GEORGIA DOT RESEARCH PROJECT 22-02 Final Report
FIELD EVALUATION OF WIRELESS ULTRASONIC THICKNESS MEASUREMENT
WITH STEEL BRIDGE MEMBERS
Office of Performance-based Management and Research
600 West Peachtree Street NW | Atlanta, GA 30308 October 2023

1. Report No. FHWA-GA-23-2202

TECHNICAL REPORT DOCUMENTATION PAGE

2. Government Accession No. 3. Recipient's Catalog No.

N/A

N/A

4. Title and Subtitle Field Evaluation of Wireless Ultrasonic Thickness Measurement with Steel Bridge Members
7. Authors Yu Otsuki (https://orcid.org/0000-0002-8202-967X), Korawat Sakpunpanom, Yang Wang, Ph.D. (https:/orcid.org/0000-0002-1031-9491)

5. Report Date October 2023
6. Performing Organization Code N/A
8. Performing Organization Report No. 22-02

9. Performing Organization Name and Address Georgia Institute of Technology 790 Atlantic Drive Atlanta, GA 30332

10. Work Unit No. N/A
11. Contract or Grant No. PI# 0017924

12. Sponsoring Agency Name and Address Georgia Department of Transportation (SPR) Office of Performance-based Management and Research 600 West Peachtree Street NW Atlanta, GA 30308

13. Type of Report and Period Covered Final Report April 2022 October 2023
14. Sponsoring Agency Code N/A

15. Supplementary Notes Prepared in cooperation with the U.S. Department of Transportation, Federal Highway Administration.

16. Abstract
This report presents the field validation of wireless ultrasonic thickness measurement systems on in-service steel bridges. The objective of this implementation is to demonstrate the feasibility of long-term corrosion monitoring. The developed system utilizes Martlet wireless ultrasonic sensing devices with two specially designed daughterboards: the high-rate ultrasonic board and the pulser board. To ensure accurate measurements, the project first derives the calibration function of the Martlet ultrasonic device. Subsequently, thickness measurements are carried out on various steel members and compared with readings obtained from a commercially available handheld thickness gauge. For preliminary validation, the system is deployed on an in-service highway bridge near the city of LaGrange, GA. Following the successful operation of the installed system, this project implements the system on the target bridge located in Douglas County, GA. Visual inspection of the bridge confirms corrosion on structural members across the bridge. A solar panel and a support structure are first installed to provide continuous power for the battery and gateway computer. Four wireless ultrasonic sensing units are installed on three pile tops and the bottom flange of a beam at the Douglas County bridge. Ultrasonic data are collected at scheduled intervals and automatically uploaded to the cloud, enabling remote monitoring of the thickness values over time. This research validates the feasibility of a continuous steel thickness-monitoring solution without the need for physical presence at the bridge location.

17. Key Words Ultrasonic thickness measurement, wireless sensors, nondestructive testing, structural health monitoring, long-term monitoring

18. Distribution Statement No restrictions. This document is available through the NationalTechnical Information Service, Springfield, VA 22161.

19. Security Classification (of this report) Unclassified

20. Security Classification (of this page) Unclassified

21. No. of Pages 22. Price

43

Free

Form DOT 1700.7 (8-72)

Reproduction of completed page authorized.

GDOT Research Project 22-02
Final Report
FIELD EVALUATION OF WIRELESS ULTRASONIC THICKNESS MEASUREMENT WITH STEEL BRIDGE MEMBERS
By
Yu Otsuki Graduate Research Assistant1
Korawat Sakpunpanom Graduate Research Assistant1
Yang Wang, Ph.D. Professor 1,2
1School of Civil and Environmental Engineering 2School of Electrical and Computer Engineering
Georgia Institute of Technology
Georgia Tech Research Corporation
Contract with Georgia Department of Transportation
In cooperation with U.S. Department of Transportation, Federal Highway Administration
October 2023
The contents of this report reflect the views of the authors, who are responsible for the facts and accuracy of the data presented herein. The contents do not necessarily reflect the official views or policies of the Georgia Department of Transportation or the Federal Highway Administration. This report does not constitute a standard, specification, or regulation.
ii

Symbol
in ft yd mi
in2 ft2 yd2 ac mi2
fl oz gal ft3 yd3
oz lb T
oF
fc fl
lbf lbf/in2

SI* (MODERN METRIC) CONVERSION FACTORS

APPROXIMATE CONVERSIONS TO SI UNITS

When You Know

Multiply By

To Find

LENGTH

inches

25.4

millimeters

feet

0.305

meters

yards

0.914

meters

miles

1.61

kilometers

AREA

square inches

645.2

square millimeters

square feet

0.093

square meters

square yard

0.836

square meters

acres

0.405

hectares

square miles

2.59

square kilometers

VOLUME

fluid ounces

29.57

milliliters

gallons

3.785

liters

cubic feet

0.028

cubic meters

cubic yards

0.765

cubic meters

NOTE: volumes greater than 1000 L shall be shown in m3

MASS

ounces

28.35

grams

pounds

0.454

kilograms

short tons (2000 lb)

0.907

megagrams (or "metric ton")

TEMPERATURE (exact degrees)

Fahrenheit

5 (F-32)/9

Celsius

or (F-32)/1.8

ILLUMINATION

foot-candles foot-Lamberts

10.76 3.426

lux candela/m2

FORCE and PRESSURE or STRESS

poundforce

4.45

newtons

poundforce per square inch

6.89

kilopascals

Symbol
mm m m km
mm2 m2 m2 ha km2
mL L m3 m3
g kg Mg (or "t")
oC
lx cd/m2
N kPa

Symbol
mm m m km
mm2 m2 m2 ha km2
mL L m3 m3
g kg Mg (or "t")
oC
lx cd/m2
N kPa

APPROXIMATE CONVERSIONS FROM SI UNITS

When You Know

Multiply By

To Find

LENGTH

millimeters

0.039

inches

meters

3.28

feet

meters

1.09

yards

kilometers

0.621

miles

AREA

square millimeters

0.0016

square inches

square meters

10.764

square feet

square meters

1.195

square yards

hectares

2.47

acres

square kilometers

0.386

square miles

VOLUME

milliliters

0.034

fluid ounces

liters

0.264

gallons

cubic meters

35.314

cubic feet

cubic meters

1.307

cubic yards

MASS

grams

0.035

ounces

kilograms

2.202

pounds

megagrams (or "metric ton")

