What ASME Section V Article 2 Actually Requires on a Radiograph
ASME Section V Article 2 governs how a radiograph is produced and judged, not whether the weld passes. It fixes written-procedure content, IQI selection and placement, geometric unsharpness limits, film density between 1.8 for X-ray or 2.0 for gamma and 4.0 maximum, density variation of minus 15 to plus 30 percent, location markers, backscatter checks and technique records. Acceptance criteria come from the referencing construction code.
Article 2 is written in the order a radiographer works, and that order is worth respecting. Surface condition and identification come first, then geometry: source size and the two distances that set geometric unsharpness. Only then does IQI selection enter, driven by radiographic thickness rather than nominal wall. Placement rules follow, then the technique choice that decides how many exposures a circumference needs. Density and IQI sensitivity are the last gates before an image may be interpreted at all. Nothing in Article 2 tells you whether a slag line is acceptable; that judgment belongs to ASME Section VIII, B31.1, B31.3, AWS D1.1 or API 1104. Crews that treat Article 2 as the shooting standard and the construction code as the acceptance standard rarely lose film. Crews that blur the two argue with the Authorized Inspector about the wrong document.
Source: ASME BPVC Section V, Article 2, Radiographic Examination — 2023 Edition, paragraphs T-210 through T-292, including Table T-221 (procedure variables), the T-274.2 geometric unsharpness table and the T-276 IQI selection table, read alongside Article 1 general requirements. Section V is reissued on a two-year cycle and the referencing construction code fixes the edition that governs your job; confirm every number against that edition before it enters a procedure. Where a requirement is described here without a figure, the figure was not verifiable at the time of writing and must be read out of the Article itself.
| Clause | What it governs | Requirement as written | How crews lose the film |
|---|---|---|---|
| T-221 | Written procedure | Procedure content plus the split of essential and nonessential variables; an essential-variable change requires requalification by demonstration | Field shot uses a different film class or screen than the qualified procedure, and nobody revises the procedure |
| T-222 | Surface preparation | Weld ripples and surface irregularities conditioned so their images cannot mask or be confused with a discontinuity | Undressed weld toe casts a line that gets called as lack of fusion, or hides one |
| T-274 | Geometric unsharpness | Ug = Fd/D, limited to 0.020 in. under 2 in. material, 0.030 in. from 2 through 3 in., 0.040 in. over 3 through 4 in., 0.070 in. over 4 in. | Source-to-object distance shortened in the field to save exposure time; unsharpness never recalculated |
| T-276 | IQI selection | IQI chosen on nominal single-wall thickness plus the estimated weld reinforcement; backing rings and strips excluded from that thickness | Reinforcement ignored, IQI one row too light, essential hole invisible |
| T-277 | IQI placement | Source side by default; film side marked with a lead F; separate blocks and shims sized so at least three edges of the IQI image show | Shim cut to the IQI outline, edges lost, placement unprovable |
| T-282 | Density | 1.8 minimum X-ray, 2.0 minimum gamma, 4.0 maximum, 1.3 minimum per film for composite viewing; area of interest within minus 15 to plus 30 percent of IQI density | Density read on convenient parent metal instead of through the IQI body |
| T-271 | Technique and coverage | Single wall by default; double-wall viewing limited to 3 1/2 in. nominal outside diameter and to a source-side IQI, with minimum exposure counts per geometry | Two elliptical shots fired on a joint that required three exposures |
| T-223 / T-284 | Backscatter | A lead B at least 1/2 in. high and 1/16 in. thick on the back of every film holder; a light B on a darker background rejects the radiograph | B omitted from holders, so backscatter can never be ruled out |
Article 2 governs the image, not the weld
Article 2 runs from T-210 through T-292 and every paragraph in it answers one question: is this radiograph good enough to interpret? Surface condition, identification, geometry, IQI sensitivity, density, backscatter and blemish freedom are all gates on the image. None of them is a gate on the weld. The Article contains no length limit for slag, no diameter limit for porosity and no rule about cracks, because those numbers belong to the construction code that invoked Section V in the first place. Get that separation right and most Article 2 disputes disappear before they start.
