Pipe Welding X-Ray Inspection: The Complete RT Guide

A 6-inch carbon steel line is welded, inspected, and cleared for service. Then somebody in the QA office asks a deceptively simple question: “Where is the radiograph?” The answer turns out to involve a 5-mesh-per-inch wire penetrameter, a 220 kV exposure, and a careful reading of a black-and-white image where a crack can hide between two pixels of slag. Radiographic testing (RT) is the oldest and still one of the most trusted ways to look inside a pipe weld without cutting it open. This guide explains exactly how that image is made, what the codes demand, how to read what you see, and how to avoid the handful of mistakes that send perfectly good welds back for rework.

What Pipe Welding X-Ray Inspection Actually Shows You

Before a pressurized pipe system is cleared for service, the most important question is never about how the weld looks from the outside, but what is hiding on the inside. This is the exact purpose of pipe welding x-ray inspection, more commonly known as Radiographic Testing (RT). By sending either X-rays or gamma rays through a welded joint, RT creates a permanent image of its internal structure. For over a century, this has made it the backbone of weld quality control, as it allows for a full internal view without having to cut the joint open.

What It Can, and Cannot, Find

The goal of RT is to detect internal discontinuities within the weld metal and its adjacent areas. According to ASNT, radiographic testing is the only common NDT method that produces a permanent, auditable record of the internal condition of a weld, complete with evidence of sensitivity through the use of image quality indicators (IQIs). It is also capable of detecting small changes in thickness and density down to about 1% along the path of the X-ray beam. With this level of sensitivity, RT is most effective at finding volumetric defects, such as porosity, slag inclusions, and certain forms of lack of fusion or incomplete penetration.

However, its strengths have their limits. RT is far less reliable for tight, surface-breaking cracks, which is why methods such as penetrant or magnetic particle testing are often paired with it. More importantly, RT cannot confirm the welding process itself. Preheat, interpass temperature, heat input, and overall weld procedure compliance are all process variables that determine weld metallurgy, so an acceptable radiograph does not by itself prove that the welding procedure was followed. This is one of the largest points of confusion for many welders and project teams.

Reading The Silhouette

To understand a radiograph, it is easiest to think of it as a silhouette. The weld is projected as an image of varying thickness, and the amount of radiation that reaches the film or detector determines the color. Less dense regions, such as voids from porosity or gaps from a crack, attenuate fewer X-rays and therefore appear dark. Denser inclusions, such as slag, block more radiation and therefore appear light. This is the opposite of what most people would expect at first glance.

Before reading these indications, it is important to know the joint itself. For pipe, the two main weld types are girth welds (circumferential joints connecting two pipe sections) and longitudinal welds (seams running along the length of the pipe). Within these joints, inspectors pay extra attention to the root (the bottom of the weld closest to the inside of the pipe) and the face reinforcement (the built up cap on the outside). From here, the approach to shooting the image comes down to geometry. Single-wall single-image (SWSI) places the source and detector on opposite sides of a single wall of the pipe. Double-wall single-image (DWSI) places both on the same side of the pipe, imaging both walls at once onto one layer of the film. Double-wall double-image (DWDI) produces two separate images of the two walls, and is utilized when access to only one side of the pipe is possible. The selection between these is based entirely off of pipe diameter and accessibility.

Who Needs To Know This, And What Governs It

This information is vital for QA inspectors, welders, project engineers, and shop foremen who need to know exactly what the radiographer is tasked to do with their pipe. These tasks are never done at random, as they are governed by established codes and standards. As stated by ASME BPVC Section V, acceptance criteria are outside of the scope of Section V itself. Instead, the acceptance limits come directly from the construction codes, such as ASME Section VIII, ASME B31.3, API 1104, and others. From specialized repairs all the way down to tasks such as aluminum boat welding, this distinction between “how to examine” and “what is acceptable” applies across every industry that utilizes RT.

With all of this in mind, the rest of this guide will walk through the physics of the rays, the equipment used to create them, procedure building, code requirements, interpretation, alternative NDT methods, common mistakes, safety regulations, and a practical workflow to put it all together. Before any exposure is made, understanding these fundamentals is the first step to obtaining a radiograph that is not only readable, but defensible.

The Physics Behind the Radiograph: How X-Rays and Gamma Rays Reveal Flaws

Before any exposure is taken, it’s important to understand the simple idea behind a radiograph: radiation is shot through the pipe weld, and the image we see is nothing more than a shadow. Denser material, such as solid steel or a heavy inclusion, blocks more radiation and comes out light on the film or detector. Less dense material, such as a gas pocket or a void, lets more radiation through and comes out dark. With that in mind, the way we create that radiation is what dictates how much of a weld we can actually see.

Creating the Beam: X-Ray Tubes Versus Gamma Sources

For most field work, there are two sources to choose from. The first is a portable X-ray tube. Inside of it, a heated filament releases electrons, which are then accelerated across a high voltage gap and slammed into a tungsten anode. That sudden deceleration produces a broad bremsstrahlung spectrum of X-rays, along with a few characteristic tungsten lines. These machines commonly operate anywhere from about 100 kV to 300 kV for field work, giving the radiographer a controllable beam at the flip of a switch.

