption by 90 if another signal is not received within the 1 Windows protocol provides that an underlined letter in the caption of a command button or other object can be used to activate or select that object by pressing the Alt key together with the key of the underlined letter 2 To check the screen size on your computer go to the Windows screen click Start Settings Control Panel and double click on the Display icon The screen size in pixels is the Desktop Area found under the Settings tab If possible choose 800x600 as the Desktop Area for your computer 15 specified interval The default value is 2 minutes To change this value enter another integer in the text box and press Return The new value is stored in a file when the system is shut down and will be used again when it is restarted Intervals between 2 and 5 minutes are recommended This setting is not used by ImpactDemo It is used only during testing with Impact E software system see Instrument Manual Par 8 12 Click OK to return to the Main Menu screen Open A Data File A new or existing te
21. calculate dimensions of the structure or depths of flaws Examine the fixed labels text boxes and command buttons within or beneath the spectrum Consider first the fixed labels the gray boxes in the upper right corner of the spectrum graph e Resolution is frequency resolution the spacing or discrete difference between frequency points on the spectrum in kHz It determines the precision with which thickness or depth can be determined The resolution is 1 nAt where n is the number of points in the waveform 1024 for this record and At is the sampling interval 2 us for this record The quantity nAt is the record length or the length of time over which the signal is recorded The resolution can be improved by increasing n or At but there are practical limits The maximum value for n is 2048 but 1024 works well for most testing Values of At between 2 and 4 us usually provide the best results Because the transient resonant vibrations induced by an impact decay very rapidly it is seldom useful to increase the record length beyond about 8 milliseconds 2048 At 4 us For more information about frequency resolution see Appendix A in this manual and Sansalone and Streett pp 64 65 32 e Depth is shown in the remaining three gray boxes in mm inches and per cent of full thickness In calculating the per cent thickness the full thickness is the value entered when testing begins or calculated later from test results It is disp
22. detect voids within plastic ducts The Response of Grouted and Ungrouted Tendon Ducts The frequency of multiple P wave reflections from a grout steel interface is given by f C 4d where Cp is the P wave speed and d is the depth from which the P waves are reflected The response from a group of strands in a fully grouted duct can be predicted based on the distance d to the nearest strands The frequency response obtained from steel post tensioning strands is usually slightly higher than predicted by this equation By comparison the frequency of P wave reflections from a concrete air interface are given by f 0 96C 2d The frequency of P wave reflections from a void will be approximately twice that of P wave reflections from steel tendons 86 The differences between P wave reflections from a concrete steel interface and those from a concrete air interface are explained in more detail in Chapter 17 of Sansalone and Streett pp 167 171 Besides being reflected from the steel strands in a grouted duct Pwaves are also refracted through the duct and reflected from the opposite surface of the plate Because the wave speeds in concrete and grout are similar and because the cross sectional area of the steel strands is relatively small the full thickness response of a structure containing a fully grouted duct is very close to that for the solid structure without a duct The schematic illustrations in Figure IX 1 summarize the three basic respo
23. etc wave speed s thickness or dimension s of the structure and description saved with the record The file name structure name and test date will appear at the top of the table This provides a condensed summary of a large number of test records in a file Activating the Send to Printer button will cause this information to be sent to a printer if one is connected to the computer Changing Parameters for Records Activating the Select Structure Type and Parameters button calls up a screen Figure III 4 that allows the user to select and change any of several parameters for a selected number of 38 records in the current file The records to be changed will be those within the range specified by numbers in the two text boxes above the button The default numbers are those of the first and last records in the file Select Structure Type and Parameter to Change Select both a structure type and a parameter to change Select Structure Type Select Parameter to Change Plate O Structure Name O Plate Overlay Wave Speed m s O Circular Column O Overlay Wave Speed m s O Square Beam Col O Thickness or Dimension mm O Rectangular Beam Col Overlay Thickness mm O Hollow Cylinder O 2nd Dimension Rect Beam Col mm C Vertical Scale Waveform Dimension in direction of impact for Rectangular Beam Column Enter new wave speed m s Change Parameter Cancel Exit Figure III 4 The Select Structure
24. experimenting with different features of the software until you are thoroughly familiar with the topics covered The next section of the tutorial describes the use of the Examine Test Data screen and explains how the user interacts with the software using command buttons text boxes moveable cursors and other interactive features 23 24 SECTION III Learning About the Software Introduction This section describes the principal features of the software through a series of steps that display and examine the important screens that appear during testing and evaluation of test data Most user interactions with Impact E or ImpactDemo can be accomplished with a mouse or from the keyboard In the instructions that follow the use of a mouse is assumed The phrase click X instructs the user to place the mouse pointer on object X and press the left mouse button Instructions for using the keyboard are shown in brackets as in Keyboard press Alt 0 Start the Program 1 On the Windows screen click Start and Programs to open the list of program groups Click the Impact Echo group and click ImpactDemo to start the program Keyboard press the Windows key not on all keyboards press P to open the list of program groups and use the up down arrow keys to select the program group named Impact Echo Press Enter to display the list of programs and use the up down arrow keys to select the program named ImpactDemo
25. factor B was unknown and the simpler equation was used See Sansalone and Streett pp 50 52 29 e Cursor to Cursor The number in the gray box in the upper right corner of the screen is the time in us separating the two cursors on the waveform It changes when either of the cursors is moved The use of this number is explained later 8 Examine the three text boxes white backgrounds within or below the waveform These are interactive boxes whose contents can be changed when the program is running e Horizontal Scale The integer in the 3 H Scale text box at the lower right of the waveform is the horizontal scale or the number of data points plotted in the waveform It is normally expressed as a power of 2 2048 1024 512 etc to facilitate the mathematical calculations that produce the spectrum When impactecho tests are performed with Impact E 2048 digitized data points are always recorded and stored as a permanent part of the test record The number in the 3 H Scale text box is the number of points plotted on the screen usually 1024 and used to calculate the spectrum Example Click the 3 H Scale text box Keyboard press Alt 3 The number in the box is selected and can be changed by entering another number Enter 2048 and press Enter The waveform is re drawn using 2048 data points and the spectrum is re calculated using this number e Vertical Scale The number in the text box labeled 2 V Scale defines th
26. is a void beneath a slab the low frequency flexural vibration can be strong as it is in record 19 or it can be weak or absent depending on the thickness of the slab and the size of the void The key to identifying the presence of a void beneath a slab on soil is the rate of decay of the signal that results from P wave reflections Record 21 is an example of a test over a void in which the flexural vibration is relatively weak and record 22 is an example of test in which it is almost nonexistent To verify that records 19 and 22 are similar in terms of the decay of reflected P waves open record 19 place the waveform cursor at about 10 kHz and click the Filter Low Freg button at the lower right of the screen This removes the low frequency components and the resulting waveform which is similar to a slowly decaying sine wave and looks very much like the waveform for record 22 5 Records 23 and 24 are from tests where the slab is in contact with soil This is indicated by the rapid decay in the P wave reflections in the waveform and by the relatively broad main peaks in the spectra To compare the peak in the spectrum of record 23 with that in record 22 for example open record 23 click the text box next to the Compare button enter 22 and press the Enter key This causes the spectrum from record 22 to be overlaid as a green dashed line on the spectrum of record 23 Note that the main peak in record 23 concrete on soil is broader than tha
27. is moved a similar cursor within this box a horizontal yellow line moves in tandem The number on this line is the depth within the plate corresponding to the frequency of the cursor 33 on the spectrum If the spectrum cursor is moved to a frequency below the thickness frequency the frequency associated with P wave reflections between the top and bottom of the plate the horizontal cursor disappears from the depth box the caption on the box changes to Cursor Out of Range and the depths indicated in the gray boxes in the upper right of the spectrum ppear in red The significance of peaks at frequencies below the thickness frequency is discussed in a later section The Control Panel at the Bottom of the Screen 15 Examine the controls at the bottom of the screen The rectangular control panel at the bottom of the screen contains a variety of labels text boxes and command buttons These will be explained starting from the left e The Number and Name text boxes These contain the number and name of the record displayed on the screen They can be used to call up records in the open file by number or name To display record 3 for example click the Number text box Keyboard press Alt b enter 5 and press Enter Record 5 is displayed A similar procedure is followed to display a record using its name Entering a number greater than the number of records in the file or a name that is not in the record list will result in an err
28. is the value of the apparent P wave speed in a plate 49 Impact echo tests were then carried out to measure the thickness using a grid of points on the wall A value of 400mm was used as a first estimate of thickness for each test 10 11 12 13 Use the Next Record and or Prev Record commands to open record 6 in the file c ImpactDemo TestFile dat An alternate method is to click the Number text box in the lower left enter 6 and press the Enter key The test results for record 6 record name J2 show that a wave speed of 3996 m s and a thickness of 400mm give an expected thickness frequency of 5 0 kHz These three numbers appear in the three text boxes labeled Thickness Wave Speed and 6 Thickness Frequency at the upper right of the screen The light blue vertical line on the spectrum marks the expected thickness frequency of 5 0 kHz The presence of a single dominant frequency of 6 3 kHz in this test record indicates that the true thickness is less than the estimated thickness of 400mm To calculate the true thickness using the known wave speed of 3996 m s and the observed thickness frequency of 6 3 kHz click the 6 Thickness Frequency text box in the upper right Keyboard press Alt 6 enter 6 3 and press Enter The Recalculate Parameters screen appears see Figure IV 6 Click the Thickness option box Keyboard press Alt T to select thickness as the parameter to recalculate and cli
29. of method 9 Impactor 10 117 Effect of size 59 Information Labels 28 J Label Peak command 35 Label Peak command 35 Layered Plate Parameter Calculations screen 80 Logo screen 14 Main Menu screen 15 Measure Crack Depth screen 66 Moveable Cursors 28 Name text box 33 Next Record command 34 Number of Samples 99 Number text box 33 Open Test Data File command 26 Open Test Data File screen 17 Overlay Asphalt on concrete 76 Concrete on concrete 74 Parameters for Data Acquisition See Data Acquisition Parameters Parameters for Data Acquisition screen 99 Plate With Overlay screen 77 Plates contact with soil 70 Solid Response 52 Two layers 74 Plates Thickness 48 Post tensioned structures 84 Power Saving 16 Prepare to Print command 35 Prev Record command 34 Print Multiple Records command 36 Printer Setup command 36 Printing 35 Adding text to graphs 36 Print Summaries of Records 37 Program Parameters 15 P Wave Speed 32 apparent P Wave speed in a plate 32 Direct Measurement 44 From tests on plate of known thickness 47 Relation to frequency 29 P waves 10 Rayleigh wave 10 Record Length 32 Record Name text box 34 Records Comparing Spectra 34 Displaying a List of Records 34 Hide Record List Command 34 Labeling Peaks on Spectrum 35 Records in File Label 34 Replacing Current Record 35 Save as New Rerd Command 34 Save Current Record as New Record 34 Saving to another file De
30. post tensioned structures 83 84 Section IX Voids in the Tendon Ducts of Post Tensioned Structures Introduction Serious problems can develop in post tensioned structures if voids are present within the grout that is used to fill the tendon ducts Voids in the grout can occur along the tendon trajectory due to blockages improper grouting procedures grout material problems and construction oversight Inadequate grouting may allow water to penetrate into the ducts causing corrosion of the steel tendons leading to failure of the structure The impact echo method can be used to detect voids in grouted tendon ducts in many but not all situations The method s applicability depends on the geometry of a structure and the locations and arrangement of tendon ducts Just as is the case for other types of flaws small voids in tendon ducts cannot be detected if the ratio of the size of the void to its depth beneath the surface is less than about 0 25 see Sansalone and Streett pp 84 85 In addition complicated arrangements of multiple dicts such as often occur in the flanges of concrete Fbeams can preclude detection of voids in some or all of the ducts In other cases portions of structures can be successfully tested and information can be gained that permits an engineer to draw conclusions about the condition of the grouting along the length of the duct The simplest case and the one described here is that of post tensioned ducts in a
31. press Help for instructions on opening an existing file Open record 16 in this file by selecting the Number box at the lower left Keyboard press Alt B entering 16 and pressing the Enter key Figure VI 6 appears on the screen showing the results of a test to determine the depth of a surface opening crack in a concrete slab A ineasg Crack Depth c imagodemo T estFile dat Structure Name ime 3711797 12 31 43 PM 4 07 0 0005v Cursor to Cursor E65 u s Distance Hl mm Distance H2 mm Distance H3 mm 4 PWave 3800 Speed m s 39 070 0003v Voltage Setting Z v Sampling Interval nn u s Calculate Crack Depth 1 Upper 2 Lower E 3 Horiz Vert Scale Vert Scale ___ Cursors Scale This Record Other Records Save This Record Printing Number Crack Depth Next Record Save as New Record Prepare to Print TEST12 Prey Record Replace Cur Record Main Menu Records In File 24 To Destination Eile Figure VI 6 Test record for crack depth measurement Record 16 in file c ImpactDemo testFile dat 69 Measuring the Elapsed Time Between P Wave Arrivals at the Two Transducers 5 Entries in the 1 Upper Vert Scale and 2 Lower Vert Scale text boxes beneath the graph can be adjusted to show more clearly the structure at the beginning of each signal The key point in the upper signal is the initial rise in voltage caused by the arrival of the direct P wave In the lower signal the key point is the initial
32. record 12 in the file e ImpactDemo TestFile dat and place the spectrum cursor at 12 2 KHz about midway between the 4 9 kHz peak and the subdued peak near 19 kHz Click 61 the Filter Low Fre q command in the lower right of the screen Click Yes in the message box that appears or press Enter The screen shown in Figure V 8 appears w Examine Test Data File cAimaaodemo TestFiledal Str Name fesiPlate Type _ Pe Date Time 5 13 94 6 51 48 AM 352 Jus Thickness mm Wave Speed m s 3744 6 Thickness 7 8 Frequency kHz 2 5 v 1 Clip Level 25 Contact Time 48 urs j mar Contact Time Change Curs Cut R Wave Cut Waveform 2 V Scale 1 25 3H Scale Resolution 749 Depth mm a 40 50 60 kHz 15 Depth 4V Scalofi 5 H Scale 60 This Record Other Records Save This Record Printing Number 12 Plate shallow flaw Large Next Record Save as New Rerd Prepare to Print Help impactor te Name Prey Record Replace Record J Label Peak Main Menu Records in File 40 Z Show Record List Compare fi To Destination File Restore Low Freg 1 20 Figure V 8 Record 12 with frequencies below 12 kHz removed by digital filtering Note the dramatic change in both the waveform and spectrum The large amplitude low frequency component has been removed leaving a 19 5 kHz component the thickness frequency of the thin layer as the dominant component This component persists for a
33. relatively long time in the waveform even though it was hardly visible beyond the second or third major peak before the low frequency component was removed Click Restore Low Freq to restore the low frequency component To gain an understanding of the effects d digital filtering click this button several times and observe the changes in both the waveform and spectrum 14 Open record 13 Place the spectrum cursor at about 12 2 kHz and click Filter Low Freq The low frequency component is removed from the signal leaving a peak at 19 kHz depth 98mm as the dominant peak The additional high frequency structure in the spectrum is the response to the high frequency stress waves produced by a 3mm diameter impactor 15 Open record 11 the result of a test over a relatively deep flaw where the low frequency peak represents a displaced thickness frequency step 6 above Place the spectrum cursor midway between the two main peaks 3 9 and 13 2 kHz and click Filter Low Freq to remove the lower frequency peak Removal of the strong low frequency component changes the waveform from a relatively complex structure to one that is dominated by a single frequency and has some resemblance to a decaying sine wave The single dominant peak in the 62 spectrum at 13 2 KHz is associated with multiple P wave reflections within the concrete layer above the crack The depth of the crack is calculated to be 142mm Unconsolidated Concrete A region of unconsol
34. shows the Begin Testing screen as it appears after a new file or an existing file that contains no records has been opened and testing is ready to begin The graphs for the waveform and spectrum the two large rectangles on the screen are blank the vertical lines within these rectangles are markers and cursors that will be explained later The large command button in the center of the lower part of the screen with the caption Trigger Armed X has green border that is flashing at intervals of about 0 5 seconds indicating that the data acquisition system is armed and ready to receive a test signal The Date Time in the upper right corner of the screen is entered automatically when an impact is made and stored as part of the test record At this point the transducer is placed in contact with the structure to be tested and an impact is made Figure II 12 The handle of the transducer is pressed down and the impactor a small steel sphere on the end of a spring rod is then tapped lightly against the surface The force of the impact is equivalent to dropping the sphere from a height of one or two meters The diameter of the sphere rather than the force of the impact is the most important factor in determining the character and behavior of the resulting stress waves The signal is only weakly dependent on the force of impact Figure I 12 Close up of impact echo test on aconcrete floor Figures II 13 and I 14 show tests being pe
