DEFECTS SIZE MEASUREMENT AND PHYSICAL LIMITATIONS OF DIGITAL DATA PROCESSING IN ULTRASONIC FLAW DETECTION BY ECHO AND DIFFRACTION METHODS

The article notes that manufacturers and suppliers of ultrasonic flaw detectors usually focus on the sensitivity of the equipment, but pay little attention to the error which the linear dimensions of defects can be measured with. Examples of representing sections with defects in the form of acoustic B-scans are provided, and it is shown that these images have changed little over the past 60 years, despite the fact that the level of technology and signal processing have developed significantly during this time. It is noted that the physical principles that underlie ultrasonic flaw detection and the accumulated practical experience of its application show that at present the dimensions of defects can be measured no more accurately than with an error of 1 mm. Therefore, it would be more accurate to talk not about measuring of defects height in the cross-section of welded seams, but about assessing this height using ultrasonic flaw detection methods. At the same time, the article shows that it is possible to qualitatively assess the defect height change by fractions of a millimeter. For this purpose, in addition to the traditionally used time sweeps of signals (A-scans) and amplitude spectra, it is advisable to pay more attention to the analysis of phase spectra of signals received from products. The results of an experimental assessment of the influence of weak anisotropy of the material on the shape of the phase spectra of bottom pulses are presented. The calculations performed showed that the phase spectra also change significantly more than other characteristics of the recorded pulses with an increase in the height of defects in the cross section of the weld. This conclusion also applies to cases where the height of the defect does not exceed the length of the ultrasonic waves used