What a Radiographer Must Know Before Shooting a Pipeline Girth Weld

Why pipeline radiography is not vessel radiography

A radiographer with five clean years in an ASME Section VIII fabrication shop can produce a technically flawless pipeline girth weld radiograph and still fail the job. The film is only half the work. The other half is knowing that API 1104 judges a weld by workmanship criteria expressed in absolute dimensions rather than as a proportion of wall thickness, and that the authority standing behind the acceptance is a federal regulation rather than an Authorized Inspector. A 1.5 in elongated slag line is a 1.5 in slag line whether the wall is 0.250 in or 0.750 in.

The consequences run in both directions and both are expensive. Welds get rejected that the code plainly accepts, which costs a spread a cut-out, a repair crew and a day of progress on a right of way where daily weld counts drive the schedule. Welds get accepted that the code rejects, which is worse, because it survives until an operator audit or an incident investigation pulls the film. Both failures share a root cause: acceptance criteria carried over from another code and applied from memory.

Training that starts with method theory and arrives at acceptance criteria in the last hour has the sequence backwards. Pipeline RT training should start with what the code accepts and what the CFR demands, then work backwards into the technique that will let a technician demonstrate it defensibly.

What 49 CFR requires, and how gas and liquid differ

For gas transmission and distribution, 49 CFR Part 192 incorporates API 1104 by reference in 192.7, requires welding procedures and welders to be qualified under it in 192.225 and 192.227, requires visual inspection by a qualified individual in 192.241, and in 192.243 governs the nondestructive testing itself: the process must clearly indicate defects, the operator must have written procedures, the personnel must be trained and qualified, and the proportion of girth welds tested escalates with class location from a floor in Class 1 up to complete testing at specified crossings, tie-ins and locations where a failure would be least tolerable.

Hazardous liquid pipelines under Part 195 arrive at coverage differently. 195.234 samples by welder and by welding day rather than by class location, and requires full-circumference testing of the sampled welds, with complete testing in specified locations. A technician who has only worked gas construction will misread a liquid project's coverage plan as arbitrary. It is not arbitrary; it is a different sampling philosophy, one aimed at catching a welder going out of control rather than at protecting a population density band.

There is a trap in the incorporation itself. Section 192.7 names a specific edition of API 1104. The published edition of API 1104 can be newer than the one the regulation adopts. A technician trained last month on the current edition can confidently apply a clause the regulated project has never adopted, and an operator's quality auditor will find it. Check the incorporated edition before the first shot, and record which edition the procedure was written to.

Source, energy and geometry selection on a live spread

Iridium-192 remains the workhorse of pipeline radiography because its energy suits the wall thicknesses that dominate transmission construction and because a source projector travels on a truck. Selenium-75 earns its place on thinner wall and small bore, where its lower energy gives better subject contrast and, just as importantly on a populated tie-in or inside a compressor station, a materially smaller exclusion zone. Cobalt-60 is reserved for heavy-wall station piping and thick fittings, and it brings a much larger barricade with it. X-ray crawlers dominate large-diameter mainline work where diameter and internal access allow.

Geometry follows access. A panoramic single-wall single-image exposure from inside the pipe is the highest-quality and by far the fastest option: one exposure covers the full circumference, the source-side IQI sits where it should, and no second wall is superimposed on the interpreted image. When there is no internal access, double-wall single-image with the source offset projects the near wall clear of the area of interest. Below roughly 3.5 in outside diameter the elliptical double-wall double-image technique takes over, two exposures at ninety degrees, with the ellipse opened enough that the two weld images can be interpreted separately.

Superimposed double-wall double-image is the technique that produces the most interpretation errors, because both walls are read together. The interpreter must be able to reason about which wall an indication belongs to using geometry and image sharpness, and must state the limitation in the report rather than guessing. A technician who has never been taught to refuse a technique is a liability on a tie-in crew.

Image quality: IQI, density, unsharpness and their digital equivalents

The image quality indicator is not decoration. It is the evidence that the technique achieved the sensitivity the code demands, and it is the first thing a competent film reviewer looks for. Wire sets to ASTM E747 and hole-type plaques to ASTM E1025 are both in use; which one, and which essential wire or hole, is read from a thickness table in the referencing code. Source-side placement is the default. Film-side placement is permitted in defined circumstances, demands greater sensitivity, and must be identified with a lead F on the radiograph. A hole-type IQI placed on the weld rather than on parent metal needs a shim of the same material and comparable thickness beneath it.

Density is the second control. The code sets a minimum through-density in the area of interest, higher for gamma sources than for X-ray, a maximum determined by the viewing equipment, and a permitted percentage variation across the interpreted length. The lead B on the back of the cassette is the backscatter check: if it appears as a light image on a darker background, the radiograph is void regardless of how good it otherwise looks. These are the checks that a reviewer applies in the first thirty seconds, and a technician who cannot pass them wastes a shift.

Geometric unsharpness is source size multiplied by material thickness, divided by source-to-object distance, and ASME Section V Article 2 caps it in bands by thickness. On computed and digital radiography the controls change identity rather than disappearing. Density is replaced by signal-to-noise ratio, and IQI sensitivity is joined by basic spatial resolution measured with a duplex wire gauge. Computed radiography systems are classified using a defined phantom and procedure; digital detector arrays have their own qualification route. A crew moving from film to digital mid-project needs retraining on the acceptance of the image, not just on the hardware.

