What Is Guided Wave Ultrasonic Testing? Modes, Range and Limits

Guided wave ultrasonic testing (GWUT) is a screening method that sends low-frequency ultrasonic waves, in the tens of kilohertz, along a pipe from one transducer ring and detects changes in its cross-section. Under 49 CFR 192 Appendix F, a pipeline GWUT assessment must detect metal loss of 5% of the cross-sectional area or less.

GWUT is also called guided wave testing (GWT) and long-range ultrasonic testing (LRUT). These are one method under three names, not competing techniques. It tells you where to look; it does not measure wall thickness.

This guide explains what guided waves are, why they run at low frequency, how a screening test is set up, how range and sensitivity are expressed, which standards apply, where the method is used, and where it fails.

What is guided wave ultrasonic testing?

Guided wave ultrasonic testing is an ultrasonic screening method in which the structure itself acts as a waveguide. A ring of transducers clamped around a pipe excites a low-frequency wave that fills the whole wall and travels along the pipe in both directions. Welds, supports, corrosion and other cross-section changes reflect part of the wave back to the ring.

ASTM E2775, the US practice for piezoelectric guided wave testing of pipe, settles the naming question in its scope. It describes guided wave testing (GWT) as "also previously known as long range ultrasonic testing (LRUT) or guided wave ultrasonic testing (GWUT)". EMA's article on guided wave vs long-range ultrasonic testing covers the terminology in more depth.

The same standard is equally clear about what the method is for:

"GWT is a screening tool. The method does not provide a direct measurement of wall thickness or the exact dimensions of defects/defected area; an estimate of the defect severity however can be provided." — ASTM E2775-16(2023), Scope 1.3

That is the key difference from conventional ultrasonic testing. A conventional 0° probe measures the wall directly under the probe, at megahertz frequencies, one spot at a time. A guided wave test covers the full pipe volume over tens of feet from one position, but it only says that something has changed and roughly how much.

For a refresher on how bulk ultrasonic waves behave, see EMA's guide to the basic principles of ultrasonic testing.

Key takeaway: guided wave ultrasonic testing finds where a pipe has changed; conventional UT then measures what has changed.

What are guided waves, and which modes are used?

Guided waves are elastic waves whose shape is set by the boundaries of the part they travel in, such as the inner and outer walls of a pipe. Instead of a narrow beam, the whole wall thickness vibrates together, so the wave can travel a long way along the part without spreading out into the bulk.

Each geometry supports its own families of guided waves, called modes:

• Lamb waves are the guided modes of a free plate. They come in symmetric and antisymmetric families (s0, a0, s1, a1 and so on). Strictly, "Lamb wave" refers to plates; it is loosely used for other shapes.
• Pipe modes are named with the convention X(n,m), where X is the mode family, n is the circumferential order and m is a counter. The ones used in screening are listed in the table below.
• Rayleigh waves are surface waves on a thick solid. They are not the waves used in pipe screening, despite claims in some older articles, including the earlier version of this page.

The two axisymmetric modes are the workhorses. 49 CFR 192 Appendix F, PHMSA's criteria for using GWUT as a pipeline integrity assessment, says plainly that "Both torsional and longitudinal waves must be used and use must be documented."

The flexural modes matter for interpretation. Demma and colleagues (2003) found that a crack confined to part of the circumference converts some of the torsional wave into the F(1,2) mode, and at lower frequencies into F(1,3). A strong flexural response alongside the torsional echo indicates a localized, one-sided feature, while a weld cap reflects symmetrically.

What is dispersion?

Dispersion means a mode's velocity changes with frequency. A test burst always contains a band of frequencies, so in a dispersive mode the burst stretches out as it travels. The echo arrives longer and lower, which costs both sensitivity and the ability to separate two features close together.

Imperial College London, where much of the pipe method was developed, states in its research impact case study that there are "over 50 at 100 kHz in a typical 6 inch diameter pipe". Only three of them are axisymmetric, and unless excitation is controlled, the received signals are too complicated to interpret. That is why commercial systems use rings of transducers spaced evenly around the pipe: the ring drives the axisymmetric T(0,1) or L(0,2) mode and suppresses most of the others.

