What is non-destructive testing (NDT) in composite materials

Non-destructive testing (NDT) is the multidisciplinary practice of evaluating the physical properties, structural integrity, and defect state of a material or component without permanently altering its function, geometry, or serviceability. Most methods work by introducing a controlled external energy field (high-frequency sound, thermal gradients, electromagnetic induction, ionizing radiation) and measuring the response.

The central rule is asymmetric. A rejectable indication means a discontinuity has exceeded code-allowable limits. A clearance never proves an absolute absence of flaws, only that none exist above the detection threshold of the equipment used. That asymmetry shapes the acceptance criteria, the sampling plan, and the certification protocol that an NDT operation runs under.

How to recognize NDT in practice#

Three features identify NDT versus destructive testing or visual inspection alone.

The part is not altered. Visual inspection at standard lighting may scratch a clearcoat; ultrasonic inspection may leave couplant on the surface; nothing permanent happens to the structure. A destructive coupon test consumes the coupon; an NDT scan returns the part to service.

The inspection couples an external energy field to a discontinuity class. Ultrasonic methods couple high-frequency sound to interlaminar separations and density variations. Thermographic methods couple a transient or modulated heat input to subsurface features that alter thermal diffusivity. Radiographic methods couple photon absorption to density variation. Each method has a defect class it reads well and a class it misses, which is why combining modalities is the standard practice for high-confidence inspection.

The output is an indication, not a verdict. An A-scan, B-scan, C-scan, thermogram, shearogram, or radiograph is a measurement of the material's response to the input field. An NDT operator interprets the indication against acceptance criteria written for the part: maximum allowable indication size, depth, location, density. The pass/fail decision rests on the comparison; the modality only produces the measurement.

The common methods, briefly#

A composite-relevant NDT toolkit covers six method families.

Visual (VT) is the baseline. Direct visual inspection, augmented by 10x loupes, raking light, borescopes, and digital microscopy, finds surface-breaking cracks, paint disruption, weave distortion, and impact dents. It misses everything subsurface, which is most of what matters on a composite.

Penetrant (PT) uses low-viscosity dye (fluorescent under UV or visible-dye against a contrast developer) that seeps into surface-breaking discontinuities. After surface cleaning, the dye trapped in the discontinuity reveals the defect's location and length. PT works on any non-porous material; it does not find subsurface defects.

Ultrasonic (UT) transmits high-frequency sound (typically 1 to 10 MHz on composites) into the part and reads the reflected signal. Pulse-echo UT measures back-wall reflection: a sound laminate returns a clean back-wall echo, while delamination or voids attenuate or shift the echo. Phased-array UT (PAUT) extends pulse-echo with electronic beam steering. UT requires couplant (gel or immersion) and careful calibration but is the workhorse for composite delamination detection.

Radiographic (RT) uses X-rays or gamma rays. The part absorbs photons in proportion to local density; a film or digital detector on the far side captures the transmission pattern. Computed radiography and X-ray CT produce 2D and 3D images respectively. RT is the gold standard for complex internal geometry and is highly effective at finding inclusions, voids, cracks transverse to the beam, and fiber misorientation. It requires radiation safety, expensive equipment, and limited part size for the high-resolution variants.

Eddy current (ET) induces alternating-current "eddy" currents in a conductive material and measures the impedance shift caused by surface and near-surface discontinuities. It is fast and non-contact but requires conductivity, so it is ineffective on glass and aramid composites; on carbon-fiber laminates it can detect breaks in the conductive fiber network.

Acoustic emission (AE) is passive. Sensors listen for the transient elastic waves released when a stressed material actively damages, with fibers breaking, the matrix cracking, plies delaminating. AE detects damage as it happens during proof loading or in service. It does not size or image static existing flaws, but it monitors for new or growing damage in real time.

Thermography (pulsed, lock-in, vibrothermography) applies a heat input (flash, modulated, or vibration-induced friction) and reads transient or steady-state surface temperature with an infrared camera. Defects that alter thermal diffusivity (voids, delaminations, inclusions, debonded patches) show as thermal anomalies. Thermography is non-contact and fast over large areas, with the trade-off that thermal-wave penetration limits depth.

Confusion points#

NDT versus visual inspection. Visual inspection is the baseline NDT method, but "NDT" in industry usage commonly implies one of the instrumented modalities (UT, RT, ET, AE, thermography). A walkaround visual inspection of a composite primary structure is necessary but not sufficient; BVID is by definition the damage class that visual inspection cannot detect, and finding it requires an instrumented modality.

NDT versus destructive testing. Destructive coupon tests (tensile, compression, interlaminar shear, compression-after-impact) consume the coupon and produce a property number. NDT measures the part in place, without consumption, and produces an indication of defect state. The two are complementary: destructive tests calibrate the acceptance criteria, NDT verifies that production parts meet them.

A clearance is not a guarantee. This is the asymmetry that drives the entire field. A rejectable indication has exceeded a known threshold; a clearance means no indication exceeded the threshold. Probability of detection (POD) curves describe the likelihood that a given defect size is detected by a given method; even at 90% POD, 10% of defects of that size will be missed. Sampling plans, certification protocols, and acceptance criteria are all built on this asymmetry.

Code-allowable versus structural significance. Code limits are conservative, set by the design authority for the part. An indication that exceeds the code limit is rejectable regardless of whether it would actually cause failure in service. An indication below the code limit is acceptable regardless of whether it represents an underlying issue. NDT enforces the code; it does not redesign the part.

Aerospace versus field-screening rigor. Aerospace NDT runs under ASNT, BINDT, and EN 4179 certified procedures with documented POD, calibrated equipment, and traceable records. Field-screening practices (the coin tap test, raking-light visual, borescope checks) are useful for triage but do not meet the documentation standard of certification-grade NDT.

Related terms#

• Phased array ultrasonic testing (PAUT): the most common ultrasonic NDT modality for composite delamination detection.
• Shearography: the full-field optical NDT method for large-area subsurface inspection.
• Barely visible impact damage (BVID): the defect class that defines why composite NDT exists at all.
• Delamination: the dominant interlaminar defect class that pulse-echo UT and active thermography are tuned to detect.