Assessment of Noise Level for Eddy Current Inspection of Steam Generator Tubes (NUREG/CR-6982)
On this page:
Download complete document
This page includes links to files in non-HTML format. See Plugins, Viewers, and Other Tools for more information.
Manuscript Completed: January 2008
Date Published: March 2009
S. Bakhtiari, D.S. Kupperman, and W.J. Shack
Argonne National Laboratory
9700 South Cass Avenue
Argonne, IL 60439
M. Stambaugh, NRC Project Manager
NRC Job Code Y6588
Office of Nuclear Regulatory Research
This report provides an overview of research activities associated with eddy current (EC) inspection of steam generator tubes that was carried out at Argonne National Laboratory (ANL) as part of the Steam Generator Tube Integrity Program sponsored by the U.S. Nuclear Regulatory Commission. Results of recent studies on assessing the influence of noise on detection of EC flaw signals are presented. In particular, methodologies for measuring the level of noise in bobbin-coil inspection data are examined. Although the primary focus is on spatially one-dimensional calculations for bobbin probe noise, limited investigations on the measurement of noise in rotating probe data are also discussed. In the first part of the report, the basic concepts and common sources of signal distortion in EC data from in-service inspection are initially reviewed. Next, various algorithms for the simulation, superposition, and measurement of noise that were developed at ANL are described. Results of studies on the effect of measurement variables (e.g., measurement window size and region of interest) on various measures of noise level are also presented. Based on these investigations, viable indicators of the noise level are identified and the associated measurement procedures are outlined. A distinction is made between the global noise levels that are important for data quality assessment and the more localized noise levels that directly affect detection capability. The second part of the report deals with measurement of the level of noise present in the tube bundle mock-up at ANL. Selected tubes were identified as having flaw signals that were borderline for detectability, i.e., they were missed by at least one of the round–robin teams or by a number of individual analysts from different teams. By analysis of these signals and the associated noise levels, inferences were made regarding the limit of detection. The implications of the results concerning the detection probability are discussed. Finally, application of the proposed methodologies for the measurement of noise level is addressed.