The second option is gamma radiography. Rather than creating radiation with electricity, isotopes such as Iridium-192 or Cobalt-60 decay spontaneously. According to ASNT, these sources require no electrical power at all, which is exactly why they are the workhorse for remote sites, offshore platforms, and anywhere extension cords simply aren’t an option. Iridium-192 has a half-life of roughly 73.8 days and emits a spectrum of roughly 0.2 to 2.0 MeV, while Cobalt-60 has a half-life of about 5.27 years and emits two discrete photons at 1.17 and 1.33 MeV. The trade off is that you can’t just “turn down” a gamma source, you have to account for its decay over time to calculate your exposure.

The Four Levers of an Exposure

Regardless of which source is used, every exposure is controlled by four primary parameters. Kilovoltage (kV) controls penetration, and therefore the required exposure time. Milliamperage (mA) controls the amount of radiation produced per unit of time. Exposure time sets the total dose delivered to the film or detector. Lastly, source-to-object distance (SOD) controls intensity and magnification through the inverse square law. This law is one of the most practical to remember: doubling the source-to-object distance cuts the intensity at the object to roughly one quarter.

All of these levers are in constant conflict with one another. A higher kV will push through thicker pipe, however it will also reduce contrast and increase scatter, which is why technique charts are built as a compromise, not a single magic number.

Geometry: Keeping the Image Sharp

Penetration alone won’t make a usable radiograph. Geometry is what determines if a tight crack can be seen, or if it’s smeared into nothing. The biggest hurdle here is geometric unsharpness (Ug). The formula to know is: Ug = source size x (object-to-image distance / source-to-object distance).

To combat this blur, a small focal spot and a long source-to-object distance are desirable. However, there is a catch. If the film or detector is placed far away from the pipe, the object-to-image distance grows, magnifying the image and increasing geometric unsharpness. The goal is a careful balance between distance, source size, and pipe size to keep the flaw as sharp as possible. These geometric considerations are just as important when working with heavier sections of alloy steel, as their increased thickness leaves very little room for error.

Density, Contrast, and Definition

With a correct exposure and geometry, the quality of that image comes down to three factors as defined by TWI: density (the overall film blackness), contrast (the difference in density between the weld and the surrounding base metal), and definition or sharpness (the ability to separate a fine indication from its surroundings). All three must be present to accurately evaluate a flaw.

To protect those three factors, lead foil screens are almost always used. These screens sit against the film and absorb backscatter radiation that would otherwise fog the film and create false indications. Lead backing is used for the same purpose, and backscatter protection is a standard element of an ASME Section V Article 2 procedure. Even the best beam can be ruined without it.

Ultimately, a good radiograph isn’t made by blasting the thickest wall possible with the highest energy. It’s made by understanding these physics. By knowing the difference between an X-ray tube and a gamma source, the effect of the inverse square law, the math behind geometric unsharpness, and the balance between density, contrast, and definition, a radiographer can reason through their technique choices before a single shot is ever fired. That knowledge is the foundation for every acceptable radiograph to come.

Choosing Your Equipment: X-Ray Tubes, Gamma Sources, Film, and Digital Detectors

The exposure is only half the battle in pipe welding x-ray inspection. The equipment behind the radiograph sets penetration power, turnaround time, licensing burden, and whether the image will be accepted by the governing specification. Choosing a setup means balancing practicality, code compliance, and image quality.

The Four Core Methods

Circumferential pipe welds are radiographed four main ways: high-frequency X-ray tubes, gamma sources, panoramic radiography cameras, and computed or digital radiography (CR/DR).

High-frequency X-ray tubes are the most common shop-based option. They are fast, produce a clean beam, and carry far fewer regulatory burdens than radioactive sources. The tradeoffs are power dependence and limited penetration. They handle thinner wall pipe easily, but as wall thickness climbs their output can fall short.

Gamma radiography uses sealed sources, most often Ir-192 or Co-60. Its main advantage is that it needs no electrical power, which keeps it preferred for offshore, in-plant, and night-shift work where generator access is difficult. Sources are self-contained and reach much thicker steel than most portable X-ray sets. The obligations are heavy: source licensing, decay-based exposure calculations, strict transport rules, and disposal logistics. With Ir-192 sources, decay runs roughly 10% per week, so exposure times must be recalculated from a decay table as the source ages. A technique that was correct on day one can demand a materially longer exposure weeks later.

Panoramic radiography cameras push efficiency further. A small X-ray tube rides a rotating arm inside the pipe, wrapping the beam around the joint to capture the full circumference in a single exposure. On long runs of straight pipe, that can eliminate the need for multiple shot exposures. The catch is rigidity. Panoramic shots require a clean, straight, unobstructed section with no adjacent fittings, valves, or supports in the way. Any obstruction blocks the beam and ruins coverage.

Film vs. Computed and Digital Radiography

Industrial radiographic film was the standard for decades. It still delivers remarkably fine detail in the right configuration and is still required by some legacy specifications and aerospace work, but the workflow has shifted toward digital.

ASNT describes two digital alternatives to film: computed radiography (CR), which uses reusable imaging plates that must be read and erased by a CR scanner, and digital radiography (DR), which uses flat-panel detectors to produce an image immediately with no external reader needed.

The shift is about more than convenience. Digital removes chemical processing entirely, eliminating a whole category of rejection causes such as chemical splash marks, static electricity marks, over-development, under-development, and fogging. Waygate/Fujifilm’s NDT resources note that digital images can be enhanced, measured, and archived without the fidelity loss film suffers when enlarged. DR also allows on-the-spot review and retakes, so the number of exposures a crew needs in the field often drops even though the capital cost is higher.