35. stress waves begin to be scattered and reflected by the natural inhomogeneous regions in concrete such as small air inclusions mortar aggregate interfaces etc with the result that there is more noise in the waveform and spectrum This is apparent in record 13 a test performed with a 3mm diameter impacting sphere Removing Low Frequencies by Digital Filtering The dominant low frequency peaks in records 12 and 13 of the file c ImpactDemo TestFile dat resulting from flexural vibrations of the thin layer above a crack or delamination Their presence is a certain indication that a flaw is present but in the case of record 12 the low frequency peak so dominates the spectrum that the other important peak at about 19 kHz has a relatively small amplitude and is easily overlooked In order to amplify a higher frequency peak in the waveform it is sometimes useful to remove low frequency components by digital filtering This is accomplished through a numerical process applied to the waveform In the ImpactDemo program this method can be used to remove frequencies below a specific value the frequency marked by the position of the cursor up to a maximum of 20 KHz Digital filtering should be used with caution In general no useful information will be lost if the frequencies removed by filtering are below the thickness frequency However if frequencies above that level are removed important information about the structure can be lost 13 Open
36. that is the information they contain cannot be changed directly by the user Example the box labeled Date Time in the upper right corner of the screen contains the date and time that the test shown in this record was performed in the format month day year hour minute second AM or PM Text Boxes boxes with a white background that contain information about the test record displayed Text boxes are interactive The information they contain can be changed by the user To enter information into a text box select the box by clicking on it with the mouse Keyboard press the Alt key plus underlined letter or number in the fixed label next to the box After the desired text is entered into the box the Enter key must be pressed to transfer new text into the computer memory Example At the lower right of the upper graph the number 1024 in the box labeled 3 H Scale is the number of digitized data points used to plot the waveform the voltage time signal Moveable Cursors vertical lines on the waveform and spectrum that can be moved horizontally across the graphs using the mouse or keyboard There are two cursors on the waveform the upper graph and one on the spectrum The numbers on the waveform cursors give the time in microseconds after the trigger point while the number on the spectrum cursor gives the frequency of the cursor position in kiloHertz kHz Example Place the mouse pointer at about the midpoint of the wavef
37. the R wave if it has been cut set the active cursor to the point where the waveform cuts the zero voltage line the horizontal line at 330 us and click Cut Waveform Keyboard press Alt W The waveform to the right of the cursor is set to zero the caption on the button changes to Restore W and the spectrum is re calculated showing a broad dome of frequencies associated with the leading high amplitude portion of the waveform caused by the R wave When the amplitude of the R wave is large compared to the remainder of the waveform this dome of frequencies becomes a prominent part of the spectrum and can obscure the important peaks See Sansalone and Streett pages 58 59 These effects are subdued in this test record For further discussion of R wave removal see the analysis of record 25 in Section VII 31 Example 2 The combination of the Cut R Wave and Cut Waveform commands can be used to determine the frequency content of any portion of the waveform Click Restore W to restore the full waveform Place the active cursor at about 600 us and click Cut R Wave Place the cursor at about 1600 us and click Cut Waveform The resulting spectrum shows the frequency content of the waveform between 600 and 1600 us after the trigger point The Spectrum 13 Examine the spectrum The spectrum lower graph is a plot of amplitude vs frequency showing the distribution of frequencies in the waveform It is calculated from the waveform by a mathe
38. to 5000 m s When the wave speed is unknown 4000 m s is a good starting estimate The accuracy of the results is determined in part therefore by the accuracy with which Cp is known P wave speed can be determined using impact generated stress waves by two different methods 1 measuring the travel time of a P wave between two transducers a fixed distance apart on a concrete surface and 2 performing an impact echo test on a solid structure of known dimensions preferably a plate of known thickness In the second method the wave speed is calculated from the equation Cp 2fd where f is frequency observed in the test d is the characteristic dimension plate thickness for example and is a known shape factor that characterizes the geometry of the structure Both methods are described in this section See Sansalone and Streett Chapter 7 ASTM Standard for Measuring P Wave Speed and Plate Thickness Using Impact Echo An ASTM Standard Practice C 1383 98a entitled Standard Test Method for Measuring the P Wave Speed and the Thickness of Concrete Plates Using the Impact Echo Method was first published in October 1998 and updated in April 1999 A copy is included as Appendix B in this manual The methods outlined here are based on this standard Setup for Direct Measurement of Wave Speed Using Two Transducers In this method the travel time of a stress wave P wave or R wave between two transducers a fixed distance apart on a
39. to a spreadsheet place the spectrum cursor for each record at the peak in the spectrum that indicates the dominant response For solid structures this will be the thickness frequency When a delamination is present the resulting flexural vibrations in the delaminated layer will produce a strong peak at a lower frequency The entry for each element of the spreadsheet will be the thickness indicated by the position of the cursor and the known wave speed If the indicated thickness at any point is substantially greater than he nominal thickness of the structure being tested 1t is likely that the structure is delaminated at that point For example when testing an inch 203mm thick bridge deck with a wave speed of 4000 m s the nominal thickness frequency will be 9 8 kHz Therefore a dominant frequency close to this value is to be expected when testing a solid portion of the deck If the spectrum of a test record is characterized by a single dominant frequency of the order of 5 kHz or less corresponding to a thickness of 400 mm or higher the response is very likely to be that of a delaminated layer See Sansalone and Streett Chapter 10 p 99 On the File Utilities screen click Save to Excel Spreadsheet to open the Excel Spreadsheet screen On this screen enter the beginning and ending record numbers for the range of records you wish to save Wave speed records and records for 42 structure types other than plates will be ignored The pat
40. to the spectrum with a maximum at about 9 kHz Next click Restore W and then Cut R Wave Cutting or removing the R wave eliminates the broad dome of frequencies it contributes to the spectrum leaving a sharp narrow peak due to P wave reflections from the delamination A similar effect can be obtained by clipping the R wave With the R wave restored click the 1 Clip Level text box Keyboard press Alt 1 enter 0 06 as the clip level and press the Enter key The advantage of using the clip level rather than removing the R wave is that during testing the clip level can be set at an appropriate level and the R wave will be automatically clipped as each test is performed The full R wave can be recovered by raising the clip level to a value a higher value Open record 34 The dominant peak at 7 3 kHz is the result of P wave reflections from a depth of 239mm beneath the asphalt surface This was found to be the result of delamination at a lower layer of reinforcing steel Open Record 35 The dominant peak at 15 6 kHz corresponds to a depth of 96mm within the asphalt layer On further inspection this was found to be the result of a debonded asphalt patch This completes Section VIII of this tutorial covering impact echo testing of plates with two layers Repeat the action steps in this section until you are thoroughly familiar with the methods described The next section will describe testing to locate voids in the grouting of tendon ducts in
41. 24 4 kHz the expected frequency of the response from a void at a depth of about 80mm x Examine Test Data e x File e imagodemo TestFile dal Str Name TestPlate Type Plate Se 5 13 94 7 11 30 AM L Jus Thickness mm Wave Speed a dl 6 Thickness 7 8 Frequency kHz v 1 Clip Level Contact Time 26 u s Contact Time Curs Cut R Wave Cut Waveform 2V Scale 0 62 3H Seale ar Resolution Esa Depth 143 mm in Ba x 0 10 20 30 40 50 60 kHz ee 1 1010000100012 50 40 30 20 15 Depth 4V 5cale 1 5 H Scale s0 This Record Other Records Save This Record Printing Number 38 Tendon duct spec Next Record Save as New Rerd Prepare to Print Help Grouted duct witendons Name Prey Record Replace Record J Label Peak Main Menu Records in File 40 Z Show Record List Compare 1 To Destination File Filter Low Freg Figure IX 2 The results of a test over a fully grouted tendon duct record 38 88 This completes Section IX covering testing to locate voids in the grouting of tendon ducts in post tensioned structures Detailed explanations and illustrations of the impact echo response of structures other than concrete plates such as hollow cylinders pipes and tunnel and mine shaft liners circular square and rectangular beams and columns and masonry structures can be found in Sansalone and Streett The final section of the manual covers practical aspects of f
42. 3 It consists of eleven command buttons with captions that are largely self explanatory The actions produced by these command buttons are explained and illustrated in this and subsequent sections When this screen first appears 7 of the 11 command buttons are disabled cannot be activated ImpactDemo Software for Impact Echo Impact Echo Testing Cannot be Performed Main Menu nue Measme Depth of Surlace Opening t Ph Drock 5 File Utitiee Program Parameter Me Figure II 3 The Main Menu screen Click Help Keyboard press Alt H to open the Help System The Help System is in standard Windows format Any topic can be accessed from any help screen Most of the material in this manual is also covered in the internal Help System The Help System is accessed through Help command buttons Close the Help System by clicking X at the upper right of the Impact Echo help screen 14 Explore the Program Parameters Settings 5 10 11 Click the Program Parameters command button with the mouse Keyboard press the Alt key plus the key of the underlined letter P or press the Enter key when the Program Parameters button has the focus indicated by a red border Use the Tab key to shift the focus from one object to another The Program Parameters selection box appears Figure II 4 Click the Show screens in black and white option and click OK Keyboard press S then
43. 5 OPEN A DATA TEE cosa A e iii 16 CREATE A NEW DATA FILE IN A NEW OR EXISTING DIRECTOR Y SUBDIRECTORY eeeennenn 17 DESCRIBE THE STRUCTURE ui A saeta 18 BEGIN IMPACT ECHOTESTING 2 en tons 20 SECTION IH LEARNING ABOUT THE SOFTWARE oooonconcnonononcososonorrononorcnconononorcoconorornonononenconononoroconorcrnonono 25 INTRODUCTION cr A A laico START THE PROGRAM cuna dica OPEN AN EXISTING DATA FILE LEARNING THE FEATURES OF THE EXAMINE TEST DATA SCREEN THE CRITICAL PARAMETERS THICKNESS P W AVE SPEED AND FREQUENCY THE WEA VEFORM nta THE SPECTRUM ii S ESAS THE DEPTH BOX FOR PLATE STRUCTURES cccsccsssssessecsecsecssessesseseeseesscessesessesssseecesceecesseasaecaecaecsacsacsaeeaesaeenees THE CONTROL PANEL AT THE BOTTOM OF THE SCREEN ccsscsssssesscescescessscsecesccceseeseesecsecsecseesecsaeseeeaeenees PRINTING W AVEFORM AND SPECTRUM ccccsscsscsscsscsseesecsecsecsscssesseseeseesecescesessssseseecesseecnseeseeaecsecaecsaeseesaesaeeeesnees THE FILE UTILITIES SYSTEM a a iaa Saving Summary Information for RecCOTrAS ooniniiiininn nionmmmnm ens 38 Changing Parameters for Records iii tdi e A EE aia 38 Copying Sets of Records from the Current File to Another File The Destination File 39 Saving Records to ASCH Fl ade 40 Saving records to a Microsoft EXCEL Spreadsheet unnnesnsessenesesenenenesenenenesenenenesenenenenenennnonen 40 SECTION IV MEASURING P WAVE SPEED AND PLATE THICKNESS usus
44. AM Lag Jus Thickness mm iT Wave Speed m s 3780 6 Thickness Frequency kHz 25 1Clip Level 25 Tau Contact Time Change Curs Restore R Cut Waveform 2 Y Scale 3H Scale Resolution Depth Cer mm 32 in 50 60 kHz 20 Depth 4V 5cale 1 5 H Scale 60 This Record Other Records Save This Record Printing Number 26 Concrete slab with Next Record Save as New Rerd Prepare to Print Help idebonded topping Name Prey Record Replace Record J Label Peak Main Menu Records in File 40 Z Show Record List Compare 1 To Destination File Restore Low Freg Figure VIII 2 Record 26 with the low frequency components filtered out and the R wave removed 3 There next four records 27 through 30 are tests on the same 2 layered slab Record 28 is from a test where the layers are bonded it shows the effects of a relatively large transducer resonance while records 27 29 and 30 show debonding at the interface Concrete With Asphalt Overlay When concrete slabs such as bridge decks have asphalt overlays the problem is one of a two layered structure in which the upper layer the asphalt layer is acoustically softer than the lower layer of concrete The P wave speed is lower in asphalt than in concrete typically 2500 3000 m s in asphalt and 3800 4500 in concrete The coefficient of reflection at an asphalt concrete interface is about 0 20 This means that if the two layers are strongly b
45. D C 1383 98A MEASURING P WAVE SPEED AND PLATE THICKNESS USING THE IMPACT ECHO METHOD sssssssnonsessssnonsenenenensesensnensnssnsnensnnenenenensensnsnsnssnsnenennen 105 Section I The Impact Echo Method The Impact Echo Method Impact echo is a method for nondestructive testing of concrete and masonry structures based on the use of impact generated stress sound waves that propagate through concrete and masonry and are reflected by internal flaws and external surfaces It can be used to determine the location and extent of flaws such as cracks delaminations voids honeycombing and debonding in plain reinforced and post tensioned concrete structures including plates slabs pavements walls decks layered plates including concrete with asphalt overlays columns and beams round square rectangular and many I and T cross sections and hollow cylinders pipes tunnels mine shaft liners tanks The method can be used to locate voids in the grouted tendon ducts of many but not all types of post tensioned structures It can provide thickness measurements of concrete slabs with an accuracy better than three percent and it can locate voids in the subgrade directly beneath slabs and pavements When properly used the impact echo method has achieved unparalleled success in locating flaws and measuring thickness in highway pavements bridges buildings tunnels dams piers sea walls and many other types of structures An ASTM Standard Pr
46. G THE ELAPSED TIME BETWEEN P W AVE ARRIVALS AT THE TWO TRANSDUCERS un 70 CALCULATING THE CRACK DEP TIE u san Aa 70 SECTION VII PLATES IN CONTACT WITH SOILS susenecsecseesensensensensenseneenesnessersersensensensensensenesnesnersersersenne 71 INTRODUCTION ner Ra E ols Cece BR ads een onen VOIDS UNDER PLATES untaranen IDENTIFYING VOIDS UNDER A CONCRETE SLAB SECTION VIII PLATES CONSISTING OF TWO LAYERG sscssssssssssssscssensensenscssenessesnesnessenncnnencencenceneeneeneans 75 INTRODUCTION rios aussen ec aa a ative eee ised rennen 76 DEBONDING AT THE INTERFACE OF A CONCRETE OVERLAY ON A CONCRETE SLAB uenenenensnensenne 76 CONCRETE WITH ASPHALT OVERLAY auseinander en a Bons S 78 SECTION IX VOIDS IN THE TENDON DUCTS OF POST TENSIONED STRUCTURES ssscsscssssessessess 85 INTRODUCTION is 86 METAL DUETS u nase ke eu te tee boas AA Es 86 THE RESPONSE OF GROUTED AND UNGROUTED TENDON DUCTS unenenensnenssseneesensensensensensenseneenanee 86 SECTION X TIPS ON SIGNAL INTERPRETATION s20s0sususesesenensenensnsnsesensnsnsnssnsnenssnensnensnsensnsnsnsensnenensenen IL INTRODUCTION A eiii EVALUATING THE WAVEFORM ee THE IMPORTANCE OF THE R WAVE a ech Per Eu INTERPRETING THE SPECTRUM ie UNDERSTANDING THE EFFECTS OF THE SAMPLING INTERVAL AND NUMBER OF SAMPLES 98 APPENDIX A SETTING THE DATA ACQUISITION PARAMETERS sssssssssssssssssssssssssessssssesessesssnseess 101 APPENDIX B ASTM STANDAR
47. II 2 c may also be present as a result of flexural vibrations of the unsupported portion of the plate above the void Flexural vibrations occur because the unsupported section above the void is restrained at its edges where it contacts the soil The response is similar to that produced by an impact above a shallow delamination see p 55 However because the thickness of the slab is relatively large the amplitude of the flexural vibrations is smaller relative to the P wave thickness response 72 Ly Concrete Soil Amplitude 10 20 30 40 50 Frequency kHz Figure VII 1 The impact echo response of a concrete slab on soil subgrade a cross section b waveform and c spectrum Concrete toa Soil Me El a E lt 10 20 30 40 50 Frequency kHz Figure VII 2 The impact echo response obtained from a concrete slab at a location where a void exists in the soil subgrade a cross section b waveform and c spectrum Identifying Voids Under a Concrete Slab 1 Start the ImpactDemo program and open the existing file c ImpactDemo TestFile Records 19 24 in this file are from tests carried out to locate voids in the soil beneath a concrete warehouse floor with a nominal thickness of 152mm 6 inches 73 2 To open record 19 click the Number box in the lower left of the screen Keyboard press Alt B enter 19 and press the Enter key Note the similarity between this record an
48. Impact Echo Instruments LLC Ithaca New York USA Impact Echo User s Manual A Self Teaching Course and Reference for the Impact Echo Method and Software Version 2 2a October 2001 Impact Echo Instruments LLC P O Box 3871 Ithaca NY 14852 3871 Fax 607 257 3390 Email wbs impact echo com Copyright 2000 All rights reserved This document may not be copied or reproduced in any form The impact echo test instruments and software described in this manual are products of Impact Echo Instruments LLC P O Box 3871 Ithaca New York 14852 3871 Table of Contents THE IMPACTIECHO METHOD ives urn ais 7 USING THIS MANUAL AND IMPACT DEMO SOFTWARE cccssessesessessescssescuecasescneeccuecaesecnecaesecnessesecaeseeseceeeeaeenens 7 THE IMPACT ECHO BOOK ii dais HOW IMPACT ECHO WORKS represie neerstort Ease e rS eret STRESS W AVES nen E EE E Ba IMPACT ECHO FIELD TEST SYSTEM ccccescsssssesesseseeceecesescescaeseesceeeeceecaeeeens STORAGE AND RETRIEVAL OF TEST DATA SECTION II SETUP AND PREPARATION FOR TESTING susussssnsnsnsosenensnssnenenensesensnsnsnnenennsnsenenenensesenenene 1 2 GETTING STARTED WITH IMPACT DEMO erer serier o EES EEEE E Erare oE er TEER 13 INSTALLING IMPACT DEMO SOFTWARE mra a a e A E A E A R E 13 START THE PROGRAM cion e e r E E r PAE E S 13 EXPLORE THE PROGRAM PARAMETERS SETTINGS cccccscessssescssescescssescuecsesecaeeeesecaesecnecaesecaeseeaecaeeeaeeeeaeeaeess 1
49. O The Main Menu screen and all other screens appear with dark lines and letters on a white background Click Program Parameters again and remove the check from the Show screens in black and white box to restore color The default unit for the display of dimensions is millimeters mm Click the Show all dimensions in inches checkbox to cause all dimensions to be displayed in inches Keyboard press Alt d If the Screen for bad signals box is checked the software Program Parameters m Show screens in black and white Show all dimensions in inches MV Screen for bad signals will automatically check the incoming waveform to determine if it has the basic characteristics of a valid impact echo signal If it does not the computer will emit a loud buzzer signal and the waveform will be plotted as a red line These signals should be ignored The screening system can be turned on and off using a command button on the testing screen r Screen Size pixels C 640x480 800x600 C 1024x768 Comm Port Number 1 2 Power Saving Interval Min The Screen Size pixels option is used to make the important screens in the program fill the entire screen on the computer being used The default is 800x600 pixels 2 For computers with a maximum Desktop Area of 640x480 pixels choose this option Figure 1 4 The Program Parameters box The Comm Port Number designates the communications port to be used
50. Press Enter to start the program Open an Existing Data File 2 On the Main Menu screen click Open Test Data File Keyboard press Alt O On the Default File screen click Continue On the Open Test Data File screen the c drive is active and the Directory box is selected In the Directory box find the directory ImpactDemo use the scroll bar if necessary and double click ImpactDemo to display the data files it contains Keyboard press Alt r to select the Directory box use up down arrows to select ImpactDemo and press Enter to display the files In the File Name box double click the file TestFile dat to open it and click OK in the message box to return control to the Main Menu screen Keyboard press Alt N to select the File Name box use up down arrows to select file named TestFile dat and press Enter and Enter again to open it Press Enter to close the message box Learning the Features of the Examine Test Data screen 5 On the Main Menu screen click Examine Test Data Keyboard press Alt E to open the Examine Test Data screen Figure III 1 The first record test point in the file is displayed The principal features of this screen are the two graphs The upper graph is the waveform the voltage time graph that describes the surface displacements from this test The lower graph is the amplitude spectrum which identifies the principal frequencies in the waveform The interpretation of these grap
