Skip to content

NDT Methods: UT, RT, MT, PT, VT

Non-Destructive Testing (NDT)

NDT techniques detect defects in materials and welds without destroying the test piece. This allows inspection of every item (unlike destructive testing, which sacrifices samples) and continued use after testing. In oil and gas piping, NDT is governed primarily by ASME B31.3 (process piping), ASME Section V (examination methods), and ASME Section VIII (pressure vessels).

The terms NDT and NDE (Non-Destructive Examination) are used interchangeably in industry. ASME codes generally use “examination,” while API and ISO standards tend to use “testing” or “inspection.”

Detectable Flaws

Defect TypeDescription
CracksSurface or subsurface fractures from manufacturing, service, or fatigue
InclusionsForeign materials embedded during production
PorosityVoids or gas pockets weakening the structure
LaminationsLayer separations in composites or rolled products
Weld defectsIncomplete penetration, lack of fusion, undercut, porosity
CorrosionMaterial loss from chemical attack
ErosionSurface wear from abrasive substances
Dimensional variationsThickness deviations from spec

Who Performs NDT?

NDT technicians require certification from recognized bodies such as ASNT (American Society for Nondestructive Testing). Certification levels (I, II, III) indicate increasing expertise and authority to interpret results.

Third-party inspection agencies (SGS, Lloyd’s, DNV, Bureau Veritas) provide independent NDT services for critical applications where unbiased results are required (oil & gas, aerospace, nuclear, and construction).

NDT vs. Destructive Testing

AspectNDTDestructive Testing
Effect on sampleNo damage; item remains usableSample destroyed
Sample rateCan test 100% of productionSample-based only
Data typeDefect detection and locationUltimate strength, failure mode
ApplicationOngoing inspection, quality controlDesign verification, material qualification

Selecting the Right NDT Method

NeedRecommended Method
Internal flaws in thick materialsUT (Ultrasonic)
Internal weld structureRT (Radiographic)
Surface cracks on ferromagnetic steelMT (Magnetic Particle)
Surface cracks on non-magnetic materialsPT (Penetrant Testing)
Thin-wall tubing inspectionET (Eddy Current)
Quick initial assessmentVT (Visual Testing)
Accurate flaw sizing for fitness-for-serviceTOFD
Cross-sectional weld imagingPhased Array UT

NDT allows inspection of every item without destroying the test piece, unlike destructive testing which sacrifices samples. The six main methods each detect different defect types: VT for surface defects, PT for surface-breaking cracks on non-magnetic materials, MT for surface and near-surface flaws on ferromagnetic steel, UT for internal flaws, RT for internal weld structure, and ET for thin-wall tubing inspection. Advanced techniques such as TOFD, phased array UT, and digital radiography offer improved accuracy, speed, and permanent records.

The 6 Key NDT Methods

MethodAbbreviationDetectsMaterials
Visual TestingVTSurface defectsAll
Liquid PenetrantPT / LPT / DPISurface-breaking cracksNon-porous only
Magnetic ParticleMT / MPI / MPTSurface and near-surfaceFerromagnetic only
UltrasonicUTInternal and surfaceMost metals
RadiographicRT / RXInternal structureDense materials
Eddy CurrentET / ECTSurface and subsurfaceConductive only

Visual Testing (VT)

The simplest and cheapest NDT method, and the first step in any inspection regime. Inspectors examine surfaces with the naked eye or magnifying tools to identify visible defects. All welds must undergo VT as a minimum requirement per ASME B31.3 para. 344.2.

RequirementDetails
StandardASME Section V, Article 9
Surface preparationClean thoroughly before examination
Defect typesCracks, porosity, undercut, misalignment, corrosion
LimitationSurface defects only; cannot detect subsurface flaws
Light requirementsMinimum 1000 lux (100 fc) at the examination surface
Follow-upIf defects found, additional NDT methods determine extent

Liquid Penetrant Testing (PT)

liquid penetrants test

PT testing detects surface-breaking defects in non-porous materials (metals, plastics, ceramics) using capillary action to draw colored dye into cracks. It is the standard surface examination method for austenitic stainless steel, duplex, nickel alloys, and aluminum where MT cannot be used.