There are caveats. Digital systems demand a large upfront investment, trained personnel, and compatibility checks against client specifications, and some owner and API specifications still require film outright. On specialized jobs in alloy steel piping where tight, legacy criteria apply, film may be the only accepted medium regardless of howadvanced the shop’s digital suite is.

A Practical Equipment Selection Guide

Site conditions usually settle the argument. When building a radiographic plan, weigh the following:

• Pipe material and wall thickness: Thicker sections tend to favor Ir-192 or Co-60 gamma sources, or higher-output X-ray sets. Thinner walls suit portable high-frequency X-ray units or digital systems.
• Diameter and geometry: Long, straight, unobstructed girth welds make panoramic cameras very efficient. Smaller spools, tight tie-ins, or pipes with adjacent flanges and elbows call for individual shot exposures with a gamma projector or a portable X-ray tube.
• Access and power: Without reliable power, with limited scaffolding, or on an offshore platform, gamma radiography is often the only realistic option. In a shop with easy access to electricity, high-frequency X-ray or DR cuts turnaround dramatically.
• Code and client requirements: Check the governing specification before booking any equipment. The contract, not convenience, decides whether CR, DR, or film is allowed and which IQI types are acceptable.
• Turnaround and volume: At high volume, the ability to review, enhance, and re-shoot digitally can pay back a DR system over time. For one-off field repairs, gamma or film can be far cheaper to mobilize.

There is no universal best setup for pipe welding x-ray inspection. A well-planned job matches the beam to the pipe, the medium to the code, and the mobilization to the reality of the site. The most efficient radiograph is not the most expensive one. It is the one that is qualified for the job, shoots correctly the first time, and produces an image that can be confidently evaluated against its acceptance criteria.

Building a Reusable Radiographic Procedure: IQIs, Exposure, and Image Quality

A radiograph is only as reliable as the procedure used to create it. For pipe welding, the goal is not to simply produce an image, but to prove that the image itself is sensitive enough to show the flaws that the code cares about. To do this, every qualified radiographic setup must be built with control, consistency, and a way to verify its own quality before any interpretation takes place.

The Purpose of the Image Quality Indicator (IQI)

The single most important tool for this verification is the image quality indicator, historically known as a penetrameter. An IQI is placed in the beam path to demonstrate that a specified change in material thickness is detectable in the final image. In practice, this means the technique must be capable of resolving a small detail that is representative of the sensitivity required by the code.

The classic call-out, ‘2-4T’, is a perfect example of this. It means a shim equal to 2% of the wall thickness with a hole four times the shim thickness. If that hole is visible on the radiograph, the setup has met the sensitivity for that thickness range.

Wire-Type vs. Hole-Type IQIs

There are two main IQI designs used for pipe welds. Wire-type IQIs consist of a set of thin metallic wires of decreasing diameter, mounted between plastic or acrylic. These are generally considered the most sensitive for detecting small defects. Hole-type IQIs, on the other hand, utilize a shim with a series of drilled holes of different diameters. They are easier to read and are used in situations where a numerical wire sensitivity requirement is not specified. Both designs are widely used within the radiographic testing field.

These IQIs are governed by a set of standards. For film, wire-type IQIs fall under ASTM E2002, while hole-type IQIs fall under ASTM E747. For digital systems, the equivalents are ASTM E2002M and ASTM E746. Most notably, EN ISO 17636-1 for radiographic examination of welded joints in metallic materials tabulates required wire IQI values — ranging roughly from W19 for very thin sections down to W3 for the thickest double-wall work — and defines image quality classes A and B.

IQI Placement: Source Side vs. Detector Side

Where the IQI is placed is just as important as which one is selected. As noted by Baker Hughes Waygate Technologies, the IQI should sit where contrast and definition are least favourable, near the extremities of the diagnostic area, as that is where geometric unsharpness is highest and film density is lowest.

With this in mind, the placement rule is straightforward:

• Single-wall single-image (SWSI): The IQI is normally placed on the source side of the pipe, over a section of equivalent thickness to the weld.
• Double-wall techniques (DWSI/DWDI): The IQI is placed on the far side of the pipe, as the radiation must pass through the additional wall to reach it.

Building the Rest of the Procedure

With the IQI in place, the remaining elements of the procedure must be controlled to ensure repeatability. For pipe welds, especially those made with alloy steel welding electrodes, surface condition is often the first failure point. Rust, mill scale, paint, grease, and spatter must be removed from the joint area, along with any insulation or coating. The joint must also be free of clamps that could obstruct the beam or cause artifacts.

From there, exposure planning comes into play. A qualified exposure chart must be used to select the correct parameters for the source-to-object distance, kilovoltage or source strength, and time. The radiographic image must extend beyond the weld reinforcement on both sides so the interpreter can see the weld toes. Additionally, the image must show the identifying information for the weld, be adequately shielded, and have backscatter protection in place. All of this must be verified using the quality control guidance found within NDE-Ed’s radiography quality control resources.

Quality control doesn’t stop at the exposure. For film, this includes film processing control, screen and cassette handling, and darkroom QC. For digital radiography, it includes digital QC checks for the imaging plate or flat panel, along with system performance checks. Most importantly, any image deemed unacceptable for image quality, coverage, or IQI visibility must be re-shot. It cannot be justified or accepted.

