From NDT Technician, Vol. 15, No. 1, pp: 8-9.

Copyright © 2016 The American Society for Nondestructive Testing, Inc.

FYI
UT Inspection of Welds in Accordance with Code
AWS D1.1: Part 2. Effect of Surface Conditions
by Heydar Alakbarov

Editor’s note: This is the second article in a two-part series. Part 1 was P-grades: P220, average particle size 68 µm (0.00254 in.);
Title
published in the October 2015 issue of The NDT Technician. P150, 100 µm 0.00378 in.); P100, 162 µm (0.00608 in.);
When performing contact ultrasonic testing (UT), the P80, 201 µm (0.00768 in.); P60, 269 µm (0.01014 in.);
by Author
surface condition of the inspected item is an important P40, 425 µm (0.01601 in.); P36, 538 µm (0.02044 in.); and
variable. Surface conditions include: cleanliness, contour and P24, 764 µm (0.02886 in.). Each step was performed using
waviness, and roughness. Requirements for surface cleanliness 70, 60, and 45° angle search units. Sanding was performed
are formulated in subdivision 6.26.3 of AWS D1.1 (AWS, in such a way that uniform isotropic roughness was created
2010). Recommended waviness value needs to be less than over the entire treated surface. The difference in sound-path
1.5 mm (0.06 in.) over a 50 mm (2 in.) span (ASNT, 2007). distances between the search unit and reflector (as I and I0
However, no criteria are given for acceptable roughness of were measured) for each step did not exceed 1%. A light oil
scanning surfaces. A rough surface affects the examination in was used as a couplant. To provide constant pressure, a fixed
many ways, including, but not limited to: causing difficulty in weight of approximately 1.4 kg (3 lb) was put on the top of
moving the search unit; reducing the energy of the refracted the search unit in all cases. The results for all of the tested
beam due to scattering and increasing thickness of the search units were the same, so no dependence of incident
couplant; causing a change in the mode transfer; and causing angle was found. Surface roughness created by sandpapers
reverberation of the sound in the pockets on the surface, P220 to P100 did not create any difference in the reflected
resulting in a wide front surface echo and increased dead signal (I0 – I = 0). Surface roughness created by sandpapers
zone. P80 and P60 decreased the reflected signal by 0.4 / 0.6 dB and
The roughness of the scanning surface of an International roughness created by sandpaper P40 decreased the reflected
Institute of Welding (IIW) calibration block is less than signal more than 1 dB.
3.17 µm (125 µin.) root mean square (rms), which is Then, sets of mild steel (ASTM A36) bars were prepared,
significantly less than the surface roughness of most structural as shown in Figure 1. The initial roughness of Face A and Face B
elements. To find practical criteria for acceptable surface (which were strictly parallel) was 3.7 µm (125 µin.) rms. A
roughness, the following experiment was conducted. 1.59 mm (1/16 in.) diameter hole was drilled in the middle
First, the effect of roughness of the contact surface of a of the bar with axis parallel to Face A and Face B, so that the
plastic shoe using angle beam search units on a receiving response from this hole for an angle beam search unit (70, 60,
signal was investigated. Initially, the surface was polished and 45° angles) was exactly the same from Face A and Face B.
using Coated Abrasive Manufacturers Institute (CAMI) Then, Face A was treated with sandpapers P220, P150, P100,
grade 400 sandpaper, with an average particle size of 24 µm P80, P60, P36, and P24 and the differences in the responses
(0.00092 in.), and the response (I0) from a 1.52 mm from Face A and Face B were recorded. Before each step, the
(0.06 in.) diameter side-drilled hole on an IIW calibration contact surface of the search unit’s plastic shoe was polished
block was recorded. Then, the surface was treated with with P220 sandpaper. To avoid any additional effects from
sandpaper with a larger average particle size and the response scratching of the plastic shoe during the measurements, first
(I) from the same hole was recorded. This step was repeated the response (IA) from Face A and then the response (IB) from
for treating the surface with sandpapers with the following Face B was recorded. In all cases, the difference between the

8 · Vol. 15, No. 1

16Jan TNT forWeb.indd 8 1/14/16 9:39 AM

 sound path to the hole from Face A and Face B did not exceed 1%.
The results are very similar to the results obtained when the effect
of roughness of the contact surface of a plastic shoe were measured.
No dependence of the incident angle was found. Surface roughness
created by sandpapers P220 to P80 did not reduce the reflected
signal (IA – IB = 0). Surface roughness created by sandpaper P60
decreased the reflected signal by approximately 0.5 dB. Surface
roughness created by sandpaper P40 decreased the reflected signal
more than 1 dB and surface roughness created by sandpaper P24
decreased the reflected signal more than 2 dB.

Face A

Face B

Figure 1. Test blocks: metal ASTM A36, thickness 2.54 cm

(1 in.), height 5.08 or 7.62 cm (2 or 3 in.).

Based on the results described in the preceding section, the

following recommendations can be made:
l The roughness of the contact surface of the plastic shoe using an
angle beam search unit should be the same during calibration and
examination. If scratching of the contact surface of a plastic shoe
during scanning above the inspected surface exceeds the roughness
created by sandpaper P80, the reference level should be rechecked.
l The roughness of the scanning surface is acceptable up to the
roughness created by sandpaper P80. If the roughness of the
scanning surface exceeds this roughness, the transfer correction
(which is defined in subdivision S6.1 of Annex S of AWS D1.1)
should be made.
All of these recommendations are applicable only for using angle
beam search units that meet the requirements of subdivision 6.22.7
of AWS D1.1. h

AUTHOR
Heydar Alakbarov: ASNT NDT Level III in UT, VT, and MT;
Inspection Services, Inc., 1798 University Ave., Berkeley, California
94703; (510) 900-2100, cell (480) 399-8365; e-mail halakbarov@
inspectionservices.net.

REFERENCES
ASNT, Nondestructive Testing Handbook, third edition: Vol. 7: Ultrasonic
Testing, American Society for Nondestructive Testing, Columbus, Ohio,
2007, p.224.
AWS, AWS D1.1/D1.1M, Structural Welding Code – Steel, American
Welding Society, Miami, Florida, 2010.

TNT · January 2016 · 9

16Jan TNT forWeb.indd 9 1/14/16 9:39 AM