StepAction
1. CleanRemove all contaminants from surface
2. Apply penetrantSpray or brush colored dye onto surface
3. DwellAllow time for penetrant to seep into defects (typically 10-30 minutes)
4. Remove excessWipe off surface penetrant with solvent/emulsifier
5. Apply developerWhite powder draws penetrant out of defects
6. InspectColored indications reveal crack locations

Penetrant types: Type I (fluorescent, viewed under UV-A light, higher sensitivity) and Type II (visible dye, red contrast, used in field conditions). Most piping fabrication shops use Type II visible dye for convenience; Type I fluorescent is specified for critical components such as high-pressure valves and nuclear piping.

ProsCons
Simple, low-costSurface defects only
Works on non-magnetic materialsNot for porous materials
Detects cracks invisible to naked eyeRequires clean, smooth surface
Portable, no power needed (visible dye)Temperature sensitive (typically 10-52 deg C)

Magnetic Particle Testing (MT)

magnetic particles test

MT testing detects surface and near-surface flaws in ferromagnetic materials only (carbon steel, low-alloy steel, ferritic/martensitic stainless steel). The component is magnetized, and magnetic particles accumulate at flux leakage points caused by defects.

StepAction
1. MagnetizeApply magnetic field using yoke, prods, or coil
2. Apply particlesSpray dry powder or wet suspension onto surface
3. InspectParticles cluster at defect locations, forming visible indications
4. DemagnetizeRemove residual magnetism (required by most specifications)

Magnetization techniques: AC yoke (most common in field), DC prods (deeper penetration but risk arc strikes), and multi-directional units (detect flaws in all orientations in one shot). For weld inspection, two magnetization passes at 90 degrees are typically required to detect defects in both longitudinal and transverse orientations.

ProsCons
Detects surface and near-surface defectsFerromagnetic materials only
Fast, cost-effectiveCannot detect deep internal flaws
Portable equipment availableComponent must be demagnetized after test
Works through thin coatings (up to ~50 microns)Requires two passes for full coverage

Ultrasonic Testing (UT)

ultrasonic test

UT uses high-frequency sound waves (typically 1-10 MHz) to detect internal flaws and measure wall thickness. A transducer probe sends ultrasonic pulses into the material; reflections from defects or back walls are analyzed to determine flaw location, size, and characteristics.

StepAction
1. SetupApply couplant (gel/oil) and position probe on surface
2. CalibrateSet range and sensitivity using reference blocks with known reflectors
3. TransmitProbe sends ultrasonic waves into material
4. ReceiveProbe detects reflected waves from defects or back wall
5. AnalyzeSoftware calculates defect location, size, and depth

Conventional UT techniques: Straight beam (0-degree, for thickness measurement and lamination detection) and angle beam (45, 60, or 70 degrees, for weld examination). Angle beam UT is the standard volumetric examination method for butt welds in process piping per ASME B31.3.

ProsCons
Detects internal and surface flawsRequires skilled operator
Accurate depth/size measurementSurface must be accessible and smooth
Portable, no radiation hazardCoarse-grained materials challenging (cast SS, Inconel)
Real-time resultsCalibration required for each setup change
Single-sided access sufficientNo permanent image record (conventional UT)

Radiographic Testing (RT)

RX test pipe

RT uses X-rays or gamma rays to create images of a component’s internal structure. Radiation passes through the material and exposes film or a digital detector on the opposite side. Defects appear as density variations in the image.

StepAction
1. SetupPosition radiation source on one side, film/detector on opposite
2. ExposeActivate source; radiation penetrates component
3. ProcessDevelop film or process digital image
4. InterpretTrained interpreter examines image for defect indications per acceptance criteria

Radiation sources: X-ray tubes (adjustable energy, better image quality, needs power supply) and gamma ray isotopes: Ir-192 (for steel up to ~75 mm) and Co-60 (for thicker sections). Gamma sources are portable and need no electricity, making them standard for field work on pipelines.

