SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

SYSTEM REFERENCE MANUAL 2008

SONIC ECHO/IMPULSE RESPONSE TEST NDE-360

NDE-360 Platform with WinTFS Software Version 2.3

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

NOTICES

Errors and Omissions

This document is believed to be accurate, but Olson Instruments, Inc. will not be responsible for errors or omissions, which may be found. Further, Olson Instruments will not be responsible for any damages resulting from any errors or omissions.

Proprietary Information

This document, as well as all software written by Olson Instruments, Inc., is proprietary to Olson Instruments, Inc. This manual may not be sold, reproduced, or used with any other product other than the Olson Instruments Sonic Echo/Impulse Response system unless approved by Olson Instruments, Inc. Further, this manual and the accompanying software may not be used by any party other than the original purchaser without prior approval of Olson Instruments, Inc.

Warranty (See Sales Contract Documents)

Copyright

Copyright 2008 by Olson Instruments, Inc. All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage or retrieval system, without prior written permission of the above named copyright holder.

Written by:

Olson Instruments, Inc. 12401 W. 49th Avenue Wheat Ridge, Colorado, USA 80033-1927 Ofc: 303/423-1212 Fax: 303/423-6071 E-Mail: [email protected] Revised: June 2008

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

1.0 INTRODUCTION

The Olson Instruments Sonic Echo/Impulse Response (SE/IR) system is commonly used for quality assurance, condition evaluation, and forensic testing of piles and deep foundations. Common applications of the SE/IR system are in determining the length of the foundations and/or locating defects within the foundation. The basic requirement for any foundations to be testable by this system is that an area of approximately 3 in by 5 in access be available on the foundation top, and that the surface to be tested be relatively smooth and flat concrete. When used for quality assurance, forensic testing, and condition evaluation of deep foundations, the SE/IR test with the NDE-360 system is a powerful, easy to handle and operate tool in verifying foundation integrity, locating defects and determining length of the foundation. The SE/IR test with the NDE-360 system allows the user to collect data easily and fast by a single operator with the light handheld device platform and a touch screen feature to operate the system.

The NDE-360 system with the SE/IR test consists of several basic components. These components include the 4 channel NDE-360 platform for data acquisition, analysis and display, an accelerometer, an instrumented 3 lb hammer, grease and cables. Details of the hardware and its usage are included in Section 2.0.

The Windows WinTFS software is a post data analysis program with a number of specialized SE/IR analysis tools built-in. This manual covers step-by-step hardware assembly, software setup, data acquisition, data analysis, and output generation.

1.1 Organization and Scope of Manual

This operation manual for the SE/IR test with the NDE-360 system manufactured by Olson Instruments includes all required instruction for the use of the hardware and software included with the system. Also included at the end of the manual is a troubleshooting guide for the system to help overcome any problems experienced or answer any questions. If any problems with the system appear that are not covered in this manual, please call Olson Instruments at the number included in the front of this manual. Note that training in the use of the system by Olson Instruments personnel is recommended for the most effective operation of this system.

1.2 Test Methodology

Sonic Echo (SE) – The SE method is a low strain integrity test conducted from the top of the shaft as shown in the Figure in the following page. Test equipment included a 3 lb hammer (instrumented or mechanical hammer), receiver (accelerometer and/or geophone) mounted on the top of the shaft, and a data acquisition platform (in this case, NDE-360 system). The test involves hitting the foundation top with the hammer to generate wave energy that travels to the bottom of the foundation. The wave reflects off irregularities (cracks, necks, bulbs, soil intrusions, voids, etc.) and/or the bottom of the foundation and travels back along the foundation to the top. The receiver measures the vibration response of the foundation to each impact. The NDE-360 system acquires, processes and displays the hammer and receiver outputs. Foundation length and integrity of concrete are evaluated by identifying and analyzing

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

the arrival times, direction, and amplitude of reflections measured by the receivers in time. The echo depth (D) is calculated by multiplying the reflection time (t) by the compression wave velocity (V) and dividing this quantity by 2 to account for the fact that the wave has gone down and reflected back, i.e., D = V*t/2 .

Analysis of the length determination and the integrity evaluation of a foundation for the SE method is based on the identification and evaluation of reflections. Test results are analyzed in the time domain for the SE test.

The SE test method is sensitive to changes in the shaft impedance (shaft concrete area * velocity * mass density where mass density equals unit weight divided by gravity), which cause the reflections of the compression wave energy. Compression wave energy (hammer impact energy) reflects differently from increased shaft impedance than from decreased shaft impedance. This phenomenon allows the type of reflector to be identified as follows. Soil intrusions, honeycomb, breaks, cracks, cold joints, poor quality concrete and similar defects (referred to herein as a neck) are identified as reflections that correspond to a decrease in the shaft impedance. Increases in the shaft cross-section or the competency of surrounding materials such as bedrock and stiffer soil strata (referred to herein as a bulb) are identified as reflections corresponding to increases in the shaft impedance. A decrease in impedance is indicated by a downward initial break of a reflection event in an SE record and frequency peaks positioned in a record such that a peak could be extrapolated to be near 0 Hz in the mobility plot. Conversely, an increase in shaft impedance is identified by an upward initial break for an SE reflector. When length to diameter ratios exceed 20:1 to 30:1 for shafts in stiffer soils/bedrock, the attenuation of compression wave energy is high and bottom echoes are weak or unidentifiable in SE test results.

Hammer Receiver Hammer Receiver Hammer Receiver

Reflection Reflection from Lower from Higher Impedance Impedance

Defects and Cracks and Bulbs and Length Intrusions Breaks Determination

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

Impulse Response (IR) - The IR method is also an echo test and uses the same test equipment as the SE method. The test procedures are similar to the SE test procedures, but the data processing is different. The IR method involves frequency domain data processing, i.e., the vibrations of the foundation measured by the receivers are processed with Fast Fourier Transform (FFT) algorithms to generate transfer functions for analyses. The coherence of the impulse hammer impact and accelerometer receiver response data versus frequency is calculated to indicate the data quality. A coherence near 1.0 indicates good quality data. For shafts in air or in relatively soft soils, the coherence will typically only be near 1.0 at frequencies for which the mobility is non-zero. In the IR records the linear transfer function amplitude is in inches/second/pound force on the vertical axis (mobility) and frequency in Hz on the horizontal axis. Because of the rod-like shape of a deep foundation, reflections are indicated by equally spaced resonant peaks that correspond to modes of vibration associated with the depth of the reflector. The inverse of the SE reflection time, t, is equal to the change in frequency, delta f, between the resonant peaks in the IR mobility plot. The reflector depth is then calculated as: D = V/(2*∆f).

SE/IR Analyses - Analysis of the length determination and the integrity evaluation of a foundation for both the SE and IR methods is based on the identification and evaluation of reflections. However, test results are analyzed in the time domain for the SE and in the frequency domain for the IR method. The reflections are shown as resonant frequency peaks in the frequency domain for IR test data. The two methods complement each other because the identifications of reflections are sometimes clearer in either the time or the frequency domain.

The SE and IR test methods are sensitive to changes in the shaft impedance (shaft concrete area * velocity * mass density where mass density equals unit weight divided by gravity), which cause the reflections of the compression wave energy. Compression wave energy (hammer impact energy) reflects differently from increased shaft impedance than from decreased shaft impedance. This phenomenon allows the type of reflector to be identified as follows. Soil intrusions, honeycomb, breaks, cracks, cold joints, poor quality concrete and similar defects (referred to herein as a neck) are identified as reflections that correspond to a decrease in the shaft impedance. Increases in the shaft cross-section or the competency of surrounding materials such as bedrock and stiffer soil strata (referred to herein as a bulb) are identified as reflections corresponding to increases in the shaft impedance. A decrease in impedance is indicated by a downward initial break of a reflection event in an SE record and frequency peaks positioned in a record such that a peak could be extrapolated to be near 0 Hz in the mobility plot. Conversely, an increase in shaft impedance is identified by an upward initial break for an SE reflector and frequency peaks positioned in an IR record such that a trough could be extrapolated to be near 0 Hz in the mobility plot. When length to diameter ratios exceed 20:1 to 30:1 for shafts in stiffer soils/bedrock, the attenuation of compression wave energy is high and bottom echoes are weak or unidentifiable in SE/IR test results. The term “weak bottom echo” is used to indicate that a bottom echo is seen, but that it is barely visible above the normal background noise. Note that it is possible for a perfectly sound shaft to have a weak bottom echo, as a weak echo can be caused by either a shallow reflector blocking energy from the shaft bottom, or from wave energy coupling into the bedrock at the shaft bottom rather than reflecting back up. Thus, the strength of the bottom echo is used as a secondary, minor consideration as to shaft condition.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

2.0 HARDWARE

The SE/IR system with the NDE-360 platform consists of several basic components packaged into a padded carrying case. The padded case stores the NDE-360 data acquisition platform, power supply, battery charger, accelerometer, instrumented hammer, grease, and cables. A description of each of these components as well as their connection and operation is included in the following sections.

2.1 Hardware Component Listing

Component Name QTY Description

1 Platform for data acquisition, analysis and display of SE/IR time domain data

1) Olson Instruments NDE-360 Platform

1 To receive the SE/IR acceleration time \ domain data (to be mounted on the foundation top)

2) Accelerometer

1 An impact source instrumnted3 lb hammer

3) Instrumented Hammer

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

Component Name QTY Description

1 4 heads for the instrumented hammer (each head has different hardness)

4) Heads for Instrumented Hammer

1 To receive the SE/IR velocity time domain data (to be mounted on the foundation top)

5) Geophone

2 BNC Extension Cables for Accelerometer and Instrumented Hammer

5) BNC Cable

1 Female to Female BNC adaptor

6) BNC Adapter

2 Connect a BNC cable and a 4 pin input channel from the NDE-360 system

7) BNC to 4 Pin Adaptor Cable

1 To connect the 3 Pin to Phone Plug Cable to the channel on the amplifier on the Freedom Data PC

8) Phone Plug to 4 Pin Adaptor

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

1 To connect the accelerometer and a BNC Cable

9) Microdot to BNC Cable

1 Power supply for the NDE-360 platform.

10) Power Supply

1 Battery charger must be connected to the power supply on one end and the NDE-360 on the other end

11) Charger

1 To transfer the SE/IR data files to the analysis computer

12) Card Reader and Compact Flash Card

1 Coupling agent between the accelerometer and foundation top

13) Coupling Grease

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

2.2 Basic Components and Maintenance of NDE-360 Platform

Battery Charging: The power supply (Item 8) must be plugged into the charger (Item 9) only. Then the charger (Item 9) must be connected to the NDE-360 system.

