Shear Wave Ultrasonic Testing: What It Is and How It Works

Shear wave ultrasonic testing is an angle-beam inspection method that sends shear (transverse) waves into a solid at a set refracted angle, usually 45°, 60° or 70°. In steel the shear wave travels at about 3,250 m/s, and it finds flaws tilted to the surface, such as weld lack of fusion and transverse defects in rail.

The shear wave is not sent straight from the transducer. A longitudinal wave crosses a plastic wedge and converts to a shear wave at the steel surface, a process called mode conversion. The wedge angle is chosen with Snell's law so that only the shear wave enters the part.

Contents: Shear vs longitudinal waves · Velocities · Liquids · How it works · Snell's law · Critical angles · Probe angles · Curved parts · Rail · Calibration · Limits · FAQ

How is a shear wave different from a longitudinal wave?

In a longitudinal wave the particles move back and forth along the direction of travel; in a shear wave they move at right angles to it. That difference sets the speed, the wavelength and the materials each wave can cross.

Because a shear wave in steel is a little over half as fast as a longitudinal wave, its wavelength at the same frequency is also a little over half as long. The shorter wavelength gives better sensitivity to small reflectors at a given frequency, which is one reason weld procedures use shear waves.

Key takeaway: a shear wave is slower and has a shorter wavelength than a longitudinal wave of the same frequency, and it needs a solid to travel in.

How fast do shear waves travel in steel?

Shear waves travel at about 3,250 m/s (0.128 in/µs) in carbon steel, against about 5,890 m/s (0.232 in/µs) for longitudinal waves. Evident, the probe manufacturer, designs its standard angle-beam wedges for steel with a shear velocity of approximately 3,250 m/s.

These are typical values. Evident notes that its table is only a general guide, because composition, heat treatment and temperature all shift velocity. For anything other than plain carbon steel, measure the velocity on a block of the same material before trusting a refracted angle or a depth reading.

Why can't shear waves travel through liquids?

A shear wave needs a material that resists being sheared, and liquids and gases have essentially no shear stiffness. Shear velocity is the square root of the shear modulus divided by density (v = √(G/ρ)). With G close to zero, as in water, oil or gel couplant, a shear wave has no restoring force and does not propagate.

This has two practical results:

• The couplant carries only longitudinal energy from the wedge to the part. The shear wave is created at the steel surface by mode conversion, not inside the gel.
• Immersion and water-column testing must also send a longitudinal wave through the water and rely on refraction to create the shear wave in the part.

Key takeaway: the shear wave in shear wave ultrasonic testing exists only in the steel; everything between the crystal and the steel carries longitudinal waves.

How does shear wave ultrasonic testing work?

Shear wave ultrasonic testing works by sending a longitudinal wave through an angled wedge so that it refracts and mode-converts into a shear wave at a chosen angle in the part, then timing and measuring the echoes that come back. The flaw detector converts each echo's arrival time into a sound path, and from the angle it calculates the flaw's depth and surface distance.

The sequence for a contact angle-beam probe is:

1. The piezoelectric element sends a longitudinal pulse into the wedge.
2. The pulse reaches the wedge-to-steel interface at the wedge's incident angle.
3. At that boundary part of the energy reflects, and part refracts into the steel. Because the incident angle is beyond the first critical angle, the refracted longitudinal wave cannot exist and only a shear wave enters.
4. The shear beam travels at its refracted angle, reflects from the far wall and continues in a "skip" path.
5. A planar flaw facing the beam returns an echo. The instrument plots its amplitude against time (an A-scan).

Evident's weld inspection tutorial describes the most common angle-beam probes as designed to generate "a refracted shear wave at standardized angles of 45, 60, or 70 degrees in the test material."

What is mode conversion?

Mode conversion is the change of part of a wave's energy from one wave type to another when it strikes a boundary at an angle. A longitudinal wave hitting the wedge-to-steel boundary obliquely generates both a refracted longitudinal wave and a refracted shear wave, each obeying Snell's law with its own velocity.

Mode conversion also happens at flaws, bevels and the back wall, and the converted waves can return echoes at unexpected times. These are one source of non-relevant indications the operator must recognize.

