Thickness determination of dual-layer coatings based on ultrasonic spectral filtering

Z. Y. Ma, W. Zhang, Z. B. Luo, L. Lin*, S. Krishnaswamy

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


In this paper, a non-destructive method for determining the thicknesses of dual-layer coatings is developed using an ultrasonic spectral filtering technique based on the theoretical investigation of wave propagation in specimens with two layers and three interfaces. The acoustic reflection coefficient of dual-layer coatings was separated from the measured signal through Fourier transformation and Wiener deconvolution. Utilising the multi-frequency windows autoregressive spectral extrapolation (MARSE) technique, the cosine signals (sine signals), including the time-of-flight (TOF) of each layer, were split from the real part (imaginary part) of the acoustic reflection coefficient. The thicknesses of the dual-layer coatings were calculated utilising the periods of the cosine signals (sine signals) and the longitudinal wave velocities of each layer. Ultrasonic experiments were carried out on the dual-layer coatings, where the thicknesses of layer 1, layer 2 and the aluminium substrate were 230 μm, 450 μm and 5 mm, respectively. The thicknesses measured by the proposed ultrasonic method were in good agreement with those observed by optical microscopy (OM) and the relative errors of layers 1 and 2 were both less than 5.0%.

Original languageEnglish (US)
Pages (from-to)200-205
Number of pages6
JournalInsight: Non-Destructive Testing and Condition Monitoring
Issue number4
StatePublished - Apr 2018


  • Dual-layer coating
  • Multi-frequency windows autoregressive spectral extrapolation (MARSE)
  • Spectral filtering
  • Thickness
  • Ultrasonic

ASJC Scopus subject areas

  • Mechanics of Materials
  • Mechanical Engineering
  • Metals and Alloys
  • Materials Chemistry

Fingerprint Dive into the research topics of 'Thickness determination of dual-layer coatings based on ultrasonic spectral filtering'. Together they form a unique fingerprint.

Cite this