Choosing T(0,1) solves most of the dispersion problem, because it has none. In a dispersive mode, however, arrival times must be converted to distance with the group velocity, not the phase velocity, and that error is worked through for rail in EMA's long-range ultrasonic rail inspection guide.

Key takeaway: GWUT works because the ring excites one or two clean axisymmetric modes, and the torsional mode does not disperse.

Why does guided wave testing use such low frequencies?

Guided wave tests run at tens of kilohertz because low frequencies keep the number of modes small and the attenuation low, which is what gives the long range. ASTM E2929, the practice for magnetostrictive guided wave testing, states that GWT uses "ultrasonic guided waves in the 10 to approximately 250 kHz range".

The price of low frequency is a long wavelength. Wavelength is velocity divided by frequency, and the guided wavelength is far longer than the defects a conventional probe can size. A long wavelength responds to how much of the cross-section is missing, not to the fine shape of a pit.

Worked example: distance and wavelength for a torsional wave

Take a steel pipe tested with the T(0,1) mode at 25 kHz. Because T(0,1) travels at the shear wave velocity of the steel, use 3,250 m/s (about 10,660 ft/s), a typical value for carbon steel at room temperature. Measure the actual velocity on a known weld before relying on it.

Step 1: locate a reflector. An echo arrives 18.46 ms after the burst. In pulse-echo the wave goes out and back, so:

d = 3,250 m/s × 0.01846 s ÷ 2 = 30.0 m (98.4 ft) from the ring

Step 2: find the wavelength.

λ = 3,250 ÷ 25,000 = 0.130 m (5.1 in.)

Step 3: compare it with the axial-length rule. ASTM E2775 (5.4) says GWT gives some information on a discontinuity's axial length only when that length is longer than "roughly a quarter of the wavelength". At 25 kHz that is 0.130 ÷ 4 = 32.5 mm (1.28 in.). A corrosion patch shorter than about 1.3 in. along the pipe gives no useful length information at this frequency.

Step 4: see what raising the frequency does. At 50 kHz the wavelength halves to 65 mm (2.6 in.), and the quarter-wavelength limit falls to about 16 mm (0.64 in.). Demma et al. found the torsional reflection from cracks rises with frequency at a given depth. Higher frequency also brings more attenuation and more modes, so the range shortens.

This trade-off is why Appendix F requires "a minimum of three frequencies" for each pipeline inspection. One frequency may miss a feature that another picks up.

Key takeaway: low frequency buys range and costs resolution, so every GWUT result is tied to the frequency it was taken at.

How does a guided wave screening test work?

A guided wave screening test clamps a transducer ring at one access point, sends bursts in both directions, and maps every reflector by distance and estimated cross-section change. The steps below follow the order a typical pipe test takes, with the Appendix F criteria noted where they apply to pipelines.

1. Prepare one access point. The ring needs a short length of bare, sound pipe, often a window cut in the insulation or a small excavation. The rest of the line can stay insulated, painted, buried or sleeved.
2. Fit the ring. Rings are solid or inflatable collars sized to the pipe. Piezoelectric rings press transducer elements onto the surface; magnetostrictive systems use a strip bonded to the pipe and a coil.
3. Run several frequencies and both mode types. The operator collects torsional and longitudinal data at several frequencies to find the best signal for the line. Appendix F requires at least three frequencies and both wave types.
4. Set the distance-amplitude correction. Girth welds are the reference reflectors. Appendix F says a conservative estimate for a weld's reflection is 25% CSA when welds can't be reached, and requires the DAC to be set for each inspection. EMA's guide to the DAC curve in ultrasonic testing explains the same principle at megahertz frequencies.
5. Account for the dead zone and near field. Next to the ring, the outgoing pulse blinds the receiver. Appendix F requires the collar to be moved, or the zone covered by B-scan UT and visual examination, and the lengths documented. The idea is the same as the dead zone in conventional UT.
6. Interpret the trace. Each reflector is classified as a known feature (weld, flange, support, branch) or an unexplained indication, with its distance, estimated CSA change and circumferential position from the flexural response.
7. Confirm range from both ends. Appendix F expects double-ended inspections, overlaid to show at least 2:1 signal-to-noise in the middle of the segment.
8. Report and prove up. Indications above the threshold go to direct examination, usually by conventional UT thickness mapping, before the pipe is cleared.