ProsCons
Permanent record of inspectionRadiation safety requirements (exclusion zones)
Detects internal voids, inclusions, porosityAccess needed to both sides of the component
Good for weld inspectionSlower than UT (film processing time)
Works on most materialsEquipment cost, licensing, and storage
Image is intuitive to interpretPoor at detecting planar defects oriented parallel to beam

(Source: Peter Smith, Piping Materials Selection and Applications, 2004)

Eddy Current Testing (ET)

Eddy current testingEddy current testing probe

Eddy current testing finds surface and subsurface defects in conductive materials only through electromagnetic induction. A probe coil generates alternating current that induces eddy currents in the test material; defects cause measurable impedance changes.

StepAction
1. SetupPosition probe coil near material surface
2. GenerateAlternating field induces eddy currents in material
3. DetectDefects/thickness changes alter impedance
4. AnalyzeSignal variations indicate defect location and size
ProsCons
No couplant requiredConductive materials only
Fast inspection speedLimited penetration depth (typically < 5 mm)
Detects small surface cracksSensitive to lift-off variations
Excellent for tube/heat exchanger inspectionInterpretation requires experience

ET testing is particularly effective for inspecting heat exchanger tubes, aircraft structures, and thin-wall tubing where rapid, automated inspection is needed. Internal Rotary Inspection Systems (IRIS) combine UT with a rotating mirror for tube inspection where ET sensitivity is insufficient.

Advanced NDT Methods

Advanced techniques offer improved accuracy, permanent digital records, and faster inspection speeds compared to conventional methods.

Time-of-Flight Diffraction (TOFD)

TOFD uses two angled ultrasonic probes (one transmitter, one receiver) positioned on opposite sides of a weld. Instead of measuring reflected pulse amplitude (like conventional UT), TOFD measures the arrival time of diffracted signals from flaw tips. This provides accurate through-wall flaw sizing independent of flaw orientation.

FeatureDetails
StandardASME Section V, Article 4; EN ISO 10863
AccuracyFlaw height sizing to +/- 1 mm
CoverageFull weld volume in a single scan pass
RecordPermanent encoded digital image (D-scan)
LimitationDead zones at OD and ID surfaces (supplemented with PAUT or pulse-echo)

Phased Array UT (PAUT)

Phased array probes contain multiple piezoelectric elements (typically 16-64) pulsed with controlled time delays. This allows electronic beam steering, focusing, and sweeping without moving the probe. PAUT produces cross-sectional images (S-scans) that are similar to RT images but without radiation.

FeatureDetails
StandardASME Section V, Article 4; ASME CC 2235
Probe elements16 to 128 elements typical
OutputS-scan (sectorial), B-scan (side view), C-scan (plan view)
Speed3-5x faster than conventional angle beam UT
AcceptanceASME, API, EN codes accept PAUT as RT alternative

Digital Radiography (DR) and Computed Radiography (CR)

Digital radiography replaces conventional film with flat-panel detectors (DR) or phosphor imaging plates (CR). Benefits include instant image availability, adjustable contrast/brightness, reduced radiation dose, and elimination of chemical film processing. Digital images are stored electronically and can be reviewed remotely.

Automated Ultrasonic Testing (AUT)

AUT uses mechanized scanners to drive ultrasonic probes around a weld, producing encoded and repeatable data. AUT typically combines PAUT, TOFD, and conventional UT channels in a single scan pass. It is the standard inspection method for cross-country pipeline girth welds under API 1104 and is increasingly specified for shop-fabricated pressure vessel and piping welds.

