Basics: A Guide to AWS
Ultrasonic Weld Inspection
How to do weld UT according to the AWS D1.1 Code
BY GORDON E.
AND UWE ASCHEMEIER
The following is the
first of a four-part series that will take us back to the basics of
weld inspection. In this first article, Gordon Smith and Uwe Aschemeier
offer a technician's guide to ultrasonic weld inspection according to
the requirements of AWS D1.1:2004, Structural Welding Code — Steel. The
code is industry's "how to do it" guide for the ultrasonic inspection
of prequalified carbon steel welds. Smith and Aschemeier are both
working AWS CWIs. Smith is an ACCP NDT Level II and III UT, who
was first certified more than 30 years ago. He is also an ASNT NDT
Level III in MT, PT, and RT. Aschemeier has been certified in Europe
and Canada, and is a CSA Weld Level III inspector. He is also a member
of the AWS D1D Subcommittee 4 on Inspection.
The 1994 North Ridge,
Calif., earthquake awakened many of us to the need for more effective
ultrasonic inspection of welds. California, which had some of the most
stringent requirements for welding and weld inspection, was found to be
lacking in the face of an earthquake. Test evaluations showed that
professed ultrasonic testing experts could not effectively and
repeatedly perform UT to the levels required. The concept of GRR, gauge
repeatability and reliability, had not been applied to our system of
structural ultrasonic weld evaluation. The obvious result was that 30
inspectors produced 30 sets of results.
The recently released
355 Final Report makes recommendations for UT weld inspectors that
require special performance testing and qualification in UT, after
being certified as an ASNT NDT Level II — UT. Performance testing is
now a requirement in the laws and regulations of many states.
Did you know that a
practical examination is also an AWS requirement? Section 6.21 of AWS
D1.1, Structural Welding Code — Steel, states: "The qualification of
the UT operator shall include a specific and practical examination
which shall be based on the requirements of the code. This examination
shall require the UT operator to demonstrate the ability to apply the
rules of this code in the accurate detection and disposition of
discontinuities." Many of you reading this article have not taken a
formal test in this area, but you must do so as a starting point for
successful inspections to comply with the requirements of AWS D1.1.
While overall our
ability to perform this type of ultrasonic inspection is poor, here and
there we find refreshing individuals who have come close to mastering
the methodology of AWS ultrasonic testing of welds. However, many do
not have the formal tests on file to support their UT activities.
So, what can you do?
turns out, a lot, but keep in mind you are limited with regard to UT
techniques to those in the D1.1 code. Using techniques beyond the
practices written in the code requires a procedure from a certified NDT
Level III — UT who is familiar with the code and its application of
ultrasonic technology. The engineer of record may also elect in general
or for specific cases to use alternate acceptance criteria. If so, for
code compliance, these criteria must be in writing and should be part
of the contract documents. (See D1.1 Section 6.8 on alternate
The Technician's Guide
What follows is an
operator's guide to ultrasonic weld inspection according to the
requirements of the D1.1 code. All references to sections, tables, and
forms are to items located in AWS D1.1. The authors' comments are in
The audit-acceptable ultrasonic weld
inspector should do the following:
- Obtain training, testing, and certification
as a NDT Level II — UT with VT of welds or an AWS CWI certification
- Pass a UT test specific to AWS D1.1 that
meets Section 6.21 requirements.
- Check out and test your equipment. You will
find the setup, field, and interval tests listed in Sections 6.22
through 6.26 and beyond. Test and document your instrument, cables, and
transducers to ensure you have the required ranges and sensitivities.
- Find the weld on the
part or structure. Identify face "A" and using
Table 6.3 find and mark (paint) the weld identity, the X and Y location
lines adjacent to the welds — Fig. 1. Note that, according to the code,
these are "shall dos." They were required 30 years ago and are still
- Identify and confirm
the AWS prequalified weld type and record this
information on a UT report form such as Form D-11, which is found on
page 330 of D1.1. Nonprequalified welds and certain combinations of
prequalified welds cannot be inspected with UT without specific NDT
Level III written instructions.
Perform visual testing to Section 6 criteria
followed by ultrasonic testing of the base metal. (See Sections 6.20.4,
6.25.4, 6.26.5, and 6.31.1.) Mark any visual defects and lamellar areas
on the weld area and record on Form D-11. The UT area should be clean
of any surface matter, slag, splatter, etc., that could hinder
inspection. It is a code requirement (Section 3, Workmanship) that the
welder clean the welds for inspection.