ASTM E2929 (1.6) warns that "the interpretation of the guided wave data obtained is complex and training is required to properly perform data interpretation." Appendix F goes further for pipelines: equipment operators must be qualified by the manufacturer or an independently assessed procedure "similar to ISO 9712", and a senior-level operator must oversee the inspection and approve the report.

Key takeaway: the ring position, frequencies, weld calibration and dead zone all decide what a GWUT trace can and can't show.

How is sensitivity measured as a percentage of cross-section?

GWUT sensitivity is expressed as the smallest change in the pipe's metal cross-sectional area (CSA) that the test can reliably detect, not as a defect depth. The reflected amplitude depends on how much steel is missing across the whole section. A narrow, deep pit and a wide, shallow patch can give similar echoes.

Research on notches in pipe shows why. Alleyne, Lowe and Cawley (1998) found that the L(0,2) reflection coefficient is "very close to a linear function of the circumferential extent of the notch, and is a stronger function of the through thickness depth". Depth and circumferential extent together set the echo, which is the CSA loss in another form.

ASTM E2775 (5.2) puts it the same way: GWT "is sensitive to a combination of the CSC and circumferential extent and axial extent of any metal loss", where CSC is cross-sectional change. The regulatory threshold for pipelines comes from PHMSA:

"The maximum threshold sensitivity cannot be greater than 5% of the cross sectional area (CSA)." — 49 CFR 192, Appendix F, IV

Worked example: what a 5% CSA threshold means on two pipe sizes

Take one corrosion patch, 0.14 in. deep and 2 in. wide around the circumference, and put it on two pipes.

Step 1: metal area of a 6 in. Schedule 40 pipe (OD 6.625 in., wall 0.280 in., ID 6.065 in.):

A = π/4 × (6.625² − 6.065²) = π/4 × (43.891 − 36.784) = 5.581 in²

Step 2: metal area of a 24 in. standard-wall pipe (OD 24.000 in., wall 0.375 in., ID 23.250 in.):

A = π/4 × (24.000² − 23.250²) = π/4 × (576.000 − 540.563) = 27.833 in²

Step 3: the patch's lost area. As a simple rectangle, 0.14 in. × 2 in. = 0.28 in².

Step 4: CSA loss on each pipe.

The patch is 50% of the wall on the 6 in. pipe and 37% of the wall on the 24 in. pipe, yet on the larger pipe it removes only 1% of the section. On a 24 in. pipe, a 5% CSA patch at half the wall depth would need to be about 7.4 in. wide.

So a GWUT "no indication" result means no cross-section loss above the threshold, not "no pitting". An isolated deep pit can be a leak risk well below 5% CSA. That is why ASTM E2775 (5.1) frames the method as locating areas where "wall loss has occurred due to corrosion or erosion", and leaves measurement to other methods.

Key takeaway: always read a GWUT threshold together with the pipe size, because the same pit is a bigger percentage of a smaller pipe.

How far can guided wave ultrasonic testing reach?

Range is set by attenuation and signal-to-noise, so it varies from a few feet to more than a hundred feet in each direction depending on the pipe's coating, contents, surroundings and fittings. For buried and cased pipelines, PHMSA gives a working figure:

"In general, the inspection range can approach 60 to 100 feet for a 5% CSA, depending on field conditions." — 49 CFR 192, Appendix F, II

ASNT's method page describes GWUT as able to inspect "tens of meters of pipe in both directions from a single sensor location". A bare, straight, above-ground line in good condition gives the longest range; a coated, buried or concrete-encased line gives the shortest.

Worked example: a simple range budget

The range for pulse-echo screening can be estimated from a decibel budget. Kwun and Kim (2005), in a PRCI report on high-power magnetostrictive guided waves for pipelines, give the range gained from extra signal as dB ÷ (2α), where α is the attenuation in dB per foot and the 2 accounts for the round trip.