Power Supply

Charger

NDE-360 System

Time required for a fully charge of battery is 8 hours. Fully charged battery will power the NDE-360 system for 8 – 10 hours.

Location of Battery: The panel housing the battery is located on the back of the NDE-360 system. Move the switch to remove the battery cover if battery needs to be replaced.

Battery Panel

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

Do not expose the NDE-360 system or the battery to water.

Channel Arrangement: The NDE-360 system is equipped with 4 channels. There are two 4 pin connectors per each channel. Channel 1 is located on the left hand side of the NDE-360 system (it is typically located near the Impact Echo channel which is not used in the SE/IR test). The top row of channels (the row toward the front screen) includes channels with highpass filter. The bottom row of channels (the row toward the back side) includes wideband channels which are typical channels used in the SE/IR test.

IE/SASW Channel – Not used for the SE/IR Test

Chs used for the SE/IR tests

Pulser for the UPV transducer (Source) – Not used for the SE/IR test

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

Data Storage: A compact flash card is used to store the SE/IR data files. In addition, the master program controlling the SE/IR test is stored in the compact flash card. The system must start up with the compact flash secured in place. The card should be located on the left side of the NDE-360 system if looking at the front screen.

Compact Flash Card

If the compact flash card is missing from the system, the system will not allow the test to continue. Do not remove the compact flash card while the system is turned on.

NDE-360 Operation Notes: The NDE-360 system is a self-contained data conditioning, collection, basic processing and data display platform usable for a number of types of NDT tests. The system was designed to make it user friendly and easy to handle and operate by a single operator. The main screen can be navigated using a touch screen feature.

Never impact the screen with any sharp object!

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

2.3 Hardware Setup

1. Connect the microdot cable to the accelerometer by aligning the pin on the microdot cable with the corresponding hole in the accelerometer and then hand-tighten the connection

2. If needed, attach the female BNC adaptor to the BNC end if the microdot cable by aligning the pin with the hole and gently connecting the two pieces. This connection will have a locking mechanism that will prevent over-tightening the connection

3. Next, if needed, attach the other end of the female BNC adaptor to a BNC cable. Follow the same procedure as before when connecting these two cables.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

4. Now attach the BNC cable to the 4 pin adaptor cable using the same procedure as before when connecting these two cables.

5. Insert the 4 pin adaptor cable into a wideband channel (located in a row of channels close to the back side of the unit). The default channel used by the transducers (accelerometer or geophone) the software for SE/IR is Channels 2 and 1.

6. Put a small amount of coupling grease on the base of the accelerometer and then mount the accelerometer on a smooth area on the foundation top. Note that the microdot cable is not shown in this picture.

Accelerometer

Coupling Grease

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

7. For the SE/IR test, an instrumented 3 lb hammer is typically used. Screw in the impact head (Item 4 in Section 2.1) to the instrumented hammer (Item 3). The most common used head for the SE/IR test is the black head. The red tip is typically used for very long foundations (greater than about 90 feet or 30 m).

8. Connect the BNC Cable (Item 5) to the BNC connector on the end of the instrumented hammer

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

9. Connect the other end of the BNC cable to the BNC to 4 pin adaptor cable (Item 7)

10. Connect the 4 pin adaptor end of the BNC to 4 pin adaptor cable to Channel 4 (wideband) on the NDE-360 platform

Hammer connecting to Channel 4 (wideband)

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

11. Connect the Phone Plug to 4 Pin Adaptor (Item 8) to the geophone (Item 5)

12. Connect the 4 pin adaptors of the extension cables to the wideband channels (in this case, channels 1, 2 or 4) on the NDE-360 platform. By default, the geophone should be connected to Channel1, the accelerometer to Channel 2 and the instrumented hammer to Channel 4.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

Connecting to Connecting to an an Instrumented Hammer accelerometer

Connecting to a Geophone

13. The picture below shows a complete hardware setup for the SE/IR test. Note that the BNC extension cable is not shown in the picture.

Geophone

Instrumented Accelerometer Hammer

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

3.0 SE/IR DATA ACQUISITION

1. Turn on the NDE-360 system (with the hardware in Section 2.0 connected). Make sure that the compact flash card is secured in place. The master software to run the system is located in the compact flash card. Press on the “Continue” button on the upper right corner of the screen. If the compact flash card is not in place, the “Continue” button will not appear on the screen.

Press on this button to continue

Press on this button to turn on the system

If the continue button disappears from the screen, turn off the system and insert the compact flash card in place and turn the system on again.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

2. Press on the “SE/IR” option. The option included in the unit may be different depending on the order. The picture below shows the screen with SE (Sonic Echo), IE (Impact Echo), SASW, Slab IR (Slab Impulse Response), SE/IR (Sonic Echo/Impulse Response) and UPV (Ultra Sonic Pulse Velocity) options.

3. Press on the “Param” button to setup the data parameter. Note that the unit is loaded up with the default parameters which are appropriate for most of the SE/IR tests.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

4. In the parameter setup screen, the user can change the following parameters by touching the button to toggle the value. Touch the “Back” button after the parameter setup is complete.

• Change Date/Time. This option allows the user to enter the correct date and time of testing • Time/Point or Sampling Rate means how often (in time domain) the system will acquire data points within a given data trace. In the case shown in the picture below, the Time/Point was set at 20 microseconds. This means the system will acquire data at 20 microsecond intervals. This parameter can not be changed after the data was taken. • Points Per Record is number of sampling points for each waveform. The higher this value, the more data is acquired in each waveform. The total time of a record is affected by both the Sample Rate and the Record Size. This parameter can not be changed after the data was taken. • Velocity. The software uses this velocity to calculate the echo depth. If velocity calibration is not possible, a typical concrete velocity of 12,000 ft/sec may be used as a default. This parameter can be changed later in the post data analysis. • # Recs. Number of Record is a total number of SE data records you want to save. This parameter can not be changed after the data was taken. • Pre-Trigger is number of points before the triggering point that data collection starts. In this case, the pre-trigger is 100 points (or 100 x 20 us = 2000 us). This parameter can not be changed after the data was taken. • Trigger Level % is the minimum signal amplitude (in terms of percentage full scale) to trigger data acquisition. In this case, the trigger level was set at 6%. Therefore, the system will start acquiring data once the absolute amplitude of the signal exceeds 0.6 volt. This parameter can not be changed after the data was taken. • Channel Setup. For the SE/IR test, three channels (connected to the accelerometer, geophone and instrumented hammer) will be used. Any of the 4 wideband channels can be used for the SE/IR test. However, the default parameter is loaded with Ch 1 for Geophone (Velocity), Ch 2 for an accelerometer (Accel) and Ch 4 for an instrumented hammer (Force). Typically the channel connecting to the instrumented hammer will be used as a trigger channel and SREF (reference) channel.

Type of Transducer

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

5. After the channel setup is complete in the parameter setup menu (Step 4) press the “Back” button to return. Next, the gain control for the channel should be set by touching the + button to increase the gain value and the – button to reduce the gain value. It is recommended that a gain of 10 - 100 is initially set for the accelerometer and velocity channels and 1 for the instrumented hammer channel.

Initial gain of 10 for both velocity and accelerometer

channels

To setup the prefix part of the filenames

6. The filename arrangement of the NDE-360 uses a fixed prefix of letters and a suffix of number. The prefix can be set at the beginning of any test. If the prefix is not changed, a higher suffix number will be added to the current prefix for the new filename. To change the prefix, press on the “Files” button and press number 5 and then number 1. Then a virtual keyboard will appear on the screen and the user can change the prefix part of the filename. Press the “A” button on either side to accept the new prefix of the filename.

Accept Button

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

7. Now, the system is ready to acquire the SE data. Press on the “Test” button to start testing.

8. The NDE-360 system will then wait for the data. Note that the voltage of the SE/IR time domain data has to exceed the trigger level in order for the system to start taking data. Hit the hammer on the foundation top next to the accelerometer and geophone.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

9. If the system does not receive data for 1 minute, the test will be canceled and the user can press any key to go back to the previous menu. This happens if the gain value is too low. Try either increasing the gain value of the instrumented hammer channel or decreasing the trigger level and re-test the file. If the system receives the SE/IR data, it will display the SE/IR time domain data on the screen with the percentage full scale of the maximum amplitude shown below the plot. The desirable percentage full scale is between 5 – 90 %. The picture below shows a percentage full scale of 58% and it means the maximum amplitude of the SE data is 5.8 volt. At this point, press the “A” button to accept the data or the “R” button to reject the data currently displayed on the screen. In this test, the “T” button (to switch to the next channel) is irrelevant for the SE test with only one channel.

Percentage full scale of the maximum amplitude of the signal

10. Repeat Steps 8 and 9 until all the records are collected. 11. To perform more SE/IR tests, repeat Steps 7 – 10.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

12. Basic analysis tools included in the NDE-360 system are filter and exponential functions. To set the filter, press on the “Select Filter” button and enter the cut off frequency for the low pass filter (typically 1,000 – 2,000 Hz)

To setup the filter

Then press on the Filter Data button to apply to current filter setup to the waveform.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

13. The data can be recalled in the NDE-360 system by pressing the “File” button on the main screen and then select Option 2. 14. The data files on the compact flash card can be moved to the analysis computer for post-data analysis.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

4.0 WinTFS Software

This section covers step-by-step instructions for post data analysis and output generation for the Windows WinTFS software.

4.1 WinTFS Software Installation

1. Prerequisite software (NIDAQ version 7.4 or higher) prior to installing the WinTFS software is required. The NIDAQ drivers can be located in the enclosed CD or downloaded at:

http://joule.ni.com/nidu/cds/view/p/id/319/lang/en

If retrieving files from the web, download the file named NIDAQmx8.1.ZIP. Unzip the file and install the NIDAQ program. Note that an account (free of charge) may be required to proceed to the download page.

