VALIDATION OF MULTI-FREQUENCY EDDY CURRENT MWM SENSORS AND MWM-ARRAYS FOR COATING PRODUCTION QUALITY AND REFURBISHMENT ASSESSMENT

摘要

Coatings for oxidation, corrosion, and thermal protection provide the required materials performance for gas turbine blades and vanes in state-of-the-art industrial gas turbines. These turbines must withstand severe operating conditions for well over ten thousand hours. Variations in the coating thickness, and increased porosity, can influence the lifetime of such coatings significantly. For components that have been removed from service, effective assessment of the aged coating and substrate condition is critical for refurbish/replace/continue-to-run decisions. A suitable device for coating thickness measurement and detection of unacceptable porosity is needed for ensuring the quality of such coatings. In this paper, we present new results on coating thickness measurements for metallic MCrAlY overlay coatings on gas turbine parts. These measurements were performed with a Meandering Winding Magnetometer (MWM~(~R)) eddy-current sensor using grid methods. This technique allows proper coating measurements even after a diffusion heat treatment for a better coating adhesive strength. The MWM technology enables measurement of the coating thickness, the absolute electrical conductivity (which may in turn be related to porosity or other properties of interest), and lift-off, which is related to surface roughness. Single-channel MWM sensors and multi-channel imaging MWM-Arrays permit capture of features of interest for a population of components. New capabilities for inspecting gas turbine components are, thus, provided. Inspection applications include metallic and non-metallic coating thickness measurements, porosity measurements, and detection of cracks on complex surfaces. Results of coating assessment for a production line of gas turbine vanes by means of a multifrequency MWM technique are presented for various combinations of coatings and base metals. A description of improved multiple frequency quantitative inversion methods is provided for simultaneous and independent measurement of multiple unknowns such as metallic bond coat thickness, metallic bond coat porosity, and top coat thickness. Ongoing research focuses on characterization of aged components using MWM sensors and imaging MWM-Arrays as well as on development of enhanced algorithms for four and five unknown coating / substrate properties. In a recent study of hot corrosion, uncoated nickel alloy specimens were characterized using an MWM sensor with grid methods. Preliminary results indicated that, within the limitations of the three-unknown single-layer model used, the method could readily identify specimens with no apparent corrosion damage, specimens with moderate corrosion damage, and specimens with severe corrosion damage.