Installation and test procedure for frequency stability under cyclic loading of metals and alloys

Abstract

Manufacture of parts operating under difficult conditions of cyclic loading, as well as products with stable dimensions necessitates the use of materials with minimal manifestations of inelastic properties. Study of these materials suggests conducting specialized narrowly focused tests using newly developed machines and installations with appropriate experimental techniques. We present the design, operation and control features of the original electromagnetic installation for testing fatigue and frequency stability and operating in a self- oscillating mode, which provides the equality of the frequency of cyclic loading and natural frequency of the sample. The system of the installation control contains two closed circuits: for excitation of self-oscillations and for stabilization of the oscillation amplitude. The sample is loaded by electromagnetic force, and unloading occurs due to the elastic forces of the material. The methodology and algorithms for calculating the stresses of samples of various geometric shapes for estimating changes in amplitude-frequency characteristics are presented. The calculated relationship between the force applied to the sample and its movement at the point of application of force is derived with subsequent determination of the stress by a known force. The results of calibration tests for the static mode of sample loading are presented and the forces acting on the sample (external, inertia, and elastic forces) are evaluated taking into account the maximum stress and maximum strain amplitude. Static and cyclic loading modes are compared. The frequency characteristics obtained during testing steel samples according to the proposed method are obtained. The experimental results of tests with interruptions in the process of cyclic loading and continuous testing are analyzed. It is shown that interruptions in cyclic tests lead to a jump-like increase in the frequency, whereas continuous tests revealed no jumps. At the same time, a comparative analysis with the results of continuous tests showed that the overall frequency deviation is approximately the same for the entire operating cycle in both cases. It is shown that an increase in the frequency after rest is random and does not depend on the number of operating cycles.