The practical consequence shows up at the viewing bench. A radiograph that meets Article 2 and shows a rejectable indication is a valid radiograph with a rejectable weld. A radiograph that shows nothing but falls outside the density band is not evidence of a sound weld; it is not evidence of anything, and it has to be reshot. Interpreters who call indications before confirming density and IQI sensitivity are working in the wrong order, and an Authorized Inspector who spots that order reversed will question every film in the package.
Section V pairs Article 2 with the rest of the method articles under a common general requirements article, and the structure repeats across methods. If you already work with Article 4 for ultrasonics, Article 6 for liquid penetrant or Article 7 for magnetic particle, Article 2 will feel familiar: procedure, essential variables, calibration, technique, evaluation, documentation, in that order. The Section V compliance overview maps how the articles sit against each other.
The written procedure and its essential variables
Radiographic examination is performed to a written procedure, and Table T-221 splits the procedure content into essential and nonessential variables. The mechanism is the same one used throughout Section V. Change an essential variable and the procedure has to be requalified by demonstration before it can be used again. Change a nonessential variable and the procedure needs a revision or an addendum, but no requalification. Either way the change is written down. A procedure that quietly drifts from what the crew actually does is the finding auditors write most often.
The variables that carry the most weight are the ones that move image quality: material type and thickness range, the source, its physical size, the film class, the screens, the processing route, and the two geometric distances. Those are the inputs that decide unsharpness, contrast and density. A single value or a range of values has to be stated for each one. As required is not a value, and a procedure written that way cannot be demonstrated against anything, because there is nothing to demonstrate against.
Demonstration, where the referencing code calls for it, means producing a radiograph under the worst-case conditions the procedure permits and showing that it meets every requirement of the Article. That is the point at which a paper procedure becomes a qualified procedure. If your procedures were written once and never revisited against the crews' actual field practice, a technical procedure development review costs less than the nonconformance it prevents.
Geometry first: source size, distance and unsharpness
Geometric unsharpness is the penumbra the source projects around every edge in the image. Article 2 defines it as the source size multiplied by the distance from the source side of the object to the film, divided by the source-to-object distance measured to the source side of the object. Three inputs, one number. The source size is the maximum projected dimension of the radiating source or focal spot, which is why an isotope camera and an X-ray tube of nominally similar output do not behave alike on the same joint.
The limits are tabled against material thickness: 0.020 in. for material under 2 in., 0.030 in. for 2 through 3 in., 0.040 in. over 3 through 4 in., and 0.070 in. above 4 in. The thickness that picks the row is the thickness on which the IQI is based, not the total path length the beam travels through the part. Getting that wrong on a double-wall shot moves you a row in the wrong direction and quietly widens the tolerance you are working to.
Unsharpness is where field improvisation does the most damage. Crews under exposure-time pressure shorten the source-to-object distance, which raises the ratio directly and immediately. Nothing on the film announces this. The only defense is that the technique record carries the source size and both distances, so a reviewer can recompute the number months later. That reconstructability is the same property that makes a radiographic report defensible when a joint is challenged years after handover.
IQI selection: radiographic thickness, not nominal wall
Article 2 recognizes two IQI designs, the hole type and the wire type, and both are selected from a table keyed to material thickness. For welds, that thickness is the nominal single-wall thickness plus the estimated weld reinforcement permitted by the referencing code. It is not the nominal pipe wall on the drawing, and it is not the caliper reading on the parent plate away from the joint. Reinforcement is part of what the beam crosses, so it is part of what selects the IQI.
Backing rings and backing strips are excluded from the thickness used for IQI selection. This trips up crews on socket and backing-ring joints constantly, because the ring is plainly in the beam and plainly adds attenuation. It still does not enter the selection. The rule exists so that IQI sensitivity is referenced to the weld being judged rather than to a consumable that is not part of the pressure boundary being qualified.
The selection table gives a source-side designation and a separate film-side designation for the same thickness row. The film-side value is lighter, because the IQI sits further from the source and closer to the film and therefore experiences less of the radiographic path. Read the row. Subtracting one designation from the source-side value is a habit rather than a rule, and on some thickness ranges that habit produces the wrong IQI and an avoidable reshoot.