Failing to install the prerequisites will result in an error when running the WinTFS software

2. Uninstall the previous version or delete “c:\program files\olsoninstruments\WinTFS\WinTFS.exe” if the older file exists 3. Run “Setup.exe” from the Olson Instruments install CD 4. Follow the default setup 5. After finishing the installation, the “WinTFS.exe” file will be found on: drive C:\Program Files\Olson Instruments\WinTFS\ . The shortcut to the WinTFS.exe can be located on the desktop/

Failing to uninstall the previous version of the WinTFS software will prevent the installation of the new version

4.2 Software Updates

For updates to software, the only file that is necessary is the WinTFS.exe file. This file must be copied into the C:\Program Files\Olson Instruments\WinTFS directory. If not, the shortcut on the desktop references the old version of WinTFS. Simply replace the existing WinTFS.exe by copying and pasting the new version into the directory.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

4.3 Software Uninstallation

The followings are steps to uninstall the WinTFS Software 1. Click on Start/Settings/Control Panel 2. Select “Add/Remove Programs” 3. Highlight WinTFS, select “Remove” 4. The uninstall process will begin automatically, removing all installed components including the shortcut.

4.4 First Time Executing the WinTFS Software

If the WinTFS software is executed for the first time, the program detects missing software key (OlsonWinTFSKey.dat). The program will display a warning that the software key file is missing and will only enable two options on the main menu. Select the “Software Key” button and then enter the software key attached to the case of the installation disk. Exit the software and restart the software.

Now the software will detect missing parameter file (SEIRdefault.prm is missing) and then will automatically generate the default file. The user will notice an applet shown below for the type of acquisition card. Select “No data Acquisition Card” option to continue.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

4.5 SE Data Analysis

1. Read the Data file: To read the data file, go to File/Read Data from NDE-360 and select the filename to analyze.

2. Accept or Reject Data: The SE time domain raw data will appear on the top trace of the plot. The second trace shows a velocity trance. For acceleration data, the software integrates the data to obtain a velocity trace and presents it on the second trace. For data from the geophone, the top and second traces are identical. The third trace (not shown in this picture) is an average of data from the second trace (velocity trace). Click on “Accept” or “Reject” button to either include or exclude the current record on the screen.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

3. Filter Function: In most of the cases, filtering is required for the SE data. Go to Analysis/Digital Filter Function or press F4 to set the desired filter. The dialog box below will appear on your screen. There are four options for digital filtering; Butterworth, Chevbyshev, Elliptic and Inverse Chevbyshev. Digital filter will be applied to all the data. To disable filtering, simply leave all the filter options unclicked.

Common filter used for theFigure SE data 5.4 -is Digital Butterworth Filter

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

4. Concrete Velocity: The next important aspect in the SE analysis is the velocity setting. To access the velocity being used to calculate the length of the shaft, simply click on Analysis/Change Velocity. The SE velocity of the tested structure should be set correctly in order for the program to calculate the echo depth of the foundation accurately. The Ultra Sonic Pulse Velocity (UPV) test can be performed across the cross section of the foundation to receive a more accurate concrete velocity.

Typical concrete velocity is 12,000 – 13,000 ft/sec or 3,658 – 3,962 m/s. Higher strength concrete may have higher velocity. Young concrete (less than 7 days old) may have slower velocity.

5. Pick the echo: At this point, the data can be analyzed. SE data is analyzed by picking the troughs in the wave form. This is done by placing the mouse cursor on the lowest point of the trough and left clicking the mouse. This should be repeated on each trough, moving from the left to the right. If the troughs are picked out of order, the depth calculation will be incorrect. To clear the picks, simply right click anywhere on the bottom plot.

Fast data recall

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

6. Open the next SE data File: For speedy data recall, simply click on the +/- button to open the next data file. 7. Data Export: The plots can be sent to MS Word for print out. Go to File/Send Plots to MS Word. Then setup the orientation and figure number and click the OK button.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

4.6 IR Data Analysis

1. With the time domain SE data on screen (if not, do Item 1 in Section 4.5), go to Analysis/IR Analysis

2. An applet in the next figure will appear on the screen. The IR analysis utilizes the same SE time domain data set with additional signal processing to increase the resolution in the frequency domain. This includes padding the time domain data with zero to make the data record longer thus smaller delta F or higher resolution in the frequency domain. Decimating helps reducing the size of the data file to reduce the calculation time of the response. Padding is strongly required to process the IR analysis. However, Decimating is optional. Click OK to proceed to the IR analysis.

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

3. The software will display the IR time domain data set (the SE data set with padding and decimating) on the top trace. Click on “Accept” or Reject to either accept or reject the data on the top trace. Note that the top trace displays the time domain IR data. The second trace displays the average coherence from the accepted data and the bottom trace displays the average mobility

Related Peer Reviewed Papers (2 papers attached)

8. After accepting/Rejecting all the recorded data, pick the peak in the frequency domain. This is done by placing the mouse cursor on the highest peak of the local peak and left clicking the mouse. This should be repeated on each trough, moving from the left to the right. If the peaks are picked out of order, the depth calculation will be incorrect. To clear the picks, simply right click anywhere on the bottom plot.

1st Peak 2nd Peak

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

4.7 Example Data

Two types of echo are expected in the SE/IR analysis: neck and bulb type echoes. The neck type echo represents a decrease in shaft diameter or defect area in the shaft or the toe of the shaft in soft soil/clay. The characteristic of the neck type echo in the SE data is a downward echo in the time domain data. For the IR data, the frequency peaks will be symmetry starting at 0 in the frequency domain.

Avg Time Domain SE Data - Velocity Trace, Depth = 14.2 meter, X = 11.06 ms, Y = -4.92e-005 0.0008

0.0006 0.0004 0.0002 0

-0.0002

-0.0004

-0.0006 Downward echo – neck type echo Representing the shaft toe -0.0008 -0.001 -0.0012 -0.0014 -0.0016

-0.0018

-0.002 Time domain SE data showing a neck type echo -0.0022 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 Time (us)

Freq Domain IR Data, Depth = 14.9 meter 6.5e-008

6e-008 5.5e-008 5e-008

4.5e-008

4e-008 3.5e-008 3e-008

2.5e-008

2e-008 1.5e-008

1e-008

5e-009 Frequency domain IR data showing a neck type echo 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 Frequency (Hz)

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

The bulb echo represents an increase in the shaft diameter or an encounter of hard bedrock. This section illustrates sample SE and IR data for the neck and bulb type echoes. The characteristic of the bulb type echo in the SE data is an upward echo in the time domain data. For the IR data, the frequency peaks will be symmetry starting at negative frequency (not shown in the plot) in the frequency domain.

Avg Time Domain SE Data - Velocity Trace, Depth = -30.1 ft, X = 0. ms, Y = 5.55e-010 0.002 Upward echo – bulb type echo

0.001

0

-0.001

-0.002

-0.003

-0.004

-0.005 Time domain SE data showing a bulbl type echo -0.006 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 Time (us)

Freq Domain IR Data, Depth = 12.7 ft 2.2e-005

2e-005

1.8e-005

1.6e-005

1.4e-005

1.2e-005

1e-005

8e-006

6e-006

4e-006

2e-006 Frequency domain IR data showing a bulb type echo 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 Frequency (Hz)

SONIC ECHO/IMPULSE RESPONSE (SE/IR)

www.olsoninstruments.com www.olsonengineering.com

NDT Diagnosis of Drilled Shaft Foundations by Larry D. Olson, P.E., Principal Engineer Olson Engineering, Inc. 5191 Ward Road, Suite 1 Wheat Ridge, Colorado 80033-1905 Tel: 303/423-1212 Fax: 303/423-6071 E-Mail: [email protected]

Marwan F. Aouad, Ph.D., Project Manager Olson Engineering, Inc. 5191 Ward Road, Suite 1 Wheat Ridge, Colorado 80033-1905 Tel: 303/423-1212 Fax: 303/423-6071 E-Mail: [email protected] and

Dennis A. Sack, Project Manager Olson Engineering, Inc. 5191 Ward Road, Suite 1 Wheat Ridge, Colorado 80033-1905 Tel: 303/423-1212 Fax: 303/423-6071 E-Mail: [email protected]

A paper prepared for presentation at the 1998 Annual Meeting of the Transportation Research Board and for publication in the Transportation Research Record

37

ABSTRACT Number of words = 6590 (including 250 words for each figure)

Nondestructive methods based on propagation of sonic and ultrasonic waves are increasingly being used in the United States and internationally for forensic investigations of existing structures and for quality assurance of new construction. Of particular interest is the quality assurance of newly constructed drilled shaft foundations. A large number of State Departments of Transportation specify NDT testing of drilled shaft foundations, particularly for shafts drilled and placed under “wet” construction conditions.

For quality assurance of drilled shaft foundations of bridges, the Crosshole Sonic Logging (CSL) and Sonic Echo/Impulse Response (SE/IR) methods are routinely used. The CSL method requires access tubes to be installed in the shaft prior to concrete placement. SE/IR measurements require that the top of the shaft be accessible after concrete placement. Discussed in this paper are proper test setups, specifications, and case studies to illustrate the advantages and disadvantages of each of these methods. Presented also are recommendations for repair when a defect is identified in a drilled shaft foundation. Based on our experience, the CSL method is more effective in locating defects than the SE/IR method.

CSL measurements are effective in determining anomalies and defects between two access tubes. However, an accurate image of the defect cannot be determined from just a CSL test. The Crosshole Tomography (CT) method uses multiple CSL logs with varying receiver locations to produce a 2-D image of the defect, and thus a better characterization of the defect. The CT method is briefly discussed in this paper along with presentation of one dataset obtained from a drilled shaft foundation. The CT method requires more time for data collection and analysis than the CSL method, and presently its use is justified only for critical drilled shaft foundations.

Key Words: Crosshole Sonic Logging, Sonic Echo/Impulse Response, Tomography, Quality Assurance, Drilled Shafts.

