NDT methods for composite materials and how to choose them

Choosing NDT for composites starts with the defect, not the instrument

NDT methods for composite materials are not interchangeable. A laminate, sandwich panel or composite pressure vessel can fail through delamination, disbonding, impact damage, porosity, fiber waviness, moisture ingress or matrix cracking. Each flaw interacts differently with sound, heat, radiation, strain and electromagnetic fields. In practice, ultrasonic testing is often selected for delamination and thickness-related indications. Thermography and shearography are effective for rapid large-area screening under the right conditions. Radiography and computed tomography can show internal geometry. Eddy current testing is mainly relevant to conductive carbon fiber composites, while acoustic emission is better suited to monitoring active damage than mapping every static flaw. A sound inspection plan normally combines suitable methods, reference standards, trained personnel and engineering acceptance criteria.

For more materials inspection topics, visit the Testing section.

Why composite materials need a different NDT strategy

Conventional NDT practice was largely shaped by metals, where inspectors often look for cracks, corrosion, lack of fusion, wall loss or inclusions in relatively homogeneous materials. Composite materials behave differently. A carbon fiber reinforced polymer, glass fiber reinforced polymer or aramid composite is anisotropic, layered and often bonded to cores, inserts or other skins. One component may contain plies with different orientations, resin-rich zones, adhesive films, honeycomb cells and local thickness changes.

This structure creates two practical consequences. First, a defect may be planar and parallel to the surface, such as a delamination. It may be difficult to see during visual inspection but still reduce compression strength. Second, the inspection signal can be hard to interpret because fibers, resin, core geometry and curvature all affect the measurement. A strong ultrasonic reflection, thermal contrast or strain anomaly has to be judged against the expected response of the design, not against a generic material model.

Published guidance reflects this need for engineering judgment. ASTM E2533-21, a guide for nondestructive examination of polymer matrix composites used in aerospace applications, states that its role is to help engineers select appropriate NDT methods and reference established practices. It does not set accept-reject criteria or approve components for service. That distinction matters outside aerospace as well. NDT provides evidence, but acceptance depends on the design basis, qualification program, risk level and service environment.

Common NDT methods for composite materials

The point of comparing methods is not to rank them in every situation. It is to understand what each method can detect, where it is reliable, and where its limits begin. The table below summarizes practical uses and limitations for common composite inspections.

ASTM E2580-17 specifically addresses ultrasonic testing of flat panel composites and sandwich core materials used in aerospace applications. ASTM E3370-22 addresses matrix array ultrasonic testing for composites, sandwich core constructions and metals used in aerospace applications. These standards show how far ultrasonic methods have been formalized, but they do not mean ultrasound is always sufficient. NASA technical materials also describe flash infrared thermography as a useful NDE approach for nonmetallic composite structures and for rapid inspection of composite honeycomb sandwich structures. The practical lesson is that method maturity and defect suitability have to be evaluated together.

Matching methods to defects and inspection scenarios

Delamination, disbonding and impact damage

Delamination is one of the main inspection targets in laminated composites. Low-velocity impact can leave little visible surface evidence while creating subsurface ply separation. Ultrasonic pulse-echo, through-transmission ultrasound and phased or matrix array ultrasonic methods are commonly used because acoustic reflections and attenuation changes can indicate internal separation. For flat panels, C-scan images can provide a useful plan view of the indication.

Thermography can also be effective, especially when the defect changes heat flow from the surface into the material. Flash or pulsed thermography briefly heats the surface and records the temperature decay pattern with an infrared camera. A delamination, disbond or void can create thermal contrast because it interrupts heat transfer. For large panels, that speed is attractive, but depth estimation is usually less direct than with ultrasound.

Shearography approaches the same problem from a stiffness response. Instead of primarily measuring transmitted sound or heat flow, it detects small surface strain changes under applied stress. When a hidden disbond or delamination locally changes stiffness, the surface deformation pattern can reveal an anomaly. This makes shearography a strong candidate for bonded panels, honeycomb structures and large-area screening, provided the loading method is repeatable and appropriate for the component.

Porosity, voids, wrinkles and inclusions

Manufacturing defects require a different inspection lens. Porosity and void content can affect mechanical performance, but detectability depends on size, distribution, laminate thickness and inspection resolution. Ultrasonic attenuation, backscatter and velocity-based measurements may indicate porosity. Calibration against known reference panels is important because resin system, fiber volume and layup also influence the signal.

Radiography and computed tomography can be valuable when the main concern is internal geometry. CT can provide three-dimensional information about voids, wrinkles, cracks, inclusions or core damage in suitable parts. It is especially useful in failure analysis, process development and small high-value components. However, CT is not automatically practical for every production composite. Component size, access, resolution, scan time and cost can all become limiting factors.

Fiber waviness and wrinkles are particularly challenging because their significance is design dependent. A local wrinkle may be critical in a compression-loaded aerospace part but less severe in a lightly loaded cover. NDT can help locate and characterize the anomaly, but the engineering team must determine whether it violates the qualified process window or structural acceptance criteria.

