Radiographic testing (RT) passes X-rays or gamma rays through a weld and records the shadow image on film or a digital detector. Where the weld is missing metal, a void, a slag pocket, or porosity, more radiation gets through and that spot reads darker on the image. The radiographer reads those density differences against the acceptance rules in the governing code to call the weld accept or reject.
RT is the volumetric NDT method most people picture when they hear “X-ray the weld.” It produces a permanent image anyone can re-examine years later, and it is excellent at finding porosity, slag, and incomplete penetration. It involves ionizing radiation, so it is done only by certified radiographers inside a controlled exclusion zone, never as a DIY check. The governing documents are ASME Boiler and Pressure Vessel Code Section V, Article 2 (Radiographic Examination) for the technique and most pressure work, and AWS D1.1 Structural Welding Code for steel structures, with the acceptance criteria set by whichever code your job is built to.
How Radiographic Testing of Welds Works
Radiation comes from one of two sources. An X-ray tube generates radiation electrically and can be switched off, which makes it the safer choice when shop logistics allow. Gamma radiography uses a sealed radioactive source, most often Iridium-192 or Cobalt-60, that is always emitting and is shielded in a “camera” until it is cranked out through a guide tube. Gamma sources are portable and need no power, so they dominate field radiography on pipelines and structures where dragging an X-ray tube and generator is impractical.
The source sits on one side of the weld. A film cassette or digital detector sits on the other. Radiation travels through the joint, and the amount that reaches the film at any point depends on how much metal it had to pass through. A void, a gas pocket, or a slag inclusion is less dense than solid weld metal, so more radiation gets through there and that area of the film comes out darker after processing. Solid sound metal blocks more radiation and reads lighter.
The film image is a two-dimensional shadow of a three-dimensional weld. That is why orientation matters so much, which comes up again under defect interpretation below.
Film, Computed, and Digital Radiography
Traditional film radiography still sets the benchmark for image quality and is what many older code editions assumed. The film is exposed, chemically processed, and read on a light box (a viewer with controlled brightness) by the radiographer.
Computed radiography (CR) uses a reusable phosphor imaging plate that gets scanned into a digital image after exposure. Digital radiography (DR) uses a flat-panel detector that produces an image directly, no plate handling. CR and DR cut out wet chemical processing, give near-real-time results, and let the reader adjust contrast and zoom on screen. Both are accepted under ASME Section V when the procedure and image-quality requirements of the applicable article are met. The physics of the shadow image is identical across all three.
What the IQI (Penetrameter) Proves
An image quality indicator (IQI), older shops still call it a penetrameter, is a small test piece of known dimensions placed on the part during the exposure. It does not inspect the weld. It proves the radiograph itself is good enough to trust. If the IQI is visible on the film at the required sensitivity, the technique resolved a known small feature, so it could resolve a real defect of similar size. If the required IQI feature cannot be seen, the radiograph is rejected and reshot regardless of what the weld looks like.
There are two common IQI styles under ASME Section V, Article 2:
| IQI Type | What It Is | How Sensitivity Is Read |
|---|---|---|
| Hole-type (plaque) | A thin rectangular plaque with three drilled holes (1T, 2T, 4T, where T is the plaque thickness) | The required hole on the required plaque must be visible |
| Wire-type | A set of parallel wires of graduated diameters in a holder | The required wire diameter must be visible across the area of interest |
The IQI is normally placed on the source side of the part, on a shim that matches the weld reinforcement so it represents the actual thickness being shot. The code ties the required IQI and the readable hole or wire to the material thickness. The takeaway for a welder reading a film report: the IQI line on the report is the proof the image was sharp enough to judge your weld fairly.
Reading the Film: What Common Defects Look Like
A radiograph reader matches the shape, density, and location of dark regions to defect types. This is interpretation, not guessing, and it is why the radiographer’s qualification matters.
| Indication on Film | Appearance | Likely Defect |
|---|---|---|
| Rounded dark spots | Distinct circular or oval dark dots, scattered or clustered | Porosity (gas pockets) |
| Irregular dark patches | Dark areas with ragged, non-uniform edges | Slag inclusions |
| Dark line down the weld centerline | Straight, fairly uniform dark band at the root | Incomplete (lack of) penetration |
| Dark line along the weld edge | Dark line following the toe or root edge | Undercut |
| Fine dark line, often branching | Sharp thin line, may be hard to see if not aligned to the beam | Crack |
The weakness of RT lives in that last row. A crack or a lack-of-fusion defect shows up only when the gap it represents lies close to parallel with the radiation beam, so the beam travels along the open flaw and finds a measurable difference in metal. A tight crack lying across the beam can pass radiography and still be there. When the failure mode you care about is cracking or sidewall lack of fusion, ultrasonic testing is the better tool, covered in detail in the ultrasonic weld testing guide. Many critical jobs run both: RT for volumetric flaws, UT for planar ones.
