ASME Section V Article 2 — Radiographic Examination

What Article 2 governs, and where its authority stops

Article 2 of ASME Boiler and Pressure Vessel Code Section V is the method article for radiographic examination. Its jurisdiction runs from T-210 through T-291 and covers how the exposure is set up, how image quality is proven, how the radiograph is identified and tied to a physical location on the part, and what has to be recorded afterwards. It applies to welds, to base material, and through its mandatory appendix on metallic castings, to cast components. It is written to be invoked by another Code section rather than to stand alone.

The boundary matters more than the content. Article 2 contains no acceptance criteria. It will insist that the radiograph carries adequate density and demonstrated sensitivity, but it will not tell you whether the elongated indication near the root is rejectable. It does not address radiation safety, source licensing, exclusion boundaries or dosimetry — those sit with the national regulator, in the United States the NRC or an Agreement State under 10 CFR Part 34, and elsewhere the equivalent competent authority. It does not certify the radiographer either; Article 1 paragraph T-120 routes that to SNT-TC-1A, CP-189 or a national equivalent, and paragraph T-150 requires the written procedure.

Most of the radiographic disputes we are asked to adjudicate as ASNT Level III consulting work turn out to be boundary disputes rather than technical ones. A fabricator applies Article 2 image quality rules to a job whose acceptance standard is AWS D1.1. A client rejects a radiograph on a density figure the referencing code never invoked. Name the referencing construction code and its edition on the technique sheet itself, and most of those arguments end before they start.

IQI selection: nominal single-wall thickness plus reinforcement

Article 2 recognises two IQI families: hole-type plaques to SE-1025 and wire-type sets to SE-747, both adopted from the corresponding ASTM standards. Table T-276 selects the designation, and the selection is made on nominal single-wall material thickness plus the estimated weld reinforcement. That phrasing does real work. It is not the measured thickness at the point where the technician happened to put the ultrasonic gauge, and in a double-wall exposure it is not the total thickness the beam passes through. Using measured thickness alone is the single most common IQI finding we see, and it always moves in the same direction — one designation too small, sensitivity quietly relaxed.

Placement is governed by T-277. The source side is the default. A film-side IQI is permitted where the source side is inaccessible, and it must then be identified by a lead letter F adjacent to or on the IQI, because a film-side IQI proves a different and easier geometry. Where the IQI cannot sit directly on the weld, it goes on a shim of radiographically similar material whose plan dimensions exceed the IQI by at least 1/8 in on all sides, and the shim plus IQI must be at least as thick as the weld including its reinforcement. For a single panoramic exposure of a complete circumference, three IQIs spaced approximately 120 degrees apart are required.

Two further placement errors recur. The first is an alloy mismatch: an IQI from a group in SE-1025 that is not radiographically similar to the item, typically a stainless plaque left on a carbon steel job. The second is a wire IQI laid so that the essential wire falls over the weld reinforcement rather than on the adjacent parent metal at the same effective thickness, which makes the wire easier to resolve than the code intends and overstates the sensitivity achieved.

Density, the reference point, and the -15/+30 band

Density under T-282 is not a general film quality statement. It is measured through the radiographic image of the body of the IQI and through the area of interest, and both must satisfy the minimum. For single film viewing that minimum is 1.8 with X-ray and 2.0 with a gamma source, with 4.0 the ceiling. For composite viewing of a double film pair, each film must reach 1.3 and the composite must stay under 4.0. Reading one film of a composite pair as though it were a single-film radiograph is a genuine technical error, not a paperwork one, because the pair was exposed on the assumption that both would be viewed together.

The variation rule in T-282.2 is misread more than any other number in the article. It states that the density through the body of the IQI must not vary from the density through the area of interest by more than minus 15 percent or plus 30 percent. It is a comparison between two specific points, not a uniformity requirement across the whole film. Where the band is exceeded, the article does not reject the radiograph outright; it requires an additional IQI for that exceptional area, which is a practical instruction that a lot of technique sheets never mention.

Instrument control sits behind all of this. Densitometers are calibrated at least every 90 days and step wedge comparison films verified at the stated interval. A lapse is not a trivial administrative miss: every density figure recorded since the last valid calibration becomes unverifiable, and on a campaign job that can be several hundred exposures. Where a client wants that exposure batch defended rather than reshot, an independent review of the inspection reports against the surviving films and the instrument history is usually the only route back.

