1. Overview

In ultrasonic testing, the three key aspects of defect evaluation are defect localization, quantification, and characterization.

Currently, research on the localization and quantitative evaluation of defects in ultrasonic testing has reached a relatively mature stage; however, the characterization of defect nature is seldom addressed. Although ultrasonic testing standards mandate the localization and quantification of defects, they do not require their qualitative assessment due to the inherent difficulties involved. This is because the ultrasonic reflection characteristics of a defect depend on its orientation, geometric shape, and its dimensions—length and thickness—in relation to the direction of ultrasonic wave propagation.

The roughness of the defect surface, the inclusions within the defect, and the nature and type of the defect, among other factors.

In practical inspection, the difficulty in determining the nature of defects often leads to the rework of products containing non-hazardous defects, resulting in unnecessary waste; at the same time, hazardous defects—such as cracks—may be overlooked, thereby posing safety risks during service. Consequently, qualitative defect analysis in ultrasonic testing is of paramount importance.

In practical applications, the challenges remain considerable; accurate defect characterization can only be achieved when the defect echo is highly characteristic and when information from multiple sources is comprehensively integrated. The following sections present some insights gained from real-world experience in defect characterization, for reference only.

Digital ultrasonic flaw detector, 5P6×6 probe, K3, front edge 5 mm, wall thickness 10 mm, V-groove bevel, scan ratio 1:1 in the horizontal direction, wave height 80%, first-wave sound path 30 mm, second-wave sound path 60 mm.

I. Cracks

1. Central root crack

The reflected wave appears at 25–35 mm along the horizontal sweep line of the fluorescent screen (approximately at the position of the bottom wave).

When the crack depth is small (less than 1.5 mm, with the crack fully closed), the first-wave tail and second-wave head detected from both sides produce symmetric reflected waves, such that the sound-path length along the horizontal sweep line on the fluorescent screen equals the horizontal distance. When the crack depth is larger (greater than 2 mm), the first-wave reflections from both sides remain symmetrical, resembling corner-reflection echoes. In this case, the sound-path length and horizontal distance of the crack echo are approximately the same as those observed in incomplete penetration defects (with a root crack opening of 0.5 mm).

When the probe is moved vertically along the weld, the crack echo sharply peaks on the horizontal sweep line of the fluorescent screen, sometimes exhibiting a multiple-peak phenomenon; when the probe is moved horizontally, the unevenness of the reflecting surface causes the crack echo amplitude to be less stable than that of the incomplete-penetration echo.

2. Root fusion line cracks

The sound path length and reflection characteristics are the same as those of a central crack; however, the horizontal distance on the defect side is approximately 2–4 mm greater than that on the non-defect side, and the sound path termination point lies near the fusion line on the root side.

3. Cracks at the top center

The reflected wave appears at 56–64 mm along the horizontal sweep line of the fluorescent screen (around the position of the leading edge).

When the crack depth is small (less than 1.5 mm, with the crack fully closed), the secondary wave tail and the tertiary wave head detected from both sides exhibit symmetrical reflection waves, and the sound path length along the horizontal sweep line on the fluorescent screen is equal to the horizontal distance. When the crack depth is larger (greater than 2 mm), the secondary wave tail and the tertiary wave head detected from both sides again show symmetrical reflection waves, similar to corner reflector echoes.

The presence of cracks can be identified by tapping the weld cap layer with a finger moistened with couplant; if the echo peak does not exhibit vertical oscillation, it indicates a crack.

When the probe is moved vertically along the weld, the crack echo sharply peaks on the horizontal sweep line of the fluorescent screen, sometimes exhibiting a multiple-peak phenomenon; when the probe is moved horizontally, the unevenness of the reflecting surface causes the crack echo amplitude to be less stable than that of the incomplete-penetration echo.

4. Top fusion line cracks (top lack of fusion)

The reflected wave appears at 55–65 mm along the horizontal scan line of the fluorescent screen (around the position of the top wave). Its reflection characteristics are identical to those of a root-center crack, except that the horizontal distance on the defect side is approximately 5–10 mm greater than that on the non-defect side, and the sound-path landing point lies near the fusion line of the cover layer.

The presence of cracks can be identified by tapping the cover layer with a finger moistened with couplant; if the echo peak does not fluctuate up and down, it indicates a crack.

II. Incomplete Penetration

1. Double-sided lack of penetration

When the root gap of the pipe weld is small (approximately 2 mm), the lack-of-fusion indications exhibit good symmetry when inspected from both sides of the weld.