NDT Method Comparison Table

CriterionVTPTMTUTRTET
Defect locationSurfaceSurface onlySurface + near-surfaceInternal + surfaceInternalSurface + near-surface
Ferromagnetic steelYesYesYesYesYesYes
Austenitic SS / NickelYesYesNoDifficultYesYes
Aluminum / Non-ferrousYesYesNoYesYesYes
Minimum detectable size~1 mm visible~1 micron width~0.5 mm~1 mm~2% of wall thickness~0.5 mm
Subsurface capabilityNoNoShallow (~3 mm)Full thicknessFull thicknessShallow (~5 mm)
Permanent recordPhotos/reportsPhotosPhotosDigital (PAUT/TOFD)Film or digital imageStrip chart/digital
Radiation hazardNoNoNoNoYesNo
Relative costLowLowLow-MediumMediumMedium-HighMedium
SpeedFastSlow (dwell time)FastMediumSlowFast
Personnel level (min.)VT Level IIPT Level IIMT Level IIUT Level IIRT Level IIET Level II

NDT Selection Guide by Defect Type

Selecting the right NDT method depends on the defect being sought, the material, and access conditions.

Defect TypePrimary MethodSecondary MethodNotes
Surface cracks (CS/alloy)MTVTMT detects tighter cracks than VT
Surface cracks (SS/nickel)PTVTMT not applicable to non-magnetic alloys
Subsurface cracksUT (angle beam)RTUT better for planar flaws
PorosityRTUTRT more sensitive to scattered porosity
Slag inclusionsRTUTVolumetric defects show clearly on RT
Lack of fusionUTTOFD/PAUTPlanar defect; RT often misses if parallel to beam
Incomplete penetrationRTUTConsistent orientation makes RT effective
Wall thinning / corrosionUT (straight beam)ETUT measures remaining thickness precisely
Laminations in plateUT (straight beam)RT cannot detect laminations (parallel to beam)
Heat exchanger tube defectsETIRIS (UT)ET is fast and automated for large tube bundles

Personnel Qualification Levels

NDT personnel must be certified per ASNT SNT-TC-1A (employer-based, North America), ASNT CP-189 (central certification), or ISO 9712 (international). Certification is method-specific, so an inspector qualified in UT is not automatically qualified in RT.

LevelAuthorityTypical Requirements
Level IPerform tests under direct supervision of Level II or III; cannot interpret or accept/rejectClassroom training + OJT hours (method-specific); pass general, specific, and practical exams
Level IISet up equipment, perform tests, interpret results, accept/reject per procedures, train Level IAdditional training + OJT hours beyond Level I; pass all exams; most project specifications require Level II minimum
Level IIIDevelop procedures, establish techniques, interpret codes, manage NDT programs, train all levelsExtensive experience; pass ASNT Level III Basic and Method exams; typically requires engineering background

ASME B31.3 Examination Requirements

ASME B31.3 specifies minimum examination requirements based on fluid service category. Project piping specifications and pipe class sheets may increase these minimums but never reduce them.

Examination Extent by Fluid Service

Fluid ServiceButt Welds (RT or UT)Branch WeldsSocket/Fillet WeldsHardness Testing
Normal (Category D)5% randomVT 100%VT 100%Per WPS
Normal5% randomVT 100%VT 100%Per WPS
Category M (lethal)100% RT or UT100% MT or PT100% MT or PT100%
High-Pressure100% RT or UT100% MT or PT100% MT or PT100%
Severe Cyclic100% RT or UT100% MT or PT100% MT or PT100%

Typical Project Examination by Pipe Class

In practice, EPC companies define examination percentages in project piping specifications that exceed B31.3 minimums based on pipe class criticality.

Pipe Class CriticalityButt Weld RT/UTSocket/Fillet MT or PTVisual
Utility (water, air)5-10%0-10%100%
Standard process10-20%10-20%100%
High-temperature / HP100%100%100%
Lethal / toxic service100%100%100%
Sour service (NACE)100%100%100% + hardness

Acceptance Criteria Overview

NDT results are evaluated against acceptance criteria defined in the applicable code. The two most commonly referenced codes for piping and pressure vessels are ASME B31.3 and ASME Section VIII Division 1.

