Professional Documents
Culture Documents
Materials Evaluation-March 2018
Materials Evaluation-March 2018
VOL. 76 • NO. 3
ASNT… CREATING A SAFER WORLD!™
ASNT
27th RESEARCH
SYMPOSIUM
Orlando, Florida
26–29 March
Women in NDT
AE Source Localization
Time of Flight Diffraction
Phased Array UT
THE AMERICAN SOCIETY FOR NONDESTRUCTIVE TESTING
ESTING
MARCH 2018
VOLUME 76 • NUMBER 3
BACK TO BASICS
310
310 PAUT TEST
Using Iterative Time Reversal for NDT
Inspections on Composite Materials
Daniel Richard
FEATURE
ASNT IN ORLANDO
8
with the Dispersive A0 Mode based on 300
7
6
the Cross-Correlation Technique and 200
5
4
Time Reversal Principle 131.8
100
3
2
Zenghua Liu, Tuocan Dong, Qiuling Peng, Cunfu He, Qiufeng Li,
1
Bin Wu 0
0 20 40 60 80 100 120 140 160 180 200
Time (μs)
383 Reliability Investigation of Sizing
Horizontal Planar Discontinuities Using a 371
New Time of Flight Diffraction Technique
B.S. Hayat Gheib, M.J. Ranjbar Naserabadi
Thanks to the sponsors and players that participated in the George Pherigo Memorial Golf
Tournament, the North Carolina Student Section was able to send a group of six students
and Section advisor Rand Ernst to the ASNT Annual Conference. From left: Sport Frazier,
Simon Trinidad, Zachary Baze, Ashley Stowe, Tamra Culp, Thomas Johnson, and Rand Ernst.
Hampton Roads
NEWPORT NEWS, VIRGINIA Attendees of the Hampton Roads Section’s 14 November meeting enjoyed a presentation on
The Hampton Roads Section met on training and a briefing on the Annual Conference.
14 November 2017 to hear LeDon C.
Watkins present a briefing on the training
he performs as a Hellier instructor. Six
members and one guest attended. Bill Via, The retrospection discussion included India
regional director, briefed the membership participation in the November 2017 Kailua CHENNAI, INDIA
on the Annual Conference. The 2016–2017 High School Career Fair and the receipt of The India Section successfully conducted
Bronze President Award was delivered at a Section rebate. The action items discus- its third technical meeting of the year on
this meeting. sion focused on the Section newsletter. A 26 November 2017 at the Courtyard by
January 2018 issue is planned. Members Marriott Hyderabad. There were nearly
Hawaii are requested to provide a roster of recipi- 30 members in attendance, and the
HONOLULU, HAWAII ents, including a list for their clients. Editor meeting was chaired by Loganathan
The Hawaii Section held its December Jayme J. Martin will provide interview Thangavel, the vice chair of the Section.
quarterly meeting on 7 December 2017, at questions to two veteran members to The secretary gave an overview of the
Buzz’s Original Steakhouse in Pearl City, answer for the newsletter. The planned 2016–2017 activities of the Section to the
Hawaii. In attendance were 10 members 2018 quarterly meetings will be on members. The technical talk was given by
and guests. The agenda included the 2017 8 March, 7 June, 6 September, and Mallu Venkata Reddy from Defence
retrospection, action items, the 2018 6 December. A January potluck picnic Research and Development Laboratory,
schedule, and the January potluck picnic. had been in the works at a Pearl City park. and was titled “Advanced Radiography
The India Section’s third technical meeting of the 2017–2018 year included a talk on radiographic testing on aerospace components.
Testing on Aerospace Components.” The of pipeline that is laid underground Mohawk Hudson
meeting was sponsored by C/MOS within North America, and how much S C H E N E C T A D Y, N E W Y O R K
Processors, a channel partner of GE. work is involved with regards to main- The Mohawk Hudson Section held a
taining it. The meeting also included a meeting on 19 December 2017 at the Van
Metro NY/Northern New Jersey raffle giveaway. Dyck Restaurant and Lounge with 35 members
ROCKVILLE CENTRE, NEW YORK and guests in attendance. The meeting
The Metro NY/Northern New Jersey Section Miami Valley included a presentation titled “Advances
held a meeting on 29 November 2017 at C I N C I N N AT I , O H I O In Wind Turbine Blade Technologies” by
the Press 195 Restaurant in Rockville The Miami Valley Section met on Shridhar Nath of GE Global Research in
Center, New York. Section Chair Toni Bailey 8 November 2017 at Miami Valley Schenectady, New York. As wind blades are
opened the meeting with officer introduc- Gaming for a very informative presentation getting longer and longer, the industry is
tions. After dinner Larry E. Mullins, Sr. titled “Advances in Resonant Ultrasound looking at new ways to design and manufac-
gave a presentation on the role of NDE in Spectroscopy for Materials and Component ture these blades to keep them cost competi-
pipeline construction and maintenance, Characterization,” by Rasheed Adebisi, tive. The talk covered new technologies being
which was an informative topic that is from the University of Dayton Research introduced for the next generation of wind
relevant to the modern NDT field. Mullins Institute. Eighteen members came out turbine blades. Materials, additive manufac-
was a great speaker, and the presentation for the presentation. turing, instrumentation, automation, and
seemed to capture everyone’s attention; inspection technologies were some of the
attendees were surprised by the mileage topics discussed.
The Mohawk Hudson Section held a meeting on 19 December 2017 that included a presentation on wind turbine blade technologies.
Narragansett Bay There were 30 members in attendance. Past President John A. Stringer were also
PROVIDENCE, RHODE ISLAND Region 11 Director Brenda L. Collins of in attendance. All attending enjoyed the
The Narragansett Bay Section held a Sherwin, Inc. gave a talk on liquid talk and asked many questions.
meeting on 28 November 2017 at the penetrant inspection and its applications.
Atwood Grill in Johnston, Rhode Island. Region 1 Director John P. Moran and ASNT North Alabama
HUNTSVILLE, ALABAMA
The North Alabama Section held a meeting
on 14 November 2017 at the Redstone
Arsenal Test Center. The meeting consisted
of a technical discussion and tour, in
which attendees were able to view and
discuss radiographic and computed
tomography images of hardware, as well
as the test facility and inspection
equipment setups.
Old Dominion
RICHMOND, VIRGINIA
On 27 November 2017, the Old Dominion
Section held a technical meeting at
Calabash Seafood in Mechanicsville,
Virginia. Section member Robert E.
Campbell, Jr. gave a presentation on
revisions to the AWS B1 publications.
Campbell, who is the chair of the B1
subcommittee, covered numerous topics,
including the revision process to comply
with ANSI, documents under revision, and
The Southern Africa Section held its last meeting of 2017 at Fahrenheit Seafood and Grill.
attempts to combine several documents
From left: Section Treasurer Johannes Stephanus Beukes, Section Vice Chair Benjamin Beetge,
into a useful publication. Section Secretary Elaine MacDonald, Section Past Chair Garth Appel, and Section Chair Simon
Neil Wilding.
Pacific Northwest
SE ATT L E , WA S H I N GTO N (from the regional sales department). 2017 at Fahrenheit Seafood and Grill in
The Pacific Northwest Section held a Maziuk and Herrera gave a presentation on Benoni. There were a lot of achievements
meeting on 13 November 2017 at the Old the current state of computed tomography during the year, including three well-
Spaghetti Factory in Tukwila, Washington. (CT) and showed how an entire car can be attended technical evenings, a great
Twenty-two attendees enjoyed a presenta- imaged to reveal all of the different family fun day, the most successful golf
tion by two speakers from VJ Technologies: materials inside. CT imaging has improved day to date, and another ASNT Bronze
Robert Maziuk (Level III RTR, who manages greatly and is also capable of finding Award. Everyone is very excited for the
the high energy DR/CT facility in East porosity in plastics and metals in addition upcoming year.
Haven, Connecticut), and Mel Herrera to identifying cracks.
The Section held a meeting on United Arab Emirates
11 December 2017 at the Old Spaghetti D U B A I , U N I T E D A R A B E M I R AT ES
Factory that included a presentation on The United Arab Emirates Section held a
NDT automation. Automation is an ever- board meeting in Dubai on 9 December
expanding part of NDT technology, 2017. The meeting was attended by seven
increasing both speed and accuracy of members. Topics covered included
jobs and changing some of the skills creating a new Section logo with the assis-
required, which is creating more opportu- tance of the ASNT International Service
nity to learn and advance in particular Center, promoting ASNT to nonmember
disciplines and fields of interest. Brian companies, and the Section yearbook. w x
Ruhl and Whitney Moon from Genesis
Systems Group, LLC made a presentation
that was both interesting and educational.
The evening was made more festive by the
Correction
Board giving each of the 32 people in
attendance a beautiful Christmas mug On page 17 of the January 2018 issue
filled with candy. Everyone enjoyed the of Materials Evaluation, Clyde W. May
was incorrectly identified as Clyde
meeting while making connections and
Mays. May gave a presentation to the
new friendships with other NDT folks.