51. V 2 Comparison of the solid response left with the response from a region with a crack of wide lateral extent Flaws of Limited Lateral Extent When the lateral dimensions of a crack are comparable to its depth stress waves are both reflected from the crack and diffracted around it As a result multiple P wave reflections occur both within the layer above the crack and across the full thickness However the full thickness frequency is lower than that of the solid plate because of the reduced stiffness in the vicinity of the crack and because the Pwaves must travel a longer path around the crack to reach the bottom surface 5 Open record 11 in the file c ImpactDemo TestFile dat This shows the results of a test on the same 240mm thick plate described in the previous two records 6 Click the Compare text box at the bottom of the screen enter 9 and press Enter to overlay the solid response from record 9 The spectrum for record 11 is very different from that of the solid response again indicating the presence of a flaw Analysis and Interpretation There are two peaks of about equal amplitude in the spectrum one at 13 2 kHz and one at 3 9 kHz Use the mouse or the Mm and keys to move the cursor between these two peaks When the cursor is at the 13 2 kHz peak the indicated depth is 142mm This is the depth of the flaw When the cursor is at the 3 9 kHz peak the 55 depth indicated in the box on the graph in red letters is 479m
52. acks and voids in concrete plates Repeat the action steps in this section until you are thoroughly familiar with the topics covered Section VI Determining the Depth of Surface Opening Cracks Introduction A surface opening crack is any crack that is visible at the surface Such cracks can be perpendicular or inclined to the surface or curved as shown in FigureVI 1 a b c Figure VI 1 Surface opening cracks a perpendicular b inclined and c curved When stress waves are generated by an impact on the surface of concrete adjacent to a surface opening crack as shown in Figure VI 2 a the pattern of wave propagation differs significantly from the pattern in solid concrete Because the crack is a concrete air or in some cases a concrete water interface it reflects the stress waves propagating outward from the impact point as shown in Figure VI 2 b If the receiving transducer is placed on the opposite side of the crack the R wave traveling along the surface and the spherical P and S waves traveling within the concrete do not reach the transducer directly When the P wave reaches the bottom Figure VI 2 Reflection and diffraction of P waves from a surface opening crack a initial P wave front generated by impact b Pc is the P wave reflected from the crack c and d P P is the P wave diffracted from the bottom of the crack 66 edge of the crack it produces a diffracted P wave label
53. actice C 1383 98a entitled Standard Test Method for Measuring the P Wave Speed and the Thickness of Concrete Plates Using the Impact Echo Method was first published in October 1998 and updated in April 1999 A copy is included as Appendix B in this manual Impact echo is not a black box system that can perform blind tests on concrete and masonry structures and always tell what is inside The method is used most successfully to identify and quantify suspected problems within a structure in quality control applications such as measuring the thickness of concrete slabs and in preventive maintenance programs such as routine evaluation of bridge decks to detect delaminations In all of these situations impact echo testing has a focused objective such as locating cracks voids or delaminations measuring thickness or checking a post tensioned structure for voids in the grouted tendon ducts Successful field work requires both an understanding of the impact echo method and knowledge about the structure being tested Using This Manual and ImpactDemo Software This manual together with the software program ImpactDemo with its accompanying files of impact echo test data provide a comprehensive introduction to the impact echo method for nondestructive evaluation of concrete and masonry It also introduces the software Impact E used with Impact Echo test systems made by Impact Echo Instruments LLC ImpactDemo is a live active version of th
54. been named 2J the suggested name would be 2K This makes 1t convenient to use a logical sequence of record names during testing If the first record is named MyTestl for example subsequent records will automatically be named MyTest2 MyTest3 MyTest4 etc unless the suggested name is changed before the record is saved After entries have been made in the Record Name and Description text boxes click Accept to save the current record as a new record or click Cancel to return without saving the record Keyboard press Alt A or Alt C To replace the current record in the file with information on the screen click Replace Record Keyboard press Alt D A Yes No text box appears Click Yes to proceed or No to cancel Keyboard press Alt Y or Alt N This command is useful for saving as a permanent part of the record new values for parameters that have been changed on the screen such as horizontal scales P wave speed thickness etc The To Destination File button is used to copy the record displayed on the screen to another file called the destination file that has been opened using the File Utilities system described later in this section At this point the To Destination File button is disabled because a destination file has not been opened This command provides a means of creating a new data file by copying test records from one or more existing files Prepare to Print See Printing Waveform and Spectrum
55. bers for the range of records you wish to save The path and name of the Excel file where the data will be saved appears in a text box on the screen To save the information in a different file enter the path and file name in the text box and press Enter An Excel file should have the suffix xls Click Save Waveforms and Sprecta to Excel Spreadsheet The wave form and spectrum for each record in the specified range will be written as vertical arrays on a single Excel Sheet with a blank column between records For wave speed measurements two waveforms are written to the spreadsheet The time required to write the waveform and spectrum for one record to a spreadsheet is 5 10 seconds therefore writing a large number of records can take several minutes After a message appears indicating that the records have been entered on the spreadsheet another group of records can be saved to the same spreadsheet by entering a second set of beginning and ending record numbers in the text boxes at the top of the screen and clicking the button again When all desired records have been saved click Finish keyboard Alt f The spreadsheet will be saved to the file named in the text box and it can be opened by the Excel software on your computer Save results to Excel Spreadsheet for records N to from current file Data will be saved to a Microsoft Excel Spreadsheet with the path and file name shown in the text box below To save the results to a differe
56. cel Accept Figure II 10 The Plate screen 19 Click on the Structure Name text box to select it Keyboard press Alt N Enter Test Plate as the Structure Name and press Enter Enter 400 for the Thickness in mm and 3996 for the P Wave Speed m s pressing the Enter key after each entry 20 Click Accept on the Plate screen to return to the Main Menu screen Fig II 8 4 For impact echo purposes a plate is defined as any structure with two parallel faces for which the lateral dimensions are at least five times greater than the thickness 19 Begin Impact Echo Testing 21 On the Main Menu screen click Begin Impact Echo Test Keyboard press Alt B A message box appears with the message This is a demo version of Impact E software Testing cannot be performed Click OK or press Enter The Begin Testing screen appears Figure II 11 is Begin Testing File __cAlmpactE Testtiles ie4uqii dal Str Name Slab Type Plate Date Time 5 13 1994 6 56 37 AM Thickness mm Wave Speed se jo 6 Thickness Frequency kHz Resolution di Depth abel43 200 Ruz Depth Depth Indicated By Cursors ANSE 5H Scale 60 Cursor Last Record Saved Number f E Save as New Rerd Help E E Main Menu Z Show Record List Compare m Signal Screening On Filter Low Freg Figure II 11 The Begin Testing screen opened after a new file has been created 20 Figure I 11
57. ck Calculate Keyboard press Alt a The Recalculate Parameters screen disappears and the calculated thickness of 317mm appears in the Thickness text box in the upper right of the screen At the same time the light blue thickness frequency line on the spectrum is placed at the frequency of the dominant peak 6 3 kHz The thickness of the plate at point J2 has been found to be 317mm Open records 7 and 8 in the current file c ImpactDemo TestFile dat and determine the wall thickness These are records from the same plate described in record 6 For record 7 Record Name K2 the thickness frequency is 5 4 kHz and the correct thickness is 370mm For record 8 Record Name L3 the thickness frequency is 11 2 kHz and the correct thickness is 178mm 14 Click Main Menu to return to the Main Menu screen and click Exit to close the program This completes Section IV of this tutorial covering the use of impact echo to measure P wave speed and plate thickness Repeat the action steps in this section until you are thoroughly familiar with the topics covered The next section describes how impact echo is used to locate and identify cracks and voids in plate structures 50 51 Section V Detecting Cracks and Voids in Plates Introduction The impact of a steel sphere on a solid concrete plate causes multiple reflections of P waves between the top and bottom surfaces characterized by a resonant frequency that is a function of P wave speed a
58. concrete surface is measured If the fixed distance is L and the travel time is At the speed is distance divided by time Cp L At Figure IV 1 is a schematic diagram of the test set up and Figure IV 2 is a photograph of a wave speed measurement in progress Figure IV 1 Schematic representation of test set up for wave speed Figure IV 2 Wave speed measurement measurement Both transducers are controlled by the same clock in the data acquisition system making it possible to measure the elapsed time between the arrival of a stress wave at transducer 1 and its arrival at transducer 2 The numbered steps below show how this is done In setting up the instrument for a wave speed test transducer 1 the transducer nearest the impact is connected to Channel A on the data acquisition card and transducer 2 to Channel B The connections are made with BNC cable connectors 45 Start the ImpactDemo program and create and open a new file named WaveSpd dat on the c drive in directory ImpactDemo The complete path and file name will appear as c ImpactDemo WaveSpd dat for instructions on opening a new file see p 16 When control is returned to the Main Menu screen after the file is opened click Measure Wave Speed to open the Set up for Wave Speed Measurement screen Figure IV 3 w Set up for Wave Speed Measurement Structure Name Test Slab 1 Distance L 300 mm Dynamic Poisson s Ratio Help Cancel Begin Test Figu
59. ct is relatively small however when the amplitude of the R wave is very much larger than the remainder of the waveform its removal has a more dramatic effect Clipping the R wave Restore the R wave click the 1 Clip Leveltext box Keyboard press Alt 1 enter 0 2 and press Enter Portions of the R wave outside the range 0 2 volts are clipped or removed from the waveform and the vertical scale in the waveform is expanded to 0 2 volts The effect on the spectrum is almost the same as removing the R wave After the transducer resonance has been filtered out and the R wave clipped the screen should appear as in Figure VII 1 There is a strong sharp peak in the spectrum at 12 2 kHz with very little additional structure This is reflected in the waveform which is clearly dominated by a single frequency is Examine Test Data IN ES BEI Eg File dNimaga TestFiedal Str N wbs3 Type Pite _Date Time 11 13 1994 12 00 00 AM L170 Ju s Thickness mm i Wave Speed m s 3780 Nfl fy y 6 Thickness Frequency kHz 2 5 v Clip Level Contact Time us 2 Ju s Contact Time Change Curs Cut R Wave Cut Waveform 2 Y Scale 3H Scale Resolution 49 Depth mm in Em x 155 0 10 20 30 40 50 60 kHz Cursor Out of 100 70 50 40 30 20 Depth Range en 17 5 HiScalel 60 This Record Other Records Save This Record Printing Number Stab with bonded topping Next Record Save as New Rerd Prepare to Print
60. d the results shown in Figure VII 2 It is a test over a void in the soil beneath the floor The P wave reflections associated with the 14 2 kHz thickness frequency decay slowly indicating that the bottom of the slab is a concrete air interface The low frequency component in the waveform which appears as the 1 5 kHz peak in the spectrum is the result of flexural vibration in the slab above the void To remove the low frequency component place the spectrum cursor at about 10 kHz and click Filter Low Freq The resulting waveform is similar to a decaying sine wave with a frequency of 14 2 kHz 3 Open record 20 The spectrum pak at 14 6 kHz indicates that the slab is 134mm 5 3 inches thick at this point The rapid decay in the waveform indicates that stress wave energy was transmitted through a concrete soil interface and lost to the soil beneath the slab When this happens the waveform contains little useful information in the region beyond where it decays to a small fraction of its original amplitude A simple test can verify this Click on the waveform with the mouse to place the active cursor at about 1000 us and click the Cut Waveform button This removes sets equal to zero that portion of the waveform to the right of the cursor and calculates the spectrum using only that portion to the left Click the button several times and note the very small change in the spectrum when the right half of the waveform is deleted or added 4 When there
61. e software with all the capabilities of Impact E except real testing which is simulated Both Impact E and ImpactDemo have extensive on line Help Systems ImpactDemo software is Windows based It must be installed on a computer with a Microsoft Windows 95 or later e g Windows 98 or Windows 2000 operating system Instructions for installing the software are in the next section Users must have at least an elementary knowledge of the Windows operating system The user can interact with the software through a mouse or keyboard Throughout this manual it is assumed that a mouse is being used The instruction click X instructs the user to place the mouse pointer on object X and press the left mouse key Alternative instructions for using the keyboard are shown in brackets as in Keyboard press Alt O This manual and the ImpactDemo software program have been designed to be used together The manual is divided into sections containing explanatory text and numbered action steps to be performed using the software When the software program is installed a data file of impact echo tests on real structures is loaded into the computer memory and used to explain how tests are performed and how the results are interpreted Follow the action steps in sequence to obtain a comprehensive introduction to the impact echo method and the software Study each section until the methods and principles it covers are thoroughly understood before proce
62. e vertical range of the graph on which the waveform is plotted It is expressed in volts and is automatically set so that the maximum absolute voltage in the signal reaches the top or bottom of the graph The vertical scale cannot be changed by the user e Clip Level The number in the text box labeled 1 Clip Level in the lower left of the graph is a voltage value that is used to clip the waveform that is to remove voltages with an absolute value greater than this number The default setting is 2 volts Example Click the 1 Clip Level text box Keyboard press Alt 1 enter 0 2 as the clip level value and press Enter Voltages above 0 2 and below 0 2 are clipped removed from the waveform and the vertical scale is reset to 0 2v The effect is similar to cutting or removing the entire R wave See examples in Sectons VII and VIII The advantage of clipping is that in testing situations where the R wave is very large relative to the remainder of the signal the clip level can be set at an appropriate level and the R waves will be automatically clipped partially removed in each test See Sansalone amp Streett pp 58 59 9 Learn to measure the contact time The Contact Time command is used to facilitate measurement of the width of the R wave the region of negative voltage at the beginning of the waveform The R wave or Rayleigh Wave is a surface wave that propagates outward from the impact point like a ripple in a
63. ed P P in c and d that travels outward along a cylindrical front centered on the bottom edge of the crack This is the first stress wave to reach the transducer In determining the depth of a surface opening crack two transducers are used as shown in Figure VI 3 The transducer on the same side of the crack as the impact is used to determine the time of the impact and the second transducer measures the time required for the stress wave to travel from point A the impact point to B the crack tip to C the transducer The two transducers and the impact point are placed on a straight line perpendicular to the line of the crack on the surface If the wave speed and the distances H H and H are known and the arrival times of the stress waves at the two transducers are measured the depth of the crack can be calculated This method called a time of flight technique is illustrated schematically in Figure VI 3 and is explained in detail in Chapter 12 of Sansalone and Streett pp 123 37 It can be used to obtain good approximations of the depths measured perpendicular to the surface of cracks that are perpendicular to the surface inclined or curved Figure VI 3 Measuring the depth of a surface opening crack a schematic of experimental test setup and b sample waveforms The two waveforms labeled 1 and 2 in Figure VI 3 b are the signals from transducers 1 and 2 in the drawing at the left The arrival of the direct P wave
64. ed X m Main Menu Records In File 24 Figure IV 4 The Wave Speed Measurement Screen as it appears when a new or empty file has been opened To see the screen as it would appear after a wave speed measurement has been performed click Main Menu to return to the Main Menu screen Click Open Test Data File and open the existing file c ImpactDemo TestFile dat see Section IM or click on Help for instructions on opening an existing file Open record 2 in this file Figure IV 5 It shows results of a wave speed measurement test on a concrete slab is Wave Speed Measurement File d Impact Echo DatafFiles TestFile dat i MEEI ma ture Name Date Time 472972000 1 30 29 PM Chrsbr to Cursor 70 5 u s Distance LL ae WaveSpd From R wave Sampling Interval 057 u s WaveSpd From P Wave 1 Upper 2 Lower a H 3 Horiz Vert Scale Vert Scale i al E Scale Voltage Setting DI v This Record Other Records Save This Record Printing ee Next Record Save as New Record Prepare to Print Help Prey Record Replace Cur Record Main Menu Records In File 4 To Destination File Figure IV 5 Test record for wave speed measurement Second record in file c ImpactDemo TestFile dat The upper waveform on the graph is from the transducer nearest to the impact point The distance L separating the two transducers is 300mm in this case shown in upper ri
65. ed to 7 8 kHz Analysis and Interpretation To understand the relationship between the 7 8 KHz peak in the spectrum and the periodic structure of the waveform use the Kk and keys to place the active waveform cursor on the first positive peak after the R wave The label on the cursor reads 258 the elapsed time in us after the trigger point Double click on the active cursor or click Change Cursor Keyboard press Alt C to make the second cursor active and use the Kk and keys to move that cursor to the second positive peak at 388 us after the trigger point The time separation between the cursors 130 us as shown in the box at the upper right of the graph is the approximate period of the waveform The frequency is the reciprocal of the period f 1 0 000130 7692 Hz or 7 7 kHz This is close to the frequency of the main peak in the spectrum 7 8 kHz Try measuring the period using other adjacent peaks in the waveform The resulting frequencies will vary slightly but the average will be about 7 8 kHz Flaws of Wide Lateral Extent 3 Open record 10 in the current data file c ImpactDemo TestFile dat This is the result of a test on the same plate described in record 9 in a region where cracks are suspected 4 To compare the spectrum of this record with that of the solid structure in record 9 click the text box at the right of the Compare button at the bottom of the screen Keyboard press the tab key repeatedly until
