Industrial X-Ray System

What is an Industrial X-Ray System

An industrial X-ray system is the equipment chain that produces a radiographic image of a part's interior. A source emits penetrating radiation, the part attenuates that radiation differentially according to its local thickness and density, and a detector on the far side records the surviving intensity as a shadow image. Dense or thick regions absorb more radiation and appear lighter; voids, cracks, and porosity absorb less and show as darker indications. This thickness-to-image mapping is what lets an inspector see internal flaws that surface methods cannot reach.

Radiographic testing (RT) is one of the four classical volumetric and surface NDT methods used across welding, foundry, aerospace, electronics, and energy industries. Unlike ultrasonic testing, which is also volumetric but presents A-scan or B-scan signal data, radiography produces a directly interpretable two-dimensional image that captures planar and volumetric defects in a single view, which is why it remains the reference method for weld qualification and casting acceptance in most pressure-equipment codes.

A complete system has three functional blocks. First, the radiation source: an electrically powered X-ray tube or linear accelerator, or a sealed radioactive isotope for gamma radiography. Second, the imaging medium: silver-halide film, a reusable computed radiography (CR) imaging plate, or a flat-panel digital detector array (DDA). Third, the supporting infrastructure: a shielded cabinet or controlled area, part manipulation and positioning, a generator and controller, and image acquisition and analysis software. In a CT system the manipulator becomes a precision rotary stage and the software performs three-dimensional reconstruction.

The physics dates to 1895, when Wilhelm Conrad Roentgen discovered X-rays and produced the first radiograph, work that earned the first Nobel Prize in Physics in 1901. Industrial radiography grew through the twentieth century on film, with gamma sources such as Cobalt-60 and Iridium-192 entering field use after the 1950s. The modern transformation came from digital detectors: computed radiography plates in the 1980s and 1990s, then amorphous-silicon flat panels, and from microfocus and nanofocus tubes that enabled industrial computed tomography. Phoenix introduced the first industrial nanofocus X-ray tube in 2001, opening sub-micrometre inspection of microelectronics.

Four engineering metrics frame the quality of any industrial X-ray result: penetration (can the beam get through the part), contrast sensitivity (can it distinguish a small density change), spatial resolution or unsharpness (can it resolve a small geometric feature), and throughput (how fast each part is imaged). These four trade against each other. Higher energy buys penetration but loses contrast; a smaller focal spot buys sharpness but loses power and speed. Selection is the discipline of matching these trades to the part and the acceptance standard, not chasing a single headline number.

Radiography Method Types

Industrial X-ray work splits along two axes: how the image is captured (film versus computed radiography versus digital radiography) and how the part is viewed (single-projection 2D versus full 3D computed tomography). Choosing the wrong capture medium is a common and expensive mistake, because it locks in throughput, resolution, and recurring consumable cost for the life of the asset. The table below compares the four mainstream imaging approaches.

Film radiography remains the historical reference. A fine-grain industrial film records the latent image, which is then chemically developed in a darkroom. Film delivers the highest spatial resolution of any medium and an archival hard record, which is why some codes still default to it for critical welds. Its drawbacks are slow turnaround, consumable cost, chemical handling, and the difficulty of digitising and searching an archive. ISO 17636-1 and ASTM E1742 set the technique rules for film weld and casting radiography.

Computed radiography replaces film with a reusable photostimulable phosphor imaging plate. After exposure the plate is read by a laser scanner that releases the stored energy as light, producing a digital image, and the plate is then erased and reused. CR is the lowest-capital path off film, fits existing exposure geometries, and removes wet chemistry, but it carries an extra scanning step and its spatial resolution is below both film and good DR. It is governed under ISO 17636-2 alongside DR.

Digital radiography with a flat-panel digital detector array captures the image directly in seconds, with no plate handling or scanning. DR gives the best signal-to-noise ratio and image quality indicator sensitivity per exposure, supports immediate review and image processing, and is the basis of automated and in-line inspection. The trade is the highest hardware cost and a native spatial resolution coarser than fine-grain film, set by the detector pixel pitch.

Computed tomography is not a separate capture medium but a method that rotates the part through hundreds of angular projections and reconstructs a full three-dimensional voxel volume. CT is the only radiographic method that yields true internal geometry, defect volume, and dimensional metrology of features no probe can reach. Microfocus and nanofocus CT systems reach resolutions from a few micrometres down to 0.2 micrometre, at the cost of long scan times and high capital expense, so CT is reserved for high-value parts rather than high-volume screening.

X-Ray and Gamma Source Technologies

The radiation source defines penetration, exposure time, and image sharpness more than any other component. Three families dominate: electrically powered X-ray tubes for the bulk of inspection, high-energy linear accelerators and betatrons for thick steel, and sealed gamma isotope sources for field and access-limited work. The table below maps the practical steel-penetration envelope of each, which is the first filter in any source decision.