Pacific Northwest Section Chair Kathy Cleveland Section on 16 October; the
Ferguson (right) presented Robert Maziuk meeting was described in the Section
(left) and Mel Hererra (center) speaker gifts Southern Africa News department.
JOHANNESBURG, SOUTH AFRICA
for their presentation on electronics and Materials Evaluation regrets this error.
high-energy computed tomography at the The Southern Africa Section held its
Section’s 13 November meeting. Christmas dinner and final meeting of
David Blosser David L. Culbertson Juan R. Diaz Jerry Fullin William H. Spaulding
50 years. He served as the Aerospace Spaulding wrote six NDT books for About the Mentoring Recognition
Level 3 Examiner for 300 global NDT Honeywell, which were study guides. He
persons, and was the director for QA also coauthored ASNT Level II Study Guide: Mentoring is an important activity that the
special processes, auditing, NDT, and Liquid Penetrant Testing Method, second Section Operations Council of ASNT honors
quality systems. He established the edition, and ASNT Level II Study Guide: by offering an award to outstanding
Honeywell “NDE NETWORK,” with all site Radiographic Testing Method, second mentors of the past and present. Realizing
Level 3 personnel reporting to a single edition. In 1998, he received a US patent that mentors have probably touched every
management. The end result was a via Honeywell for manipulator systems person in the Society at some time, this
common/standardized NDT process that used for real-time X-ray. He has also recognition may be presented to more
reduced duplication, prevented errors, and presented technical papers on computer than one honoree. The maximum number
resulted in significant cost savings. He also image enhancement on radiographic of recipients in any given year will be five.
established the Honeywell Supplier Level 3 media and advanced eddy current testing The Mentoring Recognition was estab-
certification process (now known as the of bearing components at ASNT confer- lished to recognize those people in the
Source Certifying Agent program). While ences, as well as authored numerous Society working to encourage others to
working for Honeywell, Spaulding certified trade/quality magazine articles on NDT reach goals they may have otherwise not
over 500 supplier Level 3 personnel, and technologies. sought and to offer the rest of the
he is currently approved for Honeywell to membership an example of what they
support their supplier base. could be accomplishing by acting as
mentors. w x
Although perhaps best known for its theme parks, Orlando offers visitors a mix of cultural attractions, sports and recreation, and natural
beauty (Photo credit: Susanne Pommer / Shutterstock.com).
Student Participation
To help enable students to take advantage
of the educational and networking resources
that the Research Symposium provides, the
Society has offered travel reimbursements
of $1000 each to 15 students (the deadline
to apply for this reimbursement ended in
December). The Symposium is a great Universal’s CityWalk features a variety of dining, shopping, and live entertainment (Photo
opportunity for students to connect with credit: Kamira / Shutterstock.com).
The Symposium is within walking distance to Universal Studios Resort, which includes two theme parks, a water theme park, and an
entertainment district (Photo credit: Michael Gordon / Shutterstock).
1. Call to Order — Kevin D. Smith, Chairperson of the Bylaws. Each change to the Bylaws would be voted on
Board separately.
Smith called the meeting to order at 2:10 p.m. and made Smith called for the vote on the proposed change to
introductions of the Officers, Past Presidents, and stagger the Council Directors’ terms in Article III. Votes were
Dignitaries. counted and the proposed change to stagger the Council
1.1 Quorum: Secretary/Treasurer Scott P. Cargill estab- Directors’ terms was approved.
lished the presence of a quorum, with 61 members in Smith called for the vote on the proposed change to
attendance, representing 24 sections. add the position of Immediate Past Chairperson of the
1.2 Appointment of Tellers: Smith appointed Michael V. Board. Votes were counted and the proposed change to add
McGloin and William F. Via, Jr. as tellers. the position was approved.
1.3 Recognition of Agenda Changes: No additions or dele- Smith called for the vote on the proposed change to the
tions were presented. The agenda was set. verbiage regarding the membership composition of the
Selection Committee. Votes were counted and the
2. Ratification of Election Results — Smith proposed change to the verbiage regarding the
2.1. On Saturday, 19 August 2017, membership ballots membership composition of the Selection Committee was
were tallied for the election of the 2017–2018 officers approved.
and new directors. The following individuals were Smith asked for a motion to ratify the Bylaws changes.
elected:
l Officers Motion: Savoy recommended and so moved to ratify the
Chairperson of the Board: David R. Bajula proposed ASNT Bylaws changes.
President: David A. Mandina
Vice President: Scott P. Cargill The motion was seconded by Doron Kishoni and was
Secretary/Treasurer: David E. Savoy approved.
l Directors at Large (three-year term) 4. Approval of 2016 ASNT Annual Business Meeting
John Z. Chen Minutes — Scott P. Cargill, Vice President
John C. Duke, Jr. The ASNT Board of Directors recommended that the 2016
C. Deal Moore Annual Business Meeting minutes, as published in the
The ratification of the election results is automatic. March 2017 issue of Materials Evaluation, and with no
comments having been subsequently received, be
3. Vote and Ratification of Bylaws Changes — Smith approved as published.
Proposed ASNT Bylaws amendments were presented to
members for consideration in the June 2017 issue of Motion: Cargill recommended and so moved to approve the
Materials Evaluation. All eligible voting members were 2016 Annual Business Meeting minutes as published.
either emailed or mailed a ballot from our balloting
provider, Intelliscan, on 1 August 2017 and voting The motion was seconded by Danny L. Keck and was
remained open through 15 October 2017. approved.
Smith asked for vote of members present that had not
already voted to accept the proposed changes to the
David R. Bajula
Marwan F. Basrawi
Joseph R. Clasen
Brenda L. Collins
Mark E. Dahn
Cindy Finley
Joaquin P. Gutierrez
Dick Hooper
Brad Hoskins
John J. Kinsey
Michael V. McGloin
John P. Moran
Lawrence E. Mullins, Sr.
Mark Ormrod
Yi-Cheng (Peter) Pan
Emery E. Roberts
Hussein M. Sadek
David E. Savoy
William F. Via, Jr.
TEST
Using Iterative Time Reversal
for NDT Inspections on
Composite Materials
by Daniel Richard
O
ver the last 15 years, phased array ultrasonic technology has
dramatically changed the face of nondestructive testing. This
advanced pulse-echo technique uses multiple elements in a
transducer to precisely steer, focus, and scan an ultrasonic sound
beam. It is a reliable, repeatable, and efficient way to inspect complex geome-
tries, visualize welded joints, and determine the size and type of flaws like
surface and subsurface cracks in a range of materials.
Phased array UT is especially effective in aerospace and aircraft applica-
tions where the use of carbon fiber reinforced polymer (CFRP) and other
composite materials continues to expand. The multimaterial and anisotropic
nature of CFRP is a challenge for nondestructive testing. Composite structures
and components can involve large areas with varying geometries and thick-
nesses. Discontinuities, including porosities, delamination, and foreign
bodies, can be introduced during the materials manufacturing process, and
can occur in a particular layer of carbon fiber or in resin. Furthermore,
suppliers of CFRP components face competitive pressures to meet the quality,
cost, and delivery expectations of their customers.
By and large, component manufacturers conduct ultrasonic testing in envi-
ronments where an automated scanner can be used to control the alignment
between the ultrasonic probe and the surface of the test material. Iterative
time reversal is a real-time adaptive phased array UT technique that is
designed to eliminate the effects of misalignment encounter a discontinuity, some of that energy is
between the probe and the specimen. This article will reflected back like an echo.
explain how this adaptive UT inspection technique can Transducers (or probes) come in a variety of config-
improve inspection coverage and detection capability urations and sizes, but for typical composite shapes,
while maintaining test speed and efficiency. two types of transducers are used. A linear one-dimen-
sional (1D) phased array probe is used to inspect flat
Phased Array UT surfaces. These probes typically have 32, 64, or 128
Phased array ultrasonic testing is a common NDT elements. For curved sections, arc-shaped 1D probes
inspection method for composite component manu- are used, typically with 32 or 64 elements.
facturers. With composites, discontinuities such as Alignment is simpler to achieve on flat or nearly
cracks can occur on the surface or beneath it—and in flat surfaces where the inspection is performed essen-
different layers or plies—with no predictable orienta- tially at normal incidence with the component surface.
tion. Delamination discontinuities can occur very On a component with convex or concave surfaces or
quickly and propagate, especially in components, variable thicknesses, maintaining the proper probe
where a laminate is loaded through the thickness (for orientation generally involves sophisticated and
example, at spar or stringer runouts in an airplane expensive mechanical systems and an exact
wing structure). There may be little or no warning knowledge of the specimen geometry.
between the points of fatigue and catastrophic failure.