1.103

short tons (2000 lb)

TEMPERATURE (exact degrees)

Celsius

1.8C+32

Fahrenheit

ILLUMINATION

lux candela/m2

0.0929 0.2919

foot-candles foot-Lamberts

FORCE and PRESSURE or STRESS

newtons

0.225

poundforce

kilopascals

0.145

poundforce per square inch

Symbol
in ft yd mi
in2 ft2 yd2 ac mi2
fl oz gal ft3 yd3
oz lb T
oF
fc fl
lbf lbf/in2

**SSIIisisthtehesysmymbobl ofolrftohretIhneteIrnatetironnaatlioSnyaslteSmysotfeUmniotsf. UAnpiptsro. pAripatperoropurniadtiengrosuhnodulidngbeshmoaudled tboecommapdley wtoithcoSmecptlioynw4itohf ASSeTctMioEn3480o.f ASTM E380. (R(ReevivsiesdedMMarcahrc2h00230)03)

iii

TABLE OF CONTENTS
EXECUTIVE SUMMARY .......................................................................................................... 1 CHAPTER 1. INTRODUCTION ................................................................................................ 3 CHAPTER 2. LABORATORY STUDIES OF MARTLET WIRELESS ULTRASONIC
SENSING DEVICE ................................................................................................................... 5 MARTLET ULTRASONIC WIRELESS SENSING UNIT .................................................. 5 DERIVATION OF CALIBRATION FUNCTION ............................................................... 8 COMPARISON WITH A HANDHELD THICKNESS MEASUREMENT GAUGE ..... 11 CHAPTER 3. PRELIMINARY VALIDATION OF CONTINUOUS WIRELESS THICKNESS MEASUREMENTS ON THE LAGRANGE BRIDGE................................ 16 TESTBED BRIDGE IN LAGRANGE, GA ......................................................................... 16 FIELD THICKNESS MEASUREMENTS .......................................................................... 17 LONG-TERM ULTRASONIC THICKNESS MEASUREMENTS.................................. 20 CHAPTER 4. LONG-TERM WIRELESS THICKNESS MEASUREMENTS ON THE DOUGLAS COUNTY BRIDGE............................................................................................. 23 TESTBED BRIDGE IN DOUGLAS COUNTY .................................................................. 23 INSTALLATION ................................................................................................................... 24
Solar Panel and Support Structure ................................................................................... 25 Enclosure and Devices......................................................................................................... 26 Wireless Ultrasonic Sensing Units ..................................................................................... 27 LONG-TERM THICKNESS MEASUREMENT RESULTS ............................................ 32 CHAPTER 5. CONCLUSIONS................................................................................................. 35 ACKNOWLEDGEMENTS ....................................................................................................... 36 REFERENCES............................................................................................................................ 37
iv

LIST OF FIGURES
Figure 1. Diagram. Ultrasonic thickness measurement. ................................................................. 3 Figure 2. Diagram. Functional diagram of the Martlet ultrasonic thickness measurement
device. ...................................................................................................................................... 6 Figure 3. Photographs. Printed circuit board design of the Martlet wireless ultrasonic
device. ...................................................................................................................................... 8 Figure 4. Photograph. Steel calibration block from 0.1- to 1.0-inch thicknesses. .......................... 9 Figure 5. Graph. Measured ToF vs error (without calibration). ................................................... 10 Figure 6. Photographs. Steel specimens collected from the Structures Lab at Georgia Tech.
............................................................................................................................................... 13 Figure 7. Photograph and diagram. Overview of the bridge in LaGrange, GA............................ 16 Figure 8. Photographs. Received ultrasonic signals and autocorrelation function. ...................... 19 Figure 9. Photographs. Installation of a Martlet wireless ultrasonic device for long-term
thickness monitoring.............................................................................................................. 21 Figure 10. Plot. Daily history of the web thickness measurements on the bridge in
LaGrange, GA........................................................................................................................ 22 Figure 11. Photograph and drawing. Overview of the bridge in Douglas County, GA................ 24 Figure 12. Diagrams. Location of four wireless sensing units. .................................................... 25 Figure 13. Photographs and diagrams. Design and installation of a solar panel and its
support structure. ................................................................................................................... 26 Figure 14. Photograph. Enclosure with the gateway computer and solar charging battery.......... 27 Figure 15. Photographs. Installation of four wireless ultrasonic units. ........................................ 29 Figure 16. Plots. Ultrasonic waveforms and autocorrelation function obtained from four
wireless sensing units. ........................................................................................................... 31 Figure 17. Plots. Daily history of thickness measurements on the Douglas County bridge. ........ 34
LIST OF TABLES
Table 1. Comparison of thickness measurements on various specimens without calibration. ......10 Table 2. Comparison of thickness measurements on various specimens after calibration. ...........11 Table 3. Comparison of two thickness measurement devices. ......................................................12 Table 4. Comparison of measurement results................................................................................15 Table 5. Thickness measurement results on a steel girder bridge in Span 4. ................................20 Table 6. Summary of thickness measurement results from four wireless sensing units. ..............32
v