X-ray tubes generate radiation by accelerating electrons across a vacuum gap and striking a metal anode, releasing bremsstrahlung X-rays. Output stops instantly when power is removed, and tube voltage (kV) sets the spectrum and penetration while tube current (mA) sets the intensity. Penetration roughly scales with voltage: about 25 mm of steel at 160 kV, 40 mm at 225 kV, 60 to 80 mm at 320 kV, and up to 90 to 100 mm at 450 kV. As a rule, use the lowest kV that still achieves full penetration, because lower energy produces higher image contrast and better defect sensitivity.

Tubes are further classed by focal-spot size. A standard minifocus directional tube has a focal spot of roughly 0.4 to 5.5 mm sized per EN 12543 and ISO 32543; the larger the spot, the more anode power and penetration it tolerates. A microfocus tube reaches a few micrometres and a nanofocus tube reaches 0.2 micrometre, which permits high geometric magnification by placing a small part close to the source. This magnification is the foundation of CT and electronics inspection, but small spots accept far less power, so they image thin or low-density parts only.

Linear accelerators and betatrons take over where tubes run out of energy, above roughly 100 mm of steel. Both accelerate electrons to megavolt energies to generate a hard X-ray spectrum. A 6 MeV system penetrates about 50 to 300 mm of steel and 9 MeV systems reach 300 to 500 mm, covering heavy castings, thick weldments, and rocket motors. Betatrons are more compact and lower in power draw, with 2 to 9 MeV models adjustable in roughly 0.1 MeV steps, which makes them the portable choice; linear accelerators offer higher output but cost more and need a dedicated shielded vault.

Gamma sources use a sealed radioactive isotope that emits continuously and cannot be switched off, only shielded inside an exposure device. Iridium-192 is the workhorse for field weld and pipe radiography, with discrete gamma energies near 0.3 to 0.6 MeV (comparable to a 460 kV X-ray set) and a 74-day half-life, covering 10 to 90 mm of steel per ISO 5579. Selenium-75 gives lower energy and better contrast on thin sections (5 to 40 mm). Cobalt-60 emits 1.17 and 1.33 MeV gammas with a 5.3-year half-life for thick sections, comparable to a 1.25 MeV X-ray system. Gamma needs no power and reaches tight access, but the source decays, must be replaced, and carries continuous security and licensing duties.

Detectors, IQI and Standards

The detector and the way image quality is verified determine whether a radiograph is acceptable evidence under a code, not just a pretty picture. On the digital side, the flat-panel digital detector array (DDA) is the dominant capture device. Most NDT DDAs are indirect: a scintillator layer of cesium iodide (CsI) or gadolinium oxysulfide (GOS) converts X-ray or gamma photons into visible light, which an amorphous-silicon thin-film-transistor photodiode array reads out pixel by pixel.

The headline detector spec is pixel pitch, the centre-to-centre spacing of pixels, commonly 50, 75, 100, or 200 micrometres for NDT panels. Smaller pitch gives finer detail but a larger file and lower light per pixel. Note that the panel's actual basic spatial resolution is typically somewhat coarser than the nominal pixel pitch, because scintillator light spread and electronics blur it; the true figure must be measured with a duplex wire gauge per ISO 19232-5, not read from the pitch alone. Detection areas range from compact panels to large 410 by 410 mm arrays for full weld coverage.

Image quality is proven with an image quality indicator (IQI), also called a penetrameter, a calibrated artifact placed on the part to confirm the radiograph reveals defects of a required minimum size. Two families exist. Wire IQIs (ISO 19232-1, ASTM E747) lay a set of graded wires across the part, and the thinnest wire visible sets the sensitivity. Hole or plaque IQIs (ASTM E1025) use a stepped plaque with drilled holes referenced as 2-1T, 2-2T, or 2-4T, where the number is plaque thickness as a percent of part thickness and the hole size is a multiple of plaque thickness. Sensitivity is read as a percentage, and lower is better; good radiography reaches 1 to 2 percent.

The contrast sensitivity that a digital system can achieve depends on three measurable essentials: the basic spatial resolution, the signal-to-noise ratio (SNR), and the specific contrast. Because film has very high spatial resolution but DDAs have a stronger SNR, a digital detector can often match or beat film IQI visibility at a much shorter exposure, which is the practical argument for moving a film shop to DR. Standards bodies set minimum SNR and unsharpness values that a digital technique must demonstrate before it is accepted as film-equivalent.

The standards landscape is layered, and a buyer should know which document governs the work. The table below lists the most-cited radiographic standards and what each controls, so a purchase specification can name the right document rather than a vague reference to industry practice.

Key Specification Parameters

Reading an X-ray system data sheet is a core skill for the buyer. Vendors list many figures, but only a handful drive a fit-for-purpose decision: tube voltage and current, focal-spot size, detector pixel pitch and area, achievable spatial resolution and contrast sensitivity, source-to-detector geometry and magnification, duty cycle, and shielding or safety class. Each is decoded below.