Assume the 5% CSA DAC level starts 20 dB above the noise floor just beyond the near field. These figures are illustrative assumptions, not measured values.

Step 1: required signal-to-noise. Appendix F's 2:1 minimum in amplitude is 20 × log₁₀(2) = 6.0 dB.

Step 2: available margin. 20 dB − 6.0 dB = 14.0 dB.

Step 3: range for different attenuations.

The 0.10 dB/ft row lands inside Appendix F's 60 to 100 ft figure. The point is the shape of the relationship: doubling the attenuation halves the range, which is why a test plan on a coated line includes more ring positions.

Key takeaway: quote GWUT range only with its threshold and conditions, for example "70 ft at 5% CSA on painted pipe".

Which standards govern guided wave testing?

Guided wave testing of pipe is covered by ASTM E2775 and E2929, BS 9690 Parts 1 and 2, and ISO 18211, while 49 CFR 192 Appendix F sets binding criteria for gas pipeline integrity assessments. None of the practice standards sets acceptance criteria; that comes from the contract or the regulator.

Three points help when writing a specification:

• The ASTM practices are limited to "basic pipe configuration". Both define it as straight pipe of one size and schedule, fully accessible at the test location, joined by girth welds, on simple contact supports, and free of internal or external coatings, though "the pipe may be insulated or painted". Anything else needs a procedure written for the case.
• ASTM E2775 excludes data interpretation and acceptance. Its scope says "actual interpretation of the data is excluded" and that acceptance criteria "shall be specified in the contractual agreement".
• Appendix F is a regulatory "Go-No Go" scheme. Any GWUT application that doesn't meet it is treated as "other technology", for which PHMSA's Office of Pipeline Safety must be notified 90 days before use under 49 CFR 192.921(a)(7) or 192.937(c)(7).

ISO 18211's scope, as reproduced by catalog listings, describes GWT as a fast screening method that locates probable corroded or eroded areas but does not quantify them. A 2014 NRC letter also cites NACE SP0313-2013, Guided Wave Technology for Piping Applications; check the current edition with AMPP before citing it.

Key takeaway: specify the standard, the threshold in % CSA, the frequencies and the prove-up method; the standards themselves won't supply an acceptance level.

Where is guided wave ultrasonic testing used?

GWUT is used mainly on pipe and pipelines where the steel can't be reached for point inspection without major cost, such as insulated lines, road crossings and pipe under supports. Cawley's 2024 review in NDT & E International sums up the method's niche: it is "relatively mature, and largely limited to simple structures in applications that are difficult or impossible to address by conventional point inspection methods."

Permanent monitoring is the main recent development. Cawley's review notes that sensitivity can be improved "relative to that obtained in one-off inspection by monitoring the change in signals with time using baseline subtraction". A 2017 BINDT presentation by the manufacturer Guided Ultrasonics Ltd (vendor material) says a residual of about 1% CSC needs the new reading within roughly 5 to 10 °C of its baseline.

What about non-metallic pipe? The standards above cover carbon and low-alloy steel only. Polymers and composites attenuate guided waves much more strongly, so the original article's claim that GWUT is "effective for both metallic and non-metallic materials" is not supported for screening use.

Key takeaway: GWUT earns its place where access is the expensive part of the inspection.

Can guided waves inspect rail?

Guided waves can travel along rail, but rail guided-wave testing is mainly a research and broken-rail-monitoring tool, not a replacement for conventional rail flaw testing. Rail's irregular head, web and foot support many dispersive modes, and we found no published standard for guided wave testing of rail in track.

The longest rail ranges in the literature come from pitch-catch systems that detect complete breaks, not from pulse-echo defect detection. Guided wave results also do not give the valid search that 49 CFR 213.237 requires; that search is run with conventional rail flaw testing at megahertz frequencies.

EMA's article on long-range ultrasonic rail inspection covers rail modes, published ranges, a time-of-flight example and the 213.237 position in detail. For how conventional rail testing works, see how ultrasonic testing works for railways.