38

INTRODUCTION Quality assurance of foundations, particularly drilled shaft foundations, is becoming an important part of the foundation installation process to ensure a good foundation that can transfer the applied loads to the surrounding soil or rock. Until the mid 1980's, quality assurance of driven piles and drilled shafts in the USA was performed at selected shafts and used the Sonic Echo and Impulse Response (SE/IR) test methods to identify anomalies or defects (Olson and Thompson, 1985; Davis and Dunn, 1974). The SE/IR method relies on reflection events from a change in impedance. Although the SE/IR method can be applied to identify defects, the method suffers from the following limitations: 1) the strength of the echoes depends on the surrounding soil, 2) echoes are frequently too weak to be distinguished when length to diameter ratios exceed 20:1, 3) the size and location of the defect cannot be determined, 4) defects located below a major defect cannot be identified, 5) defects at or near the bottom of the shaft cannot generally be identified, and 6) planned or unplanned diameter changes can appear to be defects even if the diameter is acceptable.

The drawbacks associated with the SE/IR method have led to the search for other alternatives and the development of the Crosshole Sonic Logging and Gamma-Gamma methods (Olson et. al, 1994) to identify defects in drilled shaft foundations by use of cast-in-place access tubes. One of the advantages of the CSL method over the Gamma-Gamma method is that a fairly complete coverage of the shaft conditions can be determined with the CSL method, while the Gamma-Gamma method determines the shaft conditions around the installed tubes in a drilled shaft. In addition, CSL is generally much faster to perform and does not use radioactive materials.

In this paper, the Crosshole Sonic Logging (CSL), Crosshole Tomography (CT) and Sonic Echo/Impulse Response (SE/IR) methods are discussed along with case studies to illustrate the use of each method. It should be mentioned that the CT method is not routinely applied to drilled shaft foundations, and its application is limited to critical structures to produce a better image of defects identified in CSL and SE/IR tests.

CROSSHOLE SONIC LOGGING METHOD The CSL method was developed in the mid 1980's for quality assurance of drilled shaft foundations, slurry walls and seal footings. The CSL method relies on direct transmission of sonic/ultrasonic waves between access tubes placed in a drilled shaft prior to concrete placement. Figure 1 shows an illustration for a CSL test setup.

The number of access tubes per drilled shaft is dependent on the diameter of the shaft, typically 1 tube per 1 ft of diameter, and the tubes are installed around the perimeter of the shaft and tied to the inside (or outside) of the cage of the shaft. The tubes are usually 38 to 50 mm (1.5 to 2.0 in.) inside diameter schedule 40 steel or PVC pipe. Tube debonding from the surrounding concrete can occur at an earlier time in PVC tube as compared to steel tubes. Most State DOT’s specify that CSL tests be performed in 10 days or less after concrete placement for PVC tubes and in 45 days or less for steel tubes to avoid problems associated with tube debonding.

To perform a CSL test, two probes (hydrophones) are lowered to the bottom of two access tubes, and are retrieved to the top of the shaft while CSL measurements are taken approximately every 50 mm (2 in.). The ultrasonic wave pulser is controlled by a distance wheel to trigger the transmission of waves at preselected vertical intervals. Automatic scanning of the collected records produces two plots, time (or velocity) and energy, versus depth. Anomalies and defects between tested tubes are manifested by time delays (or velocity decreases) and energy drops in the scanned CSL plot. Concrete velocities are calculated by simply dividing the distance between the two tubes by the time required for the wave to 39 travel from the source hydrophone to the receiver hydrophone. CSL tests are typically performed between all perimeter tubes to evaluate the concrete conditions of the outer part of the shaft and between major diagonal tubes to evaluate the concrete conditions of the inner part of the shaft. Figure 2 shows an illustration for the interpretation of a CSL log. NDT methods which could be used in conjunction with the CSL method to better identify anomalous zones include Crosshole Tomography (CT), Singlehole Sonic Logging (SSL), Gamma-Gamma Nuclear Density Logging, Downhole Sonic and/or Sonic Echo/Impulse Response (SE/IR) tests. The CT and SE/IR methods are briefly discussed below.

CROSSHOLE TOMOGRAPHY METHOD The Crosshole Tomography method uses the same equipment as the CSL method with more tests being collected (many source and receiver locations). Once a defect is identified in CSL tests, CT tests can be performed to produce an image of the defect between the test tubes. The CT tests are typically performed at depths extending few feet below and above the defect zone as shown in Figure 3.

Tomography is an inversion procedure that can provide for ultrasonic images of a concrete zone from the observation of transmitted compressional or shear first arrival energy. The CT data is used to obtain an image of the defect. The test region is first discretized into many cells with assumed slowness values (inverse of velocity) and then the time arrivals along the test paths are calculated. The calculated times are compared to the measured travel times and the errors are redistributed along the individual cells using mathematical models. This process is continued until the measured travel times match the assumed travel times within an assumed tolerance. Tomographic analysis was performed using two series expansion algorithms with a curved ray analysis from geotomography. The tomographic analysis presented herein was performed using a SIRT (Simultaneous Iterative reconstruction Technique, Herman 1980) based analysis program developed as part of a research project sponsored by the National Science Foundation to image defects in structural concrete (Olson et al, 1993).

SONIC ECHO / IMPULSE RESPONSE METHODS Sonic Echo Test Method The SE method is a low strain integrity test conducted from the surface. Test equipment includes an impulse hammer (optional, an ordinary plastic tipped hammer) and an accelerometer (or geophone) on the shaft top as shown in Figure 4. The impulse hammer has a built-in load cell that can measure the force and duration of the impact (needed for IR tests). The test involves hitting the foundation top with the hammer to generate wave energy that travels down the foundation. The wave reflects off irregularities and/or the bottom of the foundation and travels up the foundation to the foundation top. The receiver measures the vibration response of the foundation to each impact. The signal analyzer or PC processes and displays the hammer and receiver outputs. Foundation integrity is evaluated by identifying and analyzing the arrival times, direction, and amplitude of reflections measured by the receiver in time. The receiver output is usually integrated (if an accelerometer is used) and exponentially amplified with time (Koten and Middendorp, 1981) to enhance weak reflections. Digital filtering with a low-pass filter of about 2,000 Hz is usually applied to eliminate high frequency noise. In some cases, where reflections are difficult to identify, an impedance imaging procedure is used to obtain a 2-D image of the shaft (Paquet, 1991).

Impulse Response (IR) Test Method The IR method is also an echo test and uses the same test equipment as the SE method. The test procedures are similar to the SE test procedures, but the data processing is different. The IR method involves frequency domain data processing, i.e., the vibrations of the foundation measured by the receiver 40 are processed with Fast Fourier Transform (FFT) algorithms to generate transfer functions for analyses. The coherence of the impulse hammer impact and accelerometer receiver response data versus frequency is calculated to indicate the data quality. A coherence near 1.0 indicates good quality data. In the IR records the linear transfer function amplitude is in velocity/force on the vertical axis (mobility) and frequency in Hz on the horizontal axis.

SE/IR Analyses Analysis of the integrity of a foundation for both the SE and IR methods is based on the identification and evaluation of reflections. However, test results are analyzed in the time domain for the SE and in the frequency domain for the IR method. The reflections are shown as resonant frequency peaks in the frequency domain for IR test data. The two methods complement each other because the identifications of reflections are sometimes clearer in either the time or the frequency domain.

The SE and IR test methods are sensitive to changes in the shaft impedance (shaft concrete area * velocity * mass density where mass density equals unit weight divided by gravity), which cause the reflections of the compression wave energy. Compression wave energy (hammer impact energy) reflects differently from increased shaft impedance than from decreased shaft impedance. This phenomenon allows the type of reflector to be identified as follows. Soil intrusions, honeycomb, breaks, cold joints, poor quality concrete and similar defects (referred to herein as a neck) are identified as reflections that correspond to a decrease in the shaft impedance. Increases in the shaft cross-section or the competency of surrounding materials (referred to herein as a bulb) are identified as reflections corresponding to increases in the shaft impedance. A decrease in impedance is indicated by a downward initial break of a reflection event in an SE record and frequency peaks positioned in a record such that a peak could be extrapolated to be near 0 Hz in the mobility plot. Conversely, an increase in impedance is identified by an upward initial break for an SE reflector and frequency peaks positioned in an IR record such that a trough could be extrapolated to be near 0 Hz in the mobility plot.

CASE STUDIES Discussed below are results from CSL and CT tests performed by Olson Engineering on two drilled shaft foundations in California. Also discussed are results from CSL and SE/IR tests performed on a drilled shaft foundation in New Mexico.

The CSL results between tube pair 1-4 of the first shaft in California are presented in Figure 5. A significant delay in arrival times of compression waves and a significant drop in energy were observed in this CSL log at depths ranging from 2.5 to 3.4 m (8.3 to 11.2 ft) below the top of the shaft. Although the anomaly depth is well identified in Figure 5, the exact location of the defect between tubes 1 and 4 cannot be determined. For a better characterization of the anomaly indicated in Figure 5, a tomograhic dataset was obtained by Olson Engineering. For this dataset, the source was pulled from a depth of 5.5 m (18 ft) below the shaft top and ending at the top of the tubes with the receiver moved at fixed interval locations of 57 mm (2.25 in.). A velocity tomogram between tube pair 1-4 is presented in Figure 6. The anomalous zone in Figure 6 is represented as the low-velocity area (light area) which extends from a depth of 2.8 m (9.3 ft) to a depth of 3.5 m (11.6 ft) below the top of the shaft. The apparent low-velocity regions in the middle at the top and bottom of Figure 6 are artifacts resulting from a low ray density in these areas (see Figure 6 for the distribution of the ray densities). Figure 6 clearly shows that the defect occupies the entire distance between tubes 1 and 4, which cannot be inferred from the CSL results. This was confirmed by subsequent excavation.

