On Electromagnetic Testing of the Hardness Distribution in Parts Made of Structural Materials

A. V. Anfilofyev

(Presented by the scientific and technical seminar of the Department of Information and Measurement Engineering)

By the distribution of hardness one can assess the technological effectiveness of operations and predict the service capabilities of parts [1].

Numerous studies note a good correlation of hardness with a number of electromagnetic parameters. For a large number of similar parts, when the use of hardness testers is impractical or impossible, electromagnetic methods of hardness testing are widely used. However, when measuring the hardness distribution by electromagnetic methods it is necessary to ensure a locality of testing commensurate with the locality of hardness testers, and the possibility of making measurements sufficiently close to the contour of the product.

In this connection, a number of methodological questions arise, related to the choice of the sensor’s measurement base, its design, and the electromagnetic characteristic correlated with hardness.

On the surface of parts, hardness can be tested by the point-pole method from residual magnetization [2], by coercimeters of the Institute of Physics of Metals of the Academy of Sciences of the USSR according to the demagnetizing current [3], and by spectral analysis of the harmonic signal of the sensor. The indicated methods are implemented mainly with sensors of one- and two-pole design.

The sensor must ensure locality of measurements and register the macro-inhomogeneity of the metal and its gradient that arises in the course of technological operations.

On the other hand, the dimensions of the contact spot must be such that the original annealed material can be considered isotropic (or homogeneous).

Let us establish approximate minimum dimensions of the contact spot. In structural steels the grain sizes are, on average, identical and vary within the limits of 0.01–0.1 mm. A grain, considered in a first approximation as a single crystal, has the maximum possible anisotropy. In a metal the orientation of grains is random, unless special types of treatment have been applied. It is natural to assume that an area on which grains of all orientations are present will be the minimum element starting from which the material may be regarded as homogeneous. With a further increase in the area, grains with orientations already present will occur. For ferromagnetic materials having a cubic lattice, allowing for a misorientation of two grains of 1% (3.6°), the elementary area includes 5000 grains [4], which corresponds to 0.5–50 mm². Since the electromagnetic field penetrates

to a certain depth, and subsurface grains participate in the measurement; the dimensions may be reduced.

The influence of the edge of the specimen was studied using an attached electromagnet (Fig. 1). The distance between the poles of the electromagnet can be varied within 0–20 mm and, in this way, a one- and two-pole transducer can be modeled. The electromagnet is made of plates of electrical steel; the cross section of the poles is \(3 \times 5\) mm, and the magnetizing winding is \(\omega_1 = 2 \times 250\) turns. The maximum induction between the poles is 12,500 G.

Fig. 1. Design of an attached electromagnet

On a specimen \((3 \times 5 \times 150\ \text{mm})\) made of annealed Armco iron with an inhomogeneity of \(1 \div 2\ HV\), a measuring coil \(\omega_2 = 350\) turns was installed. The electromagnet was closed by a ferroprobe transducer \((\omega_1 = 2 \times 200,\ \omega_2 = 800)\). The residual induction after magnetization with the electromagnet installed and after its removal was measured with a ballistic galvanometer. When monitoring with one pole, the residual induction was calculated as the arithmetic mean of two measurements made by pulling the coil off the specimen, the coil being located on one side and on the other side of the pole.

With the electromagnet installed, the influence of the edge was not noticeable. Because of the low residual induction of the specimen, the difference in galvanometer deflections during magnetization and reduction of the current to zero amounted to 2 mm on the scale. The residual induction in the specimen and in the electromagnet are commensurate; the ferroprobe transducer did not register a change in the field. As follows from [3], as the distance between the poles decreases, the leakage flux and the influence of the residual induction of the electromagnet increase, which leads to a decrease in sensitivity. Obviously, a transducer design with such an arrangement of the ferroprobe is ineffective for measurements over small areas.

When the electromagnet is removed, the edge influence is proportional to the distance between the poles and is considerably smaller than for a single pole. The demagnetizing factor in two-pole magnetization is also smaller (Fig. 2). In an alternating field and in the steady-state mode, it is difficult to ensure the required magnetizing fields with a small-sized transducer. The possibility of hardness monitoring was studied when the field was increased from zero to the value at which the third harmonic rises sharply, which corresponds to the inflection point on the metal magnetization curve. In this case the transducer does not have time to heat up.

Fig. 2. Influence of the edge of the article on the residual induction

Figure 3 gives the measurement results for a number of steels when the sensor operates in the indicated mode.

Fig. 3. Dependence of the sensor excitation voltage at which the third harmonic rises sharply on the hardness of the metal

Conclusions

1. In studying the distribution of hardness in parts, it is advisable to use two-pole sensors with a measuring area of not less than \(0.5\ \text{cm}^2\).

2. In a constant field, testing must be carried out by the residual magnetization of the metal with the electromagnet removed.

3. In an alternating field, hardness can be tested by the magnitude of the sensor excitation field at which the third harmonic rises sharply.

LITERATURE

1. G. D. Del. Determination of Stresses in the Plastic Region from the Distribution of Hardness. Moscow, Mashinostroenie, 1971.

2. G. S. Tomilov. Magnetic testing of the structure and hardness of steel parts by measurements of the local field of residual magnetization. Defektoskopiya, 1966, No. 4.

3. V. A. Zakharov, M. N. Mikheev, V. P. Tabachnik, V. M. Frantsevich, L. A. Fridman. Influence of variation in the gap between the article and the electromagnet pole pieces on the magnitude of the demagnetizing current of a ferrosonde coercimeter with an attached electromagnet. Defektoskopiya, 1970, No. 4.

4. A. A. Ilyushin, V. S. Lenskii. Strength of Materials. Moscow, Fizmatgiz, 1959.