EXECUTIVE SUMMARY
This report presents the field validation of a wireless ultrasonic thickness measurement system on in-service steel bridges for long-term corrosion monitoring. The developed system utilizes Martlet wireless ultrasonic sensing devices with two specially designed daughterboards: the high-rate ultrasonic board and the pulser board. The compact ultrasonic sensing device is capable of highvoltage pulse excitation, filtering/amplification of the received ultrasonic signal, and high-speed analog-to-digital conversions (up to 80 MHz). To ensure accurate measurements, the project first derives the calibration function of the Martlet ultrasonic device. Subsequently, thickness measurements are carried out on 14 steel members with various thickness values. Measurements from the Martlet ultrasonic device are compared with readings obtained from a commercially available handheld thickness gauge for further validation.
Field testing is first conducted on a highway bridge in LaGrange, GA. The Martlet ultrasonic device is confirmed to provide the accurate thickness values of the web and bottom flange of a steel girder. This bridge has electricity available, allowing convenient implementation of the longterm monitoring system. A Martlet wireless ultrasonic device and a 2.25 MHz dual-element transducer are installed at the web of a steel girder as a preliminary validation of the long-term monitoring system. The thickness measurements have been reliably obtained for about seven months.
Following the successful operation confirmed on the first bridge, this project proceeds to implement the system on a second bridge, which is located in Douglas County, GA. Visual inspection of the bridge confirms corrosion on structural members across the bridge. A solar panel and a support structure are installed to provide continuous power for the battery and gateway
1

computer. Four wireless ultrasonic sensing units are installed on three pile tops and the bottom flange of a beam on the bridge. Ultrasonic data are collected at scheduled intervals and automatically uploaded to the cloud, enabling remote monitoring of the thickness values over time. Through wireless ultrasonic sensing devices and a gateway computer installed on-site, this research validates the feasibility of continuous steel thickness measurement that can be monitored remotely.
2

CHAPTER 1. INTRODUCTION
Recent advancements in wireless sensing technology have provided structural health monitoring systems with a cost-effective and efficient way to collect, analyze, and store data.[1] Wireless sensors can be installed at critical locations across a structure to monitor its condition. The collected sensor data can be wirelessly transmitted to the cloud, allowing engineers to monitor the structural condition and identify potential problems in real time.
Ultrasonic thickness measurement is a technique that can measure the thickness of a metal plate, as shown in figure 1. The technique works by sending ultrasonic waves through the material and measuring the time of flight (ToF), which is the time it takes for the waves to travel back and forth. The identified ToF is then used to calculate the thickness of the specimen. A thin layer of viscous or liquid gel (known as couplant) is typically applied between the transducer and specimen to facilitate the propagation of solid waves.

TrTarnanssdduucceerr Couplant Gel Couplant
SMpeectaiml en Specimen

Time of TimFeligohft Flight (ToF)

Initial

Back wall

pRuleseceived Sirgenflaecl tion

Figure 1. Diagram. Ultrasonic thickness measurement.

3

Regularly assessing corrosion damage on bridge structural members is crucial to making informed maintenance decisions. However, current practices primarily rely on human visual inspection, which is labor-intensive and expensive, mainly due to the difficulties in regularly accessing the underside of the bridge deck. The main goal of this project is to develop and implement long-term ultrasonic thickness measurement systems on steel members of in-service bridges using Martlet wireless sensing devices. The measured ultrasonic data are uploaded to the cloud through an onsite gateway computer and can be accessed remotely for decision-making.
4

CHAPTER 2. LABORATORY STUDIES OF MARTLET WIRELESS ULTRASONIC SENSING DEVICE
This chapter begins with a description of the Martlet wireless ultrasonic sensing device. The calibration function of the Martlet ultrasonic device is then derived, using a steel calibration block. Finally, laboratory experiments compare thickness measurement values between the Martlet ultrasonic device and a commercial handheld device. MARTLET ULTRASONIC WIRELESS SENSING UNIT This section describes the Martlet ultrasonic thickness measurement device. Figure 2 shows the functional diagram. The Martlet wireless sensing system is a low-cost platform for intelligent infrastructure monitoring.[2] As shown in figure 2, the Martlet motherboard incorporates a dualcore Texas Instruments Piccolo microcontroller, which can run up to 90 MHz. A Zigbee radio transmits data at rates up to 250 kbps. With Martlet's modular design, researchers have developed a variety of stackable daughterboards to interface with various sensors used in structural health monitoring.[3-7] Recently, the authors and collaborators have developed two daughterboards for ultrasonic thickness measurement: the high-rate ultrasonic board and the pulser board.[6, 8] These two daughterboards, together with the Martlet motherboard, form a Martlet ultrasonic thickness measurement device.
5

Transfer of sampled high-rate data through SPI interface

ADC Module Shift register

High-Rate Ultrasonic Board

Memory buffer

High-speed ADC data (up to 80MHz
sampling rate)

Receiving Module
4th order lowpass filter

Amplifier 10dB/20dB/
30dB

1st order high-pass
filter

Martlet motherboard Command signal 0~3.3V

Excitation Module (for other applications)

Driver MOSFET
5V regulator

Power MOSFET
Potentiometer to adjust peak
voltage

12V power supply

Pulser Board Short pulse up to 200V

DC-to-DC converter (0~200V)

Dual element ultrasonic transducer

Excitation port

Receiving port

Specimen

Figure 2. Diagram. Functional diagram of the Martlet ultrasonic thickness measurement device.

The high-rate ultrasonic board consists of three modules: the excitation module, the receiving module, and the analog-to-digital converter (ADC) module. The excitation module was designed for launching surface waves and is not utilized in this study for thickness measurement. Instead, the pulser board serves as the transducer excitation source for measuring specimen thickness. The receiving module conditions the signal from a transducer through a first-order high-pass filter, amplitude amplification (10dB/20dB/30dB), and a fourth-order low-pass filter. Among the commonly used standard filter types, Bessel is selected for the fourth-order low-pass filter that is critical for the signal conditioning performance. Bessel filters offer a linear phase performance and help maintain the signal waveform in the time domain, which assists in accurately identifying the ToF of the received signal.[9] The high-speed ADC module samples the filtered and amplified signal with a sampling frequency up to 80 MHz. The sampled data are transferred to the Martlet