Geometric unsharpness: the calculation people invert

Geometric unsharpness is calculated as Ug equals Fd divided by D. The definitions matter and are frequently swapped. F is the physical source size in the direction being considered — the focal spot for an X-ray set, the active pellet dimension for an isotope. The term d is the distance from the source side of the weld or object to the film. The term D is the distance from the source of radiation to the source side of the weld or object. It is not the source-to-film distance. Substituting source-to-film distance for D inflates the denominator and produces a number that is comfortably inside the limit while the real unsharpness is not.

The consequence is specific rather than abstract. Excess unsharpness smears fine, tightly closed indications — transverse cracks, lack of sidewall fusion at a steep bevel, and narrow root defects — until they read as background mottle. The radiograph will still have good density and a visible essential hole, so nothing in the routine record flags the problem. It is a defect-missed mechanism, and it survives review because every number on the report looks correct.

The limits in T-274.2 are 0.020 in for material under 2 in, 0.030 in from 2 through 3 in, 0.040 in over 3 through 4 in, and 0.070 in above 4 in, with the material thickness being the thickness on which the IQI selection was based. When access will not allow a longer source-to-object distance, the practical lever is F rather than D: a smaller focal spot, or a smaller physical source capsule such as Se-75 in place of a large Ir-192 pellet, buys back unsharpness without moving the crawler.

Technique selection, exposure count and location markers

Article 2 distinguishes single-wall viewing from double-wall viewing, and on pipe the distinction drives the exposure count. Where the double-wall double-image ellipse technique is used on pipe of 3.5 in nominal size or less, a minimum of two exposures approximately 90 degrees apart is required. Where the source and film sides are superimposed rather than offset, a minimum of three exposures approximately 60 or 120 degrees apart applies. A single panoramic exposure is only valid where an internal source position genuinely gives full circumferential coverage with the IQI requirements met.

Surface condition is a prerequisite, not a courtesy. Under T-222 surface irregularities must be removed to the degree that their images cannot mask or be confused with a discontinuity, on the inside surface as well where it is accessible. Weld ripple that would previously have been argued about becomes an easy rejection when a client's reviewer sees a linear shadow running along a cap.

Location markers are where paperwork failures become permanent. The markers must appear on the radiograph, and they must be placed on the part, not on the exposure holder or cassette, so that the position of each radiograph can be re-established on the component. A film marked from the cassette cannot be tied back to a physical location once the vessel is painted and shipped. Combined with the identification system of T-224 — permanent identification traceable to the contract, component, weld seam and part — this is what makes a radiograph an auditable record rather than a picture.

Where the acceptance criteria actually come from

Section VIII Division 1 sets both the extent and the acceptance criteria for radiography on unfired pressure vessels: UW-11 decides whether full, spot or no radiography applies to a given joint and joint efficiency, UW-51 gives the acceptance criteria for full radiography, and UW-52 governs spot radiography including the two-additional-spot retest sequence when a spot fails. Section I refers to PW-51 for power boilers. Section III routes nuclear class 1 components through NB-5320. ASME B31.3 covers process piping in paragraph 344.5 for the examination itself and Table 341.3.2 for the acceptance criteria, which differ by fluid service.

The most expensive trap is the one that lies outside ASME entirely. AWS D1.1 does not invoke Section V. A structural steel job radiographed by a crew working from a Section V technique sheet will produce films that are perfectly good and a documentation package that does not meet the contract, because D1.1 has its own radiographic requirements and its own acceptance criteria. Fabrication shops that run both pressure and structural work need two procedure sets, not one procedure with a note.

Interpretation is where the criteria meet the film, and interpreters need to be certified for the method and demonstrably current on the acceptance standard they are applying. Where a shop is bringing radiographic interpretation in-house, or lifting technicians toward interpretation authority, the qualification route runs through documented training and examination under the employer's written practice, which is what our NDT training and certification programmes are built to support.

Digital radiography under the mandatory appendices

The mandatory appendices to Article 2 extend it beyond film. Phosphor imaging plate systems, digital detector array systems and real-time radioscopy each have their own appendix, and each replaces the film density criterion with a different pair of controls: basic spatial resolution, measured with a duplex wire gauge to SE-2002, and signal-to-noise ratio. Limits on scan spot size, pixel size and image unsharpness follow from those. The overarching requirement is that the digital technique be demonstrated to give image quality at least equivalent to the film technique it replaces, on representative material and thickness — not asserted from a vendor datasheet.