Phased array UT can accurately determine the Time Reversal
depth and size of discontinuities in composite Time reversal is an adaptive phased array UT inspec-
materials using pulses of high-voltage electricity that tion technique that can compensate for the effects of
are converted to high-frequency ultrasonic energy by a misalignment between the probe and specimen in real
probe or transducer. The transducer emits pulsed time. This is achieved through a two-step process based
sound waves into the material at exact intervals (focal on measurement and compensation of flight times of
laws) and set angles; when these ultrasonic waves surface echoes of individual elements in the probe.
The first step is called surface profiling (Figure 1).
During this step, a plane wave is generated by firing all
probe elements simultaneously. When the wave meets
the specimen, it is reflected back to the probe. The
Delays reflected wave is no longer a plane wave; it is affected
by the shape of the inspected component. The variation
in the wave shape translates into a different time of
flight for the response on each individual element i.
Using the different time of flights measured in the
Sound wave
initial firing, software calculates the delay for every
individual element i that will compensate for the differ-
ences introduced by the surface profile. Equations 1
and 2 show how the corresponding emission and
Figure 1: Determination of the surface profile: (a) a plane wave is generated;
reception delays are calculated:
(b) the wave is reflected back to the probe.
1
(1) Ei =
2
(max ⎡⎣ti ⎦⎤ − ti )
where
Sound wave ti is the time of flight for the wave received by
element i.
Delays
Sound wave
(a) (b)
Figure 3: Data recording using compensated delays: (a) the principle; (b) real inspection data.
64
3 mm
(a)
Figure 6: Probe intentionally misaligned by 3 mm.
Figure 7: Amplitude C-scan and end view for arc-shaped 1D probe intentionally misaligned by 3 mm: (a) time reversal technique; (b) standard
phased array.
(a) (b)
Figure 8: Amplitude C-scan and end view for misoriented arc-shaped 1D probe: (a) time reversal technique; (b) standard phased array.
l The time reversal technique enhances discontinuity- l By mitigating the need for exact alignment and
detection capability and improves porosity assess- orientation, the proposed time reversal solution
ment by stabilizing the amplitude of the back wall. also reduces the cost of the required mechanical
l The technique enables the ability to easily scanning systems. w x
determine the fiber orientation within CFRP
material. Time reversal stabilizes back wall ACKNOWLEDGMENTS
amplitude and straightens the interface echo, This article is based on a white paper called “Iterative
Time Reversal Technique for the Inspection of Composite
making it possible to examine large areas of a Structures” from Zetec, Inc.
curved component. The bends often contain fiber
weaving that can be noteworthy for the specimen’s AUTHOR
integrity. Daniel Richard: Technology Manager, Zetec, Inc.,
8226 Bracken Place SE, Snoqualmie, WA 98065;
email drichard@zetec.com
women
Industry
Veterans
Discuss the
Opportunities, Dear Reader,
Challenges, and The following is a discussion of the trials, triumphs, and hopes of nine
women who are veterans in the field of nondestructive testing, from industry
Triumphs of the to engineering to academia.
This discourse is intended to engender open conversation and awareness
NDT Community of the challenges that many in NDT—and many other traditionally male-
dominated industries—face. It’s a timely topic: as referenced in Dr. Arny
by Materials Evaluation Editor Jill Ross Bereson’s Director’s Letter this month, ASNT is making plans to establish a
special task force to identify gaps in the education, recruitment, and retention
of women in NDT, to assess the working environment, and to create actionable
steps for the future.
What is the future of women in NDT? If these women are any reflection, I
think that it is very bright indeed. I hope upon reading these stories that you
are inspired to mentor, reach out, and advocate on behalf of others to
strengthen our industry for all of our colleagues, now and for the future.
—Jill Ross, Materials Evaluation Editor
Cheryl Dufur is an NDT industry veteran of more than Cara Leckey, Ph.D., has performed and led research
40 years. She started out working for the Spartan in the NASA Langley Nondestructive Evaluation
School of Aeronautics in Tulsa, Oklahoma, but soon Sciences Branch since 2010. Her research focus is in
afterward began taking classes where she was the computational NDE (CNDE), particularly ultrasound
only woman in a class of 25 students. After gradua- simulation code development, and in experimental
tion, she moved to Southern California and started her guided wave methods. In 2017 she was selected to
career in NDT, where her first job was in a department lead NASA Langley’s High Performance Computing
called “Fokker Bond,” a form of ultrasonics in which a
transducer was used to look for voids. Today she
works for Test Equipment Distributors LLC in sales and
for AAE Aerospace as a Level III RT/PT in composites.
She is a longtime member of the Greater Los Angeles
Section, where she served as director, chair, and vice
chair and received a lifetime achievement award for
women in NDT.
Forty-year industry veteran Cheryl Dufur reads film Cara Leckey takes a break in the laser vibrometry lab at
looking for large metallic inclusions. NASA Langley Research Center in Hampton, Virginia.
problems in NDE. Udpa served as an Associate Successful managers realize the greatest strength
Technical Editor of Materials Evaluation from of any organization is the people who make up the
2000–2017 and she continues to serve on the workforce, and in recent years, top management is
editorial board of Research in Nondestructive Evaluation. realizing that many outdated notions concerning women
She is a Fellow of ASNT, the IEEE, and the Indian in the workforce are depriving their companies of one
Society of Nondestructive Testing, and a member of half of the potential talent that could be utilized. The
Academia NDT International. same low percentage of females in the workforce is
apparent across the spectrum of traditionally male jobs.
The Discussion In the course of my NDT career, I have worked with
hundreds of welders; very few of them were women.
In 1982, female membership in ASNT was 2%. Today, Because of ASNT outreach programs, the work of
it’s estimated at less than 5% (ASNT, 2016, p. 289). sections, articles, education, and general awareness of
Does this number surprise you? Why or why not? the wonderful opportunities a career in NDT presents, I
am truly hopeful that within the next ten years we will
Bruer: The extremely low percentage does not surprise see a huge increase of women starting careers in NDT.
me in the least; it reflects the male/female ratio found
in industry. I believe this is the product of many Miceli: Unfortunately, no, this number doesn’t surprise
variables such as low active recruiting of young ladies to me. NDT is not an easy field for anyone—it requires hard
pursue mathematics or science during early education, work, dedication, and often long hours away from home.
the general discouragement of women to pursue nontra- And for a long time, I think women were not pushed to
ditional career paths, and the overall lack of enthusiasm go into STEM fields. It took me a long time to convince
for females working in heavy industry. my colleagues that I wasn’t going anywhere, even after
Miceli: Speak up, even if it costs you a job or a client. What advice would you give to young women
It’s OK to say no. Know your worth. looking to get in the field?
Major: I would have told myself, keep doing what Bailey: I would say find a mentor to help you “hang in
you’re doing! I was told several times I moved around there” and guide your choices. My first mentor was a
too much. Yet, if I had not done so, I would not have Level III consultant named Ed Dukich. He has
gained the multitude of experience I have. I would not mentored me for 22 years and was a recipient of the
have seen the methods employed in every way ASNT mentoring recognition in 2016. Everyone needs
possible. Having that under my belt allows me to think a mentor like Eddie!
outside of the box and problem solve in a manner
most cannot. I pick up new projects/jobs or technolo- Bruer: Pick your industry well. Nondestructive testing
gies easily because I’m extremely familiar with change is a fabulous career path to pursue, but working in
and advancements. The constant moving led me to aviation is vastly different than working maritime or
deal with the most diverse, unique people in the power generation or petrochemical. Each industry has
world, allowing me to have a better understanding unique attributes and challenges, but all use NDT.
when working with others, leading them, and/or Basically, I would not recommend anyone choose to
teaching them. I was always able to hold firm to my work in this exciting field in an industry they find boring.
integrity and speak my mind because I knew I could The young women of today have the opportunity and
ability to do exceptionally well in the NDT field, but working in NDE are exposed to a range of topics, such
industry choice is critical to career satisfaction and as physics of electromagnetics, ultrasonics, finite
happiness. element modeling, and signal processing, which also
enhances their marketability after graduation.
Leckey: When you attend conferences, meetings, and
workshops, don’t hesitate to approach the one other What do you think the future of women in NDT looks
woman in the room and make a connection. Additionally, like?
though you may sometimes feel strange about the fact
that you are the only woman in the room (which is often Bruer: I foresee outstanding opportunities for women
the case), don’t hesitate to speak up, give your input, in NDT! As this vitally needed skill set grows in our fast-
ask questions, and make connections. paced world of developing technologies and innovation,
the prospects for all young technicians is increasing
Miceli: Go for it. Getting to work with technologies and exponentially. Nondestructive testing prevents
help save lives—there is nothing better. I cherish the accidents, avoids system failure through inspection
days I spent climbing on stadium roofs or huge bridges monitoring, and saves lives, making it an extremely
or on the antenna of the Empire State Building or at valuable and rewarding career. It is a technically diverse
Ground Zero, helping to recover remains and keep the profession that provides women outstanding opportuni-
NYPD and FDNY and construction workers safe. The work ties to learn and advance. With the changing percep-
is exciting and rewarding, and most of the people in NDT tions in our school systems, outreach programs targeting
are some of the greatest I’ve ever known. girls, and ASNT’s efforts to ensure more women are
Get involved in ASNT or another organization where aware of nondestructive testing, I see a future that
you can contribute, not just for others but for yourself. includes more women joining me—an NDT examiner
There were many times I was frustrated with circum- who happens to also be female—in the workforce.
stances in my career, and the people and mission of
ASNT kept me engaged and brightened my day. Dufur: I believe the future for women in NDT looks very
Don’t take no for an answer. Perseverance is promising. Women need to be more informed about
important in any industry but in NDT it’s priceless. the opportunity there is for a good career in NDT, and
that there are schools that offer free classes to
Major: Go for it! Make sure you study hard, never stop women, like AATA. I highly recommend women to look
learning or striving for knowledge. It’s an ever- into these opportunities.
changing field with each method having many facets.