ASME B31.3 (Process Piping)

Examination MethodAcceptance StandardKey Limits
RTASME B31.3 Table 341.3.2; acceptance per ASME Section VIII UW-51No cracks; porosity limited by wall thickness charts; slag length limited
UTASME B31.3 para. 344.6.2; ASME Section V Article 4Reflectors exceeding reference level evaluated for acceptance
MTASME B31.3 para. 344.3.2Relevant linear indications > 1.5 mm rejected; no cracks
PTASME B31.3 para. 344.4.2Relevant linear indications > 1.5 mm rejected; no cracks
VTASME B31.3 Table 341.3.2Per workmanship standards: undercut depth, reinforcement height, misalignment

Applicable Standards Reference

StandardTitle / Scope
ASME Section VNondestructive Examination: methods, techniques, calibration
ASME B31.3Process Piping: examination requirements by fluid service
ASME Section VIII Div. 1Pressure Vessels: UW-51 (RT acceptance), UW-52 (spot RT)
ASNT SNT-TC-1APersonnel Qualification and Certification (employer-based)
ASNT CP-189Standard for Qualification and Certification (central)
ISO 9712NDT Personnel Qualification and Certification (international)
ASTM E94Standard Guide for Radiographic Examination
ASTM E164Standard Practice for Contact Ultrasonic Testing of Weldments
ASTM E165Standard Practice for Liquid Penetrant Testing
ASTM E709Standard Guide for Magnetic Particle Testing
ASTM E1416Standard Practice for Radioscopic Examination of Weldments
ASTM E2698Standard Practice for Radiographic Examination Using Digital Detector Arrays
ASTM E2373Standard Practice for Use of the UT Time-of-Flight Diffraction Technique
EN ISO 10863Welding: Use of TOFD Technique
EN ISO 13588Welding: Use of Phased Array Technique
API 1104Welding of Pipelines: includes AUT acceptance criteria

Practical Tips for NDT in Piping Projects

Plan NDT early in the project schedule. Build NDT hold points into the Inspection and Test Plan (ITP) before fabrication starts. Late NDT planning causes rework, schedule delays, and inflated costs.

Specify methods in the piping specification. Define RT/UT percentages, MT/PT requirements for branch connections and socket welds, and hardness testing requirements for each pipe class. Do not leave it to the fabricator to decide.

Coordinate RT scheduling with other disciplines. Radiation exclusion zones shut down nearby work. Schedule RT during off-hours or weekends. Some projects have switched entirely to PAUT/TOFD to avoid radiation disruptions.

Get surface preparation right. The number one cause of false PT/MT indications is inadequate surface cleaning. This wastes time on unnecessary repairs and re-examination.

Keep calibration blocks traceable. UT reference blocks must have current calibration certificates traceable to national standards. Expired calibration invalidates all examination results obtained with that block.

Document everything. NDT reports, film/digital images, operator certifications, calibration records, and procedure qualifications must be filed in the project data book. Missing documentation is treated the same as missing examination, and the weld must be re-examined.

Frequently Asked Questions

Which NDT method is best for detecting internal weld defects?

For internal defects (lack of fusion, porosity, slag inclusions), radiographic testing (RT) and ultrasonic testing (UT) are the primary methods. RT provides a permanent film record and detects volumetric defects well, but requires radiation safety precautions. UT is faster, more portable, detects planar defects (cracks) better than RT, and doesn't require evacuating the work area. For critical welds in piping, ASME B31.3 typically specifies RT or UT depending on the service class. Many modern projects use TOFD or phased array UT (PAUT) as RT replacements to eliminate radiation hazards while providing superior flaw detection and permanent digital records.

What is the difference between RT and UT for weld inspection?

RT (Radiographic Testing) passes X-rays or gamma rays through the weld to create an image on film or digital detector, and is best for volumetric defects (porosity, slag inclusions, incomplete penetration). UT (Ultrasonic Testing) sends sound waves through the material and analyzes reflections, and is better for planar defects (cracks, lack of fusion). RT provides a visual record that is intuitive to interpret but is slower and requires radiation safety zones. UT is faster, requires single-sided access, and can measure defect depth and height, but requires more operator skill and (in conventional form) does not produce an image. Many modern projects use TOFD or phased array UT as RT replacements, combining the advantages of both.

When is magnetic particle testing (MT) used instead of liquid penetrant testing (PT)?