Key takeaway: on rail, guided waves supplement the required test; they don't replace it.

What are the limits of guided wave ultrasonic testing?

GWUT can't measure wall thickness, can miss small isolated pits, loses range on coated or buried pipe, and struggles with complex geometry. These limits come from the physics, so they apply to every system.

• No thickness measurement. ASTM E2775 and E2929 both say GWT does not measure wall thickness, and E2775 (5.5) calls severity assessment "qualitative only".
• Threshold in % CSA, not depth. A deep, narrow pit can be below the threshold, as the worked example shows. ASNT warns that GWUT "may not detect very small or gradual defects, especially at greater distances from the sensor ring."
• Gradual wall loss. General thinning over a long length produces little reflection, because there is no sudden cross-section change to reflect from.
• Coatings and burial. Bitumen, thick wraps, compacted soil and bonded concrete shorten the range, sometimes to a few feet.
• Geometry. Tees, elbows, flanges, valves and branches reflect and mode-convert energy. Signals past a bend are harder to interpret, and a flange usually ends the test.
• Dead zone and near field. The pipe next to the ring is not tested from that position and must be covered by moving the ring or by conventional UT.
• Shorted casings. Appendix F (XVII) says plainly: "GWUT may not be used to assess shorted casings."
• Interpretation depends on the operator. Feature classification is a judgment, and ASTM E2775 leaves interpretation out of its scope altogether.

Key takeaway: treat a clean GWUT trace as "nothing above threshold within the tested range", and state that range and threshold in the report.

How is a guided wave call-out followed up?

A guided wave call-out is followed up by direct examination at the indicated distance, usually with conventional ultrasonic thickness measurement, visual examination and, where needed, other NDT. The call-out gives a distance from the ring, an estimated CSA change and often a clock position.

A practical prove-up sequence looks like this:

1. Locate the indication on the pipe. Measure the distance from the ring datum, allowing for the reported position accuracy, and expose a length either side.
2. Inspect visually. Remove insulation, coating or soil, and record what is found.
3. Map the wall with conventional UT. A 0° thickness scan or grid finds the minimum remaining wall that GWUT can't give. EMA's guide to the ultrasonic thickness gauge covers the instrument; encoded mapping or phased array ultrasonic testing gives a C-scan of a larger patch.
4. Check for cracking where relevant. At welds or supports, angle-beam UT or a surface method may be needed, because a crack and a corrosion patch can give similar guided wave responses.
5. Feed the result back. Compare the measured loss with the GWUT estimate. A large mismatch suggests the DAC or the feature classification needs revisiting for the rest of the line.

Under Appendix F, the timing of direct examination depends on operating pressure. Above 50% SMYS, the operator must replace or directly examine within 6 months, run an instrumented leak survey every 30 calendar days, and reduce MAOP to 80% of the operating pressure at discovery.

Key takeaway: GWUT finds the dig; conventional UT decides what the dig means.

How EMA Industries applies this

EMA Industries' part in guided wave work is conventional ultrasonic testing: ultrasonic rail flaw testing for transit agencies, short lines and industrial track owners, working from Southern California and mobilizing nationwide. Where a project includes guided wave screening by others, possible scope for EMA's NDT Level III services includes reviewing the GWUT procedure and report and planning the conventional UT follow-up of its call-outs. See EMA's ultrasonic examination services or contact EMA Industries to discuss the scope.

Related reading

• Long-range ultrasonic rail inspection
• Guided wave vs long-range ultrasonic testing
• Types of ultrasonic testing in NDT
• DAC curve in ultrasonic testing
• Dead zone in ultrasonic testing

Frequently asked questions

How far can guided wave ultrasonic testing inspect from one ring?

It depends on the pipe. PHMSA's Appendix F says the range for buried and cased pipelines "can approach 60 to 100 feet for a 5% CSA, depending on field conditions", and ASNT describes tens of meters in both directions. Bare, straight pipe gives the longest range; heavy coatings, compacted soil and bonded concrete can cut it to a few feet.