Reading API 1104 acceptance without importing ASME habits

Start with cracks, because the rule is nearly absolute: a crack is rejectable regardless of size or location, with a narrow allowance for a shallow crater or star crack under a defined short length. Everything else is dimensional. Incomplete penetration, with or without high-low, is limited by individual length and by aggregate length in a continuous twelve-inch weld length. Incomplete fusion, including cold lap between passes, follows the same logic. Elongated and isolated slag inclusions have separate width and length limits, and a cap on how many isolated inclusions may appear in a given weld length. Porosity is limited by individual pore size, evaluated against the smaller of a fixed dimension and a percentage of wall thickness, and by distribution against charts.

Then the clause that generates the most argument on a right of way: internal concavity. Under API 1104, internal concavity of any length is acceptable provided the radiographic density through the concavity does not exceed the density of the thinnest adjacent parent metal. It is a density test, not a dimensional one, and it is unlike anything in vessel practice. Radiographers arriving from fabrication reject concavity on sight and hand a spread a cut-out it never needed. Teaching this clause properly, with a densitometer in hand and film examples, pays for a course on its own.

Two further habits to unlearn. Hollow bead is a real, named pipeline defect arising from cellulosic root passes and has its own criteria; it does not exist in the Section VIII vocabulary. And API 1104's Annex A alternative acceptance standards are fitness-for-service based, requiring flaw height and an engineering critical assessment. Film does not give reliable through-wall height, which is precisely why mechanised ultrasonic testing with zonal discrimination displaced radiography on large ECA-governed spreads. A technician should know why the method changed, not just that it did.

Radiation safety, and the two certifications a pipeline radiographer needs

Industrial radiography using byproduct material is governed by 10 CFR Part 34, or by the equivalent programme of the Agreement State that has assumed regulatory authority. No one may act as a radiographer until certified through a recognised certifying entity under the criteria in Appendix A to Part 34, and that certification sits on top of employer site-specific training and documented supervised work. A radiographer's assistant operates only under the personal supervision of a certified radiographer and may not run the device alone. Dose limits come from 10 CFR Part 20: an annual occupational total effective dose equivalent, a much lower limit for a declared pregnant worker over the gestation period, and public dose limits that set where the barricade goes.

The operational discipline is unglamorous and it is where audits land. Personnel monitoring requires both a direct-reading device and a processed dosimeter, plus an alarming ratemeter. Survey instruments must be calibrated at defined intervals and after repair, and their calibration records travel with them. A physical radiation survey is required after each exposure to confirm the source has returned to its shielded position, and the single most common finding on a radiography audit is a missing or reconstructed post-exposure survey record. Transport of the exposure device brings 49 CFR hazardous materials obligations, shipping papers, labelling and, for higher-activity sources, a security plan.

The distinction that matters for training planning is this: the Part 34 radiographer certification is a safety credential permitting device operation. The ASNT SNT-TC-1A or ISO 9712 Level II in radiographic testing is a technical credential permitting technique selection, interpretation and reporting. Neither one substitutes for the other, and a project that hires one believing it has bought both discovers the gap when the film review starts.

Radiography in integrity work, where the target is wall loss

Construction radiography looks for weld flaws. Integrity radiography usually does not. On an in-service insulated line, profile or tangential radiography images the pipe wall in silhouette through the insulation and jacket, letting a crew screen for corrosion under insulation, wall loss at supports, and liquid or deposit levels without stripping cladding. The technique is geometrically different, the exposure is different, and the reading is different: the technician is measuring remaining wall from a magnified profile image, with a correction applied, not searching for slag.

The acceptance basis changes completely. API 1104 has nothing to say about a corroded run of process piping. Remaining thickness is compared against a minimum required thickness, and a local area of metal loss is evaluated under ASME B31G for a corroded pressure-containing area or under API 579-1/ASME FFS-1 for general or local metal loss. A technician who reports a profile radiograph in the language of weld defects has produced a document the integrity engineer cannot use. Reporting to the right framework is part of the method training, not an afterthought.

Equally important is teaching the limits. Stress corrosion cracking and strain-induced girth weld cracking are tight and planar, frequently orientated so that a radiographic beam passes across rather than along the flaw, and a clean radiograph is therefore weak evidence of their absence. In-line inspection, phased array and mechanised ultrasonics carry that load. A Level II who writes down that radiography was not capable of addressing the stated threat is doing the job correctly.

How a technician qualifies for pipeline radiographic work

Under the employer-based US system, ASNT SNT-TC-1A recommends classroom training hours and in-method experience hours for RT at each level, and radiography sits at the demanding end of that table alongside ultrasonics and eddy current. But the recommended practice recommends; the employer's own written practice is the controlling document, and an auditor holds the employer to the numbers it wrote. Where a client instead invokes ANSI/ASNT CP-189, the requirements become mandatory minimums and the Level III must hold an ASNT Level III certificate. Where a client invokes ISO 9712, an accredited third party examines and certifies, and the certificate follows the individual.

Certification comes through general, specific and practical examinations. For radiography the practical is the part that separates candidates: a graded set of radiographs to interpret against a stated acceptance standard, plus technique calculation, IQI selection and setup. Candidates who read the films well and write unusable reports still fail, and they should. Level II may interpret and accept or reject; Level I may perform under a written instruction and record results but may not evaluate or make the accept-reject decision, a boundary that gets quietly crossed on night shifts and shows up in audits.

The finding that recurs on pipeline projects is narrower than most people expect. It is rarely a missing certificate. It is a technician certified in RT with no documented specific examination covering API 1104, the incorporated CFR edition, or the operator's own procedure. Atlantis delivers RT training and certification preparation at Level I, II and III to ASNT SNT-TC-1A and ISO 9712, classroom, on-site or blended, and ASNT Level III consulting to write the written practice and the specific examinations that close exactly that gap. Talk to us at info@atlantisndt.com for a consultation or a scoped quote.

Request a consultation