CSL results performed by Olson Engineering on a drilled shaft foundation at the Sargent Bridge on Highway 101 in Hollister, California identified a defect at depths ranging from 4.6 to 5.2 m (15 to 17 41 ft). The anomaly was more severe between tube pair 1-3 than between other tube pairs (Figure is similar to Figure 5 and not shown here). This anomaly was further confirmed by Gamma-Gamma testing and destructive coring showed it to be a soil intrusion. Tomograhic data was obtained between tube pair 1-3 with the source pulled from 5.8 m (19 ft) below the shaft top to the shaft top and the receiver fixed at 49 locations of 57-mm (2.25-in.) separation. Figure 7 shows the velocity tomogram obtained from this tomographic dataset along with the corresponding ray density plot. The anomalous zone in Figure 7 is represented as the low-velocity area which extends from a depth of 4.6 m (15 ft) to a depth of 5.2 m (17 ft) below the top of the shaft. Figure 7 shows that the defect is centered around the tubes with good quality concrete in the interior between the two tubes as opposed to the defect shown in Figure 6 which extended through the entire distance between the two tubes.

The CSL results between tube pair 1-2 of a shaft tested in New Mexico are presented in Figure 8. A significant delay in arrival times of compression waves and a significant drop in energy were observed in this CSL log at depths ranging from 10.4 to 11.6 m (34 to 38 ft) below the top of the shaft. The CSL results between other tubes showed similar anomalies at the same depths. Another commonly used display for CSL data is the banded time versus depth (also known as Z-axis modified). In this display, a line is plotted for each point of each record for which the positive or negative signal peaks are greater than the threshold value. This results in a series of bands vertically down the plot for a shaft with no defects. A defect will be seen as a disappearance of the bands at the defect depth. Figure 9 shows this type of display for the shaft tested in New Mexico with the negative peaks as the black bands. Sonic Echo/Impulse Response tests were performed on the same shaft. Echoes from a depth of 11 m (36 ft) were identified in the SE records as shown in Figure 10. The upper trace in Figure 10 represents the accelerometer output and the lower trace represents the upper trace after integration (to velocity) and exponential amplification. The IR results showed an echo from a depth of 10.8 m (35.3 ft) as shown in Figure 11. The upper trace in Figure 11 represents the coherence function to reflect data quality and the lower trace is the mobility function which is equal to velocity divided by pound force. No echoes from the bottom of the shaft at a depth of 18.9 m (62 ft) were identified in the SE/IR records. It was then concluded that the encountered defect is a major defect since bottom echoes could not be identified. Note also the good agreement between the CSL and SE/IR results. If there were additional defects below the major encountered defect at a depth of 11 m (36 ft), they most likely would not have been identified by the SE/IR method, but could easily have been identified with the CSL method.

SOLUTIONS TO ENCOUNTERED DEFECTS When defects are encountered in drilled shafts, the design engineer is informed of the problem. The structural and geotechnical engineers usually check the axial and lateral capacity of the shaft after considering the size, depth and severity of identified defects plus actual load test results (if any), concrete strengths and volumes with appropriate safety factors. Based on the new calculations, the shaft is either accepted or rejected.

For shallow defects, the most common procedure is to excavate around the perimeter of the shaft until the defect area is exposed and repair procedures take place. For deep defects which still influence the capacity of the shaft, coring is usually performed and grout is injected using the coreholes. In other circumstances, two substitute shafts are drilled next to the defect shaft so that sufficient load can be transferred to the new shafts via a tie beam such that the defect shaft may carry no loads or reduced loads safely.

CONCLUSIONS 42

The CSL method is an excellent nondestructive method for identifications of anomalous zones in drilled shaft foundations. Many State DOT's are moving towards specifications for CSL tests on new construction of drilled shaft foundations. The method is effective at locating defects between tube pairs, defect depths and extent, but not exact locations of defects between tube pairs. The CT can be used as a complimentary method to the CSL method to determine a better characterization of the defect. Because of the much greater time required to perform tomographic analyses, the method may not gain popularity and its application will be limited to the more critical structures. The SE/IR method can be used in conjunction with the CSL method to determine the nature of encountered anomalies. A good application for the SE/IR method is when the identified anomalies in CSL testing are due to tube debonding and do not represent any defects in the drilled shafts. An SE/IR test is usually performed and if no reflections are identified in the areas where tube debonding occurred in CSL testing, the shaft is considered to be sound.

REFERENCES Olson, L.D. and R.W. Thompson, 1985. “Case Histories Evaluation of Drilled Pier Integrity by the Stress Wave Propagation Method,” Proceedings: Drilled Piers and Caisson II, American Society of Civil Engineers Spring Convention, Denver, Colorado.

Davis, A.G. and C.S. Dunn, 1974. "From Theory to Field Experience with the Nondestructive Vibration Testing of Piles," Proceedings of the Institution of Civil Engineers Part 2, 57: 571-593.

Olson, L.D., M. Lew, G.C. Phelps, K.N. Murthy, and B.M. Ghadiali, 1994. "Quality assurance of Drilled Shaft Foundations with Nondestructive Testing," Proceedings of the FHWA Conference on Deep Foundation, Orlando, Florida.

Herman, G.T, 1980. "Image Reconstruction from Projections, the Fundamentals of Computerized Tomography," Academic Press, Inc.

Olson, L.D., F. Jalinoos, M.F. Aouad, and A.H. Balch, 1993. “Acoustic Tomography and Reflection Imaging for Nondestructive Evaluation of Structural Concrete,” NSF Phase I Final Report (Award # 9260840), SBIR Industrial Innovation Interface Division, Washington, D.C. Koten, H. Van and P. Middendorp, 1981. “Testing of Foundation Piles,” HERON, Joint Publication of the Department of Civil Engineering of Delft University of Technology, Delft, The Netherlands, and Institute TNO for Building Materials and Sciences, Rigswijk (ZH), The Netherlands, Vol. 26, No. 4.

Paquet, J., 1991. “Une Nouvelle Orientation Dams le Controle D’Integute Des Pieux par Sollicitation Dynamique: Le Profil D’Inpedana,”, Frud Colloque International, Foundation Profondes, Paris, pp 1-10 (in French).

LIST OF FIGURES Figure 1- Crosshole Sonic Logging Test Setup. Figure 2- Example Crosshole Sonic Log. Figure 3- Crosshole Tomography Setup between Two Tubes of a Drilled Shaft. Figure 4- Sonic Echo/Impulse Response Test Method. Figure 5- CSL Results Between Tubes 4-1 in a Drilled Shaft in California. Figure 6- Velocity Tomography Results from the First Drilled Shaft in California. Figure 7- Velocity Tomography Results from the Second Drilled Shaft, Hollister, California. Figure 8- CSL Results Between Tubes 1-2 in a Drilled Shaft in New Mexico. 43

Figure 9- Alternative Banded Time Display of CSL Results Between Tubes 1-2 in a Drilled Shaft in New Mexico. Figure 10- Sonic Echo Test Results, Drilled Shaft in New Mexico. Figure 11- Impulse Response Test Results, Drilled Shaft in New Mexico.

44

Figure 1- Crosshole Sonic Logging Test Setup. 45

Figure 2- Example Crosshole Sonic Log.

46

Figure 3- Crosshole Tomography Setup between Two Tubes of a Drilled Shaft.

47

Figure 4- Sonic Echo/Impulse Response Test Method. 48

Figure 5- CSL Results Between Tubes 4-1 in a Drilled Shaft in California.

49

Figure 6- Velocity Tomography Results from the First Drilled Shaft in California.

50

Figure 7. Velocity Tomography Results from the Second Drilled Shaft, Hollister, California.

51

Figure 8- CSL Results Between Tubes 1-2 in a Drilled Shaft in New Mexico.

52

Figure 9- Alternative Banded Time Display of CSL Results Between Tubes 1-2 in a Drilled Shaft in New Mexico. 53

First Echo Second Echo at 6 ms at 12 ms

The lower trace shown is the same data as the upper trace after it has been integrated and exponentially amplified. T1 = 6 ms T2 = 12 ms Compression Wave Velocity = 3658 m/sec (12,000 ft/sec) Depth of Reflector = VC * T1/2 = 3658 * 0.006/2 = 11 m (36 ft) The reflection is from an anomaly located at about 11 m below the shaft top

Figure 10 - Sonic Echo Test Results, Drilled Shaft in New Mexico.

54

∆ ∆f = ∆f = f = 170 Hz 170 Hz 170 Hz

The upper trace shown is the coherence function to reflect data quality. The lower trace is the mobility function used to obtain reflector depth

∆f = 170 Hz Compression Wave Velocity = 3658 m/sec (12,000 ft/sec) Depth of Reflector = VC/∆f* 2 = 3658/ 170* 2 = 10.8 m (35.3 ft) The reflection is from an anomaly located at about 11 m below the shaft top

Figure 11- Impulse Response Test Results, Drilled Shaft in New Mexico.

55

Non-Destructive Evaluation of Installed Soil Nails

Word Count: Abstract: 158 Text: 4808 Tables 1@250: 250 Figures 9@250: 2250 Total Words: 7466

Date Submitted: November 18, 2005

Jie Gong Research Assistant Civil Engineering Department Texas Tech University, Lubbock, TX 79409-1023 Telephone No.: (806) 742-3471 ext. 245 Fax: (806) 742-4168 E-mail: [email protected]

Priyantha W. Jayawickrama Associate Professor Civil Engineering Department Director, Center for Multidisciplinary Research in Transportation Texas Tech University, Lubbock, TX 79409-1023 Telephone No.: (806) 742-3471 ext. 245 Fax: (806) 742-4168 E-mail: [email protected]

Yajai Tinkey Research Engineer & Special Project Consultant Olson Engineering Inc. 12401 W. 49th Ave. Wheat Ridge, Colorado 80033 Telephone No.: (303) 423-1212 Fax: (303) 423-6071 E-mail: [email protected]

56

ABSTRACT This paper presents the findings from a research study that evaluated Non-Destructive Test (NDT) methods that may be used for quality assurance purposes during soil nail retaining wall construction. More specifically, the new NDT procedure will be used to evaluate the integrity of the grout column surrounding the steel tendon and to measure the grouted length of the installed soil nail. A preliminary evaluation of the existing NDT methodologies was undertaken to identify candidate NDT methods. The selected NDT methods, Sonic Echo and Impact Echo methods were subjected to a detailed evaluation through three separate series of full-scale field NDT testing. The findings from the Sonic Echo tests are presented in this paper. The field studies demonstrated that the Sonic Echo method can be used successfully to detect major defects found in installed soil nails. The nail condition predictions from Sonic Echo tests were confirmed by the actual nail conditions observed after exhuming all the soil nails.