In-service monitoring and large structures

In-service composite inspection often involves constraints that factory inspection does not. Access may be one-sided, surfaces may be painted or contaminated, and the structure may be curved, thick or integrated with metallic fittings. The inspection objective may also shift from full characterization to risk-based screening. For example, maintenance teams may first use visual inspection, tap testing or thermography to identify suspect zones, then apply ultrasound for local confirmation and sizing. See also: Application.

Acoustic emission has a different role. It is generally used when the component is under load and the aim is to detect active damage events such as matrix cracking, fiber breakage or delamination growth. This can be useful during proof testing of composite pressure vessels or during fatigue evaluation. Acoustic emission should not be treated as a simple substitute for imaging methods. It is a monitoring and event-detection tool, and interpretation depends on sensor placement, noise control, loading history and source classification.

Standards, qualification and acceptance criteria

A reliable composite NDT program needs more than a capable instrument. It requires written procedures, trained personnel, calibration or reference standards, documented scan settings, environmental controls and clear reporting rules. Standards from ASTM Committee E07 and related aerospace guidance are often used to structure this work, but the acceptance decision normally belongs to the responsible engineering organization.

This is a frequent source of confusion. A standard practice may describe how to perform an inspection, verify equipment performance, collect data or report indications. It may not define whether a specific indication is acceptable for a specific aircraft part, wind blade, marine structure, pressure vessel or automotive component. Acceptance criteria require knowledge of loading, safety factors, damage tolerance, manufacturing qualification and service consequences.

Reference standards are especially important for composites because signals are material specific. A reference block or panel should represent the relevant material system, thickness, layup, surface condition and defect type as closely as practical. Artificial defects are useful for procedure development, but they may not perfectly reproduce real impact damage, kissing bonds, distributed porosity or fatigue-driven delamination. For critical parts, inspection reliability should be demonstrated through probability of detection work, round-robin studies or other qualification evidence appropriate to the risk.

A practical selection workflow

When choosing NDT methods for composite materials, start with a structured decision process rather than a preferred technique.

1. Define the material and structure. Identify whether the part is CFRP, GFRP, aramid, ceramic matrix composite or sandwich construction. Record thickness, curvature, surface finish, core type, bonding and access.
2. Define the target defect. Separate delamination, disbonding, porosity, inclusion, impact damage, moisture ingress, fiber waviness and crack growth. Each defect has a different physical signature.
3. Define the inspection stage. Incoming material inspection, in-process control, final acceptance, maintenance, proof testing and failure analysis require different sensitivity, speed and documentation.
4. Screen possible methods. Use ultrasound for depth-related laminate indications, thermography or shearography for large-area screening, radiography or CT for internal geometry, eddy current for conductive carbon fiber applications and acoustic emission for active damage monitoring.
5. Check constraints. Consider one-sided access, couplant restrictions, radiation safety, part size, scan speed, automation, operator qualification and data archiving.
6. Validate with representative standards. Confirm that the chosen procedure can detect relevant flaws in representative materials before relying on it for acceptance decisions.
7. Document limits clearly. A good report should state not only what was found, but also what the method was not capable of detecting under the inspection conditions.

For many industrial programs, the final answer is a layered inspection plan. Thermography or shearography may rapidly screen a broad area, ultrasound may size selected indications, and CT may be reserved for high-value investigations or process development. This combination is often more defensible than forcing one method to answer every inspection question.

Frequently asked questions

What is the most commonly used NDT method for composite materials?

Ultrasonic testing is one of the most established methods for composite laminates and sandwich panels, particularly when the target is delamination, disbonding or thickness-related indications. However, it is not universally the right method. Thermography, shearography, radiography, CT, eddy current and acoustic emission may be better suited depending on defect type, access and inspection purpose.

Can thermography replace ultrasonic testing for composites?

Thermography can be faster for large-area screening and is attractive for near-surface defects that create clear heat-flow contrast. It should not be treated as a direct replacement for ultrasound in every case because depth capability, surface condition, heating method and defect orientation strongly affect results. Many programs use thermography for screening and ultrasound for confirmation or sizing.

Is eddy current testing useful for all composites?

No. Eddy current testing depends on electrical conductivity, so it is mainly relevant to conductive composites such as carbon fiber reinforced polymers. It has limited value for nonconductive glass fiber composites unless there is a conductive feature or special inspection objective.

Why are acceptance criteria not included in many NDT standards?

Acceptance depends on the component design, load case, qualified manufacturing process, safety requirements and service environment. Many NDT standards describe inspection methods and reporting practices, while the responsible engineering organization defines whether an indication is acceptable for a specific application.

Which method is best for composite sandwich structures?

There is no single best method for all sandwich structures. Ultrasound can be useful for skin and bondline indications, thermography and shearography can screen large bonded areas, and radiography or CT can help with crushed core, water ingress or internal geometry. The choice depends on core type, skin thickness, access, defect depth and required confidence.