INFLUENCE OF THE MOTION OF AN ARTICLE ON THE DETECTABILITY OF ITS DEFECTS IN THE TWO-FREQUENCY INSPECTION METHOD

V. I. Urusov

(Presented by the scientific and technical seminar of the Department of Information and Measuring Technology)

The question of the influence of the motion of an article on the detectability of its defects by any electromagnetic method is very important in the development of flaw-detection equipment used in continuous-flow production. It is no accident, therefore, that so much attention is being devoted to the study of this question. Thus, for example, in [1] it is shown that signals from defects, taken with the aid of an indicator in the form of an induction coil with a ferromagnetic core, at speeds of motion up to 15 m/sec always increase in comparison with the static case. Similar results were obtained in [2] when a ferroprobe-polymer was used as the defect indicator. However, in [3] it is asserted that signals from defects, taken with a ferroprobe gradiometer under dynamic conditions, are 2–10 times smaller than the magnitudes of signals taken under static conditions.

In contrast to the magnetic-induction and ferroprobe methods, which are based on reading the magnetic leakage fields over defects, in the two-frequency method, with the aid of a high-frequency field, the change in the magnetization of the article in the defect zone, produced by a low-frequency field, is recorded. Therefore, the nature of the influence of the speed of motion of the article on the detectability of its defects in the two-frequency method may be different from that in magnetic inspection methods.

In order to determine this nature, we carried out corresponding experimental investigations on an installation whose block diagram is described in [4]. The objects of investigation were disks of carbon steels with thicknesses from 2 to 10 mm, on which surface and subsurface defects had been made. The parameters of these defects are shown in Fig. 1. The detectability of the defects was evaluated by the ratio Ac/Aш. The parameters Ac and Aш were taken from the screen of an oscilloscope whose sweep was synchronized with the rotational speed of the disk. In the investigations, absolute (Fig. 2, a) and differential self-comparison sensors (Fig. 2, b) were used, having a special winding for creating an additional constant field Ho.

As a result of the investigations it was established that the use of two-frequency sensors of absolute value for inspecting defects in moving articles has a fundamental difficulty, caused by the fact that already under static conditions the output signal of the sensor, for which the mean value of the second high-frequency harmonic was selected, ...

...is modulated at twice the frequency of the low-frequency field, which makes it difficult to isolate the voltage amplitude-modulated by the signal from the defect. Taking this difficulty into account, the principal studies aimed at determining the nature of the influence of the motion of the product on the detectability of its defects were carried out using differential self-comparison transducers.

Fig. 1. Parameters of artificial surface (a) and deep (b) defects

In this case, the measuring coils of the transducers were arranged so that the line connecting the centers of the coils was directed either parallel or perpendicular to the velocity vector of the product’s motion.

Fig. 2. Design of an absolute (a) and a differential self-comparison transducer (b)

Studies with transducers in which the line connecting the centers of the coils was directed parallel to the velocity vector showed that, as the speed of motion of the product increases, the ratio \(A_c/A_{\text{sh}}\) for surface and deep defects decreases monotonically owing to an increase in \(A_{\text{sh}}\) (Fig. 3, a). The increase in \(A_{\text{sh}}\) is explained by a violation of the identity of the conditions in which the measuring coils are located during motion of the product, caused by a displacement of the maximum of the low-frequency field near the product in the direction of motion. This leads to decompensation of the output voltages of the measuring coils, which increases \(A_{\text{sh}}\) and, consequently, worsens the detectability of defects \((A_c/A_{\text{sh}})\).

With the line connecting the centers of the measuring coils perpendicular to the velocity vector of motion, compensation of the coils’ output voltages was not observed, since in this arrangement the measuring coils are in identical conditions when the product moves relative to the probe. Studies showed that, with this arrangement, increasing the speed of motion of the product from 0 to 15 m/sec has practically no effect on the value of the ratio \(A_c/A_{\text{ш}}\) for surface and internal defects lying at a depth of up to 3–4 mm from the surface (Fig. 3, b). Therefore, when round two-frequency surface-mounted probes are used to inspect defects in moving products, precisely this arrangement of the measuring coils should be recommended.

Fig. 3. Influence of motion of the product on the detectability of its defects

It should be noted that, for both methods of arranging the measuring coils, increasing the speed of motion of the product to 15 m/sec causes a slight decrease in the amplitude of the signals \(A_c\) from surface and subsurface defects (no more than 5–7%) compared with the static case and strongly affects their duration. As for internal defects lying at a depth of no more than 3–4 mm from the surface, they exhibit a noticeable decrease in the amplitude of the signal \(A_c\) as the speed increases, and this decrease is the greater the deeper the defect is located. This phenomenon is explained by the fact that, when a moving product is magnetized by an additional constant field \(H_0\), eddy currents are induced in it, exerting a substantial influence on the process of magnetization of the deep layers of the product. The surface effect that arises does not permit the deep layers to be magnetized and, consequently, defects lying at a considerable depth from the surface to be detected. Thus, at \(V = 15\) m/sec, defect No. 13 was not detected against the background of noise, whereas under static conditions it was detected with a ratio \(A_c/A_{\text{ш}}\) of not less than four under the same inspection conditions.

Increasing the magnitude of the constant magnetizing field and the diameter of the coil producing this field makes it possible to improve the resolving capability of the two-frequency method for deep defects.

References

1. V. V. Vlasov. Studies on high-speed electromagnetic rail flaw detection. FMM, vol. 5, no. 2, 1957.

2. B. P. Dovnar, V. A. Shcherbinin. Investigation of defect fields in high-speed electromagnetic flaw detection of rails. 1956, Defektoskopiya, No. 1.

3. N. S. Savorovsky, G. S. Tropnikov. A contactless system for transverse magnetization of pipes. 1970, Defektoskopiya, No. 2.

4. V. I. Urusov. Investigation of the two-frequency method under conditions of motion of the inspected product. The present collection.