Qualification: Making it Repeatable

All of these controls culminate into one requirement: the technique must be auditable. ASME Section V Article 2 requires that the radiographic procedure be qualified, including a performance demonstration in Appendix A. This means the exact setup, IQI placement, technique, and parameters used for production must be proven to be capable of producing an acceptable image. The result is a procedure that is repeatable and auditable, rather than one that relies on operator intuition. When executed correctly, this system is what turns a single shot into a defensible, permanent record of the weld’s volumetric integrity.

What the Codes Require: ASME B31.3, Section VIII, and API 1104

When dealing with pipe welding x-ray inspection, the most misunderstood split is between the “how” and the “what”. The technique to shoot a radiograph is not the same thing as the rules to accept or reject it. To put it plainly, ASME BPVC Section V is the how-to book, while the construction codes are the ones that tell you what is acceptable. Within Section V, Article 2 governs radiographic examination, whilst Articles 4, 6, 7, and 9 cover the other common methods. Most importantly, Article 1 explicitly excludes acceptance criteria from its scope. For any defensible RT report, both must be cited: the method standard, and the acceptance standard that the job is being held to.

ASME B31.3: Process Piping

For process piping, the extent of examination is driven by ASME B31.3. Under paragraph 341.4.1(b), not less than 5% of circumferential butt and miter groove welds must be examined fully by random radiography in accordance with 344.5, or by random ultrasonic examination in accordance with 344.6. However, this is not a “one and done” percentage. B31.3 enforces a progressive escalation rule: if an unacceptable indication is found within the examined sample, the examination scope must be increased, often to a much larger portion of the welds.

For higher severity service, the requirements are raised drastically. Category M severity levels require at least 20% of circumferential butt and miter groove welds, as well as fabricated lap and branch connection welds comparable to those shown in the referenced figures, to be examined. These stricter requirements are put in place for services where failure would be far less tolerable.

Not every weld requires a full length shot either. For spot radiography, 341.5.1 lays out specific rules. For longitudinal groove welds requiring a joint factor, E_j = 0.90, radiography must cover at least 300 mm (1 ft) in each 30 m (100 ft) of weld for each welder or welding operator. For girth, miter, and branch groove welds, a single elliptical exposure encompassing the entire weld circumference is required for sizes up to DN 65 (NPS 2-1/2). For sizes larger than that, the spot must cover the lesser of 25% of the inside circumference or 152 mm (6 in.). These spot rules are calculated per welder, meaning they cannot be pooled across an entire shift to avoid coverage.

ASME Section VIII Division 1: Pressure Vessels

When it comes to unfired pressure vessels, the acceptance criteria shift over to Section VIII Division 1. Full radiography is governed by UW-51, whilst spot radiography is governed by UW-52. The philosophy here is strict: any crack, any zone of incomplete fusion, or any incomplete penetration is unacceptable under either UW-51 or UW-52, with no allowance for length. Slag, porosity, and other rounded indications are not automatically rejected, but are judged against the defined limits for the vessel’s thickness.

This distinction is vital. When inspecting critical pressure retaining welds, especially those made from materials such as those covered under our alloy steel welding electrodes selection guides, the production code is always the one that decides acceptability, not the qualification test alone.

API 1104: Longitudinal Pipeline Welds

For cross-country and transmission pipelines, API 1104 is the governing standard for welding and radiographic acceptance of longitudinal pipeline welds. Much like the other codes, API 1104 has its own acceptance tables for planar and rounded indications that differ from B31.3 and Section VIII. These differences are intentional, as pipeline construction has its own geometry, joint designs, and in-service considerations.

That being said, the code minimum is rarely the final word. Owner specifications are extremely common within the pipeline industry and will frequently add stricter requirements on top of API 1104. These are enforced through Inspection and Test Plans (ITPs), and the golden rule is this: the contract document controls wherever it is more stringent than the code.

With all of this in mind, remember one final rule that ties every one of these codes together. No matter if you’re working to B31.3, Section VIII, or API 1104, the code that is specified for construction is the acceptance standard. Section V will always tell your radiographer how to create a quality radiograph, however it will never tell you if that weld is acceptable for service.

Reading the Radiograph: Defects, Artifacts, and Acceptance Criteria

Reading a pipe weld radiograph is less about guessing and more about knowing what you’re looking at. For an interpreter, every shade on that film has a reason. The goal isn’t to find anything that looks odd, it’s to tell the difference between a true weld defect, and a man-made artifact, then measure it against the correct acceptance criteria.

The Anatomy of a Weld Radiograph

Before looking for flaws, it’s important to get oriented with the image itself. On a properly exposed girth weld radiograph, you’ll typically see the base metal, the weld metal, reinforcement on the cap and root, the root face, and slight density changes where the heat-affected zone blends into the parent pipe. You’ll also see the identification block or ID wire placed on the film to prove which joint, weld, and location you’re looking at.

Understanding density is the biggest key here. Volumetric indications appear as dark (high exposure) areas because voids or thinner material allow more radiation to pass through to the film. On the flip side, slag and other dense inclusions appear light because they absorb more radiation than the surrounding weld metal. This simple rule is what allows an interpreter to separate many indications from one another right off the bat.