How is the refracted angle calculated with Snell's law?

Snell's law states that sin θ₁ / V₁ = sin θ₂ / V₂, where θ₁ is the incident angle in the wedge, θ₂ is the refracted angle in the part, and V₁ and V₂ are the matching velocities. For a shear-wave probe, V₂ is the shear velocity of the part.

Worked example: designing a 45° shear-wave wedge for steel

Inputs: acrylic wedge, V₁ = 2,730 m/s (longitudinal); carbon steel, V₂ = 3,250 m/s (shear); target θ₂ = 45°.

1. sin θ₁ = (V₁ / V₂) × sin θ₂ = (2,730 / 3,250) × sin 45° = 0.840 × 0.7071 = 0.594
2. θ₁ = arcsin 0.594 = 36.4°
3. Check the longitudinal wave: sin θL = (5,890 / 2,730) × 0.594 = 1.28. A sine above 1 has no solution, so no refracted longitudinal wave enters the steel. Only the 45° shear wave does.

The same arithmetic for the other standard angles:

Time-of-flight check. On a 1.000 in plate with a 70° probe, the round trip to the bottom and back is 2 × 2.924 in ÷ 0.128 in/µs, or about 45.7 µs of steel time, plus the wedge delay. An operator can use this to confirm that a back-corner echo is where it should be. More checks of this kind are in our ultrasonic testing formulas guide.

Why the same wedge gives a different angle in another material

Evident notes that a wedge designed for steel produces a different refracted angle in other materials. A lower shear velocity than 3,250 m/s gives a smaller true angle than the label, so measure it on a block of the same material.

Evident's white paper gives a shortcut for this: apply Snell's law between the two metals, using the label angle in steel and the two shear velocities, so sin θ(new) = (V(new) / 3,250) × sin θ(label). The wedge material drops out of the sum.

Worked example: a steel wedge on aluminum

Inputs: shear velocity in steel 3,250 m/s; in an aluminum alloy about 3,130 m/s (a value published for duralumin; measure the real alloy); velocity ratio 3,130 / 3,250 = 0.963.

The 70° probe loses about 5°, and that throws the depth off. Say an echo comes back at a 2.00 in sound path. If the instrument is still set to 70°, it gives a depth of 2.00 × cos 70° = 0.68 in. The true depth is 2.00 × cos 64.8° = 0.85 in, so the reported depth is 0.17 in too shallow. The surface distance moves as well: it reads 1.88 in when the true figure is 1.81 in. On a thin weld that is enough to put the reflector in the wrong pass, or even outside the weld. The fix is to use a wedge built for the material (Evident lists aluminum wedges as a standard product) or to measure the true angle on an aluminum block and enter that value.

What are the first and second critical angles?

The first critical angle is the incident angle at which the refracted longitudinal wave reaches 90°; the second is where the refracted shear wave reaches 90°. Between them, only a shear wave enters the part, which is the working range of every standard angle-beam wedge.

Each critical angle comes from Snell's law with θ₂ = 90° (sin 90° = 1), so sin θcrit = V₁ / V₂.

What happens at each incident angle for an acrylic wedge on steel:

For example, a 20° acrylic wedge on steel would produce a longitudinal wave at about 48° and a shear wave at about 24° at the same time. That is why practical shear-wave wedges sit above the first critical angle. Published figures vary by a few tenths of a degree depending on the velocities used.

Key takeaway: for an acrylic wedge on carbon steel, a shear-only beam needs an incident angle between about 27.6° and 57.1°.

Which probe angle should you use: 45°, 60° or 70°?

The angle is chosen so the beam meets the expected flaw face as close to perpendicular as possible and so the beam covers the weld volume within a practical skip distance. Thickness, bevel angle and the flaw type set the choice. A code or written procedure often fixes the angle.

The beam angle should be matched to the flaw orientation, not just the plate thickness. A fusion face on a 30° bevel lies 30° from the vertical, so a 60° beam can meet a lack-of-fusion flaw on that face at close to 90°. For more on why that flaw type matters, see lack of fusion in welds.

