An eddy current array (ECA) is a non-destructive testing (NDT) probe that bundles multiple small sensing coils into a single housing, allowing large surface areas to be scanned in a single pass rather than one tiny point at a time. The technology dramatically speeds up inspections of metal components while maintaining high sensitivity to cracks, corrosion, and other flaws. ECA probes are now standard equipment in aerospace maintenance, nuclear power plant inspections, and structural weld evaluation, and the technology continues to evolve with flexible sensors, advanced imaging software, and hybrid instrument platforms.
How an Eddy Current Array Differs From a Single Probe
Traditional eddy current testing uses a single coil that must be raster-scanned back and forth across a surface in tight, overlapping passes to ensure full coverage. Missing a strip means missing a crack. An eddy current array replaces that single coil with rows of individually driven elements packed side by side. Each element generates its own localized eddy current field and measures the impedance response independently. Because many elements cover a wide swath simultaneously, the probe can follow a much coarser scanning pattern while still catching small defects, simply because so many sensors are listening at once.1NDT & E International. An eddy current array instrument for application on ferritic welds
The practical payoff is speed and reliability. An inspector using a single-element probe on a long weld seam might spend hours carefully indexing back and forth. An array probe covering the same width in one sweep cuts inspection time to a fraction of that. And because every element independently reports its signal, the data can be stitched together into a two-dimensional image of the surface, much like how a flatbed scanner builds an image line by line. That imaging capability is a major reason ECA has displaced older point-by-point methods in many industries.
Multiplexing and Signal Extraction
Running dozens or even hundreds of coils simultaneously would create electromagnetic crosstalk between neighboring elements, mudding the signals. Array instruments solve this through multiplexing, which activates subsets of coils in a rapid sequence so that nearby elements are never energized at the same moment. Early array designs used a transient excitation method to extract coil impedance parameters quickly and digitally, minimizing the analog front-end electronics needed per element. That approach was a practical breakthrough: fewer components per channel meant that scaling up to large element counts was feasible without the instrument becoming unwieldy.2NDT & E International. An eddy current array instrument for application on ferritic welds
Modern systems have pushed this further. Instruments now drive array probes at multiple frequencies simultaneously, collecting axial and transverse datasets in a single scan. One automated inspection system, for instance, acquired data at four transmission frequencies (250, 300, 400, and 450 kHz) at the same time, generating multiple impedance C-scan images from a single pass.3PubMed Central. Automated Real-Time Eddy Current Array Inspection of Nuclear Assets Different frequencies penetrate to different depths, so running several at once lets inspectors see both shallow surface defects and slightly deeper flaws without re-scanning.
Where Eddy Current Arrays Are Used
ECA technology found its footing in industries where missing a crack can be catastrophic and where the components being inspected are large, curved, or otherwise difficult to scan with a single probe.
Aerospace
Aircraft skins are held together by thousands of rivets, and fatigue cracks love to nucleate at rivet holes. Inspecting every fastener with a single-element probe is painstaking. Array probes sized to straddle a row of rivets can sweep an entire joint in one pass, flagging both cracks emanating from fastener holes and hidden corrosion between skin layers. Inspection of riveted fuselage and wing joints on training aircraft, for example, has been carried out using eddy current array methods to detect cracks and reveal hidden corrosion beneath the surface.4Transportation Research Procedia. Eddy Current Array Inspection of Zlin 142 Fuselage Riveted Joints The method is especially valuable in aging fleet management, where operators need to survey hundreds of joints on a single airframe during scheduled maintenance windows.
Nuclear Power
Steam generators in nuclear reactors contain thousands of thin-walled tubes that transfer heat from the primary coolant loop. Stress-corrosion cracking on the outer surfaces of these tubes is a well-known failure mode. In one case, a CANDU reactor was shut down for over a year after tube failures caused by outer-diameter stress-corrosion cracking in the U-bend section. Novel single-pass eddy current transmit-receive array probes (designated C3) successfully detected all significant cracks, allowing the damaged tubes to be plugged and the unit brought back online.5Proceedings of the fourth international conference on CANDU maintenance. Validating eddy current array probes for inspecting steam generator tubes When a reactor outage costs millions of dollars per day, the speed advantage of a single-pass array probe over a conventional bobbin coil is enormous.
Structural Welding
Welded steel structures in bridges, pipelines, pressure vessels, and offshore platforms require periodic crack inspection. The heat-affected zone along a weld is a prime location for fatigue cracks, and the surface there is often rough and geometrically uneven. ECA probes originally developed for ferritic welds demonstrated that deploying many elements across the heat-affected zone maintained a high probability of detection even with a coarser mechanical scan.6NDT & E International. An eddy current array instrument for application on ferritic welds
Detecting Cracks in Rough and Irregular Surfaces
One of the hardest challenges for any eddy current method is dealing with surface geometry that creates signal noise. Weld surfaces are a prime example: the bead itself is lumpy, and the transition zone between the weld and base metal produces geometric signals that can mask small cracks. Crater cracks, which form at the termination points of welds due to uneven cooling and stress concentration, are especially tricky because they tend to be small, multidirectional, and sometimes buried beneath or blended into the surface deformation.
