ASME Section V Article 4 — Ultrasonic Examination of Welds
Ultrasonic examination (UT) of welds under ASME BPVC — covers basic calibration block, DAC, scanning, IIW V1/V2 reference, and acceptance flagged at 50% DAC.
Scope
Article 4 of ASME Section V governs ultrasonic examination (UT) of welds in ferritic materials, primarily for pressure-retaining welds in Section VIII Div 1, Section I, and B31 piping. The article covers conventional pulse-echo UT with single-element transducers in shear-wave and longitudinal-wave modes. Phased-array UT (PAUT) and time-of-flight diffraction (TOFD) are addressed by Article 4 with the Mandatory Appendix V (PAUT) and Mandatory Appendix N (TOFD) — both adopted in recent editions and increasingly used as alternatives to RT. Article 4 requires a basic calibration block of the same material specification, product form, heat treatment, and approximate thickness as the production weld. It establishes the distance-amplitude correction (DAC) curve from the side-drilled hole (SDH) of the basic calibration block, scanning techniques (raster, line, zone), and acceptance criteria keyed to percentage of DAC (typically 50% DAC threshold for recording).
Code compliance is only demonstrable if the evidence behind it is: the procedure revision in force, the inspector's certification state and the instrument's calibration status at the time of test. Atlantis NDT provides ASNT Level III consulting for procedure and written-practice work against this code, training toward the certifications that reference it, and inspection management software that keeps that evidence recoverable years later. Request a consultation.
How a standard like this is applied in an inspection programme
A standard is only half of the requirement. It defines how an examination is performed and, in some cases, how results are classified — but the acceptance criteria that decide whether a component stays in service normally come from the construction or in-service code governing the item, not from the examination standard itself. Confusing the two is one of the more common findings in a procedure review: a procedure that correctly cites the examination standard but applies acceptance criteria from the wrong code or the wrong edition.
What has to be in place for compliance to be demonstrable
A written procedure qualified against this standard for the specific materials, thickness ranges and geometries in scope — not a generic procedure covering everything
Personnel certified for the method and level under ASNT SNT-TC-1A, ANSI/ASNT CP-189, NAS 410 or ISO 9712, current on the date the examination was performed
Equipment, probes and reference standards in calibration on that date, with traceability to a national standard under ISO 17025
The applicable edition of the standard recorded against the examination, so historical work stays assessed under the edition then in force
Technique sheets under the same revision control as the procedure above them — the most frequently uncontrolled document in an otherwise compliant quality system
Edition changes
When a new edition is issued, new work moves to it from a defined effective date that you set and record; work already performed stays assessed under the edition in force at the time. Retrospectively applying a new edition to historical dispositions invalidates the original acceptance decision and creates a substantially larger problem than the one being solved.
Where this usually goes wrong
Not in the technical content, but in reconstruction. An auditor picks an issued report and asks which procedure revision applied, who performed the work and whether they were qualified on that date, and whether the instrument and reference blocks were in calibration. Programmes that hold only current state can answer none of those. Binding the document revision, the qualification state and the calibration state to each inspection record as it is created turns that from an investigation into a lookup.
ASME Section V Article 4 governs ultrasonic examination of welds and the adjacent base metal. It specifies equipment, basic calibration blocks, DAC construction from side-drilled holes at 1/4T, 1/2T and 3/4T, scanning at 6 dB above reference, and 10% transducer overlap. It contains no acceptance criteria — those come from the referencing Code Section.
Article 4 sits inside a chain of authority that trips people up. Article 1 supplies the general rules — written procedures, personnel qualification, the Inspector's role. Article 4 supplies the method: instrument linearity, block geometry, calibration, scan plan, recording. The referencing Code Section — Section VIII Division 1 Mandatory Appendix 12, Section III NB-5330, B31.3 paragraph 344.6 — supplies the acceptance criteria and decides whether UT is even permitted in place of radiography. Reading Article 4 alone will tell you how to run the examination and nothing about whether the result passes. The requirements that actually bite in practice are the basic calibration block being of the same material specification, product form and heat treatment as the part, the 25°F temperature differential limit between block and examination surface, calibration checks at least every four hours, and the straight-beam base metal scan for laminar reflectors before any angle-beam work begins.