Tube voltage and current set penetration and exposure time. Voltage (kV) defines the spectrum and the maximum steel thickness the system can image; current (mA) defines photon flux, so higher mA shortens exposure but loads the anode. A 320 kV at 10 mA generator is a different machine from a 320 kV at 3 mA microfocus head even at the same voltage, because power equals voltage times current and determines both speed and the largest tolerable focal spot. Always read voltage and current together, not voltage alone.

Focal-spot size governs geometric unsharpness, the penumbra blur at the edge of every feature. Unsharpness grows with focal-spot size and with the object-to-detector distance relative to source-to-object distance, so a small spot and good geometry are needed for fine detail. Standard tubes list 0.4 to 5.5 mm spots; microfocus and nanofocus tubes list micrometre values. The figure should be stated per EN 12543 or ISO 32543, which define how the spot is measured, because vendors otherwise quote it inconsistently.

Detector pixel pitch and area set native resolution and field of view, while the achieved spatial resolution and contrast sensitivity are the figures that actually pass or fail a code. Demand the measured basic spatial resolution by duplex wire gauge (ISO 19232-5) and the IQI sensitivity, not the pixel pitch. A 100 micrometre panel may deliver a measured basic spatial resolution noticeably coarser than 100 micrometres once scintillator blur is counted.

Geometry and magnification matter most for CT and electronics. Placing a small part close to a microfocus source magnifies it onto the detector, so the effective resolution can far exceed the detector pitch, limited instead by the focal spot. Data sheets state maximum geometric magnification, minimum voxel size for CT, and the manipulator's stage accuracy. The other practical figures are listed below.

• Duty cycle: the fraction of time the generator can run at rated output. Production systems specify 100 percent duty cycle for continuous shift work; lab sets may be lower.
• Anode dissipation and cooling: rated anode power and whether cooling is air, oil, or water, which caps sustained current and tube life.
• Detector dynamic range and frame rate: bit depth (often 14 to 16 bit) and frames per second, which set CT scan speed and contrast.
• Shielding and safety class: enclosed cabinet rated for unrestricted operation versus open-beam requiring a controlled area, with leakage dose stated per the regulator.
• Software and analysis: defect recognition, porosity and wall-thickness analysis, ADR or AI assistance, and CT reconstruction and metrology modules.

A final caution on terminology: vendors sometimes blur spatial resolution, contrast sensitivity, and IQI sensitivity into a single marketing claim. They are independent properties. A system can have excellent contrast sensitivity yet modest spatial resolution, or the reverse. For a code-governed purchase, require all three as separate measured numbers tied to the relevant standard.

Selection Decision Factors

To turn the previous five chapters into a specific machine, follow the decision sequence below. Most selection mistakes come not from a single wrong figure but from settling capture medium or energy before the part and the acceptance standard are pinned down. These steps work as a fixed RFQ template.

1. Part and material first: define the maximum and minimum material thickness, the material (steel, aluminium, composite, electronics), and the smallest defect that must be detected. These three set the floor for energy, focal spot, and resolution before any vendor is contacted.
2. Acceptance standard and quality level: name the governing code, such as ISO 17636-1 or 17636-2, ASME Section V Article 2, or a customer specification, and the required IQI sensitivity. The code dictates technique, IQI type, and minimum image quality, and therefore the detector and geometry.
3. Source type and energy: choose X-ray tube, high-energy accelerator or betatron, or gamma source from the penetration table in Chapter 3. Pick the lowest energy that fully penetrates the thickest section, since lower energy gives higher contrast.
4. Capture medium: decide film, CR, or DR against throughput, archival needs, and capital budget. High-volume or in-line work points to DR; a low-capital move off film points to CR; the most critical archival welds may still specify film.
5. 2D versus CT: decide whether a projection image suffices or the application needs three-dimensional geometry, defect volume, or internal metrology. Reserve CT for high-value parts; do not buy CT throughput penalties for simple pass-fail screening.
6. Throughput and automation: estimate parts per hour, manipulation and part handling, and whether automated defect recognition is required. This separates a manual cabinet from an automated in-line cell and drives most of the system cost.
7. Radiation safety and siting: confirm cabinet versus open-beam, available floor and shielding, regulatory registration or licensing, interlocks, survey meters, and dosimetry. For gamma, add source security, leak testing, and transport licensing.
8. Total cost of ownership (TCO): capital plus installation and shielding, tube or source replacement, detector replacement, calibration, consumables for film or CR, software maintenance, and operator certification. A low purchase price with short tube life and heavy film consumption can exceed a digital system within a few years.

One dimension that buyers underweight is manufacturer serviceability and certification support: local field service and calibration, tube and detector lead times, documented compliance to the named standard, software update path, and operator training to ISO 9712 or SNT-TC-1A. A radiographic system is a multi-year capital asset whose uptime depends on parts availability and trained personnel long after purchase. Established suppliers such as Waygate Technologies (Seifert and Phoenix), Comet YXLON, Nikon Metrology, ZEISS, North Star Imaging, and Varex Imaging maintain global service and detector supply, which is decisive for production-critical lines.