INTRODUCTION Soil nailed retaining walls are commonly used by transportation agencies in both temporary and permanent applications. Unfortunately, occasional failures of soil nail walls have occurred, indicating that there are still some unresolved issues related to the construction of this type of retaining wall. In a recent forensic study that investigated failure of several Texas Department of Transportation (TxDOT) soil nailed retaining walls, it was found that incomplete grouting of soil nails was a common problem. Thus, it is clear that the ability to assess grout condition is vital for the safe performance of soil nailed retaining structures. Traditionally, TxDOT has relied on pull out testing of selected nails to verify the integrity of installed soil nails. However, TxDOT experience has shown that such destructive field testing is inefficient and does not ensure the integrity of the remaining, untested soil nails. If a non-destructive test method can be developed for integrity assessment of the grout in soil nails, then it will greatly enhance field quality control and quality assurance of soil nailed walls. Non-Destructive Testing (NDT) techniques have been used for many years to provide quality control of drilled shaft foundations and driven concrete piles. Although soil nails have characteristics that are similar to drilled shafts and piles, there has been little or no previous work related to the use of NDT methods for soil nail quality control. The purpose of the study presented in this paper is to develop an NDT procedure that can be used to evaluate the integrity of the grout column surrounding the steel tendon and measure the grouted length of the installed soil nails. A systematic search was conducted to identify candidate NDT technologies that have been used successfully in applications similar to soil nails. A field test program was implemented to further evaluate the candidate NDT techniques. 32 soil nails with various lengths (5 – 25 ft) and defect locations were installed at the test site. Three series of NDT tests were performed on the experimented soil nails. After the final NDT tests, the experimental soil nails were exhumed and observed to verify the NDT predictions. This paper describes construction problems commonly encountered during soil nail wall construction and reviews candidate NDT methods that may be used in this application. It also explains the field test procedure used in the NDT investigation and presents the findings from the study.

57

THE PRELIMINARY STUDY

Construction Problems in Soil Nail Installation One of the critical steps in soil nail construction involves grouting around the steel tendon. The grout column around the nail serves two important functions; first, it transfers the load from the tendon to the surrounding soil and secondly, it protects the steel tendon from corrosion. Incomplete or unsatisfactory grouting can occur due to a number of reasons. The most common among these are: (a) the collapse or bulging of the borehole walls, (b) the use of grout with improper consistency, (c) centralizers that obstruct grout flow, (d) use of inadequate tremie pipe length, and (e) the rapid withdrawal of the tremie pipe. These improper construction practices result in the formation of different types of grout column defects. These defects that include voids at the end of the nail, voids near the nail head (bird’s mouth defect), and half-filled holes pose a significant threat to the safe performance of soil nailed structures. Figure 1 depicts some of these problems discovered after wall failure.

Review of Candidate NDT Methods A comprehensive search was conducted to identify candidate NDT methods that would meet the requirements of this application. The criteria used in the selection were: ability to detect major grout column defects with good reliability and accuracy, low equipment cost, ease of implementation in a field setting, and short test duration. Soil nails are rod-like structures embedded in soil. They are very similar to piles and drilled shafts. A number of NDT methods, including Sonic Echo, Impulse Response, Impedance Logging, Impact Echo, Parallel Seismic, and Cross-hole Sonic Logging have been used in applications similar to soil nails. The principles and limitations of these methods can be found in ACI 228.2R-98 (1). The early work of Steinbach and Vey (2) laid the foundation for applying the Sonic Echo method in deep foundation evaluation. Davis and Dunn (3) developed the Impulse Response method as an extension of pile vibration tests. Rausche (4) and Likins (5) conducted several research and case studies on low strain pile integrity testing. Finno et al. (6) reviewed Sonic Echo, Impulse Response, Parallel Seismic and Cross-hole Sonic Logging methods in great detail and compared the performance of Sonic Echo, Impulse Response, and Parallel Seismic methods in pile integrity evaluation through full-scale field experiments. Finno and Gassman (7, 8) further studied the Sonic Echo/Impulse Response methods through numerical simulation and the use of multiple sensors. Recently, the study of wave propagation based NDT methods has been conducted in the signal quality enhancement domain by Addison et al. (9). The Impact Echo method was first developed in the early 1980s by Sansalone and Carino (10) for testing plate-like structures. Since then, continuous research efforts by researchers at Cornell University and the National Bureau of Standards have extended the Impact Echo method to a great variety of problems and geometries (11, 12). The recent work of Withiam et al. (13) presented the application of Impact Echo in assessing the condition of metal-tensional systems. Different types of defects have been found in soil nails. Deep defects, such as soil cave-in, air voids at the end or middle of soil nails, and insufficient grout length, present themselves as abrupt impedance changes in NDT test results. Sonic Echo and Impulse Response methods are able to detect these defects as long as stress waves are able to reach the defects and be reflected back. One limitation in the use of Impulse Response method in soil nail application involves the type of impactor used to create the excitation. A large sledge hammer is often used in Impulse Response method. The long contact time between the hammer and the structure head increases the low frequency content in the waves, resulting in waves with wavelengths that are too long to detect nail defects. The direct transmission methods such as Cross-hole Sonic Logging and Parallel Seismic methods offer the advantages of ease of interpretation and lower susceptibility to attenuation effects. However, they require the installation of additional access 58 tubes in the soil nail wall and therefore are impractical in routine soil nail quality control testing. Based on the above observations, the Sonic Echo technique appears to be the most suitable method for detecting deep defects. Bird’s mouth is a shallow defect occurring at the nail head. When shallow defects are present, the time interval between the generation of stress pulses and the reception of the reflected wave is so small that the determination of the arrival time of reflected waves would be difficult to accomplish in the time domain. Furthermore, as observed by Hearne et al (14), Rayleigh waves generated by hammer impact propagate along the shaft structure, creating a noisy environment in the top 10-ft of the shaft and making defect detection unreliable. Therefore, the Impact Echo method which employs frequency analysis is a more logical choice for detecting shallow defects in soil nails because the identification of peaks belonging to defects from spectral plot is much quicker and simpler than in the time domain.

BACKGROUND OF SONIC ECHO TEST The Sonic Echo (SE) test is commonly used to evaluate the integrity of the drilled shaft foundations or timber piles. The SE method is a low strain integrity test that is conducted from the top of the pier as shown in Figure 2. Test equipment for SE on foundations typically include a 3-lb impulse hammer, an accelerometer receiver mounted on the top or sides of the piers, and a PC-based data acquisition/analysis system. The impulse hammer has a built-in load cell that measures the force and duration of the impact. The test involves hitting the foundation top or a block mounted to the side with the hammer to generate wave energy that travels to the bottom of the foundation. The wave reflects off irregularities (cracks, necks, bulbs, soil intrusions, voids, etc.) and/or the bottom of the foundation and travels back along the foundation to the top. The receiver measures the vibration response of the foundation to each impact. The signal analyzer processes and displays the hammer and receiver outputs. Foundation length and integrity of concrete are evaluated by identifying and analyzing the arrival times, direction, and amplitude of reflections measured by the receivers in the time domain. The echo depth (D) is related to the reflection time (t) and the compression wave velocity (V) by the following equation:  t  = VD   (1)  2  The SE test method is sensitive to changes in the pier impedance (pier concrete area · velocity · mass density where mass density equals unit weight divided by gravity), which cause the reflections of the compression wave energy. Compression wave energy (hammer impact energy) reflects differently from increased pier impedance than from decreased pier impedance. This allows the type of reflector to be identified either as a “neck” (representing a decrease in the pier impedance indicated by a downward initial break of a reflection event in an SE record) or as a “bulb” (representing an increases in the pier impedance marked by an upward initial break for an SE reflector ).

THE EXPERIMENTAL SOIL NAIL WALL

Construction of the Experimental Soil Nail Wall An experimental soil nail wall was constructed on the Texas Tech University campus to evaluate the selected NDT methods. A 125-ft long, 6.5-ft high test wall was built at the site to accommodate 32 soil nails placed in a single row with a nail spacing of 4 ft ( Figure 3a). The design lengths of nails varied from 5-ft to 25-ft. The soil nails were 6 inches in diameter and installed at a 10° or 15° downward angle with nail heads located 3 ft from the top of the test wall. The center steel tendons consisted of epoxy coated Grade 60 steel bars with a diameter of 1 inch. Split PVC style centralizers were used because TxDOT specifications require the use of this type of centralizer. Two types of grout, a sand-cement-water 59 mixture (“sand cement grout”) and cement-water mixture (“neat cement grout”) were used in soil nails. Table 1 summarizes the designed conditions of installed experimental soil nails. Three types of artificial defects, including voids at the end of nails, voids in the middle of nails, and bird’s mouth, were simulated at specific locations by using polyethylene foam. The donut-like foam disk was attached to steel tendons at specified locations, preventing grout materials from penetrating through and creating abrupt impedance changes (Figure 3b).

Field Investigation of Candidate NDT Methods Before commencing field tests, preliminary testing using Ultrasonic Pulse Velocity (UPV) was conducted in the laboratory to measure the compression wave velocity in the grout. Ultrasonic Pulse Velocity tests were performed on sand cement, neat cement grout cylinders that were sampled from the batches of grout material used in grouting the field test nails. In addition, Sonic Echo and Impact Echo tests were conducted on a stand-alone steel tendon to measure the compression wave velocity in steel. Note that only results from Sonic Echo tests are included in this paper. Three series of field NDT tests were performed on the experimental soil nails. Necessary improvements were incorporated in the test equipment and data acquisition and processing procedures after each round of testing. Upon the completion of three series of NDT investigations, all the experimental soil nails were exhumed for direct observation of actual conditions. The direct observation data were then used to verify the predictions from the NDT investigations. The hardware system used in Sonic Echo and Impact Echo testing consisted of an impact generator, a signal detector and data collection equipment. Modulated PCB hammers (0.2 lb) with two different types of tips were used as the impact generator. The hammer with a plastic tip typically generates frequency up to 4000 Hz. The hammer with an aluminum tip typically generates higher vibration frequency up to 8000 Hz, yielding lower penetration depth but higher sensitivity to minor defects than the hammer with the plastic tip. Accelerometers were used as the sensors for recording surface accelerations. Steel washers were mounted on the nail head using grease or epoxy glue and the accelerometer was then attached to the washer. A rugged Freedom Data PC along with a data collection and analysis software from Olson Instruments were used as the data collection equipment. Two Accelerometer mounting locations, two impact locations, and two types of hammer tips were explored in Sonic Echo/Impact Echo test configurations. As shown in Figure 4, eight test configuration combinations were performed on each experimental nail.