What size of defect can GWUT detect?

GWUT detects cross-section change, not a defect depth. Pipeline assessments under 49 CFR 192 Appendix F must reach a threshold of 5% of the cross-sectional area or better. On a 6 in. Schedule 40 pipe that equals a patch half the wall deep and about 2 in. wide, while on a 24 in. pipe the same patch is only about 1%.

Does guided wave testing measure wall thickness?

Not directly. ASTM E2775 and E2929 both state that GWT "does not provide a direct measurement of wall thickness". Remaining wall is measured afterward with conventional ultrasonic thickness readings or mapping at each call-out.

Why are both torsional and longitudinal waves used?

The two modes respond differently to features and conditions. The torsional mode T(0,1) does not disperse and isn't affected by liquid in the pipe, while the longitudinal mode L(0,2) adds a second view of each reflector. Appendix F requires both wave types to be used and documented for pipeline assessments.

Is guided wave testing the same as long-range ultrasonic testing?

They are the same method. ASTM E2775 calls it guided wave testing and notes it was "previously known as long range ultrasonic testing (LRUT) or guided wave ultrasonic testing (GWUT)".

Sources

• ASTM E2775-16(2023), Standard Practice for Guided Wave Testing of Above Ground Steel Pipework Using Piezoelectric Effect Transduction: ASTM store (scope and significance sections)
• ASTM E2929-18(2022), Standard Practice for Guided Wave Testing of Above Ground Steel Piping with Magnetostrictive Transduction: ASTM store
• BS 9690-1:2011, Non-destructive testing. Guided wave testing. General guidance and principles: BSI Knowledge
• BS 9690-2:2011, Basic requirements for guided wave testing of pipes, pipelines and structural tubulars: BSI Knowledge
• ISO 18211:2016, guided wave testing of above-ground pipelines and plant piping: Genorma catalog listing
• 49 CFR Part 192, Appendix F, Criteria for Conducting Integrity Assessments Using Guided Wave Ultrasonic Testing (GWUT): eCFR (text read on Cornell LII; Amdt. 192-125, 84 FR 52255, Oct. 1, 2019, as amended by 91 FR 22017, Apr. 24, 2026)
• 49 CFR 213.237, Inspection of rail: eCFR
• ASNT, "Guided Wave Testing": asnt.org
• Cawley P. "Guided waves in long range nondestructive testing and structural health monitoring: Principles, history of applications and prospects." NDT & E International 142:103026 (2024): doi.org/10.1016/j.ndteint.2023.103026
• Demma A, Cawley P, Lowe M, Roosenbrand AG. "The reflection of the fundamental torsional mode from cracks and notches in pipes." JASA 114:611–625 (2003): doi.org/10.1121/1.1582439
• Alleyne DN, Lowe MJS, Cawley P. "The reflection of guided waves from circumferential notches in pipes." Journal of Applied Mechanics 65:635–641 (1998): doi.org/10.1115/1.2789105
• Alleyne DN, Pavlakovic B, Lowe MJS, Cawley P. "Rapid, long range inspection of chemical plant pipework using guided waves." Key Engineering Materials 270–273:434–441 (2004): doi.org/10.4028/www.scientific.net/kem.270-273.434
• Leinov E, Lowe MJS, Cawley P. "Investigation of guided wave propagation in pipes fully and partially embedded in concrete." JASA 140:4528–4539 (2016): doi.org/10.1121/1.4972118
• Kwun H, Kim SY. High-Power Long-Range Guided-Wave Inspection of Pipelines, PRCI report (July 2005): doi.org/10.55274/r0012077
• Imperial College London, "Commercialisation of Guided Wave Inspection for the Detection of Corrosion in Pipes", REF impact case study: impact.ref.ac.uk
• Imperial College London NDE Group, "The long range inspection of pipework" (archived project page): imperial.ac.uk
• Alleyne D, Pialucha T, Pavlakovic B (Guided Ultrasonics Ltd). "Guided Wave Testing and Monitoring Over Long and Short Ranges." BINDT presentation, Sept 6, 2017 (vendor material): bindt.org PDF