Weld Defects to Watch For

Not every indication is equal, and each defect tells a different story about the weld. With practice, these are the ones that get the most attention:

• Porosity: These appear as round, sharply defined dark spots. They can be randomly scattered throughout the weld, or clustered together. While small amounts can be acceptable, they are always judged by their individual size and by their total count.
• Slag inclusions: Slag tends to show up as linear, globular, or irregular lighter areas. Often, it’ll have an obvious track or tail to it. Since it’s denser than the surrounding metal, it blocks radiation and reads as a lighter indication on the film.
• Lack of fusion: One of the more concerning planar defects, lack of fusion typically shows as a dark, unfused seam that runs along the fusion line. It is often wavy or serrated, as opposed to a perfectly straight edge.
• Incomplete root penetration: This appears as a dark band at the root of the weld. What separates it from a crack is its uniform width and smooth, parallel edges, often extending for a large portion of the joint length.
• Cracks: Cracks are dark, sharp, jagged, and can be irregularly branched. These are almost always viewed with the highest scrutiny, as they are planar, can propagate, and are rejected outright under many code acceptance criteria.
• Contamination inclusions: For processes such as TIG or GMAW, tungsten or copper inclusions from contamination can appear as small, isolated, dense spots. These are usually light indications and can be very easy to miss if not looked for specifically.

With the sheer amount of work done in more advanced fabrication, such as aerospace TIG welding, the need to catch these planar and volumetric defects is taken to an even higher level of precision.

Artifacts: The False Alarms

The hardest skill for an interpreter to learn isn’t finding defects, it’s proving an indication is real. Artifacts are the product of technique, handling, or equipment, not the weld itself. The ones that cause the majority of rejected radiographs are:

• Contact issues: Double images from film or lead screens not in contact with the object, or lead screen separation.
• Handling marks: Static electricity marks from film handled in dry conditions, fingernail scratches, or chemical splash.
• Exposure errors: Light leaks, under-exposure, over-exposure, collimator cut-off leaving unexposed margins, source shadow or cone cutoff, and backscatter fogging.
• Processing and geometry: Processing stains from chemistry, and geometric distortion caused by the pipe being off-centre or the film not parallel to the joint.

Acceptance Criteria and the Discipline of Evaluation

Once an indication is deemed real, it must be measured and judged against the governing criterion. Defect size is judged against the nominal wall thickness (t) of the pipe, not the reinforcement thickness, so wall thickness records must be available to the interpreter at the time of evaluation.

For radiographic testing conducted to ASME Section VIII, some of the commonly cited summary limits are strict for a reason. No individual pore may exceed one-quarter of the nominal wall thickness and no individual pore may exceed 3.2 mm in diameter, whichever is more restrictive. For slag, the limits are tighter still. For spot radiography, slag length limits are approximately two-thirds of the wall thickness, with slag above 19 mm (3/4 in.) always rejected and slag below 6 mm (1/4 in.) always acceptable regardless of thickness.

Most importantly, cumulative slag and cumulative porosity limits apply. Many small indications that could individually pass, can still fail the joint when summed over a defined length of weld. This is where the experience of the interpreter comes in, paired with the exact tables from the governing code, as each code can vary slightly from the other.

A formal evaluation isn’t finished until the work is documented. Any measured indications must be taken with calibrated scales, photographed or recorded, and the final decision must be a clear accept or reject that references the governing criterion. Only then can the radiograph be considered a defensible, auditable record of that pipe weld.

Radiography vs. Ultrasonic, Penetrant, and Magnetic Particle Testing

Radiography became the default examination for pipe welds for one good reason: it shows the inside of the joint without opening it up. But it is not automatically the best method for every weld, material, and code clause. Picking the wrong technique is expensive, and in the worst case it produces a report that declares a weld acceptable when the defect that actually mattered was never looked for.

Method	What it finds	Access needed	Permanent record	Turnaround
Radiography (RT)	Volumetric defects, some planar indications	Both sides for single-wall	Yes, film or digital	Hours on site
Ultrasonic (UT)	Planar defects, crack-like indications	One side only	Only if data is captured	Immediate
Liquid penetrant (PT)	Surface-breaking only	One side	No	Minutes
Magnetic particle (MT)	Surface and near-surface	One side	No	Minutes

Where Radiography Wins

The American Society for Nondestructive Testing characterizes radiography as best suited to volumetric-type flaws and notes that it produces a permanent record, two qualities that are why it dominates pipe weld acceptance.

One exposure gives a complete cross-section of the joint: near wall, weld metal, root, reinforcement, and far wall. Because the image shows both walls, an interpreter can compare transmitted thickness through each side, which can reveal internal corrosion or wall thinning on the far side of the pipe. No surface method can make that comparison.

The image is also a legal document. A film in a cabinet or a digital image in a controlled repository can be pulled up years later during an audit, incident investigation, or dispute, and it still shows what the joint looked like on the day it was accepted. Aerospace work still leans on film radiography for exactly this reason, and the aerospace welding guide covers how those inspection programs fit around the welding process.

Where Radiography Falls Short

• Access. Single-wall techniques need the joint clear on both sides. On a buried line or inside a tight rack, that alone can rule RT out.
• No live systems. RT cannot be performed on an operating pressure line. The pipe must be de-pressurized, drained, and typically gas-free, and coating or insulation over the joint has to come off.
• Speed. The exposure takes seconds; setup, shot taking, film processing, and interpretation consume the morning.
• Sensitivity. RT is comparatively poor at tight cracks. A narrow, closed indication can fall below the sensitivity the IQI demonstrates.

Ultrasonic Testing: Stronger on Planar Defects

UT is generally the better choice for planar flaws and crack-like indications, especially lack of fusion and tight cracks that radiography under-reports. It works well in thick sections, needs access to only one side of the joint, and results appear on the screen as they are acquired. Those advantages are why ASME B31.3 accepts random ultrasonic examination under paragraph 344.6 as an alternative to random radiography.