Digital brings a records problem that film never had. A radiograph is its own evidence; a digital image is a file that can be windowed, filtered, sharpened and re-saved. The appendices therefore require that the image and its acquisition parameters be stored, and that any processing applied for interpretation be recorded. An image submitted for review with no visible IQI, no acquisition record and an unexplained processing history is not a radiograph in the code sense, whatever it looks like on screen.

Retention is where most digital programmes fall over three years in, when the original workstation is gone and the proprietary viewer no longer runs. Storing images and their metadata in a controlled, method-aware repository rather than on project drives is the difference between a defensible record and an archaeology exercise, and it is one of the practical reasons operators move radiographic records into an inspection data management system alongside the rest of the examination history.

Findings that recur on Article 2 records

A short list accounts for most of what auditors write up. Densitometer calibration lapsed mid-campaign. A film-side IQI used without the lead letter F. Density recorded on base metal rather than through the IQI body. Ug calculated with source-to-film distance as D. IQI selected on measured thickness rather than nominal single-wall thickness plus reinforcement. Location markers placed on the cassette so the film cannot be re-established on the part. Step wedge comparison film never verified. Interpreters whose annual vision examination expired part way through the job.

Two findings are structural rather than clerical. The first is a technique sheet that never names the referencing construction code and its edition, which leaves the acceptance basis undefined and makes every subsequent argument unwinnable. The second is a report that states the acceptance standard as Section V, which as this page has laboured is not an acceptance standard at all. Both are trivially fixed before the job and painful to fix afterwards, because correcting an acceptance basis retrospectively invites a review of every disposition already made.

The defensible position is unglamorous: one controlled procedure per referencing code, technique sheets that record the actual essential parameters used on each shot, calibration records that never gap, and an interpretation record that names the standard, the clause and the disposition. That package survives a client audit. Nothing assembled after a finding ever does, and reconstructing records after the fact is a far worse exposure than the original gap.

Frequently asked

Is a written and demonstrated procedure required for radiography under Section V?

Article 1 paragraph T-150 requires that examinations be performed to a written procedure where the referencing Code Section calls for one, and that the procedure be demonstrated to the satisfaction of the Inspector. Article 2 does not carry an essential-variable table of the kind Article 4 provides for ultrasonics, so the control on radiography sits in the technique documentation required by Article 2 together with the referencing code's procedure requirement.

What is the lead letter B for, and when does it reject a radiograph?

A lead letter B at least 1/16 in thick and 1/2 in high is attached to the back of each cassette to test for backscatter. If a light image of the B appears on a darker background of the radiograph, backscatter has fogged the film and the radiograph is unacceptable. If the B appears dark on a lighter background, that is not cause for rejection. The test is often run once per setup and then quietly dropped when the geometry changes, which defeats its purpose.

Can a film-side IQI be used instead of a source-side IQI?

Yes, where the source side is inaccessible, but it must be identified with a lead letter F adjacent to or on the IQI so that anyone reviewing the film knows which geometry was proven. Source-side placement demonstrates sensitivity through the full path the beam actually took; film-side placement demonstrates less. Using a film-side IQI on an accessible joint, or omitting the F, is a straightforward finding.

Does ASME Section V Article 2 apply to AWS D1.1 structural welding?

Not unless the contract says so. AWS D1.1 contains its own radiographic requirements and its own acceptance criteria and does not invoke Section V. A shop that runs both pressure equipment and structural steel needs distinct procedures for each; applying an ASME technique sheet to a D1.1 job produces films that are technically sound and a record package that does not meet the contract.

How long must radiographs and radiographic records be retained?

Article 2 does not set a retention period. Retention is set by the referencing construction code, the jurisdiction, and the purchase order, and for pressure equipment it is commonly tied to the life of the asset or to a stated number of years after certification. Agree the retention obligation in the purchase order before the first exposure, because film and digital archives are the first thing to disappear when a fabricator's project team disbands.

Who decides whether an indication seen on the radiograph is rejectable?

The certified interpreter applies the acceptance criteria of the referencing construction code, and the disposition is recorded on the interpretation record with the clause cited. Section V gives the interpreter a valid image; it gives no basis for accepting or rejecting what is in it. Where a client and a fabricator disagree on a disposition, the resolution is an independent Level III review of the film against the named clause, not a renegotiation of the image quality.