6

motherboard through a serial peripheral interface (SPI) connection. Ultimately, the Martlet motherboard wirelessly sends the data to a base station connected to a personal computer (PC).
When the ultrasonic solid wave propagates into the specimen, the amplitude of the ultrasonic signal decreases as the wave reflects multiple times between the two surfaces of the specimen. A larger signal amplitude is preferred to ensure a good signal-to-noise ratio, resulting in a more accurate thickness measurement. For this purpose, a compact pulser board has been developed to generate a short-pulse excitation at high voltage. The high-voltage direct current (DC)-to-DC converter increases the excitation voltage up to 200V. An onboard potentiometer can easily adjust the voltage. The metal-oxide semiconductor field-effect transistor (MOSFET) is a power amplifier that accepts a low-power command signal from a microcontroller on the Martlet and produces a high-current drive input for the gate of the MOSFET. The MOSFET initiated by the high-current drive input achieves the fast-switching time required by the pulse excitation. An example pulse excitation signal generated by the developed pulser board can provide a 200V pulse with about 1 s duration. With this high excitation voltage, the ultrasonic signal can achieve a better signal-to-noise ratio, which helps improve the accuracy of the ultrasonic thickness measurement. Note that the pulser board requires an external 12 V power supply separated from the power supply to the motherboard and the high-rate ultrasonic board.
Figure 3 shows the printed circuit board (PCB) design of the Martlet ultrasonic thickness measurement device that consists of the Martlet motherboard, the high-rate ultrasonic board, and the pulser board. The planar dimension of the motherboard is 2.5 inch 2.35 inch. The pulser and high-rate ultrasonic boards can stack on the Martlet motherboard through the four corner connectors. The device connects with a dual-element transducer (as illustrated in figure 1) with excitation and receiving ports. In this study, a 2.25 MHz dual-element transducer is selected.
7

(External comma nd input)

Pulse out put

DC-to-DC converter
Pulser board
(External Receiving excitation por t) port
Receiving module

80MHz ADC chip

High-rate ultrasonic boa rd

Bottom view of the ultrasonic board

Martlet motherboard
Figure 3. Photographs. Printed circuit board design of the Martlet wireless ultrasonic device.
DERIVATION OF CALIBRATION FUNCTION Generally, it is necessary to calibrate thickness measurement values to identify the offset value specific to the instrument, transducer type, and ultrasonic characteristics. This procedure, known as calibration, is crucial for achieving accurate ultrasonic thickness measurements. Most commercial thickness measurement devices have built-in calibration features to improve the accuracy of measurements. To establish a calibration function for the Martlet ultrasonic device, we conduct thickness measurements on a 10-step steel calibration block. The block encompasses a range of thickness values from 0.1 to 1.0 inch, as shown in figure 4.

8

Figure 4. Photograph. Steel calibration block from 0.1- to 1.0-inch thicknesses.
The calibration block is made of 1018 steel, for which the nominal velocity is set as 0.2330 inch/ s .[10] With the fixed velocity value, the calibration function is derived for the measured ToF values. Specifically, we compare the true ToF (true thickness divided by 0.2330 inch/ s ) and the ToF measured by the Martlet ultrasonic device. Table 1 summarizes the comparison of ToFs for each of the 10 thickness values. Figure 5 plots error percentages obtained from table 1. It is evident that the error percentage is more significant for smaller ToFs corresponding to thinner thicknesses (less than 0.4 inch). This observation can be attributed to the 2.25 MHz frequency of the transducer. In general, higher frequency transducers offer better resolution and can lead to smaller errors. However, it is important to note that higher frequency transducers are limited in their ability to penetrate thicker specimens. Given the practical thickness range of actual bridge members, which falls between 0.4 and 1.0 inch, the transducer with a frequency of 2.25 MHz was selected for this study.
9

Table 1. Comparison of thickness measurements on various specimens without calibration.

Calibration Block

True thickness True ToF

(inch)

(s)

0.10

0.858

0.20

1.717

0.30

2.575

0.40

3.433

0.50

4.292

0.60

5.150

0.70

6.009

0.80

6.867

0.90

7.725

1.00

8.584

Martlet Ultrasonic Device

Measured

Measured

thickness (inch) ToF (s)

Error (%)

0.0961

0.825

-3.888

0.1937

1.663

-3.159

0.2913

2.500

-2.917

0.3903

3.350

-2.431

0.4908

4.213

-1.849

0.5912

5.075

-1.460

0.6903

5.925

-1.391

0.7893

6.775

-1.339

0.8898

7.638

-1.137

0.9904

8.500

-0.975

Figure 5. Graph. Measured ToF vs error (without calibration).

Based on data in figure 5, the error percentage function is assumed to be a third-degree polynomial. The coefficients of the polynomial function are determined by linear regression using the least squares method:

() = 0.000943 - 0.05972 + 0.8519 - 4.539

(1)

10

where () is the error percentage function and is the uncalibrated measured ToF (unit: s). Finally, the calibrated thickness (unit: inch) is obtained as follows:

Calibrated Thickness =

ToF Nominal Velocity (1 - ()/100) 2

(2)

Using equation 2, calibrated thickness values of the calibration steel block are obtained, as shown

in table 2. The maximum absolute error in calibrated thickness measurements is effectively

reduced to 0.233 percent. The following section validates the calibrated thickness measurements

using different steel specimens.

Table 2. Comparison of thickness measurements on various specimens after calibration.

Calibration Block

True thickness True ToF

(inch)

(s)

0.10

0.858

0.20

1.717

0.30

2.575

0.40

3.433

0.50

4.292

0.60

5.150

0.70

6.009

0.80

6.867

0.90

7.725

1.00

8.584

Martlet Ultrasonic Device

Calibrated Calibrated thickness (inch) ToF (s)

Error (%)

0.0998

0.8570

-0.200

0.2000

1.7171

0

0..2993

2.5692

-0.233

0.3993

3.4277

-0.175

0.5003

4.2942

0.060

0.6009

5.1577

0.150

0.6999

6.0074

-0.014

0.7989

6.8573

-0.138

0.8995

7.7211

-0.056

1.0005

8.5877

0.050

COMPARISON WITH A HANDHELD THICKNESS MEASUREMENT GAUGE This section compares the measurements obtained by the Martlet ultrasonic device with a handheld thickness gauge currently on the market (table 3). Considering its price range ($1000), which is comparable to the Martlet ultrasonic device, we purchased a commercial handheld thickness-

11

measurement device. Although the two devices show similar specifications, a key difference lies in the measurement mode. The commercial handheld device can only measure in the pulse-echo mode, whereas the Martlet device is designed to measure in the echo-to-echo mode. Pulse-echo mode utilizes the time interval between the excitation pulse to the first arriving ultrasonic echo signal. In contrast, the echo-to-echo mode utilizes the time interval between the neighboring ultrasonic echoes. In general, a device that can perform echo-to-echo measurement costs between $2000 and $3000 on the market, as echo-to-echo measurement usually provides greater accuracy and can ignore the effect of paints on the thickness measurement of steel members.

