VIBRATION DIAGNOSTICS OF FATIFUE DAMAGE IN STRUCTURAL ELEMENTS OF GAS TURBINE ENGINES AT THE REPAIR STAGE

Abstract

Vibration diagnostics of damage belongs to the class of non-destructive methods, which usually do not take long time. However, the main problem of vibration diagnostics is relatively low sensitivity to the critical damage of fatigue crack type, which arises because of long time accumulation of plastic deformation. To improve the sensitivity and reliability of vibration diagnostics of damage two methods were considered. First method was based on the fact, that a characteristic feature of vibrations of structural elements with fatigue crack is the occurrence of non-linear resonances (sub- and super-harmonic) and significant non-linearity of vibration response at these resonances. Secon one – on the fact, that quite noticeable in certain cases increase of damping characteristic caused by a crack can be observed. Analytical and experimental studies of these methods were carried out as applied to the blades of aircraft gas turbine engines. As a result of the studies, the intensity of change in parameter of super-harmonic resonance and in damping characteristics at different parameters of crack was determined. Besides, the experimental techniques for vibration testing of turbine blades were developed. There was demonstrated, that the sensitivity of both considered methods of vibration diagnostics is several orders of magnitude higher than the sensitivity of conventional methods based on the change in natural frequencies and mode shapes, and they can be effectively used for the diagnostics of blades on the stage of engine repair. All the other conditions being equal, second-order superharmonic resonance demonstrates a higher sensitivity to the presence of cracks than the damping characteristic. The smallest crack with an area of 0.2% considered in the work causes a reliably recorded non-linearity of vibration response at superharmonic resonance of order of 2/1. At this the change of damping characteristic slightly exceeds the error of experiment.