Source: Based on ASME BPVC Section V, Article 1 (General Requirements) and Article 4 (Ultrasonic Examination Methods for Welds) including the mandatory appendices covering time of flight diffraction and phased array; ASME BPVC Section VIII Division 1, Mandatory Appendix 12 (Ultrasonic Examination of Welds); ASME BPVC Section III article 5000 acceptance paragraphs; ASME B31.3 paragraph 344.6 and Table 341.3.2; ASNT SNT-TC-1A and ANSI/ASNT CP-189.
Article 4 controls that produce audit findings when they are missed
Control
What Article 4 requires
Numeric limit
Failure mode when missed
Scanning sensitivity
Scan above reference level, evaluate at reference level
At least 6 dB (2x) over reference
Indications near reference are never seen during the sweep; coverage failure disguised as a clean report
Scan overlap
Each pass overlaps the previous pass
At least 10% of the transducer dimension
Unscanned lanes between passes; coverage cannot be demonstrated from the record
Scanning speed
Limited unless the system is calibrated at the higher rate
6 in./s (152 mm/s)
Pulse repetition and display update outrun the operator; small reflectors pass unseen
Block and part temperature
Calibration block and examination surface matched
Within 25°F (14°C)
Velocity and attenuation shift; the DAC no longer represents the part
Calibration check interval
Check at start, finish, examiner change, system change and during work
At least every 4 hours
Drift discovered late; all work back to the last valid check must be repeated
Amplitude drift
Recalibrate and re-examine when the reference response falls
More than 20% or 2 dB
Data retained that was recorded outside a valid calibration
Sweep drift
Correct the range and re-examine recorded areas
More than 10% of the reading or 5% of full sweep
Depths and flaw positions mis-reported in the examination record
Base metal scan
Straight beam over the angle beam travel path before scanning
At reference sensitivity
Laminations unfound; the beam is blocked and the volume behind was never examined
Instrument linearity
Screen height and amplitude control linearity verified
3 months and 12 months respectively
Verification records unavailable at audit; the whole campaign's data is challenged
What Article 4 covers, and the three things it does not
ASME Boiler and Pressure Vessel Code Section V, Article 4 is titled Ultrasonic Examination Methods for Welds. Its subject is the weld and the base metal adjacent to it, including every volume the sound beam must travel through to reach the weld. The article tells you what the instrument must be capable of, what block you calibrate on, how the reference sensitivity is established, how the scan is executed, what must be recorded and what the report must contain. Article 1 sits above it and supplies the general requirements — written procedures, personnel qualification, the Inspector's role — and applies automatically whenever Article 4 is invoked.
Three things Article 4 deliberately does not do. It does not give acceptance criteria. It does not cover ultrasonic thickness measurement or the examination of plate, forgings, castings and bolting, which belong to Article 5. And it does not decide whether ultrasonic examination may be used in place of radiography — the construction code makes that call, as Section VIII Division 1 does when it permits UT in lieu of RT for certain welds under stated conditions of procedure qualification, demonstration and documentation.
That scope boundary is the most common misunderstanding on the shop floor and the fastest way to fail a procedure review. A UT procedure that names Article 4 as its acceptance basis is defective on its face, because Article 4 contains no accept or reject thresholds to name. The procedure must cite the referencing Code Section and the specific paragraph or appendix that carries the criteria. This is the first thing we check in a procedure review during ASNT Level III consulting engagements, and it is still the finding raised most often.
The basic calibration block is the examination
Article 4 builds the entire sensitivity reference on a physical block, so the block controls the result more than any setting on the instrument. The block must be of the same material specification, product form and heat treatment condition as the material being examined, or acoustically equivalent to it. A normalised and tempered block does not represent a quenched and tempered part. A rolled plate block does not represent a forging with a different grain orientation. Where the part is clad, the block must carry the same cladding, applied by the same process.
Block thickness is tied to the weld thickness by a table, and one block covers a band of thicknesses rather than a single value, which is why a small set of blocks can serve a whole fabrication shop. The reflectors are side-drilled holes placed at one quarter, one half and three quarters of the block thickness, with surface notches for the surface-connected response, and hole diameter increases with block thickness. The block surface finish should represent the part; a machined block used to calibrate for a scan across an as-rolled or blasted surface overstates sensitivity.