60

TEST RESULTS AND DISCUSSION

Determination of Compression Wave Velocity Ultrasonic Pulse Velocity tests were performed on four neat cement and four sand cement cylinders to measure the compression wave velocity in different types of grout. Each type of cement included one in wet condition and three in dry condition. The average compression wave speed for the neat cement grout was 12,818 ft/sec with a standard deviation of 45.5 ft/sec. The average compression wave velocity for the sand cement grout was 13,910 ft/sec with a standard deviation of 156 ft/sec. The Sonic Echo test was performed on a stand-alone steel tendon with a known length of 10.28-ft to calculate the wave velocity in the steel tendon. In an actual soil nail, the steel tendon is embedded in grout. Although stress waves travel faster in steel tendons than in grout, if the grout and the tendon are bonded together well, then the stress waves will travel through the soil nail as a composite. A theoretical composite wave can be calculated based on an average steel and grout velocity using the correct volumes of grout and tendons. However, the theoretical calculation was not performed in this study. Instead, the composite velocity was obtained from direct calibrations on an actual soil nail (Nail No.1) with an assumption that Nail No.1 was in sound condition. The calibration yielded a composite wave velocity of 16,000 – 16,500 ft/sec.

Soil Nail without Defects Figure 5a shows the “as-designed” configuration of Test Nail No. 26. This nail was designed to be a 20-ft long, sand-cement nail with no defects. The Sonic Echo test result shown in Figure 5b identified the bottom echo at 20.4-ft with no early echo indicating that the nail was sound with no major defects. Figure 5c shows the nail after the exhumation for inspection. The nail was found to be 21.5-ft long and have no defects. In this case, the NDT test results showed good agreement with the actual observed condition of the nail (within 1-ft difference in length).

Soil Nail with a Void at the End Figure 6a Sonic Echo plot for shows Nail No.29 which was designed with an end defect at a length of 20 ft. Review of the figure shows double echoes at a length of 15.2 ft indicating a major defect at that location. The exhumation records of Nail No. 29 (Figure 6b) showed that the grout ended at a length of 15.3 ft. The Sonic Echo (SE) result shows good agreement with the actual condition of the nail within 0.1-ft.

Soil Nail with an Intentional Void at the Middle Figure 7a shows a schematic view of Test Nail No.30. This nail was designed to be 25 feet long and carried an intentionally created void that started at 17 feet and ended at 19 feet. The foam obstruction covered only the top half of the nail so that the grout can flow underneath the obstruction and fill the far end of the nail. The Sonic Echo test result shown in Figure 7b identified the first reflection echo at 17.0 feet that corresponds to stress wave reflections from the middle artificial void. Figure 7c shows the nail after it was exhumed for inspection. The grout had moved underneath the obstruction and filled the hole up to a length of 26.4 feet. The Sonic Echo test performed on this nail was able to detect the middle void that represented a 50% cross section area reduction, but the end of grout column was not detected by the test.

61

Soil Nail with Unplanned Defects In the test nails described above, the actual condition was found to be very close to the designed condition of those nails. However, this was not the case with many other nails. Many of them were found to contain defects that were not planned. In nearly all cases, the Sonic Echo test correctly identified the defects in these nails. The actual conditions of these nails, documented after they were exhumed and inspected confirmed Sonic Echo predictions.

An example of a soil nail with unplanned defects was Test Nail No.7. This nail was designed to be 15 feet long with no planned defects. However the SE result (as shown in Figure 8a) shows an echo at 10.4 ft indicating a major defect or discontinuity at that location. The exhumation records (Figures 8b and 8c) show that there was a significant reduction in cross section at a length of 10 ft and the grout column ended at a length of 13.6 ft. In this case, the SE test predicted the location of grout reduction accurately.

Maximum Detectable Nail Length by Using Sonic Echo Method The data collected in this research study showed that, under ideal conditions, the Sonic Echo method could detect nail lengths up to 26 feet. This conclusion is based on data recorded for Test Nail No. 27. Figure 9 shows the results from Sonic Echo tests on Nail No. 27. The actual nail condition from the exhumation records showed that Nail No. 27 joined with Nail No. 28 at 26 ft length (length to diameter ratio of 50:1). This agreed well with the predictions made by the Sonic Echo tests. In all the other cases, the reliable range of length that Sonic Echo could achieve is 15-20 feet, which corresponds to length vs. diameter ratio at 30:1- 40:1. A similar reachable length vs. diameter ratio has been documented in pile testing (15). The capability of the Sonic Echo test to reach large depths may depend on factors such as the quality of grout, the bonding condition between grout and surrounding soil, the properties of the surrounding soil, and the choice of test configuration. The data collected in this research study suggests that the Sonic Echo test performs best under following conditions: (a) Neat cement grouting (b) A soil nail which is free of minor defects at shallow depth (c) A flat and well grouted head with clear access (d) A test configuration that uses an impact on the grout column with plastic or aluminum hammer tip, and accelerometer mounted on grout column

Ability to Detect Different Types and Sizes of Defects The data collected in this research also allowed the Sonic Echo/Impact Echo methods to be evaluated with respect to their ability to detect different types and sizes of defects. As shown in Table 3, the Sonic Echo was not only able to reliably detect intentionally created large voids (representing a 50% or 100% reduction in cross-sectional area of grout column) at the middle and end nails, but also proved to be effective in detecting unplanned deep defects that represented cross-section area reductions of no less than 20%. However, the Sonic Echo method failed to identify shallow defects that were located within the top 5 feet. Conversely, the Impact Echo Test proved to be effective in detecting shallow defects such as bird’s mouth defects. The primary drawback in the impact echo method was its “over-sensitivity.” In many cases, the defects found by this method were found to be too insignificant to be of any concern.

62

Optimum Test Configuration Test Configurations 1 and 2—Impactor and Accelerometer on Steel Tendon Test configurations 1 & 2 usually did not yield useful results when the nail length exceeded 15 feet. Depending on the quality of the impact, they sometimes yielded good results for defects or length of nail within the top 15 feet (e.g. Nail No.5). The waveforms recorded from test configurations 1 and 2 were usually too noisy because impacts on steel tendon appeared to create transverse vibrations (i.e. so called “ringing effect”) and the limited space available on the steel head around the accelerometers made it difficult to apply a high quality impact. It is recommended these two test configurations not be used unless the grout is in very poor condition near the nail head so that accelerometers cannot be mounted on it. In the event that this test configuration is used in QC/QA testing, the inspector must understand that data collected will be applicable to nail condition within the top 15 ft only.

Test Configurations 3 and 4—Impactor on Steel Tendon; Accelerometer on Grout Test configurations 3 & 4 provide the best quality impacts and the most reliable response detection because steel tendons used in construction projects tend to be defect free and extend all the way to the end of nails. This made it easier to excite the entire nail with an impact upon the tendon. This test configuration allowed the Sonic Echo method to detect the ends of grout behind severe middle defects (e.g. Nail no.18), while in other test configurations Sonic Echo method tended to be “blind” to any object beyond the middle defects. It appears that, in these cases, the large middle defect blocked most of wave energy and prevented it from penetrating to greater depths.

Test Configurations 5 and 6—Impactor on Grout; Accelerometer on Steel Tendon The test configurations 3 & 4 proved to be satisfactory as far as Sonic Echo test’s ability to detect any major defects. More importantly, they were found to be the most efficient configuration for Impact Echo tests in terms of data quality and ease of defect peak identification. For a nail with bird’s mouth defect, applying impact on grout induces the nail head to behave more like a thin plate. And surface response recorded by accelerometers mounted on steel tendons would be less noisy because any disturbance from Rayleigh waves would be less pronounced.

Test Configurations 7 and 8—Impactor and Accelerometer on Grout Although test configurations 7 & 8 have the strictest requirements for the nail head conditions, these configurations proved to be the most effective in detecting defects and determining defect locations in soil nails. However, a good nail head is essential to acquire high quality data through these two test configurations. Between these two configurations, the test configuration with plastic tip generates waves with lowest frequency content or the longest wavelengths when compared to any other test configuration, yielding the deepest penetration as discussed in the previous section. This test configuration tends to overlook minor defects due to the long wavelengths. The other test configuration with aluminum tip hammer has finer resolution than the test configuration with plastic tip hammer, enabling it to detect smaller defects at the expense of penetration depth. Thus, the two configurations overcome limitations of each other and the combination works better than any one of the two.

Ease of Testing and Testing Time The Sonic Echo test investigated in this research does not require any special test equipment other than those described previously. The hardware system used is rugged and easy to mobilize. The time required for testing a soil nail using Sonic Echo/Impact Echo methods depends on the proficiency of the operator, the number of test combinations for each nail, and the number of channels which can be used 63 simultaneously in data acquisition system. As an estimate of the testing time required, the average Sonic Echo test duration for each nail from the third series of testing is provided as a reference. The test hardware system used two types of hammers and three types of accelerometers which were mounted at three locations on the nail heads, one on steel and two on grout. Four acquisition channels, three channels for accelerometers and one for impact force were used simultaneously to acquire data. Thus, with each hammer hit three test combinations could be completed. The average time for data acquisition for each nail was approximately 7 minutes. An additional 5 minutes would be required for mixing epoxy glue for attaching washers before launching the tests. Accordingly, the duration of Sonic Echo test was approximately 12 minutes per nail. Within that time, 16 test combinations were performed and 6 data sets were acquired for each test combination. This represented a total of 96 effective data sets that were recorded for one nail.

