In the determination of POD by techniques such as A-hat versus A, the slope of a plot of log (flaw response) versus log (flaw size) is determined empirically by a regression analysis. This often works quite well. However, in the analysis of ultrasonic data obtained in the detection of naturally-occurring, hard-alpha inclusions in titanium alloys, a very low slope has been obtained. Physically, this is presumably the result of the complex morphology of the flaws, which leads to a very weak dependence of flaw response of flaw area. In the context of the POD analysis, this low slope of the regression line implies that very small flaws would produce a significant response, and hence a POD curve that retains a relatively large value at these small flaw sizes, in conflict with field experience. In this paper, the causes of this apparent inconsistency are identified and a procedure to correct for it is presented. The essential idea is that there is a change in slope of the regression line when flaw sizes become sufficiently small that one enters the Rayleigh scattering regime. Hence the slope of the mean flaw response line must be much steeper in this regime than it is for the larger flaw sizes, as is derived from the regression analysis. Procedures are presented to use quantitative understanding of the physics of the Rayleigh scattering regime to make the appropriate correction.
Use of physics-based models to guide the extrapolation of aircraft engine ultrasonic POD data to small flaw sizes
Der Einsatz physikalisch basierter Modelle zur Extrapolationshilfe von Ultraschall-POD-Daten kleiner Fehlergrößen an Flugzeugmotoren
2006
8 Seiten, 7 Bilder, 11 Quellen
Conference paper
Storage medium
English
European Patent Office | 2016
|Emerald Group Publishing | 1984
ULTRASONIC FLAW DETECTOR AND ULTRASONIC FLAW DETECTING METHOD FOR COMPOSITE MATERIAL
European Patent Office | 2019
|Ultrasonic flaw detector and ultrasonic flaw detecting method for composite material
European Patent Office | 2022
|