Temperature is the requirement people record last and get caught on first. The differential between the basic calibration block and the examination surface must be within 25°F (14°C). Sound velocity and attenuation both move with temperature, so a block sitting in an air-conditioned calibration room and a vessel shell at 100°F in a yard are not the same acoustic system. Where surface condition, curvature or attenuation differ between block and part, a transfer correction keeps the reference honest; skipping it produces a distance amplitude curve that flatters the component.
Reference level, DAC, and the drift rules that force re-examination
The distance amplitude correction curve is constructed by maximising the response from each side-drilled hole at its metal path and joining those points. That curve is the reference level. Article 4 then separates two activities that field crews routinely merge: scanning is performed at a gain at least 6 dB — twice the amplitude — above the reference level, and evaluation is performed with respect to the reference level. Scanning at reference is not conservative housekeeping. It is a coverage failure, because indications that would evaluate near reference are never seen during the sweep at all.
Calibration must be performed with the complete examination system as it will actually be used: the same search unit, the same wedge, the same cable and cable length, the same couplant, the same instrument settings. Swapping a six foot cable for a twenty five foot cable after calibration is a system change, not a convenience. The couplant used in the examination must be the couplant used in calibration, and for austenitic stainless steel, titanium and nickel alloys the halide and sulfur limits on couplant constituents apply and have to be evidenced by the couplant certificate.
The check schedule and the drift rules are where records either hold or collapse. A calibration check is required at the start and finish of an examination or series, when the examiner changes, when any part of the system changes, and at least every four hours during work. If a reference reading has dropped by more than 20% or 2 dB, the system is recalibrated and all areas examined since the last valid check are re-examined. If a DAC point has moved on the sweep by more than 10% of the reading or 5% of full sweep, the range is corrected and the recorded areas re-examined. Retaining data recorded outside a valid calibration is the finding that most often forces a re-shoot, and it is exactly the pattern an independent report review is designed to catch.
Scan plan, coverage and the base metal scan everyone forgets
Before any angle beam work, the base metal through which the angle beam will travel is scanned with a straight beam at reference sensitivity to find laminar reflectors. This is a requirement, not a courtesy. A lamination in the scan path steers or blocks the beam, and the volume behind it was never examined even though the coverage record says it was. Crews under schedule pressure skip this scan more often than any other single step in the article, and it leaves no trace at all unless somebody asks for the record.
Angle beam scanning of the weld is performed from both directions perpendicular to the weld axis where access allows, with additional scanning for reflectors oriented parallel to the weld where the referencing code or the procedure requires it. Each pass overlaps the previous by at least 10% of the transducer dimension perpendicular to the scan direction. Scanning speed is limited to 6 in./s unless the system has been calibrated at the higher rate, because pulse repetition frequency and display update rate set a real physical limit on how fast a small reflector can be caught and displayed.
Where geometry, nozzle reinforcement, attachments or an unground weld crown prevent full coverage, the limitation must be recorded in the report and evaluated against the referencing code, not quietly absorbed. Auditors compare the scan plan to the physical component and to the coverage map. If the plan claims a scan surface that a nozzle or a support blocks, the whole data set becomes questionable. Recording the restriction protects both the fabricator and the examiner; concealing it converts an understandable technical limitation into a documentation failure with a much longer tail.
Where the acceptance criteria actually live
For Section VIII Division 1 vessels examined ultrasonically, the criteria sit in the mandatory appendix on ultrasonic examination of welds. It requires indications above 20% of DAC to be investigated to determine shape, identity and location, rejects cracks, lack of fusion and incomplete penetration regardless of amplitude, and rejects other imperfections whose indications exceed the reference level and whose length exceeds a thickness-dependent limit: one quarter inch for thickness up to three quarters of an inch, one third of the thickness between three quarters and two and a quarter inches, and three quarters of an inch above that.
Other referencing codes carry their own. Section III applies its NB, NC and ND article 5000 acceptance paragraphs. ASME B31.3 invokes Article 4 through its examination paragraph and then applies its own weld imperfection table for the applicable fluid service. Section VIII Division 2 permits a fracture mechanics based acceptance route supported by the flaw sizing and categorisation provisions of Article 4. Client supplementary specifications frequently tighten all of these, sometimes without saying which clause of the base code they replace.