Be the best at it that you can possibly be, and no one Miceli: I hope that as women make strides in all STEM-
will be able to stop you. Never limit yourself, let related fields and as general outreach of NDT increases,
naysayers get you down, or compromise your integrity. we will see more and more women start to see it as not
only a viable option, but as one that allows them to make
Price: Always be professional, in everything you do. the world a better place while earning a great living.
Don’t get discouraged, set a goal and stick to it, you
can do it! Bailey: I am quite optimistic about women entering
NDT in the future. As a teacher, I am seeing waves of
Roller: I would say to anyone who has a desire or women attending classes, conferences, and entering
interest in this line of work, just go do it! Work hard management positions. I can say that I have profes-
and be the best you can be, and you will succeed at sional relations with at least 10 women who are
everything you do. I hire people based on their ability, Level IIIs, and that is huge. Years ago, women would
common sense, hard work, and good work ethic. I exit this career early for various reasons, but they
have no doubt that women can do exceedingly well in mostly left due to family obligations. Now, families are
this industry, because we are not only great problem more dynamic, coparenting is more common, and
solvers, but also excellent at multitasking, brain- women are capable staying in a field that they love
storming, perfectionism, and thinking outside the box. longer. The sky’s the limit for women in NDT! w x
We can do anything we put our minds to, and don’t let
anyone tell you otherwise. REFERENCES:
ASNT, 2016, From Vision to Mission: ASNT 1941 to 2016,
Columbus, Ohio.
Udpa: When I talk to women students about my
Yoder, Brian, 2016, “Engineering by the Numbers,”
research projects, I start with how NDE can save lives. https://www.asee.org/documents/papers-and-
That is always a selling point with women. The second publications/publications/college-profiles/16Profile-Front-
strength of NDE is that it is multidisciplinary. Students Section.pdf
Ultrasonic Platform
Verasonics is committed to advancing materials science research and
innovation. Using the Vantage system, researchers have access to a highly
programmable ultrasonic platform with advanced phased array technology.
New Vantage capabilities in materials science/nondestructive testing include
time of flight diffraction for assessing welds for discontinuities; 3D data
acquisition for volume assessment of materials; closed crack detection for
improved quality control and structural health monitoring; guided wave
testing for detection and characterization of material defects; and full matrix
capture with real-time acquisitions in high resolution.
Verasonics, Inc., Kirkland, Washington
verasonics.com
Course Contacts
The following are contacts for only those organizations that offer public courses listed in this month’s Calendar.
Atlantic: Atlantic NDT Training; Gary L. LTS: Leak Testing Specialists, Inc.; Cyndi QTI: QTI Services, LLC; Michael Kowatch;
Chapman; 24 Flat Rock Rd., Branford, CT Reid; 5776 Hoffner Ave., Ste. 304, (724) 217-9575; email mike@qtillc.com;
06405; (203) 481-4041; atlanticndt Orlando, FL 32822; (407) 737-6415; fax qtillc.com.
training.com. (407) 737-6416; email cyndi.reid@leak Quality: Quality Testing Services;
ATS: Applied Technical Services; Lisa testingspec.com; website leaktesting Melissa Rankin; 2305 Millpark Dr.,
Henry; 1049 Triad Ct., Marietta, GA spec.com. Maryland Heights, MO 63043; (314)
30062; (888) 287-5227; (678) 444- Mistras: Mistras Group, Inc.; Christina 770-0607; (888) 770-0607; fax (314)
2897; fax (770) 514-3299; email Librandy; 195 Clarksville Rd., Princeton 770-0103; email training@quality
lhenry@atslab.com; atslab.com/training. Junction, NJ 08550; (609) 716-4020; fax testing.com; qualitytesting.com.
BRL: BRL Consultants, Inc.; 219 W. (609) 716-0706; email christina.librandy Snell: The Snell Group; 322 N. Main St.,
Rhapsody Dr., San Antonio, TX 78216; @mistrasgroup.com; mistrasgroup.com. Ste. 8, Barre, VT 05641; (802) 479-
(210) 341-3442; fax (210) 341-2844; Moraine: Moraine Valley Community 7100; fax (802) 479-7171; email
email info@brlconsultants.com; brl College; 9000 W. College Pkwy., Palos info@thesnellgroup.com; thesnell
consultants.com. Hills, IL 60465, M150; (708) 974-5735; group.com.
Decibel: Decibel NDE Training Institute; email ccce@morainevalley.edu; TechKnowServ: TechKnowServ
1st Floor Plainfield, Pattambi, Palakkad, morainevalley.edu/ccce/continuing/ Corporation; Subash Jayaraman;
Kerala, India 679303; 91 9387 674 nondestructive-testing. 2134 Sandy Dr., Ste. 14, State College,
153, 91 0466 2214 333, or 91 8089 NPI: NDE Professionals, Inc.; 13339 NE PA 16803; (814) 237-1031; email
000 250; email info@decibelnde.com; Airport Way, Ste. 100, Portland, OR subash@techknowserv.com; tech
decibelnde.com. 97230; (503) 287-5255; fax (503) 287- knowserv.com.
ETS: ETS Sistemi Industriali, Srl.; Alberto 5992; email training@qnpi.com; Test: Test NDT; Cathy Harvey; 193 Viking
Monici; Via S. Francesco 323, 20861 ndeprofessionals.com. Ave., Brea, CA 92821; (714) 255-1500;
Brugherio (MB), Italy; 39 039877790; Odyssey: Odyssey Technology Corp.; fax (714) 255-1580; email ndttrain
email a.monici@etssistemi.it; Carol Sansieri; 3000 Village Run Rd., @aol.com; testndt.com.
etssistemi.it. Unit 103, #149, Wexford, PA 15090; Trinity: Trinity Institute of NDT
GE: GE Inspection Academy, General (843) 817-3475; email carols@odyssey Technology; Ravi Kumar T. or Shiva
Electric, Oil & Gas: Digital Solutions; test.com. Kumar R.; Plot No. V-22a, 2nd Stage,
201 Beltway Green, Pasadena, TX PQT: PQT Services (Plumstead Training); Peenya Industrial Estate, Bangalore,
77503; (855) 232-7470; email Kim Rosa; 806 Botany Rd., Greenville, India 560058; 91 99009 29439 or
inspection.academy@ge.com; SC 29615; (864) 292-1115; fax (770) 91 98441 29439; email training
geinspectionacademy.com. 514-3299; email krosa@atslab.com; @trinityndt.com; trinityndt.com.
Kraft: Kraft Technology Resources; Karl pqt.net.
E. Kraft; 1377 Timshel St., Dayton, OH
45440; (405) 819-7786; fax (405) 691-
4342; email kraftndt@aol.com; ndtboot
camp.com.
18–20 APR
PT Level III, Kerala, India. Decibel. 22– 26 MAY 18–22 JUN
PT Level III, Houston, Texas. Kraft. RT Level III, Kerala, India. Decibel. UT Level III, Kerala, India. UT Level III, St. Louis,
1 4 – 1 6 M AY
Decibel. Missouri. Quality.
24–28 APR
UT Level III, Kerala, India. Decibel. MT Level III, Dayton, Ohio. Kraft. 30 MAY – 1 JU N 25–29 JUN
VT Level III, Houston, Texas. ET (Eddy Current) Level III,
1 4 – 1 8 M AY
30 APR–4 MAY Kraft. Houston, Texas. Kraft.
RT Level III, Houston, Texas. Kraft. ET Level III, St. Louis, Missouri.
Quality. 5–9 JUN
RT Level III, St. Louis, Missouri.
1–3 MAY Basic Level III, Kerala, India. Quality
1 6 – 1 8 M AY UT Level III, Brea, California.
VT Level III, St. Louis, Missouri. Decibel.
Quality. Basic Level III, Greenville, Test.
South Carolina. PQT. MT Level III, Kerala, India.
1–5 MAY MT Level III, Greenville, South Decibel. 26–30 JUN
Basic Level III, Kerala, India. ET Level III, Kerala, India.
Carolina. PQT. 11–15 JUN
Decibel. Decibel.