Placement: source side, film side, blocks and shims
The default is source-side placement, with the IQI on the part, on the source side, so that the IQI experiences the same radiographic path as the weld. Film-side placement is permitted only where inaccessibility prevents hand-placing the IQI on the source side, and then a lead letter F goes adjacent to or on the IQI so the interpreter knows which sensitivity applies. An unmarked film-side IQI is a technique error that invalidates the sensitivity claim, not a paperwork slip to be corrected later.
Where the IQI cannot go on the part at all, it may go on a separate block of similar material, provided the IQI is no closer to the film than the source side of the part, the block sits as close to the part as practical, and at least three edges of the IQI image are visible on the radiograph. That last condition is the proof. Three visible edges show the interpreter that the IQI sat where the technique sheet says it sat, and that nothing was cropped out of frame.
Hole IQIs on welds usually need a shim. The IQI sits beside the weld on parent metal, sees less thickness than the weld crown, and would otherwise read a higher density than the area of interest it is supposed to represent. A shim of radiographically similar material brings it up to the weld-plus-reinforcement thickness. The shim has to be larger than the IQI so that at least three sides of the IQI outline remain visible in the radiograph.
Technique and coverage: how many exposures a joint needs
Single-wall technique is the default: the radiation passes through one wall and that wall is interpreted. Everything else exists because access does not allow it. Double-wall technique divides into single-wall viewing, where the beam passes through two walls but only the film-side weld is interpreted, and double-wall viewing, where both walls are interpreted from the same radiograph. The choice is not a preference; it is what access permits, and the procedure states which techniques are qualified.
The exposure counts are explicit. Double-wall single-viewing needs a minimum of three exposures 120 degrees apart to cover a circumference. Double-wall double-viewing with the beam offset to separate the two weld images needs a minimum of two exposures 90 degrees apart. Superimposed double-wall viewing needs a minimum of three exposures at either 60 or 120 degrees. Firing two elliptical shots on a joint that required three is the coverage gap auditors find fastest, and it is unfixable after demobilization.
Double-wall viewing is bounded by size. It applies to material and welds in components of 3 1/2 in. nominal outside diameter or less, and only a source-side IQI is used. Above that diameter the unsharpness and the superimposition make the near-side image unusable, and the technique reverts to single-wall viewing. Choosing between radiographic and ultrasonic coverage on awkward geometry is a separate decision, worked through in RT versus UT for weld inspection.
Density and the measurement chain behind it
Transmitted density through the area of interest and through the body of the hole IQI adjacent to the essential hole, or beside the designated wire, must be 1.8 minimum for X-ray and 2.0 minimum for gamma on single-film viewing, with 4.0 maximum for both. Composite viewing of a multi-film set requires each individual film to reach 1.3 minimum. A tolerance of 0.05 in density is allowed for variation between densitometer readings, which is the only slack the Article gives you.
Density variation is the requirement crews forget. Density anywhere through the area of interest must not vary from the density measured at the IQI by more than minus 15 percent or plus 30 percent. Exceed it and an additional IQI is required for the exceptional area, and the radiograph is retaken. The allowable variation may be rounded to the nearest 0.1 within the permitted density range, which matters when a reading sits on the boundary.
Behind the numbers sits a measurement chain. Density is judged with a densitometer or a step wedge comparison film. Densitometers are calibrated on a defined interval, at least every 90 days in use, and step wedge comparison films are verified against a calibrated reference before first use and on the interval the Article specifies. A density call from an instrument with no current calibration record is not a density call, and the whole film package inherits that weakness.
Identification, location markers and backscatter
A system has to produce permanent identification on the radiograph, traceable to the contract, component, weld or part number, together with the manufacturer name or symbol and the date of the radiograph. The identification does not have to appear as a radiographic image, but it must not obscure the area of interest. Film that cannot be tied back to a specific weld on a specific date is not evidence, whatever the image happens to show.
Location markers are the coverage proof. Lead markers appear as images on the film and their positions are marked on the part or on a map, so that any area of interest on the radiograph can be found again on the hardware. The Article distinguishes source-side and film-side marker placement for single-wall and double-wall geometries on both flat and curved surfaces. Markers placed on the cassette rather than on the part prove nothing about coverage of the joint.
