Home
Home
16th WCNDT 2004 - World Conference on NDT
CD-ROM Proceedings, Internet Version of ~600 Papers
Aug 30 - Sep 3, 2004 - Montreal, Canada
START
Home

First 1st    previous prevWCNDT 2004 - Abstractsnext next

SESSION: SIGNAL PROCESSING
ABSTRACT:
DIGITAL SIGNAL PROCESSING OF ULTRASONIC SIGNALS FROM WELD DEFECTS A 
COMPARATIVE STUDY 
P.V.S. Suryanaryana, P V S Kurmanath, C. Subbiah, K. Viswanath
NDT/SPROB, SDSC SHAR, Sriharikota; India

In conventional pulse echo Ultrasonic Testing (UT), defects are represented as time vs. amplitude plot 
known as A scan which is used by the operator for interpretation and evaluation of the defects. Generally, in 
the routine UT method, the component is either accepted or rejected by comparing the A-scan patterns of 
defects with that of known reference reflectors like flat bottomed holes, side-drilled holes, notches, etc. The 
A-scan patterns of similar defects give rise to identical appearance on the CRT having no clue as to the 
nature of reflector. Thus difficulty arises if one intends to characterize and differentiates between defects 
such as porosity - slag or root crack -incomplete penetration etc. In this era of rapid production where 
fitness-for-purpose approach is the order of the day, there is a need for automation and visualization of 
various defects so as to take a quick decision about usefulness of a product. Towards this direction, the 
authors undertook a comparative study of digital signal processing of ultrasonic signals obtained from 
different weld defects. Transformation techniques such as cross correlation, FFT, STFT, Wavelet 
Transformations of Daubachies-439, Hartley transformation are applied in both time and spatial domains to 
extract required features. We could obtain processed signals using aforesaid transformations and it is found 
that wavelet transformations offer the most distinctive features for identical planar defects like root crack, 
side wall crack, lack of sidewall fusion and incomplete penetration. A similar differentiation could also be 
found with voluminar defects like porosity, slag inclusions. The details of analysis of signals using various 
transforms and results thereof are discussed in this paper.
Full-Text HTML-txt
Full-Text PDF (KB)
Full-Text HTML:
OPTION (MB):
MAIN AUTHOR:K. Viswanathan, Indian Space Research Organisation, India
Paper CODE: 232

© NDT.net