PT Level III, Dayton, Ohio. RT Level III, Dayton, Ohio.
MT Level III, Kerala, India. Decibel. UT Level III, Kerala, India.
Kraft. Kraft.
Decibel.
7–11 MAY UT Level III, Greenville, South
VT Level III, Kerala, India.
Basic Level III, Dayton, Ohio. Carolina. PQT.
Decibel. wx
Kraft. VT Level III, Greenville, South
Carolina. PQT.
have different types of vibration displacements and travel at obtain the arrival time of waves because the exact time of
different frequencies. Figure 1 shows the mode shapes of two generation of acoustic events is unknown (Kundu, 2014).
lamb waves in plate-like structures (Watkins and Jha, 2012). Therefore, a modified approach of source localization was
Figure 2 shows group velocity dispersion curves of lamb developed based on the arrival time difference technique. The
waves in a 3 mm thick steel plate. Here, its longitudinal wave arrival time difference is the difference of arrival times or the
velocity is 5960 m/s, its shear wave velocity is 3260 m/s, and arrival time delay among different sensors (Aljet et al., 2012;
its density is 7932 kg/m3. The dispersion curves illustrate that Xiao et al., 2014). This approach is always applied to the
the different frequency components of lamb waves travel at sensor array by measuring the velocity of AE waves and the
different velocities. The nature of lamb wave modes leads to a arrival time difference. This approach is applicable to
drastic decrease in the performance of a source localization isotropic materials, such as metallic materials. However,
system (Ciampa and Meo, 2010). In the low-frequency range uncertainties are inevitably introduced into the arrival time
(<200 kHz), only A0 and S0 modes exist. The S0 mode travels difference measurement because of random errors, systematic
faster than the A0 mode and shows a nondispersive character- errors, and AE signal characteristics (multimode and disper-
istic, whereas the A0 mode is largely dispersive. The disper- sion). To reduce these errors, different techniques have been
sion characteristic makes it difficult to analyze the arrival investigated to determine the arrival time difference.
times when localizing AE sources based on the A0 mode. In previous research, other authors (Sedlak et al., 2013)
However, in large plate-like structures, due to the attenuation presented an automatic determination technique for the first
effect, a low-frequency A0 mode can propagate a long distance arrival time based on the akaike information criterion (AIC)
with high energy (Ziola and Gorman, 1991; He et al., 2017), of AE signals for a thin multilayer plate. In the study, the first-
which indicates that the A0 mode of lamb waves has a prom- coming mode (S0 mode) was not dispersive due to the short
ising application in AE source localization. source-to-sensor distance and the small plate-like specimens
A classical way to localize the AE source is to calculate the used. Although it is still difficult to quantify the reliability of
distance between the source and the sensor by the multiplica- the arrival time, the proposed two-step AIC approach is a
tion of wave velocity with the arrival time of waves. This tech- useful tool to locate AE events by automatically assigning
nique is commonly used in the AE method for its ability to arrival time. However, in large plate-like structures, dispersion
locate the source (Kishi et al., 1991), but it is impossible to and multimode nature are two crucial factors in AE event
localization. To minimize the influence of multimode nature
and dispersion, other authors (Mohd et al., 2014) proposed a
novel technique of wavelet transform analysis and modal
location based on lamb waves theory and time-frequency
analysis and proved that this technique is more accurate for
source localization than the time of arrival (TOA) technique.
(a) (b) Also, it could be considered a more reliable and easier source
localization methodology than delta T, since no test grid and
Figure 1. Two lamb waves mode shapes in plate-like structures:
Hsu-Nielsen source for training data is needed.
(a) symmetric lamb waves mode; (b) antisymmetric lamb waves Other authors (Zarate et al., 2015) proposed a bayesian
mode. framework that considered uncertainties in wave velocities
and arrival times and adopted a continuous wavelet transform
(CWT) to calculate. They obtained a greater accuracy with
6000
the use of liquid-borne waves and obtained reliability through
the bayesian approach, which can eventually lead to signifi-
Group velocty (m/s)
TABLE 1
Related techniques used in literature review
Citations Techniques used* Key words of functions
TR CWT CC
Chen et al., 2011; Liu et al., 2011; X Signal focusing
He and Yuan, 2016; Ciampa and Meo, 2011
Chen et al., 2011; Liu et al., 2011; X Dispersion compensation
He and Yuan, 2016; Ciampa and Meo, 2011
Mustapha et al., 2014 X Signal reconstruction
Mustapha et al., 2014 X Signal filtering
Zarate et al., 2015 X Uncertainty element
Mostafapour and Davoodi, 2015 X Frequency selection
Ziola and Gorman, 1991 X Time-lag calculation
Kim et al., 2015 X Impact source localization
compare the similarity of two signals or calculate the arrival Reference Signal Construction based on Time Reversal Principle
time difference. In this study, to obtain the reference signal, The second step to construct the reference signal is to focus
two steps have been carried out. The first step is to extract the on the extracted signal in the time domain based on the time
narrowband component from the AE signals. In other words, reversal principle. In many signal-processing applications,
the dominant frequency component should be isolated from time reversal acts as a spatiotemporal matched filter to
the AE signals. The second step is to focus the narrowband compensate for the shift of signal (Tanter et al., 2000).
signal in time domain based on the time reversal principle. However, it is discovered that narrowband signals behave
Consequently, the focused signal serves as the reference better than broadband signals in minimizing amplitude
signal. In general, narrowband component extraction is dispersion (Park et al., 2009), and different frequency compo-
achieved by the CWT technique due to its abundant informa- nents in a broadband signal, such as AE signal, may be influ-
tion and high resolutions both in time and frequency enced by the time-reversal operator during time reverse
domains. The extracted narrowband signals were then time operations (Wang et al., 2004). Thus, it is necessary to extract
reversed and focused with the improved signal-to-noise ratio. the narrowband signal from AE in advance as described later.
The time reversal technique is an approach meant to mitigate
Narrowband Component Extraction dispersion effects and perform SHM without baseline data. It
The wavelets have the maximum resolutions in time and was initially aimed at increasing lamb wave resolution by
frequency, which are dependent on the mother wavelet. using time reversal mirrors (Fink and Prada, 2001) and
Different mother wavelets have been investigated (Niri and focusing lamb waves at the location of damage in the time
Salamone, 2012; Mohd et al., 2014), and the morlet wavelet domain (Ing and Fink, 1996). According to the theoretical
was chosen and applied in AE due to its admissibility as a studies of the time reversal technique, the proposed reference
wavelet of a zero mean value (Zitto et al., 2015). signal construction principle based on time reversal for AE is
The complex morlet wavelet is defined as: shown in Figures 3 and 4. Three AE monitoring sensors were
t2
coupled on the surface of a steel plate, and an artificial AE
−
1 source (Hsu-Nielson source) was simulated by a pencil-lead
(1) ψ (t) = e fb
e 2πfc jt
πf b break (PLB) placed on the surface.
where
fb and fc, respectively, represent the bandwidth frequency
and the center frequency, Forward AE sensor
is the ratio of circumference to diameter, Backward AE source
S3 (ω)
e is the Euler’s number,
j is the imaginary unit. H3 (ω)
Steel plate
Time reversed
3
E (ω) = ∑ S∗i (ω ) Si (ω)
i =1
Signals focused
P (ω) is multiplied
3) Reversed signal 4) Reference signal
Si (ω) Hi (ω)
AE sensor AE source
Steel plate
Cross correlation
Figure 4. Illustration of the basic concept of cross correlation with AE signal in a steel plate.
In Equation 6, E() can be expressed as the summation of three preamplifiers with gain sets at 40/60 dB; a four-channel
the product of Si() and Si*() with point wise multiplication. AE signal recording card integrated in computer; a steel plate
D() is the focused signal after P() is time reversed, and measuring 1000 mm 1000 mm 3 mm; and six cables. The
E() can be explained as the process on which the signal is plate was supported by a desk with both sides being free.
focused at the location of the PLB after P()*P() is time During the experiment, three sensors were affixed with vinyl
reversed. The process of adding terms on both sides of electrical tape and Nd-Fe-B magnets to the surface of the steel
Equation 5 functions as an operation strategy. Because plate, and petroleum jelly was added between the sensors and
P()*P() and the original excited AE signal, P(), have the steel plate surface for a better coupling environment. A
same frequency characteristics, no influence is introduced to 0.5 mm 2H pencil lead was broken on the plate’s surface to
the refocused signal. As E() is narrowband, so Si() is simulate the AE source. AE waves were then generated and
processed by CWT in advance. To accomplish the cross recorded by the sensors. The AE sensor outputs were pream-
correlation in the frequency domain, the AE signal Si() plified by 40 dB using one of the preamplifiers. Signals were
should be multiplied by the conjugate of the reference signal then digitized at a sampling rate of 10 MHz and stored in the
E*(). (Note that E*() is equal to E(), so no additional computer with 2048 points.
calculation is required in cross correlation.) The basic concept On the surface of the steel plate, the bottom-left corner
of cross correlation with the AE signal in the steel plate is was selected as the origin of coordinates. The sensor arrange-
illustrated in Figure 4. ment with the coordinates of S1(200,90), S2(800,90), and
S3(500,990) is shown in Figure 5. Eight different positions
Experimental Investigation were chosen as the verification points for PLBs to confirm the
repeatability of the experiment. The coordinates of PLB posi-
Experimental Setup tions are shown in Table 2.