Table 3. Comparison of two thickness measurement devices.

Martlet Wireless Ultrasonic Device

Commercial Handheld Device

Overview

Sampling frequency Excitation Transducer
Measurement mode

80 MHz 200V pulse wave 2.25 MHz dual
Echo-to-echo

120 MHz 150V square wave
2.25 MHz dual Pulse-echo

Measurements obtained from the two devices were compared using steel plates with various thickness values. We prepared 14 different sizes of structural steel at the Georgia Institute of Technology (Georgia Tech) Structural Engineering and Materials Laboratory (Structures Lab) (figure 6). All the specimens are made of carbon steel, with a nominal velocity of 0.2339 inch/s

12

and thicknesses ranging 0.175 to 0.1027 inch.[11] Before conducting measurements, we sanded the measurement surface to ensure smoothness. Each thickness was first measured using a caliper to obtain a reference value for the true thickness. Subsequently, thickness measurements were performed using the Martlet ultrasonic device with the calibrated ToF and the handheld device.

a. Specimen 1

b. Specimen 2

c. Specimen 3

d. Specimen 4

e. Specimen 5

f. Specimen 6

Figure 6. Photographs. Steel specimens collected from the Structures Lab at Georgia Tech. (Continued on the next page)

13

g. Specimen 7

h. Specimen 8

i. Specimen 9

j. Specimen 10

k. Specimen 11

l. Specimen 12

m. Specimen 13

n. Specimen 14 Figure 6. (Continued.)

Table 4 summarizes the thickness measurement results from the caliper ("True" column), Martlet ultrasonic device, and the commercial handheld device. For 10 of 14 specimens, the Martlet

14

ultrasonic device produced better accuracy than the handheld device. This result demonstrates the effectiveness of the calibration function derived in the previous section and the echo-to-echo measurement mode implemented by the Martlet device, which, in general, produces better accuracy than the pulse-echo measurement mode used by the commercial handheld device.

Table 4. Comparison of measurement results.

Specimen 1

True (inch)
0.175

Martlet Ultrasonic Device

Measured (inch)
0.1739

Error* (%)
0.63

2

0.239

0.2361

1.21

3

0.240

0.2391

0.37

4

0.275

0.2768

0.65

5

0.325

0.3183

2.06

6

0.505

0.5066

0.32

7

0.548

0.5446

0.62

8

0.549

0.5446

0.80

9

0.571

0.5667

0.75

10

0.623

0.6197

0.53

11

0.767

0.7656

0.18

12

0.950

0.953

0.32

13

1.000

1.0012

0.12

14

1.027

1.0244

0.25

* Bold indicates the smaller error value for the specimen.

Commercial Handheld

Device

Measured

Error*

(inch)

(%)

0.172

1.71

0.229

4.18

0.266

10.83

0.280

1.82

0.328

0.92

0.502

0.59

0.547

0.18

0.543

1.09

0.565

1.05

0.617

0.96

0.768

0.13

0.948

0.21

0.995

0.50

1.037

0.97

15

CHAPTER 3. PRELIMINARY VALIDATION OF CONTINUOUS WIRELESS THICKNESS MEASUREMENTS ON THE LAGRANGE BRIDGE
This chapter reports field thickness measurements and preliminary instrumentation of the longterm monitoring system on the first testbed bridge in LaGrange, GA. The chapter begins with a description of the bridge, followed by results of thickness measurements taken from the web and the flange of a girder, and finally, provides long-term thickness monitoring results obtained from the web of a steel girder over about seven months.
TESTBED BRIDGE IN LAGRANGE, GA

Span 1

Span 2

Span 3 Span 4

a. Overview

b. Elevation Figure 7. Photograph and diagram. Overview of the bridge in LaGrange, GA.
16

Figure 7 shows the overview (figure 7a), and the elevation (figure 7b) of the first testbed bridge (Structure ID 285-0067-0) investigated in this study. The bridge was built in 1977 and is located in LaGrange, GA. The condition rating by the National Bridge Inventory for this bridge is 7 Good Condition. The bridge superstructure consists of six steel beams and a reinforced concrete deck. The bridge has four spans: two simply supported end spans and two continuous middle spans.
FIELD THICKNESS MEASUREMENTS Ultrasonic thickness measurements were first conducted on the web and bottom flange of a steel girder located at Span 4. A 2.25 MHz dual element transducer was employed and connected to the Martlet ultrasonic device. The structural design documents show that Span 4 utilizes ASTM A36 carbon steel with a W36135 section. The nominal thickness of the W36135 section is 0.80 inch for the bottom flange and 0.60 inch for the web. Sand and dust have accumulated on the surface of the bottom flange, as is often the case in practice, and make the thickness measurement more challenging compared to the clean surface of the web.[12] The entire girder has paint coatings. Due to the presence of coatings, the apparent thickness is larger than the nominal thickness of the steel itself. Therefore, instead of conducting velocity calibration by caliper measurements, this study uses the nominal velocity 0.2339 inch/s for carbon steel.[13]
Figure 8a shows the received signals of the 2.25 MHz dual-element transducer sampled by the high-rate ultrasonic board for the 0.60-inch-thick web with a clean surface. The received signal includes a sequence of echoes, which are the reflections of the ultrasonic waves created by the transducer. The time interval between the neighboring echoes is the ToF. To accurately obtain the ToF, we calculate the autocorrelation function of the received signal using the following equation:
17

-

[] = [ + ][] , = 0, ... ,

(3)

=0

where [] is the autocorrelation function at discrete-time lag ; [] is the received voltage signal

at the time step ; and is the total number of data points. Based on the peak value of the

autocorrelation function, the uncalibrated ToF is identified as 5.08 s for the web. Although

alternatively one could estimate the ToF using the time difference between the first and second

peaks in the received ultrasonic signal, the autocorrelation function is generally more robust

against noise and with rough surface conditions.