Why a Weld Gets Rejected on RT
A rejection is not the radiographer’s opinion that your weld is bad. It is a finding that an indication on the film exceeds a limit written in the code your job is built to. The acceptance limits differ by code and by service:
- AWS D1.1 sets visual and RT acceptance limits for structural steel, with separate, tighter limits for cyclically loaded (fatigue) connections than for statically loaded ones.
- ASME Section VIII and similar pressure codes reference the examination method in Section V and apply their own acceptance limits for porosity, slag, and other indications.
- Pipeline work commonly runs to API 1104, which has its own RT acceptance criteria and indication-length rules.
Across all of them, a confirmed crack is rejectable at any length. That rule does not vary. Beyond cracks, the typical rejectable findings are porosity over a size or cluster limit, slag inclusions over a length limit, incomplete penetration, and incomplete fusion. The film report tells you the defect type, where it is along the weld, and the clause it was judged against.
What RT does not do is tell you the part is fit for service. Passing radiography means no rejectable indication was found on those shots, within the resolution the IQI proved, on the views that were taken. It is not a guarantee of weld fitness. Whether a flagged weld gets repaired, accepted by engineering analysis, or scrapped is a disposition decision for the engineer of record working from the governing code’s repair provisions. If you weld code work, expect a Certified Welding Inspector and the EOR, not the welder, to own that call.
Where RT Fits With the Other NDT Methods
Radiography is one of five common weld inspection methods, and on most jobs it is not the first one run. Visual inspection (VT) happens on 100% of welds before any RT is even ordered, because a weld with visible surface defects gets fixed before anyone spends money on film. Surface and near-surface methods, dye penetrant testing for any non-porous material and magnetic particle testing for ferromagnetic steel, catch surface-breaking flaws that RT may not resolve well. RT and UT are the two volumetric methods that look inside the joint.
| Factor | Radiographic Testing (RT) | Ultrasonic Testing (UT) |
|---|---|---|
| Volumetric defects (porosity, slag) | Excellent | Moderate |
| Planar defects (cracks, lack of fusion) | Poor to moderate, orientation dependent | Excellent |
| Permanent record | Yes, the film or digital image | Limited, unless encoded phased-array scans are saved |
| Safety on an active site | Radiation, requires a cleared exclusion zone | No radiation, no exclusion zone |
| Access needed | Both sides of the joint (source one side, film the other) | One side, scanning the base metal beside the weld |
| Speed and cost per weld | Slower, film and source handling add cost | Faster, real-time results, lower cost per weld |
The practical reason RT still wins certain jobs: the permanent image. A film or saved digital radiograph can be pulled and re-read by a third party long after the welder has gone home, which is exactly what an owner or a regulator wants on pressure equipment. Where the deciding factor is finding cracks fast on a live job without clearing the area, UT, especially phased array, is taking over.
Radiation Safety Is Not Optional
Industrial radiography is regulated, licensed work. The source, whether an X-ray tube or a sealed gamma emitter, puts out ionizing radiation that causes real harm at field intensities, and the exposure is silent and invisible. Certified radiographers control the work under personnel qualification programs such as ASNT SNT-TC-1A, and the operation runs inside a posted exclusion zone established with survey meters, with the crew wearing dosimetry.
For everyone else on the job site, the rule is simple. When you see radiography barricades, warning signs, or hear the survey alarm, that boundary is a stay-out line until the radiographer clears it. Do not walk a shortcut through a shoot. This is general reference information and not a substitute for radiation safety training, and any radiography on your site should be run by qualified, licensed personnel.
Common Misunderstandings About RT
Thinking a passed film means the weld is guaranteed good. It means no rejectable indication was found on the views taken, at the resolution the IQI proved. Tight cracks across the beam and flaws outside the shot coverage can still exist.
Assuming RT finds cracks reliably. It does not. Crack and lack-of-fusion detection on RT is orientation dependent. If those are the defects you fear, specify UT.
Treating the radiographer’s report as the disposition. The report says accept or reject against a clause. It does not authorize a repair or an engineering acceptance. That is the EOR’s call under the governing code.
Believing digital RT is less accurate than film. Properly set up CR and DR meet the same image-quality requirements as film under ASME Section V. The image quality is verified by the same IQI logic.
Forgetting the IQI is the proof of the test. If the required hole or wire is not visible on the radiograph, the image is rejected and reshot no matter how clean the weld looks. Always check that the report confirms the IQI was met.
Acceptance and repair of code work are decisions for the governing code and the engineer of record, not the welder, so confirm the exact criteria your job runs to before you assume a weld is in or out. For the surrounding methods, return to the weld inspection category or the welding techniques pillar for the full set of inspection guides.