CONCLUSIONS This research study examined the potential application of NDT techniques for grout condition evaluation of installed soil nails. A total of 32 experimental soil nails that represented different lengths, grout types and construction defects commonly found in soil nail installation were tested using the Sonic Echo method. It was demonstrated that the Sonic Echo method showed the greatest promise for evaluating the integrity of the grout. Sonic Echo tests were found to be particularly effective for locating major defects at greater depths, including defect at the end of nail and defects at the middle of nail. Depending on the test configurations, Sonic Echo tests were also able to detect multiple minor defects and end of steel tendons. The implementation of Sonic Echo test should be quite straightforward and economical. The limitations of Sonic Echo test on soil nail are its low sensitivity to shallow defects such as bird’s mouth and its dependency on head conditions. A hardware system for Sonic Echo technology can be used as a reliable NDT tool for evaluating installed soil nails in the field. Various impact sources that generate waves with different frequency contents could be used to increase wave penetration distance and achieve fine defect scan resolution. Special care should be given to the contact condition between sensors and nail head. Technicians or engineers should be able to use the proposed system with limited specialized training.

ACKNOWLEDGMENTS This research was sponsored and supported by the Texas Department of Transportation (TxDOT) through Center for Multidisciplinary Research in Transportation (TechMRT) of Texas Tech University. The research project was conducted as a joint effort in which researchers from TechMRT worked in partnership with research engineers at Olson Engineering, Inc., an NDT consulting company based in Denver, Colorado. The authors are grateful for the sponsorship and assistance provided by TxDOT.

REFERENCES

1. ACI Committee 228 (1998). Nondestructive Test Methods for Evaluation of Concrete in Structures. ACI 228.2R-98.

2. Steinbach, J., and Vey, E. (1975). Caisson Evaluation by Stress Wave Propagation Method. Journal of Geotechnical Engineering, ASCE, V.100, No. GT4, Apr., pp.361-378. 64

3. Davis, A. G., and Dunn, C. S. (1974). From Theory to Field Experience with Non-Destructive Vibration Testing of Piles. Proc. Inst. Of Civil Engrs., London, U.K., 571-593.

4. Rausche, F., Likins, G. E., Ren-Kung, S., September, 1992. Pile integrity testing and analysis. Proceedings of the Fourth International Conference on the Application of Stress-Wave Theory to Piles: The Netherlands; 613-617. 5. Likins, G., Rausche, F., Morgano, C.M., & Piscsalko, G. (2000b). “Advances in PDI low strain integrity test methods and equipment.” Proceedings, sixth international conference on the application of stress-wave theory to piles. Sao Paulo, Brazil. 6. Finno, R. J., and Gassman, S. L., and Osborn, P. W. (1997). Nondestructive Evaluation of a Deep Foundation Test Section at the Northwestern University National Geotechnical Experimentation Site. Federal Highway Administration, Washington, D.C. 7. Finno, R. J., and Gassman, S. L. (1998). Impulse Response Evaluation of Drilled Shafts. J. Geotech. and Geoenvir. Engrg., ASCE, 124(10), 965-975. 8. Gassman, S. L., and Finno, R. J. (1999). Impulse Response Evaluation of Foundations Using Multiple Geophones. J. Perf. Constr. Fac., ASCE, 13(2)., 82-89.

9. P.S. Addison, J.N. Watson and A. Sibbald, The practicalities of using wavelet transforms in the non-destructive testing of piles, in: 13th ASCE Engineering Mechanics Division Conf., Baltimore, MD, USA, June 13-16, 1999. 10. Sansalone, M., and Carino, N. J., Impact-Echo: A Method for Flaw Detection in Concrete Using Transient Stress Waves. NBSIR 86-3452, National Bureau of Standards, Gaithersburg, Maryland, Sept. 1986, 222 pp. 11. Lin, Y., and Sansalone, M. (1992b). Detecting Flaws in Concrete Beams and Columns Using the Impact-Echo Method. Am. Concrete Inst. Mat. J., 89(4), 394-405. 12. Lin, J., and Sansalone, M. (1993). The Transverse Elastic Impact Response of Thick Hollow Cylinder. J. Nondestructive Evaluaiton, 12(2), 139-149. 13. Withiam, J. L., Fishman, K. L. and Gaus, M. P. (2002). Recommended Practice for Evaluation of Metal-Tensioned Systems in Geotechnical Applications. NCHRP Report 477, Washington, D.C. 14. Hearne, T. M., Stokoe, K. H., and Reese, L. C. (1981). Drilled-Shaft Integrity by Wave Propagation Method. J. Geotech. Engrg. Div., ASCE, 107(10). 15. Davis, A.G. and Robertson, S.A. (1976). "Vibration Testing of Piles," Structural Engineering, June, pp. A7-A10.

65

LIST OF TABLES AND FIGURES

TABLE 1 Designed Conditions of Soil Nails in the Experimental Soil Nail Wall

FIGURE 1 Problems discovered after wall failure (a) Bird’s mouth (b) Wall failure caused by incomplete grout (c) Insufficient grout length.

FIGURE 2 Sonic Echo test; Principle of operation

FIGURE 3 Construction of field experimental site (a) The field test bed (b) Defect simulation.

FIGURE 4 Sonic Echo/Impact Echo test configurations

FIGURE 5 Results from Sonic Echo tests on experimental nail no.26: (a) Schematic representation (b) Results from Sonic Echo Test (c) Photograph of exposed nail no.26 without defects.

FIGURE 6 Results from Sonic Echo tests on experimental nail no.29: (a) Results from Sonic Echo Test (b) Photograph of exposed nail no. 29 (grout did not reach the foam but ended at 15.3 ft)

FIGURE 7 Results from Sonic Echo tests on experimental nail no.30: (a) Schematic representation (b) Sonic Echo plot (c) Photograph of exposed nail no.30 with void at the middle.

FIGURE 8 Results from Sonic Echo tests on experimental nail no.7: (a) Results from Sonic Echo Test (b) Reduction in shaft diameter observed at 10ft, (c) Grout ending at 13.6 ft.

FIGURE 9 Results from Sonic Echo tests on experimental nail no.27.

66

TABLE 1 The Designed Conditions of Soil Nails in the Experimental Soil Nail Wall

Grout Length of nails (feet) Type of defects type 5 10 15 20 25 Soil nails without defects SC #1 #4 #7 #13 #19 Soil nails without defects NC #2 #5 #8 #14 #20 Defects at the ends of soil nails SC #3 #6 #9 #15 #21 Defects at the middle parts of nails SC #10 #16 #22 Bird’s mouth SC #11 #17 #23 Defect at the middle parts of nails NC #12 #18 #24 Soil nails without defects NC #27 #25 #28 Soil nails without defects SC #26 #32 Defects at the middle parts of soil nails NC #30 Defects at the middle parts of soil nails SC #31 Defects at the ends of soil nails SC #29 [Note: SC—Sand Cement; NC—Neat Cement.] 67

(a)

(b)

(c) FIGURE 1. Problems discovered after wall failure (a) Bird’s mouth (b) Wall failure caused by incomplete grout (c) Insufficient grout length 68

FIGURE 2. Sonic Echo test; Principle of operation

69

(a)

(b)

FIGURE 3. Construction of field experimental site (a) The field test bed (b) Defect simulation.

70

Accelerometer

Grout Grout Accelerometer Hammer Hammer

Plastic or alumina tip Plastic or alumina tip Steel tendon Steel tendon Test Configuration 1 & 2 Test Configuration 3 & 4

Grout Accelerometer Grout Accelerometer

Hammer Hammer

Steel tendon Steel tendon Plastic or alumina tip Plastic or alumina tip Test Configuration 5 & 6 Test Configuration 7 & 8

FIGURE 4. Sonic Echo/Impact Echo test configurations.

(a)

Avg Time Domain SE Data - Velocity Trace, Depth = 20.4 ft, X = 5.48 ms, Y = -3.38e-005 0.0001

0

-0.0001

-0.0002 0 2000 4000 6000 8000 10000 12000 Time (us)

(b)

(c) FIGURE 5. Results from Sonic Echo tests on experimental nail no.26: (a) Schematic representation (b) Results from Sonic Echo Test (c) Photograph of exposed nail no.26 without defects.

SONIC ECHO (SE) www.olsoninstruments.com www.olsonengineering.com

Time Domain SE Data - Velocity Trace, Depth = 15.2 ft, X = 6.88 ms, Y = -4.77e-006

1e-005 0 -1e-005 -2e-005 -3e-005 -4e-005 0 2000 4000 6000 8000 10000 12000 14000 Time (us)

(a)

(b)

FIGURE 6. Results from Sonic Echo tests on experimental nail no.29: (a) Results from Sonic Echo Test (b) Photograph of exposed nail no. 29 (grout did not reach the foam but ended at 15.3 ft)

SONIC ECHO (SE) www.olsoninstruments.com www.olsonengineering.com

le he midd Void at t

(a) Avg Time Domain SE Data - Velocity Trace, Depth = 17. ft, X = 5.26 ms, Y = -1.12e-006 5e-006

0

-5e-006

-1e-005

-1.5e-005 0 2000 4000 6000 8000 10000 12000 Time (us)

(b)

(c)

FIGURE 7. Results from Sonic Echo tests on experimental nail no.30: (a) Schematic representation (b) Sonic Echo plot (c) Photograph of exposed nail no.30 with void at the middle.

SONIC ECHO (SE) www.olsoninstruments.com www.olsonengineering.com

Avg Time Domain SE Data - Velocity Trace, Depth = 10.4 ft, X = 4.26 ms, Y = -3.08e-007 5e-006

0

-5e-006

-1e-005

-1.5e-005 0 2000 4000 6000 8000 10000 12000 Time (us)

(a)

(c)

(b)

FIGURE 8. Results from Sonic Echo tests on experimental nail no.7: (a) Results from Sonic Echo Test (b) Reduction in shaft diameter observed at 10ft, (c) Grout ending at 13.6 ft

SONIC ECHO (SE) www.olsoninstruments.com www.olsonengineering.com

Avg Time Domain SE Data - Velocity Trace, Depth = 26. ft, X = 6.55 ms, Y = -1.29e-005

2e-005

0

-2e-005

-4e-005

0 2000 4000 6000 8000 10000 12000 Time (us)

FIGURE 9. Results from Sonic Echo tests on experimental nail no.27