When the reflected wave is at its strongest, the horizontal distance is approximately equal to the sound path length indicated by the horizontal sweep line on the fluorescent screen (with the horizontal distance being about 1 mm greater than the sound path length).

When the probe is moved back and forth, the lack of full penetration results in a relatively regular root face, leading to strong end-corner reflections; the reflected waves appear at 25–35 mm along the horizontal sweep line on the fluorescent screen (approximately corresponding to the position of the bottom wave).

When the probe is moved horizontally, the peak amplitude of the maximum reflection within the incomplete-penetration length remains nearly constant (continuous lack of penetration) or exhibits slow fluctuations (discontinuous lack of penetration), while the sound-path distance indicated by the horizontal sweep line on the fluorescent screen remains essentially unchanged.

2. Single-sided lack of fusion

When the probe is moved back and forth, the peak position of the incomplete penetration echo depends on the depth (or area) of the incomplete penetration.

When the probe is moved horizontally on the defect side, a large-area defect signal (continuous lack of penetration) or slow undulations (intermittent lack of penetration) are observed; when the probe is moved horizontally on the non-defect side, no lack-of-penetration wave is detected.

III. Concavity (Sunken Waist)

Reflection waves are detected on both sides of the weld, with low-amplitude concave reflection peaks appearing at 25–30 mm on the fluorescent screen. When flaw detection is performed from both sides of the weld, the concave waves exhibit good symmetry.

For the fixed-position overhead weld, the defect is located at the 5–7 o’clock position on the weld root, with a relatively smooth contour (approximately an arcuate surface; for the interface where the sound beam is incident, this surface is convex, causing the reflected waves to diverge). Consequently, the probe has a large refraction angle and a long acoustic path, resulting in much lower internal-concave reflection amplitude compared with cracks or lack-of-fusion defects. Moreover, as the probe is moved back and forth, the echo waveform remains stable.

When the probe is moved horizontally, the maximum reflected wave within the concave length changes slightly.

The positioning landing points are spaced a certain distance apart, slightly larger than the joint gap dimension; this distance is approximately equal to the joint gap plus 2–4 millimeters.

IV. Weld Beads

The weld metal overflows from the root of the weld into the pipe, forming a protruding bead; such weld bumps typically occur in overhead and uphill flat-position welding.

A. The waveform is sharp and has a relatively high amplitude. Because the curved surface of the weld bead consistently focuses and reflects the ultrasonic beam back to the probe, reflections are observed over a certain range as the probe is moved back and forth, with relatively stable variation.

Weld bumps sometimes contain multiple porosities; weld bumps with porosities typically exhibit strong, cluttered reflection signals (dense porosity echoes) in ultrasonic testing.

B. Weld bumps appear at a location slightly greater than the base material thickness (behind the bottom-wave position), with sound-path values ranging from 31 to 39 mm on the horizontal sweep line of the fluorescent screen. When inspected on both sides of the weld, they manifest as either a single sharp reflection or a double-peaked reflection, with relatively high amplitude.

Detect from both sides; the sound-path impact points intersect by approximately 4 mm.

V. Internal Undercut

The position of the internal undercut wave appears at the leading edge of the primary wave (with a sound path of approximately 27 mm on the horizontal sweep line of the fluorescent screen), and its amplitude is relatively low. If the defect wave amplitude is small, the reflection from the root weld can also be observed on the weld side.

When the probe is scanned on both sides of the weld, the sound path termination point lies in the base metal on the root-pass side. As the probe is moved parallelly, the echo amplitude exhibits fluctuations (undercut typically manifests as discontinuous transitions).

Reflection waves from internal undercut are observed on both sides of the weld. Typically, one side exhibits both an undercut wave and a small, low-amplitude bottom echo (the root-reflected bottom wave). When the probe is moved back and forth, only a very small vertical dimension is displayed; when the probe is moved horizontally, the reflection waves on the two sides differ significantly. This is attributable to the angle of incidence and the directional nature of the defect. In some cases, the reflection amplitude on one side of the weld is low or even absent.

6. Lack of Fusion at the Groove Edges

When inspected from both sides, the sound path to the reflection point lies near the fusion line; reflection signals are detected on both sides, with the signal from the non-fusion side being stronger. In this case, both the sound path and the horizontal distance are greater than those corresponding to the primary wave (i.e., the sound path and horizontal distance of the secondary-wave reflection).

