Defect Detection by Infrared Thermography (NDT): How & Why
IRT detects defects by identifying temperature differences on the surface of a material. These temperature differences arise because defects alter the material's thermal properties or impede heat flow. Defects (cracks, voids, delaminations, inclusions) often have lower thermal conductivity than the surrounding material. This creates thermal resistance, meaning heat flows less easily through the defect. Areas with defects may also heat up or cool down at different rates than the surrounding material, because the defect affects how heat is absorbed, conducted, and diffused. These temperature differences, even if small, are detected by the infrared camera and displayed as a thermal image, allowing inspectors to identify and characterize defects without damaging the material.
The "IRT Setup Classification Table" we designed below allows you to easily understand the various setups and combinations for an NDT inspection by thermography that can be realized. Simply select one item per column from column A to column G to create a specific setup:
IRT Setup classification as seen by Visiooimage Inc.
Passive IRT: This method relies on naturally occurring temperature differences. The object being inspected emits infrared radiation based on its own temperature. Defects are detected by variations in this emitted radiation. It's useful for identifying hot spots (e.g., electrical faults) or cold spots (e.g., moisture intrusion). No external energy source is used. Passive IRT is usually not considered an NDT method.
Active IRT: This method uses an external energy source to induce a temperature difference in the object being inspected. The energy source heats the object, and the infrared camera detects variations in the resulting temperature distribution. This is more sensitive than passive thermography and can detect smaller defects, even if they don't create a significant temperature difference on their own.
Static IRT: Both the infrared camera/source and the part being inspected remain stationary. This is the simplest and most common setup. It is possible to implement Static IRT with a robotic system that progresses tile by tile to cover a larger surface area, but not to be confused with Dynamic IRT below.
Dynamic IRT (Movement): Either the camera/source or the part being inspected is in motion. This is useful for inspecting large or complex objects. This is also used for inline inspections on a conveyor belt or for linescan inspection of long or large parts (e.g., composite materials, wings) using a gantry or robot.
Transmission IRT: The infrared source is placed on one side of the part, and the camera is on the other. This requires access to both sides of the part. It's useful for inspecting thicker materials.
Reflection IRT: The infrared source and camera are on the same side of the part. This is the most common and convenient method, as it only requires access to one side.
Internal Heating: Heat is generated within the part itself, typically using induction heating. This involves using an electromagnetic field to induce eddy currents within a conductive material, causing it to heat up. This is useful for inspecting materials that are difficult to heat from the surface or for detecting defects deep within the material.
Point Scanning: A focused energy source (e.g., laser) is moved across the surface.
Line Scanning: A linear energy source (e.g., a linear halogen tube) is used, often in conjunction with a moving part (linescan inspection).
Surface: A broad-area energy source (e.g., flash lamp) heats the entire surface. This is the most common method, providing a quick overview of the entire part.
Optical (Radiative): Uses electromagnetic radiation (light) to heat the surface.
Flash: A short, intense burst of heat.
Halogen: Continuous heat output.
Mechanical (Vibrothermography): The part is vibrated, and the friction inside defects (ex delamination or cracks) and surfaces generates heat.
Inductive: Uses an electromagnetic field to induce eddy currents within a conductive material, generating heat.
Pulse (Flash): A short burst of energy.
Square Pulse: Constant energy for a defined period.
Step (Continuous): Constant energy applied continuously.
Modulated: The energy source is modulated (e.g., pulsed at a specific frequency). This can improve signal-to-noise ratio and enhance defect detection.
Post-processing techniques are sometimes applied to raw infrared (IR) data to improve the clarity and visibility of defects. The goal is to enhance the signal-to-noise ratio (SNR) or, more specifically in defect detection, the contrast-to-noise ratio (CNR). Essentially, we want to make the temperature differences caused by defects stand out more clearly from the background low or high frequency noises and random variations in the thermal image.