Flexible eddy current array (FECA) sensors have shown strong results in this environment. In testing on steel structures with natural weld roughness, a FECA sensor system successfully captured abnormal signals at crater crack locations despite the background noise from the uneven surface. The crack-related signal peaks were clearly distinguishable from surface-roughness fluctuations, thanks to the high spatial resolution that comes from having many closely spaced sensing elements.7Communications Engineering. Flexible eddy current array measurement system for crack detection in weld zones of steel structures The ability to conform to a curved or rough surface while maintaining stable coil-to-surface distance across all elements is what makes flexible arrays competitive in situations where rigid probes would rock or skip, creating false signals.
Subsurface Corrosion and Penetration Depth
Standard eddy current testing is inherently a near-surface technique. Eddy currents decay exponentially with depth, and higher operating frequencies concentrate the currents closer to the surface. For surface-breaking cracks, that works perfectly. But many industrial problems involve hidden damage: corrosion growing on the far side of an aircraft skin, wall thinning inside a pipe, or material loss between layered structures.
Pulsed eddy current methods, which use a broad-spectrum excitation pulse instead of a single continuous frequency, push the effective inspection depth deeper because the low-frequency components in the pulse penetrate further into the material. Gradient-field pulsed eddy current (GPEC) probes have been developed specifically for imaging subsurface corrosion. By generating a uniform excitation field, the probe ensures that any measured signal perturbation corresponds to the corrosion itself rather than to edge effects or geometry. Researchers have shown that this approach provides high-sensitivity imaging of subsurface corrosion profiles and opening sizes in conductive structures.8PubMed Central. Imaging of Subsurface Corrosion Using Gradient-Field Pulsed Eddy Current Probes with Uniform Field Excitation The combination of pulsed excitation with array-format probes is an active area of development, aiming to bring that depth capability together with the speed and imaging advantages of multiple elements.
The Lift-Off Problem
One persistent headache in eddy current testing, whether single-coil or array, is lift-off: the gap between the probe and the metal surface. Even a fraction of a millimeter of paint, coating, or air gap changes the signal, and on a real component that gap is rarely constant. If you cannot separate the effect of the coating thickness from the effect of a defect, you risk either missing real flaws or calling out false ones.
Compensation techniques address this by combining differential and absolute coil measurements with signal-processing algorithms. One method couples differential and absolute probe outputs through a fuzzy-logic correction scheme and has demonstrated that lift-off error can be reduced to within about 0.1% of full-scale value across coating thicknesses ranging from roughly 1 mm to 5 mm.9PubMed Central. A Novel Eddy Current Testing Error Compensation Technique Based on Mamdani-Type Fuzzy Coupled Differential and Absolute Probes For array probes, lift-off compensation is even more important because each element in the array may sit at a slightly different height on a curved or rough surface. Without per-element correction, the resulting image would show a topographic map of the surface geometry rather than a defect map. Modern ECA instruments apply lift-off correction in real time as the data streams in from each channel.
Flexible and Miniaturized Sensor Designs
Traditional ECA probes are rigid assemblies, which works fine on flat plates or gently curved pipe exteriors but becomes limiting on complex geometries like turbine blade roots, fillet welds, or the inside surfaces of small-diameter tubes. The push toward flexible arrays has produced sensors built on polymer substrates that can wrap around curves or press into tight concavities.
A particularly interesting development is the integration of tunnel magnetoresistance (TMR) sensing elements into flexible eddy current probes. Unlike a conventional pickup coil, which measures inductance changes, a TMR sensor detects the magnetic field directly and can be made extremely small. A flexible eddy current TMR sensor has been proposed for monitoring internal fatigue cracks in metal joint structures, targeting the kind of hidden damage that develops deep within bolted or riveted assemblies during service.10PubMed Central. A Flexible Eddy Current TMR Sensor for Monitoring Internal Fatigue Crack Because TMR elements are tiny, many of them can be packed into a small area, pushing spatial resolution higher than what conventional wound coils can achieve.
On the electronics side, miniaturization is advancing through direct digitization techniques that eliminate the bulky analog signal-conditioning stages traditionally required between the sensor and the data acquisition system. Conventional pulsed eddy current setups require extensive front-end circuitry to amplify, filter, and convert analog coil signals, which makes the instrument heavy and power-hungry. A direct-interfacing technique that digitizes sensor signals without that intermediate circuitry has been demonstrated for pulsed eddy current thickness measurement of aluminum sheets, offering a path toward lighter, cheaper, and more portable array instruments.11IOP Publishing (Measurement Science and Technology). Directly-digitized pulsed eddy current based quantification of aluminum sheet thickness
Imaging and C-Scan Maps
One of the biggest reasons ECA has gained traction over older single-probe methods is the imaging output. When each array element records its signal as a function of position along the scan direction, the combined dataset can be rendered as a two-dimensional color map called a C-scan. Bright spots or color shifts correspond to impedance changes caused by defects, and an experienced analyst can read the shape, size, and orientation of a crack directly from the image.