UT also has real drawbacks. It needs qualified procedures, calibrated reference blocks, trained and certified personnel, and an operator who can interpret the screen. Results live in the instrument and the operator’s judgment unless the data is captured and archived, so a UT examination may not produce the permanent, transferable record an image does.

Liquid Penetrant Testing

PT is the standard surface-breach method for clean, nonporous materials. It is cheap, fast, and highly sensitive to fine surface cracks, so it remains the standard for aluminum MIG welding and other non-ferrous work where magnetic particle testing cannot be used. The limitation is that PT detects only defects open to the surface. It says nothing about porosity, slag, or lack of fusion buried inside the weld, and it needs a surface free of scale, rust, paint, and welding slag. Cleaning is part of the method, not an optional extra, and air arc gouging and grinding are often part of that prep.

Magnetic Particle Testing

MT finds surface and near-surface defects in ferromagnetic material, a strong companion to radiography on carbon and low alloy steels such as those covered in the alloy steel welding electrode guide. It does not work on austenitic stainless steels, aluminum, copper alloys, or non-metallic materials. Like PT, it leaves no permanent record and only sees defects that break the surface.

Choosing and Combining Methods

Most specifications pair radiography with visual inspection and, where appropriate, PT or MT. In thick-walled code piping, the trend has been toward random UT because it needs no double-sided access. Acceptance criteria are calibrated to the defect types a method is good at, so swapping the method without rewriting the criteria changes what a pass actually means.

Choose the method that answers the question your code is actually asking, then specify it fully. Radiography still earns its place on pipe welds because nothing else shows the whole joint at once, but UT, PT, and MT each cover ground radiography cannot reach.

Common RT Mistakes and How to Prevent Rejected Shots

The most expensive part of radiography is not the exposure. It is the third one. A wrong technique costs minutes; a rejected image costs a crew, a re-shoot, a repair cycle, and on a turnaround project, a schedule slip that lands on everybody else. Rework logs show the pattern fast: most rejected radiographs trace back to conditions that were known and correctable before the exposure was made. Cleaning, access, distance, coverage, technique selection. The weld itself is rarely the culprit.

The Mistakes That Cost the Most

Mistake	What you see on the image	Corrective action
kV too low for wall thickness	Light, washed-out image, poor contrast. A crack can hide inside the weld silhouette.	Move up the technique chart until penetration is achieved, then rebalance exposure time. Under-penetration hides the very defects you paid to find.
kV too high	Excess scatter, muddy image, loss of definition. Crack detail softens.	Drop to the charted kV and extend exposure time instead. Both errors push the image off the balance point, which is why charts exist.
Wrong exposure time or milliamperage	Film too light or dark. IQI wires washed out or merged; base metal density inconsistent.	Recalculate from the qualified chart, confirm on a test strip. Density and contrast control is the quickest sanity check.
Source-to-object distance too short	Soft, blurred weld edges. Cracks and root indications smear together.	Increase the distance. Geometric unsharpness rises as SOD falls and object-to-image distance grows, so distant long exposures buy crack definition.
Film or detector not contacting the pipe	A doubled weld edge: two parallel outlines where there should be one.	Reposition cassette and screens flat against the pipe using a contact device or tape. A double image is almost always separation, not a weld defect.
Inadequate coverage	The image stops at the reinforcement, the weld cap is cropped, or is missing.	Confirm film length and positioning against joint geometry first. Coverage must extend beyond the reinforcement on the pipe plus what the specification requires.
IQI misplaced or on different thickness	IQI missing, wrong wire number visible, or sitting on thinner pipe than the weld.	Place the IQI over material of the same thickness as the weld. An IQI on different thickness invalidates the sensitivity demonstration, so the image must be rejected even if the weld looks perfect (how IQIs demonstrate sensitivity).
Dirty pipe surface	False indications of irregular shape, plus scattered densities from mill scale, rust, paint or spatter.	Clean the joint to bright metal before the shot. On aluminum this is non-negotiable, and a dedicated aluminum cleaner for welding prep routine is worth the time.
Pipe skewed or off-centre	Apparent misalignment, elliptical distortion, or a weld that looks thicker on one side.	Re-centre the pipe relative to the beam and square the joint. That distortion is a setup error, not a fabrication error.
Backscatter	General fogging away from the part, like a hazy halo around the weld image.	Use lead foil screens, a lead backing sheet, or increase the source-to-object distance.
Film handling errors	Static crown marks, fingernail scratches, chemical splash, light leak fogging along an edge.	Handle film in controlled humidity, keep cassettes clean, never touch emulsion with bare fingers. Digital systems removed most of this category, though digital radiography still has its own handling rules.
Administrative errors	No identification block or an illegible one, an out-of-date calibration certificate, a film nobody can find later.	Verify the ID block, the lead ID wire and current calibration before exposing. Under 10 CFR Part 34, survey instruments and dosimetry must be calibrated and current. A radiograph without legible identification cannot enter the permanent record.

The symptom column reads the way a Level II would describe it out loud in front of the screen, so match it against what you are actually looking at. Two entries deserve special emphasis. Radiographers check for a doubled edge first whenever an anomaly looks doubled, because separation gets misread as a weld defect so often. And IQI placement error quietly invalidates an otherwise perfect film: if the sensitivity demonstration is not valid, the exposure proves nothing, and no interpretation step rescues it.