This is achieved through the application of various filters. These filters can be broadly categorized into two types:
Spatial Filters: These filters operate on a single IR image, modifying the pixel values based on the surrounding pixels. Common spatial filters include for example Gaussian Filters (Smooth the image, reducing high-frequency noise and blurring edges), Median Filters (replace each pixel with the median value of its neighbors, effectively removing impulsive "salt and pepper" noise and bad pixels if any).
Temporal Filters: These filters analyze a sequence of multiple IR images captured over time. They are particularly powerful for detecting subtle changes in temperature that might be invisible in a single image. Examples include the PPT (Pulse Phase Thermography): Analyzes the phase shift of temperature variations over time, and the PCT (Principal Component Thermography): a statistical technique used to reduce the dimensionality of datasets while retaining as much of the original information as possible.
Temporal filters are especially valuable for enhancing very small temperature variations or subtle evolution of temperature over time. A defect might not cause a large, immediate temperature difference, but it could slightly affect how an area heats up or cools down compared to a defect-free area. By analyzing a sequence of images, temporal filters can reveal these tiny differences, making defects visible that would otherwise remain hidden. They essentially amplify the signal related to the defect's thermal behavior, while suppressing random noise.
Post-processing techniques act as a refinement stage, transforming raw IR data into a clearer, more interpretable image, improving the reliability and accuracy of defect detection.
Examples of setups
Halogen heating by reflection
Halogen by transmission
Robotized linescan
Manual Induction
Starting IRT in Research & Development – Navigatingthe unknown
You're interested in integrating IRT into your R&D efforts. It's normal for current and future R&D projects to involve a degree of uncertainty—that's the nature of research. To help you get started with IRT despite this uncertainty, two possible pathways:
Option 1: Future-Focused Choice
Let's discuss the types of materials and defects you anticipate investigating in future projects. We can then recommend IRT configurations that are typically most effective for those scenarios. This approach proactively aligns the technology with your potential needs.
Option 2: Capability-Driven Exploration
We can provide an overview of the strengths of various IRT setups. This will help you understand which systems are best suited for detecting specific types of flaws in different materials, allowing you to identify potential applications within your broader research goals.
Validating IRT for Your Application: A Step-by-Step Approach
You’re considering IRT and want to ensure it’s a viable solution? Here’s how we recommend proceeding:
Step 1: Initial Consultation & Feasibility Discussion
You believe IRT may be applicable to your needs. Let’s discuss your specific case with a Visiooimage Engineer or Ph.D. We’ll provide an expert opinion on the realistic feasibility of IRT for your application. While 20 years of experience provides valuable insight, definitive confirmation will still require testing.
Step 2: IRT Qualification Testing
We recommend conducting a qualification test study. This isn’t about finding defects in your samples, but rather evaluating if IRT can reliably detect known defects in your materials. We’ll test your samples using IRT setups we believe are best suited, and deliver a report outlining our findings, including trade-offs between inspection speed and Probability of Detection (POD).
What’s Involved in Qualification Testing?
Sample Submission: Please send us samples (which we will return) with well-documented, known defects. This documentation should include defect specification (type, size, position, depth, etc.)
On-Site Testing (Optional): If sample shipment isn’t feasible, we can conduct testing at your facility.
Focus on IRT Performance: We evaluate how well IRT performs on known defects, not simply whether defects exist.
Why are Known Defects Crucial? Samples with unknown defects are not recommended for qualification. If we fail to detect a defect, we won’t know if it’s due to an unsuitable IRT setup or because no defect was present.
How to Create Samples with Known Defects:
Artificial Defects: Introduce controlled defects like flat-bottom holes, inserts, or cracks. Ensure you can consistently recreate these defects.
Previously Identified Real Defects: Utilize real production parts where defects have already been located using other NDT methods (e.g., radiography).
Controlled Fabrication Process: Alter your manufacturing process to reliably create specific defects (e.g., inducing porosity in a composite material).
Important Note: Focus on defects that are relevant to your production and would lead to part rejection. Don’t aim for the smallest possible defects; prioritize those that are critical to your application.
Step 3: Decision & Implementation
Based on the qualification test report, you can confidently decide if IRT is suitable for your needs. Our experience shows:
Approximately 1/3 of cases: IRT may not be effective.