Similarly, figure 8b shows the received signals and autocorrelation function for the 0.80-inch-thick bottom flange with a dusty surface. Note that we did not clean the dust on the bottom flange. Therefore, autocorrelation waveforms are slightly distorted compared to the web due to the layer of dust. However, the peak is clearly identified at 6.82 s.

18

ToF ToF

ToF = 5.08

a. 0.60-inch-thick web with clean surface

Dust on the bottom flange

ToF ToF

ToF = 6.82
b. 0.80-inch-thick bottom flange with dusty surface Figure 8. Photographs. Received ultrasonic signals and autocorrelation function. The thickness of each specimen is calculated by equation 2 and summarized in table 5. For the 0.60-inch-thick web, the estimated thickness is 0.6037 inch. The difference from the nominal thickness of 0.60 inch is only 0.0037 inch (0.62 percent). For the 0.80-inch-thick bottom flange, the estimated thickness is 0.8071 inch, only 0.0071 inch (0.89 percent) different from the nominal thickness of 0.80 inch. In general, actual thickness values may vary from the nominal value within
19

an allowable tolerance, as specified by ASTM-A6/A6M-14.[11] The measurements obtained both on the dusty web and the clean flange are well within their allowable tolerance of 0.03 inch from the corresponding nominal thickness. Therefore, this field test validates the performance of the developed Martlet ultrasonic device in the presence of coatings and accumulated dust on the steel surface.

Table 5. Thickness measurement results on a steel girder bridge in Span 4.

Uncalibrated time of flight (ToF) Calibrated time of flight (ToF) Nominal velocity Estimated thickness Difference from nominal thickness

0.60-inch-thick Web (clean) 5.08 s 5.16 s 0.2339 inch/s
0.6037 inch
0.62%

0.80-inch-thick Bottom Flange (dusty)
6.82 s 6.90 s 0.2339 inch/s
0.8071 inch
0.89%

LONG-TERM ULTRASONIC THICKNESS MEASUREMENTS The bridge happens to have electricity available, allowing convenient implementation of the longterm monitoring system in this preliminary investigation. A Martlet ultrasonic device is installed on the web of a girder located in Span 2.[14] Note that Span 2 utilizes a W135150 steel section with a nominal web thickness of 0.625 inch. The Martlet unit establishes Zigbee wireless communication with the gateway computer installed at the edge of Span 1 of the bridge. The gateway is connected to a 4G LTE network, enabling the collected data to be uploaded to the cloud for subsequent analysis.
Figure 9 shows the installed Martlet wireless ultrasonic device in Span 2. A 2.25 MHz dualelement ultrasonic transducer is installed together with a magnet mount to ensure firm contact between the transducer and the steel surface. A wireless antenna and a 12V battery are placed next
20

to the wireless unit. The 12V battery is connected to a pulser board via a relay switch, ensuring battery power is only consumed during brief measurement intervals that last a few seconds. Consequently, the battery lifespan is significantly extended. Between the measurement surface and the transducer, we applied a paste-like couplant (more viscous than commonly used gel) suitable for long-term monitoring.

Magnet mount

Transducer

12V battery

Martlet

Couplant for long-term monitoring
Antenna

Figure 9. Photographs. Installation of a Martlet wireless ultrasonic device for long-term thickness monitoring.
The thickness measurements are taken at scheduled time intervals for long-term monitoring. Figure 10 shows the daily thickness values recorded from November 21, 2022, to June 13, 2023. Overall, stable measurements are obtained around 0.64 inch, close to the 0.625-inch nominal thickness. The ultrasonic measurement system has a resolution of 0.0015 inch. Minor fluctuations in the measured thickness values, either slightly higher or lower by 0.0015 inch, are observed from December 2022 to February 2023. These variations could be attributed to the influence of cold
21

weather conditions on the couplant during the winter months or the couplant material taking time to reach a stable state until February 2023. Over the subsequent three months, the measurement values have remained stable, indicating the successful operation of the long-term thickness measurement system. This field testing validates the long-term ultrasonic thickness measurement system on a regular highway bridge and serves as the preliminary validation. Chapter 4 describes installation of the system to a second testbed bridge with corrosion.
Figure 10. Plot. Daily history of the web thickness measurements on the bridge in LaGrange, GA
22

CHAPTER 4. LONG-TERM WIRELESS THICKNESS MEASUREMENTS ON THE DOUGLAS COUNTY BRIDGE
Chapter 3 confirmed the successful operation of long-term thickness measurement on a bridge in good condition. This chapter presents the implementation of the long-term thickness measurement system on another testbed bridge with corrosion. The chapter first describes the testbed bridge in Douglas County, GA, and then provides the installation details followed by the long-term thickness measurement results from four installed wireless sensing units. TESTBED BRIDGE IN DOUGLAS COUNTY Figure 11 provides an overview of the bridge located in Douglas County, GA. The bridge consists of eight bents and five beams, forming a composite structure with a reinforced concrete deck. The middle two bents, Bents 4 and 5, are positioned above the river from the Bear Creek Reservoir, close to the bridge. These two middle columns (supporting Bents 4 and 5) are made of concrete, whereas other columns are made of wide flange steel sections. The bridge (Structure ID # 0970013-0) was constructed in 1957, making it 20 years older than the bridge investigated in chapter 3. The National Bridge Inventory condition rating for this bridge is 5 Fair Condition.
23

Bent 2

Bent 5 Bent 3
Bent 4

a. Overview
Bent 1 Bent 2 Bent 3 Bent 4 Bent 5 Bent 6 Bent 7 Bent 8

b. Elevation Figure 11. Photograph and drawing. Overview of the bridge in Douglas County, GA.
INSTALLATION Figure 12 shows four wireless sensing units (U152, U156, U164, and U166) installed on this bridge. The long-term thickness monitoring system also includes a gateway enclosure with a rechargeable battery powered by a solar panel. The details of each of these components are explained in the following subsections.
24

West

Receiver

nclosure Gateway PC Solar charging battery Arduino Modem 12V battery Relay switch etc.