As shown in Figure 5, the experimental setup consists of a In order to ensure the stability of the experiments and
pencil with 2H lead; three 150 kHz resonant frequency AE reduce the errors caused by instruments or cables, a
sensors with an operating frequency range of 50 to 400 kHz; 3 mm long pencil lead was pulled out every time and a circular
Real-time data
Preamplifier acquisition software
3 2 1
1000 mm
Steel plate
5 4
Nd-Fe-B
magnet
8 7 6
S2
S1 Cable
AE sensor AE sensor
Vinyl electrical tape
1000 mm
TABLE 3
Presets of acoustic emission software applied in experiment
Threshold Pregain Sample rate Pretrigger Length of signal
40 dB 40 dB 10 MHz 25.6 μs 2048
ring was placed on the pencil lead tip so that the angle
the two dotted lines in Figure 7 represents the dominant
(shown in Figure 5) formed by the lead and steel plate surface frequency component with a frequency of 131.8 kHz. The
remained constant. Different angle values will cause different CWT result matches well with the dispersion curve. As shown
energy so it is necessary to ensure the consistency of
. SMA in Figure 7, wave package 1 is the S0 mode and wave
cables were chosen as the connectors between the sensors and package 3 is the A0 mode. Wave package 2 arrives before the
the preamplifier to guarantee impedance matching. A A0 mode with a frequency component around 300 kHz and is
commercially available software program for real-time AE considered the shear horizontal (SH) mode according to the
data acquisition and replay was applied in the experiment. analysis (Kaphle, 2012b; Geng, 2006). The SH mode is
The presets of the software are provided in Table 3. another type of guided wave mode that propagates in plate-
like structures. It should be noted that the initial time (PLB
Determination of A0 Mode Arrival Time Difference time) of the AE waves is not at the zero point (Figure 7), so
As described in the previous section, the first step to determine the reflection signal from the plate edges arrives after 200 μs
arrival time difference is to extract the narrowband component (the length of the storage signal is 2048 points and the time
from the AE signal. Figure 6 shows an AE signal received at interval is 0.1 μs), even for the fast S0 mode. The possible
sensor 1 when the PLB was conducted at position 1. All signals multiple scattering of signals due to the existence of sensors is
were analyzed with the same process, described as follows. also considered as a negligible effect due to the large sensor-
Three major wave packages with different value of to-source distance, which is ignored during the analysis of
energies and arrival times are visible in Figure 6, namely, wave signals. In conclusion, wave packages 1 through 3 represent
packages 1 through 3. To determine the modes of waves and the direct waves of S0 mode, SH mode, and A0 mode,
the dominant frequency of the signal, the CWT technique respectively.
was used. The result is shown in Figure 7 along with the Figure 8 shows the AE signals before and after extraction,
dispersion curves from Figure 2. The obvious zone denotes based on CWT. Figures 8a and 8b show the original AE signal
the component with the highest energy. The intersection of in time domain and the normalized frequency spectrum of
1.0
Wave package 3
Wave package 1
Amplitude (mV)
0.5
–0.5
Wave package 2
–1.0
0 20 40 60 80 100 120 140 160 180 200
Time (μs)
1
1.0
Amplitude (V)
131.8 kHz
–1.0
0
(a) 0 50 100 150 200 (b) 0 0.1 0.2 0.3 0.4 0.5 0.6
Time (μs) Frequency (MHz)
1.0 1
Normalized amplitude
Amplitude (V)
131.8 kHz
0
Dominant frequency
component
–1.0
0
(c) 0 50 100 150 200 (d) 0 0.1 0.2 0.3 0.4 0.5 0.6
Time (μs) Frequency (MHz)
Figure 8. Comparison of original AE signal (broadband) and the extracted AE signal (narrowband): (a) original AE signal in time domain;
(b) normalized frequency spectrum of original AE signal using FFT; (c) wavelet coefficient signal extracted from original AE signal at 131.8 kHz
based on CWT method; (d) normalized frequency spectrum of the extracted wavelet coefficient signal by FFT.
Amplitude (V)
1
0
Sensor 1
–1
–2
0 20 40 60 80 100 120 140 160 180 200
Sensor 2 Time (μs)
Amplitude (V)
200
150
100
Sensor 3
50
0
0 50 100 150 200 0 0.2 0.4 0.6 0.8 1.0 1.2
(a) Time (μs) (b) Frequency (MHz)
Figure 9. Reference signal construction: (a) extracted wavelet coefficient signals from three AE sensor outputs when PLB was conducted at
position 1; (b) the reference signal (above) and its frequency spectrum (below).
software measurement. The principle of introducing differ- software, waves from the three paths firstly crossed the
ences by using the software measurement is illustrated in threshold line and were all set at 25.6 μs, as shown in
Figure 11. Run time represents the starting time of the Figure 11b. So the time delay of the A0 mode, as displayed in
software, and PLB time represents the time of breaking the the software, is not the actual arrival time difference of the A0
pencil lead. Nevertheless, it is impossible to record run time mode (tA0) but the relative arrival time difference of the A0
and PLB time. When the waves were captured by the mode (t). This phenomenon indicates that the actual arrival
time difference of the A0 mode (tA0) can be obtained by
adding tS0 to t. Because the S0 mode shows great charac-
0.02 teristics, such as a nondispersive characteristic in thin plates,
and t is processed by cross correlating AE signals with the
Amplitude (V)
0.5
Signal ities used in the localization algorithm were the theoretical
group velocities obtained from the dispersion curves shown in
0 Figure 2. It has been shown through other research that the
–0.5 group velocity of AE waves is frequency dependent and may
–1.0
vary with different tests. However, this was not the major
0 20 40 60 80 100 120 140 160 180 200
error to overcome in this research. For every testing point, the
velocities were determined by the dominant frequencies from
Time (μs)
dispersion curves based on CWT. For A0 mode localization,
the errors were minimized by dispersion compensation based
Figure 10. Pretrigger mode determination with the received signals
from the AE software.
on the time reversal principle. Nevertheless, errors will be
TABLE 4
Arrival time differences of A0 mode obtained by the proposed method
PLB position tS0 (μs) t (μs) tA0 (μs)
t12 t13 t12 t13 t12 t13
#1 –17.75 –72.25 –27.60 –69.30 –45.35 –141.55
#2 2.50 –62.50 –0.40 –74.00 2.10 –136.50
#3 27.50 –49.25 21.60 47.30 49.10 –96.55
#4 –24.05 –27.50 –30.60 32.30 –54.65 59.80
#5 29.45 –4.25 29.50 1.00 58.95 –3.25
#6 –35.75 15.75 –39.20 8.40 –74.95 24.15
#7 2.75 36.75 –0.60 25.90 2.15 62.65
#8 39.75 55.50 33.30 41.00 73.05 96.50
Kaphle, M., 2012b, “Analysis of Acoustic Emission Data for Accurate Mustapha, S., Y. Lu, J.C. Li, and L. Ye, 2014, “Damage Detection in Rebar-
Damage Assessment for Structural Health Monitoring Applications,” Reinforced Concrete Beams Base On Time Reversal of Guided Waves,”
Queensland University of Technology, PhD Dissertation, Ch. 3. Structural Health Monitoring, Vol. 13, No. 4, pp. 347–358.
Kim, J.H., Y.Y. Kim, Y. Park, and C.G. Kim, 2015, “Low-Velocity Impact Niri, E.D., and Salamone S., 2012, “A Probabilistic Framework for Acoustic
Localization In a Stiffened Composite Panel Using a Normalized Cross- Emission Source Localization in Plate-Like Structures,” Smart Materials
Correlation Method,” Smart Materials and Structures, Vol. 24, No. 4, and Structures, Vol. 21, No. 3, p. 035009.
p. 045036. Park, H.W., S.B. Kim, and H. Sohn, 2009, “Understanding a Time Reversal
Kishi, T., M. Enoki and H. Tsuda, 1991, “Interface and Strength in Process in Lamb Wave Propagation,” Wave Motion, Vol. 46, No. 7,
Ceramic Matrix Composites,” Materials Science and Engineering: A, pp. 451–467.