Approximately 1/3 of cases: IRT works well with some optimization.
Approximately 1/3 of cases: IRT performs exceptionally well.
Calibration & Ongoing Verification
Before full implementation, calibration standards (calibres) or Inspection Reference Panels (IRP) built with your own standards are recommended. These standards help define detection limits and are often reused in production to verify system performance and operator proficiency – a common practice in aerospace. Visiooimage can provide some standard IRP.
This structured approach minimizes risk and ensures a successful IRT implementation tailored to your specific requirements.
Example of an aircraft IRP with a honeycomb structure. The aircraft manufacturer included realistically sized defects that are representative of those potentially found in actual aircraft.
From top to bottom, the IRP contains delaminations, core unbonds, excessive adhesive, and crushed core. It also includes defects of varying sizes, with their positions and types clearly marked. The two circles at the bottom are not defects; they indicate the difference in honeycomb core thickness—0.187" on the left and 0.25" on the right
Infrared Thermography NDT results
Defect Detection Principle in Active IRT with induction heating
Visiooimage inductive generator is a great way to find defects in electrically conductive materials, in particular to find hidden corrosion. Let's see how this works.
Induction heating is a process that generates heat electromagnetically within electrically conductive materials, such as metal plates or carbon fiber reinforced polymers (CFRP), as long as they possess sufficient electrical conductivity.
Induction heating systems utilize eddy currents induced within the workpiece to generate heat. Unlike traditional methods, induction doesn’t rely on a heating element or flame for heat transfer via conduction, convection, or radiation. Instead, alternating current flows through a coil, creating an alternating magnetic field. When this field passes through the workpiece, it induces eddy currents within the material. Importantly, the coil itself remains cool. The resistance of the material to these eddy currents generates heat directly inside the material, rather than on its surface. This makes induction heating highly efficient, as minimal heat is lost. The part itself acts as the heating element, heating from within.
A defect in a heated material creates a temperature difference on the surface, even if covered by an insulating layer. These temperature differences arise from two main sources:
- Thermal Resistance/Conductivity: Similar to traditional active thermography (using halogen or flash heating), a defect or discontinuity in the material will alter heat diffusion compared to the surrounding sound material. This is because the defect possesses different thermal properties.
- Electrical Resistance: A defect has different electrical properties, specifically electrical resistance. Due to the Joule effect, the defect will generate a different amount of heat for the same current compared to the sound material. Furthermore, current tends to flow through paths of least resistance, avoiding more resistive areas. This affects local heat generation. Close defects can also electrically influence each other, creating complex heat patterns, particularly for small defects. Consequently, the observed heat pattern may not directly correspond to the defect’s shape. Therefore, caution is needed when quantifying defect size using induction heating.
ASNT’s Nondestructive Testing Handbook of Thermal & Infrared Testing is often considered as a reference in the field, with this new edition benefiting from contributions by leading IRT experts worldwide.
Visiooimage's engineers and PhD's were pleased to be invited to contribute to the writing and co-authored:
Description
The first update to the Thermal and Infrared Testing Handbook in over 20 years, this long-awaited fourth edition is the largest, most complete handbook ever published by ASNT. It is packed with revised and expanded content, including over 200 new color images contributed by practitioners around the world.
A new feature is the inclusion of QR codes that link to pertinent videos throughout the Handbook. Additions to this handbook also include step-by-step instructions for setting up a thermal and infrared testing room, as well as an exploration of:
pulsed thermography,
inductive thermography,
lock-in thermography,
and pulse phased thermographic techniques.
An enlarged chapter on data processing offers insightful discussions on thermographic signal recognition, principal component thermography, wavelet thermography, and thermal virtual waves.
In addition, the fourth edition offers more on composite applications, including wind blade inspection. Other material has been updated with the most recent information in areas such as errors and uncertainties of thermographic measurement, standards, applications, and equipment, including a section on drone-mounted infrared cameras.
See the TOC and get the book: https://source.asnt.org/ndt-handbook-ir-4th-ed/