The Prevention Framework

Rejected shots are a planning failure far more often than a welding failure, and four practices eliminate most of them. Every defect on the list has a tell that a trained eye can spot before the film is exposed, so the shot can be checked while a five-minute adjustment still costs nothing, rather than at interpretation when the only option left is stripping and redoing the joint.

• A documented, qualified procedure. Every shot traces to a written, demonstrated technique rather than operator memory: exposure charts, source-to-object distance, screen and cassette type, coverage limits, all written down before the crew leaves the shop.
• A pre-exposure checklist. Joint labelled, surface clean and dry, IQI positioned, source-to-object distance established, coverage adequate, film or detector identified, calibration certificates current, radiation survey done, witness present. Ten items, about three minutes, catching nearly every entry above.
• A properly certified radiographer. The person shooting the film should be qualified for the work in hand, with documented training and current certification. ASNT calls radiography a specialized discipline because technique decisions carry acceptance consequences.
• A pre-job briefing with the welding crew. Most often skipped, and it saves the most money. Insulating, painting or clamping over a joint that still needs radiography destroys the option to re-shoot; if a defect surfaces after coating, the only path back is stripping, gouging and re-welding, which is where carbon arc gouging for defect removal becomes a real cost. Align the crew on joint condition, access and timing before the source leaves the shop.

Certification also carries physical requirements. Radiographers need documented vision and hearing tests, because reading a fine crack in a low-contrast image and hearing a colleague over a noisy plant floor are both part of the job. A Level II also needs hands-on documented training on the equipment and techniques actually in use, not simply a certificate on a wall.

Then there is the client’s own requirement, which frequently exceeds the code minimum. Aerospace work, for instance, often demands personnel qualified to additional NASA or industry standards on top of ASNT or ISO certification, and aerospace welding standards and careers guidance is worth a read if that sector is on your radar. Check the contract and the inspection and test plan before the shot, not after the rejection.

The Safest Radiograph Is the One You Do Not Have to Repeat

Here is the practical link between safety and quality. Every reshoot means more dose, more time inside the controlled area, more handling of the source or the X-ray set, and more exposure for the crew standing nearby. A rejected image shot again on bad technique doubles everyone’s exposure and leaves two records to reconcile instead of one.

Getting the technique right the first time is a safety measure as much as a quality measure. Treat every exposure as the one you would have to defend to an inspector, a regulator, and an owner, because in most shops, somebody will.

A Practical RT Workflow, Checklist, and Final Thoughts

Everything so far has been preparation for a single moment: the shot. This section turns the guide into a repeatable process for your next job.

The Workflow, In Order

The first three steps are paperwork, and they separate a defensible radiograph from a picture.

1. Fix the governing documents. Confirm the code, the acceptance criteria, and the extent of examination before equipment is booked: how many joints, what percentage, which welders, which repair cycles.
2. Confirm the weld is ready. Check welder qualification, WPS and PQR compliance, and confirm any required post-weld heat treatment is complete. RT follows PWHT, because heat treatment can generate new indications and the examination should reflect the joint’s final condition.
3. Verify the joint is safe and accessible. Depressurize, drain, and confirm safe. The surface must be clean and free of clamps or supports that block the beam path.
4. Select the technique and source for the material and wall thickness.
5. Establish the exposure parameters from a qualified technique chart and confirm source-to-object distance against the chart before energizing.
6. Place the IQI over a section of equivalent thickness. The IQI shot must be on the same section of the joint as the weld image, not a nearby butt or another wall thickness.
7. Make the exposure and check coverage and image quality before releasing the crew. This is the most valuable habit in the process.
8. Process and evaluate the image against the correct acceptance criterion.
9. Document, retain, and file the result so the work can be verified later.

Two details deserve expansion. The wall thickness used in the acceptance criteria should be the specified nominal wall for the material grade, because porosity and slag limits scale with it, so a mill-certificate value that differs from nominal shifts every limit in the table. A report should cite the method standard and the acceptance standard, for example “RT per ASME Section V Article 2, evaluated to ASME B31.3 Table 341.3.2.” Section V Article 1 requires a qualified procedure, qualified personnel, and recorded results.

A Pre-Shot Checklist You Can Copy

#	Check	Confirm
1	Joint identified and labelled	Matches the weld map
2	Surface clean and dry	Scale, rust, paint, spatter removed
3	IQI placed	Right type, size, source side
4	Source-to-object distance set	Matches the technique chart
5	Coverage adequate	Extends past reinforcement both ways
6	Film or detector identified	ID block readable, no unlabelled frames
7	Calibration current	Source, instrument, IQI certificates in date
8	Radiation survey done	Completed before the exposure
9	Witness present	Client or inspector available to observe

Run this before every shot, and add one line for film handling, which causes more rejects than anything else: screen and cassette condition, plus controlled transport to the darkroom or digital workstation.

What Actually Drives Cost and Turnaround

Fabricators underestimate what radiography costs, and the drivers are not the length of the weld.

• Pipe diameter and wall thickness. A 36-inch line at 1.5-inch wall is a different job from a 4-inch line at 0.375-inch wall.
• Exposures per joint. Scales with circumference and multiplies setup time, IQI shots, film use, and evaluation.
• Access and scaffolding. Often the largest cost line, and the one that decides shop or field work.
• Crew size and shift. A confined-space crew on nights costs more per hour than a shop crew on days.
• Source type. A gamma source needs licensing, decay-based timing, and transport logistics; an X-ray set needs power but moves easily.
• Site versus shop. Field work carries weather, access, and shutdown time; shop work is faster and cleaner.
• Rework from one rejected image. Excavation, re-welding, re-heat treatment, re-shooting, and possibly another cycle.

