Imaging composite material using ultrasonic arrays

Chuan Li*, Damien Pain, Paul D. Wilcox, Bruce W. Drinkwater

*Corresponding author for this work

Research output: Contribution to journalArticle (Academic Journal)peer-review

110 Citations (Scopus)


As carbon fibre composite becomes more widely used for primary structural components in aerospace and other applications, the reliable detection of small defects in thick-sections is increasingly important. This article describes an experimental procedure for improving the detectability of such defects based on modifications to the Total Focusing Method (TFM) of processing ultrasonic array data to form an image. First the TFM is modified to include the directional dependence of ultrasonic velocity in an anisotropic composite laminate, and practical procedures for measuring the direction-dependent velocity profile are described. The performance of the TFM is then optimised in terms of the signal to noise ratio for Side-Drilled Holes (SDHs) by tuning both the frequency-domain filtering of data and the maximum aperture angle used in processing. Finally an attenuation correction is applied to the image so that the background structural noise level is uniform at all depths. The result is an image where the sensitivity (i.e. the signal to noise ratio) to a particular feature is independent of depth. Signals from 1.5 mm diameter SDHs in the final image at depths of 4, 10 and 16 mm are around 15 dB above the root-mean-square level of the surrounding structural noise. In a standard TFM image, the signals from the same SDHs are not visible above the structural noise. (C) 2012 Elsevier Ltd. All rights reserved.

Original languageEnglish
Pages (from-to)8-17
Number of pages10
JournalNDT and E International
Early online date22 Aug 2012
Publication statusPublished - Jan 2013

Structured keywords

  • Composites UTC


  • Anisotropic
  • Direction-dependant velocity
  • Full matrix capture
  • Total focusing method


Dive into the research topics of 'Imaging composite material using ultrasonic arrays'. Together they form a unique fingerprint.

Cite this