66. eding to the next section The Impact Echo Book Detailed information about the physics and mathematics of the impact echo method can be found in the book Impact Echo Nondestructive Evaluation of Concrete and Masonry by Mary J Sansalone and William B Streett 1997 Bullbrier Press Ithaca NY Address inquiries to Bullbrier Press R R 1 Box 332 Jersey Shore PA 17740 telephone and FAX 570 769 7345 A copy of the book is included with each impact echo instrument It is a useful companion to this manual and ImpactDemo software Elsewhere in this manual the book is referred to as Sansalone and Streett How Impact Echo Works Impact echo is based on the use of transient stress waves generated by elastic impact A diagram of the method is shown in Figure L1 A short duration mechanical impact produced by tapping a small steel sphere agamst a concrete or masonry surface generates low frequency 70 kHz or less stress waves that propagate into the structure and are reflected by flaws and or external surfaces Surface displacements caused by the arrival of reflected waves at the impact surface are recorded by a transducer which produces an analog signal of voltage vs time called a waveform This signal describes transient local vibrations caused by multiple reflections of stress waves within the structure The dominant frequencies in these vibrations are related to the depths from which stress waves are reflected within the structure The anal
67. een Keyboard press the Windows key not on all keyboards and press R Enter a setup on the line labeled Open in the Run box that appears and click OK Keyboard press Enter 2 Follow the instructions on the screen The software will be er installed in a directory named ImpactDemo The installation will Fr create a program group called Impact Echo containing an icon GEO for the ImpactDemo program Figure II 1 The icon will also be ImpactDemo installed on the computer s Desktop screen Figure II 1 The ImpactDemo icon Start the Program 3 Double click the ImpactDemo icon on the Desktop screen or on the Windows screen click Start and Programs Click the Impact Echo group and click ImpactDemo to start the program Keyboard press the Windows key not on all keyboards press P to open the list of program groups and use the up down arrow keys to select Impact Echo Press Enter select ImpactDemo and press Enter to start the program When the program is started the Logo screen Figure II 2 appears for several seconds The box at the bottom contains the Version Number of the software and its release date ImpactDemo_ gt Software for Impact Echo Copyright 2000 AU Rights Reserved Impact Echo Instruments LLC Ithaca New York Version 2 1 Release Date 20 April 2000 Figure 11 2 The Logo screen 13 After the Logo screen disappears the Main Menu screen appears Figure II
68. en the wave arrivals at the two transducers is 71 us which appears in the box at the upper right of the graph Calculating the Crack Depth 6 With the two cursors positioned as described in the preceding paragraph click the Calculate Crack Depth button at the right of the graph Keyboard press Alt P A message box appears on the screen indicating that the crack depth is 155mm or 6 1 inches 7 The next two records in the open file records 17 and 18 are also crack depth measurements In record 17 the upper and lower cursors are positioned at 72 0 0 0011v and 95 0 0 0025y respectively and the crack depth is found to be 31mm or 1 2 inches In record 18 the cursors are positioned at 52 0 0 0042v and 107 0 0 0024v and the crack depth is 121mm or 4 7 inches This completes Section VI of this tutorial covering determination of the depth of surface opening cracks Repeat the action steps in this section until you are thoroughly familiar with the methods described The next section will describe the testing of concrete plates in contact with soil 70 Section VII Plates in Contact With Soils Introduction A common problem in the evaluation of plate structures is detecting voids under or behind concrete slabs in contact with soils The absence or loss of subgrade support for a horizontal slab can result in damage when it is subjected to heavy loads such as equipment on warehouse floors trucks on highway pavements or airplanes
69. encies of interest At the same time the record length nAt where n is the number of samples must be sufficiently long to identify the important frequencies in the response and to provide a reasonable level of frequency resolution but short enough to avoid multiple reflections from the side boundaries of the structure It often appears desirable to maximize the frequency resolution by setting n to the maximum level of 2048 and At to 4 us or higher Although this provides a frequency resolution four times better than the recommended starting values of 1024 for n and 2 us for At a very real risk exists that unless the test is being performed on a thick structure or a very large plate a record Ength of 8192 micro seconds 2048 x 4 will allow multiple reflections of both R and P waves from the side boundaries to become part of the waveform adding frequency peaks that complicate the response A time interval between samples of 1 microsecond is equivalent to a sampling rate of 1 MegaHertz MHz or one million cycles per second The minimum time interval is 0 5 micro seconds equivalent to a sampling rate of 2 MHz The default value set by the Impact E software for impact echo testing is 2 microseconds a value suitable for most testing For wave speed measurements the time between samples is automatically set to 0 5 microseconds and the system samples simultaneously on two channels Frequency Resolution When the waveform is processed by means of a fas
70. equencies from the waveform Its proper use is explained later in this manual Printing Waveform and Spectrum If your computer has a printer connected to the printer port copies of the waveform and spectrum can be printed from records in the open file 16 Click Prepare to Print to open the Prepare Graphs for Printing Screen Figure III 2 This screen permits the user to preview the waveform and spectrum add text to the graphs and send images to a printer connected to the computer In addition to the Help button explained earlier there are six command buttons and one interactive text box on the screen e Click Printer Setup Keyboard press Alt N to open a screen that provides several options for adjusting the size and character of the printed graphs These options are self explanatory w Prepare Graphs for Printing Preview Graphs Add Text Printing Waveform Enter Text and drag Printer Setup to graph Spectrum Print This Record Print Multiple Records Record Number 1 Erom N Ta pa Figure III 2 The Prepare Graphs for Printing screen 36 e Click Waveform Keyboard press Alt W to preview the waveform before printing e Click Spectrum Keyboard press Alt S to preview the spectrum before printing e After one or both graphs have been previewed the Print This Record button is enabled Click Print This Record Keyboard press Alt P to print the graphs that have been previewed
71. esponse obtained from a concrete bridge deck with an asphalt overlay a waveform with clipped R wave and b spectrum 97 Interpreting the spectrum Spectra are valid only if waveforms are valid Repeated tests taken in the same location with the same test set up should produce similar patterns in spectra If the spectrum of a test cannot be reproduced the test is probably invalid The bllowing sequence of steps and questions are suggested as an approach to analyzing spectra a b c d e Examine the overall pattern in the spectrum Is it a typical solid response for the type of structure being tested If not does it include a peak or set of peaks consistent with the presence of a flaw If the fundamental solid response frequency is shifted downward but a distinct higher frequency response is not observed try repeating the test using a smaller impactor This will introduce stress waves with more energy in high frequencies which will help amplify any high frequency response produced by a flaw at a shallow depth If one or more frequency peaks are present determine their frequencies and the corresponding depths Are the depths consistent with any known or expected flaws in this structure If a secondary peak or peaks are present but the fundamental solid response is unchanged check to see if the secondary peak is consistent with P wave reflections from a concrete steel interface such as a reinforcing bar or tendo
72. ests on a 180mm thick concrete bridge deck with an asphalt overlay having a nominal 80 thickness of 135mm Wave speeds in the concrete and asphalt were found to be 4500 and 3000 m s respectively Open record 31 Figure VIII 5 Analysis and Interpretation On this screen the spectrum has two thickness lines a blue line at 5 9 kHz that marks the composite thickness frequency associated with reflections from the bottom of the concrete at a depth of 315mm beneath the asphalt surface and a green line at 11 2 kHz that marks the expected frequency of reflections from the asphalt concrete interface at a nominal depth of 135mm Note also that the cross section box at the lower right of the screen contains two layers with the darker layer at the top representing the asphalt and the layer below the concrete In this record the main peak in the spectrum lies at 6 3 kHz corresponding to a depth of 286mm or 29mm above the nominal depth of the bottom of the concrete layer This could be the result of a crack or delamination at that depth but it could also mean that the asphalt and or concrete have less than the nominal thickness at this point Because there are two wave speeds and two thicknesses that are variables in this case finding the correct combination of wave speed thickness and frequency is not as straightforward as in the case of a simple plate w Examine Test Data File c imagodemo TestFile dadl Str Name _ Type Plate Overlai Da
73. frequency slowly decaying component is the hallmark of flexural vibrations in a thin or delaminated layer It is apparent from the shape of the beginning of the waveform that a smaller amplitude higher frequency component is present indicated by the closely spaced peaks superimposed on the low frequency component The spectrum is dominated by large peak at 4 9 kHz although there is a hint of a smaller peak around 20 kHz 8 Click the Compare text box at the bottom of the screen enter 9 and press Enter to overlay the solid response on the spectrum The current record exhibits a strong low frequency peak below the thickness frequency a certain indication of a flaw The absence of strong peaks in the spectrum above that frequency suggests that a smaller impactor might be needed to generate stress waves covering a broader range of frequencies Using the R Wave to Determine Contact Time and Frequency Content 9 The maximum useful frequency in an impact echo test can be determined from the contact time the length of time the sphere is in contact with the concrete surface during an impact With record 12 as the open record click the Contact Time command button beneath the waveform to expand the leading part of the waveform for measurement of the contact time The screen shown in Figure V 5 appears 58 File c magodemo TestFile dad Str Name TestPlate Type Pie Date Time 1 Clip Level is Contact Time 1 Make certain cu
74. ght of screen Each waveform has one cursor labeled with two numbers separated by a slash The 47 first number is the time in us from the trigger point and the second is the voltage of the transducer signal at that point Note that the vertical scales for both graphs have been set to 0 02 volts for the purpose of identifying the first arrival of a signal the passage of the P wave at each of the transducers The structure in the leading part of each waveform is electrical noise in the system Measuring the Travel Time of a P Wave Between Two Transducers 3 The first step is to position the cursor to mark the P wave arrival on the upper waveform The arrival of the direct P wave at a transducer is the point at which the voltage first rises above the horizontal line the zero level of voltage To position the cursor on the upper graph use the Ll key to move the cursor several digital points to the left and then move it slowly back to the right using the key As the cursor traverses the flat portion of the waveform the voltage oscillates between 0 0002 and 0 0001 volts this is low level noise in the system until the rising portion of the waveform is reached The first point at which the voltage increases significantly is 87 5 0 0001 v indicating that at 87 5 microseconds after the trigger point the voltage suddenly increased to 0 0001v Leave the cursor at this point which marks the P wave arrival at the fi
75. h and name of the Excel file where the data will be saved appears in a text box on the screen To save the information in a different file enter the path and file name in the text box and press Enter An Excel file should have the suffix xIs Click Save Thickness Delamination Data from 2 D Grid to Excel Spreadsheet A message will appear stating that the 2D array has been created on the spreadsheet this process may take 10 to 20 seconds for a large array To save a different set of records as part of the same array enter the beginning and ending record numbers and click the button again When all desired records have been saved click Finish The spreadsheet will be saved in the file named in the text box and can be opened by the Excel software on your computer This completes Section III of this tutorial covering the Examine Test Data screen and controls available to the user to examine test records and print copies of the waveform and spectrum Repeat the action steps in this section until you are thoroughly familiar with the topics covered The next section of the ImpactDemo program describes how impact echo is used to measure P wave speed and plate thickness 43 Section IV Measuring P Wave Speed and Plate Thickness Introduction In impact echo testing the equations used to calculate dimensions and flaw depths express these quantities as linear functions of P wave speed Cp Wave speeds in concrete typically vary from about 3500
76. hape factor that depends on the geometry of the structure being tested d is a characteristic dimension Cp is the P wave speed and fis frequency For plate structures B 0 96 and the key frequency called the thickness frequency is the vibration frequency induced by multiple P wave reflections between the top and bottom surfaces If the thickness d of a solid concrete plate is known and 7 In the early work on impact echo the existence of the shape factor B was unknown and the quantity 0 96Cp was used as the wave speed for plates It is sometimes called the apparent P wave speed in a plate or Cp plate 48 the thickness frequency f is measured in an impact echo test the wave speed Cp can be calculated The following steps illustrate this procedure 7 Open record 5 in the file c ImpactDemo TestFile dat It shows the results of an impact echo test on a slab that is known to be 240mm thick A value of 4000 m s has been used as a first estimate of the wave speed If this were the correct wave speed the thickness frequency would be 8 3 KHz shown in the text box in upper right and marked by the position of vertical blue line on spectrum However the results of the impact echo test show that the measured thickness frequency is 7 8 KHz the frequency of the main peak in the spectrum the lower graph 8 To calculate the correct wave speed click the 6 Thickness Frequency text box in the upper right of the screen
77. he data acquisition system is bipolar which means that 1t responds to both positive and negative voltages The possible settings for the voltage range from 0 1 to 5 0 volts With current technology each data point is recorded as a 14 bit number It follows that the voltage resolution the uncertainty introduced by the digitizing process is V 2 or one part in 16384 of the absolute value of V For the ranges 0 5 2 0 and 5 volts for example the uncertainties in the voltage are 0 00003 0 0001 and 0 0003 volts respectively or 0 006 of the maximum 102 voltage In normal impact echo testing it is rare for the output signal from the transducer to exceed 2 volts therefore a voltage range of 2 0 volts is the default value automatically set by the software The waveform for an impact echo test on a 250mm thick concrete plate with the voltage range set at 2 5 volts is shown in Figure A 2 The zero point in voltage is at the center of the vertical axis and the extremes of 2 5 and 2 5 volts are at the top and bottom o gt lt lt amp n S o gt Time ms Figure A 2 Waveform from impact echo test on a solid plate 250mm thick In this example the maximum voltage in the waveform is about 1 5 volts and the uncertainty is 0 0001 volts or about 0 007 percent The maximum voltage in impact echo signals is usually less than 2 volts and often less than 1 volt so these values are suitable for routine testing
78. he ducts were determined using a magnetic cover meter The approximate depth of the duct 87 beneath the surface was about 80mm or a little over 3 inches The wave speed was measured from impact echo tests on a region of known thickness and found to be 3900 m s Using the equations given above the thickness frequency of the solid plate was calculated to be 7 8 kHz the frequency associated with P wave reflections from a fully grouted duct was estimated to be 12 2 kHz and the frequency for a void was estimated to be 24 4 kHz The latter two estimates are approximate because the depth of a void or of the tendons varies according to their location within the duct 1 Open the file c ImpactDemo TestFile dat see p 26 and open record 37 This record from a test on a solid portion of the web is typical of the response of a solid plate and exhibits a thickness frequency peak at the expected value of 7 8 kHz 2 Open record 38 This record is from a test over a fully grouted tendon duct Figure IX 2 The peak at 13 7kHz is due to P wave reflections from the grout steel tendon interface Its frequency is slightly higher than the estimated value of 12 2 kHz which is common for P wave reflections from multiple steel tendons 3 Open record 39 This record is from a test over a duct that is ungrouted The peak associated with the thickness frequency is reduced to 3 9 kHz well below the solid thickness frequency of 7 8 kHz and there is a peak at
79. hs is explained in detail in later sections See also Chapters 4 and 5 in Sansalone and Streett The objects on this screen command buttons text boxes labels etc are described in the following paragraphs In addition to the two graphs there are five classes of objects on this screen that display information and or allow the user to interact with the program 26 e Command Buttons gray rectangles with 3D borders These are standard Windows command buttons At the base of the upper graph are command buttons labeled Contact Time Change Curs Cut R Wave and Cut Waveform In the large rectangle at the bottom of the screen are command buttons labeled Next Record Save as New Record Prepare to Print Help etc A command is activated by clicking the button with the mouse or from the keyboard by pressing the Alt key together with the key of the underlined character in the caption on the button gt Example At the bottom of the screen is a command button labeled Z Show Record List Click this button with the mouse Keyboard press Alt Z A list of all records in the open file appears on the screen and the caption on the button changes to Z Hide Record List Click the button again to hide the list The functions of other command buttons are explained later in the tutorial ii ETE Test Data E E x File c imagodemo TestFile dal Str Name ltunnel Type Plate o ee 9 8 1997 4 29 05 PM Thickness mm Wave Speed ll 6 Thickness
80. icking Cut R Wave Keyboard press Alt R The result is shown in Figure V 9 In this figure the horizontal scale on the spectrum has been changed to 30 kHz to show more clearly the structure in the region below that level This change is made by clicking on the 5 H Scale text box Keyboard press Alt 5 entering 30 and pressing the Enter key The three peaks between 14 6 kHz and 21 0 kHz have been marked by placing the cursor on each one and activating the J Label Peak command Keyboard press Alt J 63 w Examine Test Data ragode estFile da Str Name TesPlate Type Plate Date Time 3494 7 01 28 AM Thickness mm Wave Speed Had 6 Thickness 78 Frequency kHz 12 5 v 1 Clip Level Contact Time Change Curs Restore R Cut Waveform 2V Scale 1 25 3H Scale Resolution Depth 400 250 30 kHz Cursor Out of Depth Range 30 ee 5 H Scalej30 Other Records Save This Record Printing This Record Number Test plate 2 Next Record Ss New Rerd P to Print H ps ndice ext Recon Save as New Rer repare to Pr Help Name Prey Record Replace Record J Label Peak Main Menu Records in File 40 Z Show Record List Compare To Destination kile Filter Low Freg Figure V 9 Record 15 with the R wave removed and the horizontal scale on the spectrum changed from 60 to 30 kHz This completes Section V of this tutorial covering the use of impact echo to identify and locate cr
