Modeling of a Giant Magnetoresistance Eddy Current Testing (GMR-ECT) Array Probe for Detection of Small Flaws in Complex Geometries, 14-9400Printer Friendly Version
Inclusive Dates: 09/09/03 - 10/24/03
Background - In the aircraft industry, there is a strong need for detection of tiny surface flaws in turbine engine parts. Eddy current testing (ECT) is presently used for this purpose; however, it is necessary to improve spatial resolution, sensitivity, and inspection coverage of such probes. Because conventional wire-wound coils are near the limits of current technology, a new approach is needed. The Applied Physics Division has externally proposed a technology involving an array of giant magnetoresistance (GMR) sensors. GMR sensors individually are able to very sensitively detect the magnetic leakage field of current flowing around a tiny crack. An array of GMR sensors would enable significant coverage of the engine part without significantly increasing inspection time. To assist with design, software had to be developed that would quantitatively assist with the design of such sensors. The purpose of this quick-look project was to put SwRI in a position where it would have the necessary software developed.
Approach - The problem involves being able to (1) compute the eddy current flow generated by a horizontal coil and (2) then compute the perturbation in the magnetic field as the current flows around a tiny surface crack. The problem is essentially a two-part problem. In the first part, an analytic solution, originally developed by Dodd and Deeds, was adapted to the geometry of a horizontal coil by resolving the N-turn coil into a geometry of N concentric circular coils. The resulting current density in the plate at different depths inside the plate could then be computed for coils of different radii and different liftoffs and coil excitation currents of different magnitudes and frequencies. In the second part, the computed current density from the first part was used in a different analytical computation, originally developed by Beissner, Sablik, and Teller, to compute the magnetic perturbation field (leakage flux density) caused by current flowing around a surface crack with half-penny-shaped cross section and an oblate spheroidal shape three dimensionally. The mathematics was adapted to the problem at hand, with appropriate approximations. Thus, for a given initial current density, the flux density components could be computed for cracks of different sizes.
Accomplishments - The computed results compare very favorably with experimental test results and should thus be useful in guiding design of the GMR array probes to be developed under an external project. One example of what was learned from the computation is that the GMR detector should be positioned approximately three-quarters of the way out from the center of the coil, where the current density in the specimen peaks.