Fig. 1 — A —
Close-up of a weld X-Y line; B — weld labeled and lined for inspection.
you, the inspector, can fail the welds for this, it is also your
responsibility to ensure that the UT area is acceptable for testing
(see Section 6.26.3). The authors feel that the contour of the
transducer scanning area and the back side are important. Rolling
random contours exceeding ±1 deg are not acceptable for
testing as the sound path will be refracted unpredictably.
Fig. 2 — Acceptable screen displays for
ultrasonic weld testing. A — Soundthickness display for two legs. Weld
grooves are drawn on the screen for reference. path display for three legs; B — weld
- Apply couplant along the weld to be
inspected using a brush to prewet large areas. This will increase your
ability to remain in good sound contact.
- If part of a weld is found to be
inaccessible to testing in accordance with the requirements of Table
6.6, due to laminar content or adjacent welds, the UT is to be
conducted using an approved alternative procedure as necessary to
attain full weld coverage.
- Welds are to be tested using an angle beam
search unit conforming to the requirements of 6.22.7 with the
instrument calibrated in accordance with 6.25.5 using the angle as
shown in Table 6.6 and the scanning motions in Section 6.24. Following
calibration and during testing, the only instrument adjustment
permitted is the sensitivity level adjustment with the calibrated gain
control (attenuator). Note that use of the display zoom function is not
considered an instrument adjustment.
- The reject (clipping or suppression) control
is to be turned off for UT inspection.
- Sensitivity must be increased from the
reference level for weld scanning in accordance with Tables 6.2 or 6.3,
- All butt joint welds "shall be tested from
each side of the weld axis."
- Corner and T-joint welds are primarily
tested from only one side of the weld axis.
- All welds are to be tested using the
applicable scanning pattern or patterns shown in Fig. 6.24 as necessary
to detect both longitudinal and transverse flaws.
- It is intended that, as a minimum, all welds
be tested by passing sound through the "entire volume of the weld and
the heat-affected zone in two crossing directions," wherever practical.
- Discontinuity indications shall be evaluated
as specified in Sections 126.96.36.199 through 188.8.131.52. Indication level
"a" is the decibel (dB) difference between the reference level "b" and
the indication. Reference level "b" can be found in Section 6.25, and
the attenuation factor "c" is obtained by subtracting 1 in. (25 mm)
from the total sound path. The indication rating "d," from Annex D,
Form D-11, represents the algebraic difference in decibels between the
indication level and the reference level with correction for
attenuation as indicated in the following expressions:
Instruments with gain in dB: a b c = d Instruments with
attenuation in dB: b a c = d
Alternately, indications exceeding the reference levels from a
distance amplitude correction (DAC) curve within written instructions
from an ASNT NDT Level III using AWS D1.1 Appendix K are considered
rejectable if the flaw length exceeds acceptance levels in accordance
with 6.31.2 and 6.26.8.
- The length of flaws is determined in
accordance with procedure 6.31.2. This is often called the 6-dB drop
method. An alternative name is the K9.3 length.
The Importance of
the 6-dB Drop in Flaw Evaluation
Fig. 3 — The
6-dB drop. A sound beam is shown traversing a flaw.
As the UT transducer beam is passed over a
flaw, a reflection or UT indication may be produced. With the flaw
indication maximized, the reflective area is centered in the UT beam.
By moving the UT transducer in either the X or Y direction, the beam
will decrease. When the decrease is equal to 6 dB, the beam has moved
halfway off of the defect's reflective area. By traversing the defect
and measuring 6-dB drop dimensions, the flaw reflector dimensions may
be approximated for evaluation.
Note that the surface of most flaws — with
the possible exception of an unwetted braze or weld surface — is not a
flat perfect sound reflector and that flaws only return a frequency
response in relation to their composition and structure. When the flaw
is very small in relation to the scanning sound beam wavelength, the
reflector will have no discernable shape in any dimension. When
inspecting steel with a 2.25-mHz transducer, it would be hard to give
flaws below 0.025-in. diameter a specific dimension. This should be far
below the reference that is specified. Figure 3 illustrates the 6-dB
Fig. 4 — A
comparison of flaw types and flaw reflectors.
- Ultrasonic testing indications found to be
within ± 1/16 in. (4 mm) of a weld fusion plane are to be
evaluated with 45 to 60 or 70-degree transducers to characterize the
type of flaw — Fig. 4. Do not forget to do this from the opposing weld
inspection face. Your comments would then be recorded in the comments
section of Form D-11.
Each weld discontinuity is accepted or rejected on the basis of its
indication rating and its length, in accordance with Table 6.2 for
statically loaded structures or Table 6.3 for cyclically loaded
structures, whichever is applicable.
that are rejectable need be recorded on the test report, except for
welds designated in the contract documents as being fracture critical.