81. idated concrete typically consists of large numbers of small interconnected voids commonly referred to as honeycombing Such areas include many small concrete air interfaces over a range of depths and often they do not have a well defined external boundary The interaction of P waves with regions of honeycombing is discussed in Chapter 11 of Sansalone and Streett pp 115 122 The usual response of a honeycombed region to an impact echo test includes a displaced thickness frequency that is a strong peak at a frequency below that of the solid plate and one or more additional peaks representing P wave reflections from a range of depths within the unconsolidated region 16 Open record 15 in the file e ImpactDemo TestFile dat This record is from a test on a 240mm thick plate with a region of honeycombing at depths from about 100mm to 135mm Analysis and Interpretation The peak at 4 9 kHz is the displaced thickness frequency peak displaced downward from the solid thickness frequency of 7 8 kHz The peaks at 21 0 17 6 and 14 6 kHz are the result of P wave reflections from depths between about 93mm and 133mm Some of the small peaks at higher frequencies are a result of the very large R wave produced by a strong impact close to the transducer In this case it is useful to remove the R wave before the spectrum is calculated This is done by placing the active waveform cursor at the right of the R wave at about 220 us and cl
82. ield testing and interpretation of impact echo test results 89 90 Section X Tips on Signal Interpretation Introduction To use the impact echo method effectively one must learn to interpret and classify graphical patterns in the waveforms and spectra This section provides advice and insights on this topic Evaluating the waveform The waveform is the raw response of an impact echo test It contains all of the information provided by the test but in a form that often makes it difficult to extract key features of the response such as the transient resonant frequencies associated with multiple P wave reflections Nevertheless it is the features of the waveform that allow one to determine when a recorded test is valid A common mistake by new users of impact echo is to ignore the waveform altogether and to base an interpretation solely on the spectrum The waveform always contains useful information and it should be examined as part of the interpretation of each test Valid and Invalid Waveforms It is very important to recognize when recorded data are invalid that is when something has gone wrong with the test and the data should be discarded A valid waveform indicating a successful test will consist of horizontal or zero voltage segment at the beginning followed by a distinct Rwave except in the case of surface opening cracks and a periodic displacement pattern caused by multiple reflections of stress waves Many examples
83. il the box is highlighted enter the number 5 and click Compare or press Enter The spectrum for record 5 is overlaid on that of the open record as a green dashed line and the caption on the button changes to Remove This is useful for comparing records during routine testing Once the response of the solid structure is known it can be used as a comparison record for subsequent tests Significant 34 departures from the solid response are an indication of a flaw More about this later Click Remove Keyboard press Alt 0 to remove the comparison spectrum Click Save as New Rerd Keyboard press Alt S to open the Save to Data File screen used for adding records to the open data file This is most often used during testing but 1t can also be used when examining test data The information at the top of the Save to Data File screen informs the user of the path and name of the open data file and the number that will be assigned to this record 1f it is saved to the file Saved records are always added to the end of the file and numbered sequentially To select a text box or command button press Tab until the desired object is highlighted A suggested record name appears in the Record Name text box This suggested name is produced by incrementing the letter or number at the end of the name of the current record In this case the current record is named J2 and the suggested name for the record to be saved is J3 If the open record had
84. iption of files in terms of paths drives directories and file na filing system For example consider the storage of a test result a Aaa lama pla sar c impactdemo TestFile dat Think of the computer as a file cabinet and t directory impactdemo as a folder in the drawer the file TestFile dat as Open New File E numbered entry on the document The drawer can hold multiple folders di documents files and each document can contain multiple entries records Subdirectories are folders within folders 17 15 Click the Directory Subdirectory text box Keyboard press Alt t enter the directory name ImpactDemo in the box and press Enter Always press the Enter key to transfer information entered into a text box into the computer memory Enter the file name MyPlate in the File Name box The suffix dat is added automatically and the complete file designation appears in the gray box below figure II 6 Press Enter to enter the file name into memory 16 Click Open New File Keyboard press Alt N to create and open the new file A message box appears stating that the file has been opened and that it contains no records Click OK on the message box Figure II 7 The New File portion of the Open Test Date File screen Keyboard press Enter to close it Control is returned to the Main Menu screen and all command buttons are enabled Figure II 8 Main Menu Open Test Data File i Describe Structure
85. l cursor in the cross section box at the right It should be clear from this exercise that frequencies between the green and blue lines represent P wave reflections from within the concrete layer while frequencies to the right of the green line represent P wave reflections from within the asphalt layer In searching for cracks or delaminations within the concrete look for peaks between 5 9 and 11 2 kHz Open record 32 The dominant peak in the spectrum is at 11 2 kHz the frequency associated with P wave reflections from the asphalt concrete interface This indicates debonding at the interface In this case no information can be obtained about the concrete layer because the P waves are not transmitted across the debonded interface Open record 33 The dominant peak at 9 8 kHz indicates P wave reflections from a depth of 162mm or about 27mm below the asphalt concrete interface the approximate depth of the upper layer of reinforcing steel in the concrete Delamination in the concrete at that level was confirmed by coring Removing or dipping the R Wave In the waveform of Figure 33 the magnitude of the R wave is very large compared to the remainder of the signal To gain a clear picture of the frequencies contributed to the spectrum by the R wave place the active waveform cursor to the right of the Rwave at 302 us and click Cut Waveform The result is a spectrum showing that the large R wave in this signal contributes a broad dome of frequencies
86. landing on runways The impact echo method can be used to detect voids under concrete slabs in these and similar situations This application follows from concepts introduced in Chapter 3 of Sansalone and Streett pp 42 45 based on the reflection and refraction of P waves at an interface between two solid layers with different acoustical properties The case considered here is that of a concrete plate upper layer in contact with soil where the depth of the soil is great enough that no wave reflections from within that layer reach the transducer Only wave reflections from within the concrete or from the concrete soil boundary are significant The acoustic impedance of concrete is several times larger than that of typical soils As a result when a stress wave encounters a concrete soil interface some of the wave energy is transmitted and some is reflected For more details see Chapter 13 of Sansalone and Streett pp 135 142 Voids Under Plates Detecting voids under concrete plates is one of the simplest applications of the impact echo method It relies on the clear and easily recognizable difference between waveforms and spectra obtained from plates in contact with soil on the one hand and plates in contact with air a void under the slab on the other Figure VII 1 shows a typical set of results obtained from an impact echo test on a concrete plate in contact with soil The waveform shows periodic displacements caused by P wave reflectio
87. layed in the text box at the upper right of the screen Example 1 The relationship between depth d frequency f and wave speed Cp is d BCp 2f where B is a shape factor characteristic of the geometry For a plate structure B 0 96 and the quantity 0 96Cp is sometimes called the apparent P wave speed in a plate or Cp plate see footnote 6 on page 29 The reciprocal relationship between d and f causes the uncertainty in depth due to the limited resolution in the spectrum to increase sharply at low frequencies To illustrate this place the spectrum cursor on the peak at 6 3 kHz and note that the corresponding depth is 315mm Moving the cursor one digital point above and below this peak to 6 8 and 5 9 kHz using the gt and lt keys shows the corresponding depths to te 292mm and 341mm respectively Therefore the uncertainty in depth at this level due to the resolution of 0 24 kHz is about 24mm or 7 per cent of the nominal depth of 315mm Now move the spectrum cursor to 20 kHz The depth corresponding to this frequency is 100mm and the uncertainty due to the frequency resolution of 0 49 kHz is 23mm or 2 3 per cent of the nominal depth This 1s a subtle point that is sometimes difficult to grasp Experiment with moving the cursor back and forth and observe how the change in depth associated with a change of one digital point in the spectrum increases as the frequency decreases Example 2 There is a Depth scale beneath
88. length of about 2 milliseconds is similar to that obtained in step 5 Click Restore W to restore the waveform Click the 5 H Scale text box enter 2048 in place of 1024 and press the Enterkey The result is a spectrum based on a record length of 2048 x 4 us or about 8 milliseconds In this case the effects of multiple wave reflections within the bounded structure have become dominant and the thickness frequency peak at 7 8 kHz is almost lost in the noise Place the active waveform cursor at about 1000 us and click Cut Waveform The resulting spectrum based on a record length of about 1 millisecond shows a single dominant peak at 7 8 kHz The four cycles in the waveform following the R wave indicate that only 4 P wave reflections were needed to obtain a clear response Place the cursor at 2000 3000 4000 and 6000 ms and cut the waveform at each of those points The results show spectra based on record lengths of 2 3 4 and 6 milliseconds It is clear in this example that increasing the record length beyond about 2 milliseconds complicates the picture Open record 41 This is from a test on a railway tunnel wall with a thickness of 300 to 400mm The lateral dimensions were very large relative to the thickness so that over the 4 millisecond record length there were no effects from wave reflections from the lateral boundaries The result is a waveform that closely approximates a simple sine wave To confirm this place the active waveform curs
89. ll be created for each record saved as an ASCII file These files will be identified as follows diimpactdemoiTestFileiRcrdi asc Cancel Figure III 5 Screen for saving test records as ASCII files For each record saved a new file is created in the same directory as the test data file The file name of the test data file becomes a subdirectory name and the ASCII file name is Rerd where is the record number of the record being saved The suffix asc added to the file name to identify it as an ASCH file For example writing an ASCII file for record number 7 from the file c ImpactDemo TestFile dat will create an ASCII file of the waveform and spectrum and save that file as e ImpactDemo TestFile Rerd7 asc ASCII files can be read directly into spread sheets or into other software programs for graphing or other purposes Saving records to a Microsoft EXCEL Spreadsheet You can create Microsoft Excel Spreadsheets using Impact E or ImpactDemo software but you must have Microsoft Excel software installed on the computer on which you open and use these files There are three options for saving data from individual test records to an Excel Spreadsheet 1 Saving a Summary of Records for all or part of a data file On the File Utilities Screen which is opened from the Main Menu enter the beginning and ending numbers of the records to be included in the Summary of Records the default setting includes all records in the file Click Sa
90. lt B The screen shown in Figure II 15 will appear on your computer This shows the results of a test on a concrete slab as they would appear immediately after the test has been performed The analysis and interpretation of test results will be discussed in later sections of this tutorial 22 is Begin Testing 9 8 1997 4 29 Thickness mm Wave Speed 6 Thickness Frequency kHz Str Name tunnel Type Plate _Date Time 1 Clip Level y F Contact Time Change Curs Cut R Wave Cut Waveform 2Y Scale 0 82 3H Scale Resolution Depth GEJ mm in m 0 10 20 30 40 50 60 kHz 10070 50 40 30 20 15 Depth 10 Cursors 4 Mae 5 H Scale 60 Last Record Saved Number ps Plate thickness y Save as New Rerd Help e Trigger rmedX tee Z Show Record List Compare 1 Signal Screening On Filter Low Freg Figure I 15 The results of a test to determine the thickness of a plate as they appear on the computer screen immediately after the impact 27 Click Main Menu at the lower right of the screen Keyboard press Alt a and on the Main Menu screen click Exit Keyboard press Alt x to close the program This completes Section II of this tutorial covering the procedure for setting up the impact echo test system opening the software program initializing the system and preparing for an impact echo test Repeat the numbered action steps in this section
91. m and the caption on the depth box at the right reads Cursor out of range The 3 9 kHz peak is a displaced thickness frequency that is the solid thickness frequency of 7 8 kHz has been displaced to a lower frequency 3 9 kHz because of the reduced stiffness of the slab in the vicinity of a flaw and because of the increased path length taken by the stress waves that travel around the flaw to reach the bottom of the slab This is a common pattern in impact echo testing The presence of a significant peak in the spectrum at a frequency below the thickness frequency for the solid plate is a certain indication of a flaw This is illustrated schematically in Figure V 3 When the depth of the flaw is greater than about 10 cm 4 inches the response from multiple P wave reflections within the layer above the flaw is relatively strong the 13 2 kHz peak in record 11 If the depth of the flaw is less than 10 cm flexural vibrations in the thin layer are often excited and the response is dramatically different This case is treated in the next paragraph Amplitude oO 3 5 E 3 a lt Frequency kHz Frequency kHz Figure V 3 Comparison of solid response left with response in the vicinity of a flaw of limited lateral extent 56 Shallow Flaws and Delaminations A Special Case If a crack is less than about 100mm 4 inches below the surface of a concrete structure an impact excites flexural vibrations in the thin layer
92. matical function called a Fourier transform Because the waveform is stored in the computer as a two dimensional array of voltage vs time the transformation is carried out numerically using a technique known as a Fast Fourier Transform or FFT See Sansalone and Streett Chapter 5 Frequency Analysis Calculating the spectrum is sometimes referred to as transforming the signal from the time domain the time displacement graph called the waveform into the frequency domain an amplitude frequency graph called the spectrum The significance of the spectrum can be explained as follows If one were to assemble a group of sine waves with amplitudes and frequencies represented by points on the spectrum and add all these together the result would be the waveform In other words the height of the spectrum at any frequency gives the amplitude of a sine wave of that frequency that is a component of the waveform If the waveform is a single sine wave of frequency fand amplitude A the spectrum is a spike of height A at frequency f A waveform produced by multiple reflections of stress waves using the impact echo method contains periodic components but is not a pure harmonic function like a sine wave When the waveform is transformed into the frequency domain by the FFT the result is a continuous distribution of frequencies in which the dominant periodic components appear as sharp peaks The frequencies of these peaks are used with the P wave speed to
93. meters are thickness T P wave speed Cp and frequency f These Parameters are displayed in text boxes in the upper right corner of the screen For a test on a solid portion of a plate these three parameters are related through the equation T 0 96Cp 2f 5 This is the fundamental equation of impact echo testing for simple plate structures If two of the three parameters are known the third can be calculated The Waveform 6 Learn the features of the waveform The waveform upper graph is a plot of the voltage time signal from the transducer Because the voltage output of the transducer is proportional to displacement normal to the surface this graph describes displacements at the point of contact between the transducer and the surface These displacements are caused by the arrival of stress waves generated by the impact and reflected from the external boundaries or flaws within the structure Multiple reflections of stress waves within a structure give rise to periodic displacements The frequencies of these displacements appear as peaks in the amplitude spectrum the lower graph and they provide information about the dimensions of the structure and depths of flaws It is important to understand the features of the waveform and to learn how to interpret the information it contains Examine the fixed labels gray boxes that contain information about this test record e Voltage The number in the gray box in the lower left corner is the vol
94. named c ImpactDemo TestFile dat containing impact echo test results from projects 16 in the USA and elsewhere is included with the ImpactDemo software program When the software is installed this file is installed in the ImpactDemo directory that is created on the e drive the hard disk during the installation 3 Choose Hard Disk Default or Other Drive Se MICRON Create a New File isting Fi File Name asttex dat Directory Subdirectory MAR nn GI BRMFLPRO GC DAILY G DemoManual File Name J Exchange G FIDELITY E Fowin CO Gage CI GRS L Open New kile Cancel Help Open Existing File Figure II 6 The Open Test Data File screen Create a New Data File in a New or Existing Directory Subdirectory 14 Choose the drive for a new file by clicking the drop down drive box at the top of the screen Keyboard press Alt o to select the drive box followed by the letter of the desired drive e a d etc The e drive hard disk is the default When the Open Test Data File screen is opened Figure H 7 the Open New File and Open Existing File buttons are disabled When the file to Create a New File be opened has been identified one of these buttons will be enabled At this point a new data file will be opened in the ImpactDemo directory for impact echo tests on a plate Directory Subdirectory imagodemo File Name MyPlate 3 The descr
95. nd plate thickness Patterns of stress wave propagation and reflection and the resulting frequencies are changed by the presence of flaws These changes appear in the waveforms and spectra obtained from impact echo tests and they provide both qualitative and quantitative information about the flaws This section focuses on behavior associated with cracks and voids in plates including the special case of a shallow crack or delamination and the response of plates containing unconsolidated concrete honeycombing Although the discussion is focused on plate structures the interaction of stress waves with flaws and the resulting changes in the waveform and spectrum can be generalized to any geometry For a more detailed discussion of these topics see Sansalone and Streett Chapters 9 12 Cracks and Voids A crack or void within a concrete structure forms a concrete air interface Laboratory experiments have shown that cracks with a minimum width crack opening of about 0 08mm 0 003 inches cause almost total reflection of a P wave The responses from cracks and voids are similar because stress waves are reflected from the first concrete air interface encountered Thus a crack at a depth d will give the same response as a void whose upper surface nearest to the impact surface is at the same depth Figure V 1 Figure V 1 A crack at depth d gives the same response as a void at that depth Determine the Solid Response of the Plate Struc
96. ns within the concrete plate but because energy is lost to the soil each time a P wave is incident on the concrete soil interface the amplitude of the dsplacements indicated by the signal voltage decays rapidly The corresponding spectrum shows a single peak corresponding to the frequency of P wave reflections from the concrete soil interface Note however that the peak is somewhat rounded and is broader than those obtained from plates in contact with air for example see record in the open file or Figure MI 1 in this manual In Figure VII 1 only a few wave reflections were recorded before the signal decayed to an undetectable level For comparison Figure VI 2 shows a typical result obtained from an impact echo test on the same plate at a location where a void exists in the soil just below the plate In this case P wave reflections occur from a concrete air interface at the bottom of the plate Because virtually all of the wave energy is reflected at a concrete air interface surface displacements caused by the arrival of reflected P waves decay more slowly compared to those reflected from a concrete soil interface The response is essentially the same as that obtained from a simple concrete plate in contact with air The spectrum exhibits a very sharp high amplitude peak corresponding to the P wave thickness frequency If the concrete slab is relatively thin about 150mm or less a lower frequency lower amplitude peak labeled faex in Figure V