The practical consequence is a reporting rule. Every UT report must name the acceptance standard by document, edition and paragraph, alongside the procedure number and revision. When a report says only that a weld was examined to ASME V Article 4 and accepted, there is no defensible statement of what accepted means. Keeping the procedure, revision, acceptance standard and examiner certification linked to each report is precisely the record-keeping problem that inspection management software exists to solve, and it is the difference between an audit that takes an hour and one that takes three days.
Essential variables, procedure qualification and demonstration
Article 4 lists the essential and nonessential variables for an ultrasonic procedure in a table. Essential variables include the weld configurations and thickness range covered, the surface condition, the couplant, the search unit types, frequencies and angles, the wedge, the scanning technique including manual versus automated, the directions and extent of scanning, the method of demonstrating coverage, and the computer software where used. A change to any essential variable requires the procedure to be revised and requalified. A change to a nonessential variable requires only that the procedure be revised.
Where the referencing code requires demonstration, the procedure is proven on a qualification block containing the flaw types of interest at the relevant depths, and the demonstration record becomes part of the procedure package rather than a loose attachment. Encoded and automated techniques carry a heavier burden, because coverage and positional accuracy have to be shown rather than asserted. Keeping the qualification records physically with the controlled procedure — instead of in a personal folder on somebody's laptop — is what makes the package survivable when the person who ran the demonstration has left.
Personnel qualification runs on a parallel track. The employer's written practice, built on SNT-TC-1A, defines training hours, experience, examinations and the certification levels for each method. A UT Level II must be qualified to the specific procedure they are using, not merely certified in the method in general. Where an organisation needs its examiners brought up to a defensible standard, structured NDT training and certification support and a written practice that matches what the examiners actually do in the field are two halves of the same problem.
Encoded techniques: TOFD and phased array under Article 4
Time of flight diffraction is addressed by a mandatory appendix to Article 4. It measures the arrival time of signals diffracted from flaw tips rather than the amplitude reflected from specular surfaces, which makes it strong on through-wall sizing and weak in the near surface dead zone beneath the lateral wave. That dead zone is a genuine limitation, and a TOFD-only scan on a thin section without a complementary pulse-echo or phased array scan will not see near surface flaws. The procedure has to state how the near surface volume is covered.
Phased array is addressed by further mandatory appendices covering linear scanning with electronic and sectorial scans and raster examination techniques. The essential variable list expands to cover the array, aperture, element count, focal laws, wedge, and the sensitivity and wedge delay calibration approach. Time corrected gain replaces the manually plotted curve in many setups, but the underlying obligation is unchanged: a demonstrated reference sensitivity, a demonstrated coverage of the required volume, and a check schedule that catches drift before a shift's worth of data is compromised.
Encoded data is evidence. The raw files, the scan plan, the calibration files and the instrument setup are what let a third party reconstruct months later what the examiner actually saw. Contracts increasingly require the native data rather than a PDF of screenshots, and clients are right to insist on it. If the native files cannot be produced on request, the examination is effectively unverifiable no matter how competent the technique was on the day it was performed.
The findings that recur, and how to close them before the audit
The same handful of findings appear on Article 4 audits across fabricators and service companies alike: a basic calibration block whose heat treatment does not match the part; no record of block and part temperatures; missing four hour calibration checks on a shift that ran ten hours; scanning performed at reference level; no record of the straight beam base metal scan; a report citing Article 4 as its acceptance standard; a procedure never requalified after the wedge or angle changed; and instrument linearity verifications past their interval.
None of these are exotic. Every one is closed by a document, a record or a habit, and every one is far cheaper to fix before an auditor arrives than afterwards. A Level III review reads the procedure against the essential variable table, checks the block certification against the material actually being examined, samples completed reports against the raw data behind them, and looks at the calibration log for the intervals rather than for the presence of entries. Auditors do the same thing, with less patience.
The output of that review should be a defect list with owners and dates, not an opinion. Where procedures need rewriting, qualification records need rebuilding, or a written practice needs to be brought into line with what examiners actually do, that is technical authority work rather than an audit exercise, and a short scoping conversation through contact is usually enough to size it before anybody commits to a programme.
Does ASME Section V Article 4 contain weld acceptance criteria?
