- since 1996 -
|NDT.net Issue - 2014-12 - NEWS |
Full matrix capture (FMC) imaging through a non-planar surface is a time-intensive task, due to the need to calculate the time of flight from each transmit/receive element combination to a given pixel in the region of interest through the refractive boundary. Due to the computational complexity, data is often post-processed after the inspection. TWI Technology Centre (Wales) has developed a real-time FMC imaging system capable of generating full-focused imagery through a component of complex geometry where no prior knowledge of the profile is provided.
Leading innovation in non-destructive testing
FMC is a relatively new ultrasonic non-destructive testing (NDT) technique, used to image sub-surface flaws in test components. The main advantage of FMC over existing phased array techniques is its ability to generate fully focused imagery of components. It achieves this by firstly acquiring the full time-domain signal for every possible transmit/receive combination, and then discretising the region of interest into a finite grid. Every cell in the grid is treated as a focal point and represents a pixel in the final image.
This ability to acquire the full time-domain signal from every possible transmit/receive element allows for changes in focal requirements to be determined dynamically and after acquisition, as there is no ultrasonic focusing within the component (as in the case of phased array).
This work developed algorithms which:
The ultrasonic and signal processing techniques developed were:
To illustrate the effectiveness of the technique a part was manufactured with a profile of complex geometry, with which the FMC imaging system was provided with no prior knowledge. This component is shown in Figure 1. Custom FMC software was developed where the FMC imaging algorithm was used to image the component before and after accounting for the boundary. These results are shown in Figure 2.
Figure 1 Component of complex geometry
Figure 2 FMC imaging (left) before accounting for boundary and (right) after dynamically accounting for boundary in real-time
The technique shows promise to deliver fast and accurate defect detection and characterisation, beyond what is currently achievable. Future work includes extending the capabilities of current auto-focus algorithms; converting algorithms for commercial application; and developing a robust, portable hardware system for on-site inspection.