Automated real-time ECA inspection systems now generate these C-scan images live as the probe moves across the surface. By acquiring data at multiple frequencies simultaneously, the system produces several overlaid images, each optimized for a different depth range or defect type.12PubMed Central. Automated Real-Time Eddy Current Array Inspection of Nuclear Assets An operator watching the screen sees surface-breaking cracks pop out in the high-frequency image while deeper corrosion thinning shows up more clearly in the low-frequency image. This multi-frequency imaging is analogous to how medical imaging uses different contrast settings to highlight different tissue types.
The imaging capability also creates a permanent, auditable record. Regulatory bodies in the nuclear and aerospace industries increasingly require that inspection data be stored and reviewable, not just noted on a paper form. A C-scan file tied to a GPS-tagged position on the component gives future inspectors a baseline to compare against, making it possible to track how a known indication evolves over time.
Combining Eddy Current Arrays With Ultrasonic Testing
No single NDT method catches everything. Eddy current arrays excel at finding surface and near-surface cracks and corrosion thinning, but they lose sensitivity beyond a few millimeters of depth in most materials. Ultrasonic testing, particularly phased-array ultrasonic testing (PAUT), penetrates deep into the material volume and can locate internal voids, inclusions, and delaminations that eddy currents simply cannot reach.
Researchers have been developing integrated inspection platforms that house both an eddy current array and a phased-array ultrasonic transducer in a single instrument, fusing the two data streams. The logic is straightforward: the eddy current array handles the surface-to-near-surface zone rapidly and with high resolution, while the ultrasonic array covers the deeper material volume. Together they provide what has been described as a “surface plus internal” volumetric inspection, addressing the blind spots that either method would have alone.13International Journal of Applied Electromagnetics and Mechanics. High efficient nondestructive testing based on instrument integration and signal fusion of array eddy current testing and phased array ultrasonic testing method
This kind of fusion is especially relevant for thick-walled components like pressure vessels or heavy structural steel, where surface cracks and internal hydrogen-induced cracking can coexist. Running both methods in a single scan pass, rather than deploying two separate teams with two separate instruments, cuts inspection time and eliminates the positioning errors that come from trying to register two independent datasets after the fact.
Common Misconceptions About ECA Limitations
People sometimes assume that eddy current testing only works on non-ferromagnetic metals like aluminum and stainless steel. It is true that ferromagnetic materials (carbon steel, for instance) complicate eddy current signals because the material’s own magnetic permeability creates large background signals that can dwarf defect responses. But array instruments designed for ferritic specimens have been in use since the early development of the technology, using techniques like magnetic saturation or specialized signal processing to suppress the permeability effect.14NDT & E International. An eddy current array instrument for application on ferritic welds Carbon steel welds, pipeline walls, and reactor vessel surfaces are all inspectable with appropriately designed ECA probes.
Another misconception is that ECA is a replacement for all other surface inspection methods, including magnetic particle testing and dye penetrant testing. In practice, ECA occupies a different niche. It does not require consumables, works through thin coatings, and produces digital records, but it demands a trained operator, relatively expensive equipment, and careful calibration. Magnetic particle testing remains faster and cheaper for simple go/no-go surface crack detection on ferromagnetic parts, particularly in field conditions where portability matters more than imaging. ECA earns its place when you need speed over large areas, permanent digital records, sensitivity to subsurface flaws, or the ability to inspect through coatings without stripping them.
Calibration Standards and Reference Blocks
Before any ECA inspection, the instrument must be calibrated against a reference standard containing machined notches or electric-discharge-machined (EDM) slots of known dimensions. The operator runs the array over these artificial flaws and adjusts gain, phase rotation, and threshold settings so that the instrument produces a clear, repeatable signal from a defect of the minimum reportable size. This step is critical because different materials, thicknesses, and surface conditions all change the baseline impedance, and without calibration the operator has no way to interpret the C-scan image.
In regulated industries, these reference standards are themselves controlled documents. Aerospace maintenance programs specify the notch depth, length, and orientation required for each inspection procedure. Nuclear inspection codes define the reference-standard geometry and the minimum signal-to-noise ratio the instrument must achieve before the inspection data is considered valid. The reference block is, in effect, the link between the colorful C-scan image on screen and a real-world crack size that an engineer can make a disposition on.
Getting this calibration wrong is one of the most common sources of inspection error. If the reference block’s surface condition does not match the component being inspected, or if the probe-to-surface coupling differs between calibration and field use, the threshold settings derived from calibration will not translate accurately. Array probes reduce some of this risk compared to single-element probes because the image format makes anomalies visually obvious, but they do not eliminate it. An inspector who calibrates on a polished reference block and then scans a corroded, paint-coated pipe section is comparing apples to oranges unless lift-off and surface-condition corrections are properly applied.