When detected from the opposite side, the reflected signal is weaker, and both its acoustic path length and horizontal distance are shorter than those of the primary wave (i.e., the primary-wave reflection).

VII. Porosity

1. Single pore

They are randomly distributed within the weld, with no discernible pattern; the waveforms are sharp, and the wave roots are relatively narrow.

As the probe is moved back and forth, when the leading edge of the sound beam encounters a porosity, the echo amplitude increases rapidly with further forward movement of the probe, then rises more gradually; after reaching a maximum, further forward movement of the probe causes the echo to decay more quickly.

2. Chain-like stomata

The waveform is similar to that of a single porosity; as the probe is moved parallel to the weld, the wave peaks continuously rise and fall; when the probe is rotated, the waveform alternates between high and low, with a slight change in sound path.

When one pore wave subsides and another immediately follows, only when there are three or more such pore waves is it referred to as a chain-like pore pattern.

3. Dense porosity (reheat cracking, hydrogen-induced white spots)

When multiple reflected waves appear simultaneously, there is often one relatively high wave flanked by several smaller ones, with a clear waveform that resembles a flower.

When probing on both sides of the weld, a cluster of irregular reflection waves can be observed, with waveforms that fluctuate unpredictably. As the probe is moved parallel to the weld, the pattern of these reflections alternates in intensity and appearance.

8. Linear Slag Inclusions

Slag inclusion in single-sided, double-sided weld joints often occurs at the groove root angle between the root pass and the fill pass.

The waveform is serrated, with low-amplitude reflections and small peaks along the edges of the main peak.

When the probe is moved parallelly, defect echoes are present over a relatively large area, with little variation in echo amplitude but noticeable changes in echo peak position, indicating a certain indication length; however, when the probe is rotated, the echo waveform disappears more rapidly.

The wave amplitude is different when scanning on both sides of the weld.

9. Normal weld root reflection wave

The reflected wave amplitude at the root of a normal weld is low (the bottom echo at 30 mm), and sometimes no echo is present; when the probe is moved laterally, the wave peak exhibits only minimal variation.

When inspecting on both sides of the weld, the horizontal distance locators of the reflected waves intersect, with their positions on the fluorescent screen offset toward the side opposite the weld centerline (i.e., the root fusion line on the non-probe side), and their depth values are slightly greater than the base material thickness.

X. Reflection Waves from the Weld Reinforcement Surface Caused by Beam Diffusion

Because the wall of the probe is relatively thin, when the diverging sound beam following the primary beam from the transducer reflects off the bottom surface and reaches the weld reinforcement face, it will generate a reflected wave from the cover layer, which can easily be misinterpreted as a defect in the upper part of the weld. This can be identified by tapping the weld cover layer with a finger while applying couplant, based on the probe position and the corresponding sound path; if the waveform peak oscillates up and down, it indicates a cover-layer echo.

XI. Deformation Waves Caused by Misalignment (or Weld Beads)

When the probe’s refraction angle and the orientation of the misalignment (or weld bead) are properly matched, in addition to the misalignment (or weld bead) echo, transformed longitudinal or transformed transverse waves will also be generated and propagate to the weld reinforcement surface.

Echo signals are generated behind the main beam.

The cover layer can be identified by tapping it with a finger coated with coupling agent; if the waveform exhibits upward and downward oscillations, this indicates a mode-converted wave.

Summary:

In practical ultrasonic testing, due to limitations imposed by site conditions, it is often difficult to accurately characterize defects based solely on their echo signals.

For welds with good material weldability and well-established welding procedures, defects are typically caused by improper operation, such as incomplete penetration, slag inclusions, and porosity. As long as the relevant standards are strictly followed—specifically, assessing weld quality based on the amplitude of defect reflection waves, their indicated length, and their area at the specified sensitivity levels to determine whether they exceed the acceptance limits—quality can be reliably evaluated.

For critical components, the material must exhibit high strength but poor weldability and a strong tendency toward hardening in the weld zone. In addition to detection according to the sensitivity levels specified in the relevant standards, particular attention should be paid to the identification of hazardous defects—such as those with reflection amplitudes within acceptable limits but exhibiting a certain length—to prevent the omission of such defects.

All content in this article is collected from the internet and is intended solely for technical exchange and sharing. If any infringement occurs, please contact us, and we will promptly remove the offending material.