NDTnet 1998 Aug, Vol.3 No.8

The Performance of Radiography for Large Planar Defects in Thick-Section Welds.
R.K. Chapman, G.S. Woodcock - Nuclear Electric. A.B. Wooldridge - Magnox Electric. I.J. Munns, G.A. Georgiou - TWI. UK.
Keywords: Welds, Radiography
Abstract
The UK nuclear licensees have recently been investigating the reliability of radiography for detecting large planar defects in thick-section butt welds (thickness 50-114 mm). The main part of the work has been an extensive experimental study in which test specimens were manufactured containing 19 realistic large planar welding defects, typically at least 15 mm through-wall and 70 mm long. These specimens were radiographed using a variety of techniques, including angled shots and the use of spacer plates to increase the thickness, giving 132 radiographs in all. The specimens were then sectioned to provide detailed information on the sizes, orientations and gapes of the defects. Such sectioning data was used to confirm that the defects were representative, to aid interpretation of the experimental results and to provide data for theoretical modelling. Fig. |
The results of the study show that radiography has the capability to detect a wide range of planar manufacturing defects of the size which might be of structural concern. Overall, one radiographic interpreter detected about 87% of the defects radiographed, the other about 93%. The results have been analysed to provide evidence on the variation of detectability with such parameters as defect type, through-wall size, orientation and gape (see figure for an example).The experimental results compare encouragingly with the predictions of the simple Pollitt model of radiography, which assumes a uniform parallel-sided slot. Further work on a more comprehensive model is currently in progress to provide more precise comparisons between theory and experiment.
Various auxiliary studies have provided supplementary evidence using other test specimens, including results on thinner material. Another study is currently underway aimed at including the effects of human factors in radiographic interpretation, by using a larger sample of interpreters and a greater variety of radiographs.
Abstract Source:
Book of Abstracts, 7th European Conference on Non-Destructive Testing, 26-29 May 1998, ISBN: 87-986898-0-00
Full-Text Source:
Proceedings of the 7th European Conference on Non-Destructive Testing, 26-29 May 1998, ISBN:
Publication Contact:
7th ECNDT, Park Alle 345, DK-2605 Broendby, Denmark, Fax: +45 46 26 70 11, Email: 101373.3414@compuserve.com
© 1998 NDT.net, info@ndt.net