97. ns in a fully grouted post tensioning duct In situations where the R wave has a large amplitude relative to the amplitude of displacements produced by P wave reflections check to see if the key features of the spectrum can be clarified by removing or clipping the R wave in the waveform With experience in using the impact echo method in a variety of applications the above steps and questions become automatic The response patterns obtained from different types of structures become easily recognizable Understanding the Effects of the Sampling Interval and Number of Samples The choice of sampling interval and number of samples for an impact echo test is discussed briefly on p 18 and in more detail in Appendix A on data acquisition parameters The effect of changing these numbers is illustrated with two records in the test data file 3 Open the test data file ImpactDemo TestFile dat and open record 43 This record shows the results of a test on a solid 250mm thick concrete plate with a wave speed of 3900 m s The plate was a laboratory test specimen with a length and width of about 1 5 meters The sampling interval was 2 us sampling rate 500 kHz and 1024 data points were used to plot the waveform The record length the time period over which the signal is recorded is the product of the number of data points n and the sampling interval At In this case nAt 1024 x 2 2048 us or about 2 milliseconds The waveform has the general
98. nses for a plate containing a single post tensioning duct in any particular cross section a the response of the solid plate b the response over a fully grouted duct and c the response over a duct containing a void In b the full thickness frequency peak fr is at or only very slightly below the frequency of the response from a solid plate a and the response due to wave reflections from the steel tendons is given by the equation fu Cp 4d In c the thickness frequency fr is shifted downward due to the additional path length traversed by P waves diffracted around the void and by the reduced stiffness of the plate in the vicinity of the void The frequency of Pwave reflections from the void is given by f 0 96C 2d which means that fwia exceeds fstee by a factor of about two even though the void and the steel are about the same distance beneath the surface fr 0 96Cp 2T Impact steel Cp 4d fT shifted f voii a 0 96Cp 2d Frequency Figure IX 1 Schematic diagram showing the three types of response for a plate containing post tensioning ducts a solid plate b grouted duct and c duct containing a void These responses are illustrated using several records from impact echo tests on a post tensioned girder n a concrete bridge The web of the girder behaved as a simple plate lateral dimensions greater than five times the thickness It had a thickness of 0 25m or about 10 inches Positions of t
99. nt Excel file new or existing enter the path and name in the text box and press the Enter key cAlmpact E TestFile xls Save Waveforms and Spectra to Excel Spreadsheet l Save Thickness Delamination Data from 2 D Grid to Excel Spreadsheet Cancel Einish Help Figjre III 6 The Excel Spreadsheet screen 41 Saving an array of points from a 2 D grid This method is designed to permit a 2 dimensional array of data points from a rectangular grid to be saved as a 2 d array in an Excel Spreadsheet It is useful for example for recording thickness in a concrete slab or tunnel wall or in recording the results from tests on a bridge deck to locate areas where a relatively thin layer of concrete is delaminated separated from the main slab To use this method the array must be laid out using letters and numbers as follows Al A2 An B1 B2 Bn Cl C2 Cn etc This means that the name of each record must be a letter followed by a number If the first test is named Al the computer will automatically name succeeding records A2 A3 etc To begin column B enter Bl as the record name for the first test in that column It is not necessary for the named points to be recorded in sequence in the computer As long as they are named according to points in the array they will be properly sorted when saved to an Excel Spreadsheet Up to 26 lettered columns can be used A to Z and any number of rows Before saving data
100. of valid waveforms can be found in figures in this manual and in the accompanying test records in the file e ImpactDemo TestFile dat That portion of the waveform following the R wave contains the important information on resonant frequencies Impact E software has the capability to automatically screen or check incoming waveforms to determine whether they have the characteristics of a valid impact echo test See Program Parameters Option in Section I or see Screening for Bad Signals in the on board Help System for Impact E Software If an incoming signal does not have the characteristics of a valid impact echo test the computer will emit a loud buzzer signal and the waveform will be plotted as a red line Invalid waveforms can result from rough surfaces on the concrete dirt or other foreign material on the surface premature triggering loss of contact between the transducer and the surface accidental movement of the transducer during the test or a host of other causes Examples of bad waveforms resulting from some of these causes are shown in Figure X 1 The waveform in Figure X 1 a is the result of a stray electrical signal that triggered the data acquisition system prior to an impact Stress waves generated by movement of the impact echo hand held unit generated the invalid waveform in b Stress waves caused by the vibration or impact of heavy equipment jackhammers etc which are in contact with the structure being te
101. og signals are digitized by an analog digital data acquisition system and transferred to the memory of a computer Through a mathematical operation the voltage time signal is transformed into an amplitude frequency graph a spectrum Peaks in the spectrum identify the dominant frequencies in the waveform which are used to calculate thickness and or the depths of flaws Sansalone and Streett Chapter 4 pp 47 52 Transducer Data Acquisition System and Computer Waveform Spectrum Amplitude Frequency Figure I 1 Schematic diagram of the impact echo method The patterns present in the waveforms and the dominant frequencies identified by the spectra provide information about the depth of flaws or the dimensions of the structure such as the thickness of a pavement For each of the common geometrical forms encountered in concrete structures plates circular and rectangular columns rectangular I and T beams hollow cylinders etc impact echo tests on a solid structure produce distinctive waveforms and spectra in which the dominant patterns especially the number and distribution of peaks in the spectra are easily recognized For solid structures the frequencies of the dominant peaks provide information about the thickness of the structure If flaws are present cracks voids delaminations etc a different set of key frequencies is recorded providing qualitative and quantitative information about the existence and loca
102. on File and Copy Records allow sets of records within the range specified in the two text boxes above the buttons to be copied from the Current File the file identified at the top of the screen to a second file The second file called the Destination File is opened by activating the Open Destination File button which calls up the Open Test Data File screen see p 17 The Destination File can be a new or existing file After the Destination File has been opened activating the Copy Records button causes a Yes No message box to appear giving the names of the Current and Destination Files and the number of records to be copied Selecting Yes will cause the copying to take place This enables 39 the user to create new files from individual records or groups of records selected from different test data files After a Destination File has been opened the To Destination File button on the Examine Test Data screen will be enabled allowing the user to add to the Destination File the record currently displayed on that screen Saving Records to ASCII Files A sequence of records can be saved as ASCII files by entering the beginning and ending record numbers in the boxes above the Save As ASCII Files button and activating the button Another screen appears Figure III 5 The drop down box at the top allows the user to choose the drive Hard Disk Floppy Disk etc to which the files will be saved Choose Drive Se MICRON y A new file wi
103. onded P wave reflections due to the acoustic mismatch will be too weak to be detected see p 44 in Sansalone and Streett and the P wave reflections will occur from the bottom of the concrete layer The layered structure will respond as a single plate with the added complication of different 78 wave speeds in the to layers In this case the response due to P wave reflections from the bottom of the concrete is called the composite thickness response Experience has shown however that even when the two layers are well bonded there is sometimes a significant fraction of unbonded area due to the coarseness of the asphalt and this is sometimes sufficient to cause P wave reflection see Chapter 15 of Sansalone and Street pp 151 158 The Pwaves are reflected from a large number of small air inclusions at the interface rather than from the area of contact between asphalt and concrete Thus the basic response of a concrete slab with a bonded asphalt overlay can consist of two components 1 a relatively weak response caused by multiple P wave reflections between the impact surface the top of the asphalt overlay and the asphalt concrete interface and 2 a dominant composite thickness response caused by multiple P wave reflections between the impact surface and the concrete air interface at the bottom surface of the concrete If the fraction of unbonded area at the interface is low the peak associated with the first response will be weak or
104. onl is for a case where there is a full thickness frequency peak as well as a weaker due to P wave reflections from the interface Because there is only one distinct peak in record 31 Option2 is selected This option presents the user with several choices of variables to change and adjust If the single peak present is assumed to be the composite thickness frequency peak then the wave speeds and thickness of each layer are adjustable parameters If two or more of these are not known with certainty it is difficult to proceed We assume in this case that the wave speeds have been separately measured and are known to be accurate This means that the discrepancy between the expected and observed composite thickness frequencies 5 9 and 6 3 kHz is due to incorrect values of thickness for one or both of the layers In this case we assume that it is the asphalt thickness that is not accurately known Having entered 6 3 as the full thickness frequency select Overlay Thickness as the variable to be calculated and click Calculate make certain that Option2 on the left has been selected The overlay thickness is recalculated as 118mm Click Accept to return to the Examine Test Data screen and note that the full thickness frequency marker is now at the main peak 6 3 kHz The frequency associated with reflections from the interface has been automatically recalculated as 12 7 kHz and the green marker line on the spectrum has been moved accordingly If the
105. or at 1000 2000 and 3000 us and cut the waveform at each of those points The resulting spectra are remarkably similar Place the active cursor at any point on the waveform and click Cut R Wave Place the cursor at a point at least 800 us to the right of the first point and click Cut Waveform The resulting spectrum shows the frequency content of the waveform between those two points It always 99 displays a dominant peak at about 6 3 kHz confirming that a single frequency dominates the entire waveform 100 Appendix A Setting The Data Acquisition Parameters Data acquisition parameters are parameters that define the conditions under which the analog signal from the hand held transducer unit is digitized for transfer to the computer memory For routine testing the data acquisition parameters are set automatically by the software However it is possible to override the default values by accessing the Parameters for Data Acquisition screen shown in Figure A 1 This screen is accessed from the Main Menu screen arameters for Data Acquisition Parameters for the Data Acquisition Card Number of Samples N Voltage O 512 O 2048 O 0 1 Volts 1024 O 0 2 Volts O 7 0 4 Volts Time Between Samples t O j 1 0 Volts O 2 0 Volts O 5 0 Volts 0 2 Micr Sec 3 Micr Sec O 0 5 Micr Sec O 4 Micr Sec O 1 MicrSec O 5 Micr Sec 2 Micr Sec O 10 Micr Sec Trigger Voltage Volts Frequency Resolution Fr Res
106. or message e The information label Records In File n at the lower left of the control panel indicates the total number of records in the open file It is not interactive e The Description Text Box The large text box to the right of the number name text boxes contains a brief description of the record 40 characters maximum This description is normally entered when the test record is saved but it can be changed when the record is examined Click this text box to open it Keyboard press the Tab key repeatedly until the box is highlighted by a red border A new or modified description can be entered Press Enter to save it to the data file e Click the Z Show Record List command button Keyboard press Alt Z A list of all records in the file appears on the screen and the caption changes to Z Hide Record List The record list that appears is interactive To open any record on the list double click that record Keyboard use the up down arrow keys to select the desired record and press Enter The record list disappears and the selected record is displayed To remove the record list without opening a new record click Z Hide Record List Keyboard press Alt Z e Click the Next Record and Prev Record buttons Keyboard press Alt X or Alt V to access the next record or previous record in the data file e Comparing Records Click the text box next to the Compare button Keyboard press the Tab key repeatedly unt
107. orm and click the left button The active yellow cursor moves to that position Double click on the active cursor or click the Change Curs button Keyboard press Alt C to make the other cursor active yellow Place the mouse pointer at another point on the waveform and click the left button The active cursor moves to that position From the keyboard use the key at the right of the 3rd row from the bottom of the keypad to move the active cursor from peak to peak to the right Use the Kk key to move it peak to peak to the left The two keys in between LI and move the active cursor left or right in small steps Holding any of these keys down causes continuous movement Do not use the Alt key with these keys to move the cursor The cursor on the spectrum is moved in the same way The four keys used to move the spectrum cursor are just below those used to move the waveform cursors To move the spectrum cursor peak to peak use the Mm and keys to move it in small steps use the lt and gt keys Although it is easy to position the cursors using the mouse there are times when the keyboard must be used It is important to experiment with moving the cursors with the keyboard and changing the active cursor on the waveform to become familiar with this feature of the display 28 The Critical Parameters Thickness P Wave Speed and Frequency In impact echo tests on plate structures the three critical para
108. permanent part of this test record This will cause the contact time to appear at the bottom of the waveform graph when the record is opened as shown for record 12 Click Exit to return to the Examine Test Data screen without saving the contact time Use the R wave to determine the frequency content of the impact The surface wave or R wave which appears as a deep trough at the beginning of the waveform is a mirror image of the force time function of the impact and it provides useful information about the distribution of amplitudes and frequencies in the resulting stress waves See Sansalone and Streett Chapter 3 Place the active waveform cursor at the right side of the R wave elapsed time 244 us and click Cut Waveform to set equal to zero that portion of the waveform to the right of the cursor The result is a spectrum showing the distribution of amplitudes and frequencies produced by the impact Figure V 6 59 10 11 lip Level Contact Time 48 us Em 1 Contact Time TEF Curs Cut R Wave Restore W 2Y Scale 1 25 Resolution Esa Depth mm in m Figure V 6 The R wave upper graph and associated spectrum lower graph In the spectrum the height of the curve at any point on the horizontal axis is a measure of the relative amount of energy in the stress waves at that frequency The graph shows that there is very little energy at frequencies above about 26 kHz and that the energy level is relatively
109. plate structure such as a bridge deck or the web of a large girder in which there is only one duct directly beneath the surface at any point In all cases the impact echo method is restricted to situations where the walls of the ducts are metal rather than plastic Effective use of the impact echo method for detecting voids in grouted tendon ducts requires knowledge of the location of the ducts within the structure This information is typically obtained from plans and or the use of magnetic or eddy current cover meters to locate the centerlines of the metal ducts Once the duct locations are known impact echo tests can be performed to search for voids Metal Ducts Tendon ducts in post tensioned structures are typically made of steel with a wall thickness of about Imm 0 04 inches The space not occupied by tendons inside the duct is or should be filled with grout which has an acoustic impedance similar to that of concrete Because the wall thickness of a duct is small relative to the wavelengths of the stress waves used in impact echo testing and because a steel duct is a thin layer of higher acoustic impedance between two materials of lower acoustic impedance concrete and grout it is transparent to propagating stress waves Therefore the walls of thin metal ducts are not detected by impact echo tests In contrast plastic ducts have a lower acoustic impedance than concrete or grout and they are not transparent complicating attempts to
110. pond The width of the R wave the time between the two points where it crosses the zero voltage line is a measure of the contact time the length of time that the impacting sphere is in contact with the concrete surface The contact time is a function mainly of the diameter of the sphere and it determines the maximum useful frequency in the signal See Chapters 3 4 and 24 in Sansalone and Streett 30 10 11 12 Example Click Contact Time Keyboard press Alt T A gray box appears with information about he contact time the sphere diameter and implications for testing The software attempts to place the two cursors at the points where the R wave intersects the horizontal line but it is not always successful To adjust the cursors more precisely click the Adjust Cursors button and click Change Curs to select one of the cursors to be moved Keyboard press Alt A and Alt C Move the cursor with the mouse or keyboard as explained under Moveable Cursors above The contact time for this record is about 50 us After positioning the cursors click Contact Time Keyboard press Alt T to return to the previous display Click Exit Keyboard press Alt X to return to the Examine Test Data screen without saving the contact time or click Save Contact Time Keyboard press Alt S to save the contact time as a permanent part of the record and display it in the lower portion of the waveform Double click on
111. re IV 3 The Set up for Wave Speed Measurement screen Click the Structure Name text box and enter Test Slab 1 as the structure name Leave unchanged the default values of 300mm for Distance L and 0 18 for Dynamic Poisson s Ratio Click Begin Test to open the Wave Speed Measurement screen Figure IV 4 Default values for the sampling interval and voltage setting see Appendix A are 0 5 micro seconds and 0 1 volts If other values have been chosen a message will appear asking the user to verify that other than the default values are to be used The Wave Speed Measurement screen contains a single graph on which two waveforms are plotted when a wave speed measurement test is performed Because a new file has been created and opened the graph is blank The green border around the Trigger Armed X button is flashing indicating that the system is ready to receive the signals from a wave speed measurement test 46 is Wave Speed Measurement File e imagodemo TestFile dat Structure Name Test Slab 1 Date Time 8 17 93 12 00 00 AM Cursor to Cursor us Distance L mm Poisson s Ratio Wayespd From R wave Voltage Setting 2 v Sampling Interval Z u s WaveSpi From PR Wave 1 Upper 2 Lona 3 Horiz Vert Scale 0 29 Vert Scale 7 Cursors Scale Last Record Saved Number Concrete slab in contact E Save to Data File Help with soil Trigger Arm