Vol. 143, Nos. 1–2, pp. 103–110. Poggi, V., D. Fah, and D. Giardini, 2013, “Time-Frequency-Wavenumber
Kishimoto, K., H. Inoue, M. Hamada, and T. Shibuya, 1995, “Time Analysis of Surface Waves Using the Continuous Wavelet Transform,”
Frequency Analysis of Dispersive Waves By Means Of Wavelet Trans- Pure and Applied Geophysics, Vol. 170, pp. 319–335.
form,” Journal of Applied Mechanics-trans. ASME, Vol. 62, No. 4, Pullin, R., D.C. Carter, and M. Holford, 1999, “Damage Assessment in
pp. 841–846. Steel Bridges,” Key Engineering Materials, Vol. 167, pp. 335–342.
Kundu, T., 2014, “Acoustic Source Localization,” Ultrasonics, Vol. 54, Sedlak, P., Y. Hirose, and M. Enoki, 2013, “Acoustic Emission Localization
No. 1, pp. 25–38. In Thin Multi-Layer Plates Using First-Arrival Determination,” Mechanical
Li, Q.F., Y. Wang, R.M. Liu, W. Gu, and F. Ao, 2015, “Ultrasonic Systems and Signal Processing, Vol. 36, No. 2, pp. 636–649.
Computed Tomography Imaging Method Of Concrete Materials Based on Sohn, H., H.W. Park, K.H. Law, and C.R. Farrar, 2007, “Damage Detection
Simulated Annealing Genetic Algorithm,” Trans. Nanjing University of in Composite Plates by Using An Enhanced Time Reversal Method,”
Aeronautics and Astronautics, Vol. 32, No. 3, pp. 341–347. Journal of Aerospace Engineering, Vol. 20, No. 3, pp. 141–151.
Liu, S.T., L. Liu, and F.G. Yuan, 2011, “Defect Imaging Using Time- Tanter, M., J.L. Thomas, and M. Fink, 2000, “Time Reversal and the
Reversal Technique,” Proceedings SPIE Smart Materials + Nondestructive Inverse Filter,” Journal of the Acoustical Society of America, Vol. 108, No. 1,
Evaluation and Health Monitoring, Vol. 7981, p. 79811R. pp. 223–34.
Liu, Z.H., H.T. Yu, C.F. He, and B. Wu, 2013a, “Delamination Damage Wang, C.H., J.T. Rose, and F.K. Chang, 2004, “A Synthetic Time-Reversal
Detection Of Laminated Composite Beams Using Air-Coupled Ultrasonic Imaging Method for Structural Health Monitoring,” Smart Materials and
Transducers,” Science China: Physics, Mechanics and Astronomy, Vol. 56, Structures, Vol. 13, No. 2, pp. 415–423.
No. 7, pp. 1269–1279.
Watkins, R., and R. Jha, 2012, “A Modified Time Reversal Method For
Liu, Z.H., F.X. Yu, R. Wei, C.F. He, and B. Wu, 2013b, “Image Fusion Lamb Wave Based Diagnostics of Composite Structures,” Mechanical
Based On Single-Frequency Guided Wave Mode Signals for Structural Systems and Signal Processing, Vol. 31, pp. 345–354.
Health Monitoring in Composite Plates,” Materials Evaluation, Vol. 71,
No. 12, pp. 1434–1443. Wevers, M., 1997, “Listening To The Sound of Materials: Acoustic
Emission For The Analysis Of Material Behavior,” NDT & E International,
Liu, Z.H., Q.L. Xu, Y. Gong, C.F. He, and B. Wu, 2014, “A New Multi- Vol. 30, No. 2, pp. 99–106.
channel Time Reversal Focusing Method for Circumferential Lamb Waves
and Its Applications for Defect Detection in Thick-Walled Pipe With Xiao, D.X, T. He, Q. Pan, X.D. Liu, J. Wang, and Y.C. Shan, 2014, “A
Large-Diameter,” Ultrasonics, Vol. 54, pp. 1967–1976. Novel Acoustic Emission Beamforming Method With Two Uniform
Linear Arrays On Plate-Like Structures,” Ultrasonics, Vol. 54, No. 2,
Liu, Z.H., H.T. Yu, J.W. Fan, C.F. He, and B. Wu, 2015a, “Baseline-Free pp. 737–45.
Delamination Inspection in Composite Plates by Synthesizing Non-
Contact Air-Coupled Lamb Wave Scan Method and Virtual Time Reversal Zarate, B. A., A. Pollock, S. Momeni, and O. Ley, 2015, “Structural Health
Algorithm,” Smart Materials and Structures, Vol. 24, No. 4, p. 045014. Monitoring Of Liquid-Filled Tanks: A Bayesian Approach For Location Of
Acoustic Emission Sources,” Smart Materials and Structures, Vol. 24, No. 1,
Liu, Z.H., J.W. Fan, Y.N. Hu, C.F. He, and B. Wu, 2015b, “Torsional Mode p. 015017.
Magnetostrictive Patch Transducer Array Employing a Modified Planar
Solenoid Array Coil for Pipe Inspection,” NDT & E International, Vol. 69, Zhong, Y.T., S.F. Yuan, and L. Qiu, 2015, “Multi-Impact Source Localiza-
pp. 9–15. tion On Aircraft Composite Structure Using Uniform Linear PZT Sensors
Array,” Structure and Infrastructure Engineering, Vol. 11, No. 3, pp. 310–320.
Mohd, S., K.M. Holford, and R. Pullin, 2014, “Continuous Wavelet Trans-
form Analysis and Modal Location Analysis Acoustic Emission Source Ziola, S.M., and M.R. Gorman, 1991, “Source Location in Thin Plates
Location For Nuclear Piping Growth Monitoring,” AIP Conference Using Cross-Correlation,” Journal of the Acoustic Society of America, Vol. 90,
Proceedings, Vol. 1584, No. 61, pp. 61–68. No. 5, pp. 2551–2556.
Mostafapour, A., and S. Davoodi, 2015, “Continuous Leakage Location In Zitto, M. E., R. Piotrkowski, A. Gallego, F. Sagasta, and A. Benavent-
Noisy Environment Using Modal and Wavelet Analysis With One AE Climent, 2015, “Damage Assessed by Wavelet Scale Bands and B-Value in
Sensor,” Ultrasonics, Vol. 62, pp. 305–311. Dynamical Tests of a Reinforced Concrete Slab Monitored With Acoustic
Emission,” Mechanical Systems and Signal Processing, Vols. 60–61,
pp. 75–89.
ABSTRACT Introduction
The reliability of a new time of flight diffraction Achieving an optimal configuration is one of the most impor-
technique for finding horizontal planar discontinu- tant steps in designing and establishing a test method. In this
regard, new nondestructive testing (NDT) methods have to
ities (H-TOFD) has been investigated using “â satisfy all criteria considered for the quality of the test proce-
versus a” probability of detection (POD) evalua- dure, which is evaluated by the procedure’s accuracy, sensi-
tion. A finite element model (FEM) was developed tivity, reliability, and other factors.
to simulate the ultrasonic response from horizontal Time of flight diffraction (TOFD) is one of the known
techniques in the ultrasonic method of NDT that can accu-
discontinuities at different depths. Several key rately measure discontinuities in parts with a moderately high
variables were also considered to represent the thickness (Silk, 1979; Temple, 1983). Reliability assessments
modeling procedure, including removing the of conventional TOFD techniques have been applied for flaw
unwanted reflections by implementing infinite sizing in many different parts, including the following: lack of
fusion, lack of penetration, and undercut at a pipeline girth
elements and using equivalent mechanical weld (Carvalho et al., 2008); surface-breaking cracks at a
damping for transducer wedges. Simulated data complex geometry like a rotor shaft (Nath et al., 2010); and
was then generated for evaluating the “â versus a” fatigue root cracks at a pipeline girth weld (Bloom et al.,
statistical model. Based on the statistical model 2010). Today, codes and standards have approved TOFD
techniques for conventional inspections as an alternative to
results, all tests were found to fit within the radiographic testing for sections with at least 12 mm of thick-
bounds of the POD model, demonstrating the ness (Verkooijen, 1995). In recent years, some researchers
H-TOFD technique is reliable for detecting and (Baskaran et al. 2006; Subbaratnam et al., 2011) have worked
sizing horizontal planar discontinuities. on new techniques to reduce inspection thickness. Nonethe-
less, it is still a noteworthy issue. Another limitation of TOFD
KEYWORDS: time of flight diffraction (TOFD), is that conventional trigonometric calculations and formula-
probability of detection (POD), finite element, tions of TOFD only provide a vertical measure of discontinu-
horizontal planar discontinuities, H-TOFD ities, and the real size of discontinuities, especially horizontal
discontinuities, are not able to be measured exactly. A new
trigonometric formula for the assessment of horizontal
discontinuities by a novel TOFD technique, called H-TOFD,
using phased array transducers was presented previously
(Bagheri et al., 2014). Recently, it was developed (Seyed
Hayat Gheib and Sodagar, 2016) in order to be used with
TOFD probes (wedge technique). However, the reliability of
the proposed technique for detecting and sizing such disconti-
nuities is not yet evaluated. In this paper, the probability of
sizing (POS) as a function of horizontal discontinuity size â is
* Department of Technical Inspection Engineering, Petroleum University of investigated. According to the Programme for the Inspection
Technology, Abadan, Iran of Steel Components Phase III (PISC-III) program (Visser,
† Department of Technical Inspection Engineering, Petroleum University of 2002), POS curves can be developed in the form of POD
Technology, Abadan, Iran; email m.ranjbar@put.ac.ir
curves appropriately. Therefore, finding the reliability of this
new technique in terms of POD is the aim of this paper.