U164

U166

U152

Bent 1

Bent 2

(abutment)

Bent 2

Bent 3
Bent 3

Bent 4
South
Bent 4 (concrete)

Bent 5

U152 U156 U164

U166

Bent 6

Beam 2
Beam 3 East
Beam 4

Bent 7

Beam 5 Bent 8 (abutment)

Figure 12. Diagrams. Location of four wireless sensing units.

Solar Panel and Support Structure The long-term thickness monitoring system requires a continuous power supply for the operation of the gateway computer. However, unlike the previous bridge in LaGrange, which had electricity available, the bridge in Dougals County lacks such a power source. Consequently, a solar panel and solar charging battery system are installed on this bridge as part of the current project. We select a 200W solar panel to meet the power requirements for the continuous operation of the system. A solar panel support structure is designed to mount the solar panel in the middle of the bridge. Figure 13 shows the installed solar panel and support structure in the middle of the bridge between Bents 4 and 5. This location provides an unobstructed view for the panel, ensuring optimal exposure to sunlight. The solar panel is oriented in the south direction for maximum charging efficiency.

25

South

Solar panel support structure between Bent 4 and 5
North

Figure 13. Photographs and diagrams. Design and installation of a solar panel and its support structure.
Enclosure and Devices As shown in figure 12, a steel enclosure has been installed at the west end of the bridge. Figure 14 shows photographs of the enclosure securely attached to the pole sign, housing various devices inside. A modem is housed in a waterproof case and attached to the pole outside to receive a reliable 4G LTE network for uploading collected ultrasonic data to the cloud. Within the enclosure is a large 2400Wh battery that is charged by the solar panel during the daytime. A charging controller is connected between the panel and the battery to manually monitor charging efficiency and protect the panel from electrical damage. The gateway PC is powered by the battery through an inverter. The enclosure is equipped with a built-in lock and an external padlock for security purposes.
26

Gateway PC

Enclosure Modem

Charging controller
Solar charging battery

Figure 14. Photograph. Enclosure with the gateway computer and solar charging battery.
Wireless Ultrasonic Sensing Units Prior to the installation in April 2023, the Georgia Tech and Georgia Department of Transportation (GDOT) team conduct a visual inspection in October 2022 to evaluate the level of corrosion on structural members throughout the bridge. During the inspection, corrosion was observed on the west part of the bridge. In response to these findings, four wireless sensing units, namely U152, U156, U164, and U166, are installed across Bent 2 to Bent 4, as illustrated in figure 12. This subsection provides a summary of the installation locations, pictures, collected ultrasonic waveforms, and thickness measurement values for each sensor unit upon installation.

27

Figure 15 shows photographs of the four installed wireless ultrasonic sensing units. Wireless unit U152 is installed on the top of Pile 4 on Bent 2, as shown in figure 15a. The installed location exhibits relatively healthy conditions. In the meantime, unit U156 is installed on the top of Pile 1 on Bent 3 (figure 15b), which is one of the more corroded pile tops on the bridge, as we can confirm a through-hole on its top. The motivation of installation at this location is to monitor the rate of corrosion on this corroded pile over time. Figure 15c illustrates unit U164 installed on the top of Pile 3 on Bent 3. Finally, unit U166 is installed on the bottom flange of Beam 3 on Bent 4 (figure 15d), as this bottom flange is one of the more corroded flanges on the bridge.
28

a. U152 installed on the top of Pile 4 on Bent 2
b. U156 installed on the top of Pile 1 on Bent 3 Figure 15. Photographs. Installation of four wireless ultrasonic units.
(Continued on the next page) 29

c. U164 installed on the top of Pile 3 on Bent 3
d. U166 installed on the bottom flange of Beam 3 on Bent 4 Figure 15. (Continued). 30

a. U152 installed on the top of Pile 4 on Bent 2
b. U156 installed on the top of Pile 1 on Bent 3
c. U164 installed on the top of Pile 3 on Bent 3
d. U166 installed on the bottom flange of Beam 3 on Bent 4 Figure 16. Plots. Ultrasonic waveforms and autocorrelation function
obtained from four wireless sensing units. Upon installation, ultrasonic waveforms are collected to validate the measurements, as shown in figure 16. The autocorrelation function is calculated to automatically identify the peak, which
31

corresponds to the uncalibrated ToF. We confirm that ultrasonic signals are reliably obtained from the installed four units. The waveforms obtained from U166 are slightly distorted due to the rough surface conditions at the bottom flange.

Corresponding to ultrasonic waveforms in figure 16, thickness values are calculated using equation 1 and the nominal velocity 0.2339 inch/s. Table 6 provides a summary of the thickness measurement values from the four wireless ultrasonic sensing units. As the pile tops exhibit relatively healthy conditions, the thickness values measured by U152, U156, and U164 closely match the nominal thickness of 0.435 inch. The differences between the measured and nominal thicknesses are confirmed to be within the manufacturer's tolerance of 0.03 inch. On the other hand, for the corroded bottom flange measured by U166, the remaining thickness is reduced by nearly 30%, which can also be visually confirmed from figure 16d.

Table 6. Summary of thickness measurement results from four wireless sensing units.

Unit Number
U152 U156 U164 U166

Calibrated ToF () 3.943
3.541
3.692
4.507

Measured Thickness
(inch) 0.4611
0.4140
0.4317
0.5270

Nominal Thickness
(inch) 0.4350
0.4350
0.4350
0.7450

Difference from the Nominal Thickness
(inch) 0.026 (6.0%)
-0.021 (-4.8%)
-0.013 (-0.8%)
-0.198 (-29.3%)

LONG-TERM THICKNESS MEASUREMENT RESULTS In this section, we present the long-term thickness measurement results obtained from the four wireless ultrasonic sensing units. The gateway computer is configured to initiate ultrasonic data collection at scheduled intervals for each sensor. The collected ultrasonic data are then uploaded to the cloud.

