Ultrasonic Measuring Of The Thickness Of NiCoCrAlYTa Six-element-coating

Turbine engines experience more harsh working environment, with the flying speed requirements continue to increase. In order to improve the reliability and extend the life of the turbine engine, People often used coating technology to improve the working conditions of turbine blades. The reliability of the turbine blade will be affected because of the types of defects and uneven coating thickness in the preparation process of the coating. Therefore, the coating thickness testing is particularly important.In this paper, the NiCo Cr AlYTa regard as the research object. NiCo CrAlYTa composed by the Ni,Co,Cr,Al,Y,Ta.. With a strong resistance to high temperature oxidation and hot corrosion. For spraying this coating often used in thermal spraying or low pressure plasma spraying.It used destruct method of coating thickness sampling in the practical engineering. To achieve low-cost, high-efficiency automatic coating thickness detection,Ultrasonic Testing has been studied in this paper.Since the coating thickness is less than the half-wave length of the ultrasonic, the coating thickness can not be detected by conventional ultrasonic. Immersion auto-scan equipment has been designed and use it to collect the signals of coating. Study use44-scale Marr wavelet analysis and Hilbert-Huang Transform to detect mutated position of coating time-domain signals after analyzing the difference of physical properties between the coating material and the base material and signal will produce abrupt change at the interface between coating and substrate. Using ABAQUS software to calculate longitudinal wave velocity of coating material and substrate material. The systematic error is corrected by experiment. The coating thickness was calculated by the time interval of mutation and the corrected longitudinal wave velocity of the coating. Then, the outcomes of wavelet analysis and Hilbert Huang transform were compared by the results of metallographic measuring thickness and verify the reliability of the two coating time-domain signal processing methods.The results of the study indicate :Longitudinal wave velocity is not consistent which influenced by coating process and other factors. The uneven coating thickness phenomenon exists in the test samples by metallography measurement. Thickness of the coating in the range of 65μm~120μm. The average error of the coating thickness is less than 5% by wavelet analysis, and the average error is about 10% by Hilbert-Huang transform. The error mainly comes from the Hilbert Huang transform which has theeffect of filtering and reducing the sampling rate. Two signal processing methods can be used to detect the Mutation position of the time-domain signal, Hilbert Huang transform thickness measurement data is generally larger than the wavelet analysis method,the thickness measured by Wavelet analysis is close to the metallography. In this paper, the wavelet analysis method is applied to the measurement of the thickness of the turbine blade. An effective method is provided for the nondestructive measurement of the coating thickness.