112. rete slab 71 Under Plates 70 Voltage 29 101 Trigger Voltage 102 Wave Speed 44 47 Direct Measurement 44 P wave 44 Wave Speed Measurement Screen 45 Waveform 29 90 Cutting the Waveform 97 Waveform command 36 Waveforms Valid and Invalid 90 WaveSpd from P Wave command 47 Z Hide Record List command 27 34 Z Show Record List command 27 34 119
113. rformed on a concrete beam and on the wall of a ventilation shaft for a railway tunnel Figure Il 13 Testing a concrete beam football Figure I 14 Testing a concrete wall in a stadium USA ventilation shaft for a railroad tunnel 21 Within one or two seconds after the impact the results appear on the computer screen a waveform the voltage time signal from the transducer in the upper graph and the amplitude spectrum a plot of amplitude vs frequency in the lower graph A normal signal will be accompanied by a bell sound on the computer If a loud buzzer sound is emitted the signal is of questionable value and should be ignored The results of a test to determine the thickness of a plate are shown in Figure II 15 as they would appear on the screen immediately after the impact To view this screen on your computer perform the following steps 22 On the Begin Testing screen click Main Menu Keyboard press Alt a to return to the Main Menu screen 23 Click Open Test Data File Keyboard press Alt 0 24 Double click ImpactDemo directory in the Directory box Keyboard press Alt r select ImpactDemo using the up down arrow keys and press Enterl 25 Double click on the file TestFile dat in the File Name box Keyboard select TestFile dat using up down arrow keys press Enter and press Enter again to open the file 26 On the Main Menu screen click Begin Impact Echo Test Keyboard press A
114. rigger eoe Voltage 0 10 negative slope 0 100 200 300 Time u sec Figure A 3 Leading portion of waveform illustrating the significance of the trigger voltage When the data acquisition system is activated by the computer it begins recording continuously into its on board memory reading data points digitized voltages from the transducer into one end of a buffer and pushing them out the other end Prior to the impact there is no signal from the transducer input signal is zero and a string of zeros is fed into the buffer When the stress waves arrive at the transducer a voltage is generated and the early part of the signal is used to initiate or trigger the data acquisition process The Trigger Voltage instructs the system to begin recording data points when the input signal reaches a set value typically 0 05 to 0 25 volts and the slope of the voltage time curve is negative voltage decreasing with time This insures that triggering occurs just after the arrival of the R wave which produces a sharp drop in voltage If there are mechanical vibrations in the structure being tested they will sometimes trigger the system before or after the impact If this occurs the Trigger Voltage should be assigned a value in the range 0 15 to 0 25 The data acquisition system is set to record approximately 196 data points prior to the trigger point This insures that the earliest wave arrival at the transducer is part of the data stream
115. rsors are properly positoned to measure contact time at the two points where the R wave trough intersects the horizontal axis Adjust Cursors 2 The indicated contact time is 48 micro seconds Sphere diameter approximately 11 mm 3 The maximum useful frequency is about 26 kHz 4 This impact is suitable for detecting flaws at depths greater than about 72 mm or flaws with lateral dimensions greater than 144 mm To locate flaws closer to the surface or smaller in lateral extent use a smaller impactor Save Contact Time Figure V 5 The Contact Time screen with the R wave drawn on an expanded scale 10 The two cursors should be positioned at the points where the R wave trough crosses the zero voltage line the horizontal line The width of the R wave between these two points is equal to the contact time the length of time the impacting sphere is in contact with the concrete surface during the impact If necessary click Adjust Cursors and use the mouse or keyboard to properly position the cursors and click Contact Time again The text on the Contact Time screen gives the measured contact time it is about 48 us in this case the approximate diameter of the impacting sphere 11mm the maximum useful frequency in the signal 26 kHz and information about sizes and depths of flaws that can be detected For more information see Table 15 p 302 of Sansalone and Streett Click Save Contact time to save the measured contact time a
116. rst transducer The second step is to position the second cursor to mark the P wave arrival on the lower waveform Follow the procedure outlined above using the lt key to move the cursor to the left and the gt key to move it to the right On the flat portion of the waveform leading up to the wave arrival the voltage remains steady at 0 0006v The first point at which it rises significantly is 157 5 0 0007v Leave the cursor at this point The Cursor to Cursor box in the upper right of the graph indicates that the elapsed time between the P wave arrival at the two transducers is 157 5 87 5 70 0us Click WaveSpd from P Wave to calculate the P wave speed Keyboard press Alt P A text box appears showing two P wave speeds The speed of 4286 m s given for Beams and Columns is the true P wave speed and the second value 4114 m s is the apparent P wave speed in a plate 0 96 times the true P wave speed used for all tests on plate structures see Sansalone and Streett pp 51 52 Records 3 and 4 in the file that has been opened c ImpactDemo TestFile dat show additional measurements of P wave speed in concrete Follow the instructions given above to determine the P wave speed from these records The plate P wave speeds determined from these records should be around 4100 m s P Wave Speed From Impact Echo Tests on a Plate of Known Thickness The fundamental equation of impact echo is d BCp Qf where B is a s
117. rtifact resulting from excitation of the natural resonance of the piezoelectric transducer element and the brass block to which i is attached This transducer resonance at 1 0 kHz is often excited when an impact is made close to the transducer and it is especially strong when the concrete surface being tested is very smooth It is responsible for the low amplitude very low frequency component in the waveform To remove this component by digital filtering place the waveform cursor at a frequency slightly above the 1 0 kHz peak 2 3 kHz for example and click Filter Low Fre q Keyboard press Alt Q Click this button several times to add and remove the low frequencies and note the change in the waveform and spectrum Removing the R wave In this waveform the amplitude of the R wave is large compared to the remainder of the waveform In cases such as this it is sometimes desirable to remove the R wave from the waveform to reduce its contribution to high frequencies in the spectrum To observe this effect place the active waveform cursor at the point where the R wave overshoot returns to zero voltage at about 246 us after the trigger point indicated by the label on the cursor and press Cut R wave Keyboard press Alt R Click this button several 76 times and observe the change in the waveform and spectrum The level of high frequency noise in the spectrum is reduced when the R wave is removed In this case the effe
118. s in identifying areas where the overlay the top layer of a slab poured in two layers or a concrete patch for example is debonded or only weakly bonded to the parent concrete The test records used here are from tests on a concrete floor that was poured in two layers with effective bonding in some areas and little or no bonding in others The floor had a nominal thickness of 154mm about 6 inches and had been poured in two layers of approximately equal thickness An impact echo test on a solid slab of known thickness was used to determine the P wave speed of 3780 m s Using the equation f BCp 2d the frequency of P wave reflections from the full thickness areas where the two layers are strongly bonded was found to be 12 2 kHz while the frequency of reflections from an unbonded interface at a depth of about 77mm was found to be 24 5 kHz 1 Start the ImpactDemo program and open the file e ImpactDemo TestFile dat provided as part of the software for this tutorial Records 25 30 in this file are from tests on the 2 layered floor described above Open record 25 click the Number text box at the lower left of the screen Keyboard press Alt B enter 25 and press the Enter key The dominant frequency peak is at 12 2 kHz indicating that the response is from a point where the two layers are bonded and the composite thickness is 155mm or about 6 1 inches Recognizing Transducer Resonance The low frequency peak at 1 0 kHz is an a
119. shown in c and d The important response although discernible in b is very clear in d Record 33 in the test data file c ImpactDemo TestFile dat is very similar to the test shown in Figure X 8 96 Amplitude 10 20 30 Frequency kHz Amplitude AN Aaaa 10 20 30 40 Frequency kHz Figure X 8 An impact echo response obtained from a concrete bridge deck with an asphalt overlay a and b show a signal dominated by a large amplitude R wave c and d show the same results with the R wave removed Clipping the R wave In cases where the R wave is dominant in every test its contribution to the spectrum can be reduced electronically by clipping the R wave signal The clip level is a voltage beyond which signals are ignored and not used in calculating the spectrum Figure X 9 is the waveform shown in Figure X8 a except that the R wave has been clipped The effect on the spectrum is similar to that produced by cutting the waveform altogether Figure X 7 c and d though not as dramatic Clipping is often a convenient method for reducing the contribution of R waves to spectra because it can be done automatically during data acquisition whereas removing the R wave from the waveform must be done by the user after waveform has been plotted Use record 33 in the test data file to observe the effects of setting different values for the clip level Amplitude mu Time us Frequency kHz Figure X 9 The impact echo r
120. st data file must be opened each time the program is started The results of a single impact echo test are stored as a record in a random access file A data file is specified by a path designation consisting of the drive identification c for hard drive a for floppy disk etc followed by the directory name subdirectory name optional and file name separated by back slashes 13 Click Open Test Data File on the Main Menu Keyboard press Alt 0 The Default File screen appears Figure II 5 The Default File screen prompts the user to open a file on the c drive in the directory named ImpactDemo subdirectory TestFiles with a file name consisting of the current date ddmmmyy followed by the suffix dat This creates files with names based on the dates on which the files are created If a different file name is desired it can be entered in the text box 14 Click OK to open the file named in the text box The default test data file for this session will be placed in the directory c impactdemo TestFiles The default file name based on today s date is HR dat Enter another file name if desired Click OK to open this file OK Click CONTINUE to create or open other data files Continue Fig II 5 The Default File screen 15 To open an existing file or a new file in a different directory subdirectory click Continue to bring up the Open Test Data File screen Figure II 6 A test data file
121. sted can also cause the data acquisition system to trigger in the absence of an impact if the stress waves generated are in the range over which a transducer is sensitive typically about 500 Hz to 100 KHz The waveform in c is an example of such a response The spectra produced by invalid waveforms especially the waveform in c are sometimes much like normal spectra but are invalid because they are derived from an invalid source Invalid waveforms are common and should be rejected and the test repeated 92 Amplitude 10 20 30 40 Frequency kHz o ye 2 a E lt 10 20 30 40 Frequency kHz Amplitude 10 20 30 40 Frequency kHz Figure X 1 Examples of bad waveforms left and their spectra right Another example of an invalid waveform is shown in Figure X 2 This type of waveform is often observed when testing thin structures or mainly over very shallow delaminations The response is generated by an impact of too large energy content typically caused by using too large an impactor Large surface displacements cause over ranging of the transducer the transducer produces a voltage greater than the maximum range of 2 5 volts and intermittent loss of contact between the surface and the transducer A normal waveform can often be obtained by reducing the force of the impact Sometimes a bad waveform such as this provides evidence of the presence of very shallow delaminations Over Range Amplitude Vol
122. stination File 35 Saving to Ascii Files 39 Show Record List Command 34 Remove command 34 Replace Record command 35 Resolution 32 Restore Waveform 31 See Cut Waveform R wave 10 92 Clipping the R wave 95 Cutting the R wave 31 Cutting the R wave 74 81 97 Irregular 93 Irregular R wave 93 Separated 93 Separated R wave 93 Sampling Interval 29 96 100 Sampling Rate 29 99 100 Save as New Rerd command 34 Save to Data File screen 34 Saving records to a Microsoft EXCEL Spreadsheet 39 Screen for bad signals 16 Setup for Crack Depth screen 66 Shape Factor 32 44 47 Show all dimensions in inches See Program Parameters Show Record List command 27 34 Show screens in black and white See Program Parameters Solid Response 52 Spectrum 31 Interpretation 96 118 Spectrum command 36 Steel Tendons 84 85 Stress Waves 10 Structure Name 20 Structure Name text box 45 Structure Type 19 plate 19 Surface Displacements 26 29 56 Rate of Decay 70 Surface Opening Crack 67 Determining Depth 64 65 Tendon Ducts 84 Grouted 84 Test Records See Records TestFile dat 26 Thickness Plate 48 Thickness Frequency 47 53 Time Between Samples 96 99 To Destination File command 35 transducer 10 Transducer 21 Resonance 74 Trigger Armed X command 21 Trigger Voltage 101 102 Unconsolidated Concrete 61 Vertical Scale 30 33 Vibrations Flexural 56 72 Voids 52 Tendon Ducts 84 Under conc
123. t Fourier transform FFT the result is a table of amplitude vs frequency called an amplitude spectrum Sansalone and Streett Chapter 5 The points in this spectrum are calculated for a fixed number of discrete frequencies Frequency resolution is the frequency difference between successive points in the calculated spectrum If n is the number of samples in the waveform and Af is the time between samples the spectrum calculated by the FFT will contain n 2 points and the frequency interval between them the frequency resolution will be 1 nAt The frequency resolution is displayed on the Parameters for Data Acquisition screen and also on the screen with the results of a test where it appears in the upper right corner of the spectrum graph See for example Figure IX 2 At first glance it might appear desirable to improve decrease the frequency resolution by increasing n or At however it can be misleading to consider the expression 1 nAf in isolation For routine testing of plates beams columns and other common structures sampling intervals of 1 to 4 microseconds and sample numbers of 1024 to 2048 are recommended The resulting frequency resolutions range from 0 12 to 0 98 kHz For more information about choosing n and Af for specific applications and about the effect of frequency resolution on the precision of depth and thickness measurements in impact echo tests see Chapter 6 Digital Signals in Sansalone and Streett Voltage T
124. t for record 22 void beneath the concrete The sharp narrow peak in record 22 is characteristic of P wave reflections from a concrete air interface at the bottom of the slab This completes Section VI of this tutorial covering the detection of voids beneath plates in contact with soils Repeat the action steps in this section until you are thoroughly familiar with the methods described The next section will describe the testing of plates consisting of two layers such as concrete patches and concrete with asphalt overlays 74 Section VIII Plates Consisting of Two Layers Introduction The impact echo response of structures consisting of two layers is described in detail in Chapters 14 15 and 16 of Sansalone and Streett pp 143 166 In this section of the tutorial two cases are examined 1 a concrete overlay on a concrete slab and 2 an asphalt overlay on a concrete slab Concrete on concrete is a special case of a simple plate because the two layers have the same acoustic impedance Asphalt on concrete presents a different and more complicated case because the acoustic properties of asphalt including wave speed and acoustic impedance differ from those of concrete Debonding at the Interface of a Concrete Overlay on a Concrete Slab In the case of a concrete overlay on a concrete slab wave reflections from the concrete concrete interface occur only if the two layers are debonded or weakly bonded The principal use of impact echo i
125. tage ed Over Range 10 20 30 40 Frequency kHz 93 Figure X 2 Response produced over ranging of the transducer and intermittent loss of contact with the surface o Us E a E lt Time us Frequency kHz Figure X 3 A waveform mildly affected by electrical noise Sometimes impact echo signals are affected but not invalidated by stray electrical noise Figure X3 shows a test in which the waveform contains a low amplitude high frequency component which produces a jagged appearance in an otherwise smooth curve The effects of the noise are especially visible in the portion of the waveform prior to the arrival of the R wave In this case the amplitude of the electrical noise is too low relative to the R wave and P wave signals to produce a visible peak in the spectrum and the test is valid Figure X4 is an example of a test in which the electrical noise is more regular and has a larger amplitude In this example which was obtained from a test on a solid 250 mm thick plate the thickness frequency fr is 7 8 kHz f 44 9 noise Amplitude 10 20 30 40 Frequency kHz Figure X 4 A test result with electrical noise of relatively high amplitude The electrical noise appears in the spectrum as a sharp peak at 44 9 kHz In this impact echo test the impact duration was approximately 46 micro seconds and the maximum frequency of useful energy in the impact generated stress waves is about 27 kHz This means
126. tage range selected when the data acquisition parameters were set see Appendix A It is the maximum absolute voltage accepted by the data acquisition system To take full advantage of the 14 bit resolution of the data acquisition system the voltage should be set just above the maximum value expected in the signal In most cases a voltage range of 1 volt works well For very smooth concrete surfaces which produce strong responses a setting of 2 volts is sometimes appropriate e Sampling Interval The gray box in the lower right comer gives the sampling interval or time between samples in microseconds us This is the time interval at which the voltage signal from the transducer is digitized by the data acquisition system The sampling rate is the reciprocal of this number If the sampling interval is 2 us for example the sampling rate is 1 0 000002 500 000 samples per second usually stated as 500 kHz kiloHertz or 0 5 MHz MegaHertz The time between samples is set on the Parameters for Data Acquisition screen Appendix A The general equation is T BCp 2f where B is a shape factor dependent upon the geometry of the structure For simple plate structures B 0 96 and it is customary to use the quantity 0 96Cp as the apparent P wave speed in a plate sometimes written Cp plate so that the equation reduces to T Cp plate 2f This is done in part because in early work on impact echo the existence of the shape
127. te Time 3 12 92 12 00 00 AM Overlay Thickness mm Overlay Wave Speed m s 6 Overlay Thick Frequency kHz 7 Plate Thickness mm 8 Plate Wave 25 1 ClipLevel 25 Contact Time 42 us 2 ju Speed m s 2 Scale 1 25 3H Scale 9 Composite Contact Time e hange Curs Cut R Wave Cut Waveform Thick Freq kHz 0 3 Resolution Depth mm 113 in 286 mm as 50 60 kHz 4V Scale t 5 H Scale 60 This Record Other Records Save This Record Printing Number Full thickness response Next Record Save as New Rerd Prepare to Print Help Name Prev Record Replace Record J Label Peak Main Menu m in File _ Z Show Record List Show Record List Compare 1 To Destination Ele Filter Low Freg Figure VIII 5 A full thickness response for a 180mm thick concrete slab with a 135mm asphalt overlay The proximity of the main peak in the spectrum 6 3 kHz to the expected full thickness frequency of 5 9kHz makes it reasonable to assume that one or more of our estimates of thickness and wave speed are incorrect To adjust one or more parameters click the text box labeled 9 Composite Thick Freq Keyboard press Alt 9 Enter the observed 81 composite thickness frequency of 6 3 and press Enter The Layered Plate Parameter Calculations screen appears Figure VIII 6 Analysis and Interpretation The option box on the left presents the user with two choices Opti