What AWS D1.1 expects from the search unit

AWS D1.1/D1.1M is the structural welding code for carbon and low-alloy constructional steels; the current edition is D1.1/D1.1M:2025, released in April 2025 (AWS announcement). Procedures written to the code commonly require:

• a search unit that produces a beam within ±2° of 70°, 60° or 45° in steel
• a transducer frequency of 2 to 2.5 MHz
• the frequency, nominal angle and index point marked on the search unit
• the IIW block as the standard for distance and sensitivity calibration, with other blocks permitted when they give equivalent results

These points come from public procedures based on the code, not from the code text itself. Confirm the clause and table numbers in the edition named in your contract. How D1.1 then rates an indication (the d = a − b − c rating) is covered in our AWS D1.1 weld inspection guide.

When does a shear-wave probe need a contoured wedge?

A flat wedge on a curved part touches it only along a line. As the gap grows on either side, less sound gets in and the couplant layer adds noise. Evident cites the IIW Handbook on the Ultrasonic Inspection of Welds, which recommends a contoured wedge "whenever the gap between the wedge and the test surface exceeds 0.5 mm" (about 0.020 in).

Evident turns that into a rule of thumb: contour the wedge when the part radius R is less than W² ÷ 4. W is the wedge width when the beam points along the pipe axis, as when scanning a girth weld. It is the wedge length when the beam points around the circumference, as when scanning a long seam. The formula comes from the 0.5 mm gap limit, so it works in millimeters. A flat face of width W on radius R stands off the surface by about W² ÷ (8R) at its edges. Setting that gap to 0.5 mm gives R = W² ÷ 4.

Worked example: a girth weld on 6 in pipe

Inputs: NPS 6 pipe, outside diameter 168.3 mm, so R = 84.2 mm; wedge width 20 mm; beam pointed along the axis to scan the girth weld.

1. Limit: W² ÷ 4 = 400 ÷ 4 = 100 mm.
2. R = 84.2 mm, which is less than 100 mm, so the rule calls for a contoured wedge.
3. Check the gap: 400 ÷ (8 × 84.2) = 0.59 mm, which is over the 0.5 mm limit.

With a 12 mm wide wedge the limit falls to 36 mm, so a flat wedge would pass on this pipe. Evident also suggests a smaller wedge as an alternative to contouring, where the procedure allows it. Recheck the index point and refracted angle with the contoured wedge fitted, because reshaping the face changes both.

How is shear wave UT used on rail?

Rail flaw detection uses shear waves to find cracks that cross the rail, because a straight beam cannot see a flaw lying in the same plane as its own travel. The 70° shear-wave probe is the standard tool for transverse defects in the rail head.

A typical rail search unit combines several beams. One U.S. patent for a real-time rail test system (US 5,777,891, Dapco Industries) describes the set this way:

Probe sets vary by manufacturer. The internal defects that count as rail service failures under federal track rules are a transverse fissure, detail fracture, compound fissure or vertical split head, as defined in 49 CFR 213.237(j). The same section requires search equipment to reach the rail under joint bars:

"Inspection equipment shall be capable of detecting defects between joint bars, in the area enclosed by joint bars." (49 CFR 213.237(f))

At joints, the angled beams are what reach under the bars. Our guides to the transverse fissure in rail and bolt hole cracks show how each defect looks to these probes.

Which calibration blocks are used for shear wave UT?

The IIW reference block, and smaller blocks such as the AWS DSC block, are used to set range, find the beam index point, measure the true refracted angle and set a sensitivity reference before scanning. Evident's angle-beam guidance states that test procedures generally require these checks on "an IIW reference block or similar reference standard before beginning testing."

Calibration blocks are made in several materials, commonly 1018 carbon steel, 4340 steel, 304 stainless and 7075-T6 aluminum. The block material must match the acoustic properties of the part, or the angle and range checks are wrong.

A practical calibration order for a shear-wave probe

1. Index point: peak the echo from the block's radius and mark the beam exit point on the wedge.
2. Range and zero: set the screen from known radius echoes so that sound path reads correctly in steel.
3. Refracted angle: peak the echo from the side-drilled hole and read the true angle from the block scale. A worn wedge can shift the angle by a degree or more.
4. Sensitivity: set the reference level on the reflector the procedure names, then build a DAC or TCG if required. See our DAC curve guide.