32

Figure 17 displays the daily history of thickness measurement values obtained from April 4 to July 23, 2023. Three units U152, U156, and U164 installed at pile tops have provided consistent measurements over the last three months (up to the time of this report). However, reliable measurements from U166 have ceased near end of May. Soldering quality on the prototype circuit board is suspected to be the cause; replacement of the circuit board would resolve the issue. Some data points are missing due to the challenge with the solar charging battery. Specifically, the battery was configured by the manufacturer to occasionally shut off, resulting in data gaps. To resolve this issue, we replaced the battery with one from a different manufacturer on July 10, 2023. The battery issue was successfully resolved after this replacement, and measurements are now being obtained consistently.
33

Figure 17. Plots. Daily history of thickness measurements on the Douglas County bridge. 34

CHAPTER 5. CONCLUSIONS
This project implemented long-term wireless ultrasonic thickness measurement systems on two testbed bridges in Georgia. The summary and conclusions are made as follows:
The developed system consists of the Martlet wireless ultrasonic device, a gateway computer, and a solar charging battery. The system is configured to collect ultrasonic thickness data of steel bridge members and upload it to the cloud automatically. The developed system allows the remote monitoring of thickness values over time without physically accessing measurement locations.
On the bridge in LaGrange, thickness values were recorded over about seven months. Although the measurements stabilized in the last three months, we observed slight variations for the first few months of installation. These variations could be attributed to cold weather conditions during the winter months or the couplant material requiring time to achieve a stable state.
A second testbed bridge in Douglas County exhibits corrosion conditions on structural members. Three wireless sensing units have obtained reliable measurements since the installation and are still functioning properly till the time of this report's preparation. Due to the short duration of the project, the measurement values have not yet shown any changes in thickness values.
35

ACKNOWLEDGEMENTS The project team thanks the Georgia Department of Transportation for its support and resources provided toward the success of this project. Specifically, we appreciate the technical supervision, feedback, and insights provided by Mr. Donn P. Digamon and Mr. Rabindra Koirala, as well as the project management and support by Mr. Brennan A. Roney. In addition, many thanks are owed to the GDOT specialized bridge inspection team and district traffic control crews who helped make the field installation and testing possible. This includes, but is not limited, to Matt Butler, Dale Goff, Matthew Owen, and Lee Stevenson.
36

REFERENCES
1. Lynch, J. P. and Loh, K. J. (2006). "A Summary Review of Wireless Sensors and Sensor Networks for Structural Health Monitoring." The Shock and Vibration Digest, 38(2), pp. 91128. Available online: https://doi.org/10.1177/0583102406061499.
2. Kane, M., Zhu, D., Hirose, M., Dong, X., Winter, B., Hckell, M., Lynch, J. P., Wang, Y., and Swartz, A. R. (2014). "Development of an Extensible Dual-Core Wireless Sensing Node for Cyber-Physical Systems." Proceedings of SPIE 9061, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems, San Diego, CA, USA. Available online: https://doi.org/10.1117/12.2045325.
3. Dong, X., Zhu, D., Wang, Y., Lynch, J. P., and Swartz, A. R. (2014). "Design and Validation of Acceleration Measurement Using the Martlet Wireless Sensing System." Proceedings of the ASME 2014 Smart Materials, Adaptive Structures and Intelligent Systems (SMASIS), Newport, RI, USA. Available online: https://doi.org/10.1115/SMASIS2014-7611.
4. Lander, P., Fahed, N., and Wang, Y. (2022). "Martlet Wireless Sensing System for Full Scale Bridge Weigh-in-Motion." Proceedings of SPIE 12046, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems, Long Beach, CA, USA. Available online: https://doi.org/10.1117/12.2612610.
5. Liu, X., Dong, X., Wang, Y., Stewart, L., Dodson, J., and Joyce, B. (2018). "High-G Shock Acceleration Measurement Using Martlet Wireless Sensing System." Proceedings of the 36th International Modal Analysis Conference (IMAC XXXVI), Orlando, FL, USA.
6. Otsuki, Y., Lander, P., and Wang, Y. (2021). "Ultrasonic Thickness Measurement Using the Martlet Wireless Sensing System." Proceedings of the 2021 IEEE International Instrumentation & Measurement Technology Conference (I2MTC), Glasgow, Scotland. Available online: https://doi.org/10.1109/I2MTC50364.2021.9460094.
7. Otsuki, Y., Nguyen, S. T., La, H. M., and Wang, Y. (2023). "Autonomous Ultrasonic Thickness Measurement of Steel Bridge Members Using a Climbing Bicycle Robot." ASCE Journal of Engineering Mechanics, 149(8), p. 04023051. Available online: http://dx.doi.org/10.1061/JENMDT.EMENG-7000.
8. Otsuki, Y., Lander, P., and Wang, Y. (2022). "Field Validation of a Compact Martlet Wireless Ultrasonic Thickness Measurement System." Proceedings of the 8th World Conference on Structural Control and Monitoring (8WCSCM), Orlando, FL, USA.
9. Horowitz, P. and Hill, W. (1989). The Art of Electronics (2nd ed). Cambridge University Press, Cambridge, UK.
10. NDT Supply.com Inc. (2023). "Ultrasonic Longitudinal Velocity Table for Thickness Gauging." Available online: https://ndtsupply.com/.
37

11. ASTM-A6/A6M-14 (2014). Standard Specification for General Requirements for Rolled Structural Steel Bars, Plates, Shapes, and Sheet Piling. ASTM International, West Conshohocken, PA, USA.
12. Kulicki, J. M., Prucz, Z., Sorgenfrei, D. F., Mertz, D. R., and Young, W. T. (1990). Guidelines for Evaluating Corrosion Effects in Existing Steel Bridges, National Cooperative Highway Research Program Report 333, Transportation Research Board, Washington, DC, USA. Available online: https://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_rpt_333.pdf.
13. Aquil, A. and Leonard, J. B. (2018). ASM Handbook, Nondestructive Evaluation of Materials, vol. 17. ASM International, Materials Park, OH, USA.
14. Otsuki, Y., Sakpunpanom, K., and Wang, Y. (2023). "Long-Term Ultrasonic Thickness Measurement on an in-Service Bridge through a Wireless Sensor Network." Proceedings of SPIE 12486, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems, Long Beach, CA, USA. Available online: https://doi.org/10.1117/12.2657802.
38