MT is preferred for ferromagnetic materials (carbon steel, alloy steel, ferritic stainless); it detects both surface and near-surface defects, is faster than PT, and works through thin paint or coatings up to about 50 microns. PT is used for non-magnetic materials (austenitic stainless steel, duplex, aluminum, nickel alloys) where MT cannot work, and for detecting surface-breaking defects only. PT is also used when surface finish quality must be verified. Both methods are typically applied to welds after visual testing (VT). Rule of thumb: if the material is magnetic, use MT; if it is non-magnetic, use PT.

What NDT qualifications do inspectors need?

NDT personnel must be certified to ISO 9712 (international) or ASNT SNT-TC-1A / CP-189 (North American). Certification levels: Level I can perform tests under direct supervision; Level II can set up, perform, and interpret tests independently and accept/reject per written procedures; Level III can develop procedures, train others, interpret codes, and manage NDT programs. Most project specifications require Level II minimum for performing and interpreting inspections. Certification is method-specific, so an inspector must be qualified separately for each NDT method they use. Certifications have defined validity periods and require renewal through re-examination or continued activity documentation.

What percentage of welds must be examined per ASME B31.3?

ASME B31.3 examination percentages depend on fluid service category. Normal fluid service requires a minimum of 5% random RT or UT of butt welds, with 100% VT. Category M (high toxicity/lethal) and severe cyclic conditions require 100% RT or UT of butt welds plus 100% MT or PT of all other welds. High-pressure piping (Chapter IX) also requires 100% volumetric examination. Project specifications often exceed these minimums; for example, specifying 10-20% RT for standard process piping and 100% for high-temperature, sour service, or high-pressure classes. When a random examination reveals a defect, ASME B31.3 para. 341.3.4 requires progressive examination of additional welds by the same welder.

Leave a Comment

Have a question or feedback? Send us a message.

Your comment will be reviewed and may be published on this page.

Previous Comments

Yolandawakers

Our company uses the most common method for NDT, such as Visual and Optical Testing (VT).

Md istekhar aalam

Type of crack

Eli Richardson

It was awesome to read that ultrasonic testing is portable, which makes it useful for rural areas. One of my uncles requires a non-destructive testing service, but I'm not sure which method will work best for his project. I will help him find a professional for more options, so thanks for the description.

Projectmaterials

Thank you for your appreciated comment. Should you need further information on some specific topics, kindly send us an email to [email&#160;protected]. To submit an RFQ for piping materials, please visit this page: https://projectmaterials.com/submit-rfq-mto. Best regards, Projectmaterials

Sarah White

Very informative post! In my opinion, ultrasonic testing is one of the most fascinating NDT methods because it is so multifaceted. I actually just read another blog post about ultrasonic testing and it is very advanced but very accurate like you mention.

Scott M. Norwood

It was great to peruse that ultrasonic testing is compact, which makes it valuable for provincial zones. One of my uncles requires a non-ruinous testing administration, yet I don't know which strategy will work best for his undertaking. I will assist him with finding an expert for more alternatives, so a debt of gratitude is in order for the depiction.

srin

Good Info, Anyone has idea on DX inspection process?

PCTE Industrial

Very informative post about non-destructive testing types. Images and videos are very helpful to understand. Thanks for sharing this with us!

PCTE

The types of non-destructive testing expressed in the post is incredible. Thanks for sharing such valuable information!

kavin kumar

Informative information Thanks for sharing

inco special alloys

Nice Blog,also check Flange Supplier in Dubai , Flange Supplier in Abu Dhabi and Flange Supplier in Sharjah and visit our website Inco Special Alloys

Ananka Fasteners

Excellent Post, Also read our blogs on Fasteners Manufacturers,High Tensile Fasteners Manufacturer, Inconel Fasteners Manufacturer, Monel Fasteners Manufacturer, also visit Ananka Groups

PCTE

Great blog post on the different types of non-destructive testing (NDT)! The explanations were clear and concise, and the post provided a valuable overview of the applications and benefits of each NDT technique. Highly informative and educational!