Signal Invariance to Probe Characteristics

Abstract

The reproducibility of eddy current coils and probes is a serious concern in the field of eddy current non-destructive inspections. The Advanced NDI Group at Wyle Laboratories, Inc develops fully automated eddy current inspections for use on the United States Air Force Retirement for Cause (RFC) engine rotor inspection program. Probability of Detection (POD) is established from system response versus flaw size ( â vs. a ) data per methodology outlined in MIL-HDBK-1823 using two or a small number of probes. For the POD analysis to be meaningful on fielded inspections, it is necessary to consistently produce a response similar to that from which the POD is based. This is ensured through both calibration and acceptance testing. As part of an ongoing effort to better characterize the actual response of eddy current probes, data was taken using several differential eddy current coils in the “Split-D” configuration. Contrary to expectations, it was found that the pulse width from the differential coil did not depend on the width of the coil. The pulse width is dependent on the rate of travel of the probe, and is expected to relate to the physical construction of the coil, but only a slight correlation to the coil width was observed. Additionally, it has been found that the coil could be rotated within the coil plane +/15 degrees without significantly affecting the response on an EDM calibration notch or on reliability specimen cracks in transverse orientation. It was further observed that the signal from transverse cracks remains remarkably similar when the probe is rotated through the entire 90 degree range from ideal to non-ideal orientation.