No. Article 4 specifies how the ultrasonic examination is performed — equipment, calibration blocks, reference sensitivity, scanning and recording — but carries no accept or reject thresholds. The criteria come from the referencing Code Section: Section VIII Division 1 through its ultrasonic examination appendix, Section III through its 5000 series acceptance paragraphs, B31.3 through its weld imperfection table. A procedure citing Article 4 as its acceptance basis is incomplete and will be written up.
What must the basic calibration block be made of?
The same material specification, product form and heat treatment condition as the material being examined, or a material acoustically equivalent to it. Where the component is clad, the block carries the same cladding applied by the same process. Reflectors are side-drilled holes at one quarter, one half and three quarters of the block thickness, with surface notches. Block thickness is selected from a table covering a band of weld thicknesses, so a small set of blocks serves a fabrication shop.
How often must the ultrasonic calibration be checked during an examination?
A calibration check is required at the start and finish of each examination or series, whenever the examiner changes, whenever any part of the examination system changes, and at least every four hours during examination. If a reference response has fallen by more than 20% or 2 dB, the system is recalibrated and everything examined since the last valid check is re-examined. Sweep movement beyond 10% of the reading or 5% of full sweep triggers the same re-examination of recorded areas.
What scanning gain and overlap does Article 4 require?
Scanning is performed at a gain at least 6 dB — twice the amplitude — above the reference level, while evaluation is performed with respect to the reference level itself. Each scanning pass overlaps the previous by at least 10% of the transducer dimension perpendicular to the scan direction. Scanning speed is limited to 6 in./s, roughly 152 mm/s, unless the system has been calibrated at the higher speed. Scanning at reference level is a coverage failure, not conservatism.
When does a UT procedure have to be requalified?
When any essential variable changes. Article 4 tabulates them: weld configurations and the thickness range covered, surface condition, couplant, search unit type, frequency and angle, wedge, scanning technique including manual versus automated, the directions and extent of scanning, the method used to demonstrate coverage, and computer software where applicable. Changing a nonessential variable requires only a procedure revision. Changing a wedge angle and continuing under the old qualification is among the most common procedure findings.
Can ultrasonic examination be used in place of radiography?
Only where the construction code allows it. Section VIII Division 1 permits ultrasonic examination in place of radiography for certain welds subject to stated conditions covering procedure qualification, demonstration, personnel and documentation, with design and marking consequences that follow from the choice. Article 4 defines how the ultrasonic examination is performed but has no authority to authorise the substitution. That decision, its conditions and its record sit with the referencing Code Section and the user's design specification.
Frequently asked
What is the difference between ASME Section V Article 4 and Article 5?
Article 4 covers ultrasonic examination of welds and the adjacent base metal. Article 5 covers ultrasonic examination methods for materials and fabrication — plate, pipe and tubular products, forgings, castings, bolting and thickness measurement. A single job often needs both: Article 5 for the plate, Article 4 for the seam welded from it. Citing the wrong article in a procedure is a straightforward and avoidable finding.
What temperature difference is allowed between the calibration block and the part?
The differential between the basic calibration block and the examination surface must be within 25°F (14°C). Beyond that, velocity and attenuation differences mean the block is no longer a valid acoustic model of the part and the DAC misrepresents the true response. Record both temperatures on the calibration record; an auditor who finds the requirement stated in the procedure but never evidenced will treat it as not done.
Does Article 4 cover phased array and TOFD?
Yes, through mandatory appendices — time of flight diffraction in its own appendix, and phased array through appendices covering linear electronic and sectorial scanning and raster techniques. Each brings additional essential variables such as array, aperture, focal laws, wedge and the time corrected gain approach, and a heavier demonstration burden. A TOFD-only scan must state how the near surface dead zone beneath the lateral wave is covered.
What has to appear on a compliant Article 4 ultrasonic report?
Procedure number and revision; the acceptance standard by document, edition and paragraph; examiner name and certification level; instrument, search unit and wedge identification with calibration status; calibration block identification; calibration and check times; couplant; surface condition; the extent of scanning achieved including any coverage limitation; and every recorded indication with its location, amplitude relative to DAC and length.
Can an outside Level III write and qualify our UT procedures?
Yes. Procedure development and qualification is a standard outside Level III service, and the employer retains the resulting procedure as its own controlled document. What matters at audit is that the procedure is qualified, controlled at a revision, matched to the revision cited on the reports, and that the examiners using it are certified under the employer's written practice to use that specific procedure.