For calibration requirements in rail testing contracts, see ultrasonic testing calibration for rail inspection RFPs.

Which standards and qualifications apply?

The method is governed by the code or specification for the product, and the people doing it are qualified to their employer's written practice.

ASNT's June 2026 advisory confirms that SNT-TC-1A certifications "should only be issued by employers as part of their internal qualification system" (ASNT advisory). Our SNT-TC-1A requirements guide explains the levels.

What are the limits of shear wave ultrasonic testing?

Shear wave ultrasonic testing is weak on flaws that lie parallel to the scanning surface, on rounded flaws, and in coarse-grained or anisotropic materials, and it produces no image on a conventional A-scan. An inspection plan should say which method covers each gap.

A scan plan should show that every part of the weld or rail head is covered by at least one beam at a useful angle.

How EMA Industries applies this

EMA Industries uses shear-wave angle-beam testing in ultrasonic rail flaw testing for transit agencies, short lines and industrial track owners, in rail weld inspection, and in AWS D1.1 structural weld UT. We also provide NDT Level III services. We're based in Southern California and mobilize nationwide. To scope rail testing, see our ultrasonic testing services or contact EMA Industries for weld UT and Level III support.

Related reading: How ultrasonic rail flaw detection works · Near field vs far field in UT · Ultrasonic couplant for rail inspection

Frequently asked questions

Is shearwave UT the same as angle beam testing?

In practice, nearly always. Most angle-beam testing in steel uses a wedge between the two critical angles, so the beam in the part is a shear wave. The exception is refracted longitudinal-wave probes, which are angled but used mainly for austenitic welds.

Why do angle-beam wedges convert sound to shear waves?

The wedge is cut beyond the first critical angle, about 27.6° for acrylic on steel. At that angle the refracted longitudinal wave cannot enter the steel, so the energy that crosses the boundary does so as a shear wave.

Can shear waves find flaws parallel to the surface?

Not reliably. A flaw parallel to the scanning surface reflects an angled beam away from the probe. Straight-beam (0°) longitudinal testing is the right method for laminations and horizontal splits.

What frequency is used for shear wave weld testing?

AWS D1.1 structural work commonly uses 2 to 2.5 MHz. Higher frequencies such as 4 or 5 MHz give better resolution on thin sections, and lower frequencies help in coarse-grained or thick material, as the written procedure allows.

How often should the refracted angle be checked?

Check it on an IIW or similar block before testing and at the intervals the procedure sets. Recheck it after any wedge change or visible wear, because wear changes the incident angle and moves the index point.

Sources

• Material Sound Velocities, Evident (Olympus IMS) ultrasonic tutorial, accessed October 2026
• An Introduction to Angle Beam Assemblies, Evident white paper (steel shear velocity 3,250 m/s, mode conversion, IIW verification, the steel-wedge-on-other-metals formula, aluminum wedges, and the contoured-wedge rule citing the IIW Handbook)
• Ultrasonic wedges study, Ultragarsas (Ultrasound), Kaunas University of Technology (duralumin shear velocity 3,130 m/s)
• Angle Beam Inspection, Evident flaw detection tutorial, section 5.2
• Test Blocks, Evident catalog (IIW Type 1, ISO 7963, DSC, DC, SC, RC blocks and materials)
• ASTM E164-24, Standard Practice for Contact Ultrasonic Testing of Weldments, ASTM International
• AWS Announces Release of D1.1/D1.1M:2025, Structural Welding Code – Steel, American Welding Society, April 17, 2025
• ISO 2400, Non-destructive testing — Ultrasonic testing — Specification for calibration block No. 1, ISO (2012 edition; revised 2025)
• 49 CFR 213.237, Inspection of rail, eCFR; text checked on Cornell LII
• US 5,777,891, Method for real-time ultrasonic testing system, Dapco Industries, 1998
• Advisory on SNT-TC-1A Certification Practices and Use of ASNT Credentials, ASNT, June 11, 2026