128. that the 44 9 kHz signal could not have come from wave reflections within the structure The uniform nature of the oscillations throughout the waveform including prior to the arrival of the R wave and the distinct sharp high frequency response are characteristic of a stray electrical signal which is easily recognized In this case the validity and usefulness of the impact echo test results are not affected by the electrical noise even though it is a prominent part of the recorded signal 94 The Importance ofthe R wave Recognizing valid waveforms often centers on recognizing the features of the initial R wave Ifa clearly defined R wave does not appear at the beginning of the waveform the signal will be identified as a bad signal provided that the Signal Screening system is on In general these signals should be discarded The most distinctive feature of a normal R wave is a relatively deep well caused by the downward displacement of the surface as the R wave passes the transducer The width of this well is a good estimate of the contact time or duration of the impact t Several examples of normal R waves are shown in Figure X 5 The small voltage rise ahead of the sharp drop caused by the Rwave is due to the arrival of the P and S wavefronts which have a higher speed than the R wave Note that the signal shown in c was generated by a shorter duration impact than the signals shown in a and b Ay AN N tt R wave 1
129. the active yellow cursor or click Change Curs Keyboard press Alt C to make the other cursor active on the waveform Only the active cursor can be moved using the mouse or keyboard Learn to use the Cut R Wave command This is useful in removing the R wave when its amplitude is large relative to the remainder of the waveform Example Using the mouse or keyboard place the active cursor on the zero in the waveform at about 330us and click Cut R Wave Keyboard press Alt R That portion of the waveform to the left of the active cursor resulting from the passage of the Rayleigh or surface wave is removed and the frequency content of the remainder is calculated and displayed in the spectrum It is the initial negative trough that mirrors the force time function of the impact The large positive peak that follows is an inertial effect The caption on the button changes to Restore R Clicking this button again restores the full waveform and recalculates the spectrum Click this button several times and observe the changes in the spectrum When the leading high amplitude portion of the waveform associated with the R wave is removed the main peak in the spectrum becomes sharper For an example of useful R wave removal see the discussion of record 25 in Section VIII Learn to use the Cut Waveform command used to remove set equal to zero that portion of the waveform to the right of the active yellow cursor Example 1 Restore
130. the spectrum in light blue with a small yellow cursor that moves in tandem with the cursor on the spectrum The position of the cursor on this scale indicates the per cent depth Placing the spectrum cursor at 20 kHz for example shows that a peak at this frequency would correspond to a depth of about 25 of the full thickness e Vertical and Horizontal Scales The spectrum is a plot of amplitude vertical scale vs frequency horizontal scale With the vertical scale set at 1 the spectrum is normalized to make the height of the highest peak equal to 80 of the height of the graph The default value of the horizontal scale is 60 kHz There is seldom a need to change the vertical scale If the important frequencies in the spectrum are at low values the horizontal scale can be decreased to 20 to 40 kHz to show more detail e Cursors command button When this button has the focus it is highlighted by a red border and the cursors on both the waveform and spectrum can be moved with the mouse or the keyboard See Moveable Cursors above The Depth Box for Plate Structures 14 Examine the depth box for plate structures When the structure being tested is a plate or plate with overlay a gray rectangle appears to the right of the spectrum representing a cross section of the structure The top of this box represents the impact surface depth 0 and the bottom a depth equal to the full thickness 400mm in this example When the spectrum cursor
131. the text box is highlighted enter 9 and click Compare or press the Enter key The spectrum from record 9 is overlaid on the current spectrum as a dashed green line The response in record 10 is dramatically different from the solid response of record 9 indicating the presence of a flaw Analysis and Interpretation Place the spectrum cursor on the main peak at 12 2 kHz The indicated depth is 153mm shown in the upper right of the spectrum and also by the horizontal cursor in the thickness box at the right of the screen The existence of a single dominant peak in the spectrum indicates that the response is due to multiple P wave reflections in a 153mm thick layer above a wide crack Place the waveform cursors on the third and fourth peaks after the R wave 390 and 472 us after the trigger point This gives an approximate period of 82 us corresponding to a frequency of 12 2 kHz in agreement with the frequency of the main peak in the spectrum Similar frequency values are obtained from the period measured between adjacent peaks near the end of the waveform The vertical scale of the waveform can be set to a lower voltage 0 2 volts for example to make the structure of the latter part of the waveform easier to see 54 The difference between the solid response and the response when a crack of wide lateral extent is present is illustrated in Figure V 2 ora Time us Amplitude Amplitude Frequency kHz Frequency kHz Figure
132. thickness of each layer is known and the wave speed in one layer is uncertain this method can be used to calculate the true wave speed The point to remember is that in order to proceed with certainty three of the four adjustable variables thickness and wave speeds must be known to obtain a reliable estimate for the fourth variable w Layered Plate Parameter Calculations Calculating wave speeds and thicknesses from impact echo measurements for plates with overlays Overlay Thickness mm 3 Two frequency peaks are observed The overlay and plate thicknesses O Overlay Wave Sp m s 4 are known but the wave speeds are ia uncertain Enter the observed frequencies and press Calculate O Ovelay Thk Freg kHz 5 12 7 to determine the wave speeds O Plate Thickness mm 6 One frequency peak is observed Option2 full thickness Change any one of O Plate Wave Sp m s Zz the variables and highlight a second one to be calculated Press Calculate Composite Thk Freq kHz 8 6 3 Cancel Calculate Accept Change wave speeds for all Plate Overlay records in this file to the values shown on this screen Change Wave Speeds Figure VIII 6 The Layered Plate Parameter Calculations screen 82 8 10 11 12 Use the mouse to place the spectrum cursor at frequencies between the two thickness lines at 6 3 and 12 7 kHz and to the right of the 12 7 kHz line and observe the level of the horizonta
133. tion of the flaws Stress Waves Two types of elastic waves propagate within solids 1 dilatational waves called primary waves or P waves and 2 distortional waves called secondary waves or Swaves A third type of elastic waves known as Rayleigh waves or R waves propagates along the surface of a solid The impact echo method is based mainly on the effects produced by P waves and R waves See Sansalone and Streett Chapter 3 pp 29 46 Impact Echo Field Test System Figure F2 is a photograph of a typical impact echo field test system It has five main components e A hand held transducer unit cylindrical unit on the left that produces a voltage signal in response to surface 000 Figure 1 2 A typical impact echo test system displacements caused by reflected stress waves e A set of small hardened steel spheres in front of computer called impactors for producing impact generated stress waves e A high speed analog to digital data acquisition system left of computer that receives and digitizes the signal from the transducer and transfers it to the computer memory e A notebook computer that receives stores and processes the digitized signal from the data acquisition system and displays the results in numerical and graphical form e A software program Impact E that monitors and controls each test and processes the data to produce output displays that provide information about the structure being tested
134. tioning 68 Cursors command button 33 Cut R Wave command 31 Cut Waveform command 31 Cutting the R Wave 81 Data Acquisition Parameters 99 Changing Parameters Using File Utilities System 38 Clip Level 102 Frequency Resolution 100 Time Between Samples 96 100 Trigger Voltage 101 Voltage 100 data acquisition system 10 Data File Open a new data file 18 Opening an existing file 26 Date Time 28 Depth Box for Plate Structures 33 Description text box 34 Desktop area See footnote 2 p 12 Destination File 35 39 Digital Filtering 60 Directories and Subdirectories 17 Directory box 26 Displacements Surface 29 See Examine Test Data screen 26 EXCEL Spreadsheet 39 Field Test System 10 File Name box 26 File Utilities System 37 Change Parameters for Records 38 Copying Records Between Files 39 Print Summaries of Records 37 Saving Records to ASCII Files 39 Filter Low Fre q command 35 61 Filtering low frequencies 74 Fixed Labels 28 Flaws Delaminations 55 Limited lateral extent 54 Shallow 55 Wide lateral extent 53 Flexural Vibrations 56 72 Frequency Filtering low frequencies 35 74 Thickness Frequency 47 Frequency Content 57 Frequency Resolution 32 100 Frequency Spectrum 31 Help System 15 Hide Record List command 27 honeycombing 61 Horizontal Scale 30 33 ImpactDemo 8 14 Installing the Software 14 Impact E 8 10 16 20 26 30 39 90 100 impact echo 8 Impact echo explanation
135. transferred to the computer Clip Level Clip level specifies a voltage beyond which signals are cut off and ignored Itis used primarily to cut or clip very large R waves that would otherwise dominate the signal and obscure other features Examples of the use of the clip level are given in Section VII of this manual For a more detailed discussion of the use of clip level including examples see pages 314 15 in Sansalone and Streett Cancel and Accept Buttons Activating either of these buttons returns control to the Main Menu screen If changes have been made in the data acquisition parameters they are not saved if the Cancel button is activated 104 Appendix B ASTM Standard C 1383 98a Measuring P Wave Speed and Plate Thickness Using the Impact Echo Method 105 106 107 108 109 110 111 112 113 114 115 116 Index Alt key 27 Asphalt Overlay on concrete 76 bad signals 16 Begin Impact Echo Test command 20 Begin Testing screen 20 Calculate Crack Depth command 68 Change Cursor command 31 Clip Level 30 75 81 102 Comm Port 16 Command Buttons 27 communications port 16 Compare command 34 53 Comparing Records 34 Composite thickness frequency 79 Concrete With asphalt overlay 76 Contact Time 30 57 93 Save Contact Time 58 Copying Records 39 Cp plate 32 Crack Depth Measurement 65 Cracks 52 Surface opening 64 Critical Parameters 29 Cursor to Cursor 29 Cursors 28 Posi
136. ture When tests are carried out to locate flaws in a plate structure the first step is to determine the response of the solid structure This is accomplished by performing tests in a region where the structure is known to be solid If the thickness and wave speed are known the thickness frequency can be calculated and tests can be performed until a solid response is obtained In this section test records for a 240mm thick plate structure will be examined 1 Open the existing data file c ImpactDemo TestFile dat and open record Record 9 shows the results of a test on a solid portion of a 240mm thick plate using an estimated wave speed of 4000 m s The expected thickness frequency for this wave speed is 8 3 KHz text box at upper right of screen However the observed frequency the main peak in the spectrum is 7 8 kHz The presence of a single dominant peak with a frequency near the expected thickness frequency indicates that this is a solid response The observed thickness frequency of 7 8 kHz can be used to calculate the correct wave speed 53 2 Click the 6 Thickness Frequency text box in the upper right of the screen enter the observed frequency of 7 8 kHz and press Enter In the box that appears select P Wave Speed as the parameter to calculate and click Calculate The correct plate P wave speed of 3744 m s appears in the box at the upper right and the vertical blue line that marks the thickness frequency on the spectrum is mov
137. tween files and copy test records to ASCII files or directly to Excel Spreadsheet The File Utilities screen is illustrated in Figure III 3 37 The current file is c Impact E TestFile dat it contains 45 records Save summary information for records BD to 45 trom current file Send to Printer Save to Excel Spreadsheet Change parameters for records to in current file Select Structure Type and Parameters Copy record to from current file to destination file Open Destination File Copy Records Save Records to from current file as ASCII Files Save As ASCII Files Save Records from current file to Excel SpreadSheet Save to Excel Spreadhseet Cancel Help Exit Figure III 3 The File Utilities screen The name of the current file which can be acted upon by the commands on this screen is shown at the top of the screen Sending Summary Information for Records to Printer or Excel Spreadsheet Activating the Save to Excel Spreadsheet button will open another screen that allows a summary of all records in the current file to be sent to an Excel Spreadsheet Microsoft Excel software must be installed on the computer The summary is in the form of a table that includes the following information for each of the records in the range covered by the numbers in the text boxes above the button record number test point name structure type plate plate overlay circular column
138. ures X 7 a and b 2 On the waveform place the active cursor at the right side of the R wave at 242 us and click Cut R Wave The resulting changes are similar to those shown in Figures X 7 c and d Removal of the R wave removes the broad dome of high frequencies centered at about 35 kHz in b caused by the abnormal shape of the R wave Voltage Amplitude 10 20 30 Frequency kHz o 2 a E lt Voltage Time us Figure X 7 Response from a test with a separated R wave a and b the waveform and spectrum with the R wave included c and d the waveform and spectrum with the R wave removed It is sometimes useful to remove cut an R wave that has a normal shape if its amplitude is much greater than that of the periodic signal that follows because it can introduce frequencies that diminish or obscure the important P wave resonances This often occurs for example in tests on concrete with asphalt overlays particularly if testing is carried out on a warm day A waveform and spectrum from a test on a concrete bridge deck with an asphalt overlay is shown in Figures X 8 a and b In this case the large amplitude Rwave response consists of both the initial well caused by the passage of the R wave and immediately following an overshoot response due to inertial effects in the transducer The large amplitude R wave has a significant effect on the spectrum The waveform and spectrum with the R wave removed are
139. usessnenensensnenensenensnensensnenensenene 44 INTRODUCTION a cita 45 ASTM STANDARD FOR MEASURING P W AVE SPEED AND PLATE THICKNESS USING IMPACT ECHO 45 SETUP FOR DIRECT MEASUREMENT OF W AVE SPEED USING TWO TRANSDUCERS MEASURING THE TRAVEL TIME OF A P W AVE BETWEEN TWO TRANSDUCERS unse P W AVE SPEED FROM IMPACT ECHO TESTS ON A PLATE OF KNOWN THICKNESS MEASURING PLATE THICKNESS WHEN THE P W AVE SPEED IS KNOWN ccsssessescesseseesecsscsecsseseeseeseesaeenees SECTION V DETECTING CRACKS AND VOIDS IN PLATES usssssnsnsosesensnssnenenensesenenensenensnsnsnssnenensnsessnenene D 2 INTRODUCTION salia 53 CRACKS AND VOD des E Ki 53 DETERMINE THE SOLID RESPONSE OF THE PLATE STRUCTURE e cesescssssseseeesesecseecneceesecaeeeesecaeeeeaeeeeeeaeess FEAWS OF WIDE LATERAL EXTENT un essen aan een erde FLAWS OF LIMITED LATERAL EXTENT Dirion a n T RE EER E N RA SHALLOW FLAWS AND DELAMINATIONS A SPECIAL CASE nenenenenenensensensensenee USING THE R W AVE TO DETERMINE CONTACT TIME AND FREQUENCY CONTENT USING A SMALLER IMPACTOR TO AMPLIFY HIGH FREQUENCY COMPONENTS REMOVING LOW FREQUENCIES BY DIGITAL FILTERING ccscsssssssssessesssssesssssssesnssscsnessessesseseeseeseaseaneaneaneaneesss UNCONSOLIDATED CONCRETE 85 u a ausweisen ia SECTION VI DETERMINING THE DEPTH OF SURFACE OPENING CRACKS cusensensensenseneonesnersersensenne 65 INTRODUCTION A EEAO SETUP FOR CRACK DEPTH MEASUREMENT MEASURIN
140. using the Waveform and Spectrum commands e To print a consecutive group of records from the open file click the From and To text boxes Keyboard press AIt F and AIt 0 and enter the beginning and ending numbers of the records to be printed Click Print Multiple Records Keyboard press Alt M to initiate printing The waveform and spectrum for each record will be printed each record on a separate page e To add text to a graph displayed on the screen click the text box beneath the caption Enter Text and drag to graph Keyboard press Alt T and enter the desired text in the box Place the mouse pointer anywhere in the box and hold down the left mouse button while the box is dragged to the graph above Ignore the box outline When the mouse button is released the text will be placed on the graph with the lower left corner of the text at the tip of the mouse pointer If you make a mistake click Waveform or Spectrum again and start over Text added to the graphs will appear in printed copies e Click Return Keyboard press Alt R to return to the Main Menu screen when printing is completed The File Utilities System The File Utilities screen is accessed by activating the File Utilities button on the Main Menu screen It permits the user to send summaries of sets of test records to an Excel Spreadsheet or to a printer change parameters such as wave speed or thickness for sets of test records copy records be
141. ve to Excel Spreadsheet in the block at the top of the screen to open the Excel Spreadsheet screen The path and name of the Excel file where the information will be saved appears in a text box on the screen To save the information 40 to a different Excel file enter a path and file name in the text box and press Enter An Excel file should have the suffix xls Click Save Summary of Records to Excel Spreadsheet A summary for each record including the Record Number Record Name Structure Type Wave Speed Thickness and Description is written to the designated Excel spreadsheet which is created if it does not already exist The data file is identified at the top of the spreadsheet by the path and file name the structure name if any and the date on which the file was created You can open the spreadsheet to view it without shutting down the Impact E program For test records that are wave speed measurements the wave speed calculated from the test results appears in the summary For tests on plates with overlay such as concrete with asphalt overlay a second wavespeed and thickness are given for the overlay For measurements of the depth of surface opening cracks the crack depth calculated from the test results appears in the summary Saving waveforms and spectra On the File Utilities screen click Save to Excel Spreadsheet to open the Excel Spreadsheet screen Figure III 6 On this screen enter the beginning and ending record num
142. voltage drop caused by the arrival of a P wave diffracted from the crack tip a tension wave Position the Cursors Position the cursor on the upper waveform to mark point at which the voltage first starts to rise This is done using the mouse or the Ll and keys to move the cursor The label on each cursor is in the format ttt t x xxxxv where ttt t is the time in us after the first recorded data point and X xxxx is the signal voltage at that instant Start with the cursor to the left of the first voltage rise and move it slowly to the right using the key In the flat portion of the waveform the voltage oscillates between 0 0005 and 0 0007 volts due to low level noise in the system The first significant increase in voltage occurs where the label on the cursor reads 84 0 0 0019v indicating that at 84 us after the first recorded data point the direct P wave reached transducer 1 Similarly the cursor on the lower waveform is moved using the lt and gt keys In the flat or leading portion of the waveform the voltage oscillates between 0 0009 and 0 0003 volts The first indication of a significant drop in voltage occurs when the label on the cursor reads 155 0 0 0021v This indicates that at 155 us after the first recorded data point there was a sudden drop in the voltage recorded by transducer 2 signaling the arrival of the tension wave diffracted from the bottom edge of the crack The elapsed time betwe

Impact-Echo User`s Manual

Contents

Download Pdf Manuals

Related Search

Related Contents