Acceptable ratings within 6 dB, inclusive, of the minimum rejectable
rating are recorded on the test report.
- Identification of
Rejected Area. You should indicate each rejectable discontinuity on the
weld by placing a mark directly over the entire length of the
discontinuity. The depth from the surface and indication rating are
then noted on the nearby base metal.
ultrasonic testing are to be repaired by the methods permitted in
Section 5.26. Repaired areas are then retested ultrasonically with
results tabulated on the original form (if available) or additional
Preparation and Disposition of Reports
At the time of inspection,
the ultrasonic inspector must complete a report form that clearly
identifies the work and the area of inspection. The report form for
welds that are acceptable need contain only sufficient information to
identify the weld, the inspector (signature), and the acceptability of
the weld. As mentioned previously, Form D-11, which is found on page
330 of AWS D1.1, is an example of such a form.
Before a weld subjected to ultrasonic
testing is accepted, all report forms pertaining to the weld, including
any that show unacceptable quality prior to repair, are to be submitted
to the owner's inspector.
A full set of completed report forms,
including any that show unacceptable quality prior to repair, are to be
delivered to the owner upon completion of the work.
For specifics, you should
always refer to the AWS D1.1 code and project contract documents and
drawings; however, in general (static), acceptability is as follows:
- "Class A
(large discontinuity): Any indication in this category shall be
rejected regardless of length.
- "Class B
(medium discontinuity): Any indication in this category having a length
greater than 3/4 in. shall be rejected.
- "Class C
(small discontinuity): Any indication in this category having a length
greater than 2 in. shall be rejected.
- "Class D
(minor discontinuity): Any indication in this category shall be
accepted regardless of length or location in the weld."
If you know the
transducer angle and the plate thickness, you can easily calculate the
- 45 deg, the
skip distance = 2 ¥ the thickness = 2 legs
- 60 deg, the
skip distance = 3.5 ¥ the thickness = 2 legs
- 70 deg, the
skip distance = 5.5 ¥ the thickness = 2 legs
from the weld centerline is proportional to the depth of the flaw.
from the centerline of the weld to the transducer index point is
proportional to the indication depth.
To find flaw
divide the distance from the weld centerline by the skip distance
factor by 2.
Example: T = 1
angle = 70 deg, SF = 5.5, distance = 1.875 in. Therefore, flaw depth =
(1.875/(5.5/2)) = 0.682 in.
Checklist for Personnel and Equipment
for AWS Field UT Inspection
1. NDT Level II UT
(AWS qualified) certified operator.
2. Instrument meeting code requirements.
3. Tested cable/transducer assemblies with current documentation of
sensitivity and range, and with wedge couplant (not defined in AWS
4. AWS D1.1 code acceptance criteria (static/dynamic).
5. Written approved engineering UT test requirements.
6. Written approved instructions for "special" UT inspection conditions.
7. AWS D1.1 instructions or written practice.
8. Calibration block traceable to an accepted master IIW block with
offset dB value, if any.
9. Inspection couplant meeting AWS requirements. (Special couplants
require NDT Level III approval/written procedure for use.)
10. Marking pens, ruler, tape measure, paper, reporting forms.
11. Drawing with weld identification numbers and locations.
12. Cleanup materials such as rags, soap, paper towels, etc.
Additional suggested equipment: digital camera and laptop computer.
Doing the UT inspection
Locate welds to be inspected, complete heading information on report
2. Paint identification on each weld joint.
3. Paint/mark the long axis and "skip" distances on each weld.
4. Test the base metal for laminations from the appropriate faces.
Note that butt joints are to be tested from two
Mark on the base metal all rejected areas.
Sketch reject base metal areas and attach the
sketch to the weld
inspection report. 5. Select the appropriate angle transducer to begin
6. Perform setup calibrations; record on UT report form.
7. Perform in-process UT system checks on schedule. Record system
checks on UT report form.
8. Switch to opposite side for welds in butt joints.
9. Evaluate flaw reflectors with little or no transducer movement using
10. Evaluate flaw reflectors with definite transducer movement using
the 6-dB drop method to define the edges of the reflector in both the X
and Y dimensions.
11. Record/evaluate reflectors as applicable on report form.
12. Mark reflector areas on base/weld metal. Include reflector IDs with
each marked reflector.
13. Evaluate reflectors for location on weld interface.
If on the weld interface, reevaluate with other
Use most critical amplitude measurement angle to
Determine the most likely flaw type(s) from
indicate this on the report. 14. Mark depth of reflector(s) on the
base/weld metal; record on report.
15. Complete the UT report form; submit for review and approval, if