Transmitter Receiver
PCS=2S
A
d1
B
d2
Back-wall wave
(b)
L2
L1
Longitudinal
diffracted wave
Longitudinal
lateral wave
(a) D-scan (c)
B-scan
Figure 2. H-TOFD technique: (a) schematic configuration; (b) expected signals for receiver A; (c) expected signals for receiver B.
E = 9 GPa, v = 0.35, and = 1190 kg/m3 for the acrylic glass where
wedges. By considering the above properties and an incident f is the pulse frequency,
angle at 26.5°, the wave refracted angle while hitting the target N is the number of cycles.
will be 50°. Four-node plain strain finite elements, CPE4R,
were employed for analyzing the main parts, and also four- A 5 MHz central frequency with one cycle is set for the
node plain strain infinite elements, CINPE4, which accounted pulse excitation.
for simulating an infinitely long steel piece by absorbing
reflections (Riahi et al., 2014b). The size of elements was set POD Modeling
as 0.35 mm. Figure 3 shows the geometry of the test model. In The “â versus a’’ model is a type of test regarding signal
this study, the discontinuity sizes were considered from 5 to response that shows the correlation between the measured
30 mm with a step of 0.5 mm, which were placed horizontally value, â, and the real discontinuity size, a (Nath et al., 2010).
with respect to the top surface. According to MIL-HDBK-1823A (US Department of
Defense, 2009) and ASTM E3023-15 (ASTM, 2015), this
correlation is referred to a simple linear regression model
Appropriate expressed as
geometry
- Part (7) yˆ = β0 + β1 x
- Discontinuity
where
Input parameters
ŷ is the expected response of given x,
- Material properties
0 and
1 are regression coefficients.
- Material damping
- Frequency and bandwidth
In terms of POD, ŷ and x may represent ŷ = â or ŷ = ln(â)
Model validation and x = a or x = ln(a) and random errors may be included.
Herein the linear regression model is defined as Equation 8:
- Signal-to-noise convergency
Figure 4. Simulation-based probability of detection (POD) process for POD ( a) = probability ( aˆ > aˆdec )
the H-TOFD technique. (9)
= 1 − probability ( aˆ ≤ aˆdec )
Transverse Longitudinal
lateral lateral
wave wave
Longitudinal
diffraction
Planar refracted
transverse wave
Planar refracted
longitudinal
Transverse wave
diffraction
Figure 5. The result of diffractions from the left discontinuity tip at 11.4 μs using the finite element model.
Transverse Longitudinal
lateral lateral
wave wave
Longitudinal
diffraction
Transverse
diffraction
Planar refracted
longitudinal wave
Figure 6. The result of diffractions from both discontinuity tips at 13.8 μs using the finite element model.
30
β̂0 = –24.107 (1.277)(8.66e–72)
0.050
β̂1 = 14.502 (0.45512)(1.85e–209)
Amplitude
τ̂ = 2.204666
0
25 âthreshold = 3.99
Discontinuity tip diffraction âdecision = 15
–0.075 ntotal = 102
0 5 10 15 20 25 ntargets = 51
Response, â
20
Time (μs)
0.015
Amplitude
15
0
Discontinuity tip diffraction
–0.025 10
0 5 10 15 20 25
Time (μs) Confidence bound
Predicted bound
5 Test A
Test B
Figure 7. The discontinuity tip diffracted echoes received by the
probes: (a) probe A; (b) probe B.
2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9
100 101 102
Size, a (mm)
waves, which can be diffracted, refracted, or reflected in
different directions. By recording the longitudinal diffracted Figure 8. The â versus log (a) plot showing the â scatter within
confidence and prediction bounds for the tests A and B.
wave signals, the simulated discontinuities are measured.
The corresponding A-scan signal of a 15 mm discontinuity
embedded at a 25 mm depth is shown in Figure 7. The longi- and 25 mm depths, which are introduced as test A and test B,
tudinal wave diffraction can be seen from the left edge of the respectively. Deeper discontinuities (test B) have fewer errors
discontinuity with probe A (see Figure 7a). In the signal and are easier to find, while low scattering is a notable feature
received by probe B, shown in Figure 7b, the first echo is the of both. As the analysis relies on a generalized linear model
longitudinal lateral wave, the second echo is the diffraction based on ordinary regression, the decision discontinuity size is
wave from the right edge of the discontinuity, and the last echo set as 15 mm from where the data are linearly distributed.
is the longitudinal reflection from the discontinuity’s surface. It POD(a) function curves, with their 95% confidence
should be noted that the back-wall reflection is omitted due to bounds, are demonstrated in Figure 9. The slope area of the
using absorbing boundaries. A comparison of the experimental
results obtained from photoelastic visualization in the TOFD
testing (Ginzel et al., 2008) and the experimental results a90/95
presented by other authors (Riahi et al., 2014a) indicates a50 a90
consistency with the simulated wave diffraction patterns 1.0 a50 = 14.83
resulted in this study. However, relative to the conventional
Probability of detection (POD), a
a90 = 18.02
0.9
TOFD sizing of vertical discontinuities, H-TOFD is sensitive a90/95 = 18.92
0.8 μ̂ = 2.697
to perfect alignment of the probes with the discontinuity, along σ̂ = 0.15203
with choosing appropriate PCS when testing the materials. 0.7 POD covariance matrix
0.000467 –2.1e–05
To obtain precise estimates of the POD function, 0.6 (–2.1e–05 0.000276 )
choosing at least 40 target sizes is suggested for a quantitative 0.5
target response, â. In order to plot â versus a data, the discon-
0.4
tinuity sizes are chosen in uniformly spaced extents and are
plotted on a log scale, which result in narrower confidence 0.3
bounds on the POD(a) curve (US Department of Defense, 0.2 Test A
2009). Hence, 51 data were collected for every test. Figure 8 0.1
Test B
Estimated POD
demonstrates signal response results of the simulated discon- POD 95% confidence bound
0
tinuities detected by the presented technique. The signal 2 3 4 5 6 7 8 9
response is the measured discontinuity size calculated from 101 101
the difference of the measured time of the signal diffracted Size, a (mm)
from both discontinuity tips. Two datasets have been
obtained for different discontinuity lengths placed at 20 mm Figure 9. POD curve for the tests A and B.
l The April issue of Materials American Aerospace Technology Academy aatatraining.org 322
Evaluation is a special technical Applus applus.com 325
focus issue on the emerging Carestream carestream.com 277
topic of the NDT of Additive Curtis Industries curtis-test.com 307
Manufacturing. Ultrasonic
Doppler cndoppler.com IBC
testing equipment will be
featured in the Product DÜRR NDT GmbH & Co. KG duerr-ndt.de 285
Spotlight. Contact the adver- Eddyfi teletestndt.com 279
tising supervisor to reserve Fujifilm fujifilmndt.com IFC
your space today!
Guided Ultrasonics guided-ultrasonics.com 273
l In May, Materials Evaluation Hellier hellierndt.com 282
will publish research papers on Jesse Garant Metrology Center jgarantmc.com 301
Infrastructure NDT. Ground
Laser Technology laserNDT.com 280
penetrating radar equipment
Lavender lavender-ndt.com 303
will be featured in the Product
Spotlight. This issue also M2M m2m-ndt.com 338
includes our semiannual Met-L-Chek met-l-chek.com 330
NDTMarketplace product guide MFE Enterprises mfeenterprises.com 336
for the nondestructive testing
NDT Boot Camp ndtbootcamp.com 358
industry. Don’t miss your
chance to be included! Olympus olympus-ims.com BC
PCC Structurals precast.com 290
Looking for a low cost, highly
Physical Acoustics express.physicalacoustics.com 335
visible advertising program?
Quality Testing training@qualitytesting.com 299
Consider placing a directory ad
in Materials Evaluation. Build Spec sourceproduction.com 321
product or service recognition Spectronics spectroline.com 274
with this popular program. Techna NDT technandt.com 321
Technology Design technologydesign.com 316
Use the Coming Attractions
information to help plan your Test Equipment Distributors tedndt.com 325
advertising schedule. Contact TecScan tecscan.ca 294
the Advertising Supervisor at The Phased Array Company ThePhasedArrayCompany.com 333
(800) 222-2768 X209 or by
Thermo Fisher Scientific thermofisher.com 351
email at doen@asnt.org.
University of Ultrasonics universityofultrasonics.com 293
UniWest uniwest.com 305
Varex vareximaging.com 309
Virtual Media Integration starrview.com 295
Volume Graphics volumegraphics.com 327