Backscatter is checked with a lead letter B, at least 1/2 in. high and 1/16 in. thick, attached to the back of every film holder. If the B shows as a light image on a darker background, protection from backscatter was insufficient and the radiograph is unacceptable. A dark B on a lighter background is normal and is not cause for rejection. Omitting the B does not make backscatter go away; it makes backscatter undetectable, which is worse.
Film quality, viewing conditions and the record
Radiographs must be free of mechanical, chemical and processing blemishes to the extent that they cannot mask or be confused with a discontinuity in the area of interest. Fogging, streaks, water marks, chemical stains, scratches, crimps, static marks, finger marks and loss of detail from poor screen-to-film contact all qualify. A blemish outside the area of interest is not automatically a rejection. A blemish lying across the weld image is, because it destroys the interpretation.
Viewing conditions are written as performance rather than as a lux value. Background lighting is subdued enough that reflections, shadows and glare do not interfere with interpretation, and the viewer provides a variable light intensity sufficient to see the essential hole or designated wire across the density range in use. Light spilling around the edge of the film or through thin areas must not interfere either. Article 2 sets no numeric illumination figure here, which surprises people who expect one.
The technique record ties it together. The documentation carries the identification, the date, material type and thickness, film and screen details, the source, and critically the source size and both geometric distances, so the unsharpness calculation can be reconstructed by someone who was not there. Article 2 does not set a retention period; the referencing construction code does. Where an organization needs a Level III of record to own these procedures and reviews, outsourced ASNT Level III consulting is the usual route: affordable, accessible, fully customizable, quote on request.
Does ASME Section V Article 2 tell you whether a weld is acceptable?
No. Article 2 governs how the radiograph is produced, judged for image quality and documented. Whether porosity, slag, a crack or incomplete fusion is acceptable comes from the referencing construction code: ASME Section VIII Division 1, B31.1, B31.3, AWS D1.1 or API 1104. Reading Article 2 for acceptance limits is the single most common misuse of the Article, and it produces arguments with the Authorized Inspector about the wrong document.
What film density does Article 2 require through the area of interest?
Single film viewing requires 1.8 minimum for X-ray sources and 2.0 minimum for gamma sources, with 4.0 maximum for either. Composite viewing of multiple film exposures requires each film of the set to reach 1.3 minimum. Density is read through the body of the hole IQI adjacent to the essential hole, or immediately beside the designated wire, and through the area of interest, never through convenient parent metal.
How is geometric unsharpness calculated and limited under Article 2?
Geometric unsharpness equals the source size multiplied by the source-side-of-object to film distance, divided by the source to object distance. The ceiling is 0.020 in. for material under 2 in. thick, 0.030 in. from 2 through 3 in., 0.040 in. over 3 through 4 in., and 0.070 in. above 4 in. The thickness that selects the row is the thickness on which the IQI is based.
When may an IQI be placed on the film side instead of the source side?
Only when inaccessibility prevents hand placing it on the source side. The film-side IQI sits in contact with the part and a lead letter F is placed adjacent to or on it, so the interpreter knows which sensitivity row applies. The IQI selection table lists a separate film-side designation for each thickness range, so read the row rather than subtracting a designation from the source-side value.
How many exposures does a circumferential weld need under Article 2?
Double-wall single-viewing needs a minimum of three exposures taken 120 degrees to each other for complete coverage. Double-wall double-viewing with the beam offset to separate the two weld images needs a minimum of two exposures 90 degrees apart. Superimposed double-wall viewing needs a minimum of three exposures at either 60 or 120 degrees. Double-wall viewing itself is limited to components of 3 1/2 in. nominal outside diameter or less.
What does the lead letter B on a radiograph mean?
A lead B at least 1/2 in. high and 1/16 in. thick is attached to the back of every film holder to test for backscatter reaching the film. If the B appears as a light image on a darker background, backscatter protection was insufficient and the radiograph is unacceptable. A dark B on a lighter background is normal and is expressly not cause for rejection.