That is why shooting in the shop is worth fighting for when the pipe can be fabricated there. Access is easier, a joint can be prepared and re-shot in minutes rather than hours, and a failed shutdown inspection costs far more than the premium on controlled shop radiography. For stainless and alloy work especially, filler metal and consumable control pays off, because a chemistry issue that shows up as inclusions on the film is expensive to correct after the pipe is in the rack.

Retakes and Records

Retaking an image for a better result is normal practice. Shaky geometry, a scratched film, a missed IQI shot: you shoot again. But the original must be retained and the retake reason documented. A rejection cannot be discarded and re-shot until it passes, and that is the rule an auditor looks for first.

One habit worth building in: compare as-welded radiographs against each welder’s qualification record. A pattern of small porosity in one welder’s work is a process problem, not a random one, and it keeps showing up until somebody changes gas flow, wire feed, or shielding setup. You will find more of that in all-position welding rod selection problems than anything the codes catch.

Final Thoughts

Back to the question that opened this guide: where is the radiograph? In the best case, it is a defined, indexed image tied to a joint on a weld map, evaluated against a stated criterion by a named Level II radiographer, with a retake history and a calibration certificate behind it. The weld is sound, the record is defensible, and the paperwork holds up three years later when somebody asks.

That outcome is not the product of one good exposure. It is the product of three things working together: technique, image quality, and criteria. Radiographic testing is a system, not a single shot. Get all three right and the radiograph holds up; get one wrong and you are back on the scaffold with a grinder.

Frequently Asked Questions

What is the difference between X-ray and gamma radiography for pipe welds?

Both are radiographic testing (RT) and produce the same kind of image; the difference is the source. X-ray machines accelerate electrons into a tungsten anode to generate X-rays, giving fast, controllable exposures but requiring power and practical penetration limits. Gamma sources use radioactive material such as Ir-192 or Co-60, need no power, reach thicker steel, but require licensing and decay-based exposure-time calculation.

How much steel thickness can a pipe weld X-ray inspection cover?

It depends on the equipment and the geometry. Portable X-ray sets typically work effectively up to roughly 25–40 mm (1–1.5 in.) of carbon steel, while Ir-192 gamma sources commonly handle up to about 100 mm (4 in.) and Co-60 extends to roughly 200 mm (8 in.). Thicker sections usually need a panoramic camera, computed tomography, or an in-service ultrasonic technique instead.

Is radiography required on every pipe weld?

No. The extent of examination comes from the governing code, the fluid service category, and the owner or contract specification. ASME B31.3, for example, requires a minimum of 5% random radiography or random UT for many service categories, with higher percentages for more severe service, while specific code rules require 100% radiography on certain welds such as high-energy or high-temperature service lines.

What is an IQI, and why is one required on every radiograph?

An image quality indicator (IQI), historically called a penetrameter, is a set of stepped wires or a shim with drilled holes placed in the X-ray beam path. It proves the image is sensitive enough to reveal the defects the code cares about — a call-out such as 2-4T means a shim of 2% of the wall thickness with a hole four times that thickness. Without a valid IQI demonstration, the image has no defensible sensitivity and cannot be accepted.

Can radiography detect surface cracks in a pipe weld?

Only unreliably. Radiography is strongest on volumetric defects such as porosity, slag, and lack of fusion. Tight surface-breaking cracks may show only as a broken weld edge or wire image, which is why specifications pair radiography with liquid penetrant or magnetic particle testing, and why ASME Section VIII rejects any indication that is clearly a crack once identified.

What makes a radiograph unusable, even when the weld is perfect?

Technical quality problems make the image unreadable: insufficient penetration, excessive contrast loss from over-penetration, a double image from film not in contact with the pipe, collimator cut-off leaving unexposed margins, backscatter fogging, static marks, chemical splash or scratches, and missing or illegible identification data. In those cases the image is retaken — the fix is technique and handling, not the weld.

Should I choose radiography or ultrasonic testing for pipe welds?

Radiography gives a permanent image of the full weld cross-section and is ideal for thinner pipe, forgings, and castings, but it needs access to both sides of the joint and takes longer. Ultrasonic testing detects planar defects better, works on thicker pipe with access to only one side, and gives near-instant results, but it is harder to qualify and interpret. Many codes accept random UT in place of random RT for exactly this reason.

Who is allowed to shoot an X-ray of a pipe weld, and what does it cost?

The radiographer must hold the required certification — Level II for performing examinations under an approved procedure, Level III for authorising them — under a scheme such as ASNT SNT-TC-1A, ASNT CP-189, or ISO 9712. Gamma radiography in the US additionally requires an NRC licence under 10 CFR Part 34 with named qualified radiographers. Cost scales with pipe diameter and wall thickness, the number of exposures, access and scaffolding, crew size, and site versus shop work.

How long are pipe weld radiographs kept, and can a failed one be re-shot?

Retakes are allowed and routine, but the original image and the reason for the retake must be retained and documented — you cannot discard a rejection and shoot again until it passes. Retention of the accepted radiograph and the associated report is typically for the life of the component or as specified by the owner, the code, or the contract, so the records package should be treated as a permanent deliverable.